Nivolumab in RCC Phase II Trial

On Saturday, the major oral presentations on kidney cancer were given, covering the Phase II results for nivolumab, a Phase I trial using nivolumab in combination with sunitinib or pazopanib, and a discussion of the immunomodulatory activity of nivolumab. Interspersed among these was a report on Genentech’s entry in the immunotherapy race, a PD-L1 inhibitor. As I’ve said before, PD-1 is the hot topic of the conference and 3200 people attended the session. Thomas Powles, who is from Barts Cancer Institute in London, and who presented the bladder cancer study, said that he never saw so many people attending a bladder cancer presentation. I’ll summarize the Phase II nivolumab study.

Dr Robert Motzer from Memorial Sloan Kettering reported on the results of a Phase II trial for the anti-PD-1 drug nivolumab. I wrote up the mechanism of the drug last year and posted it on SmartPatients calling it PD-1 Defined, so I won’t repeat it here. Only to say that nivolumab is a monoclonal antibody that binds to the PD-1 receptor on the T-cell preventing the tumor cell from doing so, which would disable the T-cell.

In this trial, 168 patients were randomized into three arms, .3 mg/kg, 2 mg/kg, and 10 mg/kg. To be eligible for the trial, patients had to have a clear cell component in their tumor and have had at least one prior anti-angiogenic therapy, like sunitinib, but not more than three priors. The patients were categorized with the following risk status: 33% favorable, 42% intermediate, and 25% poor. 90% of the patients were from the U.S., with the others from Canada, Finland, and Italy. This was a heavily pre-treated population with 30% having one prior, 37% having two priors, and 33% having three priors or more. 74% had taken sunitinib, 34% everolimus, 27% pazopanib, and 23% Interleukin-2.

A patient was kept on treatment even if they progressed if they tolerated the treatment and they had clinical benefit – 22% of the patients. 6% of patients came off treatment due to toxicity, with 75% coming off due to progression.

The progression-free survival (pfs) for the three arms was: .3 mg/kg – 2.7 mos; 2 mg/kg – 4.1 mos; 10 mg/kg – 4.2 mos, with no significant difference. Likewise, the objective response rate ORR (CR and PR) was 20%, 22%, and 20%, respectively. These figures don’t seem to be impressive, except for the following: these patients were heavily treated before entering the trial; many of the partial responses were called “deep responses”, or over 75% with a sprinkling of complete responses; and of the original 35 responders, 20 are still responding, and 14 of these are still responding over two years later. If we were compare the results of this trial to those of the Phase III trial of axitinib (Inlyta) versus sorafenib (Nexavar), the pfs for axitinib was 6.7 mos and sorafenib 4.7 mos. But this trial only included patients who had had one prior therapy, while for many of the patients on the this trial, nivolumab was 3rd or 4th line. Of course, this comparison is not really scientifically valid – it is what is called hypothesis building. I only include it because Dr. Motzer himself compared overall survival to this and other trials, with axitinib/sorafenib being the best.

Overall survival was: 3 mg/kg – 18.2 mos; 2 mg/kg – 25.5 mos; 10 mg/kg – 24.7 mos. in a separate analysis, the 2 and 10 mg/ kg doses tended towards greater OS. 18 patients are still on trial while 97 of the original 168, or 58%, have died. Dr. Motzer stratified the median OS by risk group and number of previous treatments, as follows: favorable – not yet reached; intermediate – 20.3 months; and poor – 12.5 months. Also, 1 previous treatment – not yet reached; >= 2 previous treatments – 18.7 months. He also compared the results with other trials, the best of which was axitinib versus sorafenib: 15.2 mos and 16.5 mos.

Finally, there was no dose-dependent relationship vis-a-vis pfs.

The Phase III trial of nivolumab versus everolimus is fully recruited. This trial was initiated even before the Phase II results of nivolumab were released indicating that Bristol-Myers Squibb wants to fast track this drug (nivolumab received Breakthrough Therapy designation for Hodgkin Lymphoma by the FDA). A new Phase III trial using nivolumab as a first-line therapy in RCC is being planned.

Lauren Harshman from Dana Farber commented on the results of this trial. She said that the pfs of 2.7 – 4.2 mos was disappointing as compared to the 7 mos. pfs for the Phase I trial of nivolumab. Also, the ORR of 20-22% was lower than the 29% rate of the first trial. She listed a few possibilities for the difference, one being that the people entering Phase I trials usually have a more indolent tumor load. But she was impressed by the OS.

She noted the new paradigm for the immune system’s response to cancer, which consists of three stages. The first stage is “Elimination” whereby the body’s T-cells and Natural Killer (NK) cells eliminate nascent tumor cells, which are not even discernible to radiology exams. The second stage is “Equilibrium” where tumor cells co-exist with immune cells, in eternal battle where the immune cells suppress the tumor cells that attempt to grow and proliferate. The third stage is called “Escape” where the tumor cells grow, proliferate, and invade. The latter stage could be due to suppression of the immune system, e.g., PD-1 or other receptor suppression. The objective of these anti-PD-1 molecules is to block the suppression, or reactivate the T-cells to attack the tumor cells and return the tumor’s microenvironment back to Equilibrium, or even Elimination in the case of complete responses.

So what is the take-home from this trial? These are my thoughts.

1. If you go on an anti-PD-1, make sure your oncologist is experienced in handling the side effects, especially pneumonitis.
2. These drugs can be late acting so the first new lesion should not be grounds for removal from a trial, especially if the drug is well tolerated and the rest of your tumor burden is stable or even decreasing. There should be at least two scans to evaluate progression properly.
3. The pfs and even OS data are not outstanding and sunitinib, which in its Phase III seminal trial showed an OS of 26.4 months and more recently of 32 months. But, to be fair, the patients on this nivolumab trial were heavily pre-treated, contained poor prognosis patients as well as favorable and intermediate, and for some patients showed durable responses. We’ll have to wait further trials to really evaluate its effectiveness.
4. The low side effect profile lends nivolumab to be combined with other therapies, and that’s already happening with TKIs and other immune therapies like ipilimumab. Plus, immunotherapies that block other suppressors are being developed. As one investigator here said, no single therapy will cure kidney cancer, there will have to be a combination to therapies.