Objective:To compare the prognostic value of different stratification methods in the era of rituximab for diffuse large B-cell lymphoma;To verify whether the Kyoto prognostic index(KPI)can be more effective in identifying and assessing the prognosis of high risk patients with diffuse large B cell lymphoma(DLBCL)than international prognostic index(IPI),national comprehensive cancer network-IPI(NCCN-IPI)and grupo espa(?)ol de linfomas/transplante de médula (?)sea-IPI(GELTAMO-IPI)in the era of rituximab;To guide the prognostic model direction that can effectively identify high risk diffuse large B-cell lymphoma in the era of rituximab.Methods:The clinical data of 127 patients with DLBCL admitted to The First Hospital of Lanzhou University and the 940th Hospital of Joint Logistics Support force of the Chinese People’s Liberation Army from January 2012 to January 2021 were analyzed retrospectively.IPI、KPI、NCCN-IPI and GELTAMO-IPI re-stratified these patients according to the data at initial diagnosis and compared with the risk stratification and prognostic ability of the four prognostic indexes in DLBCL.The Kaplan-Meier method was used to analyze the progression-free survival and overall survival of the stratified groups with the four models.The Log-rank test was used for intergroup comparison.The Harrell’s C statistics were used to compare the prognostic stratification ability of four prognostic models.And descriptively analyzed the prognostic stratification of IPI,KPI,NCCN-IPI in different molecular characteristics DLBCL.Results:(1)In the untransplanted group,survival analysis showed that IPI,KPI,NCCN-IPI and GELTAMO-IPI all had prognostic stratification(P<0.05),and the 2-year OS of the high risk group was 63.2%,50%,58.3%,44.4%(n=69),and the 2-year PFS was 47.4%,40%,50%,and 33.3%(n=69),respectively;The comparison between the Log-rank test showed that only the prognosis difference between the low-intermediate risk group and the high-intermediate risk group of IPI was obvious(P<0.05),while the prognostic difference between the high-intermediate risk group and the high risk group of IPI,KPI,NCCN-IPI and GELTAMO-IPI was not obvious(P>0.05);However,the 2-year OS differences between high-intermediate risk group and high risk group of IPI,KPI,NCCN-IPI and GELTAMO-IPI were 9.5%,30%,16.7%,35.6%(n=69),respectively,and the 2-year PFS differences were-1.9%,20%,7.1%and 46.7%(n=69),respectively.Harrell’s C statistical analysis showed that the 2-year OS consistency index predicted by IPI,KPI,NCCN-IPI and GELTAMO-IPI was 0.664,0.629,0.681,0.633(n=69),respectively,the predicted 2-year PFS consistency index was 0.665,0.627,0.628,0.569(n=69),respectively,which were low accuracy.These results indicated that all stratification methods had predictive effect and the ability to identify high risk patients were not ideal,but the identification ability of KPI for high-risk patients was better than that of IPI and NCCN-IPI,and GELTAMO-IPI showed good potential in identifying high-risk patients.(2)In the transplantation group,KPI could better distinguish between the risk groups,and the prognostic stratification ability of IPI and NCCN-IPI was not as good as that of KPI;The comparison between groups showed that the prognosis of the low-intermediate risk group and the high-intermediate risk group of KPI was significantly different(P<0.05).(3)Subgroup survival analysis showed that KPI was not as good as IPI and NCCN-IPI in the ability of risk stratification and prognosis for GCB-DLBCL in PFS,but the risk stratification ability for non-GCB patients was better than that of IPI and NCCN-IPI.(4)Subgroup survival analysis showed that the risk stratification and prognosis of KPI was more in line with the actual prognosis for non-DHL/THL patients than IPI and NCCN-IPI.(5)Subgroup survival analysis showed that the risk stratification and prognostic value of KPI and NCCN-IPI for patients with TP53 mutation or without TP53 mutation was consistent and better than IPI.Conclusion:1.In the era of rituximab,IPI、KPI、NCCN-IPI and GELTAMO-IPI all have risk stratification capabilities,and the ability to identify low risk patients is comparable,but none of them can effectively identify high-risk patients;2.Compared with IPI and NCCN-IPI,KPI has stronger ability to identify high risk patients,but the overall risk stratification ability is not as good as NCCN-IPI.Compared with IPI and NCCN-IPI,KPI has better risk stratification ability for patients who received autologous hematopoietic stem cell transplantation or patients who are non-GCB or patients who are not DHL/THL.3.It can improve the risk stratification ability of the prognostic model by subdividing the original factors on the basis of IPI and incorporating new prognostic factor to IPI. |