The predictive value of serum bilirubin in the post-liver transplant de novo malignancy: A data mining approach

Abstract

Post-transplantation de novo malignancy in immunosuppressed organ recipients has become a major source of death, making early cancer surveillance through diagnosis and detection important in drastically improving survival rates. The focus of this work is on predicting de novo malignancy after liver transplantation using machine learning. Patients were chosen as those having developed malignancy after liver transplantation, with no history of cancer prior to transplantation, with donors being cancer-free as well. We analyzed a large volume of patients with post-transplant malignancy from the US Organ Procurement and Transplantation Network (OPTN). Several popular data mining methods were employed to characterize de novo malignancy after liver transplantation. Recipient’s bilirubin, creatinine, weight, gender, number of days in wait on the transplant list, Epstein Barr Virus (EBV), International normalized ratio (INR), and ascites were found to be among the most important factors affecting post-liver-transplantation de novo malignancy occurrence.

Keywords

• Organ transplantation
• Liver transplantation
• De novo malignancy
• Machine learning
• Data mining
• Random forest