Project Supported by Higher Education Institutions, 2016 - 2017
Aim: Diffuse large B-cell lymphoma is a biologically heterogeneous disease that can be cured in less than half of patients. In the gene expression profiling studies performed by cDNA microarray method, diffuse large B-cell lymphomas were divided into three different subgroups: germinal center B cell-like, active B cell-like and Type 3 (unclassified). Among these subgroups, patients with germinal center B cell-like phenotype had more favorable clinical course. New targeted treatment regimens that have been recently reported are effective in cases of active B cell-like phenotype. Due to financial and technical limitations, the use of gene expression profiling technology for routine clinical purposes is challenging. In an attempt to translate gene expression profiling classification into a manageable set of measurable proteins, several algorithms suggesting various immunohistochemical stains have been proposed in recent years. Among these, the Hans algorithm is the most commonly used in the daily practice.
The aim of our study was to investigate the extend to which the germinal center B-cell immunohistochemical markers GCET1, HGAL and LMO2 reflected the follicular center cell phenotype in the follicular lymphoma case group. We also aim to demonstrate the distribution of these antigens in the diffuse large B-cell lymphoma case group and to discuss their role in determining the germinal center cell phenotype and their contribution in the context of the Hans algorithm.
Materials and methods: We retrieved 40 follicular lymphoma cases consisting of 20 high-grade, 20 low-grade and 65 diffuse large B cell lymphoma, not otherwise spesified cases from our archive. We reviewed all the archive glass slides, revised the diagnosis of the cases and selected a proper paraffin block for immunohistochemical analysis for each case. Immunohistochemical studies were performed using GCET1, HGAL and LMO2 antibodies.
Results: In the follicular lymphoma case group, GCET1, HGAL and LMO2 were positive in 85%, 77.5% and 100% of the cases, respectively. Three of the 4 CD10 negative cases were positive with one or more of these three antigens. Similarly the cases negative for BCL2 (n=6) were all positive for GCET1, HGAL and LMO2.
In the diffuse large B-cell lymphoma case group, GCET1, HGAL and LMO2 were positive in 35.4%, 53.8% and 63.1% of the cases, respectively. In the cases with germinal center
B-cell phenotype according the Hans classification, GCET1, HGAL and LMO2 were found to be 88%, 84% and 100% positive, respectively. In the cases with non-germinal center B-cell phenotype according the Hans classification, GCET1, HGAL and LMO2 were positive in 2.5%, 35% and 40% of the cases, respectively. The expression difference of all three antigens between these subgroups was statistically significant (p<0.001). In the CD10 positive and MUM1 positive cases only, the spesificity and positive predictive power of GCET1 in determining germinal center B-cell phenotype were 100%. HGAL and LMO2 immunoexpression had false positivity of 10.5% and 9.1%, respectively. From these findings, other groups of cases were evaluated according to the expression of these antigens. Nine of the 20 CD10 negative, BCL6 positive, MUM1 positive cases and 4 of the 5 CD10, BCL6 and MUM1 negative cases, formerly classified in non-germinal center B-cell group according the Hans algorithm could join the germinal center B-cell group.
Conclusion: GCET1, HGAL and LMO2 are highly sensitive markers for determining follicular center cell phenotype. It would be helpful to add these three markers to the immunohistochemical panel in the diagnosis of follicular lymphoma especially in CD10 negative cases. In order to properly classify diffuse large B-cell lymphomas, the use of these three markers or other immunohistochemical algorithms containing these can be recommended, especially for cases exhibiting CD10 negative, BCL6 positive, MUM1 positive and triple negative immun profile.