| ObjectivePrimary gastrointestinal diffuse large B-cell lymphoma(PGI-DLBCL)is a gastrointestinal manifestation of non-Hodgkin’s lymphoma that may present clinically as abdominal surgical symptoms and undergo surgery,and the impact of surgery on the long-term survival of patients with PGI-DLBCL is unclear.To evaluate the prognostic implications of surgical therapy on patients with PGI-DLBCL and to develop a prospective model to provide new perspectives and ideas for clinical practice.MethodsDiagnosis and therapy information for patients who were diagnosed with diffuse large B-cell lymphoma in the National Cancer Institute Surveillance,Epidemiology,and End Results Database(SEER database)from 1998 to 2015 and with sites of gastric,duodenal,jejunal,ileal,cecum,appendix,ascending colon,transverse colon,sigmoid colon,rectum,and recto-anal canal was obtained and Kaplan-Meier survival analysis was conducted,and prognostic elements were screened using Cox regression models,and the results were considered statistically significant at P<0.05 and presented graphically.calibration curve)and C-index for the effectiveness test.Results1.KM survival analysis was performed for cases from 1998-2002 and cases from 2003-2015 grouped by surgery or not.The p-values were less than 0.0001 for surgical treatment relative to non-surgical treatment for patients with PGI-DLBCL with primary sites in the stomach,duodenum,jejunum,ileum,and colon and rectum,and the differences were statistically significant.2.The diagnosed cases from 2003-2015 were divided into stomach group,large intestine group and small intestine group according to the primary site,and KM survival analysis was performed,and the results are shown in Figure 2 and Table 2,which Indicated that the mean survival time of patients in the stomach group t(surgical treatment)=113 months,t(non-surgical treatment)=87 months,P=0.04;the mean survival time of patients in the large intestine group t(surgical treatment)=116 months,t(non-surgical treatment)=97 months,P=0.0071;mean survival time of patients in the small intestine group t(surgical treatment)=110 months,t(non-surgical treatment)=86 months,P=0.00067,P-values are.3.Among patients with PGI-DLBCL whose primary site was the small intestine,patients with ileal DLBCL had the lowest prognostic survival risk and the highest survival rate with HR=0.639(95%CI 0.483-0.846),p=0.0024.2003-2015 case Cox multifactorial results showed that age and stage p-values were less than 0.05 in patients with PGI-DLBCL at different primary sites,which were prognostically significant influencing factors.5.In the Cox multifactorial analysis of the small bowel group,non-surgical treatment relative to surgical treatment HR=1.584(95%CI 1.205-2.083),p=0.001,which shows that surgical treatment is a significant prognostic influencing factor in patients with duodenum,jejunum and ileum.6.Among the three models established in this study,C index(stomach)=0.668287,C index(large intestine)=0.716681,and C index(small intestine)=0.6950466,which indicated that the above models performed better in internal validation and could be applied to clinical practice.Conclusion1.Surgical treatment can effectively prolong survival and improve survival rate for PGI-DLBCL patients originating in the stomach,duodenum,jejunum,ileum and colorectum;2.Age and stage are significant factors affecting the prognosis of PGI-DLBCL patients,and the overall trend is that age and stage are inversely proportional to prognosis,that is,the older and higher the stage,the worse the prognosis;3.In patients with PGI-DLBCL whose primary site is the small intestine,ileal DLBCL has the lowest prognostic survival risk and the highest survival rate;4.Surgical treatment is a significant prognostic factor for patients with PGI-DLBCL with primary sites in the duodenum,jejunum,and ileum.Unlike the current guidelines for non-surgical treatment strategies,we recommend aggressive surgical treatment for patients with primary sites in the small intestine once the diagnosis is confirmed.5.Patients with PGI-DLBCL whose primary sites are the stomach and the colon and rectum,there is no remarkable difference in the findings between surgical and non-surgical treatments in the absence of life-threatening conditions such as perforation and bleeding;however,if serious complications such as obstruction,perforation,and hemorrhage occur,surgery should be performed immediately.6.The column line graphs constructed in this study require clinical data that are easily accessible,portable and operable,and slightly less accurate,and can be applied in clinical prediction models. |