Assessment of CD52 Expression in "Double-Hit" and "Double-Expressor" Lymphomas: Implications for Clinical Trial Eligibility

High-grade B-cell lymphomas (HGBCL) with MYC and BCL2 and/or BCL6 rearrangements (i.e. "double/triple-hit lymphomas" [DHL/THL]) and the subset of diffuse large B-cell lymphomas (DLBCL) and HGBCL, not otherwise specified (NOS), with MYC and BCL2 protein over-expression (i.e. "double-expressor lymphomas" [DEL]) represent distinct subsets of mature B-cell lymphomas with aggressive clinical course, poor response to conventional chemotherapy and high relapse rates. Compared with other lymphoma subtypes, DHL exhibits a higher frequency of bone marrow (BM) involvement, which in turn is associated with adverse outcome. The BM microenvironment contributes to this negative prognostic impact by promoting chemoresistance through several mechanisms, including the inhibition of antibody-mediated phagocytosis by BM macrophages.

The antibody-refractory nature of the BM microenvironment can be temporarily abrogated through the synergistic effects of high-dose cyclophosphamide (CTX), which inhibits secretion of prostaglandin E2 and induces secretion of IL8, TNFα, VEGF, and CCL4 by leukemic cells, resulting in macrophage recruitment, activation and phagocytosis (Pallasch, Cell 2014). The potential of this therapeutic strategy has recently been extended to human-derived xenografts taken from patients with relapsed/refractory DHL (Lossos, Blood 2016 [Abstract]). The humanized monoclonal antibody alemtuzumab (Campath) targets CD52, a GPI-linked glycoprotein frequently expressed by mature lymphocytes. The success of this agent in the proof-of-principle studies referenced above has provided inspiration for a new phase I clinical trial investigating the use of alemtuzumab and low-dose CTX for the treatment of DHL/THL and DEL (ID: NCT03132584).

Previous studies have shown significant heterogeneity in CD52 expression by several of the more aggressive mature B-cell lymphomas (e.g. DLBCL, Burkitt lymphoma), with 25% of cases exhibiting negligible expression by immunohistochemistry (IHC) (Rodig, Clin Cancer Res 2006). As such studies predated our current conception of DHL/THL and DEL, the prevalence of CD52 expression within these newer diagnostic categories has remained speculative. To eliminate this knowledge gap and provide decision support for clinical trial enrollment, we assessed the frequency, intensity and uniformity of CD52 expression within 98 cases of DHL/THL and DEL. Our study included 35 DHL (25 with MYC and BCL2 rearrangements, 10 with MYC and BCL6 rearrangements), 5 THL (with MYC, BCL2 and BCL6 rearrangements), 7 HGBCL, NOS, with double-expressor phenotype, and 51 DLBCL, NOS, with double-expressor phenotype (17 of GCB origin, 34 of ABC origin).

The majority of DHL/THL and DEL exhibited convincing cytoplasmic and/or membranous CD52 expression (75% in both groups). There was no significant difference in CD52 intensity (scored from 0-3) or predominant staining pattern (cytoplasmic vs. membranous) between DHL/THL and DEL, however DHL/THL were more likely to show uniform (vs. variable) staining [OR = 4.8; P = 0.0026]. Consistent with their predilection for BM, DHL were more likely to be derived from BM biopsies than DEL [OR = 12.1; P = 0.022], however the frequency of CD52 expression within BM specimens was not significantly different than those from other bodily locations. Assigned morphologic group (blastoid vs. DLBCL vs. DLBCL/BL) and the quantitative results of additional IHC studies (BCL2, MYC, Ki67) were similarly non-predictive of CD52 expression status. Within DHL, the presence of BCL2 versus BCL6 translocation had no bearing on CD52 expression, although there was a trend towards increased CD52 expression in the presence of BCL2 translocation [OR = 3.45; P = 0.099]. Among DLBCL with double-expressor phenotype, the Hans cell-of-origin classification (GCB vs. ABC) had no apparent impact on CD52 expression.

In summary, our results indicate that CD52 is expressed by the majority of DHL/THL and DEL, suggesting that alemtuzumab-based therapy may be appropriate for most patients. Target validation must be performed on a case-by-case basis, however, as CD52 expression is not readily predicted by histomorphologic, immunophenotypic or cytogenetic findings.