Incidence, survival of primary colorectal lymphoma over 40 years and a nomogram based on SEER database and an independent validation

Background: The purpose of this study was to clarify trends in the incidence, survival, and prognosis of patients with primary colorectal lymphoma (PCL) and to establish a nomogram for primary colorectal diffuse large B-cell lymphoma (PC-DLBCL). Methods: Patients diagnosed with PCL from 1975 to 2018 in the Surveillance, Epidemiology, and End Results (SEER) database were included for assessment of temporal trends in incidence. Clinical data of patients from 1975 to 2015 were extracted for survival and prognostic analysis. Patients diagnosed with PC-DLBCL from 2009 to 2017 in our department were designated as an independent validation dataset. Univariate and multivariate Cox regression analyses were used to identify independent prognostic factors affecting overall survival (OS) and disease-specific survival (DSS). A nomogram was constructed specifically for PC-DLBCL and evaluated by C-index, calibration curves, receiver operating characteristic (ROC) curves, and decision-curve analysis (DCA) plots. Results: The study included a total of 2644 patients diagnosed with PCL. From 1975 to 2018, the incidence of PCL increased from 1.13/1000000(adjusted to the US standard population in 2000) to 2.54/1000000, With an obvious growth trend (annual percentage change = 2.91, 95% confidence interval 2.29–3.94, P < 0.05). The incidence of PCL increased steadily before 2005 and decreased annually since then. The survival outcome of PCL improved over four decades (P < 0.0001). Both the OS and DSS were affected independently by age, marital status, pathological subtype, Ann Arbor stage, surgery and chemotherapy. Besides, the nomogram for PC-DLBCL was constructed to predict 1-year, 3-year, 5-year, and 10-year OS and DSS with robust and accurate performance. Moreover, risk stratification based on the nomogram score was accomplished. Conclusion: The incidence of PCL presented with a parabola-like trend, along with improved survival outcomes over the past 40 years. The nomogram for PC-DLBCL was capable of predicting the 1-, 3-, 5- and 10-year OS and DSS with high accuracy and help identify high-risk patients who might need more intensive medical care.