Concurrent Expression Of MYC/BCL2 Protein In Newly Diagnosed DLBCL Is Not Associated With An Inferior Survival Following EPOCH-R Therapy

While several studies have now demonstrated that patients with MYC-rearranged DLBCL have a worse outcome following treatment with CHOP -with or without rituximab - recent reports suggest that concurrent expression of MYC and BCL2 protein by immunohistochemistry (IHC), independent of a MYC rearrangement, is associated with an inferior clinical outcome following R-CHOP (Johnson et al. JCO 2012 Oct 1; 30 (28): 3452-9). We previously reported that DLBCL patients with a MYC rearrangement have a similar outcome to MYC negative cases (4 year EFS of 83% versus 76% respectively: p=0.46) following dose-adjusted (DA)-EPOCH-R (Dunleavy et al. Ann Oncol 2011 (22) Suppl 4 # 71).

To assess the prognostic role of MYC and BCL2 protein expression, we performed immunohistochemistry for MYC and BCL2 in patients with newly diagnosed DLBCL who received DA-EPOCH-R or short-course (SC)-EPOCH-RR (if HIV infected). Primary mediastinal B-cell lymphoma (PMBL) cases were excluded as they have a different biology. Tumors were scored positive for MYC if ≥ 40% of cells stained positive. For BCL2, the staining intensity was compared with that in control T cells present in the tumor samples – tumor cells were considered positive if they stained the same or more intensely than T cells. We used the Hans algorithm, as previously described, to predict cell of origin as germinal center B-cell like (GCB) or non-GCB type and correlated this with outcome.

Of 66 patients enrolled on study, characteristics were median (range) age: 48 (18-76) years; male sex 47 (71%); IPI score low versus intermediate/high 18 (27%) versus 48 (73%); HIV positive 22 (33%). IHC was positive for MYC in 28/48 (58%) cases and positive for BCL2 in 24/51 (47%). 36/51 (71%) and 15/51 (29%) of cases were of GCB and non-GCB origin respectively. At a median follow-up time of greater than 10 years, progression-free survival (PFS) and overall survival (OS) were 67.5% and 75% for all patients. We compared survival in the 4 groups: MYC+/BCL2+, MYC+/BCL2-, MYC-/BCL2+ and MYC-/BCL2-. PFS and OS were not significantly inferior in any group (global p value=0.5 (PFS) and 0.8 (OS)). Cell of origin did however predict outcome and at 10 years follow-up, PFS was 78% (GCB) versus 43% (non-GCB) (p=0.016) and OS 80% (GCB) versus 65% (non-GCB) (p=0.24). There was a significant association between MYC+/BCL2+ cases and non-GCB subtype (p=0.01) but as 42% were of GCB origin, MYC+/BCL2+ status alone did not predict a poor outcome.

In DLBCL patients treated with DA-EPOCH-R and SC-EPOCH-RR, concurrent expression of MYC and BCL2 protein did not correspond with a worse clinical outcome. However, cell of origin (GCB versus non-GCB) was predictive of outcome. MYC+/BCL2+ cases segregate with the non-GCB subtype. The overall survival of PMBL cases was 97% at 5 years follow-up. We are currently prospectively studying the DA-EPOCH-R regimen in MYC-rearranged DLBCL in a multicenter study (NCT01092182).

No relevant conflicts of interest to declare.