PF529 PROGNOSTIC CORRELATION OF CELL‐OF‐ORIGIN AND MYC, BCL‐2 AND BCL‐6 STATUS IN HIV‐ASSOCIATED DIFFUSE LARGE B‐CELL LYMPHOMAS

Abstract

Background:Diffuse Large B‐cell lymphoma (DLBCL) can be classified by gene expression profiling (GEP) into 3 subtypes according to the putative Cell‐Of‐Origin (COO): the germinal center B‐cell type (GCB), the activated B‐cell type (ABC) and the unclassifiable type (UC). Immuno‐histochemistry (IHC) algorithms, being Hans's the mostly used, and NanoString (NS) 20‐genes assay can surrogate adequately GEP in COO assignment. In DLBCL HIV negative population treated with rituximab‐CHOP, GCB type is associated with a better outcome, while gene rearrangements of MYC plus BCL‐2 and/or BCL‐6 detected by fluorescence in situ hybridization (FISH) confer a worse prognosis. In HIV positive DLBCL, limited data are available on COO subtypes and genes rearrangements incidences and on their prognostic significance.Aims:To evaluate in HIV‐associated DLBCL: the proportion of COO subtypes defined by IHC and NS assay and the concordance of these methods; the prognostic impact of COO and of MYC, BCL‐2 and BCL‐6 status.Methods:We retrospectively analyzed 66 HIV positive patients (pts) with newly diagnosed DLBCL from 2000 to 2016. Histological samples were centrally reviewed for diagnosis confirmation, COO assignment according to Hans and MYC, BCL‐2 and BCL‐6 protein expression. NS Lymph2Cx assay was applied for COO identification and FISH using break‐a‐part probes for MYC, BCL‐2 and BCL‐6 gene analysis.Results:Pts characteristics at DLBCL diagnosis were: 73% male, median age 45 yrs (range: 28–83), 80% advanced stage, 34% high‐intermediate or high IPI score and median CD4+ cell count 188/microL (range: 8–1172). An anthracycline‐containing first‐line chemotherapy was delivered to 97% of pts with the addition of rituximab in 86%. By IHC COO assigned was GCB in 31 (47%) and non‐GCB in 35 (53%) cases. COO by NS was determined in 60 pts: 34 pts were GCB (57%), 11 ABC (18%) and 15 UC (25%). Hans algorithm and NS assay concordantly assigned COO subtypes in 80% of cases (Cohen's kappa = 0.604; p < 0.0005). After a median follow‐up of 50 months, 2‐yrs progression‐free survival (PFS) and overall survival (OS) of the entire series were 56% and 61%, respectively. No statistically significant survival difference was found according to IHC and NS COO assignment, although a trend for a superior OS in GCB subtype defined by NS was observed (2‐yrs OS: 71% vs 53%, p = 0.078). MYC, BCL‐2, BCL‐6 proteins expression and gene rearrangements are summarized in Table 1. BCL‐2 overexpression was detected in 22 pts (34%), was not sustained by BCL‐2 rearrangement and was associated with an inferior OS (2‐yrs OS 48% vs 67%, p = 0.047, HR: 2.04). MYC protein overexpression and MYC, BCL‐2, BCL‐6 gene rearrangements, both single or combined, did not impact survival. Stage IV and elevated LDH were associated with inferior PFS (p = 0.024 and p = 0.017) and OS (p = 0.014 and p = 0.027). A baseline CD4+ count >200/microL had a protective role for both progression and death (2‐yrs PFS: 63% vs 42%, p = 0.013, HR: 0.40; 2‐yrs OS: 72% vs 42%, p = 0.002, HR: 0.29). At multivariate analysis, only CD4+ count >200/microL was associated with superior PFS and OS.Summary/Conclusion:In HIV‐associated DLBCL we found an high concordance between Hans algorithm and NS in the COO subtypes allocation, with subgroups distribution by NS similar to what described in the HIV negative population. The trend for superior outcome in GCB type by NS needs to be confirmed in a larger cohort, as well as the prognostic role of MYC, BCL‐2, BCL‐6 status. DLBCL's adverse clinical features and severe immundeficiency correlated with a dismal prognosis, as expected.image