Dear Editor,
The real-world study by Chon et al., titled “Sorafenib vs. Lenvatinib in advanced hepatocellular carcinoma after atezolizumab/bevacizumab failure: A real-world study,” highlights important observations on the differences between second-line treatments for advanced hepatocellular carcinoma (HCC) after first-line treatment with atezolizumab plus bevacizumab (ATE+BEV [
1]. Although the study is important, certain issues could be addressed in future studies to improve the accuracy and applicability of the results in clinical practice.
Firstly, the impact of specific tumor characteristics, such as macrovascular invasion and extrahepatic metastasis, on treatment outcomes could be explored in more depth. These parameters have a clear impact on the prognosis of advanced HCC. For instance, previous trials have reported that the grade of macrovascular invasion is a complex parameter, which on one hand offers a poorer prognosis and potentially affects systemic therapies like tyrosine kinase inhibitors (TKIs) such as sorafenib and lenvatinib [
2]. Patient stratification by these characteristics may offer a personalized method to identify subgroups more likely to benefit from lenvatinib compared with sorafenib.
The second issue concerns the propensity score matching process. Propensity score matching is widely utilized in real-world study designs [
3]. Although the authors attempted to adjust for baseline characteristics (i.e., patients with comparable treatment histories) using propensity score matching, a significant difference remained in the number of cycles the two groups received before ATE/BEV after matching (
P=0.007, as indicated in Table 1 of the Chon et al. study1). The number of cycles could impact subsequent therapy due to the development of upfront resistance mechanisms [
4]. A deeper exploration of the impact of preceding ATE+BEV exposure on second-line therapy response would better inform these findings.
Finally, the study suggests that larger and possibly longer-term studies are needed, with a focus on how lenvatinib or sorafenib should be chosen based on tumor molecular profiling, if any. Current evidence suggests that molecular profiling may even differentiate those patients with HCC who will benefit from TKIs [
5]. Identifying these detectable predictive markers could eventually allow future studies to be more targeted, resulting in more desirable outcomes for patients.
In conclusion, Chon et al.’s study offers valuable insights into second-line treatments for advanced HCC but would benefit from further exploration of tumor characteristics, refined propensity score matching, and the potential of molecular profiling. These improvements could enhance the clinical applicability and personalization of treatment strategies.