Woo-Hyoung Kang, Shin Hwang, Deok-Bog Moon, Ki-Hun Kim, Chul-Soo Ahn, Tae-Yong Ha, Gi-Won Song, Dong-Hwan Jung, Gil-Chun Park, Young-In Yoon, Byeong-Gon Na, Sang-Hoon Kim, Sung-Min Kim, Sung-Gyu Lee
ABO-incompatible (ABOi) living donor liver transplantation (LDLT) requires rituximab-based desensitization and intensified early immunosuppression, and whether this promotes hepatocellular carcinoma (HCC) recurrence remains controversial. We hypothesized that this effect depends on tumor burden, assessed morphologically (Milan criteria) and biologically (ADV score). In this single-center study, we analyzed 1,573 adults transplanted for HCC (ABOi, n = 316; ABO-compatible [ABOc], n = 1,257); the primary endpoint was recurrence-free survival (RFS). Crude recurrence was similar (21.8% vs. 20.7%; P = 0.64) despite a lower tumor burden in ABOi recipients. In a propensity score-matched cohort, RFS did not differ overall (hazard ratio 1.11, 95% CI 0.82-1.48), but beyond Milan recurrence was higher in ABOi recipients (53.7% vs. 38.5%; adjusted HR 1.45, 95% CI 1.00-2.09), whereas within Milan it was identical (13.3% vs. 13.4%). The same pattern held for tumor biology: recurrence was higher in ABOi only at ADV ≥5log (64.0% vs. 44.6%; P = 0.022). Higher early tacrolimus exposure in ABOi recipients was associated with recurrence. Overall survival was similar (P = 0.91). ABO-incompatibility did not compromise survival. Recurrence was higher in ABOi recipients with a high tumor burden by either measure, although the formal interaction tests were not significant (P = 0.23 and P = 0.088). These exploratory subgroup findings are hypothesis-generating and support closer early surveillance.