Printer Friendly

Next-Generation Sequencing and Immunotherapy Biomarkers: A Medical Oncology Perspective.

It would not be an act of an optimistic oncologist's hyperbole to state that the explosion in genomic information and the development of immunotherapy agents have revolutionized the care of cancer patients, with more advances undoubtedly to continue to be realized in the next few years. However, even with these significant accomplishments, much work needs to be done to further refine our knowledge and translate scientific discovery of driver mutations and host-tumor immune interactions into lasting benefits for our patients. In this review, I will discuss some of the issues facing clinicians as they attempt to use these scientific advances at the bedside.

Since the publication in 2004 of findings showing that activating mutations in the epidermal growth factor receptor (EGFR) correlated with clinical responsiveness to a small molecule tyrosine kinase inhibitor, (1) it has been hoped that the discovery of a tumor's driver mutations would in many cases lead to the rational development of targeted agents that would enhance response and decrease toxicity when compared with standard indiscriminate cytotoxic agents. And, in fact, many randomized trials have borne out the validity of that strategy in both EGFR-mutated and anaplastic lymphoma kinase (ALK) translocated adenocarcinomas of the lung. (2-5) This in turn has led expert panels to recommend that all new diagnoses of advanced adenocarcinoma of the lung be tested for EGFR and ALK. (6)

The search for druggable genomic lesions has led to a number of clinical trials where genomic analyses were fused with the use of various targeted therapies. These broader trials have led to a number of mixed results, highlighting the difficulties and possible benefits of expanding precision medicine to the care of patients with advanced lung cancer. One of the early trials was MD Anderson Cancer Center's Biomarker-integrated Approaches of Targeted Therapy for Lung Cancer Elimination (BATTLE) trial, in which patients with previously treated non-small cell lung cancer (NSCLC) were evaluated for a number of biomarkers and then randomized to 1 of 4 treatment arms of small molecule-targeted therapies. (7) Patients were allowed into the trial even if they had a number of prior chemotherapy regimens, as long as they had a good performance status. They agreed to be rebiopsied in order to get fresh tissue upon which to perform biomarker analysis. In the initial trial, samples were tested for a number of markers, such as EGFR, KRAS, vascular endothelial growth factor, and retinoid x receptor and cyclin D1 proteins.

The study proved that obtaining tissue in previously treated lung cancer patients and assaying them for biomarker discovery was feasible and in some cases led to improved clinical results. However, the study also emphasized that biomarkers only help if available therapeutics are effective for a particular genomic group. A number of the arms contained therapies that still remain experimental and have not been approved for general clinical use (vandetanib and sorafenib lack US Food and Drug Administration approval for use in NSCLC). Contrast those results with the identification of ROS1 as a driver mutation and the demonstration of the remarkable efficacy of crizotinib for treatment of advanced-stage disease. (8) Although ROS1 mutations represent only 1% of patients with lung cancer, considering the number of patients receiving a diagnosis of that disease, there is a significant clinical benefit in identifying patients who are candidates for crizotinib therapy. The response rate in metastatic disease was 72% and the median duration of response was an amazing 17.6 months.

The point of the contrast is not to lessen the results of the BATTLE trial; rather, it is to demonstrate that the worth of biomarker testing outside of exploratory clinical trials is closely tethered to the efficacy of available agents. KRAS would be a case in point; we have been able to test for KRAS mutations in lung cancer for a long time, but the negligible progress to date in developing effective therapeutic strategies has limited the use of that biomarker. (9) KRAS has little utility as a prognostic biomarker; the strategy of using it in a testing algorithm (because it is often mutually exclusive with certain other driver mutations) has fallen as testing for EGFR, ALK, and ROS1 have become more widespread and less expensive, and we do not yet have the therapeutic tools to capitalize on the demonstration of KRAS mutation in patients with lung cancer.

Recently, Lopez-Chavez et al (10) reported the results of another basket trial (the CUSTOM or Molecular Profiling and Targeted Therapies in Advanced Thoracic Malignancies trial) in patients with advanced thoracic malignancies. The trial enrolled 647 patients with recurrent or advanced thoracic tumors, including NSCLC, small cell lung cancer, and thymic cancers. The patients had genomic testing and then were randomized to either therapy with standard of care or targeted therapies selected on the basis of the driver evaluation: erlotinib for EGFR mutations; selumetinib for KRAS, NRAS, HRAS, or BRAF mutations; MK2206 for PIK3CA, AKT, or PTEN mutations; lapatinib for ERBB2 mutations or amplifications; and sunitinib for KIT or PDGFRA mutations or amplifications.

The frequencies of mutations of EGFR and KRAS were in line with what has been reported before. Not surprisingly, EGFR-mutated adenocarcinoma patients had the best survival rates. Many of the other arms, however, had small numbers of patients, and some arms were felt to not be feasible to complete accrual. The story for KRAS was consistent with prior data, in that KRAS was predominantly found in adenocarcinoma patients with a tobacco exposure history and the responses to selumetinib were poor, with an overall response rate of 11% (1 partial response out of 11 patients). However, the total number of patients in individual baskets was low, suggesting that future basket trials have to be performed in a multi-institutional setting with a large number of patients accrued in order to make sure that the statistical power to detect a signal of response is present.

The recalcitrance of KRAS-driven lung cancers to the development of effective therapies is fortunately not the only story emerging from more detailed genomic analyses. BRAF- and RET-mutated lung cancers are found at a rate of approximately 1% each and have available targeted therapies that are showing what appear to be meaningful responses in clinical trials. Recently, Planchard et al11 reported on 24 patients with BRAF-mutated adenocarcinoma who had prior chemotherapy and were then treated with dual therapy with dabrafenib and trametinib. The overall response rate was 63% (15 of 24) and the disease control rate at 12 weeks was 88%. Obviously, although further data will need to be collected, BRAF-mutated lung cancer seems to be quite responsive to targeted therapy. Similarly, RET has emerged as a potential target as well. Drilon and colleagues (12) reported on a single-institution trial of cabozantinib in patients with advanced RET-rearranged lung cancers. Like many driver-mutated lung cancers, this population tends to have adenocarcinoma histology and be mostly female light to never smokers. A total of 20 patients were treated; the disease control rate was 72% (13 of 18) and the median progression-free survival was 7 months.

The emergence of recognizable genomic subgroups of advanced NSCLC has led the National Comprehensive Cancer Network to list suggested markers that are reasonable to test in patients. The guidelines now list emerging targeted agents for patients with particular genetic mutations: crizotinib for ROS1-and MET-amplified tumors, afatinib and trastuzumab for HER2-mutated lung cancers, cabozantinib for RET-arranged tumors, and vemurafenib and dabrafenib for BRAF lung cancers. In reality, if this testing is done at all it will most likely be in patients who have failed initial therapy with chemotherapy, or failed EGFR- and ALK-targeted therapy, because those are the mutations that are routinely looked for currently in clinical practice when advanced disease is diagnosed. Many patients with lung cancer in this country do not get their care at major tertiary cancer centers where panels have been or are being developed. However, with the availability of companies that can perform next-generation sequencing (NGS) on archival tissue, more patients should have the opportunity to have their tumors analyzed and have treatments offered on the basis of which driver mutations are present.

So should patients with advanced adenocarcinoma of the lung have NGS panel studies performed upon diagnosis? Very few trials have looked at upfront genomic analysis in a comprehensive fashion. Many trials, such as BATTLE or CUSTOM, have studied second-line patients. There is also the issue that many patients in trials at tertiary cancer centers are a self-selected population--they tend to be wealthier, they tend to be light smokers to nonsmokers, and there are fewer minorities. This is the population that is a priori expected to have a higher incidence of driver mutations: there is a significant need to look at the results of NGS in minority patients and patients with a more extensive tobacco history.

Drilon et al (13) recently presented their experience with broad hybrid-capture NGS in patients with adenocarcinoma seen at Memorial Sloan Kettering Cancer Center. They enrolled lung adenocarcinoma patients with a [less than or equal to] 15 pack-year smoking history whose tumors previously tested "negative" for alterations in 11 genes (mutations in EGFR, ERBB2, KRAS, NRAS, BRAF, MAP2K1, PIK3CA, and AKT1, and fusions involving ALK, ROS1, and RET). They were able to find actionable mutations in 26% of patients who had negative test results with other means. Most of the abnormalities were in EGFR, ALK, and RET. The authors postulated a number of possible reasons why NGS was picking up actionable driver mutations missed on initial testing, such as tumor heterogeneity or more complex rearrangements. In addition, many patients required repeat biopsies--sometimes multiple ones--to acquire enough tissue to run NGS. It is not clear whether repeated biopsies, especially when multiple, are feasible outside of tertiary cancer center hospitals, and they add expense and potential patient morbidity. Nevertheless, the findings add data to the growing feeling that upfront NGS testing would capture more patients with driver mutations that are druggable and would make better use of limited tissue, a perennial problem with lung cancer patients in particular.

Does NGS testing open up different therapies that would not have been chosen? Kris et al (14) recently reported on the experience of the Lung Cancer Mutation Consortium (LCMC) with testing 1000 patients with adenocarcinoma; data were available for the first 617 patients. (14) Patients had to have adenocarcinoma, stage IIIB or IV, have a performance status of 0 to 2, and have available tissue on which to run tests. A total of 52% of patients were found to have an oncogenic driver, although the most common alterations (KRAS in 22% and MET expression in 54%) still pose formidable therapeutic challenges. A total of 16% of patients were able to be treated with a targeted therapy or go on to a trial. The experience of the LCMC clearly demonstrates that it is possible to acquire timely genomic information that can help guide patients with adenocarcinoma into particular, rational therapeutic pathways. Still, most patients with NSCLC (comprising squamous cell carcinoma, large cell carcinoma, and adenocarcinomas without druggable mutations) will still need conventional cytotoxics and/or trials of immunotherapeutic approaches, despite our growing ability to acquire the genomic information internal to each patient's unique malignancy.

Looking ahead, will NGS testing of lung cancers become the standard of care? A number of obstacles remain, including the fact that, as mentioned before, many of the trials have not necessarily looked at testing upon initial diagnosis in treatment-naive patients. Turnaround time remains a significant issue because many patients and families can feel uncomfortable about delaying therapy for 2 weeks or more while waiting for the results to become available. Lastly, cost-effectiveness analysis of genomic testing will eventually need to be performed given the climate in which we practice. Clearly, better therapeutics against KRAS-mutated lung cancer, which have been a significant clinical goal for so long, would go a long way in justifying NGS testing upfront because most patients currently with actionable driver mutations (EGFR and ALK) can be detected without resorting to more extensive NGS sequencing.

Clearly, finding actionable mutations in patients with advanced squamous cell lung cancer remains a serious unmet need. Unlike many of the advances seen with adenocarcinomas--US Food and Drug Administration-approved targeted therapies for actionable mutations, demonstration of the efficacy of maintenance therapy, and the ability to use bevacizumab with cytotoxic chemotherapy--squamous cell carcinoma has therapeutically lagged behind. The Southwest Oncology Group's Lung Cancer Master Protocol (Lung-MAP) trial is attempting to change that by using an adaptive basket trial designed to look at particular mutations in advanced squamous cell carcinomas (PI3K, FGFR, CDK4/CDK6, HGF, and PD-L1). (15) This trial, unlike the aforementioned CUSTOM trial, uses a design whereby multiple simultaneously running phase 2/3 trials can independently open or close without affecting the other arms of the study. Second-line patients will be tested with NGS and then randomized into a substudy of a targeted agent versus standard of care. It is hoped that this trial will open new therapeutic options for patients with metastatic squamous cell carcinomas by demonstrating the efficacy of targeted therapies in molecular subgroups.

It would not be too much of an overstatement to point out that the development of immunotherapy to treat a number of patients with advanced solid tumor is part of a significant paradigm shift in the treatment of cancer patients. Since the recognition that we can unlock the immune system and effect clinical benefit for patients rather than attack cancer cells with cytotoxic agents or tumor mutations with targeted therapies, there has been an explosion of clinical interest in pursuing this approach. (16,17) Still, many patients do not benefit from immune checkpoint inhibitors and there is a concerted effort to look for biomarkers that would help guide patient selection or direct patients on to clinical trials to look at treatments that would enhance immune checkpoint efficacy.

The excitement for immune checkpoint inhibitors is not difficult to understand. (18) Overall, these agents are well tolerated and have a different pattern of toxicities than cytotoxics. They can cause fatigue and some autoimmune issues, and there are rare reports of pneumonitis. Still, many patients tolerate them well and can have a fairly good quality of life while on these agents. But the appeal is not just the tolerability; there is a subset of patients treated with these agents who have dramatic and prolonged response to these agents, which often lasts many years.

Benefit has been shown in patients with advanced NSCLC who have failed therapy with cisplatin-based chemotherapy regimens. This clinical benefit has been demonstrated recently in two randomized trials in both squamous cell and adenocarcinoma histologies, both using docetaxel as the comparison arm. (19,20) For adenocarcinomas, nivolumab provided a median overall survival of 12.2 months versus 9.4 months; the 1-year survival was 50.5% versus 39%. For patients with squamous cell histology, nivolumab also had a superior 1-year survival rate (42% versus 24%), and progression-free survival at 1 year strongly favored nivolumab (21% versus 6%). As has been seen in a number of other immune checkpoint trials, responding patients often had durable benefit from therapy.

As exciting as the data just reviewed are, however, and despite the fact that immunotherapy is clearly superior to second-line cytotoxic therapy in patients with metastatic disease, many patients do not benefit from either therapy. There is considerable interest in developing biomarkers to help guide clinicians in determining who can benefit from immunotherapy and who, unlikely to benefit, can go onto clinical trials. However, to date the discovery of a reliable biomarker remains elusive. There remains little consensus on the threshold level of PD-L1 expression, and to further complicate matters, some companies are developing their own proprietary tests. Some reports have found that PD-L1 expression on tumor cells is predictive of response, whereas others have suggested that PD-L1 expression on infiltrating immune cells correlates with response. (21) Different studies are reporting testing with different antibodies and different scales of expression of the marker, clouding a clear picture of utility. In a recent study in treatment-naive patients with metastatic lung cancer with the PD-1 inhibitor pembrolizumab, reporting PD-L1 levels as measured by the 22C3 antibody, patients with PD-L1 expression of greater than 50% on their tumors had the best overall response rate. (22) However, in the aforementioned nivolumab trial in patients with squamous cell cancer, there was not a correlation between PD-L1 expression and response. Whether these discordant results are because the latter trial enrolled patients previously treated with cytotoxic therapy that might have altered PD-L1 expression remains an active issue of clinical investigation. As of now, there are no data that would suggest that medical oncologists should be ordering PD-L1 testing on their patients' tumor samples in order to decide on using an immune checkpoint inhibitor. Ultimately, expert panels of clinicians and pathologists will need to weigh in on guidelines as further data clarify the utility of testing.

Beyond looking at PD-L1 expression, there has been significant interest in whether the mutational landscape affects patients' response to immunotherapy, and a number of recent trials have suggested that there is a connection. Snyder et al (23) looked at nonsynonymous mutations in patients with metastatic melanoma treated with the CTLA-4-blocking drug ipilimumab and found a strong correlation with response. Similarly, Rizvi et al (24) recently published data looking at mutation rate and response to PD-L1-blocking antibodies in advanced lung cancer and found a similar picture. For patients whose tumors exhibited a high nonsynonymous tumor burden and some degree of tumor expression of PD-L1, the durable clinical benefit rate was 91%. In patients with some degree of PD-L1 staining but low mutation burden, the durable clinical benefit rate was much lower: only 10%. The number of patients looked at was small, and thus this needs to be looked it in a larger number of patients, but it is possible that a treatment algorithm will be developed that will look at a combination of PD-L1 expression, mutational landscape, and histologic factors (such as immune cell infiltrate patterns).

Similarly to effective biomarkers for specific mutations, the other clinical issue that will determine the utility of immunotherapy biomarkers will be the efficacy of alternative therapies. If the choice is between an immune checkpoint inhibitor or a cytotoxic drug for patients who have already failed frontline chemotherapy and if the anticipated response rate is around 10% to 15% (ie, roughly equivalent), then patients and their physicians might choose to proceed with immunotherapy, given its favorable toxicity profile. In that case, a biomarker will not be needed because it will not make much of an impact on clinical decision making. More than likely, these biomarkers will eventually have their place in deciding between upfront therapy: high mutational load/high PD-L1 expressors could conceivably go on to frontline immunotherapy, and low expressors without targetable driver mutations could go on to chemotherapy or go on to clinical trials to test strategies to improve tumor immune responsiveness.

Given the high cost of these agents and the fact that they are active in common malignancies, it is imperative that trials be designed in order to maximize the benefit to patients who have advanced disease, but hopefully limiting exposure to patients unlikely to benefit. Although it is hoped that the advent of computed tomography screening for lung cancer will allow us to catch earlier-stage disease, metastatic NSCLC will not disappear anytime soon. Thus, it is imperative that we imaginatively design trials to capture the best genomic and immunologic knowledge and learn how that will guide us to make the most out of available therapies. With a collaborative effort from clinicians, pathologists, and basic scientists working together to acquire meaningful study data, we should be entering an age where hyperbole becomes the expected reality.


(1.) Lynch TJ, Bell DW, Sordella R, et al. Activating mutations in the epidermal growth factor receptor underlying responsiveness of non-small-cell lung cancer to gefitinib. N Engl J Med. 2004; 350(21):2129-2139.

(2.) Mok TS, Wu YL, Thongprasert S, et al. Gefitinib or carboplatin-paclitaxel in pulmonary adenocarcinoma. N Engl J Med. 2009; 361(10):947-957.

(3.) Wu YL, Zhou C, Hu CP, et al. Afatinib versus cisplatin plus gemcitabine for first-line treatment of Asian patients with advanced non-small-cell lung cancer harbouring EGFR mutations (LUX-Lung 6): an open-label, randomised phase 3 trial. Lancet Oncol. 2014; 15(2):213-222.

(4.) Sequist LV, Yang JC, Yamamoto N, et al. Phase III study of afatinib or cisplatin plus pemetrexed in patients with metastatic lung adenocarcinoma with EGFR mutations. J Clin Oncol. 2013; 31(27):3327-3334.

(5.) Shaw AT, Kim DW, Nakagawa K, et al. Crizotinib versus chemotherapy in advanced ALK-positive lung cancer. N Engl J Med. 2013; 368(25):2385-2394.

(6.) Lindeman NI, Cagle PT, Beasley MB, et al. Molecular testing guideline for selection of lung cancer patients for EGFR and ALK tyrosine kinase inhibitors: guideline from the College of American Pathologists, International Association for the Study of Lung Cancer, and Association for Molecular Pathology. Arch Pathol Lab Med. 2013; 137(6):828-860.

(7.) Kim ES, Herbst RS, Witsuba II, et al. The BATTLE trial: personalizing therapy for lung cancer. Cancer Discov. 2011; 1(1):44-53.

(8.) Shaw AT, Ou SI, Bang YJ. Crizotinib in ROSl-rearranged non-small-cell lung cancer. N Engl J Med. 2014; 371(21):1963-1971.

(9.) Roberts PJ, Stinchcombe TE. KRAS mutation: should we test for it, and does it matter? J Clin Oncol. 2013; 31(8):1112-1121.

(10.) Lopez-Chavez A, Thomas A, Rajan A, et al. Molecular profiling and targeted therapy for advanced thoracic malignancies: a biomarker-derived, multiarm, multihistology phase II basket trial. J Clin Oncol. 2015; 33(9):1000-1007.

(11.) Planchard D, Groen HJM, Kim TM, et al. Interim results of a phase II study of the BRAF inhibitor (BRAFi) dabrafenib (D) in combination with the MEK inhibitor trametinib (T) in patients (pts) with BRAF V600E mutated (mut) metastatic non-small cell lung cancer (NSCLC). Paper presented at: 51st Annual Meeting of the American Society of Clinical Oncology; May 29-June 2, 2015; Chicago, IL. Abstract 8006.

(12.) Drilon AE, Sima CS, Somwar R, et al. Phase II study of cabozantinib for patients with advanced RET-rearranged lung cancers. Paper presented at: 51st Annual Meeting of the American Society of Clinical Oncology; May 29-June 2, 2015; Chicago, IL. Abstract 8007.

(13.) Drilon A, Wang L, Arcila ME, et al. Broad, hybrid capture-based next-generation sequencing identifies actionable genomic alterations in lung adenocarcinomas otherwise negative for such alterations by other genomic testing approaches. Clin Cancer Res. 2015; 21(16):3631-3639.

(14.) Kris MG, Johnson BE, Kwiatkowski DJ, et al. Migration to next-generation sequencing and the identification of RET and ROS1 rearrangements plus PTEN and MET protein expression in tumor specimens from patients with lung adenocarcinomas. Paper presented at: 51st Annual Meeting of the American Society of Clinical Oncology; May 29-June 2, 2015; Chicago, IL. Abstract 8094.

(15.) Gandara DR, Hammerman PS, Sos ML, et al. Squamous cell lung cancer: from tumor genomics to cancer therapeutics. Clin Cancer Res. 2015; 21(10): 2236-2243.

(16.) Brahmer JR, Tykodi SS, Chow LQM, et al. Safety and activity of anti-PD-L1 antibody in patients with advanced cancer. N Engl J Med. 2012; 366(26):2455-2465.

(17.) Topalian SL, Hodi FS, Brahmer JR, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. N Engl J Med. 2012; 366(26):2443-2454.

(18.) Postow MA, Callaghan MK, Wolchok JD. Immune checkpoint blockade in cancer therapy. J Clin Oncol. 2015; 33(17):1974-1982.

(19.) Brahmer J, Reckamp KL, Baas P, et al. Nivolumab versus docetaxel in advanced squamous-cell non-small-cell lung cancer. N Engl J Med. 2015; 373(2): 123-135.

(20.) Paz-Ares L, Horn L, Borghaei H, et al. Phase III, randomized trial (CheckMate 057) of nivolumab (NIVO) versus docetaxel (DOC) in advanced non-squamous cell (non-SQ) non-small cell lung cancer (NSCLS). Paper presented at: 51st Annual Meeting of the American Society of Clinical Oncology; May 29-June 2, 2015; Chicago, IL. Late-breaking Abstract 109.

(21.) Herbst RS, Soria JC, Kowanetz M, et al. Predictive correlates of response to the anti-PD-L1 antibody MPDL3280A in cancer patients. Nature. 2014; 515(7528):563-567.

(22.) Rizvi NA, Garon EB, Leighl N, et al. Optimizing PD-L1 as a biomarker of response with pembrolizumab (pembro; MK-3475) as first-line therapy for PDL1-positive metastatic non-small cell lung cancer (NSCLC): updated data from KEYNOTE-001. Paper presented at: 51st Annual Meeting of the American Society of Clinical Oncology; May 29-June 2, 2015; Chicago, IL. Abstract 8026.

(23.) Snyder A, MakarovV, MerghoubT, et al. Genetic basis for clinical response to CTLA-4 blockade in melanoma. N Engl J Med. 2014; 371(23):2189-2199.

(24.) Rizvi NA, Hellman MD, Snyder A, et al. Mutational landscape determines sensitivity to PD-1 blockade in non-small cell lung cancer. Science. 2015; 348(6230):124-128.

Eric Bernicker, MD

Accepted for publication August 4, 2015.

From the Cancer Center, Houston Methodist Hospital, Houston, Texas.

The author has no relevant financial interest in the products or companies described in this article.

Presented at the Biennial Meeting of the Pulmonary Pathology Society; June 3-5, 2015; San Francisco, California.

Reprints: Eric Bernicker, MD, Cancer Center, Houston Methodist Hospital, 6445 Main Street, Floor 24, Houston, TX 77030 (email:
COPYRIGHT 2016 College of American Pathologists
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2016 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Bernicker, Eric
Publication:Archives of Pathology & Laboratory Medicine
Date:Mar 1, 2016
Previous Article:Pulmonary Langerhans Cell Histiocytosis: An Update From the Pathologists' Perspective.
Next Article:Update on Pulmonary Fibrosis: Not All Fibrosis Is Created Equally.

Terms of use | Privacy policy | Copyright © 2019 Farlex, Inc. | Feedback | For webmasters