BCL-2 AND KI-67 AS INDEPENDENT PREDICTORS OF POOR-RISK IPI GROUP OF PATIENTS WITH DIFFUSE LARGE B-CELL LYMPHOMA
(Abstract release date: 05/18/17)
EHA Library. Suljovic Hadzimesic E. 05/18/17; 182792; PB2078
Emina Suljovic Hadzimesic
Contributions
Contributions
Abstract
Abstract: PB2078
Type: Publication Only
Background
Diffuse large B cell lymphoma (DLBCL) is heterogeneous disease ina terms of clinical behaviour, morphology, phenotype and genetics. Gene expression profiling has made a distinction between two entities germinal center B-phenotype (GCB), activated B-center phenotype (ABC). Use of immunohistochemical algorithms for identification of these phenotypes has been translated into clinically feasible approach defining groups as GCB, non- GCB. These algorithms do not provide completely accurate prognostic information so the International Prognostic Index (IPI) which identifies poor- and good-risk patients with diffuse large B cell lymphoma (DLBCL) is still part of all current diagnostic guidelines; however, the majority of patients have an intermediate IPI, with an uncertain prognosis.
Aims
In this study, we investigated the impact of bcl-2, bcl-6, CD10, MUM1 and Ki-67 on IPI as well as impact of GCB and non-GCB subclassification according to Hans and Muris algorithm on IPI risk stratification.
Methods
We have analyzed 50 patients with DLBCL for the expression of bcl-2, bcl-6, CD10, MUM1 and Ki-67 Patients were divided into two groups, the non-GCB, GCB group or favorable group 1 and unfavorable group 2, according to Hans's algorithm and Muris's algorithm. Clinical-pathological, biochemical parameters of disease have been correlated with subgroups of DLBCL and biomarkers individually. The impact of the expression of bcl-2, bcl-6, CD10, MUM1 and Ki67 on IPI-highest score in multiple regression analysis, afterwards in regression equation and variance analyse
Results
Group with GCB phenotype (defined by expression of bcl-2, bcl-6, CD10 and MUM1) according to Hans’s and Muris's algorithm showed positive correlation with good-risk patients identified by IPI. Multiple regressional analysis proved impact of biomarkers on IPI. Following this analysis, bcl2 i Ki67 are independent predictors of poor-risk IPI group of patients, (bcl-2: p 0,0107, Ki67: p 0,0377). The value of F-ratio 2,9845 proves that there is a linear connection between models including all variables bcl-2, bcl-6, CD10,MUM1 and variable depended on the value (IPI)(p 0,0210). The mutual impact of bcl-2, bcl-6, MUM1, Ki67 is significantly related to poor-risk IPI patients
Conclusion
Multiple regressional analysis proved impact of biomarkers on IPI. Ki67 and bcl-2 are independent predictors of poor-risk IPI group of patients. Sequential addition of bcl-2 expression, Ki67 and GCB phenotype into the IPI significantly improves risk stratification in DLBCL. These finding can be part of treatment strategies that should be considered in future trials.
Session topic: 18. Non-Hodgkin & Hodgkin lymphoma - Biology
Keyword(s): International prognostic index, DLBCL, BCL2, Ki-67
Abstract: PB2078
Type: Publication Only
Background
Diffuse large B cell lymphoma (DLBCL) is heterogeneous disease ina terms of clinical behaviour, morphology, phenotype and genetics. Gene expression profiling has made a distinction between two entities germinal center B-phenotype (GCB), activated B-center phenotype (ABC). Use of immunohistochemical algorithms for identification of these phenotypes has been translated into clinically feasible approach defining groups as GCB, non- GCB. These algorithms do not provide completely accurate prognostic information so the International Prognostic Index (IPI) which identifies poor- and good-risk patients with diffuse large B cell lymphoma (DLBCL) is still part of all current diagnostic guidelines; however, the majority of patients have an intermediate IPI, with an uncertain prognosis.
Aims
In this study, we investigated the impact of bcl-2, bcl-6, CD10, MUM1 and Ki-67 on IPI as well as impact of GCB and non-GCB subclassification according to Hans and Muris algorithm on IPI risk stratification.
Methods
We have analyzed 50 patients with DLBCL for the expression of bcl-2, bcl-6, CD10, MUM1 and Ki-67 Patients were divided into two groups, the non-GCB, GCB group or favorable group 1 and unfavorable group 2, according to Hans's algorithm and Muris's algorithm. Clinical-pathological, biochemical parameters of disease have been correlated with subgroups of DLBCL and biomarkers individually. The impact of the expression of bcl-2, bcl-6, CD10, MUM1 and Ki67 on IPI-highest score in multiple regression analysis, afterwards in regression equation and variance analyse
Results
Group with GCB phenotype (defined by expression of bcl-2, bcl-6, CD10 and MUM1) according to Hans’s and Muris's algorithm showed positive correlation with good-risk patients identified by IPI. Multiple regressional analysis proved impact of biomarkers on IPI. Following this analysis, bcl2 i Ki67 are independent predictors of poor-risk IPI group of patients, (bcl-2: p 0,0107, Ki67: p 0,0377). The value of F-ratio 2,9845 proves that there is a linear connection between models including all variables bcl-2, bcl-6, CD10,MUM1 and variable depended on the value (IPI)(p 0,0210). The mutual impact of bcl-2, bcl-6, MUM1, Ki67 is significantly related to poor-risk IPI patients
Conclusion
Multiple regressional analysis proved impact of biomarkers on IPI. Ki67 and bcl-2 are independent predictors of poor-risk IPI group of patients. Sequential addition of bcl-2 expression, Ki67 and GCB phenotype into the IPI significantly improves risk stratification in DLBCL. These finding can be part of treatment strategies that should be considered in future trials.
Session topic: 18. Non-Hodgkin & Hodgkin lymphoma - Biology
Keyword(s): International prognostic index, DLBCL, BCL2, Ki-67
{{ help_message }}
{{filter}}