Abstract
<title>Abstract</title> <p> <bold>Background</bold> :Pulmonary metastases frequently develop in hepatocellular carcinoma (HCC) patients. Treatment strategies remain controversial. We conducted a retrospective comparison of all-trans retinoic acid (ATRA) plus FOLFOX4 chemotherapy versus radiotherapy combined with lenvatinib and sintilimab. <bold>Methods</bold> :This study enrolled 107 HCC patients with lung metastases. The cohort was divided into two groups: RA (n=68) and RT+S+L (n=39). Inverse probability of treatment weighting (IPTW) based on propensity scores addressed baseline imbalances. Multiple imputation by chained equations (MICE) handled missing data across 20 iterations. Overall survival (OS), objective response rate (ORR), disease control rate (DCR), and adverse events were evaluated. <bold>Results</bold> :Median OS was 25 months in the RT+S+L group and 17 months in the RA group (unweighted), with the upper confidence bound not reached for either group. After IPTW weighting, the HR was 3.12 (95% CI: 1.11–8.72, P=0.030). ORR showed no significant difference: 23.0% (RT+S+L) versus 27.9% (RA), P=0.607. DCR was 74.8% (RT+S+L) versus 57.4% (RA), P=0.104. RA treatment correlated with higher gastrointestinal toxicity, whereas RT+S+L caused more liver enzyme elevations. <bold>Conclusions</bold> . The RT+S+L regimen demonstrated superior OS benefits. RA preserved liver function better. Treatment selection should consider patient characteristics, liver reserve, and survival expectations. </p>