973P - Atezolizumab plus bevacizumab or lenvatinib versus sorafenib as first-line therapy for advanced hepatocellular carcinoma: Overall survival using real-world data from TrinetX platform

Background

Atezolizumab plus bevacizumab (A+B) and lenvatinib (L) are approved first line systemic therapies for advanced hepatocellular carcinoma (HCC). They demonstrated superiority to sorafenib in clinical trials, but assessment from big real world data (RWD) studies are lacking. Here we present results of overall survival (OS) using the TrinetX platform.

Methods

Using TriNetX Global Collaborative Network we identified patients with advanced HCC from 66 healthcare organizations treated with S, L or A+B as first line systemic therapy. Treatment start happend between January 2015 and December 2022. Propensity score (PSM) was used to balance cohorts on age, gender and race. Kaplan-Meier analysis was used to compare the 5-year overall survival (OS) between the cohorts.

Results

We identified 7038 patients: 1030 treated with A+B, 1674 with L, and 4334 with S. Mean age was 65 years old and 78% were male. Second line therapy was received by 1682 patients (23.8%), 48% of them with immunotherapy. When analysing raw data, OS was significantly better with A+B (mOS 449 days) compared to S (mOS 374 days) (HR 0.88, 95% CI 0.80-0.97, p=0.047), and with L (mOS 460 days) compared to S (HR 0.90, 95% CI 0.84-0.98). After matching cohorts by sex, age, and race OS difference maintained with A+B vs S (HR 0.83, 95% CI 0.73-0.93, p=0.002) but it was not significant for L vs S (HR 0.92, 95% CI 0.84-1.01, p=0.12).

Conclusions

In this RWD setting, A+B maintains a robust OS benefit when compared to S as first line therapy for advanced HCC. L seems to be superior to S but difference in OS is not significant when patients data are matched. These results support the use of A+B as standard treatment.

Clinical trial identification

Editorial acknowledgement

Legal entity responsible for the study

The authors.

Funding

Has not received any funding.

Disclosure

All authors have declared no conflicts of interest.