Impact of the geographic redistribution of MMaT-3 policy on regional and socioeconomic disparities in hepatocellular carcinoma transplantation.

Abstract

e23242 Background: The MMaT-3 (Median MELD at Transplant minus 3) policy was implemented in 2019 to reduce geographic variation in liver transplant access by prioritizing candidates whose MELD scores approached local transplant thresholds. Although intended to promote regional equity, its impact on geographic performance and socioeconomic disparities among hepatocellular carcinoma (HCC) candidates remains unclear. Methods: Using UNOS/OPTN registry data, we conducted a retrospective analysis of adult HCC candidates listed during symmetric 21-month periods before (n = 4,630) and after (n = 4,049) MMaT-3 implementation. Time to deceased donor liver transplantation (DDLT) was assessed using adjusted restricted mean failure time (RMFT). Policy-associated changes in wait time were compared across geographic regions, race/ethnicity, and insurance status, using White race, private insurance, and the Southwest as reference groups. Results: Overall, MMaT-3 was associated with reduced system efficiency. Adjusted RMFT increased for White candidates (+25.5 days, p < 0.001) and privately insured candidates (+23.7 days, p < 0.001) compared with the pre-policy period. Regional effects were heterogeneous. Geographic disparities narrowed primarily due to worsening outcomes in historically high-performing regions rather than improvements in the most disadvantaged areas. New England (+68.3 to +78.7 days, p < 0.001) and New York (+61.1 to +52.7 days, p < 0.001) did not significantly improve and remained the most disadvantaged regions. The Southeast (approximately −160 to −60 days; policy-associated prolongation > 100 days, p < 0.001) and Midwest (approximately −80 to −20 days; prolongation ~60 days, p < 0.001) experienced substantial deterioration. The Mid-Atlantic (−29.8 to +14.4 days, p < 0.01) and Texas (−3.9 to +51.6 days, p < 0.001) shifted from neutral or favorable to significantly longer wait times. Despite regional stagnation, sociodemographic disparities improved. Pre-policy disadvantages among Black (+15.9 days) and Hispanic (+27.8 days) candidates were eliminated post-policy ( p ≤0.006), as were disparities among publicly insured candidates (+16.8 to +1.6 days, p < 0.001). Among Asian candidates, a substantial pre-policy disadvantage (+65.8 days) was reduced, though a residual disparity persisted post-policy (+39.7 days, p < 0.001). Conclusions: MMaT-3 was associated with reductions in racial and insurance-based disparities among HCC candidates but achieved geographic equity largely through deterioration in previously advantaged regions, with limited benefit for the most disadvantaged areas. These findings highlight trade-offs inherent in spatial allocation reforms and inform subsequent policy modifications.