Oncology · Medical Literature
[Articles] Addition of ipilimumab to atezolizumab plus bevacizumab in advanced hepatocellular carcinoma (PRODIGE 81-FFCD 2101-TRIPLET HCC): phase 2 results from a randomised, multicentre, open-label, phase 2–3 trial
Abstract
The addition of ipilimumab to atezolizumab plus bevacizumab did not show any benefit in patients with previously untreated, unresectable hepatocellular carcinoma. These results do not support the addition of low-dose (1 mg/kg) ipilimumab to atezolizumab plus bevacizumab as a first-line treatment in this setting.
Clinical relevance
Does the addition of low-dose ipilimumab (1 mg/kg, anti-CTLA-4) to standard atezolizumab plus bevacizumab improve the objective response rate at 24 weeks in previously untreated patients with unresectable hepatocellular carcinoma, sufficiently to warrant progression to a phase 3 superiority trial? __SNAPSHOT__ {"studyDesign":"Randomised, multicentre, open-label, phase 2–3 adaptive trial (pre-planned phase 2 non-comparative analysis reported here; phase 3 progression contingent on meeting a pre-specified response threshold)","setting":"36 hospitals in France (PRODIGE/FFCD academic cooperative group network)","sampleSize":"n=229 randomised; n=226 received at least one dose (intervention: n=113, control: n=113)","population":"Adults ≥18 years with unresectable HCC (BCLC-B or BCLC-C), no prior systemic therapy, at least one measurable lesion per RECIST v1.1, Child–Pugh class A liver function, ECOG performance status 0–1; 91% male, 9% female","intervention":"Atezolizumab 1200 mg + bevacizumab 15 mg/kg + ipilimumab 1 mg/kg Q3W (ipilimumab for up to 4 doses); all agents given IV; total treatment duration up to 2 years","comparator":"Atezolizumab 1200 mg + bevacizumab 15 mg/kg Q3W for up to 2 years (standard doublet)","primaryOutcome":"Investigator-assessed objective response rate (ORR; complete or partial response) within the first 24 weeks of treatment by RECIST v1.1; pre-specified threshold of ≥35 responders in the experimental arm required to advance to phase 3","followUp":"Planned treatment duration up to 2 years, with 2-year post-enrolment follow-up per protocol; median follow-up duration at time of phase 2 analysis not reported in available abstract/sources","fundingSource":"Non-profit/academic — Fédération Francophone de Cancérologie Digestive (FFCD)"}
Key findings
The pre-specified phase 2 efficacy threshold was not met. In the experimental (triple) arm, 34 patients achieved an objective response within 24 weeks (ORR 30%; 80% CI 24–36%), narrowly falling short of the pre-specified threshold of ≥35 responders required to progress to phase 3. In the control (doublet) arm, 31 patients achieved an objective response (ORR 27%; 80% CI 22–34%). As the phase 2 analysis was designed as non-comparative, no formal between-arm statistical comparison (p-values, relative risks, or confidence intervals for the difference in ORR) was pre-specified or reported. Absolute difference in ORR between arms: approximately 3 percentage points (calculated), but this comparison was not powered or pre-specified. Duration of response, overall survival, and progression-free survival were assessed as secondary endpoints; median OS and PFS were reported as not reached or not evaluable in available data. Regarding safety, serious adverse events occurred in 55 patients (49%) in the triple-therapy arm versus 48 patients (42%) in the doublet arm. Treatment-related deaths occurred in 6 patients (5%) in the triple arm and 0 patients in the doublet arm — a clinically meaningful safety signal. Grade 3–4 treatment-related adverse events of note in the triple arm included arterial hypertension, asthenia, colitis, and confusional syndrome. NNH for treatment-related death (triple vs. doublet): approximately 1 in 19 patients treated with the triple regimen experienced a treatment-related death versus none in the control arm (calculated).
Practice implications
This phase 2 trial established that the addition of low-dose ipilimumab (1 mg/kg) to atezolizumab plus bevacizumab did not produce a sufficient improvement in 24-week ORR to meet the pre-specified bar for progression to phase 3 (34 responders achieved vs. 35 required). The increment in response over the doublet was negligible (~3 percentage points, calculated) and was accompanied by a meaningfully higher burden of toxicity — including a 5% rate of treatment-related death in the triple arm versus 0% in the doublet arm. The trial did not establish benefit for the triple regimen on any efficacy endpoint. Patients represented were exclusively treatment-naïve, Child–Pugh A, ECOG 0–1, predominantly male (91%), and enrolled only at French academic centres, limiting generalisability to patients with more advanced hepatic impairment, poorer performance status, or diverse ethnic and geographic backgrounds. Because the full publication is available but the current summary is based on published abstract data and secondary source reporting, complete secondary endpoint data (including median OS, PFS, duration of response, and subgroup analyses) may carry additional nuance not fully captured here. This should be noted when interpreting secondary outcomes. The results do not support adoption of this triple regimen in clinical practice. __LIMITATIONS__ Non-comparative phase 2 design: the study was powered and pre-specified for a single-arm efficacy threshold in the experimental arm, not for a formal head-to-head comparison; between-arm differences must be interpreted with extreme caution.,Single-country study (France only): all 36 sites were French academic hospitals within the PRODIGE/FFCD network, limiting generalisability to community settings and non-European populations.,Highly selected population: Child–Pugh A only, ECOG 0–1, treatment-naïve — excludes patients with intermediate hepatic impairment or declining performance status who are frequently encountered in practice.,Strong sex imbalance (91% male): limits applicability to female patients with HCC.,Surrogate primary endpoint: 24-week ORR by RECIST v1.1 is a surrogate; while ORR has been proposed as a correlate of OS benefit in HCC immunotherapy, OS benefit is not directly established from this phase 2 analysis.,Open-label design introduces potential investigator bias in response assessment.,Ipilimumab dose (1 mg/kg) may have been insufficient for meaningful CTLA-4 blockade; the dose selected was chosen to minimise toxicity, and the results cannot rule out benefit at higher doses.,Short follow-up for survival endpoints at the time of phase 2 analysis; OS and PFS data were immature (not reached/not evaluable).,No industry sponsorship (FFCD-funded), which reduces commercial bias but also means limited resources for correlative biomarker or translational substudies whose findings might identify a responsive subpopulation.,Summary based on abstract and secondary source reporting; complete statistical details, subgroup analyses, and biomarker data require review of the full publication. __BOTTOM_LINE_CATEGORY__ not_ready_for_clinical_application __BOTTOM_LINE__ The TRIPLET-HCC phase 2 trial found that adding low-dose ipilimumab (1 mg/kg) to atezolizumab plus bevacizumab produced no meaningful gain in 24-week objective response rate (30% vs. 27%) while increasing serious adverse events and causing treatment-related deaths in 5% of patients in the triple arm versus none in the doublet arm. The pre-specified efficacy threshold was not met, and the trial will not advance to phase 3. Clinicians should not add ipilimumab to atezolizumab plus bevacizumab in the first-line treatment of unresectable HCC; the established doublet remains the standard of care in this context.