The current state of molecular testing in the treatment of patients with solid tumors, 2019

The world of molecular profiling has undergone revolutionary changes over the last few years as knowledge, technology, and even standard clinical practice have evolved. Broad molecular profiling is now nearly essential for all patients with metastatic solid tumors. New agents have been approved based on molecular testing instead of tumor site of origin. Molecular profiling methodologies have likewise changed such that tests that were performed on patients a few years ago are no longer complete and possibly inaccurate today. As with all rapid change, medical providers can quickly fall behind or struggle to find up-to-date sources to ensure he or she provides optimum care. In this review, the authors provide the current state of the art for molecular profiling/precision medicine, practice standards, and a view into the future ahead.

Molecular Testing and Its Evolution

Comprehensive molecular profiling of patient tumors has been widely studied over the last few years in a variety of cancers, leading to the development of a new discipline termed “personalized” or “precision” medicine. Molecular profiling is becoming standard practice for most patients with advanced disease, replacing the historical treatment paradigm of prescribing standard chemotherapy based upon the tumor’s organ of origin, histology, and stage. This approach has allowed oncologists to reorganize the way they think about cancer and to make treatment recommendations based upon genomic drivers of tumorigenesis. In some cases, this has produced dramatic, positive outcomes, including complete remissions, even in the setting of treatment-refractory disease, delighting both patients and their caregivers.

The molecular profiling field is evolving rapidly. We are now shifting our focus from a few small, predictive, disease-specific, evidence-based tests—chosen “a la carte”—to broader panel testing that measures levels of changes in myriad “genes or gene products.” These genomic changes can serve as biomarkers of both response prediction (indicating tumor and patient outcome/response to a specific therapy) and a patient’s prognosis (describing innate tumor aggressiveness, which aligns with patient survival regardless of treatment received). Increasing numbers of biomarkers have been identified for which targeted drugs are being discovered and exploited therapeutically. Scientific advances go hand-in-hand with technological advances, which lead to improved therapeutic choices, all of which have garnered US Food and Drug Administration (FDA), Centers for Medicare & Medicaid Services (CMS), and insurance company attention. Growing acceptance of evidence-based biomarker testing for the purpose of targeting treatment to solid tumors has ensued. Notably, Foundation Medicine’s FoundationOne CDx assay, which tests for several well-known markers using next-generation sequencing (discussed later in this review), was recently approved by the FDA and concurrently accepted by the CMS.1, 2

To facilitate cancer therapy, it is important to distinguish between germline abnormalities and somatic abnormalities. A very good example of this is the recently incorporated BReast CAncer gene () germline testing for all patients with pancreatic cancer. Germline testing involves an extensive coverage of , whereas current somatic testing covers only certain regions of that gene. As mutation analysis evolves into whole exome sequencing, coverage of germline and somatic testing will be similar if not identical. Given the increased need for somatic testing in patients with pancreatic cancer, it is possible that whole exome sequencing will replace germline testing in guidelines to come. As these “standard” tests evolve, they make the choices facing patients and providers more complex while providing hope that harnessing this knowledge will translate into substantial benefits for patients, including cancer cures and prevention.

Molecular Profiling and Its Methodology

Molecular profiling refers to the assessment of DNA, RNA, and/or proteins within an individual patient’s cancer using cells obtained from a tumor biopsy or through the capture of tumor cells circulating in the bloodstream, with the latter being less well established as a methodology. The term “molecular profiling” was initially applied to DNA analysis but evolved with advances in technology to take on a broader meaning to encompass analyses of RNA and proteins. DNA-level alterations do not necessarily lead to biological alterations, thus making examination at the “multiomic” (transcriptome and proteome) level imperative. This multipronged analysis results in the generation of an inordinate amount of data that can be processed only with the help of bioinformatic methodology. Bioinformaticians combine a host of scientific and mathematical data to create a computer infrastructure that assists in the analysis and interpretation of biological data and picks out correlations between certain gene mutations and response to a specific therapy.3 Currently used molecular profiling techniques are as follows:

DNA and RNA

• (PCR) is used to amplify and detect DNA and RNA sequences. Standard PCR involves the amplification of one or more copies of a chosen DNA sequence to produce millions of copies and enable detection and analysis. Reverse transcription PCR converts RNA templates into complementary DNA for molecular analysis.
• (ISH) localizes and determines a specific DNA or RNA sequence in a tissue section (in situ) or in circulating tumor cells using a labeled complementary DNA, RNA, or modified nucleic acid strand probe. This technique detects gene deletions, amplifications, translocations, and fusions. Gene fusions commonly occur in epithelial cancers as a result of genomic rearrangements or abnormal mRNA processing. ISH techniques include chromogenic ISH and fluorescence in situ hybridization (FISH).
  • (CISH) uses brightfield microscopes for label detection.
  • uses fluorescence microscopes for label detection.
• examines strands of DNA to identify mutations by analyzing long, contiguous sequencing reads. This DNA sequencing takes place according to the selective incorporation of chain-terminating dideoxynucleotides by DNA polymerase during in vitro DNA replication. This was the primary sequencing method used for well over 20 years and, although it is still widely used, next-generation sequencing (NGS) is now preferred for multigene/variant assessment.
• is a high-throughput technique that rapidly examines and more broadly detects DNA mutations (often used for circulating tumor DNA), copy number variations (CNVs), and gene fusions (using an RNA sequencing panel) across the genome. NGS can be performed on a range of cancer types using blood, solid tissue, and bone marrow samples. Precise tissue collection and workup are necessary for accurate results. Laboratory regulatory agencies constantly provide updated guidance documents pertaining to the design, development, and use of NGS-based tests, recognizing the importance of NGS in cancer diagnostics and therapeutics.
• detects and quantifies mutations, methylation, etc, through sequencing by synthesis—a method that performs DNA sequencing by detecting the nucleotide that is incorporated by DNA polymerase.
• detects changes in DNA (eg, the length of a specific DNA sequence) or RNA to indicate the presence or absence of an inserted or deleted genomic sequence.

Protein

• uses the principles of antibody binding to proteins to determine the levels of protein expression in tissue samples. Tumor-related proteins of interest can include tumor-specific antigens, protein products of oncogenes and tumor suppressor genes, tumor cell proliferation markers, and enzymes.

Molecular Profiling Assays and Why Physician Oncologists and Pathologists Should Be Familiar With Them

Modern approaches to tumor profiling assess DNA, RNA, and proteins to form a detailed molecular map to guide more precise and individualized treatment decisions. Because the field of molecular profiling is continually evolving, physician education is vital. Clinical oncologists and pathologists benefit greatly from an understanding of the technology involved, possibly even gaining hands-on experience in molecular profiling assays and their interpretation. Any treating physician should know what, when, and how to test and how to make subsequent informed, patient-personalized treatment decisions.4, 5 Correct interpretation of profiling results is critical; many fear that overinterpretation or misinterpretation will lead to treatment of patients with ineffective but expensive therapies, negatively affecting not only patient lives but also the health care budget. Laboratories offering broad molecular profiling services should be suitably Clinical Laboratory Improvement Amendments (CLIA)–certified for this exact purpose to ensure quality control (see CLIA-approved laboratories offering molecular panel analysis, below). However, even CLIA-certified laboratories do not use identical methodologies and techniques, which can still lead to variable results. Reproducibility is key, and the rationale behind assay cutoff limits should be strong. Even before a patient sample is submitted for profiling, the pathologist or the treating physician—whoever plays the lead role in any particular institution—must ensure quality-controlled tissue sample collection.6, 7 Reputable molecular testing laboratories will advise on the exact set of tumor profiling tests to perform, how to process samples, and how to interpret the final generated report, which is created to inform physicians of treatment choices for their patients. Still, physician education is key to such a critical set of processes.

Biomarker Testing in the Clinic

Targeted therapies are showing efficacy in the right subgroups of patients. Of course, these subgroups must be defined, and this process is becoming more accurate and efficient with evolving molecular testing methods and broader use in research and in the clinic. As this process improves, treatment options will improve for an increasing number of patients while eventually emerging as a more cost-effective, generally beneficial option compared with the currently accepted trial-and-error treatment model.

The biomarker information within Tables 1 through 2.12 is based mainly on the National Comprehensive Cancer Network (NCCN) Clinical Practice Guidelines in Oncology (NCCN Guidelines), NCCN Biomarkers Compendium (NCCN.org, Accessed February 6, 2019), and FDA recommendations and approvals (for the full definitions of all genes, please see the Supporting Information). Although the NCCN Biomarkers Compendium details not only predictive but also prognostic, diagnostic, screening, monitoring, and surveillance markers, the focus of this current review is on predictive biomarkers that can be used to guide treatment decisions. Within Tables 1 through 2.12, the classifications in the “evidence” columns are based on the level of clinical evidence available and the degree of consensus among NCCN panel and other experts. In some cases, clinical evidence comes from large, well-designed, randomized controlled trials, but in many cases, it is mostly based on data from indirect comparisons among randomized trials, phase 2 or nonrandomized trials, multiple smaller trials, retrospective studies, or merely clinical observations. In some cases, substantial clinical data are lacking and evidence comes from clinical experience alone. On the basis of all these factors and how compelling the data are, the evidence is rated as:

Table 1.

Predictive Microsatellite Instability/Mismatch Repair Testing for Any Solid Tumor

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

MMR

Expression

See in the text

IHC

dMMR and MSI-H tests on available tissue are recommended to predict response to pembrolizumaba

Lower level; wide acceptance

All

, , , or

Mutation (= dMMR expression)

NGS

Where applicable, dMMR and MSI-H tests are used together to identify whether a patient should undergo further mutation testing for Lynch syndromeb

MSI

Testing (changes in short repeated DNA sequences)

See in the text

PCR, NGS

dMMR and MSI-H tests on available tissue are recommended to predict response to pembrolizumaba

Lower level, wide acceptance

All

Where applicable, dMMR and MSI-H tests are used together to identify whether a patient should undergo further mutation testing for Lynch syndromeb

Table 2.1.

Currently Recommended Predictive Molecular Testing for NSCLC

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

ALK

Gene fusion

Metastatic workup

FISH, NGS, RT-PCRa

Response to oral ALK TKIs; alectinib has improved efficacy over crizotinib in first line

High-level, wide acceptance

Adenocarcinoma, large cell, NSCLC NOS

Fusion protein expression

IHCb

Together with testing in “never smokers” or small/mixed histology specimens

Response to oral ALK TKIs, eg, crizotinib

Lower level, wide acceptance

Squamous cell

T790M

Mutation

Metastatic workup

NGS, multiple mutation testing

Resistant to EGFR TKIs

High-level, wide acceptance

Adenocarcinoma, large cell, NSCLC NOS

exon 21 (L858R, L861), exon 20 (S768I), exon 18 (G719X, G719)

Mutation

Metastatic workup

NGS, multiple mutation testing

Sensitive to EGFR TKIs

High-level, wide acceptance

Adenocarcinoma, large cell, NSCLC NOS

Lower level, wide acceptance

Squamous cell

exon 19

Deletion

Metastatic workup

NGS, multiple mutation testing

Sensitive to EGFR TKIs

High-level, wide acceptance

Adenocarcinoma, large cell, NSCLC NOS

Lower level, wide acceptance

Squamous cell

exon 20 7p12

Insertion mutation

Metastatic workup

NGS, multiple mutation testing

Likely resistant to EGFR TKIs

High-level, wide acceptance

Adenocarcinoma, large cell, NSCLC NOS

Lower level, wide acceptance

Squamous cell

Fusion rearrangement

Metastatic workup

NGS, FISH, RT-PCR

Responsive to ROS1 TKIs

Lower level, wide acceptance

Adenocarcinoma, large cell, squamous cell, NSCLC NOS

PD-L1

Protein expression ≥50%

Metastatic workup

NGS, multiple mutation testing

Response to pembrolizumab in first-line; FDA approved treatment15

Lower level, wide acceptance

Adenocarcinoma, large cell, NSCLC, squamous cell NOS

Mutation

Metastatic workup

Gene sequencing

Resistance to EGFR TKIs. Gives poor prognosis compared with wt

Lower level, wide acceptance

All NSCLC

Mutation, V600E

Metastatic workup

NGS, pyrosequencing, AS-PCR

Emerging targeted agents19

Lower level, wide acceptance

All NSCLC

Mutation

Any time

NGS, multiple mutation testing

Emerging targeted agents20

Lower level, limited acceptance

All NSCLC

Amplification, mutation

Any time

NGS, FISH

Emerging targeted agents21

Lower level, wide acceptance

All NSCLC

Fusion, rearrangement

Any time

NGS, FISH, RT-PCR

Emerging targeted agents22, 23

Lower level, wide acceptance

All NSCLC

Table 2.2.

Currently Recommended Predictive Molecular Testing for Colon and Rectal Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

/a

Mutation

Workup for metastatic disease (suspected or proven)

NGSb

Avoid cetuximab or panitumumab treatment in patients who have tumors with and mutations (exons 2, 3, and 4 in both)

NCCN indicate , but many believe classification is

Metastatic synchronous adenocarcinoma (any T, any N, M1), suspected or documented;

Metachronous metastases by CT, MRI, and/or biopsy, documented

a

Mutation V600E

Workup for metastatic disease (suspected or proven)

NGS, pyrosequencing, AS-PCRb

Cetuximab or panitumumab treatment is not recommended in patients who have tumors with V600E mutations unless given with a BRAF inhibitor such as vemurafenib

NCCN indicates , but many believe classification is

Metastatic synchronous adenocarcinoma (any T, any N, M1), suspected or documented;

The use of irinotecan in combination with cetuximab or panitumumab plus vemurafenib is recommended in all patients with previously treated mCRC

Metachronous metastases by CT, MRI, and/or biopsy, documented

Table 2.3.

Currently Recommended Predictive Molecular Testing for Gastric, Esophageal, and Gastroesophageal Junction Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

Gene amplification

Workup any time

(F)ISH

Particularly if trastuzumab therapy is being considered

Lower level, wide acceptance

Gastric, esophageal, and gastroesophageal junction cancers

PD-L1 (CD274) and HER2 protein

Expression

Workup any time for suspected or documented, inoperable, locally advanced, recurrent, or metastatic adenocarcinoma

IHC, FISH

HER2-negative status corresponds with higher PD-L1 expression rates; together with MMR, HER2 is a potential biomarker for anti–PD-L1 therapy

Lower level, wide acceptance

Gastric, esophageal, and gastroesophageal junction cancers

Table 2.4.

Currently Recommended Predictive Molecular Testing for Pancreatic Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

and

Mutation (somatic and germline)

Initial workup if the patient has a strong family history on initial diagnosisa

NGS

A known germline mutation could help guide therapy (eg, PARP and other DDR enzyme inhibitors). In October 2018, olaparib was approved for the treatment of patients with germline -mutated, metastatic pancreatic cancer that has not progressed after first-line, platinum-based chemotherapy

Nine percent of pancreatic cancers harbor a germline or somatic or mutation, and this has an impact on response to therapy. testing in patients who are still responsive to cytotoxic therapy is becoming standard practice. The use of PARP inhibitors, specifically olaparib, in these patients is an option

Pancreatic adenocarcinoma

Table 2.5.

Currently Recommended Predictive Molecular Testing for Prostate Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

and

Mutation (somatic and germline)

Initial workup: If the patient has a strong family history on initial diagnosisa

NGS

A known germline mutation could help guide therapy (eg, PARP and other DDR enzyme inhibitors)

Lower level; wide acceptance

Prostate cancers

If the patient has metastatic, castration-resistant disease

Germline mutation

Initial workup showing strong family history

If patient has metastatic castration-resistant disease

NGS

NCCN guidelines recommend inquiring about known / mutations in a patient’s family for prostate cancer early detection6163/2 and is recommended

Lower levelb

Prostate cancers

Known / and germline mutations could help guide therapy with PARP and other DNA damage–response enzyme inhibitors

Table 2.6.

Currently Recommended Predictive Molecular Testing for Endometrial Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

ESR1 (ER)

Expression

In the stage III, IV, and recurrent disease settings

IHC

ER positivity predicts response to endocrine therapy

Lower level, wide acceptance

Uterine neoplasms, endometrial carcinoma

PMS2 (Lynch syndrome, MMR gene)

Expression

Upon diagnosis or upon recurrence if not previously tested

IHC

Loss of positivity indicates MMR, possible Lynch syndrome, and susceptibility to checkpoint inhibitors

Recommended by SGO Clinical Practice Statement

Uterine neoplasms, endometrial carcinoma

Table 2.7.

Currently Recommended Predictive Molecular Testing for Ovarian Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

and

Mutation

Recurrent disease; initial workup if the patient has a strong family history on initial diagnosis

NGS

Include other homologous recombination pathway genes and MSI or DNA MMR; helps guide therapy (eg, PARP or other DDR enzyme inhibitors; chemotherapy response)

Lower level, wide acceptance

Ovarian cancer

Mutation

Recurrent disease; initial workup if the patient has a strong family history on initial diagnosis

NGS

Include other homologous recombination pathway genes and MSI or DNA MMR; helps guide therapy (eg, PARP or other DDR enzyme inhibitors; chemotherapy response)

Lower level, wide acceptance

Ovarian cancer

Mutation

Recurrent disease; initial workup if the patient has a strong family history on initial diagnosis

NGS

Include other homologous recombination pathway genes and MSI or DNA MMR; helps guide therapy (eg, PARP or other DDR enzyme inhibitors; chemotherapy response)

Lower level, wide acceptance

Ovarian cancer

Mutation

Recurrent disease; initial workup if the patient has a strong family history on initial diagnosis

NGS

Include other homologous recombination pathway genes and MSI or DNA MMR; helps guide therapy (eg, PARP or other DDR enzyme inhibitors; chemotherapy response)

Lower level, wide acceptance

Ovarian cancer

Mutation

Recurrent disease; initial workup if the patient has a strong family history on initial diagnosis

NGS

Include other homologous recombination pathway genes and MSI or DNA MMR; helps guide therapy (eg, PARP or other DDR enzyme inhibitors; chemotherapy response)

Lower level, wide acceptance

Ovarian cancer

,

Mutation

Recurrent disease; initial workup if the patient has a strong family history on initial diagnosis

NGS

Include other homologous recombination pathway genes and MSI or DNA MMR; helps guide therapy (eg, PARP or other DDR enzyme inhibitors; chemotherapy response)

Lower level, wide acceptance

Ovarian cancer

Table 2.8.

Currently Recommended Predictive Molecular Testing for Breast Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

ER

Expression

Noninvasive, invasive, early stage, metastatic

IHC

Predictor of response to endocrine therapy

High-level, wide acceptance

Noninvasive and invasive breast cancer, stage I-IV

PR

Expression

Invasive, early stage, metastatic

IHC

Predictor of response to endocrine therapy

High-level, wide acceptance

Invasive breast cancer, stage I-IV

Gene amplification

Invasive, early stage, metastatic

ISH

Predictor of response to HER2-targeted therapy such as trastuzumab, pertuzumab, lapatinib, or trastuzumab emtansine

High-level, wide acceptance

Invasive breast cancer, stage I-IV

HER2 (ERBB2)

Protein expression

Invasive, early stage, metastatic

IHC

Predictor of response to HER2-targeted therapy such as trastuzumab, pertuzumab, lapatinib, or trastuzumab emtansine

High-level, wide acceptance

Invasive breast cancer, stage I-IV

and

Germline mutation

Metastatica

NGS

Predictor of response to PARP inhibitor

High-level, wide acceptance

Invasive breast cancer, stage IV

Oncotype Dx

Gene expression

Hormone receptor-positive, HER2-negative

RT-PCR

Prognostic for recurrence in lymph node–negative ER-positive/HER2-negative; predictive of chemotherapy benefit in lymph node–negative ER-positive/HER2-negative

High-level, wide acceptance

Stage I, II ER-positive/PR-positive, HER2-negative

Table 2.9.

Currently Recommended Predictive Molecular Testing for Central Nervous System Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

1p19q Co-deletion75

Chromosome deletion

Preadjuvant therapy

FISH, PCR

Also prognostic and diagnostic: helps confirm the oligodendroglial character of tumors with ambiguous histologic features; the 1p19q co-deletion provides a good prognosis and predicts response to alkylating chemotherapy alone and in combination with radiation

Lower level, limited acceptance

Adult, low-grade, infiltrative, supratentorial astrocytoma/oligodendroglioma; anaplastic gliomas/glioblastoma

,

Mutation

Preadjuvant therapy glioma workup

NGS, IHC

Also prognostic and diagnostic: and mutations are related with a favorable prognosis and help in clinical trials; infiltrative gliomas that are wild-type or are likely aggressive tumors; or mutations are associated with survival benefit when treated with radiation or alkylator chemotherapy and are commonly associated with promoter methylation

Lower level, limited acceptance

Adult, low-grade, infiltrative, supratentorial astrocytoma/oligodendroglioma; anaplastic gliomas/glioblastoma

76

Promoter methylation

Preadjuvant therapy

Methylation-specific PCR, pyrosequencing

Also prognostic: strongly associated with 75 wild-type tumors; used for risk stratification in clinical trials; used in treatment decisions for elderly patients with high-grade gliomas (grades III-IV); any patient with an promoter-methylated glioblastoma obtains greater benefit from treatment with temozolomide than patients without promoter methylation

Lower level, limited acceptance

Adult, low-grade, infiltrative, supratentorial astrocytoma/oligodendroglioma; anaplastic gliomas/glioblastoma

Table 2.10.

Currently Recommended Predictive Molecular Testing for Sarcomas

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

, a

Amplification

At diagnosis

NGS

Possible clinical trial with CDK4/CDK6 inhibitor

Wide acceptancea

Well-differentiated liposarcoma, dedifferentiated liposarcoma

a

Mutation

At diagnosis

NGS

Possible trial with IDH1 inhibitor

Wide acceptancea

Chondrosarcoma

Table 2.11.

Currently Recommended Predictive Molecular Testing for Head and Neck Cancers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

PD-L1

Protein expression

Metastatic workup

IHC

Recurrent, unresectable, or metastatic (with no surgery or radiation therapy option); second-line or subsequent therapy options: pembrolizumab for PD-L1–positive disease

Lower level, wide acceptance

Cancer of the nasopharynx

Table 2.12.

Currently Recommended Molecular Testing for Melanomas: Predictive and Risk Assessment Biomarkers

BIOMARKER

TEST DETECTS

WHEN

TECHNOLOGY

RECOMMENDATIONS

EVIDENCE

CANCER TYPE

BRAF

Gene mutation, V600E

Workup any time

NGS, pyrosequencing, AS-PCR

Not recommended for patients with cutaneous melanoma who otherwise have no evidence of disease (except to guide therapy)

Lower level, wide acceptance; PREDICTIVE

Melanoma

Protein expression

IHC

Gene mutation

Workup for metastatic or recurrent disease

NGS, pyrosequencing, AS-PCR

Ascertain alterations in and from either biopsy of the metastases (preferred) or archival material if targeted therapy is under consideration

Lower level, wide acceptance; PREDICTIVE

Melanoma

Protein expression

IHC

Mutation

Workup for metastatic or recurrent disease

NGS

Ascertain alterations in and from either biopsy of the metastases (preferred) or archival material if targeted therapy is under consideration

Lower level, wide acceptance; PREDICTIVE

Melanoma

Mutation

Follow-up; risk assessment (predisposing mutation)

NGS

If the individual has personal or familial incidence of 3 or more cases of invasive melanoma or a mix of invasive melanoma, pancreatic cancer, and/or astrocytoma, consider testing with other genes that can harbor melanoma-predisposing mutations (eg, , , , and )

Lower level, wide acceptance; RISK ASSESSMENT

Melanoma

1. Based upon high-level evidence, there is uniform NCCN and other expert consensus that the intervention is appropriate (high-level, wide acceptance).

2A. Based upon lower level evidence, there is uniform NCCN and other expert consensus that the intervention is appropriate (lower level, wide acceptance).

2B. Based upon lower level evidence, there is some NCCN and other expert consensus that the intervention is appropriate (lower level, limited acceptance).

Infrequent but Important Site-Agnostic Biomarkers

Microsatellite instability-high tumors and DNA mismatch repair

Microsatellite instability (MSI) is the result of inactivation of the DNA mismatch repair (MMR) system and is characterized by a high frequency of frameshift mutations in microsatellite DNA. In a portion of tumors, MSI is caused by germline mutations in one of the MMR genes (, , , or ), which results in hereditary Lynch syndrome. However, the majority (80%) of MSI cases are sporadic, often because of hypermethylation of the gene promoter.8, 9

MSI-high (MSI-H) has been found in as many as 24 primary cancer types, most of which are displayed in Table 3,10, 11 and appears to be a generalized cancer phenotype in about 4% of all adult cancers. Tumor MSI-H status is prognostic (patients with early-stage cancers that are MSI-H have a better prognosis than those with microsatellite stable tumors) as well as predictive—many MSI-H tumors are exquisitely sensitive to PD-1/PD-L1 inhibitors.12, 13

Table 3.

Frequency of MSI-H Status Across Cancer Types

% MSI-H (NO./TOTAL NO.)

CANCER TYPE

VANDERWALDE 201810

BONNEVILLE 201711

All cancer types

3.0 (342/11,348)

3.8 (425/11,139)

NSCLC (adenocarcinoma/squamous cell carcinomab )

0.6 (12/1868)

0.5-0.6 (6/1065b )

Colorectal adenocarcinoma

5.7 (80/1395)

–

Colon adenocarcinoma

–

19.7 (85/431)

Rectal adenocarcinoma

–

5.73 (9/157)

Pancreatic adenocarcinoma

1.2 (6/518)

0.0 (0/183)

Esophageal and esophagogastric junction carcinoma

0.0 (0/189)

1.6 (3/184)

Gastric adenocarcinoma

8.7 (16/184)

19.1 (84/440)

Liver hepatocellular carcinoma

2.7 (2/73)

0.8 (3/375)

Gastrointestinal stromal tumors (GIST)

0.0 (0/52)

–

Ovarian surface epithelial carcinoma (serous cystadenocarcinomac )

1.1 (17/1517)

1.37 (6/437c )

Nonepithelial ovarian cancer

1.8 (1/56)

–

Endometrial carcinoma

17.6 (155/879)

31.4 (170/542)

Cervical cancer (squamous cell carcinoma/endocervical adenocarcinomad )

3.6 (6/168)

2.6 (8/305d )

Breast carcinoma

0.6 (6/1024)

1.5 (16/1044)

Prostatic adenocarcinoma

2.1 (4/191)

0.6 (3/498)

Bladder cancer

0.0 (0/143)

0.5 (2/412)

Glioblastoma (multiforme)

0.7 (3/427)

0.3 (1/396)

(Skin cutaneous) melanoma

0.0 (0/345)

0.6 (3/470)

Head and neck squamous carcinoma

0.0 (0/111)

0.8 (4/510)

Sarcoma

–

0.78 (2/255)

At present, the FDA has granted approval for practitioners to administer the PD-1 inhibitor pembrolizumab for the treatment of patients with unresectable or metastatic, MSI-H or MMR-deficient (dMMR) solid tumors (site-agnostic). Currently, the approval is for patients with tumors that have progressed after prior treatment who have no satisfactory alternative treatment options, as well as for patients with MSI-H or dMMR colorectal cancer (CRC) after progression on a fluoropyrimidine, oxaliplatin, and irinotecan, and in the first line for non–small cell lung cancer (NSCLC).14, 15 In 2017, the FDA granted accelerated approval of single-agent nivolumab, another PD-1 inhibitor, for the treatment of adult and pediatric patients older than 12 years with MSI-H or dMMR CRC. Subsequently, in 2018, the FDA granted accelerated approval to a combination of nivolumab plus ipilimumab (a CTLA-4 inhibitor) for treatment of the same set of patients.16, 17 See Table 1 for MSI/MMR biomarker testing recommendations.

Neurotrophic receptor tyrosine kinase

Members of the neurotrophic receptor tyrosine kinase (NTRK) fusion oncogene family, , are most prevalent in rare adult cancer types and in several pediatric cancers, although they can occur in a very small proportion (approximately 1%) of commonly occurring cancer types in adults, including NSCLCs, CRCs, head and neck cancers, thyroid cancers, bladder cancers, gliomas, and malignant melanomas (Table 418). , , and fusions and the proteins they encode (neurotrophin receptor kinase A [TRKA], TRKB, and TRKC, respectively) are observed at an increased frequency in highly aggressive cancers such as glioblastoma multiforme, and recognition of their potential oncogenic activity led to the use of this fusion family as a predictive biomarker as well as a drug target.25

Table 4.

NTRK Frequencies in Selected Cancers

NTRK GENE

TUMOR TYPE

FUSION PARTNERSa

FREQUENCY (NO./TOTAL NO.)b
, n = 7

Gliomas

, ,

0.3% (3/982)

Colorectal carcinoma

0.2% (2/1272)

Cervical carcinoma

1.5% (1/68)

Lung adenocarcinoma

0.0% (1/4073)

, n = 10

Gliomas

, , , , , (n = 2), (n = 2),

0.9% (9/982)

Lung adenocarcinoma

0.0% (1/4073)

, n = 8

Gliomas

,

0.2% (2/982)

Lung adenocarcinoma

0.0% (2/4073)

Secretory carcinoma (breast)

0.1% (1/769)

Uterine sarcoma

0.2% (1/478)

Cancer of unknown primary

0.4% (2/227)

Larotrectinib is an oral and highly selective TRK inhibitor that was granted accelerated approval by the FDA on November 26, 2018, for the treatment of adult and pediatric patients with metastatic or unresectable solid tumors that have an fusion without a known acquired resistance mutation ( kinase domain mutations, including solvent front mutations). Patients must have a cancer that has progressed after treatment and/or have no satisfactory alternative treatment for their disease.26 The approval of larotrectinib is the second tissue-agnostic FDA approval, after pembrolizumab, for the treatment of cancer.

Another TRK inhibitor named entrectinib (RXDX-101) was granted a breakthrough therapy designation by the FDA in 2017, although it has not yet been approved for use as a treatment for adult and pediatric patients who have -positive, locally advanced or metastatic solid tumors that have either progressed after prior therapies or have no acceptable standard therapy options.27, 28

fusion testing has evolved massively over the last year or 2, and new discoveries are constantly being made using a range of different assays. The fusions displayed in Table 418 are taken from a study that was originally published in 2018. Although comprehensive at the time, this table does not contain the complete list of fusions known today, in 2019. IHC has been used as an initial screening tool to inform highly sensitive but less available and more expensive molecular testing methodologies.29–31 However, it is now clear that IHC does not have sufficient sensitivity to detect all existing fusion-encoded proteins, meaning that tumor samples should certainly be assayed using FISH or NGS from the get-go.18 In conclusion, clinicians need to be aware of all 3 TRK targets and arrange adequate testing for all of them.

Germline alterations and their testing

Gene mutations can be somatic or germline; the former spontaneously occur after birth, and the latter are inherited (ie, present at birth). Tumor genetic (somatic) testing detects mutations that may actually be germline alterations, but germline alterations require confirmation in matched normal samples (eg, DNA extracted from white blood cells, buccal swabs, or cultured skin fibroblasts) from the tumor-bearing host. Suspected germline mutations and genetic testing are relevant to cancer treatment and prevention. There is potential for patients to develop tumors at other sites or for family members to develop cancer, particularly early-onset malignancies.

Table 532 lists the somatic mutations that may be germline. This table indicates the cancer types for which germline testing should be carried out if the specified somatic mutations are found in a patient’s tumor profile.

Table 5.

Best Known Somatic Mutations That Could Also Be Germline Mutations

GERMLINE OR SOMATIC MUTATION

RARE GERMLINE-ASSOCIATED SYNDROME

MAIN CANCER APPLICABILITY

Li-Fraumeni

Sarcomas, and cancers of the breast and brain

, , , ,

Lynch

Cancers of the GI tract (particularly colorectal), endometrium, ovary, brain, breast, and renal pelvis

,

Hereditary breast, ovarian, prostate, and pancreatic cancers

Cancers of the breast, ovary, prostate, and pancreas

Cowden

Cancers of the breast, endometrium, and thyroid gland

,

Familial adenomatous polyposis

Cancers of the colon and rectum, small intestine, stomach, brain, bone, and skin

Hereditary diffuse gastric cancer

Cancers of the stomach and breast

,

Familial atypical multiple mole melanoma

Melanoma, pancreatic adenocarcinoma, and cerebral astrocytoma

Werner

Pancreatic endocrine cancer and pituitary gland tumors

Retinoblastoma

Eye cancer, pinealoma, osteosarcoma, melanomas, and soft-tissue sarcomas

Multiple endocrine neoplasia type 2

Medullary thyroid cancer, and pheochromocytoma

Von Hippel-Lindau

Kidney cancers and multiple noncancerous tumors

Peutz-Jeghers

Cancers of the breast, colon and rectum, pancreas, and stomach and hamartomas

, ,

Familial paraganglioma

Paragangliomas and pheochromocytomas

Birt-Hoge-Dube

Chromophobe renal cell cancers

,

Tuberous sclerosis

Angiofibromas, angiomyolipomas, giant cell astrocytomas

Neurofibromatosis type 1

Optic gliomas and neurofibromas

Neurofibromatosis type 2

Schwannomas, meningiomas, gliomas, neurofibromas

Gorlin

Childhood primitive neuroectodermal tumors, skin basal cell carcinomas

,

Juvenile polyposis

Multiple noncancerous growths in the colon

There are 3 main categories of tumor genetic modifications with wide variation in the expectation that these reflect germline changes. The first comprises common tumor mutations associated with rare germline alterations. For example, mutations in are found in greater than 60% of lung cancers.33 Although mutations can be inherited in the Li-Fraumeni syndrome, such familial syndromes are rare. It is believed for the most part that there is little need for germline testing unless the personal or family history is suggestive of such a syndrome. The second category comprises moderately common somatic mutations that may be associated with familial syndromes. For example, in colon cancer, dMMR is found by routine MSI or IHC testing in about 12% of tumors.34 Molecular germline testing demonstrates that about one-quarter of these dMMR alterations are inherited. Hence tumor testing should lead to germline confirmation in patients and possibly further evaluation of family members. The final category comprises uncommon tumor mutations that often reflect germline mutations. As an example, patients with breast and ovarian cancers regularly have germline testing done for and , especially if the personal or family history is suggestive. With routine molecular genetic tumor testing, / mutations are being found in patients with other tumors where it is less expected. An analysis of 100 patients with pancreatic cancer found that 7 had mutations in , 4 of which were in the germline.35 Finding mutations in the tumor may aid in choosing therapy but requires germline testing for confirmation and consideration of genetic counseling for the family.

It has generally been considered that germline testing is not always needed if somatic tumor testing has been done. However, it must be kept in mind that molecular genetic tumor testing can miss a small percentage of inherited cases, where mutations are outside the hotspots covered in the somatic panel or large-scale deletions and duplications have occurred. Conversely, larger gene panel profiling may actually identify previously unknown, clinically relevant alterations that are germline, either de novo or inherited from parents, despite a lack of associated clinical history.36

In conclusion, taking into consideration the increasing availability of germline testing and whole exome sequencing to identify inheritable mutations, as well as the personal and family history of cancer and the potential need for genetic counseling, medical teams can help provide better treatment selection for patients with some types of cancer and help to create a systematic approach to hereditary risk.

Disease-Specific Biomarkers

In the subsections below and in Tables 2.1 through 2.12, we address the currently accepted genes or gene products that act as predictive biomarkers (and risk assessment markers in some cases) for each specific solid tumor. Details on when in the disease course the presence or levels of these markers should be assessed are also included. Under each of the following subsections, we also include some description of pertinent biomarkers in research. Compelling evidence suggests that these biomarkers will be listed in the NCCN “recommended” biomarker category in the foreseeable future.

Lung cancers

Lung cancer therapy continues to follow the genomic testing paradigm (see Table 2.115, 19–24). All patients with NSCLC should be tested for , , , , and PD-L1 at baseline before treatment. Patients with uncommon mutations of may also be treated with tyrosine kinase inhibitor therapy. Other recommended markers of interest include insertion 20 mutations, rearrangements, and exon 14 mutations. All of these targets are still being actively investigated in clinical studies and hold potential for patient treatment.

Gastrointestinal cancers

Colon and rectal cancers

Oncologists now recommend the assessment of several predictive markers in patients with CRCs (see Table 2.2). The ideal time to perform genomic testing for treatment purposes is a matter of some controversy and varies depending on disease stage. At the time of initial diagnosis of a stage I, II, or III tumor, it is reasonable to perform MSI testing. Patients with MSI-H, locally confined tumors have a better prognosis, and recommendations are for patients with MSI-H stage II tumors to forgo adjuvant therapy.37, 38 Additional evidence suggests that 5-fluorouracil (5-FU) and related agents, such as capecitabine, can actually worsen outcomes when delivered as single agents to patients with early-stage MSI-H CRCs.39, 40 Treatment with an oxaliplatin regimen is the standard of care recommended for MSI-H stage III CRCs. Finally, guidelines now recommend universal MSI testing in all stages of CRC to determine whether patients have a germline mutation indicative of Lynch syndrome.41 If both the tumor DNA and the patient’s germline DNA harbor an MMR defect, this indicates that the patient has Lynch syndrome. Oncologists need to refer these patients for genetic counseling and a discussion about potential testing of relatives. Such individuals should have screening for Lynch syndrome–associated cancers at an earlier age, and more intensive screening is called for than is recommended for individuals without such a cancer susceptibility mutation. There is additional evidence that the use of aspirin can reduce premalignant polyp formation in patients and their relatives with MSI-H tumors.43 Aspirin has also been associated with improved outcomes in patients with tumors that harbor mutations, suggesting a potential value for assessment of mutations in that gene.44 MSI testing is also an eligibility requirement for the current US intergroup trial of combined 5-FU, leucovorin, and oxaliplatin (FOLFOX) with or without atezolizumab, a PD-1 inhibitor, in patients with MSI-H stage III colon cancer.

In patients with advanced CRC, MSI testing is also indicated at diagnosis. Mutations in or overexpression of additional genes that are predictive of outcomes include and . Often, other than for mutations, the optimal time for this testing is when tumors become refractory to standard chemotherapy so that the assessment reflects the current status of the disease. Patients with MSI-H tumors are now eligible for therapy with PD-1–targeting, PD-L1–targeting, and/or CTLA-4–targeting immunotherapies after their disease becomes refractory to standard chemotherapy. Those with fusions are candidates for treatment with larotrectinib.26 Individuals whose tumors harbor a mutation are insensitive to treatment with and should not receive an anti-EGFR–targeted monoclonal antibody such as cetuximab or panitumumab.45 It is likely that additional genomic analyses that are currently underway or to be evaluated in future studies, involving whole genome or whole exome sequencing in cohorts of patients with known outcomes, will identify other mutations that have either prognostic or predictive utility.

as a CRC prognostic factor

mutational status is used as a strong predictor for overall survival (OS) at all stages of disease; patients with -mutated CRC have a generally poor prognosis.46–52 V600E is the best known mutation assessed using NGS.53 Compared with patients who have CRC with wild-type tumors, patients whose tumors manifest a mutation are generally older and more likely to be female. Such patients commonly have higher grade cancers at diagnosis, with a primary tumor that is more likely to be right-sided and to have a higher number of cancer-involved lymph nodes. These -mutated tumors are also more likely to be MSI-H.54

Gastric, esophageal, and gastroesophageal junction cancers

See Table 2.3.42

Pancreatic cancers

See Table 2.4.

Genitourinary cancers

Bladder cancers

In The Cancer Genome Atlas extended 2017 study carried out by Robertson et al, findings from the complete cohort of 412 muscle-invasive bladder cancer cases revealed that mutations in the DNA repair genes (n = 57; 14%) and (n = 40; 10%), and deletions in (n = 10; 2%) were significant.55

It was found that all nonsilent somatic mutations were missense, and many could be mapped within the conserved helicase domain. Dominant negative effects on ERCC2 function were observed.56 Thus, bladder cancer missense mutations in were associated with improved response to cisplatin-based chemotherapy. However, mutations are distributed across the gene, and the functional impact of most individual mutations is unknown. Recently, Li et al reported developing a microscopy-based assay that measures the nucleotide excision repair function of clinically observed mutations. Most helicase domain mutations impaired the function. In addition, a preclinical -deficient bladder cancer model showed that loss was sufficient to drive cisplatin sensitivity. Thus, was concluded to be a predictive biomarker in bladder cancer. Moreover, this study underscores the importance of combining genomic and functional approaches in a co-clinical trial to guide precision oncology for conventional chemotherapy agents. Current evidence presented here supports the idea that and are potentially useful markers in muscle-invasive bladder cancer.55–57

Prostate cancers

It was recently reported that patients with metastatic castration-resistant prostate cancer (mCRPC) harboring germline mutations in and have superior clinical outcomes after first-line treatment with abiraterone and enzalutamide (see Table 2.5).58 The authors suggested that this improved response is likely driven by mutations in , , and . Because these conclusions were based on only 9 patients harboring germline mutations and the study was not entirely prospective, these findings require prospective validation in larger patient cohorts. A separate, small, retrospective study found that all responders to poly(adenosine diphosphate [ADP]-ribose) polymerase (PARP) inhibitor therapy harbored mutations, whereas nonresponders did not.59 However, it was agreed that the functional relevance of mutations in DNA repair genes other than should be considered before committing to PARP inhibitor therapy.

At the American Society of Clinical Oncology 2018 meeting, De Bono et al60 reported preliminary findings from the KEYNOTE-199 phase 2 trial comparing responses to the immune checkpoint inhibitor pembrolizumab in patients who had mCRPC with or without tumor expression of PD-L1. Thus, pembrolizumab showed antitumor activity and disease control with acceptable safety in patients with docetaxel-refractory mCRPC, regardless of PD-L1 status. Of note, the response rate was numerically higher in patients with somatic or mutations (12%), indicating that these could be predictive markers of response to checkpoint inhibitors. It can be seen from Table 2.561 that testing of is NCCN recommended. Testing of is also suggested but not yet NCCN recommended.

Gynecologic cancers

Endometrial cancers

As noted above (see Microsatellite instability high tumors and DNA mismatch repair), the presence or absence of MSI should be determined through universal tumor molecular testing in every patient with uterine cancer (see Table 2.6).62 Approximately 2% to 5% of uterine cancers are because of Lynch syndrome, caused by germline mutations in , , , or . Abnormalities in should prompt hypermethylation testing, as this can also cause tumors to be MSI-H in the absence of a germline mutation. The detection of a germline mutation affects subsequent screening for colon and ovarian cancer and prompts cascade testing to identify other affected family members. The presence of MSI-H because of either a germline mutation or hypermethylation provides an indication for pembrolizumab in the setting of recurrent uterine cancer, based on site-agnostic FDA approval granted in 2017.13 Women with -aberrant endometrial cancers demonstrate a favorable prognosis and may require less aggressive therapy, although this remains theoretical at present. Identification of hotspot mutations in genes such as , , , and may correlate with biological behavior but are not yet targetable. Phase 2 data demonstrate activity of mTOR inhibitors in endometrioid carcinoma of the uterus, but these trials were not assay-directed to determine whether molecular testing can select for potential activity.64

Ovarian cancers

The presence of pathogenic mutations in -related genes identify an important subset of high-grade serous epithelial ovarian cancers that have a specific biology, natural history, and susceptibility to platinum and PARP inhibitors. The spectrum of mutations in this category includes those in , , , , , , , , , , , and (see Table 2.7).65–67 Patients with these mutations have an improved prognosis with a higher likelihood of platinum sensitivity and long-term survival. Homologous recombination (HR)–deficient (HRD) tumors act similarly to tumors that have -related mutations and may serve as a surrogate for platinum sensitivity. Identification of these mutations directly affects therapy, as patients should be considered for treatment with PARP inhibitors immediately after upfront therapy with platinum and a taxane, based on the improved progression-free survival (PFS) observed in the SOLO-1 trial (Olaparib Maintenance Monotherapy in Patients With BRCA Mutated Ovarian Cancer Following First Line Platinum Based Chemotherapy; ClinicalTrials.gov identifier NCT01844986).68 This international superiority trial showed a 70% reduction in risk of ovarian cancer progression in women with germline or somatic mutations who received maintenance olaparib after primary therapy with paclitaxel and carboplatin. Conversely, patients without -related mutations may be better served by antiangiogenic therapy with bevacizumab concurrent with upfront platinum and taxane therapy followed by maintenance bevacizumab therapy (Gynecologic Oncology Group study 0218 [GOG-7]).69, 70

In the recurrent setting, PARP inhibitors (olaparib and rucaparib) as monotherapy were first approved for ovarian cancer patients with mutations or HRD. This indication has now been expanded to include olaparib, rucaparib, and niraparib as switch maintenance therapy for patients with platinum-sensitive ovarian cancer who have responded to platinum in the second-line or third-line setting.65–67, 71

The identification of -related gene mutations is also necessary to perform cascade testing on family members to identify affected family members who may be candidates for risk-reducing surgery and surveillance to prevent subsequent ovarian, tubal, peritoneal, and breast cancer.

Evaluation of PD-1 and PD-L1 status is useful in patients with ovarian cancer because pembrolizumab is approved for patients with MSI-H tumors based on a site-agnostic label. Single-agent activity for PD-1 inhibitors has been limited in patients with ovarian cancer, but checkpoint inhibitors are under study in the JAVELIN trials. The combination of PARP inhibitors with checkpoint inhibitors has been investigated, and initial response rates of 25% to 30% have been noted. The larger ATHENA trial (A Study in Ovarian Cancer Patients Evaluating Rucaparib and Nivolumab as Maintenance Treatment Following Response to Front-Line Platinum-Based Chemotherapy; ClinicalTrials.gov identifier NCT03522246) of maintenance rucaparib and nivolumab therapy is currently accruing patients with ovarian cancer who have responded to front-line, platinum-based chemotherapy.

Although initial trial results using MEK inhibitors in the treatment of patients with low-grade serous carcinomas have been disappointing, multiple studies are ongoing investigating MEK inhibitor monotherapy and combination therapy. Other rare ovarian cancers have different molecular profiles, but targeted therapies remain largely unstudied.

Cervical cancers

The treatment of patients with recurrent cervical cancer has been problematic, and their prognosis is dismal. Bevacizumab was approved for recurrent disease in combination with platinums, taxanes, and topotecan; however, no molecular markers have yet been found that can predict patient treatment response. Pembrolizumab was FDA-approved in 2018 for patients with recurrent and metastatic cervical cancer who had disease progression on or after chemotherapy and whose tumors expressed PD-L1, based on a 14% objective response rate seen in KEYNOTE 158 (Study of Pembrolizumab [MK-3475] in Participants With Advanced Solid Tumors; ClinicalTrials.gov identifier NCT02628067). Promising data also exist for single-agent nivolumab, which demonstrates a 26% response rate in the recurrent setting (ClinicalTrials.gov identifier NCT02488759). Trials evaluating combination therapy with nivolumab and ipilimumab are currently underway.

Breast cancers

The well-established biomarkers that drive treatment decisions for patients with breast cancers are estrogen receptor (ER) expression, progesterone receptor (PR) expression, and human epidermal growth factor receptor-2 (HER2) overexpression or amplification in the tumor (see Table 2.8). Determination of ER, PR, and HER2 status is recommended for all newly diagnosed invasive breast cancers and for any recurrences when feasible. These are routinely used to predict response to therapy and guide treatment planning for patients with breast cancer.

Some new markers that show promise for future use in breast cancer are the androgen receptor (AR), , and PD-L1. Overexpression of AR occurs in a subset of triple-negative breast cancers (TNBC).72 Clinical trials of AR-targeted treatments have shown promising preliminary results in patients with metastatic, AR-positive TNBC.73 Mutations in occur in the ligand-binding domain of the ER and can lead to a ligand-independent, constitutively active form of the ER. This is a potential mechanism of resistance to aromatase inhibitors. De novo mutations have been most commonly detected during or after treatment with aromatase inhibitors for hormone receptor-positive breast cancer.74 The treatment implication is to consider using selective ER downregulators that target ER directly in the setting of an mutation. The role of PD-L1 as a predictive biomarker for the treatment of patients with breast cancer using checkpoint inhibitors will be further delineated with several maturing trials evaluating immune checkpoint blockade in the treatment of breast cancer. In addition, multiparameter genomic assays, such as Oncotype DX (Table 2.8), MammaPrint, and Prosigna (formerly called PAM 50), are being used routinely for decision making in early-stage breast cancer. MammaPrint and Prosigna are prognostic for recurrence of tumors that are lymph node negative, have 1 to 3 positive lymph nodes, or are ER-positive but HER2-negative. Additional multigene assays used for consideration of adjuvant therapy in patients with breast cancer are EndoPredict and the Breast Cancer Index.

Central nervous system cancers

Although broad panels are often appropriate and especially meaningful in the metastatic setting when conventional therapy has failed, more limited panels may be a consideration (see Table 2.975, 76). This can be exemplified by central nervous system tumors, in which genetic alterations are not just prognostic or predictive, but diagnostic. Before 2016, the World Health Organization (WHO) classification relied strictly on histologic features to differentiate tumors of astrocytic and oligodendroglial lineage.77 Although patients have significantly different treatment paradigms and survival depending on which of these tumor lineages they harbor,78 occasionally features from both lineages can be found within the same tumor, resulting in a diagnosis of a “hybrid” oligoastrocytoma. This is further compounded by high interobserver discordance; thus, some institutions diagnose this entity more frequently than others.79 By combining both genotype and classical histologic findings, it is now possible to diagnose nearly all of these tumors to be compatible with either oligodendroglioma or astrocytoma. This has resulted in modifications to the WHO classification in 2016 to include both histologic phenotype and molecular genotype with consideration of mutation, 1p19q codeletion, loss, and mutation when diagnosing gliomas.75 Furthermore, epigenetic silencing of the promoter of the methyl-guanine methyl transferase gene by gene promoter methylation is frequently tested because it is highly prognostic and also predictive, correlating with a response to or benefit of alkylating agent chemotherapy.76

It has been found that most glioblastomas have potential actionable genomic alterations.80, 81 A recent NGS analysis using a 315-gene panel found that, of 43 patients, 95% had at least 1 therapeutically actionable genomic alteration of a median of 4.5 genomic alterations per patient. The most common genomic alteration detected was in (40%). Genotype-directed treatments were prescribed in 13 patients, representing a 30% treatment decision impact. Treatment with targeted agents—including everolimus as a single agent and in combination with erlotinib, afatinib, palbociclib, trametinib, and BGJ398—elicited no response.82

A fusion between Brevican () and is a potent oncogenic driver of high-grade gliomas and confers sensitivity to entrectinib.83 A case report of a – fusion in glioneuronal tumors highlights its clinical importance as a novel, targetable alteration,84 and an open-label, multicenter, global phase 2 basket study of entrectinib for the treatment of patients with locally advanced or metastatic solid tumors that harbor , , or rearrangements (ClinicalTrials.gov identifier NCT02568267) is currently recruiting glioma patients.

For pediatric low-grade gliomas, V600E is a potentially highly targetable tumor mutation, which was detected in 17% of patients who exhibited poor outcome on receipt of chemotherapy treatment.85 In a recent evaluation of dabrafenib in a phase 1/2 trial that included 32 children with relapsed or refractory, low-grade gliomas, findings of an objective response rate of 38% and stable disease in another 44% of patients are extremely exciting. It is encouraging that these drugs could be effective agents that allow us to replace chemotherapy entirely for pediatric glioma.86

Other central nervous system types for which molecular profiling has a role include ependymoma ( fusion), diffuse midline cerebellar gliomas (histone 3 mutations), medulloblastoma ( vs activated), and ependymoma ( amplification). Although many of these tumors inevitably recur and a broader panel may be useful at some point in the course of the disease to define clinical trial options, obtaining a limited panel that contains the molecular alterations considered within the WHO criteria remains a reasonable option.

We certainly see the potential implication of molecular profiling for a routine part of therapeutic decision making beyond classification and prognostic prediction for patients with glioma. Of other mutations tested, the epidermal growth factor gene variant vIII encodes a promising molecular target. amplification could be useful in the treatment of glioblastomas. However, agents targeting EGFR signaling pathways have displayed limited or no therapeutic efficacy in glioblastoma clinical trials. ABT-414 (an investigational, anti-EGFR monoclonal antibody drug conjugate) alone87 or in combination with temozolomide showed a trend toward improved survival and was safely administered with radiation therapy.88, 89

The V600E mutation, which is analyzed using NGS, is predictive and prognostic for low-grade pediatric glioma. This mutation is frequently found in gangliogliomas and in about two-thirds of grade II xanthroastrocytomas. It is assumed that this alteration constitutively activates the RAS/RAK/MEK/ERK kinase pathway. When BRAF kinase inhibitor treatment effects are validated within low-grade glioma, the drug could transform the V600E mutation from a diagnostic marker to predictive marker of response to therapy.90

Sarcomas

Sarcomas are heterogeneous cancers comprising over 50 diverse histological subtypes (see Table 2.10). As a group, they have a low occurrence incidence and are considered rare cancers. Although there is some crossover, pediatric and adult sarcomas have distinctly different histologies as well as different genetic drivers. The majority of genomic variations (translocations, CNVs, complex karyotypes, etc) provide important predictive diagnostic information rather than potential therapeutic targets. Thus, (for Ewing sarcoma), (alveolar rhabdomyosarcoma), (monophasic synovial sarcoma), (biphasic synovial sarcoma), and (myxoid liposarcoma) fusions are diagnostic markers that should be tested at initial workup using RNA sequencing techniques, particularly FISH. There are several prominent exceptions to this diagnosis-only rule in gastrointestinal (GI) stromal tumors, in which mutations in (particularly exon 11) and are notable biomarkers for therapeutic intervention with the tyrosine kinase inhibitors imatinib and sunitinib.

Head and neck cancers

PD-1 is highly expressed in head and neck squamous cell carcinomas (HNSCCs) and, in 2016, the PD-1 inhibitors nivolumab91 and pembrolizumab92 were both approved for the treatment of HNSCC that has metastasized or recurred on or after treatment with platinum chemotherapy (see Table 2.11). PD-L1 testing is now recommended in patients undergoing workup for metastatic HNSCC, with the intention of offering pembrolizumab as a treatment option to those with PD-L1–positive tumors.

is reportedly overexpressed in between 90% and 100% of HNSCCs.93 Accordingly, cetuximab is an approved targeted therapy for this disease and is usually administered regardless of mutation testing. HNSCCs can develop resistance to cetuximab. Activation of other EGFR family members (HER2, HER3) can play a role in this resistance, as can c-MET, insulin growth factor receptor (IGFR), and PI3K.94 is frequently mutated in HNSCC and plays a key role in the progression of HNSCC.95, 96 Targeted agents against all these markers have been developed and have undergone or are undergoing various phases of clinical testing. Human papillomavirus (HPV)-related HNSCCs are increasing in incidence and have different oncogenic processes compared with HPV-unrelated HNSCCs. Patients with HPV-positive HNSCCs respond better to treatment and have a better prognosis than their HPV-negative counterparts. Therefore, for the sake of disease diagnosis, treatment, and management, it is useful to accurately discriminate between the HPV-positive and HPV-negative HNSCCs, which can be done through tumor P16 testing by IHC; thus, P16 positivity corresponds to HPV positivity.97

Melanomas

To date, the only FDA-approved predictive biomarker in patients with advanced melanoma is genotyping (see Table 2.12). Approximately one-half of melanomas that originate from cutaneous primary sites will harbor a V600 mutation.98 This leads to constitutive activation of the MAPK pathway and increased cell proliferation, metastasis, and survival mechanisms. Vemurafenib, dabrafenib, and encorafenib are BRAF-targeted therapies that preferentially inhibit cells harboring the V600 mutation. It is important to be aware that selective inhibitors of BRAF encoded by mutant V600 can cause paradoxical activation of the MAPK pathway in cells that are V600 wild-type (particularly if they harbor a mutation). This effect occurs through RAF dimerization, leading to increased cell proliferation rather than inhibition.99 The combination of selective BRAF inhibitors with MEK1/MEK2 inhibitors is now FDA approved only for patients with V600-mutant melanoma. In patients with resected, stage III, V600E/V600K–mutant melanoma, dabrafenib plus trametinib improves relapse-free survival by 53%.100 Similarly, dabrafenib plus trametinib and other BRAF/MEK inhibitor combinations have demonstrated objective response rates of up to 68% in patients with unresectable advanced V600E/K mutant melanoma.101

Other oncogenic driver mutations have been identified in melanomas for which targeted therapies have demonstrated clinical activity. mutations (and amplifications) have been identified in up to 20% of patients with advanced melanoma, particular those with chronic sun-damaged, acral, or mucosal melanoma subtypes.102, 103 Of note, mutations are often seen across multiple exons, and hotspot mutations are not typically observed. This patient population has a reported response rate to imatinib of 21% to 29%.104–106 Higher response rates were seen in individuals whose melanoma harbored exon 11 and 13 mutations. Another important oncogene, , is mutated in approximately 20% of melanomas—most commonly at the Q61 position.107 Direct targeting of NRAS has proven difficult, but clinical activity has been demonstrated by targeting the downstream MAPK pathway with MEK1/MEK2 inhibitors. The MEK inhibitor binimetinib showed superior clinical outcomes compared with dacarbazine.108 However, the objective response rate of binimetinib was only 15%, and this agent has not yet been approved by the FDA for this indication.

With regard to predictive biomarkers for immune checkpoint therapies in melanoma, several have shown enrichment for greater clinical activity, mostly in post hoc or retrospective analyses, but have not been approved by the FDA for routine clinical use.109 These include positive PD-L1 IHC, immune gene expression profiles, and high tumor mutational burden (TMB) (see The role of TMB—an emerging biomarker, below) by targeted exome sequencing. For example, response rates to anti-PD-1/PD-L1 therapy were 81%, 36%, and 10% for patients whose tumors had >23.1 mutations per megabase (MB), 3.3 to 23.1 mutations per MB, and <3.3 mutations per MB, respectively.110 However, patients with low or negative biomarkers can still benefit from immune checkpoint therapies, and some studies have shown marginal differences between groups. The PD-L1 IHC analyses from the Checkmate 067 study demonstrate this concept well: response rates were 43% and 58% with nivolumab monotherapy in patients whose tumors had <5% PD-L1 staining versus >5% PD-L1 staining, respectively.111 Biomarkers may be useful in the application of nivolumab/ipilimumab combination therapy over nivolumab monotherapy for patients with melanoma based on Checkmate 067 study data. Improvement in PFS with the combination approach was best seen in patients whose tumors harbored a V600 mutation or had <1% PD-L1 staining (hazard ratios, 0.62 and 0.68, respectively).

The Gray Area Between Research and Clinical Practice

Core clinical markers in current use by expert molecular profiling laboratories, the frequencies of these markers in a range of tumor lineages, and assay types used for their assessment can be found in Table 6 (Caris Molecular Intelligence). Many of the markers and their assays in Table 6 are essentially still classified as belonging in a “research” category, and the NCCN has not yet recommended universal testing for these genes. Nevertheless, they have been reported as actionable and useful by a general consensus of experts in the research community. Because the use of broader gene panels and full-scale NGS is still in the gray area between research and clinical practice, it comes burdened with benefit-to-cost ratio controversies. The field is also evolving rapidly, with fluidity existing in the classification of genes as clearly, possibly, or unlikely to be relevant to treatment considerations.

Table 6.

Caris Molecular Intelligence Core Clinical Marker Frequencies for Tumor Lineages

NSCLC

ANAL CANCER

COLORECTAL CANCER

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

MSI

NGS

10

1893

0.5

TML

NGS

6

91

33.0

NGS

1744

3427

50.9

TML

NGS

712

4557

15.6

PD-L1

IHC

63

191

33.0

NGS

146

3427

4.3

PD-L1 (22c3)

IHC

2541

5658

44.9

NGS

22

92

33.0

NGS

285

3427

8.3

ALK

IHC

10

4606

0.2

NGS

10

92

33.0

NGS

600

3427

17.5

NGS

84

4606

1.8

MSI

NGS

0

40

0.0

NGS

14

3424

0.4

CNV

84

4606

1.8

PERITONEAL CANCER

Fusion

32

1272

2.5

Fusion

84

4606

1.8

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

MLH1

IHC

5840

6132

95.2

ISH/fusion

10

4606

0.2

NGS

42

87

48.3

PMS2

IHC

5719

6097

93.8

NGS

10

4606

0.2

NGS

24

87

27.6

MSH2

IHC

6011

6121

98.2

Fusion

19

4164

0.5

NGS

10

87

11.5

MSH6

IHC

5947

6104

97.4

Fusion

14

4186

0.3

NGS

2

87

2.3

MSI

NGS

83

1346

6.2

Fusion

19

4164

0.5

NGS

4

87

4.6

TML

NGS

247

3400

7.3

PTEN

IHC

7792

12,618

61.8

NGS

1

87

1.1

HER2

IHC

161

8708

1.8

NGS

165

4606

3.6

NGS

3

87

3.4

CISH

129

4460

2.9

(L858R)

NGS

174

4659

3.7

MLH1

IHC

190

192

99.0

PD-L1

IHC

196

6405

3.1

(exon 19 del)

NGS

260

4659

5.6

PMS2

IHC

189

192

98.4

PTEN

IHC

5334

11,407

46.8

T790M

NGS

62

4773

1.3

MSH2

IHC

189

192

98.4

SMALL BOWEL CANCER

tertiary

NGS

3

4659

0.1

MSH6

IHC

187

192

97.4

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

NGS

70

4606

1.5

MSI

NGS

1

27

3.7

NGS

100

183

54.6

NGS

1319

4606

28.6

TML

NGS

1

87

1.1

NGS

3

183

1.6

Fusion

3

3797

0.1

PD-L1 (SP142)

IHC

7

211

3.3

NGS

14

183

7.7

NTRK

IHC

35

2031

1.7

PTEN

IHC

252

393

64.1

NGS

26

183

14.2

NGS

219

4606

4.8

NGS

1

183

0.5

NGS

0

4606

0.0

MLH1

IHC

88

96

91.7

NGS

9

4761

0.2

MSH2

IHC

90

96

93.8

SMALL CELL LUNG CANCER

MSH6

IHC

89

96

92.7

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

PMS2

IHC

89

96

92.7

MSI

NGS

0

88

0.0

MSI

NGS

10

71

14.1

TML

NGS

13

205

6.3

TML

NGS

18

182

9.9

PD-L1

IHC

182

206

88.3

HER2

IHC

10

482

2.1

NGS

99

205

48.3

CISH

11

269

4.1

NGS

11

206

5.3

PD-L1

IHC

25

342

7.3

NGS

0

206

0.0

PTEN

IHC

264

526

50.2

NGS

0

206

0.0

Fusion

6

57

10.5

NGS

0

49

0.0

GASTROESOPHAGEAL CANCER

CHOLANGIOCARCINOMA

BLADDER CANCER

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

NGS

31

922

3.4

NGS

39

474

8.2

MSI

NGS

4

152

2.6

NGS

5

921

0.5

NGS

14

474

3.0

TML

NGS

58

361

16.1

NGS

50

921

5.4

NGS

35

474

7.4

NGS

21

364

5.8

NGS

6

922

0.7

NGS

23

474

4.9

NGS

12

364

3.3

NGS

3

921

0.3

NGS

82

474

17.3

NGS

17

364

4.7

NGS

15

921

1.6

NGS

16

474

3.4

NGS

0

364

0.0

NGS

2

921

0.2

NGS

13

474

2.7

NGS

0

364

0.0

NGS

44

921

4.8

NGS

30

474

6.3

NGS

43

364

11.8

NGS

7

921

0.8

MLH1

IHC

70

73

95.9

NGS

55

364

15.1

NGS

686

922

74.4

PMS2

IHC

69

73

94.5

NGS

23

364

6.3

NGS

30

922

3.3

MSH2

IHC

72

73

98.6

PD-L1

IHC

162

788

20.6

NGS

58

922

6.3

MSH6

IHC

72

73

98.6

PROSTATE CANCER

NGS

17

921

1.8

MSI

NGS

1

177

0.6

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

PD-L1

IHC

180

1804

10.0

TML

NGS

15

469

3.2

MSI

NGS

6

151

4.0

MSI

NGS

15

389

3.9

HER2/

IHC

40

1301

3.1

TML

NGS

15

416

3.6

TML

NGS

47

916

5.1

CISH

38

618

6.1

NGS

22

422

5.2

HER2

IHC

75

1431

5.2

PD-L1

IHC

78

934

8.4

NGS

6

365

1.6

CISH

168

1323

12.7

Fusion

2

131

0.0

NGS

26

365

7.1

NGS

34

462

7.4

Fusion

1

149

0.7

NGS

35

440

8.0

NGS

78

922

8.5

Fusion

7

149

4.7

PD-L1

IHC

25

811

3.1

PANCREATIC CANCER

HEPATOCELLULAR CARCINOMA

AR

IHC

1127

1213

92.9

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

Fusion

57

198

28.8

MSI

NGS

6

456

1.3

NGS

53

166

31.9

TML

NGS

18

1149

1.6

NGS

56

166

33.7

NGS

942

1164

80.9

NGS

8

166

4.8

NGS

2

1164

0.2

MSI

NGS

1

69

1.4

NGS

18

1164

1.5

TML

NGS

4

166

2.4

NGS

229

1164

19.7

PD-L1

IHC

26

366

7.1

HER2

IHC

38

2096

1.8

ISH

26

4063

0.6

PD-L1

IHC

222

2682

8.3

NGS

42

1164

3.6

NGS

13

1031

1.3

NGS

36

1031

3.5

RENAL CANCER

ENDOMETRIAL CANCER

BREAST CANCER (CONTINUED)

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

MSI

NGS

1

107

0.9

TML

NGS

282

1469

19.2

NGS

59

1960

3.0

TML

NGS

3

327

0.9

MSI

NGS

163

796

20.5

NGS

88

1960

4.5

NGS

33

333

9.9

MSI

Frag analysis

267

1831

14.6

NGS

160

2164

7.4

NGS

2

332

0.6

NGS

42

1841

2.3

NGS

32

2169

1.5

NGS

1

332

0.3

NGS

1137

1843

61.7

NGS

159

2164

7.3

NGS

6

333

1.8

NGS

823

1843

44.7

NGS

3

2164

0.1

NGS

60

332

18.1

NGS

745

1841

40.5

NGS

4

2164

0.2

NGS

0

332

0.0

NGS

21

1843

1.1

NGS

86

2169

4

NGS

46

332

13.9

CNV

61

1867

3.3

NGS

27

2164

1.2

NGS

5

332

1.5

MSH6

IHC

2273

2295

99

NGS

143

2169

6.6

NGS

2

332

0.6

MSH2

IHC

2301

2306

99.8

NGS

15

2164

0.7

NGS

153

333

45.9

PMS2*

IHC

1887

2293

82.3

NGS

1203

2169

55.5

PD-L1

IHC

117

630

18.6

MLH1*

IHC

1912

2298

83.2

GLIOBLASTOMA

OVARIAN CANCER

PD-L1

IHC

460

4502

10.2

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

CERVICAL CANCER

NGS

161

851

18.9

MSI

NGS

23

1484

1.5

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

NGS

3

851

0.4

TML

NGS

63

3434

1.8

MSI

NGS

2

116

1.7

NGS

17

851

2.0

NGS

238

2924

8.1

TML

NGS

20

305

6.6

MSI

NGS

1

265

0.4

NGS

180

2922

6.2

NGS

56

307

18.2

TML

NGS

28

847

3.3

NGS

63

3463

1.8

NGS

92

307

30.0

NGS

59

851

6.9

NGS

290

3463

8.4

PD-L1

IHC

191

802

23.8

NGS

18

851

2.1

NGS

158

3463

4.6

BREAST CANCER

NGS

8

851

0.9

NGS

253

3463

7.3

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

Methylation

828

1773

46.7

NGS

170

3463

4.9

MSI

NGS

5

779

0.6

1p19q

FISH

52

779

6.7

NGS

279

3458

8.1

TML

NGS

76

2146

3.5.0

vIII

Fusion

87

630

13.8

NGS

19

3452

0.6

PD-L1

IHC

389

5519

7

CNV

257

828

31

ER

IHC

7947

16,384

48.5

AR

IHC

5698

11,114

51.3

PD-L1

IHC

209

1346

15.5

PR

IHC

4698

16,353

28.7

ER

IHC

5598

9404

59.5

Fusion

8

404

2.0

PD-L1

IHC

748

8693

8.6

PR

IHC

3806

9362

40.7

Fusion

11

404

2.7

HER2

IHC

900

9299

9.7

Fusion

3

404

0.7

CISH

726

6087

11.9

Fusion

12

404

3.0

NGS

50

2169

2.3

Fusion

2

404

0.5

Fusion

1

404

0.2

SARCOMA

MELANOMA

UVEAL MELANOMA

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

MSI

NGS

6

350

1.7

MSI

NGS

0

293

0.0

MSI

NGS

0

22

0.0

TML

NGS

32

1025

3.1

TML

NGS

289

774

37.0

TML

NGS

3

97

3.1

NGS

63

1034

6.1

NGS

313

782

40.0

NGS

49

99

49.5

NGS

51

1036

4.9

NGS

20

801

2.0

NGS

43

99

43.4

NGS

1

601

0.2

NGS

27

782

4.0

NGS

48

99

48.5

NGS

0

1036

0.0

NGS

189

782

24.0

NGS

11

99

11.1

Fusion

3

28

10.7

NGS

197

782

25.0

PD-L1

IHC

21

139

15.1

Fusion

1

469

0.2

NGS

111

782

14.0

Fusion

3

470

0.6

NGS

47

782

6.0

Fusion

1

469

0.2

CNV

16

749

2.0

Fusion

1

28

3.6

PD-L1

IHC

452

1381

33.0

PD-L1

IHC

505

2386

21.2

HNSCC

BIOMARKERS

TECHNOLOGY

POSITIVE NO.

TOTAL NO.

PERCENT

MSI

NGS

1

164

0.6

TML

NGS

24

324

7.4

NGS

201

326

61.7

NGS

43

326

13.2

PD-1

IHC

250

340

73.5

PD-L1

IHC

255

721

35.4

Until recently, approved genetic testing involved a small group of genetic tests carried out in patients with specific cancers for a specific therapeutic purpose. Firmly established examples of mutational status being key to treatment recommendations include pan- testing (, , and ) in patients with CRC to direct the use of anti-EGFR therapies cetuximab and panitumumab45 and HER2 testing in patients with breast cancer to direct the use of anti–HER2-targeted therapies, such as trastuzumab,112 and tyrosine kinase inhibitor therapies, such as lapatinib.113 It took many years and a large number of trials involving many patients before pan- and HER2 testing became a standard treatment-predictive approach. Other examples of molecular testing used in standard clinical practice are detailed above (see Disease-Specific Biomarkers). However, as genetic testing evolved into whole genome sequencing, advances in computer technology allowed small-capacity assays to evolve into automated, high-throughput assays with large-scale data collection, classification, storage, and analysis. Thus, real-time, broad gene panel testing combined with relevant patient clinical data are now providing an unprecedented wealth of information. However, the interpretation of the meaning of results is limited by the finding that relatively small pools of evidence are available to validate most markers and their paired targeted therapies. Larger studies and collaborative efforts are certainly needed to further and more widely validate these broader panel markers and gene expression profiles and to integrate them and their targeted therapies into clinical practice (see Absence of Randomized, Controlled Clinical Trials, below). The immediate goal of testing is to translate genetic findings into potentially effective therapy decisions for today’s patients. Meanwhile, numerous proof-of-principle trials currently are in progress or in development. One key to accelerating the application of this knowledge is real-time national and international partnerships between cancer researchers and pharmaceutical companies to perform broad-panel profiling and elucidate targeted patient therapies. Concurrently, data pooling is mandatory using universal data-sharing capabilities to maximize the utility of these findings and generate large pools of evidence (see Data Sharing, below). Successes and failures alike will provide a more complete picture, and the result will take us steps closer to effective cancer treatment—and cures. This model is already in practice in the form of basket trials.

Oncology Basket Trials and Precision Medicine

Current oncology basket trials test therapies across a range of populations using biomarker-driven designs. Such trials choose biomarkers, which must have a clinically feasible assay, to attempt to enrich responses to a particular targeted therapy. The gathering of efficacy data across a range of populations translates to only one primary outcome endpoint, which simplifies the situation while increasing deductive power. These large-scale and small-scale, broad-panel molecular profiling trials include the National Cancer Institute’s Molecular Analysis for Therapy Choice (NCI-MATCH) trial, the American Society of Clinical Oncology (ASCO) Targeted Agent and Profiling Utilization Registry (TAPUR) study, and the European Organization for Research and Treatment of Cancer–Screening Patients for Efficient Clinical Trial Access (EORTC-SPECTA) program. These studies attempt to expand the boundaries of precision medicine and build evidence supporting the use of molecularly tailored therapy.

National Cancer Institute’s Molecular Analysis for Therapy Choice Trial

The novel, phase 2 NCI-MATCH (Molecular Analysis for Therapy Choice114) trial was initiated in August 2015 and is bringing public and private sectors together to enable access of physician researchers to investigational agents (in addition to approved agents) in an attempt to build the much sought-after evidence supporting the effectiveness of matching targeted therapy to patient molecular profiles. The primary aim of the NCI-MATCH study is to evaluate the proportion of patients with objective responses (ORs) to targeted therapies predicted to be mechanistically effective based on individual tumor genomic profiling. If the response rate to any mutation-matched therapy is at least 25%, this match will be tested in larger phase 2 trials. There are well over a thousand study locations across the United States, and pharmaceutical and biotechnology companies are providing targeted agents to enrolled patients across these sites. Patients are treated according to their profile (Table 7) and regardless of tissue origin or cancer type. New drugs of interest can be added to the “master” trial at any time. The trial is running under ClinicalTrials.gov identifier NCT02465060, where up-to-date information can be obtained.

Table 7.

Broadening Molecular Profiling Boundaries—Biomarker-Targeted Therapy Matches

TARGETED MUTATION

DRUG

NCI-MATCH trial: NCT02465060a
activating mutation

Afatinib

activating mutation

Afatinib

or mutations

Adavosertib (AZD1775)

FGFR pathway aberrations

AZD4547

, , mutation

Binimetinib

mutation

Capivasertib (AZD 5363)

mutation

Copanlisib

mutation

Copanlisib

loss

Copanlisib

amplification

Crizotinib

exon 14 deletion

Crizotinib

translocation

Crizotinib

translocation or inversion

Crizotinib

V600E/V600R/V600K/V600D mutation

Dabrafenib + trametinib

S768R, I638F, or L239R mutation

Dasatinib

inactivating mutation

Defactinib

mutation or deletion and PTEN expression

GSK2636771 (PI3Kβ inhibitor)

loss

GSK2636771 (PI3Kβ inhibitor)

mutation or fusion

Erdafitinib

amplification

Erdafitinib

, , gene fusions

Larotrectinib (LOXO-101)

Loss of or (by IHC)

Nivolumab

T790M or rare activating mutation

Osimertinib

, , amplification & Rb expression

Palbociclib

or amplification and Rb protein

Palbociclib

amplification ≥7 copy numbers

Pertuzumab + trastuzumab

or mutation

Sapanisertib

mutation

Sapanisertib

exon 9, 11, 13, or 14 mutation

Sunitinib

mutation

Taselisib

/ mutation

Trametinib

fusion or non-V600 mutation

Trametinib

mutation

Trametinib

amplification

Trastuzumab emtansine

/ mutation

Vismodegib

TAPUR trial: NCT02693535b
mutation, amplification or overexpression

Axitinib

, , , mutations

Bosutinib

, , mutations

Crizotinib

, , and (all wild type)

Cetuximab

, , , , , , , mutations

Dasatinib

inactivating mutations; mutations/deletions

Olaparib

MSI-high, high TML, and others

Nivolumab and ipilimumab

, , amplifications

Palbociclib

/ mutations; high TML

Pembrolizumab

1, 2, 3, , , , , mutations/amplifications

Regorafenib

, ,

Sunitinib

, mutations

Temsirolimus

amplifications

Trastuzumab and pertuzumab

V600E mutations

Vemurafenib and cobimetinib

Targeted Agent and Profiling Utilization Registry

The TAPUR study is an ongoing, nonrandomized, multicenter clinical trial that opened in 2016.115 This trial is testing the use of drugs already approved by the FDA that target a specific tumor mutation in individuals with advanced cancer outside of the drug’s approved indication. Patients with a range of solid tumors as well as lymphomas and multiple myelomas are eligible for enrollment. As with NCI-MATCH, treatment assignment in this study is based on an existing tumor mutation and not the organ from which the cancer originated. The study aim is to observe the real-world use of targeted therapies in any patient whose tumor tests positive for a selected genomic alteration that is known to be a drug target or has been shown to predict sensitivity to a drug available in this study. The primary outcome measure is the objective response rate (defined as the percentage of participants in a cohort with a complete or partial response at 8 weeks postbaseline or with stable disease at 16 weeks or later postbaseline according to RECIST (Response Evaluation Criteria in Solid Tumors) (for solid tumors), international uniform response criteria (for multiple myeloma),116 and Lugano criteria (for non–Hodgkin lymphoma).117, 118 Table 7 details the genomic alterations (biomarkers) and targeted therapies of interest at the time of submission of this article for publication, although markers and therapies are continually being refined as the study progresses. It is currently anticipated that TAPUR will enroll over 2500 patients in total. The trial is running under ClinicalTrials.gov identifier NCT02693535, where up-to-date information can be obtained.

European Organization for Research and Treatment of Cancer–Screening Patients for Efficient Clinical Trial Access

EORTC-SPECTA is a collaborative European molecular screening program that coordinates several disease-specific platforms with the aim of identifying actionable mutations and offering specific targeted therapy to patients (ClinicalTrials.gov identifier NCT02834884).119, 120 This is a large-scale basket trial that operates through one entry point that provides access to multiple studies and to high-quality, annotated material for research purposes and provides longitudinal follow-up of patients to understand progression patterns.121

Targets of Special Interest: Emerging and in Current Practice

There are several novel biomarkers of great interest, many of which have found a niche in common practice but are continuing to reveal exciting connections and uses. We expand on several these markers below.

Immune Markers and Immunotherapy

Programmed death–ligand 1 expression

122123124125, 12612742128129, 130

• First, performing a PD-L1 IHC assay on a single tumor site at one time point does not take into account the intrapatient tumor heterogeneity that can exist and the variability in PD-L1 expression that can occur over time.131132
• A second issue is the range of antibody assays that have been developed124133, 134
• A final issue is related to PD-L1 IHC scoring: how does one define a PD-L1–positive from a PD-L1–negative tumor? PD-L1 staining of immune cells in the tumor microenvironment, such as macrophages, gives a signal that is erroneously included in tumor PD-L1 assessment. This is the case for 22C3 and SP142 assays. To date, no approaches or thresholds reach sufficient sensitivity or specificity to be predictive of a high likelihood of response to a given drug. Providers need to be familiar with the individual PD-L1 assays and scoring used for each agent and tumor type when making patient decisions based on PD-L1 results.

The immune checkpoint PD-1/PD-L1 axis is a well-described inhibitory pathway that leads to T-cell exhaustion in the tumor microenvironment.Typically, PD-1 on tumor-infiltrating cytotoxic T cells interacts with PD-L1 on tumor cells, causing dampening of antitumor immunity (an adaptive immune response).Tumor types known to be immunogenic typically have relatively high rates of PD-L1 positivity.However, although greater clinical activity of anti–PD-1 agents (nivolumab and pembrolizumab) and PD-L1 agents (avelumab, atezolizumab, and durvalumab) has been consistently observed in patients with PD-L1–positive disease,some clinical trials have found that patients with low PD-L1–expressing tumors can derive significant benefit from anti–PD-1/PD-L1 agents. Therefore, PD-L1 IHC score alone is insufficient for patient selection in many tumor types. Assays have been developed to test for PD-L1 expression, including the PD-L1 IHC assay with 28-8 Dako (developed for nivolumab), 22C3 Dako (developed for pembrolizumab), SP142 Ventana (atezolizumab), SP263 Ventana (durvalumab), and 73-10 Dako (avelumab). These assays can be used as a tool for physicians to assess which patients might have the largest chance of benefitting from anti–PD-1/PD-L1 agents. However, because each inhibitor requires its own individual PD-L1 IHC assay, it is useful to have an upfront working knowledge of which targeted therapy is going to be used; otherwise the laboratory is required to run 5 different IHC tests, which raises costs and inefficiencies. There are several scenarios in which the FDA has mandated that PD-L1 positivity is required before anti–PD-1 agents are usable within approved indications. For example, patients with advanced, metastatic NSCLC can be treated in the front line with pembrolizumab monotherapy only if their PD-L1 tumor proportion score (TPS) is >50% (Table 2.1 ).In the second-line setting, pembrolizumab is FDA approved for adult patients with tumors (eg, gastric tumors) (Table 2.3 ) that have a lower positive TPS score (>1%).Interestingly, nivolumab has been approved as second-line therapy for select cancers regardless of their PD-L1 status.There is controversy over the use and reliability of PD-L1 IHC as a predictive biomarker. This is because of multiple factors:

Microsatellite instability and deficient MMR

Microsatellites are lengths of DNA sequence that contain single nucleotide (mononucleotide) or sections of 2 or more nucleotide (dinucleotide, trinucleotide, tetranucleotide, or pentanucleotide) repeats (see Microsatellite instability-high tumors and DNA mismatch repair). When microsatellites contain a clonal change in several repeated DNA nucleotide units, this results in MSI (tumors with such MSI are characterized as MSI-H, and this occurs when at least one of the MMR genes—, , , and —are inactivated, causing dMMR).10 Since MSI-H was established as a possible biomarker, the MSI status of a tumor has always required microdissection and PCR-based detection strategies. For practical purposes, MSI is equivalent to the loss of staining by IHC of at least one of the MMR genes because any lack of normal MMR protein expression signifies an abnormality in MMR and thus MSI. A sensitive and specific MSI assay by NGS has recently been developed that is comparable to the existing gold standard of PCR-based methods without requiring matched samples from tumor and normal tissues.10 MSI appears to be a generalized cancer phenotype in about 4% of all adult cancers in total. MSI-H tumors are associated with an improved prognosis in early-stage cancers. In Table 3,10, 11 MSI-H frequency data for several different cancer types are compared between 2 studies. In both studies, patient DNA was originally sequenced by NGS; however, the study by Bonneville et al11 obtained and retrospectively assessed sequencing data from The Cancer Genome Atlas (TCGA), the Therapeutically Applicable Research to Generate Effective Treatments (TARGET) database, and 444 other studies, whereas the study by Vanderwalde et al10 retrospectively assessed data from commercial comprehensive sequencing profiles performed on patient tumors by Caris Life Sciences. It can be readily observed from Table 310, 11 that the rate of MSI-H in tumors from different tissue types is not always consistent between studies. In particular, this can be observed for gastric cancers, endometrial cancers, breast cancers, and CRCs versus colon and rectal cancers. These study differences could be explained for the most part by sampling bias, the use of different data analysis techniques, or statistical variance. Vanderwalde et al10 certainly assessed a very sick patient population that was undergoing tumor profiling because of a bad prognosis and lack of obvious therapeutic options, whereas the patient population examined by Bonneville et al11 was not described as such and possibly consisted of patients with variable disease stages and prognoses; MSI-H patients tend to have a better prognosis than their microsatellite stable counterparts do, which would explain the lower percentage of MSI-H patients in the Caris data set compared with the TCGA data set. In addition, Vanderwalde et al10 combined colon and rectal cases in one analysis, possibly yielding a lower percentage of MSI-H than that seen by Bonneville et al11 for colon cancer alone. This highlights the potential for variability in biomarker assessment because of different assay types and technologies, not just for MSI but also across the biomarker board.

The role of TMB—an emerging biomarker

TMB is certainly an interesting marker, and evidence of its importance is growing. However, methodologies assessing TMB are not widely available at present, and most clinical laboratories do not yet offer this assessment in their assay repertoire. Immunogenicity is certainly associated with mutation load, suggesting that an increase in the number of somatic mutations present in tumor cells increases potential recognition by the immune system.135 Indeed, the presence of mutations in the tumor generates neoantigens (not expressed by normal cells), and the more mutations there are, the more the tumor is likely to be immunogenic. Furthermore, emerging evidence suggests that a high TMB is associated with increased clinical activity of immune checkpoint inhibitors.136–140 TMB was shown to predict immune therapy efficacy in patients with melanoma,138 NSCLC,140 and GI cancers.141 There is ongoing discussion regarding the definition of “high” TMB for predictive purposes. Most studies report ≥17 mutations per MB as high TMB, which is based on comparing TMB with MSI in patients with CRC. This, of course, also shows that high TMB is in strong concordance with MSI-high in CRC. However, although TMB is associated with dMMR, not all tumors with a high TMB are actually associated with MSI-H, and future studies should address this aspect.

MSI and MMR, TMB, and PD-L1

The relationship between TMB, MSI, and PD-L1 has recently been explored in a broad range of cancer types (Fig. 1).10 There is some overlap of all 3 markers in a few cancers. However, in most cancers, overlap is infrequent or does not exist at all, and 69.5% of all cancer cases were negative for all 3 biomarkers (7890 of 11,348 tested). A population of tumors exhibiting MSI-H status but low TMB and no PD-L1 expression was identified. Since MSI/MMR status alone or in combination with PD-L1 positivity became an accepted predictive marker in the FDA indication for checkpoint inhibitors, the finding that patients can test positively for only one of these markers obviously means that the number of patients now eligible to receive and hopefully benefit from checkpoint inhibition has been broadened. Until more is understood about how MSI, PD-L1, and TMB work together and how this interaction is clinically relevant, the only reasonable option is to continue to assay for all 3 markers and ensure that the number of patients who are given the chance to benefit from these drugs is maximized.

Figure 1Open in figure viewer

PowerPoint

10

Venn Diagram of the Relationships Between High Tumor Mutational Burden (TMB), High Microsatellite Instability (MSI-H), and High Programmed Death–Ligand 1 (PD-L1) for All Cancer Types.

Polybromo 1

PBRM1 is a non–MSI/PD-L1/TMB marker that could be predictive for response to checkpoint inhibitors. For example, clear cell renal cell carcinoma (ccRCC) responds to immune therapy but, unlike many other responsive human tumors, harbors a low burden of somatic mutations. Even so, ccRCC has relatively high immune cytolytic activity and a microenvironment with high immune and T-cell infiltration scores. In the past, large-scale sequencing studies demonstrated that loss of function (LOF) alterations are present in a large portion (up to 41%) of ccRCC tumors. Patients whose tumors had loss in both gene copies had significantly prolonged OS and PFS and manifested reduced tumor burden in response to immune checkpoint therapy compared with patients without loss (log-rank  = .0074 and  = .029, respectively). Miao et al summarized that, given the high prevalence of LOF in ccRCC, this genetic mutation has important implications as a molecular tool for considering immune therapy responsiveness in ccRCC and possibly across other cancer types.142

The Role of HRD as an Emerging Biomarker

Repair of DNA double-strand breaks by cells is mediated by the HR pathway or nonhomologous end-joining. HR is a complex DNA repair pathway involving multiple steps and has been reviewed extensively.143, 144 The and genes are critical for efficient double-strand DNA repair via HR and play an important role in the development and clinical progression of many cancers.145, 146 If a cell carries LOF mutations, it loses the ability to repair double-strand breaks by HR and is termed the HRD pathway. Such HRD cells are highly sensitive to DNA-damaging agents, such as platinum-based chemotherapies and other cytotoxic agents that can cause DNA strand breaks.147, 148 PARP plays a major role in DNA strand break repair. If PARP is inhibited, then cells ultimately die. Thus combining cytotoxic therapy with a PARP inhibitor can cause cell lethality.

Apart from mutations in the and genes, there are several other mechanisms associated with HRD. Defects in HR repair can be because of epigenetic changes such as promoter methylation, somatic mutations in key HR-related genes, and frequent copy number alterations.149 In addition, mutations in other genes may result in HR-defective tumors and include but are not limited to and .150–153

The most common approach to test for HRD is genomic testing for alterations in and on the basis that and germline and somatic mutations are known to cause HRD. Testing for additional genes involved in DNA damage repair through HR can also be done through commercial resources. Several other approaches have been developed to measure tumor DNA repair function.154 The myChoice HRD test is an NGS assay that uses DNA extracted from formalin-fixed, paraffin-embedded or frozen tumor tissue. A tumor can be characterized as HR-deficient or HR-nondeficient by combining the HRD score that it generates and its mutation status. HRD is defined as an HRD score ≥42 or the presence of a mutation in . As an example of its accuracy, the myChoice HRD assay was seen to identify 100% of BRCA-mutated tumors and 57% of non–BRCA-mutated tumors that had HR deficiencies in patients with platinum-sensitive, high-grade, serous or BRCA-mutated, recurrent ovarian cancer.65

The FoundationFocus CDx (Foundation Medicine, Inc) assay was used to detect both germline and somatic mutation types associated with response to PARP inhibitor therapy.155, 156 This modified NGS-based assay determined the percentage of genomic loss of heterozygosity, mutations in , and other HR genes in tumor tissue of patients with ovarian cancers taking part in the ARIEL PARP inhibitor rucaparib trial. A prespecified cutoff of ≥14% for high loss of heterozygosity was determined. FoundationFocus CDx is the first FDA-approved companion diagnostic assay for rucaparib for the treatment of advanced ovarian cancer.

As we understand more about HRD in various cancer types, the indications for the use of PARP inhibitors will likely be broadened. Certain cancers, including ovarian, fallopian tube, breast, primary peritoneal, and GI (specifically a subgroup of pancreatic adenocarcinomas and gastric/esophageal cancers), have been shown to harbor aberrations in genes involved in the HRD pathway. Mutations are seen not only in and but also in other relevant genes, such as , or .157, 158 Several PARP inhibitors have been FDA approved for the treatment of specific types of ovarian (olaparib, rucaparib, and niraparib), fallopian tube (olaparib and niraparib), breast (olaparib), primary peritoneal (olaparib and niraparib), and pancreatic (olaparib) cancers, but not yet for other GI cancers. However, at present, HRD testing before PARP-inhibitor therapy is not necessary.

Other Hot Markers in Research

Although it is not by any means an exhaustive list, some exciting new biomarkers and their targeted therapies are discussed below.

and Entrectinib

Entrectinib (RXDX-101) was granted a breakthrough therapy designation by the FDA in 2017 for use as a treatment for adult and pediatric patients with -positive, locally advanced or metastatic solid tumors who have either progressed after prior therapies or who have no acceptable standard therapy options (see also Neurotrophic receptor tyrosine kinase, above).27, 28 A trial studying the treatment of patients with solid tumors (breast cancer, cholangiocarcinoma, CRC, head and neck neoplasms, melanoma, neuroendocrine tumors, NSCLC, ovarian cancer, pancreatic cancer, papillary thyroid cancer, primary brain tumors, renal cell carcinoma, and sarcomas) that harbor an , , or fusion is ongoing (ClinicalTrials.gov identifier NCT02568267). In this trial, patients are assigned to different baskets according to tumor type and gene fusion. The primary outcome of the study will be the objective response rate to entrectinib.27

fusions may act as actionable targets in conjunction with other potentially targetable alterations, such as PD-L1–positive or MSI-H status, meaning that therapeutic combinations (TRK inhibitors plus immune checkpoint inhibitors, for example) are a promising strategy.159

and Erdafitinib

The fibroblast growth factor receptor (FGFR) family comprises part of a tyrosine kinase signaling pathway that plays a role in oncogenesis through gene amplification, activating mutations, or translocation in several tumor types. Erdafitinib is an orally administered FGFR family tyrosine kinase inhibitor. Earlier this year, the FDA granted Breakthrough Therapy Designation for erdafitinib in the treatment of urothelial cancer, which is based on data from a multicenter phase 2 clinical trial focused on evaluating the efficacy and safety of erdafitinib in the treatment of adult patients with locally advanced or metastatic urothelial cancer harboring specific mutations.160 The overall response rate was 42% in 59 patients for whom data were available.160 Erdafitinib is also under investigation in the NCI-MATCH trial as a treatment for patients with tumors that have an mutation, fusion, or amplification (Table 7).

Also in the NCI-MATCH trial, 5 of 50 patients with an aberrant FGFR pathway had a partial response to AZD4547 (another FGFR tyrosine kinase inhibitor).161 Two of these patient’s tumors had point mutations in , and 2 others had fusions, suggesting that these particular types of mutation have increased sensitivity to the drug, which warrants further study in this patient subtype.

Amplification and Exon 14 and Crizotinib

Aberrant activation of MET receptor tyrosine kinase signaling occurs in various cancer types as result of various alterations, including amplification and an exon 14 mutation. Crizotinib is an ALK/ROS1/MET inhibitor that is already FDA approved in -positive or -positive NSCLC but also has proven clinical activity in cases of exon 14 alterations and amplification. Preclinical studies have shown that inhibition of MET using crizotinib resulted in the inhibition of growth of cancer cells that possessed amplification both in vitro in cell lines and in vivo in preclinical models.162 In an updated phase 1 analysis of crizotinib in patients with low, medium, and high levels of amplification in advanced NSCLC, patients with high amplification showed clinically meaningful antitumor activity with rapid and durable responses. Crizotinib was generally well tolerated163 and is currently under study in the ASCO TAPUR trial for patients with tumors that have , , or mutations and in the NCI-MATCH trial as a treatment for patients with tumors that have a amplification, exon 14 mutation, translocation, or translocation or inversion (Table 7).

mTOR and Sapanisertib (TAK-228)

The mammalian target of rapamycin (mTOR) is a kinase encoded in humans by . mTOR exists as a core component in 2 distinct multiple-protein complexes, TORC1 and TORC2. These complexes regulate several different cellular processes, including cell proliferation, cell motility, cell survival, protein synthesis, autophagy, and transcription. Sapanisertib (TAK-228) demonstrated a reasonable safety profile as well as promising preliminary antitumor activity in a range of tumor types with aberrant .164 Tuberous sclerosis complex 1 and 2 ( and ) mutations are also observed in certain tumor subtypes and may be targeted by sapanisertib. The agent is under investigation in the NCI-MATCH trial as a treatment for patients with tumors that have or mutations (Table 7).

and Taselisib

In the NCI-MATCH trial, 65 patients with a mutated phosphatidylinositol 3-kinase gene () were treated with taselisib (a PIK3CA inhibitor) and, although there were no ORs to the drug, 24% of patients had prolonged stable disease for more than 6 months. Further research in selected cancer types is warranted.165

and Palbociclib

The cyclin-dependent kinases (CDKs) CDK4 and CDK6 play a crucial role in the G1-S phase transition during cell cycling. Palbociclib, an inhibitor of aberrant CDK4/CDK6, is FDA approved for the treatment of hormone receptor–positive, HER2-negative, advanced or metastatic breast cancer in combination with an aromatase inhibitor as initial endocrine-based therapy in postmenopausal women.166 Its effect on certain GI tumors is under investigation in the clinic.167 Palbociclib is also under investigation in the NCI-MATCH trial as a treatment for patients with tumors that have or amplification or , , or amplification (and Rb expression/protein in both study arms). The ASCO TAPUR trial is also investigating palbociclib in the treatment of patients with tumors that harbor , , or amplifications (Table 7).

and Dasatinib

DDR2 is a transmembrane receptor tyrosine kinase that plays a role in cancer progression by regulating the interactions of tumor cells with their surrounding collagen matrix. mutations are seen in several tumor types, including lung cancer, breast cancer, brain cancer, gynecologic cancer, and prostate cancer.168 The multikinase inhibitor dasatinib blocks DDR2 kinase activity to various degrees and is under investigation in the treatment of patients with tumors that possess a S768R, I638F, or L239R mutation (NCI-MATCH). The agent is also under investigation in the treatment of patients with tumors that harbor , or mutations (TAPUR trial) (Table 7).

Emerging Techniques

The Liquid Biopsy: Circulating Tumor Cells and Exosomes

Peripheral blood samples are a biomarker source by way of circulating tumor cells (DNA) and circulating nucleic acids or associated extracellular vesicles or exosomes.169–171 The use of liquid biopsy profiling has proven useful in selected clinical scenarios but, to date, despite its potential in the management of patients with most metastatic solid tumors, this technique has not established a firm role in standard practice. Obvious advantages include ease of access to the tissue through a simple blood draw. An additional advantage is that circulating samples may help reduce the problem of tumor heterogeneity as it reflects the sum total of the tumor. However, when blood and tumor tissue are concurrently collected and analyzed, results from circulating tumor cell analyses do not always match those obtained from tumor tissue analyses and thus are not always reliable.

Regarding extracellular vesicles or exosomes, a newly developed, minimally invasive ADAPT Biotargeting System characterizes complex biological systems in their inherent state(s) and relies on the fact that a large number of cells in the body secrete extracellular vesicles into the circulation, and the molecular composition of these “exosomes” correlates with the cell of origin. Through intercellular communication, exosomes play a part in controlling many tumor progressive processes, including immune evasion, angiogenesis, and metastasis.171 The ADAPT assay has been shown to have potential for biomarker identification and therapeutic use across most cancer types.171, 172

Community Hospital Molecular Testing and Assessment Program

Because 85% of cancer care is delivered in a community setting, it is imperative that the programmatic decisions concerning molecular testing for cancer include standardization, physician engagement, and application to point of care. Hoag Hospital (a large community hospital in Newport Beach, California) decided to embark on an initiative within the context of precision oncology and identified the need for molecular testing. A committee of interested physicians (including molecular pathologists) and administrators was formed, and a request for proposal was developed and put to several CLIA-certified vendors to outsource genomic testing. Certain specific qualifications were emphasized, such as price point; turnaround time; results reporting and support structure; portfolio of testing, including NGS, CNV, and IHC for protein analysis and fusion genes (all preferably using disease-specific panels); and preauthorization and billing services. Several meetings were required to reach a final decision. Once the vendor was selected and contracting was optimized, all tumor molecular testing was standardized, and all ordering and tissue processing was sent through the central pathology laboratory. It is important to note that this process had immediate benefit for the cancer programs because, before this arrangement, molecular testing was haphazard. Tumor tissue samples were being sent by individual physicians to multiple different laboratories/vendors without standardization of tissue collection and processing, and subsequent result reports were generally faxed and unavailable when needed for valuable treatment assessment in the standard clinical or research settings.

After vetting and education concerning this new molecular testing and assessment program at disease site committee and tumor board meetings, it was further decided that genomic testing would be reflexed and ordered by pathology for selected clinical stages of solid tumors. The initial pilot project for this reflex testing was in NSCLC stages IB through IV. Before this initial pilot of genomic reflex testing in lung cancer, approximately 50% of the tumors in patients with advanced lung cancer were being tested for even the minimal NCCN guideline biomarkers. Now, over 95% of advanced lung cancers undergo genomic profiling evaluation that has resulted in pervasive use at the point of care. An illustrated example of this was a recent patient’s lung cancer demonstrating an actionable mutation in . The success of this pilot study in reflex profiling has expanded to other cancer disease sites, such as advanced head and neck cancers, ovarian cancers, glioblastoma multiforme, sarcoma, and rare tumors. Over the past 18 months, more than 300 tumors have undergone clinical grade genomic profiling, and those data are readily available to the treating physician through web access. Importantly, physician education is a key component to the establishment of a comprehensive cancer molecular genomics program. Initiation of the more routine use of molecular markers and genomic profiling has stimulated interest and participation in the expanding clinical applications. Along with the assistance of molecular pathology, presentation of genomic data is now frequently requested, and this provides points of discussion in the cancer disease site tumor board meetings. In addition, this information has led to optimal patient selection and a definitive increase in referrals to the phase 1 clinical trials program.

Despite the successful implementation of this intuitional program in cancer molecular testing, challenges remain. These fall into several categories, such as reimbursement issues, evaluating the tumor genomic data for potential germline testing, paring results to clinical trial opportunities, and collecting and collating the genomic data to clinical information and outcomes, as well as the incorporation of new opportunities such as sequencing cell-free DNA or routine pharmacogenetic testing. The overall goal of this program was to provide added value and physician engagement in precision oncology. This is now a work in progress, but we are already evaluating the growth opportunity to enhanced patient care.

Challenges and Open Questions in Molecular Profiling

Absence of Randomized, Controlled Clinical Trials

One of the major challenges with the use of large panels or whole genome sequencing is the absence of randomized clinical trials demonstrating benefit. Although some retrospective analyses have appeared promising, in general, the field demands more evidence given the expense of drugs and genomic tests and potential harm from exposing patients to toxicity of drugs without proof of efficacy. A properly performed randomized clinical trial requires a large number of patients and needs to be a national study requiring significant resources to cover costs of testing, cost of drugs, and data analysis. It would be problematic to randomize patients to genomic profiling versus no testing. It is also complicated to compare standard therapy with molecularly assigned therapy in terms of comparing apples to apples and defining what the valid endpoints should be. It is clear that the gold standard is OS, but this may be affected by multiple lines of therapy and ultimate use of targeted therapy beyond a particular study. PFS is a reasonable endpoint but, if the randomization occurs at a time when there is effective standard therapy, then the impact of molecularly targeted therapy may be underestimated. With the era of precision oncology, there is opportunity to break new ground in trial design. In this regard, pooling N-of-one data that account for other factors described below, such as tumor heterogeneity, may allow for useful evidence to help patients even if it does not rise to randomization. In addition, the PFS ratio as defined by Von Hoff remains a good metric to determine benefit of molecularly assigned therapy.

Lack of Drugs

Another major challenge is the lack of availability of drugs for numerous drivers of various cancers. Examples of major drivers for which there are currently no approved drugs include mutated β-catenin, mutated , or mutated , among others. On the optimistic side, as drugs are discovered and developed, they can be offered retrospectively to patients who have actionable mutations. A related issue is the lack of drugs that effectively target emergent resistance mechanisms. There are some exceptions with mutated , , or .

Tumor Heterogeneity

It is clear that advanced cancers, especially those that have been treated, harbor significant tumor heterogeneity. This includes intralesion heterogeneity, interlesion heterogeneity, and interpatient heterogeneity, all of which complicate treatment recommendations and outcomes of studies.

Platform Heterogeneity

In clinical practice, there are several different available platforms for molecular profiling; each test has its own sensitivity and specificity. The scope of the testing varies in the number of genes, whether RNA or protein expression is assessed, and whether actionable fusions are readily detectable. Such heterogeneity makes it difficult to pool data from different platforms. The various commercial platforms or those performed within academic institutions are continuously evolving, further complicating the issue of platform heterogeneity. Thus older platforms that did not capture actionable targets for which there are effective therapies lead to data sets that may underestimate the value of genomic testing with respect to patient outcomes. Moreover, no study to date has shown that larger panels are worth doing over smaller targeted gene panels that are part of standard of care (eg, , , , and MSI in CRC). However, it is not unreasonable to expect that the many genomic changes representing the tail end of the curve of drivers may affect patient outcomes, especially if there are available drugs that target their pathways. With the emergence and popularity of liquid biopsy, this yet further adds complexity to the platforms. Of note, in the TAPUR study, which models real-life situations, liquid biopsy is acceptable for the identification of actionable targets to allow enrollment in a treatment arm.

CLIA-approved laboratories offering molecular panel analysis

Before a patient sample of any kind can be tested, the assay in question must be validated in a CLIA-certified laboratory. CLIA defines a clinical laboratory as any facility that performs laboratory testing on specimens derived from humans for the purpose of providing information for the diagnosis, prevention, or treatment of a disease or impairment. The CMS regulates all laboratory testing performed on humans in the United States through the CLIA program. In total, CLIA covers approximately 260,000 laboratory entities. The Division of Clinical Laboratory Improvement and Quality, within the Quality, Safety, and Oversight Group under the Center for Clinical Standards and Quality, has responsibility for implementing the CLIA program. The objective of the CLIA program is to ensure quality laboratory testing. Although all clinical laboratories must be properly certified to receive Medicare or Medicaid payments, CLIA has no direct Medicare or Medicaid program responsibilities.

Clinical laboratories must constantly evolve their test offerings to support the most recent advances in cancer care. For NGS tumor profiling assays, there are multiple commercially available kits with similar claims for gene content and sensitivity. Many factors contribute to the decision of which assays or customized solutions will best meet the needs of the laboratory, clinicians, and patient population. Kit costs, capital equipment expenditures, complexity of workflow, and turnaround time are all important factors that can be relatively easily compared and assessed between assay systems. However, the more important parameters for determining effective, personalized treatment for patients are accuracy, reproducibility, and sensitivity of the assay, and these can be much more challenging to critically evaluate but must be rigorously validated.173–176 The use of a highly multiplexed, consistent, and well-characterized reference material greatly facilitates the comparison of assay systems.177

Data Sharing

Given the numerous challenges described above and others, including limitations of electronic medical records, issues with intellectual property, commercial interests, Health Insurance Portability and Accountability Act of 1996 (HIPAA) regulations, and quality of clinical outcomes data, there are major difficulties with data sharing to expand data sets through larger sample size. In this regard, the Caris Precision Oncology Alliance has been addressing some of the issues through the CODE database, which is increasing the number of patients for whom analysis of clinical outcomes is possible as a function of clinical, genomic, or drug use. Other initiatives of national prominence include Orion, Project GENIE, the WIN Consortium, the Precision Medicine Exchange Consortium, and the Memorial Sloan Kettering (MSK) Cancer Alliance.

Timing

The ideal time to perform genomic testing to maximize its therapeutic value to individual patients with cancer remains a matter of controversy, and the evidence base on which to make recommendations is still evolving. Initially, physicians mainly ordered testing in patients with advanced disease who had exhausted standard-of-care treatment options to help inform choices for treatment with experimental agents. Trials are now underway in groups of patients with earlier stages of disease. For example, the newly activated, NCI-sponsored intergroup stage III colon cancer adjuvant therapy trial randomizes patients to standard adjuvant therapy with FOLFOX or FOLFOX plus experimental treatment with a PD-1 inhibitor. Only patients with MSI-H, stage III colon cancer will be eligible for randomization, and the eligibility determination mandates genomic testing for defective MMR in patients with localized disease. Commercial genomic testing now includes MSI testing in addition to a battery of genes relevant to tumor progression. The addition of MSI testing to these panels is a consequence of the recent approval of pembrolizumab and nivolumab. The agents are currently approved for the treatment of refractory tumors of any histology that exhibit defective MMR.141 In patients with advanced cancers, it is clear that the tumor genome continues to evolve with time and with the pressure exerted on cells by treatments that selectively favor the growth of treatment-resistant subclones.178 Investigators remain concerned about clonal evolution and often will recommend rebiopsy of tumors when patients have refractory disease to ensure that the genomic analysis used to make informed decisions about clinical trials with targeted agents is reflective of the tumor’s current biology.179 Others suggest testing at the first sign of advanced disease, as the efficacy of conventional chemotherapy is variable and strategies for salvage therapy may be required sooner rather than later. The use of “liquid biopsies” either on circulating tumor cells or on cell-free DNA has been touted as a method of assessing the tumor genome without the need for a repeat, invasive biopsy. This remains a research tool at this time and is not generally a part of clinical practice. Continued data analyses are urgently needed and will inevitably occur as more samples are tested and the practical application of these assessments are translated into treatment decisions.

Discussion

In 2019, optimum cancer care requires state-of-the-art molecular diagnostics, a solid knowledge base to interpret and apply the results, and a nearly constant awareness of changes on the horizon. The field is moving that quickly. Comprehensive testing performed on our patients at the beginning of 2019 is likely to be incomplete today. Drug approvals are no longer based solely on large phase 3 trials; these late-stage trials are being replaced by “basket” and “umbrella” trials, allowing us to ensure that the right drug is given to the right patient faster. Subsequent new regulatory and payer approvals seem to come daily. Precision medicine is now a part of our standard practice, but with this comes many new challenges. How do we deal with tumor heterogeneity, and will liquid biopsies satisfactorily replace tissue-based testing? Are we justified in, and can we afford, testing a large sample of patients, knowing that we will only rarely find the sought after “needle in the haystack?” How do we manage our patients’ expectations when there is so much press and hype surrounding our new discoveries? Can we afford to develop and ultimately to pay for increasingly expensive therapies targeting increasingly smaller proportions of patients? There is, of course, no turning back, but there is much work ahead.

Acknowledgements

Marion L. Hartley, PhD, provided invaluable writing and editing support, table and bibliography creation, author coordination, and article organization. These medical writing activities were supported by Caris Life Sciences.

Supporting Information

Filename
Description

caac21560-sup-0001-SI.docxapplication/docx, 26.1 KB
 

Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

References

• 1

  Centers for Medicare & Medicaid Services

  .

  CMS finalizes coverage of next generation sequencing tests, ensuring enhanced access for cancer patients [press release]

  .

  Baltimore, MD

  : Centers for Medicare & Medicaid Services;

  2018

  . cms.gov/newsroom/press-releases/cms-finalizes-coverage-next-generation-sequencing-tests-ensuring-enhanced-access-cancer-patients. Accessed February 6, 2019.

• 2

  US Food and Drug Administration

  .

  FoundationOne CDx-P170019

  .

  Silver Spring, MD

  : US Food and Drug Administration;

  2017

  . fda.gov/medicaldevices/productsandmedicalprocedures/deviceapprovalsandclearances/recently-approveddevices/ucm590331.htm. Accessed February 6, 2019.

• 3

  Bhuvaneshwar K

  ,

  Belouali A

  ,

  Singh V

  , et al.

  G-DOC Plus—an integrative bioinformatics platform for precision medicine

  .

  2016

  ;

  17

  :

  193

  .

• 4

  Ciardiello F

  ,

  Adams R

  ,

  Tabernero J

  , et al.

  Awareness, understanding, and adoption of precision medicine to deliver personalized treatment for patients with cancer: a multinational survey comparison of physicians and patients

  .

  2016

  ;

  21

  :

  292

  –

  300

  .

• 5

  Verma M

  .

  Personalized medicine and cancer

  .

  2012

  ;

  2

  :

  1

  –

  14

  .

• 6

  Juhl H

  .

  Preanalytical aspects: a neglected issue

  .

  2010

  ;

  242

  :

  63

  –

  65

  .

• 7

  Lange N

  ,

  Unger F

  ,

  Schoppler M

  ,

  Pursche K

  ,

  Juhl H

  ,

  David KA

  .

  Identification and validation of a potential marker of tissue quality using gene expression analysis of human colorectal tissue

  .

  2015

  ;

  10

  :

  e0133987

  . doi:10.1371/journal.pone.0133987

• 8

  Boland CR

  ,

  Goel A

  .

  Microsatellite instability in colorectal cancer

  .

  2010

  ;

  138

  :

  2073

  –

  2087.e3

  .

• 9

  Lynch HT

  ,

  de la Chapelle A

  .

  Hereditary colorectal cancer

  .

  2003

  ;

  348

  :

  919

  –

  932

  .

• 10

  Vanderwalde A

  ,

  Spetzler D

  ,

  Xiao N

  ,

  Gatalica Z

  ,

  Marshall J

  .

  Microsatellite instability status determined by next-generation sequencing and compared with PD-L1 and tumor mutational burden in 11,348 patients

  .

  2018

  ;

  7

  :

  746

  –

  756

  .

• 11

  Bonneville R

  ,

  Krook MA

  ,

  Kautto EA

  , et al.

  Landscape of microsatellite instability across 39 cancer types

  .

  2017

  ;

  1

  :

  1

  –

  15

  . doi:10.1200/PO.17.00073

• 12

  Chang L

  ,

  Chang M

  ,

  Kautto HM

  , et al.

  Microsatellite instability: a predictive biomarker for cancer immunotherapy

  .

  2018

  ;

  26

  :e15-e21. doi:10.1097/PAI.0000000000000575.

• 13

  Ott PA

  ,

  Bang YJ

  ,

  Berton-Rigaud D

  , et al.

  Safety and antitumor activity of pembrolizumab in advanced programmed death ligand 1-positive endometrial cancer: results from the KEYNOTE-028 Study

  .

  2017

  ;

  35

  :2535-2541.

• 14

  US Food and Drug Administration

  .

  FDA grants accelerated approval to pembrolizumab for first tissue/site agnostic indication.

  . Silver Spring, MD: US Food and Drug Administration;

  2017

  . fda.gov/drugs/informationondrugs/approveddrugs/ucm560040.htm. Accessed February 6, 2019

• 15

  Pai-Scherf L

  ,

  Blumenthal GM

  ,

  Li H

  , et al.

  FDA approval summary: pembrolizumab for treatment of metastatic non-small cell lung cancer: firstline therapy and beyond

  .

  2017

  ;

  22

  :

  1392

  –

  1399

  .

• 16

  Broderick JM

  .

  FDA Approves Nivolumab/Ipilimumab for MSI-H/dMMR Colorectal Cancer Wednesday, Jul 11, 2018

  . onclive.com/web-exclusives/fda-approves-nivolumabipilimumab-for-msihdmmr-colorectal-cancer.

• 17

  Overman MJ

  ,

  Lonardi S

  ,

  Wong KYM

  , et al.

  Durable clinical benefit with nivolumab plus ipilimumab in DNA mismatch repair-deficient/microsatellite instability-high metastatic colorectal cancer

  .

  2018

  ;

  36

  :

  773

  –

  779

  . doi:10.1200/JCO.2017.76.9901

• 18

  Gatalica Z

  ,

  Xiu J

  ,

  Swensen J

  ,

  Vranic S

  Molecular characterization of cancers with NTRK gene fusions

  .

  2019

  ;

  32

  :

  147

  –

  153

  .

• 19

  Planchard D

  ,

  Smit EF

  ,

  Groen HJM

  , et al.

  Dabrafenib plus trametinib in patients with previously treated BRAF (V600E)-mutant metastatic non-small cell lung cancer: an open-label, multicentre phase 2 trial

  . .

  2017

  ;

  18

  :

  1307

  –

  1316

  .

• 20

  Li BT

  ,

  Shen R

  ,

  Buonocore D

  , et al.

  Ado-trastuzumab emtansine in patients with HER2-mutant lung cancers: results from a phase II basket trial

  .

  2018

  ;

  36

  :

  2532

  –

  2537

  .

• 21

  Paik PK

  ,

  Drilon A

  ,

  Fan PD

  , et al.

  Response to MET inhibitors in patients with stage IV lung adenocarcinomas harboring MET mutations causing exon 14 skipping

  .

  2015

  ;

  5

  :

  842

  –

  849

  . doi:10.1158/2159-8290.CD-14-1467

• 22

  Lee SH

  ,

  Lee JK

  ,

  Ahn MJ

  , et al.

  Vandetanib in pretreated patients with advanced non-small cell lung cancer-harboring RET rearrangement: a phase II clinical trial

  .

  2017

  ;

  28

  :

  292

  –

  297

  .

• 23

  Drilon A

  ,

  Rekhtman N

  ,

  Arcila M

  , et al.

  Cabozantinib in patients with advanced RET-rearranged non-small-cell lung cancer: an open-label, single-centre, phase 2, single-arm trial

  . .

  2016

  ;

  17

  :

  1653

  –

  1660

  .

• 24

  Lindeman NI

  ,

  Cagle PT

  ,

  Aisner DL

  , et al.

  Updated molecular testing guideline for the selection of lung cancer patients for treatment with targeted tyrosine kinase inhibitors: guideline from the College of American Pathologists, the International Association for the Study of Lung Cancer, and the Association for Molecular Pathology

  .

  2018

  ;

  20

  :

  129

  –

  159

  .

• 25

  Fuse MJ

  ,

  Okada K

  ,

  Oh-Hara T

  ,

  Ogura H

  ,

  Fujita N

  ,

  Katayama R

  .

  Mechanisms of resistance to NTRK inhibitors and therapeutic strategies in NTRK1-rearranged cancers

  .

  2017

  ;

  16

  :

  2130

  –

  2143

  .

• 26

  US Food and Drug Administration

  .

  FDA approves larotrectinib for solid tumors with NTRK gene fusions

  .

  Silver Spring, MD

  : US Food and Drug Administration;

  2018

  . fda.gov/Drugs/InformationOnDrugs/ApprovedDrugs/ucm626720.htm. Accessed February 6, 2019.

• 27

  Drilon A

  ,

  De Braud FG

  ,

  Siena S

  , et al.

  Entrectinib, an oral pan-Trk, ROS1, and ALK inhibitor in TKI-naïve patients with advanced solid tumors harboring gene rearrangements

  . Presented at: The 2016 AACR Annual Meeting; April 16-20, 2016; New Orleans, LA. Abstract CT007.

• 28

  FDA Grants Entrectinib Breakthrough Designation for NTRK+ Solid Tumors

  . onclive.com/web-exclusives/fda-grants-entrectinib-breakthrough-designation-for-ntrk-solid-tumors Published: Tuesday, Accessed May 16, 2017

• 29

  Hechtman JF

  ,

  Benayed R

  ,

  Hyman DM

  , et al.

  Pan-Trk immunohistochemistry is an efficient and reliable screen for the detection of NTRK fusions

  .

  2017

  ;

  41

  :

  1547

  –

  1551

  .

• 30

  Hung YP

  ,

  Fletcher CDM

  ,

  Hornick JL

  .

  Evaluation of pan-TRK immunohistochemistry in infantile fibrosarcoma, lipofibromatosis-like neural tumour and histological mimics

  .

  2018

  ;

  73

  :

  634

  –

  644

  .

• 31

  Murphy DA

  ,

  Ely HA

  ,

  Shoemaker R

  , et al.

  Detecting gene rearrangements in patient populations through a 2-step diagnostic test comprised of rapid IHC enrichment followed by sensitive next-generation sequencing

  .

  2017

  ;

  25

  :

  513

  –

  523

  .

• 32

  Lartigue J

  .

  Blurring the lines between germline and somatic mutations in cancer

  . .

  2017

  ;

  18

  . onclive.com/publications/oncology-live/2017/vol-18-no-13/blurring-the-lines-between-germline-and-somatic-mutations-in-cancer. Accessed February 6, 2019.

• 33

  Ding L

  ,

  Getz G

  ,

  Wheeler DA

  , et al.

  Somatic mutations affect key pathways in lung adenocarcinoma

  .

  2008

  ;

  455

  :

  1069

  –

  1075

  .

• 34

  Yurgelun MB

  ,

  Kulke MH

  ,

  Fuchs CS

  , et al.

  Cancer susceptibility gene mutations in individuals with colorectal cancer

  .

  2017

  ;

  35

  :

  1086

  –

  1095

  .

• 35

  Waddell N

  ,

  Pajic M

  ,

  Patch AM

  , et al.

  Whole genomes redefine the mutational landscape of pancreatic cancer

  .

  2015

  ;

  518

  :

  495

  –

  501

  .

• 36

  Schrader KA

  ,

  Cheng DT

  ,

  Joseph V

  , et al.

  Germline variants in targeted tumor sequencing using matched normal DNA

  .

  2016

  ;

  2

  :

  104

  –

  111

  .

• 37

  Sinicrope FA

  ,

  Sargent DJ

  .

  Clinical implications of microsatellite instability in sporadic colon cancers

  .

  2009

  ;

  21

  :

  369

  –

  373

  .

• 38

  Ribic CM

  ,

  Sargent DJ

  ,

  Moore MJ

  , et al.

  Tumor microsatellite-instability status as a predictor of benefit from fluorouracil-based adjuvant chemotherapy for colon cancer

  .

  2003

  ;

  349

  :

  247

  –

  257

  .

• 39

  Sargent DJ

  ,

  Marsoni S

  ,

  Monges G

  , et al.

  Defective mismatch repair as a predictive marker for lack of efficacy of fluorouracil-based adjuvant therapy in colon cancer

  .

  2010

  ;

  28

  :

  3219

  –

  3226

  .

• 40

  Sinicrope FA

  ,

  Foster NR

  ,

  Thibodeau SN

  , et al.

  DNA mismatch repair status and colon cancer recurrence and survival in clinical trials of 5-fluorouracil-based adjuvant therapy

  .

  2011

  ;

  103

  :

  863

  –

  875

  .

• 41

  Evaluation of Genomic Applications in Practice and Prevention (EGAPP) Working Group

  .

  Recommendations from the EGAPP Working Group: genetic testing strategies in newly diagnosed individuals with colorectal cancer aimed at reducing morbidity and mortality from Lynch syndrome in relatives

  . .

  2009

  ;

  11

  :

  35

  –

  41

  .

• 42

  Fuchs CS

  ,

  Doi T

  ,

  Jang RWJ

  , et al.

  KEYNOTE-059 cohort 1: efficacy and safety of pembrolizumab (pembro) monotherapy in patients with previously treated advanced gastric cancer [abstract]

  .

  2017

  ;

  35

  (

  15 suppl

  ):

  4003

  .

• 43

  Burn J

  ,

  Gerdes AM

  ,

  Macrae F

  , et al.

  Long-term effect of aspirin on cancer risk in carriers of hereditary colorectal cancer: an analysis from the CAPP2 randomised controlled trial

  .

  2009

  ;

  378

  :

  2081

  –

  2087

  .

• 44

  Liao X

  ,

  Lochhead P

  ,

  Nishihara R

  , et al.

  Aspirin use, tumor PIK3CA mutation, and colorectal-cancer survival

  .

  2012

  ;

  367

  :

  1596

  –

  1606

  .

• 45

  Karnoub AE

  ,

  Weinberg RA

  .

  Ras oncogenes: split personalities

  .

  2008

  ;

  9

  :

  517

  –

  531

  .

• 46

  Di Nicolantonio F

  ,

  Martini M

  ,

  Molinari F

  , et al.

  Wild-type BRAF is required for response to panitumumab or cetuximab in metastatic colorectal cancer

  .

  2008

  ;

  26

  :

  5705

  –

  5712

  .

• 47

  Passiglia F

  ,

  Bronte G

  ,

  Bazan V

  ,

  Galvano A

  ,

  Vincenzi B

  ,

  Russo A

  .

  Can KRAS and BRAF mutations limit the benefit of liver resection in metastatic colorectal cancer patients? A systematic review and meta-analysis

  .

  2018

  ;

  99

  :

  150

  –

  157

  .

• 48

  Richman SD

  ,

  Seymour MT

  ,

  Chambers P

  , et al.

  KRAS and BRAF mutations in advanced colorectal cancer are associated with poor prognosis but do not preclude benefit from oxaliplatin or irinotecan: results from the MRC FOCUS trial

  .

  2009

  ;

  27

  :

  5931

  –

  5937

  .

• 49

  Roth AD

  ,

  Tejpar S

  ,

  Delorenzi M

  , et al.

  Prognostic role of KRAS and BRAF in stage II and III resected colon cancer: results of the translational study on the PETACC-3, EORTC 40993, SAKK 60-00 trial

  .

  2010

  ;

  28

  :

  466

  –

  474

  .

• 50

  Schirripa M

  ,

  Bergamo F

  ,

  Cremolini C

  , et al.

  BRAF and RAS mutations as prognostic factors in metastatic colorectal cancer patients undergoing liver resection

  .

  2015

  ;

  112

  :

  1921

  –

  1928

  .

• 51

  Sinicrope FA

  ,

  Shi Q

  ,

  Smyrk TC

  , et al.

  Molecular markers identify subtypes of stage III colon cancer associated with patient outcomes

  .

  2015

  ;

  148

  :

  88

  –

  99

  .

• 52

  Van Cutsem E

  ,

  Kohne CH

  ,

  Lang I

  , et al.

  Cetuximab plus irinotecan, fluorouracil, and leucovorin as first-line treatment for metastatic colorectal cancer: updated analysis of overall survival according to tumor KRAS and BRAF mutation status

  .

  2011

  ;

  29

  :

  2011

  –

  2019

  .

• 53

  Carter J

  ,

  Tseng LH

  ,

  Zheng G

  , et al.

  Non-p.V600E BRAF mutations are common using a more sensitive and broad detection tool

  .

  2015

  ;

  144

  :

  620

  –

  628

  .

• 54

  Tran B

  ,

  Kopetz S

  ,

  Tie J

  , et al.

  Impact of BRAF mutation and microsatellite instability on the pattern of metastatic spread and prognosis in metastatic colorectal cancer

  .

  2011

  ;

  117

  :

  4623

  –

  4632

  .

• 55

  Robertson AG

  ,

  Kim J

  ,

  Al-Ahmadie H

  , et al.

  Comprehensive molecular characterization of muscle-invasive bladder cancer

  .

  2017

  ;

  171

  :

  540

  –

  556, e525

  .

• 56

  Li Q

  ,

  Damish A

  ,

  Frazier ZJ

  , et al.

  ERCC2 helicase domain mutations confer nucleotide excision repair deficiency and drive cisplatin sensitivity in muscle-invasive bladder cancer

  .

  2019

  ;

  25

  :

  977

  –

  988

  . doi:10.1158/1078-0432.CCR-18-1001

• 57

  Van Allen EM

  ,

  Mouw KW

  ,

  Kim P

  , et al.

  Somatic ERCC2 mutations correlate with cisplatin sensitivity in muscle-invasive urothelial carcinoma

  .

  2014

  ;

  4

  :

  1140

  –

  1153

  .

• 58

  Antonarakis ES

  ,

  Lu C

  ,

  Luber B

  , et al.

  Germline DNA-repair gene mutations and outcomes in men with metastatic castration-resistant prostate cancer receiving first-line abiraterone and enzalutamide

  .

  2018

  ;

  74

  :

  218

  –

  225

  .

• 59

  Lu E

  ,

  Thomas GV

  ,

  Chen Y

  , et al.

  DNA repair gene alterations and PARP inhibitor response in patients with metastatic castration-resistant prostate cancer

  .

  2018

  ;

  16

  :

  933

  –

  937

  .

• 60

  De Bono JS

  ,

  Goh JCH

  ,

  Ojamaa K

  , et al.

  KEYNOTE-199: pembrolizumab (pembro) for docetaxel-refractory metastatic castration-resistant prostate cancer (mCRPC) [abstract]

  .

  2018

  ;

  36

  (

  15 suppl

  ):

  5007

  .

• 61

  Carroll PR

  ,

  Parsons JK

  ,

  Andriole G

  , et al.

  NCCN Guidelines Insights: Prostate Cancer Early Detection, Version 2.2016

  . .

  2016

  ;

  14

  :

  509

  –

  519

  .

• 62

  Society of Gynecologic Oncology (SGO)

  .

  SOG Clinical Practice Statement: Screening for Lynch Syndrome in Endometrial Cancer

  .

  Chicago, IL

  : Society of Gynecologic Oncology;

  2014

  . sgo.org/clinical-practice/guidelines/screening-for-lynch-syndrome-in-endometrial-cancer. Accessed February 6, 2019.

• 63

  Na R

  ,

  Zheng SL

  ,

  Han M

  , et al.

  Germline mutations in ATM and BRCA1/2 distinguish risk for lethal and indolent prostate cancer and are associated with early age at death

  .

  2017

  ;

  71

  :

  740

  –

  747

  .

• 64

  Slomovitz BM

  ,

  Jiang Y

  ,

  Yates MS

  , et al.

  Phase II study of everolimus and letrozole in patients with recurrent endometrial carcinoma

  .

  2015

  ;

  33

  :

  930

  –

  936

  .

• 65

  Mirza MR

  ,

  Monk BJ

  ,

  Herrstedt J

  , et al.

  Niraparib maintenance therapy in platinum-sensitive, recurrent ovarian cancer

  .

  2016

  ;

  375

  :

  2154

  –

  2164

  .

• 66

  Coleman RL

  ,

  Oza AM

  ,

  Lorusso D

  , et al.

  Rucaparib maintenance treatment for recurrent ovarian carcinoma after response to platinum therapy (ARIEL3): a randomised, double-blind, placebo-controlled, phase 3 trial

  .

  2017

  ;

  390

  :

  1949

  –

  1961

  .

• 67

  Ledermann J

  ,

  Harter P

  ,

  Gourley C

  , et al.

  Olaparib maintenance therapy in platinum-sensitive relapsed ovarian cancer

  .

  2012

  ;

  366

  :

  1382

  –

  1392

  .

• 68

  Moore K

  ,

  Colombo N

  ,

  Scambia G

  , et al.

  Maintenance olaparib in patients with newly diagnosed advanced ovarian cancer

  .

  2018

  ;

  27

  :

  2495

  –

  2505

  . doi:10.1056/NEJMoa1810858

• 69

  Burger RA

  ,

  Brady MF

  ,

  Bookman MA

  , et al.

  Incorporation of bevacizumab in the primary treatment of ovarian cancer

  .

  2011

  ;

  365

  :

  2473

  –

  2483

  .

• 70

  Oza AM

  ,

  Cook AD

  ,

  Pfisterer J

  , et al.

  Standard chemotherapy with or without bevacizumab for women with newly diagnosed ovarian cancer (ICON7): overall survival results of a phase 3 randomised trial

  .

  2015

  ;

  16

  :

  928

  –

  936

  .

• 71

  Ledermann J

  ,

  Harter P

  ,

  Gourley C

  , et al.

  Olaparib maintenance therapy in patients with platinum-sensitive relapsed serous ovarian cancer: a preplanned retrospective analysis of outcomes by BRCA status in a randomised phase 2 trial

  .

  2014

  ;

  15

  :

  852

  –

  861

  .

• 72

  Niemeier LA

  ,

  Dabbs DJ

  ,

  Beriwal S

  ,

  Striebel JM

  ,

  Bhargava R

  .

  Androgen receptor in breast cancer: expression in estrogen receptor-positive tumors and in estrogen receptor-negative tumors with apocrine differentiation

  .

  2010

  ;

  23

  :

  205

  –

  212

  .

• 73

  Gucalp A

  ,

  Tolaney S

  ,

  Isakoff SJ

  , et al.

  Phase II trial of bicalutamide in patients with androgen receptor-positive, estrogen receptor-negative metastatic breast cancer

  .

  2013

  ;

  19

  :

  5505

  –

  5512

  .

• 74

  Schiavon G

  ,

  Hrebien S

  ,

  Garcia-Murillas I

  , et al.

  Analysis of ESR1 mutation in circulating tumor DNA demonstrates evolution during therapy for metastatic breast cancer

  .

  2015

  ;

  7

  :

  313ra182

  .

• 75

  Louis DN

  ,

  Perry A

  ,

  Reifenberger G

  , et al.

  The 2016 World Health Organization Classification of Tumors of the Central Nervous System: a summary

  .

  2016

  ;

  131

  :

  803

  –

  820

  .

• 76

  Stupp R

  ,

  Brada M

  ,

  van den Bent MJ

  ,

  Tonn JC

  ,

  Pentheroudakis G

  ;

  ESMO Guidelines Working Group

  .

  High-grade glioma: ESMO Clinical Practice Guidelines for diagnosis, treatment and follow-up

  . .

  2014

  ;

  25

  (

  suppl_3

  ):

  iii93

  –

  iii101

  .

• 77

  Louis DN

  ,

  Ohgaki H

  ,

  Wiestler OD

  , et al.

  The 2007 WHO Classification of Tumours of the Central Nervous System

  .

  2007

  ;

  114

  :

  97

  –

  109

  .

• 78

  Donahue B

  ,

  Scott CB

  ,

  Nelson JS

  , et al.

  Influence of an oligodendroglial component on the survival of patients with anaplastic astrocytomas: a report of Radiation Therapy Oncology Group 83-02

  .

  1997

  ;

  38

  :

  911

  –

  914

  .

• 79

  van den Bent MJ

  .

  Interobserver variation of the histopathological diagnosis in clinical trials on glioma: a clinician’s perspective

  .

  2010

  ;

  120

  :

  297

  –

  304

  .

• 80

  Ramkissoon SH

  ,

  Bi WL

  ,

  Schumacher SE

  , et al.

  Clinical implementation of integrated whole-genome copy number and mutation profiling for glioblastoma

  .

  2015

  ;

  17

  :

  1344

  –

  1355

  .

• 81

  Tabone T

  ,

  Abuhusain HJ

  ,

  Nowak AK

  ,

  Erber WN

  ,

  McDonald KL

  .

  Multigene profiling to identify alternative treatment options for glioblastoma: a pilot study

  .

  2014

  ;

  67

  :

  550

  –

  555

  .

• 82

  Blumenthal DT

  ,

  Dvir A

  ,

  Lossos A

  , et al.

  Clinical utility and treatment outcome of comprehensive genomic profiling in high grade glioma patients

  .

  2016

  ;

  130

  :

  211

  –

  219

  .

• 83

  Cook PJ

  ,

  Thomas R

  ,

  Kannan R

  , et al.

  Somatic chromosomal engineering identifies BCAN-NTRK1 as a potent glioma driver and therapeutic target

  . .

  2017

  ;

  8

  :

  15987

  .

• 84

  Alvarez-Breckenridge C

  ,

  Miller JJ

  ,

  Nayyar N

  , et al.

  Clinical and radiographic response following targeting of BCAN-NTRK1 fusion in glioneuronal tumor

  .

  2017

  ;

  1

  :

  5

  .

• 85

  Lassaletta A

  ,

  Zapotocky M

  ,

  Mistry M

  , et al.

  Therapeutic and prognostic implications of BRAF V600E in pediatric low-grade gliomas

  .

  2017

  ;

  35

  :

  2934

  –

  2941

  .

• 86

  Dabrafenib effective in pediatric glioma

  . .

  2017

  ;

  7

  :

  OF5

  .

• 87

  van den Bent M

  ,

  Gan HK

  ,

  Lassman AB

  , et al.

  Efficacy of depatuxizumab mafodotin (ABT-414) monotherapy in patients with EGFR-amplified, recurrent glioblastoma: results from a multi-center, international study

  .

  2017

  ;

  80

  :

  1209

  –

  1217

  .

• 88

  Reardon DA

  ,

  Lassman AB

  ,

  van den Bent M

  , et al.

  Efficacy and safety results of ABT-414 in combination with radiation and temozolomide in newly diagnosed glioblastoma

  .

  2017

  ;

  19

  :

  965

  –

  975

  .

• 89

  Lassman AB

  ,

  van den Bent MJ

  ,

  Gan HK

  , et al.

  Safety and efficacy of depatuxizumab mafodotin + temozolomide in patients with EGFR-amplified, recurrent glioblastoma: results from an international phase I multicenter trial

  .

  2019

  ;

  21

  :

  106

  –

  114

  . doi:10.1093/neuonc/noy091

• 90

  Schindler G

  ,

  Capper D

  ,

  Meyer J

  , et al.

  Analysis of BRAF V600E mutation in 1,320 nervous system tumors reveals high mutation frequencies in pleomorphic xanthoastrocytoma, ganglioglioma and extra-cerebellar pilocytic astrocytoma

  .

  2011

  ;

  121

  :

  397

  –

  405

  .

• 91

  US Food and Drug Administration

  .

  Nivolumab for SCCHN

  .

  Silver Spring, MD

  : US Food and Drug Administration;

  2016

  . fda.gov/drugs/informationondrugs/approveddrugs/ucm528920.htm. Accessed February 6, 2019.

• 92

  US Food and Drug Administration

  .

  Pembrolizumab (KEYTRUDA)

  .

  Silver Spring, MD

  : US Food and Drug Administration;

  2016

  . fda.gov/drugs/informationondrugs/approveddrugs/ucm515627.htm. Accessed February 6, 2019.

• 93

  Agarwal V

  ,

  Subash A

  ,

  Nayar R-C

  ,

  Rao V

  .

  Is EGFR really a therapeutic target in head and neck cancers?

  . .

  2019

  ;

  119

  :

  685

  –

  686

  . doi:10.1002/jso.25386

• 94

  Wheeler DL

  ,

  Dunn EF

  ,

  Harari PM

  .

  Understanding resistance to EGFR inhibitors-impact on future treatment strategies

  .

  2010

  ;

  7

  :

  493

  –

  507

  .

• 95

  Blaszczak W

  ,

  Barczak W

  ,

  Wegner A

  ,

  Golusinski W

  ,

  Suchorska WM

  .

  Clinical value of monoclonal antibodies and tyrosine kinase inhibitors in the treatment of head and neck squamous cell carcinoma

  .

  2017

  ;

  34

  :

  60

  .

• 96

  Jung K

  ,

  Kang H

  ,

  Mehra R

  .

  Targeting phosphoinositide 3-kinase (PI3K) in head and neck squamous cell carcinoma (HNSCC)

  .

  2018

  ;

  3

  :

  3

  .

• 97

  Kobayashi K

  ,

  Hisamatsu K

  ,

  Suzui N

  ,

  Hara A

  ,

  Tomita H

  ,

  Miyazaki T

  .

  A review of HPV-related head and neck cancer

  .

  2018

  ;

  7

  :

  E241

  .

• 98

  Salama AK

  ,

  Flaherty KT

  .

  BRAF in melanoma: current strategies and future directions

  .

  2013

  ;

  19

  :

  4326

  –

  4334

  .

• 99

  Gibney GT

  ,

  Messina JL

  ,

  Fedorenko IV

  ,

  Sondak VK

  ,

  Smalley KS

  .

  Paradoxical oncogenesis—the long-term effects of BRAF inhibition in melanoma

  .

  2013

  ;

  10

  :

  390

  –

  399

  .

• 100

  Long GV

  ,

  Hauschild A

  ,

  Santinami M

  , et al.

  Adjuvant dabrafenib plus trametinib in stage III BRAF-mutated melanoma

  .

  2017

  ;

  377

  :

  1813

  –

  1823

  .

• 101

  Long GV

  ,

  Flaherty KT

  ,

  Stroyakovskiy D

  , et al.

  Dabrafenib plus trametinib versus dabrafenib monotherapy in patients with metastatic BRAF V600E/K-mutant melanoma: long-term survival and safety analysis of a phase 3 study

  .

  2017

  ;

  28

  :

  1631

  –

  1639

  .

• 102

  Beadling C

  ,

  Jacobson-Dunlop E

  ,

  Hodi FS

  , et al.

  KIT gene mutations and copy number in melanoma subtypes

  .

  2008

  ;

  14

  :

  6821

  –

  6828

  .

• 103

  Bastian BC

  ,

  Esteve-Puig R

  .

  Targeting activated KIT signaling for melanoma therapy

  .

  2013

  ;

  31

  :

  3288

  –

  3290

  .

• 104

  Guo J

  ,

  Si L

  ,

  Kong Y

  , et al.

  Phase II, open-label, single-arm trial of imatinib mesylate in patients with metastatic melanoma harboring c-Kit mutation or amplification

  .

  2011

  ;

  29

  :

  2904

  –

  2909

  .

• 105

  Carvajal RD

  ,

  Antonescu CR

  ,

  Wolchok JD

  , et al.

  KIT as a therapeutic target in metastatic melanoma

  .

  2011

  ;

  305

  :

  2327

  –

  2334

  .

• 106

  Hodi FS

  ,

  Corless CL

  ,

  Giobbie-Hurder A

  , et al.

  Imatinib for melanomas harboring mutationally activated or amplified KIT arising on mucosal, acral, and chronically sun-damaged skin

  .

  2013

  ;

  31

  :

  3182

  –

  3190

  .

• 107

  Fedorenko IV

  ,

  Gibney GT

  ,

  Smalley KSM

  .

  NRAS mutant melanoma: biological behavior and future strategies for therapeutic management

  .

  2017

  ;

  32

  :

  3009

  –

  3018

  .

• 108

  Dummer R

  ,

  Schadendorf D

  ,

  Ascierto PA

  , et al.

  Binimetinib versus dacarbazine in patients with advanced NRAS-mutant melanoma (NEMO): a multicentre, open-label, randomised, phase 3 trial

  .

  2017

  ;

  18

  :

  435

  –

  445

  .

• 109

  Gibney GT

  ,

  Weiner LM

  ,

  Atkins MB

  .

  Predictive biomarkers for checkpoint inhibitor-based immunotherapy

  .

  2016

  ;

  17

  :

  e542

  –

  e551

  .

• 110

  Johnson DB

  ,

  Frampton GM

  ,

  Rioth MJ

  , et al.

  Targeted next generation sequencing identifies markers of response to PD-1 blockade

  .

  2016

  ;

  4

  :

  959

  –

  967

  .

• 111

  Hodi FS

  ,

  Chiarion-Sileni V

  ,

  Gonzalez R

  , et al.

  Nivolumab plus ipilimumab or nivolumab alone versus ipilimumab alone in advanced melanoma (CheckMate 067): 4-year outcomes of a multicentre, randomised, phase 3 trial

  .

  2018

  ;

  19

  :

  1480

  –

  1492

  .

• 112

  Gianni L

  ,

  Dafni U

  ,

  Gelber RD

  , et al.

  Treatment with trastuzumab for 1 year after adjuvant chemotherapy in patients with HER2-positive early breast cancer: a 4-year follow-up of a randomised controlled trial

  .

  2011

  ;

  12

  :

  236

  –

  244

  .

• 113

  Baselga J

  ,

  Bradbury I

  ,

  Eidtmann H

  , et al.

  Lapatinib with trastuzumab for HER2-positive early breast cancer (NeoALTTO): a randomised, open-label, multicentre, phase 3 trial

  .

  2012

  ;

  379

  :

  633

  –

  640

  .

• 114

  McNeil C

  .

  NCI-MATCH launch highlights new trial design in precision-medicine era

  .

  2015

  ;

  107

  :

  djv193

  .

• 115

  Mangat PK

  ,

  Halabi S

  ,

  Bruinooge SS

  , et al.

  Rationale and design of the Targeted Agent and Profiling Utilization Registry (TAPUR) Study

  .

  2018

  ;

  2018

  . doi:10.1200/PO.18.00122

• 116

  Durie BG

  ,

  Harousseau JL

  ,

  Miguel JS

  , et al.

  International uniform response criteria for multiple myeloma

  .

  2006

  ;

  20

  :

  1467

  –

  1473

  .

• 117

  Cheson BD

  ,

  Fisher RI

  ,

  Barrington SF

  , et al.

  Recommendations for initial evaluation, staging, and response assessment of Hodgkin and non-Hodgkin lymphoma: the Lugano classification

  .

  2014

  ;

  32

  :

  3059

  –

  3068

  .

• 118

  Lister TA

  ,

  Crowther D

  ,

  Sutcliffe SB

  , et al.

  Report of a committee convened to discuss the evaluation and staging of patients with Hodgkin’s disease: Cotswolds meeting

  .

  1989

  ;

  7

  :

  1630

  –

  1636

  .

• 119

  European Organization for Research and Treatment of Cancer (EORTC)

  .

  SPECTA (Screening Patients for Efficient Clinical Trial Access). Recent Developments of the EORTC Collaborative Program Towards Precision Medicine

  . eortc.org/app/uploads/2017/05/SPECTA-flyer-2015.pdf. Accessed February 6, 2019.

• 120

  Lacombe D

  ,

  Tejpar S

  ,

  Salgado R

  , et al.

  European perspective for effective cancer drug development

  .

  2014

  ;

  11

  :

  492

  –

  498

  .

• 121

  European Organization for Research and Treatment of Cancer

  .

  The European Organization for Research and Treatment of Cancer Screening Patients for Efficient Clinical Trial Access (EORTC-SPECTA) program

  . eortc.org/specta/. Accessed February 6, 2019.

• 122

  Yao S

  ,

  Zhu Y

  ,

  Chen L

  .

  Advances in targeting cell surface signalling molecules for immune modulation

  .

  2013

  ;

  12

  :

  130

  –

  146

  .

• 123

  Chen DS

  ,

  Mellman I

  .

  Oncology meets immunology: the cancer-immunity cycle

  .

  2013

  ;

  39

  :

  1

  –

  10

  .

• 124

  Patel SP

  ,

  Kurzrock R

  .

  PD-L1 expression as a predictive biomarker in cancer immunotherapy

  .

  2015

  ;

  14

  :

  847

  –

  856

  .

• 125

  Mahoney KM

  ,

  Atkins MB

  .

  Prognostic and predictive markers for the new immunotherapies

  .

  2014

  ;

  28

  (

  suppl 3

  ):

  39

  –

  48

  .

• 126

  Garon EB

  ,

  Rizvi NA

  ,

  Hui R

  , et al.

  Pembrolizumab for the treatment of non-small-cell lung cancer

  .

  2015

  ;

  372

  :

  2018

  –

  2028

  .

• 127

  Reck M

  ,

  Rodriguez-Abreu D

  ,

  Robinson AG

  , et al.

  Pembrolizumab versus chemotherapy for PD-L1-positive non-small-cell lung cancer

  .

  2016

  ;

  375

  :

  1823

  –

  1833

  .

• 128

  Herbst RS

  ,

  Baas P

  ,

  Kim DW

  , et al.

  Pembrolizumab versus docetaxel for previously treated, PD-L1-positive, advanced non-small-cell lung cancer (KEYNOTE-010): a randomised controlled trial

  .

  2016

  ;

  387

  :

  1540

  –

  1550

  .

• 129

  Borghaei H

  ,

  Paz-Ares L

  ,

  Horn L

  , et al.

  Nivolumab versus docetaxel in advanced nonsquamous non-small-cell lung cancer

  .

  2015

  ;

  373

  :

  1627

  –

  1639

  .

• 130

  Brahmer J

  ,

  Reckamp KL

  ,

  Baas P

  , et al.

  Nivolumab versus docetaxel in advanced squamous-cell non-small-cell lung cancer

  .

  2015

  ;

  373

  :

  123

  –

  135

  .

• 131

  Mansfield AS

  ,

  Murphy SJ

  ,

  Peikert T

  , et al.

  Heterogeneity of programmed cell death ligand 1 expression in multifocal lung cancer

  .

  2016

  ;

  22

  :

  2177

  –

  2182

  .

• 132

  Chen J

  ,

  Jiang CC

  ,

  Jin L

  ,

  Zhang XD

  .

  Regulation of PD-L1: a novel role of pro-survival signalling in cancer

  .

  2016

  ;

  27

  :

  409

  –

  416

  .

• 133

  Hirsch FR

  ,

  McElhinny A

  ,

  Stanforth D

  , et al.

  PD-L1 immunohistochemistry assays for lung cancer: results from phase 1 of the Blueprint PD-L1 IHC Assay Comparison Project

  .

  2017

  ;

  12

  :

  208

  –

  222

  .

• 134

  Rimm DL

  ,

  Han G

  ,

  Taube JM

  , et al.

  A prospective, multi-institutional, pathologist-based assessment of 4 immunohistochemistry assays for PD-L1 expression in non-small cell lung cancer

  .

  2017

  ;

  3

  :

  1051

  –

  1058

  .

• 135

  Alexandrov LB

  ,

  Nik-Zainal S

  ,

  Wedge DC

  , et al.

  Signatures of mutational processes in human cancer

  . .

  2013

  ;

  500

  :

  415

  –

  421

  .

• 136

  Champiat S

  ,

  Ferte C

  ,

  Lebel-Binay S

  ,

  Eggermont A

  ,

  Soria JC

  .

  Exomics and immunogenics: bridging mutational load and immune checkpoints efficacy

  .

  2014

  ;

  3

  :

  e27817

  .

• 137

  Zibelman M

  ,

  Ramamurthy C

  ,

  Plimack ER

  .

  Emerging role of immunotherapy in urothelial carcinoma—advanced disease

  .

  2016

  ;

  34

  :

  538

  –

  547

  .

• 138

  Snyder A

  ,

  Makarov V

  ,

  Merghoub T

  , et al.

  Genetic basis for clinical response to CTLA-4 blockade in melanoma

  .

  2014

  ;

  371

  :

  2189

  –

  2199

  .

• 139

  Van Allen EM

  ,

  Miao D

  ,

  Schilling B

  , et al.

  Genomic correlates of response to CTLA-4 blockade in metastatic melanoma

  .

  2015

  ;

  350

  :

  207

  –

  211

  .

• 140

  Rizvi NA

  ,

  Hellmann MD

  ,

  Snyder A

  , et al.

  Cancer immunology. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer

  .

  2015

  ;

  348

  :

  124

  –

  128

  .

• 141

  Le DT

  ,

  Uram JN

  ,

  Wang H

  , et al.

  PD-1 blockade in tumors with mismatch-repair deficiency

  .

  2015

  ;

  372

  :

  2509

  –

  2520

  .

• 142

  Miao D

  ,

  Margolis CA

  ,

  Gao W

  , et al.

  Genomic correlates of response to immune checkpoint therapies in clear cell renal cell carcinoma

  .

  2018

  ;

  359

  :

  801

  –

  806

  . doi:10.1126/science.aan5951

• 143

  Helleday T

  .

  Homologous recombination in cancer development, treatment and development of drug resistance

  .

  2010

  ;

  31

  :

  955

  –

  960

  .

• 144

  Li X

  ,

  Heyer WD

  .

  Homologous recombination in DNA repair and DNA damage tolerance

  .

  2008

  ;

  18

  :

  99

  –

  113

  .

• 145

  Lord CJ

  ,

  Ashworth A

  .

  BRCAness revisited

  .

  2016

  ;

  16

  :

  110

  –

  120

  .

• 146

  Krejci L

  ,

  Altmannova V

  ,

  Spirek M

  ,

  Zhao X

  .

  Homologous recombination and its regulation

  .

  2012

  ;

  40

  :

  5795

  –

  5818

  .

• 147

  Turner N

  ,

  Tutt A

  ,

  Ashworth A

  .

  Hallmarks of ‘BRCAness’ in sporadic cancers

  .

  2004

  ;

  4

  :

  814

  –

  819

  .

• 148

  Silver DP

  ,

  Livingston DM

  .

  Mechanisms of BRCA1 tumor suppression

  .

  2012

  ;

  2

  :

  679

  –

  684

  .

• 149

  Watkins JA

  ,

  Irshad S

  ,

  Grigoriadis A

  ,

  Tutt AN

  .

  Genomic scars as biomarkers of homologous recombination deficiency and drug response in breast and ovarian cancers

  .

  2014

  ;

  16

  :

  211

  .

• 150

  Antoniou AC

  ,

  Foulkes WD

  ,

  Tischkowitz M

  .

  Breast-cancer risk in families with mutations in PALB2

  .

  2014

  ;

  371

  :

  1651

  –

  1652

  .

• 151

  Antoniou AC

  ,

  Casadei S

  ,

  Heikkinen T

  , et al.

  Breast-cancer risk in families with mutations in PALB2

  .

  2014

  ;

  371

  :

  497

  –

  506

  .

• 152

  Song H

  ,

  Dicks E

  ,

  Ramus SJ

  , et al.

  Contribution of germline mutations in the RAD51B, RAD51C, and RAD51D genes to ovarian cancer in the population

  .

  2015

  ;

  33

  :

  2901

  –

  2907

  .

• 153

  Ramus SJ

  ,

  Song H

  ,

  Dicks E

  , et al.

  Germline mutations in the BRIP1, BARD1, PALB2, and NBN genes in women with ovarian cancer

  .

  2015

  ;

  107

  :

  djv214

  .

• 154

  den Brok WD

  ,

  Schrader KA

  ,

  Sun S

  , et al.

  Homologous recombination deficiency in breast cancer: a clinical review

  . .

  2016

  ;

  2017

  . doi:10.1200/PO.16.00031

• 155

  Moschetta M

  ,

  George A

  ,

  Kaye SB

  ,

  Banerjee S

  .

  BRCA somatic mutations and epigenetic BRCA modifications in serous ovarian cancer

  .

  2016

  ;

  27

  :

  1449

  –

  1455

  .

• 156

  Hennessy BT

  ,

  Timms KM

  ,

  Carey MS

  , et al.

  Somatic mutations in BRCA1 and BRCA2 could expand the number of patients that benefit from poly (ADP ribose) polymerase inhibitors in ovarian cancer

  .

  2010

  ;

  28

  :

  3570

  –

  3576

  .

• 157

  McCabe N

  ,

  Turner NC

  ,

  Lord CJ

  , et al.

  Deficiency in the repair of DNA damage by homologous recombination and sensitivity to poly(ADP-ribose) polymerase inhibition

  .

  2006

  ;

  66

  :

  8109

  –

  8115

  .

• 158

  Farmer H

  ,

  McCabe N

  ,

  Lord CJ

  , et al.

  Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy

  .

  2005

  ;

  434

  :

  917

  –

  921

  .

• 159

  Gatalica Z

  ,

  Xiu J

  ,

  Swensen J

  ,

  Vranic S

  .

  Molecular characterization of cancers with NTRK gene fusions

  .

  2019

  ;

  32

  :

  147

  –

  153

  .

• 160

  Loriot Y

  ,

  Necchi A

  ,

  Park SE

  , et al.

  Erdafitinib (ERDA; JNJ-42756493), a pan-fibroblast growth factor receptor (FGFR) inhibitor, in patients with metastatic or unresectable urothelial carcinoma (mUC) and FGFR alterations: phase 2 continuous versus intermittent dosing [abstract]

  .

  2018

  ;

  36

  (6 suppl):

  411

  .

• 161

  Chae YK

  ,

  Vaklavas C

  ,

  Cheng HH

  , et al.

  Molecular Analysis for Therapy Choice (MATCH) arm W: phase II study of AZD4547 in patients with tumors with aberrations in the FGFR pathway [abstract]

  .

  2018

  ;

  36

  (

  15 suppl

  ):

  2503

  .

• 162

  Kawakami H

  ,

  Okamoto I

  ,

  Okamoto W

  ,

  Tanizaki J

  ,

  Nakagawa K

  ,

  Nishio K

  .

  Targeting MET amplification as a new oncogenic driver

  .

  2014

  ;

  6

  :

  1540

  –

  1552

  .

• 163

  Camidge DR

  ,

  Otterson GA

  ,

  Clark JW

  , et al.

  Crizotinib in patients (pts) with MET-amplified non-small cell lung cancer (NSCLC): updated safety and efficacy findings from a phase 1 trial [abstract]

  . .

  2018

  ;

  36

  (

  15 suppl

  ):

  9062

  .

• 164

  Burris HA 3rd

  ,

  Kurkjian CD

  ,

  Hart L

  , et al.

  AK-228 (formerly MLN0128), an investigational dual TORC1/2 inhibitor plus paclitaxel, with/without trastuzumab, in patients with advanced solid malignancies

  . .

  2017

  ;

  80

  :

  261

  -273.

• 165

  Krop IE

  ,

  Jegede O

  ,

  Grilley-Olson E

  , et al.

  Results from molecular analysis for therapy choice (MATCH) arm I: taselisib for PIK3CA-mutated tumors [abstract]

  . .

  2018

  ;

  36

  (

  15 suppl

  ):

  101

  .

• 166

  US Food and Drug Administration

  .

  Palbociclib (IBRANCE)

  .

  Silver Spring, MD

  : US Food and Drug Administration;

  2017

  . fda.gov/Drugs/InformationOnDrugs/ApprovedDrugs/ucm549978.htm. Accessed February 6, 2019.

• 167

  Al Baghdadi T

  ,

  Halabi S

  ,

  Garrett-Mayer E

  , et al.

  Palbociclib (P) in patients (Pts) with pancreatic cancer (PC) and gallbladder or bile duct cancer (GBC) with CDKN2A alterations: results from the Targeted Agent and Profiling Utilization Registry (TAPUR) study [abstract]

  .

  2018

  ;

  36

  (

  15 suppl

  ):

  2532

  .

• 168

  Xu C

  ,

  Buczkowski KA

  ,

  Zhang Y

  , et al.

  NSCLC driven by DDR2 mutation is sensitive to dasatinib and JQ1 combination therapy

  .

  2015

  ;

  14

  :

  2382

  –

  2389

  .

• 169

  Bettegowda C

  ,

  Sausen M

  ,

  Leary RJ

  , et al.

  Detection of circulating tumor DNA in early- and late-stage human malignancies

  .

  2014

  ;

  6

  :

  224ra224

  .

• 170

  Diaz LA Jr

  ,

  Bardelli A

  .

  Liquid biopsies: genotyping circulating tumor DNA

  .

  2014

  ;

  32

  :

  579

  –

  586

  .

• 171

  Domenyuk V

  ,

  Zhong Z

  ,

  Stark A

  , et al.

  Plasma exosome profiling of cancer patients by a next generation systems biology approach

  .

  2017

  ;

  7

  :

  42717

  . doi:10.1038/srep42741

• 172

  Domenyuk V

  ,

  Gatalica Z

  ,

  Santhanam R

  , et al.

  Poly-ligand profiling differentiates trastuzumab-treated breast cancer patients according to their outcomes

  .

  2018

  ;

  9

  :

  1219

  .

• 173

  Aziz N

  ,

  Zhao Q

  ,

  Bry L

  , et al.

  College of American Pathologists’ laboratory standards for next-generation sequencing clinical tests

  .

  2015

  ;

  139

  :

  481

  –

  493

  .

• 174

  Gargis AS

  ,

  Kalman L

  ,

  Berry MW

  , et al.

  Assuring the quality of next-generation sequencing in clinical laboratory practice

  .

  2012

  ;

  30

  :

  1033

  –

  1036

  .

• 175

  Jennings LJ

  ,

  Arcila ME

  ,

  Corless C

  , et al.

  Guidelines for validation of next-generation-sequencing-based oncology panels

  .

  2017

  ;

  19

  :

  341

  –

  365

  .

• 176

  Schrijver I

  ,

  Aziz N

  ,

  Farkas DH

  , et al.

  Opportunities and challenges associated with clinical diagnostic genome sequencing: a report of the Association for Molecular Pathology

  .

  2012

  ;

  14

  :

  525

  –

  540

  .

• 177

  Sims DJ

  ,

  Harrington RD

  ,

  Polley EC

  , et al.

  Plasmid-based materials as multiplex quality controls and calibrators for clinical next-generation sequencing assays

  .

  2016

  ;

  18

  :

  336

  –

  349

  .

• 178

  Gerlinger M

  ,

  Rowan AJ

  ,

  Horswell S

  , et al.

  Intratumor heterogeneity and branched evolution revealed by multiregion sequencing

  .

  2012

  ;

  366

  :

  883

  –

  892

  .

• 179

  Thierry AR

  ,

  Pastor B

  ,

  Jiang ZQ

  , et al.

  Circulating DNA demonstrates convergent evolution and common resistance mechanisms during treatment of colorectal cancer

  .

  2017

  ;

  23

  :

  4578

  –

  4591

  .