In selected patients with locally advanced hepatocellular carcinoma (HCC), liver resection after initial atezolizumab plus bevacizumab delayed treatment failure by nearly nine months compared with continued drug therapy, according to a randomized phase 3 trial published in The Lancet.
“A response-adaptive treatment strategy needs to be integrated into clinical practice,” senior author Jia Fan, MD, PhD, FACS, of the Liver Cancer Institute at Zhongshan Hospital, Fudan University, Shanghai, China, told GI & Hepatology News. “Previously, the emphasis was primarily on engaging multidisciplinary teams (MDT) at initial diagnosis of hepatocellular carcinoma to determine the primary treatment modality. Following this study, continuing systemic therapy should no longer be the default option for patients who respond favorably. Instead, they require dynamic reassessment by an MDT during treatment to evaluate their eligibility for potentially curative interventions.”
The open-label TALENTOP trial enrolled 489 treatment-naive patients with HCC at 24 hospitals in China between April 2021 and July 2024. All had tumors with macrovascular invasion but no extrahepatic spread, Child-Pugh class A liver function, and an Eastern Cooperative Oncology Group performance status of 0 or 1.
Patients first received three cycles of intravenous atezolizumab plus bevacizumab, followed by one cycle of atezolizumab alone. Each cycle lasted three weeks.
After induction, 201 patients (41%) who had a partial response or stable disease and whose tumors were judged resectable were randomly assigned to liver resection followed by 12 months of atezolizumab plus bevacizumab (n = 101) or to continued atezolizumab plus bevacizumab until loss of clinical benefit or unacceptable toxicity (n = 100). Most patients who were not randomized had progressive disease.
The primary endpoint was time to treatment failure, defined as the time from randomization to local recurrence or progression, extrahepatic spread, or death from any cause, as assessed by an independent review facility. The analysis included all randomly assigned patients.
After a median follow-up of 18.4 months, median time to treatment failure was 20.4 months with surgery vs. 11.8 months with maintenance therapy (hazard ratio, 0.60). Treatment failure occurred in 37% of patients assigned to surgery and 59% assigned to maintenance therapy. The results come from a prespecified interim analysis; the trial is ongoing.
The investigators reported that the benefit was generally consistent across prespecified subgroups, including patients whose tumors shrank by at least 10% and those with less or no shrinkage, and that the result held when the 18 patients assigned to surgery who did not undergo the procedure were excluded.
Among the 83 patients who underwent resection, 93% had an R0 resection — complete tumor removal with microscopically negative margins — and 28% had a pathological complete response, with no viable tumor cells in the surgical specimen.
Overall survival data were immature. At data cutoff, 15% of patients assigned to surgery and 23% assigned to maintenance therapy had died. Survival at 12, 18, and 24 months was 86%, 86%, and 83%, respectively, with surgery, compared with 91%, 78%, and 67% with maintenance therapy. The difference was not statistically significant. The authors attributed the early crossing of the survival curves to data immaturity and possibly to acute risks of the surgical strategy, including one perioperative death.
Grade 3 or 4 treatment-related adverse events occurred in 39% of patients who underwent surgery and received postoperative therapy vs. 21% who received maintenance therapy; serious treatment-related adverse events occurred in 23% vs. 7%. The most common grade 3 or 4 events were elevated alanine aminotransferase levels, 8% vs. 1%; low platelet counts, 8% vs. 3%; and proteinuria, 4% vs. 7%.
Clavien-Dindo grade 3 or higher surgical complications occurred in 16% of patients who underwent resection. Two treatment-related deaths occurred in the surgery group and none in the maintenance group. One was due to abnormal liver function considered related to atezolizumab, and the other to liver failure considered related to atezolizumab, bevacizumab, or surgery.
The authors acknowledged several limitations: all patients were enrolled in China and were younger than those in the IMbrave150 trial; induction therapy was given for a fixed four cycles, precluding assessment of optimal resection timing; overall survival data are still maturing; and patient-reported outcomes, including quality of life, were not assessed. About 90% of patients had hepatitis B virus–related HCC, which the authors said reflects global epidemiology and should not limit the strategy’s applicability. They wrote that multidisciplinary team evaluation and careful patient selection are crucial to ensure perioperative safety, and that longer follow-up is needed to confirm whether the improvement in time to treatment failure translates into a survival benefit.
The investigators concluded that patients whose disease responds or remains stable after initial systemic therapy “should undergo multidisciplinary team assessment for resectability and be considered for liver resection.”
Shanghai Roche Pharmaceuticals and the Ministry of Science and Technology of China funded the study. Dr. Fan and several coauthors reported research funding, speaker fees, consulting fees, honoraria, travel support or advisory roles involving Roche and other pharmaceutical companies.
Expert Insight
Dr. Fan shared his perspective on the study with GI & Hepatology News.
Based on these findings, which patients with advanced hepatocellular carcinoma should be considered for liver resection after responding to systemic therapy?
Dr. Fan: In this study, patients who had non-progressive disease after 4 cycles of systemic therapy and were eligible for R0 surgical resection underwent randomization. In clinical practice, we generally evaluate the feasibility of surgical resection in patients who show a favorable treatment response (radiologically assessed as partial response, or at least tumor shrinkage). Surgical resectability is evaluated by surgeons based on criteria including sufficient residual liver volume after future liver resection, adequate liver function, and satisfactory general conditions.
How should physicians weigh the longer time to treatment failure with surgery against the higher risk of serious adverse events?
Dr. Fan: We indeed need a balance between clinical benefits and adverse events. On the one hand, surgery offers the potential for curative resection and disease-free survival, along with the possibility of systemic treatment cessation after one year of postoperative therapy. On the other hand, surgical-related adverse events do place patients at risk and overall survival benefit currently remains unclear. Therefore, a thorough preoperative risk-benefit evaluation and the patient’s individual preference are both critical factors in this decision-making process.
Are these results strong enough to change practice now, or should physicians wait for mature overall survival data?
Dr. Fan: I believe the current data are already strong enough to change clinical practice. Given the limited second-line treatment options available following first-line immuno-combination therapy, such as atezolizumab-bevacizumab, in advanced hepatocellular carcinoma, the benefit observed in time-to-treatment failure (TTF) is highly likely to translate into an overall survival advantage. Furthermore, for patients who underwent surgical resection, a TTF event means tumor recurrence and is always with a relatively low tumor burden. In contrast, in patients who did not undergo surgery, a TTF event usually coincides with a much higher tumor burden. We are therefore confident that the TTF benefit will translate into an overall survival benefit. Nonetheless, further long-term follow-up data for overall survival remain essential.