PerspectiveIn the Journals Plus

Genomic testing for non-small cell lung cancer underused in community-based practices

Show Citation

July 17, 2017

Recommended genomic profiling for EGFR and ALK appeared inconsistent and underused among patients diagnosed with non-small cell lung cancer at community-based practices, according to published findings.

Genomic testing presented logistical challenges for community-based practices, including coordination of sampling handling, long turnaround times, test reimbursement, access to targeted therapies, insufficient tissue and patient harm from repeat biopsies due to insufficient tissue.

“The guidelines recommend broad genomic testing in all patients with advanced non-small cell lung cancer, but adherence is often impractical in the community setting because of the challenges of working with small tissue biopsies,” Martin E. Gutierrez, MD, director of the Drug Discovery and Phase I Unit at John Theurer Cancer Center at Hackensack University Medical Center, said in a press release.

Lung cancer biomarker guidelines from the College of American Pathologists, International Association for the Study of Lung Cancer and Association for Molecular Pathology require EGFR and ALK genomic testing for patients with NSCLC. In 2014, the National Comprehensive Cancer Network broadened their molecular profiling guidelines to also include testing for BRAF, HER-2, MET, RET and ROS1.

Many studies have shown that targeted therapy improves outcomes over chemotherapy for NSCLC. However, challenges remain in the genomic evaluation of the disease to identify the appropriate targeted therapy.

Gutierrez and colleagues evaluated genomic testing patterns among patients diagnosed with NSCLC in a U.S. community-based oncology practice and barriers to adherence of biomarker guidelines.

The researchers used the COTA Inc. database — which extracts and organizes demographic, diagnostic, treatment and quality data from electronic health records — to identify 814 patients with nonsquamous NSCLC (median age, 67 years) treated by 89 oncologists at clinical sites within The Regional Cancer Care Associates network of New Jersey and Maryland between January 2013 and December 2015.

Among all patients, 479 (59%) underwent EGFR and ALK biomarker testing per guideline recommendations. Only 63 patients (8%) underwent comprehensive genomic profiling for alterations in all seven genes from the NCCN guidelines.

In total, 128 patients harbored EGFR (13%) or ALK (3%) alterations, 73% of whom received first-line matched therapies. Twelve patients harbored NCCN-recommended mutations, two of whom received off-label matched therapies.

Among patients not tested for EGFR and ALK, 52% received chemotherapy without documented reasons for no testing, 32% did not undergo antineoplastic therapy and 13% had inadequate tissue for genotyping.

Twenty-two of the 335 patients (7%) who did not meet biomarker guidelines died within 30 days of diagnosis, compared with nine of the 479 (2%) who underwent required testing (P < .001)


Median OS was 31.8 months for patients who received targeted therapy, compared with 12.7 months for patients who received cytotoxic chemotherapy and 5.1 months for patients who received only supportive care (P < .001).

Gender, age, race, site of care and practice size did not impact genomic profiling frequency.

Genomic testing appeared less common among current smokers than never or former smokers (P < .01), those with stage IIIb disease than stage IV disease (P < .05), and patients who died within 30 days of diagnosis (P < .001).

Among all the patients, 53 had insufficient tissue for testing, 23 of whom underwent a second diagnostic biopsy, leading to 16 successful genomic analyses. Thus, 30 patients did not have testing after failure of the first sample, and seven underwent repeat biopsy without ever undergoing testing.

“The lack of integration of biomarker testing into routine pathology practice and uncertainty about reimbursement create additional barriers,” the researchers wrote, adding that liquid biopsies and other new technologies may be useful when there is insufficient tissue for testing. – by Melinda Stevens

Disclosure: Gutierrez reports no relevant financial disclosures. Two other researchers report employment with stock ownership in Guardant Health.

itj+ Infographic

itj+ Perspective

Geoffrey R. Oxnard

Geoffrey R. Oxnard

There is no question that tumor genotyping has transformed lung cancer care. This has given patients an opportunity to receive treatment with highly active and highly tolerable oral agents targeting driver mutations in genes like EGFR, ALK and others. But, practically, targeted therapies are only as effective as the molecular testing used to identify the targets. When tumor genotyping is completed, it creates treatment options for patients with lung cancer; when tumor genotyping is incomplete, treatment outcomes invariably suffer. In this context, the work by Gutierrez and colleagues reminds us how far we must go if we want all patients to receive tumor genotyping.

In their cohort of 814 patients, 59% underwent genotyping of EGFR and ALK, and only 8% received testing for the seven variants listed by the NCCN as potentially targetable in NSCLC. Barriers to genotyping included inadequate tissue specimens, as well as poor patient prognosis.

Fortunately, testing rates improved over the years studied, 2013 to 2016; however, more progress is still needed. One important solution could involve assays for genotyping of plasma DNA, such as the FDA-approved EGFR assay, an intuitive alternative when tissue genotyping is not possible. Liquid biopsy technologies continue to improve and, in some cases, can allow broad lung cancer genotyping without the logistical challenges of an invasive biopsy, although assay sensitivity is imperfect and tumor genotyping still the reference standard. With an ever-improving toolbox of targeted therapies — including FDA-approved regimens for four genes (EGFR, ALK, ROS1, BRAF) doctors must acknowledge that genotyping is something all patients deserve, and should be feasible for all.

Geoffrey R. Oxnard, MD
Physician at Dana-Farber Cancer InstituteAssistant professor of medicine at Harvard Medical School

Disclosure: Oxnard reports he has received consulting fees or honoraria from AstraZeneca, Ariad/Takeda, BioRad, Chugai, Ignyta, Inivata, Novartis and Sysmex.

Matthew Gubens

This study is an important and sobering reality check that we are falling short in efforts to detect actionable genetic alterations in patients with nonsquamous non-small cell lung cancer. When EGFR was the only mutation with an approved agent, the practice of profiling our patients to select whom to test was common. But, in 2017, aggressive genetic testing should never be limited to young nonsmokers. Indeed, BRAF V600E — now with an FDA-approved treatment — and MET exon 14 skipping mutations both appear to be more prevalent in current and former smokers, and are not more prevalent in the young.

Genetic testing also should not be abandoned in the face of inadequate initial specimens. We must work with our interventional colleagues to ensure larger samples and core biopsies at diagnosis. Multiplex testing should be encouraged, and funded by insurers, to allow us to be more parsimonious with our tissue, more timely with our results and more complete in our search, not just for currently actionable mutations, but also for those being evaluated in clinical trials or amenable to treatment with off-label drugs. Liquid biopsy technology should continue to be developed to complement the use of tissue biopsy.

We as a community should redouble our efforts to test every eligible patient, and this study suggests some of the barriers to this goal. In an era of precision medicine, to do less is to fail our patients, and to deprive many of treatments that would meaningfully improve length and quality of life.

Matthew Gubens, MD, MS

University of California, San Francisco

Disclosure: Gubens reports he is a consultant for AbbVie, AstraZeneca, Bristol-Myers Squibb, Genentech, Mersana and Novartis; and has received research funding to his institution from Celgene, Merck, Novartis, OncoMed and Roche.