重症新型コロナウイルス感染症における動的溶性プログラム死リガンド1の軌跡とその死亡率との関連の探索的特徴付け

Persistent immune checkpoint activation is a recognized feature of critical coronavirus disease 2019 (COVID-19). However, the temporal behavior and clinical utility of soluble programmed death-ligand 1 (sPD-L1) remain unclear. We investigated the longitudinal changes in sPD-L1, its relationship with organ dysfunction markers, and their prognostic value when combined with machine learning (ML) models.

In this single-center observational study, we included 40 adults with severe COVID-19 pneumonia admitted to the intensive care unit (ICU) (April 2021-December 2022), and 23 healthy volunteers as controls. We measured plasma sPD-L1 on ICU day 1, 5, 7, 14, and 21. Routine biochemistry, complete blood counts, and arterial blood gas analyses were conducted in parallel. Cox regression was used to identify independent predictors of hospital mortality, the primary outcome. Eight ML classifiers were trained using admission variables and sPD-L1 levels from ICU day 1, 5, and 7. Discrimination was assessed using stratified fivefold cross-validation, and feature importance was evaluated using Shapley Additive Explanations (SHAP).

Of 40 patients, 10 died during hospitalization. Overall, sPD-L1 levels declined during the ICU stay but remained persistently high in non-survivors. ICU day 5 and 7 values differed significantly between survivors and non-survivors (p = 0.023 and 0.001, respectively). In multivariable Cox analysis, ICU day 7 sPD-L1 levels and arterial lactate levels on admission independently predicted mortality. ICU day 7 sPD-L1 levels correlated positively with creatinine, C-reactive protein, and fibrinogen levels (all p < 0.05) in cross-sectional correlation analyses. Among ML models, the support vector machine achieved the highest discriminative accuracy (mean area under the curve = 0.917). ICU day 5 sPD-L1 was designated as the primary predictor of mortality based on SHAP analysis, with lactate contributing minimally.

Sustained sPD-L1 elevation during the first ICU week is strongly associated with early organ dysfunction and independently predicts death in critical COVID-19. Incorporating serial sPD-L1 measurements into bedside ML models significantly enhances risk discrimination. These findings support sPD-L1 as an integrative biomarker of the immune-renal-coagulation interplay, warranting validation in larger multicenter cohorts and exploration as a potential companion marker for immune-modulatory interventions.