Abstract 984: Optimizing first-line systemic therapy selection in advanced hepatocellular carcinoma (HCC): A biomarker approach using circulating cell-free DNA (cfDNA) (atezolizumab plus bevacizumab vs. lenvatinib vs. sorafenib)

Abstract Background: With the emergence of new immune checkpoint inhibitors and tyrosine kinase inhibitors for advanced HCC, determining the optimal first-line therapy is challenging due to the lack of precise biomarkers. This underscores the urgent need for biomarker-based strategies for treatment decisions. Methods: This prospective biomarker study analyzed plasma cfDNA from 267 advanced HCC patients treated with first-line sorafenib (S) (n=212), lenvatinib (L) (n=32), or atezolizumab plus bevacizumab (AteBev) (n=23). Low-depth whole-genome sequencing was performed on cfDNA to identify genome-wide copy number alterations (CNAs). The 'I-score' was established to quantify genomic instability, calculated as the sum of absolute Z-scores of sequenced reads for each chromosome. The treatment outcomes of the three different therapies were compared based on the I-score. The I-score cutoff to differentiate high from low genomic instability was set the level separating progressive disease (PD) from non-PD at the first response evaluation. Results: Patient demographics and disease characteristics were predominantly male (90. 6%) with a median age of 59 years. The majority had an ECOG performance status of 1 (55. 4%), HBV etiology (81. 3%), and were classified as Child-Pugh class A (88. 8%), with most presenting at BCLC stage C (95. 1%). The high I-score group (n=164) exhibited a higher prevalence of macrovascular invasion (56. 8% vs. 24. 6%; p0. 001), lymph node metastasis (39. 6% vs. 27. 2%; p=0. 038), and bone metastasis (14. 6% vs. 4. 9%; p=0. 012) but lower peritoneal metastasis (5. 5% vs. 12. 6%; p=0. 039) compared to the low I-score group (n=103). The high I-score group showed a worse disease control rate (50. 6% vs. 74. 8%; p0. 001), shorter time to progression (TTP) (median, 2. 1 vs. 4. 3 months; p0. 001), and poorer overall survival (OS) (6. 0 vs. 18. 5 months; p0. 001) compared to the low I-score group. Treatment efficacy varied within the high I-score group, with AteBev showing the most favorable outcomes: the median TTP was 6. 0 months for AteBev vs. 4. 1 months for L vs. 1. 8 months for S (p=0. 005), and the median OS was not reached for AteBev vs. 7. 6 months for L vs. 5. 4 months for S (p=0. 036). However, in the low I-score group, there were no significant differences in TTP and OS among the treatment types: the median TTP was 9. 4 months for AteBev vs. 3. 9 months for L vs. 4. 2 months for S (p=0. 665), and the median OS was not reached for AteBev vs. 9. 9 months for L vs. 18. 8 months for S (p=0. 412). Conclusions: The I-score, reflecting genome-wide CNA in cfDNA, has potential as a prognostic biomarker for outcomes in advanced HCC patients receiving first-line treatments. It may be instrumental in guiding the selection of the most appropriate therapy. Citation Format: Sook Ryun Park, Bo Hyun Kim, Ye-eun Lee, Ju Hyun Shim, Danbi Lee, Jonggi Choi, Eun-Hae Cho. Optimizing first-line systemic therapy selection in advanced hepatocellular carcinoma (HCC): A biomarker approach using circulating cell-free DNA (cfDNA) (atezolizumab plus bevacizumab vs. lenvatinib vs. sorafenib) abstract. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 1 (Regular Abstracts) ; 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84 (6Suppl): Abstract nr 984.