Inferior prognosis of gastric involvement in patients with gastrointestinal Burkitt Lymphoma

Deep learning model for predicting the survival of patients with primary gastrointestinal lymphoma based on the SEER database and a multicentre external validation cohort

PURPOSE

Due to the rarity of primary gastrointestinal lymphoma (PGIL), the prognostic factors and optimal management of PGIL have not been clearly defined. We aimed to establish prognostic models using a deep learning algorithm for survival prediction.

METHODS

We collected 11,168 PGIL patients from the Surveillance, Epidemiology, and End Results (SEER) database to form the training and test cohorts. At the same time, we collected 82 PGIL patients from three medical centres to form the external validation cohort. We constructed a Cox proportional hazards (CoxPH) model, random survival forest (RSF) model, and neural multitask logistic regression (DeepSurv) model to predict PGIL patients' overall survival (OS).

RESULTS

The 1-, 3-, 5-, and 10-year OS rates of PGIL patients in the SEER database were 77.1%, 69.4%, 63.7%, and 50.3%, respectively. The RSF model based on all variables showed that the top three most important variables for predicting OS were age, histological type, and chemotherapy. The independent risk factors for PGIL patient prognosis included sex, age, race, primary site, Ann Arbor stage, histological type, symptom, radiotherapy, and chemotherapy, according to the Lasso regression analysis. Using these factors, we built the CoxPH and DeepSurv models. The DeepSurv model's C-index values were 0.760 in the training cohort, 0.742 in the test cohort, and 0.707 in the external validation cohort, which demonstrated that the DeepSurv model performed better compared to the RSF model (0.728) and the CoxPH model (0.724). The DeepSurv model accurately predicted 1-, 3-, 5- and 10-year OS. Both calibration curves and decision curve analysis curves demonstrated the superior performance of the DeepSurv model. We developed the DeepSurv model as an online web calculator for survival prediction, which can be accessed at http://124.222.228.112:8501/ .

CONCLUSIONS

This DeepSurv model with external validation is superior to previous studies in predicting short-term and long-term survival and can help us make better-individualized decisions for PGIL patients.

Inferior prognosis of gastric involvement in patients with gastrointestinal Burkitt Lymphoma

Due to limited information reported on the clinical characteristics and outcomes of Burkitt lymphoma (BL) patients with gastrointestinal (GI) involvement, here we used the Surveillance, Epidemiology, and End Results (SEER) database to perform our study in a population‐based scale. Extranodal GI involvement was categorized into gastric and intestinal primary sites. A total of 477 BL patients with GI involvement extracted from the SEER database between 2004 and 2015 were included in this study, 112 (23.5%) with the stomach and 365 (76.5%) with the intestine. Our study demonstrated that gastric involvement, older age, male gender, black race, advanced‐stage III/IV, no‐chemotherapy, and earlier years of diagnosis were associated with a significantly worse overall survival (OS) in GI BL patients after adjustment in multivariate analysis, whereas marital status did not significantly influence OS. Notably, BL Patients with gastric involvement had a significantly inferior 5‐year OS in both univariate and multivariate analysis, as compared to those with intestinal involvement (37.8% vs. 70.2%; Univariate: HR = 2.637, P < .001; Multivariate: HR = 1.489, P = .016). In subgroup analysis, we demonstrated that gastric BL patients had a consistently worse OS than intestinal patients regardless of gender, clinical stage and year of diagnosis. Hopefully, with the advances in modern therapy, improved survival has been found in BL patients with GI involvement as a whole, specifically those with gastric involvement (HR = 0.529, P = .011) in recent years of diagnosis. In conclusion, despite the improved survival achieved in recent years, the prognosis of BL patients with gastric involvement is still poor. Novel personalized therapies and better access to intensive care remain to be needed.