麻豆传媒

SAN FRANCISCO -- An immunotherapy combination significantly improved overall survival (OS) versus sorafenib (Nexavar) as initial therapy for unresectable hepatocellular carcinoma (HCC), a large randomized trial showed.

Patients treated with a single priming dose of tremelimumab followed by durvalumab (Imfinzi) had a median OS of 16.4 months as compared with 13.8 months with sorafenib. Survival curves began to separate after about 9 months and continued to separate thereafter. Single-agent durvalumab led to a median OS of 16.6 months, which met statistical criteria for non-inferiority versus sorafenib but not superiority, reported Ghassan K. Abou-Alfa, MD, MBA, of Memorial Sloan Kettering Cancer Center in New York City, at the Gastrointestinal Cancers Symposium.

"The [current] was a large phase III study that included a global, heterogeneous population, represented of patients with unresectable HCC," Abou-Alfa said. "STRIDE [single tremelimumab regular-interval durvalumab] appeared to provide a long-term survival benefit, with a landmark 36-month overall survival of 30.7%."

"Overall survival with durvalumab monotherapy was non-inferior to sorafenib with a favorable benefit-risk profile. The STRIDE regimen and durvalumab monotherapy may represent new treatment options for patients with unresectable HCC," he stated.

Unresectable HCC

The IMbrave150 trial set the current high-water mark among phase III trials in unresectable HCC with a median of 19.2 months with atezolizumab (Tecentriq) and bevacizumab (Avastin) versus 13.4 with sorafenib, making it the current standard of care, said invited discussant Anthony B. El-Khoueiry, MD, of USC Norris Comprehensive Cancer Center in Los Angeles. However, the combination led to bleeding events in 25% of patients and severe bleeding in 6.4%.

The STRIDE regimen avoids the bleeding risks associated with VEGF inhibitors, which could be an advantage given that most subsequent regimens target VEGF. The STRIDE regimen led to more immune-mediated adverse events (IMAEs), 20% of which required management with steroids, El-Khoueiry noted. In contrast, in the trial of nivolumab (Opdivo) and ipilimumab (Yervoy) in HCC previously treated with sorafenib, 57% of patients treated with high-dose ipilimumab required steroids to manage IMAEs.

"STRIDE ... is a new first-line treatment option for advanced HCC patients," said El-Khoueiry. "STRIDE has a differentiated toxicity profile from VEGF-containing combinations and does not have VEGF-related toxicities."

The results with single-agent durvalumab are similar to those obtained with nivolumab monotherapy versus sorafenib in the , El-Khoueiry continued. Although the trial did not demonstrate nivolumab superiority, the median OS of 16.4 months was nearly identical to the 16.6 median OS in HIMALAYA.

"The constellation of data does offer a first-line treatment option for selected patients who may not be candidates for combination therapy or who have VEGF contraindications," he said.

Investigators hypothesized that a single priming dose of a CTLA-4 inhibitor, followed by anti-PD1/L1 immunotherapy, would improve survival in unresectable HCC and avoid adverse effects associated with CTLA-4 inhibition. Such a regimen would also avoid the bleeding risks associated with VEGF inhibition. The hypothesis was supported in a of tremelimumab and durvalumab, which produced a median OS of 18.7 months with the STRIDE regimen, setting the stage for the phase III HIMALAYA trial.

HIMALAYA Results

HIMALAYA included 1,324 patients with untreated, unresectable HCC, randomized to the STRIDE regimen, single-agent durvalumab, or single-agent sorafenib. The primary endpoint was OS for the STRIDE regimen versus sorafenib, and the key secondary objective was OS for single-agent durvalumab versus sorafenib.

The patients had a median age of 64-65, men accounted for 83%-84% of the total patient population, about 40% of patients were from Asia (excluding Japan), and 55%-60% of the patients had hepatitis-associated HCC.

The trial met the primary objective, as patients assigned to the STRIDE regimen had a 22% reduction in the survival hazard as compared with sorafenib (95% CI 0.65-0.92, P=0.0035).

"The hazard ratio to 9 months was 0.87 but then improved all the way to 0.70 after 9 months," said Abou-Alfa. "This priming of the T cells [with tremelimumab] takes time and, if anything, continues to propagate at landmark analyses at 18 months, 24 months, and 36 months. By 36 months, 30% of the patients on [STRIDE] were still alive as compared to 20% with sorafenib."

Subgroup analysis showed a consistent benefit for the combination versus sorafenib for almost all of the prespecified groups.

The 16.6-month median OS associated with single-agent durvalumab proved to be noninferior to sorafenib (HR 0.86, 95% CI 0.73-1.03) but did not achieve statistical superiority. Landmark analyses showed 18-month OS values of 47.4% versus 41.5%, 24-month values of 39.6% vs 32.6%, and 36-month values of 24.7% versus 20.2%, all favoring durvalumab.

Median progression-free survival was similar with STRIDE (3.78 months), single-agent durvalumab (3.65 months), and sorafenib (4.07 months). Response rates favored the STRIDE (20.1%) and durvalumab (17.0%) groups over the sorafenib control group (5.1%).

Grade 3/4 AEs occurred more often with STRIDE (50.5%) and sorafenib (52.4%) than with single-agent durvalumab (37.1%). Treatment-related AEs (TRAEs) were more common with sorafenib (36.9%) than with STRIDE (25.8%) or durvalumab monotherapy (12.9%). Serious TRAEs occurred more often with the combination (17.5%), versus 8.2% with durvalumab and 9.4% with sorafenib. Rates of TRAEs leading to discontinuation were 11.0% with sorafenib, 8.2% with STRIDE, and 4.1% with single-agent durvalumab.

Comment