Ghassan Abou-Alfa, MD, medical oncologist at Memorial Sloan Kettering Cancer Center, expands on findings that durvalumab plus tremelimumab demonstrated an overall survival benefit at 4 years for certain patients with unresectable hepatocellular carcinoma.
According to a study led by Ghassan Abou-Alfa, MD, medical oncologist at Memorial Sloan Kettering Cancer Center, a single tremelimumab regular interval durvalumab (STRIDE) regimen reduced the risk of death by 22% compared with sorafenib. These findings are applicable to patients with unresectable hepatocellular carcinoma who have not received prior systemic therapy and are ineligible for localized treatment.
Can you give an overview of the HIMALAYA phase 3 trial findings?
The HIMALAYA study was an evaluation for 2 immunotherapies, or what you call checkpoint inhibitors, assessed vs standard care, which was a tyrosine kinase inhibitor [TKI], or what we call a biologic therapy called sorafenib. This study evaluated close to 1200 patients worldwide; these patients have advanced liver cancer that is in need for systemic therapy. This was first-line treatment before they get any other treatment, understandably, with appropriate liver performance, what we call Child-Pugh score A. The patients, however, came independent of what etiology or risk factor they had for liver cancer. So we had, really, a superb representation of liver cancer throughout the world, independent of etiology or any components like ethnicity or background.
This study evaluated those patients for the convention of the anti–PD-L1 called durvalumab and the anti–CTLA-4 called tremelimumab, and they were evaluated, as I mentioned, vs sorafenib. The fascinating part is—right before we get to the results—the tremelimumab was given only once. It's a quite innovative approach. It's like jump-starting a car, you really put the engine on motion with the tremelimumab 1 dose—imagine 1 dose only in a patient's lifetime—and then durvalumab was given on a repetitive basis once every month.
The study ultimately led to the finding of improvement of median overall survival close to about 16.4 months for the durvalumab-tremelimumab vs 13.8 months for the sorafenib, which was exactly as expected for the sorafenib. The study also did evaluate durvalumab as a single agent, and this was assessed vs sorafenib in what we call noninferiority—we're trying to make sure that they are equal. Why did [we] do that? There was a reason, because from prior studies and efforts on other checkpoint inhibitors, we did not see that an anti–PD-L1 by itself will do more than what a TKI can do like, for example, sorafenib.
As such, we're looking to noninferiority, and this was positive for that purpose. It showed, again, about 16.6 months in survival for the durvalumab single agent and sorafenib 13.8 months, and they are totally equal and there was noninferiority for that purpose. So the study, if anything, met its 2 needs, the primary point being superiority for the durvalumab-tremelimumab vs sorafenib, and noninferiority for the durvalumab single agent vs sorafenib.
Can you provide more insight into the methodology used in the trial?
The study, as we mentioned, the HIMALAYA study, evaluated durvalumab-tremelimumab, and now it has a short name called the STRIDE regimen; [we] evaluated this vs sorafenib. As I mentioned already, the eligibility were to be fit in. It was a rather pretty broad cast, almost close to real life, because patients, independent of etiology [and] independent of ethnicity—as long as they have advanced disease in need for systemic therapy and, at the same time, never got any systemic therapy—were randomized in a fashion that was to go either to the doublet of durvalumab-tremelimumab vs the single-agent durvalumab vs sorafenib.
As we just mentioned, the patients, of course, were evaluated for presumed adverse events; at the same time, any events of progression and discontinuation of therapy in those cases or sadly if any patient passed away. Ultimately, we evaluated the study for survival and to other end points...all the expected ones, like progression-free survival response rate, etc.
This transcript has been lightly edited for clarity.