Neoadjuvant immunotherapy in hepatocellular carcinoma: A systematic review and metaanalysis

e16232 Background: The addition of immunotherapy in neoadjuvant hepatocellular carcinoma (HCC) has been studied recently in early phase clinical trials due to high rate of recurrence after surgery. We aim to systematically review the evidence of efficacy and safety of neoadjuvant immunotherapy treatment. Methods: We carried out a systematic search on MEDLINE, Cochrane, and ASCO meetings library. Screening was done to include clinical trials with HCC patients who underwent any combination of neoadjuvant immune checkpoint inhibitors (ICIs). Quality assessment was performed using methodological index for non-randomized studies (MINORS) tool. The primary efficacy outcomes evaluated included complete pathological response (pCR), major pathological response (MPR), 1 year recurrence free survival (RFS) and overall response rate (ORR). Treatment related adverse events (TRAE) were assessed as a safety outcome. Predefined subgroup analysis was performed based on studies including of tyrosine kinase inhibitors (TKIs) in regimen. Data was analyzed using R version 4.2.1. Results: 5 phase I/II clinical trials were included consisting of 96 patients. Trials administered neoadjuvant immunotherapy agents nivolumab (with TKI cabozantinib) (N = 1), camrelizumab (with TKI apatinib) (N = 1), cemiplimab (N = 1) and nivolumab plus ipilimumab (N = 2). 78(81.2%) of patients were males. 56(59.3%) of patients had viral etiology. 14(14.5%) patients had surgeries delayed or cancelled due to any cause per trial protocols. Meta-analysis of 96 patients from 5 trials showed a pooled pCR rate of 14.49% (95% CI: 6.82 to 23.93%) with low heterogeneity (I 2 = 0%, p value 0.55). Pooling the MPR resulted in an estimate of 35.30% (95% CI: 18.80% to 53.57%) with significant heterogeneity (I 2 = 58.8%, p value 0.04). Three trials with total 45 patients reported a 12-month RFS rate of 51.08% (95% CI: 40.59 to 70.67%) with significant heterogeneity (I 2 = 84.2%, p value 0.002). The overall response rate was 13.7% (95% CI: 6.9 to 22.1%) (I2 =0%, p =0.8). Overall rate of TRAE of any grade was 73.8% (95% CI: 63.1 to 82.5%) (I2 = 83.4%, p <0.001) among all trials and grade 3 adverse events was 21.21% (95% CI: 11.4 to 32.7%) (I2 = 33.8%, p = 0.159). 17(17.8%) patients had elevation in liver enzymes. On subgroup analyses, no difference in pathological outcomes, ORR and grade 3 adverse events was observed with addition of TKIs. Exploratory analysis showed no difference in pCR rates with single ICI vs combination with ipilimumab. Conclusions: The addition of ICIs in neoadjuvant settings appears to be a safe and feasible option. The addition of TKIs to neoadjuvant ICIs does not appear to affect pathological and safety outcomes. Further studies are needed to identify ICI based regimens to improve pCR rates and validate pathological outcomes in HCC after correlating with recurrence and survival.