Single nucleotide variants in immune-response genes and the tumor microenvironment composition predict progression of mantle cell lymphoma
Guilherme Rossi Assis-Mendonça,André Fattori,Rafael Malagoli Rocha,Gustavo Jacob Lourenço,Márcia Torresan Delamain,Suely Nonogaki,Vladmir Cláudio Cordeiro de Lima,Gisele Wally Braga Colleoni,Cármino Antonio de Souza,Fernando Augusto Soares,Carmen Silvia Passos Lima,José Vassallo
DOI: https://doi.org/10.1186/s12885-021-07891-9
IF: 4.638
2021-03-01
BMC Cancer
Abstract:Abstract Background There is evidence to consider that the tumor microenvironment (TME) composition associates with antitumor immune response, and may predict the outcome of various non-Hodgkin lymphoma subtypes. However, in the case of mantle cell lymphoma (MCL), a rare and aggressive disease, there is lacking a detailed study of the TME components, as well as an integrative approach among them in patients’ samples. Also, from the genetic point of view, it is known that single nucleotide variants (SNVs) in immune-response genes are among important regulators of immunity. At present, it is uncertain whether SNVs in candidate immune-response genes and the TME composition are able to alter the prognosis in MCL. Methods We assessed a detailed TME composition in 88 MCL biopsies using immunohistochemistry, which was automatically analyzed by pixel counting (Aperio system). We also genotyped SNVs located in candidate immune-response genes ( IL12A , IL2 , IL10 , TGFB1 , TGFBR1 , TGFBR2 , IL17A , IL17F ) in 95 MCL patients. We tested whether the SNVs could modulate the respective protein expression and TME composition in the tumor compartment. Finally, we proposed survival models in rituximab-treated patients, considering immunohistochemical and SNV models. Results High FOXP3/CD3 ratios ( p = 0.001), high IL17A levels ( p = 0.003) and low IL2 levels ( p = 0.03) were individual immunohistochemical predictors of poorer survival. A principal component, comprising high quantities of macrophages and high Ki-67 index, also worsened outcome ( p = 0.02). In the SNV model, the CC haplotype of IL10 ( p < 0.01), the GG genotype of IL2 rs2069762 (p = 0.02) and the AA+AG genotypes of TGFBR2 rs3087465 ( p < 0.01) were independent predictors of outcome. Finally, the GG genotype of TGFB1 rs6957 associated with lower tumor TGFβ levels ( p = 0.03) and less CD163+ macrophages ( p = 0.01), but did not modulate patients’ survival. Conclusions Our results indicate that the TME composition has relevant biological roles in MCL. In this setting, immunohistochemical detection of T-reg cells, IL17A and IL2, coupled with SNV genotyping in IL10 , TGFBR2 and IL2, may represent novel prognostic factors in this disease, following future validations.
oncology