Cure models, survival probabilities, and solid organ transplantation for patients with colorectal cancer

Eric A Engels,Soutrik Mandal,Douglas A Corley,Christopher D Blosser,Allyson Hart,Charles F Lynch,Baozhen Qiao,Karen S Pawlish,Gregory Haber,Kelly J Yu,Ruth M Pfeiffer

DOI: https://doi.org/10.1016/j.ajt.2024.08.018

2024-08-23

Abstract:A previous cancer diagnosis can preclude patients from consideration for solid organ transplantation. Statistical models may improve candidate selection. We fitted statistical cure models and estimated five-year cancer-specific survival (5yCSS) for colorectal cancer patients in the United States using registry data. The median cure probability at cancer diagnosis for patients in the general population was 0.67. Among 956 colorectal cancer patients who underwent solid organ transplantation, the median time since diagnosis was 6.3 years and the median 5yCSS at transplantation was 0.96. Patients with a 5yCSS below 0.90 had increased posttransplant cancer-specific mortality (hazard ratio 3.31, 95% confidence interval 1.52-7.21). Compared with recently published guidelines, our models suggested shorter wait times for some groups of colorectal cancer patients (e.g., stage IIA cancers) and longer wait times for others (stages IIB, IIIB, IIIC, IV). In conclusion, colorectal cancer patients undergoing solid organ transplantation had excellent prognoses, reflecting selection incorporating existing guidelines and clinical judgement. Nonetheless, 5yCSS probabilities estimated from cure models offer additional prognostic information for patients considered for transplantation and identify situations where current guidelines might be revised. We developed a web-based tool for clinicians to calculate 5yCSS probabilities for use in transplant evaluation for individual colorectal cancer patients (https://dceg.cancer.gov/tools/risk-assessment/calculator-of-colorectal-cancer-survival-probability).