The optimal frontline treatment strategy for patients with metastatic renal cell carcinoma could look dramatically different in the next few years, as studies assess combination strategies and predictive biomarkers for immunotherapy and targeted therapies.
Sumanta Kumar Pal, MD
The optimal frontline treatment strategy for patients with metastatic renal cell carcinoma (RCC) could look dramatically different in the next few years, as studies assess combination strategies and predictive biomarkers for immunotherapy and targeted therapies, according to Sumanta Kumar Pal, MD.
For the past decade, VEGF-targeted therapies were considered the optimal frontline treatment for patients with metastatic clear cell RCC, Pal said during a presentation at the 2016 GU Cancers Symposium. However, this paradigm could be disrupted by the recent wave of immuno-oncology agents, which are gaining approval across indications for patients with cancer.
“There's a lot of hope here for perhaps the best of both worlds, with higher response rates and longer progression-free survival with VEGF therapy combined with more durable responses with the PD-1 and PD-L1 directed agents,” said Pal, co-director of the Kidney Cancer Program at City of Hope in Duarte, California.
In November 2015, the first immune checkpoint inhibitor, nivolumab (Opdivo), was approved as a second-line therapy for advanced RCC. This decision was based on the CheckMate-025 study, which showed a median overall survival (OS) of 25.0 months with nivolumab compared with 19.6 months for everolimus (HR, 0.73; P = .002).1
“There is a subset of patients who seem to derive an exceptional response. I think we need to find out who these exceptional responders are. There has been an emerging body of evidence to characterize these individuals,” said Pal. “I don't think nivolumab in the frontline setting is appropriate at this point.”
Anecdotal findings suggest that neoantigens are associated with better responses to immunotherapy, providing a potential tool for tailoring therapy. A correlation between mutational load and responses has also been observed. However, both of these observations still need to be vetted in larger trials, explained Pal.
The need for an effective biomarker also exists for VEGF therapies. A number of studies have hinted at markers of response with various agents, although one indicator has not yet conclusively emerged. Currently, early data suggest KDM5C alterations could be associated with long and durable responses to VEGF-targeted therapy, although these findings still need to be validated.The leading frontline VEGF tyrosine kinase inhibitors (TKIs) currently used for RCC are sunitinib (Sutent) and pazopanib (Votrient). These therapies were compared in the phase III 1110-patient COMPARZ trial, which demonstrated similarity between the two agents.2 In the study, pazopanib was shown to be noninferior to sunitinib for progression-free survival (PFS; HR, 1.05; 95% CI, 0.90-1.22) and OS (HR, 0.91; 95% CI, 0.76-1.08).
In addition to established therapies, new agents have recently entered the RCC arena. Although not yet approved, the multikinase inhibitor cabozantinib (Cometriq) demonstrated superiority to everolimus in the phase III METEOR study.3 A rolling submission of data from this trial was completed in December 2015, with a decision from the FDA expected within the next 6 to 8 months.
After a minimum of 11 months of follow-up in the METEOR study, median PFS with cabozantinib was 7.4 months compared with 3.8 months with everolimus (HR, 0.58; 95% CI, 0.45-0.75; P <.001). At the interim analysis, a trend toward improvement in OS was observed; however, this did not yet pass a high bar for statistical significance (HR, 0.67; 95% CI, 0.51-0.89; P = .005).
The ongoing phase II CABOSUN trial is comparing frontline cabozantinib with sunitinib for patients with clear cell RCC. The study enrolled 150 participants to assess the dual primary endpoints of OS and PFS. Early data from this study are anticipated later this year.Early-phase clinical trials exploring VEGF TKIs in combination with anti—PD-1 agents have shown prohibitive hepatic and gastrointestinal toxicity, explained Pal. These studies looked at nivolumab with sunitinib or pazopanib or at the PD-1 inhibitor pembrolizumab (Keytruda) with pazopanib.
Despite these setbacks, promising data were seen for the combination of the VEGF TKI axitinib (Inlyta) and pembrolizumab in a small, 11 patient study.4 In this trial, the most common grade 3/4 adverse events were hypertension (27.3%), and diarrhea and increased alanine transaminase (9.1% each).
In addition to a manageable toxicity profile, axitinib plus pembrolizumab demonstrated sustained and prolonged responses. Overall, 6 patients had confirmed partial responses and 5 patients had stable disease with tumor shrinkage.
“This bodes well for a clinical trial that is currently under way exploring axitinib and avelumab,” said Pal, referring to an emerging PD-L1 inhibitor. “This particular trial has already inspired a phase III trial that will randomized patients to either sunitinib or axitinib and avelumab. We are very excited to see how that study turns out.”
Outside of TKIs, other opportunities for combination strategies exist as treatments for patients with advanced RCC. Historically, the VEGF-targeted antibody bevacizumab (Avastin) has paired well with other novel agents, according to Pal.
At this point, phase I trials have demonstrated sustained responses and tolerable safety for the combination of bevacizumab and the anti-PD-L1 agent atezolizumab (MPDL3280A), leading to the initiation of a phase III study comparing the combination with sunitinib. The current endpoint for the study is PFS and the targeted enrollment is 830 patients (NCT02420821).
Building on success in other areas, the combination of nivolumab and the CTLA-4 inhibitor ipilimumab (Yervoy) is also being assessed for patients with RCC (NCT02231749). In a phase I study exploring the combination, the objective response rate was between 43% and 48%, depending on the dose of each medication utilized.5 Additionally, these responses were durable, according to Pal.
In patients receiving a higher 3 mg/kg dose of ipilimumab and a 1 mg/kg dose of nivolumab, the adverse events associated with the combination were unacceptably high, noted Pal. However, when nivolumab was given at 3 mg/kg and ipilimumab was administered at 1 mg/kg, the combination was far more tolerable, he said.
“This is the dose that has made its way into the phase III experience comparing nivolumab and ipilimumab with sunitinib therapy,” according to Pal.
A number of studies are currently exploring various combination strategies. Among these studies are those assessing cabozantinib with ipilimumab plus nivolumab or nivolumab alone. Also, a study is examining bevacizumab with nivolumab for patients with RCC.As studies continue to assess novel immunotherapies as single-agents and in combination with VEGF-targeted therapy, the role for high-dose interleukin-2 (IL-2) grows increasingly unclear, noted Pal. Even in the absence of the latest explosion in immunotherapy for patients with RCC, IL-2 was a less common frontline option.
“Experts are always going to remain divided in their opinion on high-dose interleukin-2,” said Pal. “The criteria for selecting the appropriate high-dose interleukin-2 therapy will always remain inconsistent.”
The SELECT trial explored high-dose IL-2 in 120 patients with an ECOG performance status of 0 to 1, measurable and clearly progressive disease, and adequate organ function. By independent review, the objective response rate (ORR) was 25% with IL-2, which compares favorably with historical cohorts (14%).6
The complete response rate was 2.5% and the partial response rate was 22.5%. Importantly, 10.8% of patients in the study had a PFS time that lasted >3 years, which is the goal when using high-dose IL-2, said Pal.
The adverse events associated with high-dose IL-2 are hypotension, renal failure, and hyperbilirubinemia. There were two treatment-related mortalities in the study, which is important to keep in mind, cautioned Pal.
“What we saw in the context of SELECT is that clinical factors did not seem to predict outcomes to high-dose interleukin-2. In addition to that, biomarkers, such as carbonic anhydrase 9 (CA-9) and fibronectin, also did not predict clinical outcomes. This is an important lesson for us,” said Pal.
Interestingly, the SELECT trial did provide insight into PD-L1 and B7-H3 as biomarkers for immunotherapy. In those with PD-L1-positive tumors, the ORR with IL-2 was 50% compared with 19% in those with negative disease (P = .01). In the B7-H3—positive group, the ORR with IL-2 was 29% compared with 11% in the negative population (P = .08).6
“There were some interesting biomarker findings in SELECT. We see that PD-L1 was associated with a higher response rate, and this is quite striking given that data that we've seen with nivolumab therapy in the second- and third-line setting,” said Pal.There are selection biases for utilizing high-dose IL-2, which are inherent with the adverse events associated with the medication, explained Pal. In general, clinicians with an established infrastructure and prior experience with the medication are most likely to consider using the agent.
“High-dose IL-2 remains confined to specialized centers, and I really wanted to go to these specialized centers and figure out what their opinions were of high-dose IL-2,” said Pal.
To achieve this goal, Pal conducted an electronic survey asking participants at large centers in the United States and Europe whether or not they used high-dose IL-2.7 Overall, 33% of respondents said they did not use the immunotherapy.
Those who used IL-2 (67%) identified clear characteristics for treatment selection. In general, IL-2 was considered most frequently for those who were young, had clear cell histology, a good performance status, and no comorbidities.
Interestingly, for young patients at good risk with lung-only metastases, a majority of the respondents (64%) said they would consider a clinical trial exploring a checkpoint inhibitor or VEGF-directed therapy rather than administer IL-2, according to the survey.
“There's really no consensus at this point, after two decades of clinical research, regarding who the appropriate patient for high-dose IL-2 therapy actually is,” said Pal. “There is actually one question that did generate a consensus, and that was the question of whether there are appropriate biomarkers to identify ideal IL-2 patients. And here the answer was a resounding, 'No.'”