Patients with unresectable hepatocellular carcinoma (HCC) and poor prognostic factors, including impaired liver function, reduced performance status, or main trunk portal vein thrombosis, are frequently excluded from phase 3 clinical trials. The phase 3b SIERRA study evaluated the safety and efficacy of the STRIDE regimen (Single Tremelimumab Regular Interval Durvalumab) in these underrepresented patient populations.
On October 6, 2026, the Journal of Hepatology published the primary analysis of the study, titled “First-line durvalumab and tremelimumab in poor-prognosis unresectable hepatocellular carcinoma: Phase IIIb single-arm SIERRA study.”
The study was conducted by Stephen L. Chan, Masatoshi Kudo, Bruno Sangro, Robin Kate Kelley, Jee Hyun Kim, Binh Pham, Jung Yong Hong, Lorenza Rimassa, and colleagues. The analysis provides updated safety findings and the first comprehensive efficacy results from the SIERRA trial (NCT05883644).
Why SIERRA Was Needed
Treatment decisions in HCC depend on both the extent of the cancer and the patient’s underlying liver function and general condition. The phase 3 HIMALAYA trial established STRIDE as a first-line treatment option for unresectable HCC, demonstrating an overall survival benefit compared with sorafenib. However, HIMALAYA included patients with Child-Pugh class A liver function and Eastern Cooperative Oncology Group performance status (ECOG PS) 0–1, while patients with main trunk portal vein thrombosis (Vp4) were excluded.
As a result, prospective evidence for STRIDE in patients with Child-Pugh class B liver function, ECOG PS 2, or Vp4 has remained limited. SIERRA was designed to complement the HIMALAYA findings by investigating STRIDE in these specific populations. Earlier results had described the initial safety profile of the regimen, while the current primary analysis includes additional safety follow-up and efficacy outcomes.
Study Design and Patient Population
SIERRA was a multicenter, single-arm phase 3b trial enrolling adults with unresectable HCC who had not received prior systemic therapy for HCC and were ineligible for locoregional therapy. A total of 111 patients received STRIDE across three mutually exclusive cohorts defined by the study’s eligibility criteria:
- Child-Pugh B7/B8 cohort (n = 44): Patients with impaired liver function, ECOG PS 0–1, and no Vp4.
- ECOG PS 2 cohort (n = 47): Patients with Child-Pugh class A liver function and no Vp4.
- Vp4 cohort (n = 20): Patients with Child-Pugh class A liver function, ECOG PS 0–1, and chronic main trunk portal vein thrombosis.
Patients received a single dose of tremelimumab 300 mg plus durvalumab 1,500 mg, followed by durvalumab 1,500 mg every four weeks until disease progression, unacceptable toxicity, or other treatment discontinuation criteria. The co-primary endpoints were the incidence of grade 3 or 4 adverse events possibly related to study treatment within six months of treatment initiation and investigator-assessed confirmed objective response rate (ORR), according to RECIST 1.1.
Secondary endpoints included overall survival (OS), progression-free survival (PFS), disease control rate (DCR), duration of response (DoR), and additional safety outcomes. The median age of participants was 70 years, and 86.5% were male. At the September 22, 2025, data cutoff, the median follow-up was 9.17 months.
Safety Outcomes
The incidence of grade 3 or 4 adverse events possibly related to treatment within six months was 19.8% (22/111 patients; 95% CI, 12.9–28.5) in the overall population.
Rates across the three cohorts were:
- Child-Pugh B7/B8: 18.2% (95% CI, 8.2–32.7).
- ECOG PS 2: 19.1% (95% CI, 9.1–33.3).
- Vp4: 25.0% (95% CI, 8.7–49.1).
Adverse events of any grade were reported in 94.6% of patients, while grade 3 or 4 adverse events from any cause occurred in 47.7%. Adverse events possibly related to treatment occurred in 70.3% of patients. The most frequently reported were pruritus (22.5%), hypothyroidism (11.7%), and diarrhea (10.8%). Serious adverse events occurred in 36.0% of patients, and adverse events led to discontinuation of study treatment in 12.6%. Immune-mediated adverse events were reported in 28.8%.
Adverse events with a fatal outcome occurred in 13 patients (11.7%). Two patients (1.8%), both in the ECOG PS 2 cohort, experienced fatal events assessed as possibly related to treatment, involving drug-induced liver injury and increased blood bilirubin. The investigators considered the overall safety findings broadly consistent with those previously reported for STRIDE in HIMALAYA. However, differences in study populations, treatment exposure, and adverse-event assessment periods prevent direct comparisons between the trials.
Antitumor Activity
STRIDE demonstrated antitumor activity across the SIERRA population, with an investigator-assessed confirmed ORR of 14.4% (95% CI, 8.5–22.4). Response rates varied across the three cohorts:
- Child-Pugh B7/B8: 6.8% (95% CI, 1.4–18.7).
- ECOG PS 2: 21.3% (95% CI, 10.7–35.7).
- Vp4: 15.0% (95% CI, 3.2–37.9).
Overall, 16 patients achieved a confirmed objective response, including two complete responses (1.8%) and 14 partial responses (12.6%). Stable disease was observed in 51 patients (45.9%). The disease control rate at 24 weeks was 44.1% (95% CI, 34.7–53.9), while the median duration of response was 9.6 months (95% CI, 7.39–not calculable).
These results provide prospective evidence of antitumor activity with STRIDE in patients with poor prognostic features, although the differences in response rates between cohorts should be interpreted descriptively given their small sample sizes and distinct baseline characteristics.
Progression-Free and Overall Survival
At the primary analysis, the median PFS was 3.8 months (95% CI, 3.12–5.42) in the overall study population. Median OS was 10.6 months (95% CI, 6.70–16.72). The estimated overall survival rates were:
- 12-month OS: 45.5% (95% CI, 35.71–54.70).
- 24-month OS: 35.3% (95% CI, 24.91–45.83).
Although these findings provide initial survival estimates, the investigators emphasized that OS data remained immature at the primary analysis. The median follow-up was 9.17 months, with shorter follow-up in the Child-Pugh B7/B8 and Vp4 cohorts. Longer observation is therefore needed to better characterize survival outcomes.
In a post hoc exploratory analysis, the study also assessed changes in albumin-bilirubin (ALBI) scores over time. Mean ALBI scores remained generally stable among patients with available assessments, although the decreasing number of patients at later timepoints limits interpretation of these findings. Importantly, because SIERRA did not include a comparator arm, the reported survival outcomes cannot establish an overall survival advantage over alternative treatment approaches.
Clinical Implications and Study Limitations
The SIERRA study provides prospective safety and efficacy evidence for STRIDE in patients with unresectable HCC and poor prognostic factors who have been underrepresented in pivotal phase 3 trials. However, the findings should be interpreted in the context of several study limitations.
The single-arm, open-label design limits comparative conclusions, and the individual cohorts were relatively small, particularly the Vp4 cohort, which included only 20 patients.
In addition, eligibility criteria required the three prognostic groups to be mutually exclusive. Patients with overlapping poor prognostic characteristics, such as impaired liver function combined with ECOG PS 2 or Vp4, were therefore not specifically evaluated.
Patients with Child-Pugh B9 or C liver function, ECOG PS 3, and recent gastrointestinal bleeding were also excluded. These restrictions are important when considering the applicability of the findings to the broader population of patients with advanced HCC, particularly those with multiple risk factors or more severe underlying liver dysfunction.
Looking Ahead
The primary analysis of the phase 3b SIERRA trial reported a 19.8% incidence of grade 3 or 4 adverse events possibly related to treatment within six months and an objective response rate of 14.4% in patients with unresectable HCC and poor prognostic factors. The investigators concluded that the safety and efficacy findings indicate a favorable risk-benefit profile for STRIDE in this population with a high unmet need, supporting its further evaluation as a potential first-line treatment option.
Longer follow-up, including updated survival data and individual cohort assessments, will be important to better define the clinical outcomes of STRIDE in patients with Child-Pugh B7/B8 liver function, ECOG PS 2, or Vp4.
The full article is available in Journal of Hepatology.
Find more videos on OncoDaily YouTube TV.
