Implications of Next-Generation Sequencing for Kidney Cancer Treatment Selection

Dec 14, 2015


By Shaheen Alanee, MD, MPH, MBA, AFACS
Simmons Cancer Institute, Southern Illinois University School of Medicine

Renal cell carcinoma (RCC) is the eighth most common cancer in the United States, affecting approximately 65,000 patients every year. In addition, metastatic RCC (mRCC) is known to be a highly lethal tumor, leading to an estimated 13,000 patient deaths annually.

If RCC can be detected in early stages, it is highly curable by surgical removal, but there is no curative treatment for advanced-stage disease, and the 2-year survival rate of patients with mRCC is less than 20%. This high mortality is related to this disease being resistant to radiotherapy and chemotherapy, and having a minimal response to immunotherapy. Therefore, further investigation of the genomic alterations in mRCC could be essential for optimizing treatment.

One powerful tool for genomic characterization of cancer is next-generation sequencing (NGS). As such, there is an increasing interest in the role of NGS in improving the management of mRCC. In this issue of ASCO Connection, Dr. Sumanta K. Pal and Dr. Nicholas J. Vogelzang discuss, in a concise and enriching manner, the implications of NGS in selecting effective therapy for mRCC.

Dr. Alanee is a surgical oncologist at Simmons Cancer Institute and Director of Urologic Oncology at Southern Illinois University. He has been an ASCO member since 2013.

Next-Generation Sequencing for Renal Cell Carcinoma: Ready for Prime Time?

By Sumanta K. Pal, MD
City of Hope Comprehensive Cancer Center

Nicholas J. Vogelzang, MD
US Oncology Research Comprehensive Cancer Centers

A decade ago, the U.S. Food and Drug Administration (FDA) approval of targeted therapies caused a marked paradigm shift in the management of metastatic renal cell carcinoma (mRCC). The dismal survival estimates associated with conventional immunotherapy— approximately 1 year with interferon-α—were supplanted by estimates double to triple of that using sequential therapy with VEGF and mTOR inhibitors.1 What ensued was a series of drug approvals centered on agents with a similar mechanism of action. Today, a total of four VEGF inhibitors and two mTOR inhibitors are used in the clinic.2 However, these drugs are used in exclusivity and are not compared to alternative treatments. This redundancy has led to a plateau in clinical outcomes for patients with mRCC.

This plateau will hopefully be short-lived. A press release from a phase III trial comparing the PD-1 inhibitor nivolumab to the mTOR inhibitor everolimus indicated improved overall survival (OS) in the nivolumab arm.3 A second press release from a phase III trial comparing the dual VEGF receptor 2 (VEGFR2)/MET inhibitor cabozantinib to everolimus showed improved progression-free survival (PFS) with cabozantinib, with a trend towards improved OS.4 With both studies achieving their primary endpoint, FDA approval can be anticipated in the coming months. Envisioning a landscape in which 10 FDA-approved agents exist for mRCC, the investigative community will have to devote a great deal of attention to therapeutic sequencing. Market data indicate that a meager 20% of patients receive third-line treatment for mRCC, implying that the vast majority of agents will go unused.5 Furthermore, the sole randomized phase III trial in mRCC (dovitinib vs. sorafenib) demonstrated a median survival in both groups of 11 months, implying limited opportunities for third- or fourth-line treatments.6

How might one identify the optimal therapy for an individual patient? The answer more than likely lies in tailoring treatment to disease biology—and it is next-generation sequencing (NGS) tools that have enhanced the feasibility of this approach.

NGS has been applied in several recently completed trials in mRCC, perhaps most notably the randomized, phase II RECORD-3 trial. The study randomly assigned 471 patients to receive either the VEGF inhibitor sunitinib or the mTOR inhibitor everolimus in the first-line setting, followed by a switch to the opposing therapy at the time of progression.7 The study failed to meet its primary endpoint of improved PFS with first-line everolimus; in fact, PFS was modestly prolonged with sunitinib therapy (10.7 vs. 7.9 months; hazard ratio [HR] 1.4, 95% CI [1.2, 1.8]). However, the study proved to be a veritable gold mine for correlatives, with archival tissue available in 261 patients. At the 2015 ASCO Annual Meeting, James Hsieh, MD, PhD, of Memorial Sloan Kettering Cancer Center, presented NGS data derived from these tissues. Mutations in two genes involved in epigenetic regulation, KDM5C and PBRM1, predicted improved clinical outcome with sunitinib and everolimus, respectively. Although prospective validation is necessary, these could represent the first reliable biomarkers to distinguish activity of VEGF and mTOR inhibitors in mRCC. Other genes, such as BAP1, appear to have substantial prognostic value across multiple series, but the predictive role of such mutations has yet to be established.8

Of course, the decision between a VEGF and an mTOR inhibitor may become less relevant when agents such as nivolumab and cabozantinib enter the clinic. Profiling of MET alterations using NGS could theoretically identify patients who are more sensitive to cabozantinib; such data have been produced in the context of other MET inhibitors.9 For nivolumab, a higher overall mutational burden (rather than the presence of specific mutations) might predict responsiveness. This has been elegantly demonstrated with the PD-1 inhibitor pembrolizumab in lung cancer in the presence of mismatched repair deficient tumors and with the cytotoxic T-cell lymphocyte antigen-4 (CTLA4) inhibitor ipilimumab in melanoma.10-12

Beyond therapeutic selection for clear cell mRCC (for which most of the aforementioned therapies are applicable), NGS could uncover novel targets in non-clear cell histologies. Efforts to define the mutational landscape in papillary, chromophobe, and collecting duct RCC have revealed distinct and potentially actionable mutations. For example, in a series of 17 patients with collecting duct RCC, a high frequency of NF2 and CDKN2A mutations were identified, possibly implying sensitivity to everolimus and palbociclib, respectively.13 Prospective studies based on these findings are currently being planned. Such studies offer an opportunity to change the dismal prognosis for metastatic collecting duct RCC, typically estimated at less than 1 year with cytotoxic chemotherapy.

Through new applications of NGS there may also be an opportunity to better understand and treat toxicities related to existing therapies. Bacteriomic profiling (typically involving sequencing of bacterial 16S rRNA) has found applications in gastrointestinal disorders, such as Crohn’s disease and short gut syndrome. In one of the first applications of bacteriomic profiling in oncology, the stool bacteriome in patients receiving VEGF-directed therapies for mRCC was characterized.14 Specific bacteria (Bacteroides spp and Prevotella spp) appear to modulate the risk of treatment-related diarrhea. As this field matures, NGS of the gut flora may guide use of probiotics or similar interventions to decrease the risk of gastrointestinal toxicity.

In summary, with multiple new therapies soon to enter the clinic for mRCC, NGS may serve a critical role in defining treatment allocation. Although prospective studies would be helpful in validating this approach, currently available data—for example, data on biomarkers that define VEGF and mTOR inhibitor activity in RECORD-3—are compelling enough to guide therapy today. Use of NGS in rare histologies of RCC should also be pursued as a means of defining unique therapeutic targets. Finally, although it currently sits in the experimental realm, novel NGS techniques such as bacteriomic profiling could be applied to characterize and treat selected toxicities.

While many cite the cost of NGS as being prohibitive to widespread clinical application, it is important to keep in mind that the cost of therapeutics far exceeds that of diagnostics. Unguided use of targeted agents for mRCC will add to the economic burden of cancer care by permitting use of ineffective therapies. Using NGS to identify the right treatment for the right individual is a concept that should be attractive to both patients and payers.

Dr. Pal is an assistant professor in the Department of Medical Oncology & Experimental Therapeutics at City of Hope Comprehensive Cancer Center. An ASCO member since 2007, he has served on the Cancer Communications Committee, the Social Media Working Group, and the Scientific Program Committee.

Dr. Vogelzang is a medical oncologist at US Oncology Research Comprehensive Cancer Centers. An ASCO member since 1980, he has served on the Cancer Education Committee, the Cancer Communications Committee (Past Chair), the Prostate Cancer Advisory Panel, and the Genitourinary Cancers Symposium News Planning Team.

1. Motzer RJ, Bacik J, Murphy BA, et al. J Clin Oncol. 2002;20:289-96.
2. Pal SK, Williams S, Josephson DY, et al. Mol Cancer Ther. 2012;11:526-37.
3. Bristol-Myers Squibb. Published July 20, 2015.
4. Exelixis. Published September 25, 2015.
5. Pal SK, Malangone E, Bhurke S, et al. J Clin Oncol. 2013;31 (suppl 6; abstr 390).
6. Motzer RJ, Porta C, Vogelzang NJ, et al. Lancet Oncol. 2014;15:286-96.
7. Hsieh J, Chen D, Wang P, et al. J Clin Oncol. 2015;33 (suppl; abstr 4509).
8. Joseph RW, Kapur P, Serie DJ, et al. Cancer. 2014;120:1059-67.
9. Choueiri TK, Vaishampayan U, Rosenberg JE, et al. J Clin Oncol. 2013;31:181-6.
10. Snyder A, Makarov V, Merghoub T, et al. N Engl J Med. 2014;371:2189-99.
11. Rizvi NA, Hellmann MD, Snyder A, et al. Science. 2015;348:124-8.
12. Le DT, Uram JN, Wang H, et al. N Engl J Med. 2015;372:2509-20.
13. Pal SK, Choueiri TK, Wang K, et al. Eur Urol. In press.
14. Pal SK, Li SM, Wu X, et al. Clin Cancer Res. Epub 2015 July 7.


Back to Top