Role of Molecular Subtype in Predicting Outcome of AIDS-Related Diffuse Large B-Cell Lymphoma

We read with interest the study by Chadburn et al 1 that examined the prognostic significance of molecular subtype in AIDS-related diffuse large B-cell lymphoma (DLBCL).Herein, the authors obtained DLBCL biopsies from two AIDS Malignancy Consortium studies and examined multiple immunohistochemical (IHC) markers to identify tumors derived from germinal center B cells (GCB) or postgerminal center B cells and correlated these with outcome. [1]2][3] In contrast to what has been demonstrated in HIV-negative DLBCL, the authors concluded that molecular subclassification by IHC may not be clinically relevant in HIV-associated DLBCL. [4]][6] The role of molecular subtype in predicting outcome in HIVnegative DLBCL is now well established across many independent studies, both in the pre-and postrituximab era, and by IHC and gene expression profiling. [4]5][6] In this regard, the authors' findings are surprising and also contrary to our own prospective findings in HIVassociated DLBCL, where IHC analysis of DLBCL subtype predicts both lymphoma-specific and overall survival. 7Furthermore, we are unaware of a biologic basis to suggest that the tumor histogenesis of HIV-associated DLBCL is fundamentally different from HIVnegative DLBCL.Indeed, studies show that immune deficiency and stimulation as well as Epstein-Barr virus activation are risk factors in both HIV-positive and -negative patients with DLBCL. 8 In our opinion, the negative results of the Chadburn et al study may be related in part to its retrospective analysis of samples from two different studies that have incorporated four different treatments (cyclophosphamide, doxorubicin, vincristine, and prednisone [CHOP]; rituximab [R] plus CHOP; dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, doxorubicin [DA-EPOCH] with sequential rituximab; and DA-EPOCH with concurrent rituximab).The follow-up time is also relatively short and highly variable among the various treatment groups.Understanding the role of tumor biology in HIV-associated lymphoma is a prerequisite for identifying and targeting novel and specific pathways to improve the outcome of these biologically diverse subtypes of DLBCL. 9In HIV-negative DLBCL, for example, the activated B-cell (ABC) subtype (which constitutes most cases identified as non-GCB by IHC) has been associated with an inferior prognosis.The poor outcome of this subtype may relate to the constitutive activation of the nuclear factor-B pathway, which has been ascribed to a signaling cascade involving CARD-11, BCL-10, and MALT-1 leading to activation of I--B-kinase. [10][1][12] Inhibition of nuclear factor-B and blocking its ability to inhibit apoptosis in ABC cell lines is toxic, and recent clinical evidence suggests that the ABC type of DLBCL can be preferentially targeted (over the GCB type) by strategies that block I-B degradation through proteasome inhibition. 9,10,13 These studies highlight the pivotal role of tumor biology in the rational design of targeted therapies in DLBCL and the importance of well-designed prospective studies.