Manuela Schmidinger , MD

Are we seeing a great change in paradigm in the first-line treatment of metastatic renal cell carcinoma? Vascular endothelial growth factor (VEGF) has been recognized as one of the major drivers of clear-cell metastatic renal cell carcinoma and, more than a decade ago, VEGF inhibitors were established as the new standard of care in metastatic renal cell carcinoma. European Society for Medical Oncology treatment guidelines recommend VEGF inhibitors in first-line treatment for patients with favorable or intermediate risk. These agents received approval based on PFS and ORR, and not on OS. OS would have confounded data — for instance, see the final OS analysis of sunitinib vs. interferon alpha by Motzer and colleagues; the CALGB 90206  trial on bevacizumab (Avastin, Genentech) plus interferon vs. interferon monotherapy by Rini and colleagues; bevacizumab with interferon alpha-2a vs. placebo and interferon alpha-2a in the AVOREN trial; and pazopanib (Votrient, Novartis) vs. placebo phase 3 data from Sternberg and colleagues — because patients could either cross over or have access to a VEGF inhibitor after treatment failure.  

Comparative phase 2 or phase 3 trials between established agents, such as sunitinib or pazopanib, had primary endpoints of noninferiority PFS or patient’s preference, but not superiority in terms of PFS or OS. So far, sunitinib has never been defeated by any other agent in a randomized phase 3 superiority trial. This suggests sunitinib is ideal and, I must say, CheckMate-214 was a very brave comparative trial. This study showed a statistically significant and clear benefit in response rates and OS with nivolumab and ipilimumab; although PFS was not statistically significant, the PFS improvement with ipilimumab and nivolumab vs. sunitinib can be regarded as clinically meaningful. Therefore, this combination indeed can be considered a new standard of care in this patient population. In addition, there were some quite remarkable aspects of this study. The response rates were among the highest ever reported in this patient population, and the complete response rate (9% vs. 1%) was the highest I’ve ever seen.

Does this imply that VEGF inhibitors are no longer the primary choice in the first-line treatment of metastatic renal cell carcinoma? What could prompt us to perceive nivolumab and ipilimumab as a new standard of care rather than the standard of care? When seeing these results, I ask myself if it is about the timing of first-line nivolumab and ipilimumab, or about nivolumab plus ipilimumab in general. Is the population addressed for the primary endpoints a clear-cut population? Are there other strategies for intermediate- and poor-risk patients that we may be tempted to use instead of nivolumab and ipilimumab? Should we always consider a VEGF receptor tyrosine kinase inhibitor in favorable-risk patients, or is nivolumab-ipilimumab an option? How relevant are the results showing the association of PD-L1 expression and outcome?

Firstly, without any doubt, the combination of nivolumab and ipilimumab is highly efficacious. In the CheckMate-214 study, patients had access to this highly efficacious combination only if assigned to it. Patients in the sunitinib arm never had access to nivolumab and ipilimumab throughout their entire course of disease. How would they have performed if they would have been given the chance to cross over after sunitinib failure? Crossover is a well-known confounder of OS, but so could be the lack of crossover. Can the OS be confounded in this trial because sunitinib patients never had access to nivolumab and ipilimumab?

There also is rationale for both prior and deferred TKI treatment. The benefits of receiving a TKI first is that several treatments, including sunitinib, modulate the tumor microenvironment by reducing regulatory T cells, improving tumor-infiltrating lymphocyte expansion and reducing the intratumoral content of myeloid-derived suppressor cells. A TKI first may also optimize outcome from immuno-oncology treatment, and this is enough to rationalize trials combining immuno-oncology and TKIs. Conversely, starting a TKI first could induce a hypoxic microenvironment with the occurrence of resistance and epithelial-to-mesenchymal transition. Antiangiogenic therapy also elicits increased local invasion and distant metastases.

Secondly, is a clear-cut population addressed for the primary endpoints? Current guidelines recommend treatment based on different populations, and this would mean favorable or intermediate vs. poor risk. The CheckMate-214 trial combined intermediate- and poor-risk patients and evaluated favorable-risk patients separately. The favorable-risk patients did better on sunitinib, whereas the intermediate-risk patients did better on nivolumab and ipilimumab. In the patients with synchronously diagnosed primary tumor and metastasis, the primary tumor may have been there undetected for years; in such a case, the disease biology could be closer to favorable than intermediate.         

Third, are there strategies for intermediate- to poor-risk patients other than nivolumab-ipilimumab? CABOSUN, a randomized phase 2 trial in intermediate- and poor-risk patients, demonstrated a PFS and ORR benefit from cabozantinib (Cometriq/Cabometyx, Exelixis) vs. sunitinib, but the trial was not designed to test for differences in OS. However, more patients had bone metastases in that trial (cabozantinib, 37%; sunitinib, 36%) than patients in this trial assigned nivolumab and ipilimumab (21%). Also, the patient population had mostly poor ECOG scores.

Fourth, should we always consider VEGF receptor TKI in favorable-risk populations? Favorable-risk patients did much better on sunitinib than nivolumab and ipilimumab. It could be a matter of PD-L1 expression — in the favorable group, only 11% had PD-L1 greater than 1%, whereas more than twice as many expressed PD-L1 in the intermediate to poor population. In an analysis by Choueiri and colleagues, PD-L1 expression was shown to be associated with shorter OS in patients undergoing TKI therapy.

That leads to our final and fifth question: Should we choose nivolumab-ipilimumab rather than sunitinib based on PD-L1 status? An exploratory endpoint of the study revealed that PD-L1-positive patients derived particular benefits from nivolumab-ipilimumab. But, considering the majority of the study patients were PD-L1-negative, should this be perceived as hypothesis generating, as the role of PD-L1 as a predictive marker remains unclear? Individual immune phenotype may identify appropriate treatment for each individual patient. Addressing the disease biology of patients by whole transcriptome profiling has been shown to identify patients who will derive benefit from sunitinib, or atezolizumab (Tecentriq, Genentech) or a combination of atezolizumab and bevacizumab (Avastin, Genentech).

In conclusion, the nivolumab and ipilimumab combination induces a high rate of objective response, and the quality of this response is highlighted by the rate of complete remissions, the duration of response and its translation into OS benefit.

This first-line treatment is a new standard of care with a potentially massive impact, but we don’t have a final picture yet. Once we can properly address the biology of a patient’s individual tumor, we may pick out the best individual treatment among various first-line options like sunitinib, nivolumab plus ipilimumab, immunotherapy plus a TKI, or an anti-VEGF monoclonal antibody. The current data are encouraging news for those conducting phase 3 trials with immunotherapy-VEGF inhibitor combination vs. sunitinib.

References:

Choueiri TK, et al. Clin Cancer Res. 2015;doi:10.1158/1078-0432.CCR-14-1993.

Escudier B, et al. J Clin Oncol. 2010;doi:10.1200/JCO.2009.26.7849.

Motzer RJ, et al. J Clin Oncol. 2009; doi:10.1200/JCO.2008.20.1293.

Rini BI, et al. J Clin Oncol. 2010;doi:10.1200/JCO.2009.26.5561.

Sternberg CN, et al. Eur J Cancer. 2013;doi:10.1016/j.ejca.2012.12.010.