Cell-of-Origin Subtype Classification of Diffuse Large B-Cell Lymphoma Using the Lymph2Cx Assay Retains Relevance in the Context of BCL2 and MYC Expression Status

In 2000, gene expression profiling (GEP) studies discovered two distinct, predominant genetic profiles associated with either a germinal center B-cell (GCB) or activated B-cell (ABC) cell-of-origin (COO) contributing to the morphologically and clinically heterogeneous nature of diffuse large B-cell lymphoma (DLBCL) (Alizadeh et al NEJM 2000). Follow-up GEP and sequencing studies confirmed the validity of the COO subgroup classification. Recently, we established the Lymph2Cx assay, which uses 20 gene probes, Nanostring technology (Seattle, WA), and formalin-fixed paraffin-embedded tissues (FFPET) to classify DLBCL into reproducible COO subgroups (Scott et al Blood 2014). Expression of the anti-apoptotic BCL2 and proliferative MYC oncogenes are also highly associated with prognosis, such that double protein positive cases have a significantly worse outcome (Johnson et al J Clin Oncol 2012). However, in at least one study BCL2/MYC was reported to over-ride the prognostic significance of COO (Hu S et al Blood 2013).

We further investigated the relationship between COO, as determined by Lymph2Cx, with BCL2 and MYC protein status. Fifty-six DLBCL FFPET the from Scott et al, Blood 2014 series were constructed on a tissue microarray (TMA) including 20 ABC, 27 GCB, and 9 unclassifiable. For BCL2 staining, we used the standard BCL2 clone (124, Ventana Medical Systems, Tucson, AZ) and rabbit monoclonal antibody (SP66, Spring Biosciences), which reportedly is more sensitive for IHC (Kendrick et al Hum Pathol 2014 in press). The EP121 clone (Epitomics) was used for MYC. All cases were successfully stained with both BCL2 antibodies and MYC on the BenchMark¨ XT instrument. The stained TMAs were independently scored in increments of 10% by two hematopathologists (K.T. and L.M.R) and positivity for BCL2 and MYC were assessed at cut-offs of ³50% and ³40%, respectively, as previously described (Johnson et al J Clin Oncol 2012). All patients received R-CHOP therapy and overall survival (OS) and progression free survival (PFS) were estimated with the Kaplan-Meier method using date of lymphoma diagnosis. Significant differences were determined using the log-rank test and a P-value of less than 0.05. The University of Arizona Institutional Review Board in accordance with the Declaration of Helsinki approved the use of human tissues and clinical data for this study.

The ABC-DLBCL displayed worse OS and PFS compared to GCB-DLBCL using the Lymph2Cx assigned COO with PFS reaching significance (P=0.007). These correlations were slightly more pronounced compared to analysis with COO determined from the gold standard (Lenz et al NEJM 2008), which achieved a PFS difference with a P=0.03. BCL2 expression was detected in a higher frequency of ABC cases either alone (70% vs 56%) or concurrently (50% vs 30%) with MYC using SP66, and to a lesser extent with 124. More importantly, BCL2 when co-expressed with high levels of MYC portends a poor clinical outcome (Figure 1). When BCL2+/MYC+ cases were examined according to subtype, ABC-DLBCL displayed a worse OS and PFS, although there were not enough cases to reach statistical significance (Figure 2). The SP66 antibody was superior for BCL2 detection and demonstrating the prognostic significance of BCL2+/MYC+ cases in the total cohort. The Lymph2Cx COO designations also reflected the predominant correlation of BCL2 t(14;18) translocations with GCB-DLBCL (38%, 9/24 cases vs 0% 0/18 ABC cases, P = 0.005) within the cohort displaying the typical overall frequency (19%, 9/47 cases).

The Lymph2Cx assay provides a reliable, robust and straightforward assessment of DLBCL COO that overcomes limitations of large scale GEP and is consistent with the well known BCL2 and MYC subtype associations. The COO remains relevant even in the context of BCL2 and MYC status indicating that both of these factors are important for prognosis; however, at this time there are not enough cases for a multivariate comparison. With the advent of targeted therapies and precision medicine, the ability to distinguish DLBCL COO will have strongly impact clinical trials and patient management at time of diagnosis.

Figure 1.

View largeDownload slide

Overall and progression free survival of DLBCL subtypes as assigned by Lymph2Cx according to BCL2 and MYC protein status.

Figure 1.

View largeDownload slide

Overall and progression free survival of DLBCL subtypes as assigned by Lymph2Cx according to BCL2 and MYC protein status.

Close modal

Figure 2.

View largeDownload slide

Overall and progression free survival of concurrent BCL2 and MYC protein positive cases according to DLBCL subtypes as assigned by the Lymph2Cx.

Figure 2.

View largeDownload slide

Overall and progression free survival of concurrent BCL2 and MYC protein positive cases according to DLBCL subtypes as assigned by the Lymph2Cx.

Close modal