Okechinyere Achilonu ^1* George Obaido ² Blessing Ogbuokiri ³ Kehinde Aruleba ⁴ Eustasius Musenge ¹ June Fabian ⁵

• ¹ University of the Witwatersrand, Johannesburg, South Africa
• ² Other, Berkeley, United States
• ³ York University, Toronto, Ontario, Canada
• ⁴ University of Leicester, Leicester, East Midlands, United Kingdom
• ⁵ Wits University Donald Gordon Medical Centre, Johannesburg, South Africa

In South Africa, between 1966 and 2014, there were three kidney transplant eras defined by evolving access to certain immunosuppressive therapies defined as \textit{Pre-CYA }(before availability of cyclosporine), \textit{CYA} (when cyclosporine became available), and \textit{New-Gen} (availability of tacrolimus and mycophenolic acid). As such, factors influencing kidney graft failure may vary across these eras. Therefore, evaluating the consistency and reproducibility of models developed to study these variations using machine learning (ML) algorithms could enhance our understanding of post-transplant graft survival dynamics across these three eras. This study explored the effectiveness of nine ML algorithms in predicting 10-year graft survival across the three eras. We developed and internally validated these algorithms using data spanning the specified eras. The predictive performance of these algorithms was assessed using the area under the curve (AUC) of the receiver operating characteristics curve (ROC), supported by other evaluation metrics. We employed local interpretable model-agnostic explanations to provide detailed interpretations of individual model predictions \textcolor{blue}{and used permutation importance to assess global feature importance across each era.} Overall, the proportion of graft failure decreased from 41.5\% in the \textit{Pre-CYA} era to 15.1\% in the \textit{New-Gen} era. Our best-performing model across the three eras demonstrated high predictive accuracy. Notably, the ensemble models, particularly the Extra Trees model, emerged as standout performers, consistently achieving high AUC scores of 0.95, 0.95, and 0.97 across the eras. \textcolor{blue}{This indicates that the models achieved high consistency and reproducibility in predicting graft survival outcomes}. Among the features evaluated, recipient age and donor age were the only features consistently influencing graft failure throughout these eras, \textcolor{blue}{while features such as glomerular filtration rate and recipient ethnicity showed high importance in specific eras}, resulting in relatively poor historical transportability of the best model. Our study emphasises the significance of analysing post-kidney transplant outcomes and identifying era-specific factors mitigating graft failure. The proposed framework can serve as a foundation for future research and assist physicians in identifying patients at risk of graft failure.