Kinetics of Regulatory T-Cells Are Highly Influenced by Treatment In Patients with Gastric MALT Lymphoma

Tumor microenvironment has been shown to play a critical role in the biology and treatment response of some types of non Hodgkin's lymphomas. The significance of the presence of Treg cells (FOXP3+) in MALT lymphomas of the stomach remains unknown. The main aim of this study was to analyze by immunohistochemistry the presence of Treg cells in the microenviroment of gastric MALT (gMALT) lymphomas at the time of diagnosis and during follow-up after treatment, and to evaluate their possible impact in outcome.

Patients were included in the study if they had gMALT lymphoma diagnosed according to the 2008 WHO criteria and diagnostic paraffin-embedded blocks were available for review. Thirty-three patients met the criteria for inclusion in the study. Sections were immunostained for CD20, CD3, CD4, CD8, CD68, FOXP3, PD-1, bcl-10 and Ki67. t(11;18) was studied by FISH and/or PCR and t(1;14) by FISH. Clonality study of the B cell receptor was done according to the BIOMED-2 protocol. The number of CD20+ tumor cells and FOXP3+ infiltrating cells was quantified using a micrometric ocular (WPK 10 × mn) that has a 10 mm linear scale divided to 100 parts.

The median age was 63 y (range 32–83) with 52% being male. Stage: I in 66%, II in 25% and IV in 9%; B-symptoms in 6%. Translocation t(11;18) was present in 9 (27%) pts and 2 additional cases had strong nuclear expression of BCL-10 without t(11;18) or t(1;14). At diagnosis, the mean (± standard deviation) number of CD20+ tumor cells and FOXP3+ infiltrating cells was 680±232 cells/cm² and 30±29 cells/cm², respectively. The mean number of CD20+ tumor cells and FOXP3+ infiltrating cells was similar between gMALT with or without t(11;18).

Number of treatments analyzed was 37 (treatment not available in 2 pts and 5 pts received more than 1 treatment). Treatment regimens included: eradication therapy (n=11); interferon+ribavirin (n=1); single or combined agent chemotherapy without rituximab (n=5); rituximab alone or CHOP-like with rituximab (n=4); fludarabine (n=8); fludarabine or bendamustine with rituximab (n=8). The first response evaluation showed an overall clinical response rate of 84% (CR 76%) with persistence of residual disease by histology or molecular (IGHV) studies in 31% and 56%, respectively.

Overall, the mean (±SD) CD20+ tumor cells and FOXP3+ infiltrating cells was significantly reduced after treatment (730±208 vs 285±343, p<0.0001; 34±41 vs 21±36, p=0.016, respectively). The mean number of CD20+ tumor cells and FOXP3+ infiltrating cells decreased in responders (722 vs 210, p<0.001 and 37 vs 24, p=0.063, respectively) but not in non responders (p=0.279 and p=0.249). Treatment with fludarabine or bendamustine with or without rituximab induced a quick and profound depletion in the number of CD20+ cells in comparison with the other treatments (8-fold vs 2-fold reduction), but FOXP3+cell remained relatively unchanged (1.2-fold vs 2-fold reduction). The monitoring of CD20+ and FOXP3+ during follow-up and its value are under analysis (follow-up: median 4 years (range 1–8)).

Based on these preliminary findings, there is a significant change of CD20+ cells and moderate reduction of Treg cells in responding patients after treatment for gastric MALT lymphoma. Interestingly, regimens with fludarabine or bendamustine with or without rituximab result in quicker and deeper reduction of CD20+ tumor cells while infiltrating FOXP3+ cells are only moderately reduced.

No relevant conflicts of interest to declare.