Immunosuppression Efficacy via Nonlinear Modeling and Probabilistic Analysis of Human Leukocyte Antigen Match and...

Human Leukocyte Antigen (HLA) compatibility is a key driver of kidney allograft outcomes, yet how matching interacts with immunosuppressant dosing remains unclear. This study evaluates whether low-resolution HLA match fractions affect mean dosing and quantified the joint impact of match quality and dose on rejection risk.

In 519 transplants grouped by HLA match (0-1.0 in 7 strata), full data completeness and variance homogeneity is confirmed, followed by the usage of one-way ANOVA to compare mean doses. Ridge-penalized logistic regression models was fitted at each match level to predict rejection probability across low- and high-dose regimens, deriving absolute and relative risk reductions.

ANOVA showed no significant dosing differences by match (F_6,512 = 0.93, p = 0.47). Logistic models revealed that Absolute Risk Reduction rose from ~0.75% at zero match to ~1.25% at perfect match, while mid-range match groups (0.2-0.6) yielded the highest total prevented rejections under high-dose therapy.

Uniform baseline dosing eliminates bias, allowing clear attribution of benefit to immunologic compatibility. These findings endorse a risk-tailored immunosuppression approach intensifying doses where match is low and potentially reducing them in highly compatible transplants and identify match ranges that maximize rejection prevention.