Subgrouping and outcome prediction of diffuse large B-cell lymphoma by immunohistochemistry / 中华病理学杂志

ABSTRACT

<p><b>OBJECTIVE</b>To categorize diffuse large B-cell lymphoma (DLBCL) into germinal center B cell-like (GCB) and non-germinal center B cell-like (non-GCB) subgroups by immunohistochemistry; and to investigate the underlying prognostic significance.</p><p><b>METHODS</b>Immunohistochemical study for CD10, bcl-6 and MUM1 was performed on 133 cases of DLBCL. The cases were then categorized into GCB and non-GCB subgroups. The 5-year overall survival and 5-year progression-free survival rates were compared between the GCB and non-GCB groups, and among the cases with different immunohistochemical expression or with different IPI.</p><p><b>RESULTS</b>Amongst the 133 case studied, CD10 was expressed in 33.1%, while bcl-6 was positive in 34.6% and MUM1 in 45.1%. CD10 expression had a favorable impact on 5-year overall survival (P=0.041) and 5-year progression-free survival (P=0.031). On the other hand, bcl-6 expression had a favor able impact on 5-year progression-free survival (P=0.044). Expression of MUM1 carried an adverse effect on 5-year overall survival (P=0.031) and 5-year progression-free survival (P=0.028). GCB immunophenotype was demonstrated in 40.6% of the cases, while 59.4% showed a non-GCB profile. GCB DLBCL had a significantly longer 5-year overall survival (P=0.004) and 5-year progression-free survival (P=0.003), as compared with the non-GCB group. When dividing the cases into two groups according to their IPI score (IPI=0 to 1 and IPI=2 to 5), it turned out that the 5-year overall and progression-free survival rates of the GCB group were significantly higher than those of the non-GCB group (P=0.019 and 0.014 respectively in cases with IPI of 0 to 1 and P=0.006 and 0.009 respectively in cases with IPI of 2 to 5). The non-GCB cases with a IPI of 2 to 5 had the poorest prognosis.</p><p><b>CONCLUSION</b>DLBCL subgrouping by immunohistochemistry and analysis of the subgrouping with IPI is feasible and useful in predicting clinical outcome.</p>