Are we overestimating success of salvage hepatectomy in unresectable hepatocellular carcinoma?

Abstract

The Zhang et al’s study addresses an important clinical question of timing and role of salvage surgery post-downstaging procedures in patients with advanced hepatocellular carcinoma wherein different modalities like trans arterial chemoembolization, tyrosine kinase inhibitors, and anti-programmed cell death 1 antibodies have been used as downstaging procedure. Although proper selection of patients is a pre-requisite for salvage related liver failure.

Key Words: Salvage hepatectomy; Pathological complete response; Tyrosine kinase inhibitors; Unresectable hepatocellular carcinoma; Locoregional modalities

Core Tip: The primary treatment modality for advanced hepatocellular carcinoma is systemic therapy. However, downstaging can be done with use of locoregional modalities like trans arterial chemoembolization, transarterial radioembolization, radiation therapy etc. along with systemic therapies (tyrosine kinase inhibitors and immune checkpoint inhibitor) followed by salvage surgery results in improved survival with limited adverse effects, if patient selection is judiciously made.

TO THE EDITOR

We commented Zhang et al[1] for addressing a timely and clinically significant question: The role of salvage resection following conversion therapy in initially unresectable hepatocellular carcinoma. The integration of transarterial chemoembolization, tyrosine kinase inhibitors, and anti-programmed cell death 1 (PD-1) antibodies represents an evolving therapeutic paradigm. However, we respectfully submit that several methodological and contextual concerns warrant critical scrutiny before these findings can be reliably applied in clinical decision-making.

The authors report that only 28 of 117 patients (23.9%) who received conversion therapy proceeded to salvage surgery. This limited surgical cohort significantly underpowers survival and recurrence analyses and introduces selection bias. Crucially, the absence of a non-surgical comparator group restricts causal inference regarding surgical efficacy. In contrast, larger studies like Wu et al[2] (n = 70, multicenter) utilized robust designs to validate outcomes. The reported 50% pathological complete response (PCR) rate substantially exceeds established benchmarks; Wu et al[2] reported 41.4% PCR after triple therapy, and Huang et al[3] documented 42.85% PCR in a multicenter cohort. Recent evidence from Jia et al[4] and Lau and Lai[5] also showed that even under optimized protocols, PCR rates following conversion therapy typically range between 20%-35%, underscoring the outlier nature of Zhang et al’s findings[1]. This discrepancy is particularly notable given the median of one cycle of PD-1 blockade in the current study, which challenges biological plausibility and raises concerns about histopathological assessment or staging artifacts.

The 2-year overall survival (OS) rate of 87.6% exceeds rates in comparable studies; Wu et al[2] reported 94.4% 2-year OS, while long-term data (e.g., Lau and Lai[5]) show 5-year OS as low as 24.9%-57%. The median follow-up of 15 months is insufficient to assess late recurrences, a limitation not observed in studies with longer follow-up, such as the 5-year data presented by Lau and Lai[5]. Postoperative complications occurred in 71.4% of patients, with 14.3% being Clavien-Dindo grade III-V. This contrasts sharply with Wu et al[2], where severe morbidity was 7.1%. The current study lacks granular analysis of predictors (e.g., liver reserve, surgical approach) for these complications. In this context, Li et al[6] emphasize the importance of salvage timing and early progression patterns as determinants of postoperative safety and long-term outcomes.

Aggregating outcomes across heterogeneous regimens (multiple tyrosine kinase inhibitors, PD-1 inhibitors, transarterial chemoembolization cycles) obscures protocol-specific efficacy. Details on agents, duration, and rationale for postoperative therapy (administered to 57.1% of patients) are omitted, blurring the attribution of survival benefits to surgery vs systemic control. Preoperative albumin’s association with recurrence-free survival likely reflects hepatic decompensation rather than surgery-specific effects. This lacks mechanistic validation and may not be generalizable. As Zhang et al[7] highlighted, hepatic function reserve and tumor biology - rather than serum protein levels alone - are more reliable predictors of post-conversion surgical outcomes.

Prospective trials are needed to address these limitations. Future studies should standardize conversion and surgical protocols, include matched non-surgical control arms, and stratify by liver function (e.g., albumin-bilirubin or model for end-stage liver disease score), tumor biology, and etiology (hepatitis B virus vs hepatitis C virus or nonalcoholic steatohepatitis). Extended follow-up beyond two years is essential to capture late recurrences and true long-term survival.

In conclusion, while Zhang et al[1] contribute meaningful data, the study’s limitations-particularly the small sample size, outlier PCR and OS rates, and deviation from benchmarks in the existing literature-suggest the findings should be interpreted as hypothesis-generating. Rigorous, multicenter prospective studies are essential before these results can inform clinical guidelines.