Abstract
Abstract 543
Background:
Non-Hodgkin lymphomas (NHL) that show some but not all of the morphologic, phenotypic or cytogenetic features that define Burkitt lymphoma (BL) have been controversial throughout the history of lymphoma classification systems. In particular, MYC translocations are found in essentially all cases of BL, but are also found in some diffuse large B-cell lymphoma (DLBCL). The 2008 WHO classification introduced a category for “B-cell lymphoma, unclassified, with features intermediate between DLBCL and BL” (BCLU), but the clinical utility of this diagnostic category is limited because the category is known to be heterogeneous, the diagnostic criteria remain vaguely defined, and the most appropriate management of such cases is unclear. A recently available monoclonal antibody has allowed for detection of MYC protein by immunohistochemistry (IHC).The spectrum of clinicopathologic features associated with MYC dysregulation in non-Burkitt, diffuse aggressive NHL is not yet clear. We therefore examined the clinical significance of high grade (i.e., “Burkitt-like”) morphology and MYC protein expression in a series of DLBCL and BCLU.
Design:
370 eligible patients were enrolled on SWOG S9704, a phase III randomized study of diffuse aggressive NHL treated by CHOP±R for 5 cycles and randomized to either 3 additional cycles of CHOP±R or autotransplant. Exclusion of T-cell neoplasms, BL, follicular lymphoma, lymphoblastic lymphoma and mantle cell lymphoma by 2008 WHO criteria resulted in 260 cases of diffuse aggressive B-cell NHL. Each case was reviewed for morphologic features including blastoid cytology, intermediate cell size, and/or starry sky background. MYC protein and phenotypic profile (GC vs. non-GC per Hans algorithm) were evaluated by IHC. Where sufficient tissue was available, FISH studies for MYC and/or BCL2 translocations were performed.
Results:
31/260 cases (12%) showed high grade morphology, consistent with BCLU. Cases with high grade morphology did not show distinct clinical features at presentation and were phenotypically heterogeneous [13/27 (48%) GC, 14/27 (52%) non-GC]. In multivariate analysis including IPI and use of rituximab, high grade morphology did not correlate with outcome. MYC IHC was positive in 27/198 (14%) cases. MYC positivity was associated with CD10 [10/17 (59%) vs 25/80 (31%), p=0.032], BCL2 [21/26 (81%) vs 42/79 (53%), p=0.013], and MYC translocations by FISH [7/14 (50%) vs 6/54 (11%), p<0.001]. MYC IHC and MYC FISH were more frequently positive in cases with high grade morphology [6/21 (29%) vs 21/177 (12%), p=0.035 and 8/24 (33%) vs. 8/53 (15%), p=0.068, respectively]. In multivariate analysis, MYC positivity was associated with poor OS in all patients, in randomized patients, and in patients with high IPI. In exploratory subset analysis, survival estimates suggested a benefit for MYC+ patients in the transplant arm, and OS appeared to be worse in MYC IHC+, BCL2 IHC+ cases compared to MYC IHC+, BCL2 IHC- cases; however, small patient numbers precluded definitive assessment of these subsets.
Conclusions:
This study shows that aggressive B-cell lymphomas with high grade morphology do not show distinct clinical features at presentation, are phenotypically heterogeneous, have variable MYC expression, and do not show significantly different outcome compared to other aggressive DLBCL. These findings question whether the WHO category of BCLU should be maintained, or whether such cases are better considered part of the spectrum of DLBCL. In addition, these findings confirm recent reports that MYC immunohistochemistry is associated with poor outcome in cases of DLBCL, and for the first time extend these observations to cases with high grade morphology. The possibility that MYC+ cases benefit from transplant should be further explored in clinical trials. MYC IHC, which can be performed in most laboratories, is suggested for prognostic evaluation in routine practice.
Disclosures:
No relevant conflicts of interest to declare.
Among diffuse large B-cell lymphomas, T-cell-rich/histiocyte-rich BCL and CD30 + anaplastic B-cell subtypes exhibit distinct clinical features