Appropriate Management of Molecular Subtypes of Diffuse Large B-Cell Lymphoma
Appropriate Management of Molecular Subtypes of Diffuse Large B-Cell Lymphoma
In recent years, we have made huge strides in our understanding of the molecular complexity of diffuse large B-cell lymphoma (DLBCL). New technologies, such as gene expression profiling, RNA interference screening, and DNA sequencing, have identified several new signaling pathways and therapeutic targets for drug development. While we once considered DLBCL to be a single disease entity, recent insights have helped identify the existence of at least three distinct molecular diseases: a germinal center B-cell–like subtype, an activated B-cell–like subtype, and a primary mediastinal B-cell lymphoma subtype. All three subtypes originate from different stages of B-cell differentiation and are characterized by distinct mechanisms of oncogenic activation. This classification of DLBCL has laid the foundation for the development of new agents and novel strategies that target individual subtypes.
The classification of diffuse large B-cell lymphoma (DLBCL) has been significantly refined as a result of novel insights into the biology of lymphoid tumors. Although it has been known for some time that DLBCL is a clinically and biologically diverse disease, new diagnostic technologies, such as gene expression profiling (GEP), have defined a new molecular taxonomy for DLBCL and led to the identification of driver mutations and druggable targets. DLBCL can now be divided into at least three molecular subtypes that correspond to distinct stages of B-cell differentiation. It is critical to understand and consider these pathobiologic distinctions in the context of novel targets and strategies in DLBCL.
The most recent World Health Organization classification of tumors of hematopoietic and lymphoid tissues divides DLBCL into four major groupings—and these are further subdivided according to clinicopathologic and molecular characteristics. DLBCL not otherwise specified is the most common group; on the basis of GEP results, it can be further subdivided into the germinal center B-cell–like (GCB) and activated B-cell–like (ABC) subgroups (Figure 1).[2,3] Genes that are associated with the GCB subtype include markers of germinal center differentiation, such as CD10 and the BCL6 gene. In the ABC type, the nuclear factor kappa B (NF-κB) pathway is constitutively active, with high expression of NF-κB target genes. Although both GCB and ABC subtypes express B-cell lymphoma 2 (BCL2)—which is induced > 30-fold during peripheral B-cell activation—its expression is > 4-fold higher in most ABC DLBCLs compared with GCB DLBCLs.[5,6] The distinct genetic characteristics of the GCB and ABC subtypes suggest their derivation from different stages of B-cell differentiation, with the GCB subtype arising from germinal center B cells and the ABC subtype from post–germinal center B cells that are blocked during plasmacytic differentiation (Figure 2). Note that in patients who undergo standard immunochemotherapy with R-CHOP (rituximab + cyclophosphamide, doxorubicin, vincristine, and prednisone), overall survival is significantly worse for those with the ABC subtype than for those with the GCB subtype (Figure 3).
The third molecular subtype of DLBCL, primary mediastinal B-cell lymphoma (PMBL), arises from a thymic B cell. This disease entity occurs predominantly in girls and young women and shares many clinical features with classic Hodgkin lymphoma of the nodular sclerosis type (CHL-NS).[1,8] GEP studies have demonstrated that PMBL is a distinct biologic entity that shares many molecular similarities with CHL-NS. Although B-cell transcription factors such as OCT-2 and BOB-1 are expressed in PMBL, immunoglobulin production is defective, in contrast to other subtypes of DLBCL.
Standard Chemotherapy Platforms in DLBCL
In the early 1970s, the addition of doxorubicin to cyclophosphamide, vincristine, and prednisone (CVP)—CHOP—resulted in the first curative regimen for DLBCL and underlined the importance of anthracyclines in DLBCL therapeutics. Subsequently, the empiric addition of drugs to CHOP did not improve outcomes in patients with DLBCL, as shown in the landmark randomized study that compared CHOP with second- and third-generation regimens.
Further attempts were made to improve the curative ability of CHOP. In the Deutsche Studiengruppe für Hochmaligne Non-Hodgkin Lymphome (DSHNHL) four-arm studies, CHOP was administered every 14 or 21 days, without or with etoposide (CHOEP), to patients aged > 60 years and to low-risk patients aged ≤ 60 years. The results demonstrated the benefits of CHOEP-21 in younger patients and CHOP-14 in older patients.[11,12]. However, similar trials showed that these survival benefits were lost when rituximab was added (results of the Groupe d’Etude des Lymphomes de l’Adulte [GELA] study, which demonstrated superior survival with R-CHOP compared with CHOP in patients aged ≥ 60 years).[13-15] The DSHNHL also performed a randomized study (RICOVER-60) of 6 vs 8 cycles of CHOP-14, with or without rituximab, in elderly patients with DLBCL. They found no difference in the outcomes of patients who received 6 vs 8 cycles of treatment, but based on a historical comparison, the researchers suggested that R-CHOP-14 should be the new standard. However, when R-CHOP-14 was compared with R-CHOP-21 in two randomized trials, no benefit of R-CHOP-14 was confirmed; hence, R-CHOP-21 continues to be the standard chemotherapy platform.[16-18]
A randomized study compared dose-intense R-ACVBP (rituximab + doxorubicin, cyclophosphamide, vindesine, bleomycin, and prednisone) with R-CHOP-21 in patients aged < 60 years with a low-risk International Prognostic Index score. At 3-year follow-up, the progression-free survival (PFS) of patients who received R-ACVBP (87%) was significantly superior to the PFS of those who received R-CHOP (73%); however, significant hematologic toxicity limited the use of R-ACVBP in younger patients. Although this study demonstrates that the R-CHOP platform can be improved, the clinical limitations of R-ACVBP and the absence of information on its activity within DLBCL molecular subgroups restrict its use as a universal platform replacing R-CHOP. Other dose-intensity approaches, including autologous stem cell transplant, have been studied as initial therapy for DLBCL but have not shown a clear benefit over R-CHOP alone.
The current National Comprehensive Cancer Network guidelines include both R-CHOP and DA-EPOCH-R (dose-adjusted etoposide, prednisone, vincristine, cyclophosphamide, and doxorubicin, with rituximab) as suggested regimens for the front-line treatment of DLBCL.[17,18] Although R-CHOP-14 is included along with R-CHOP-21, recent randomized studies support the use of the latter.
The DA-EPOCH-R regimen was developed from in vitro modeling of drug resistance and drug pharmacodynamics and employs infusional drug scheduling, topoisomerase II targeting, and pharmacodynamic dosing. A multicenter Cancer and Leukemia Group B (CALGB) cooperative group study of 69 patients who received the regimen reported a 5-year time to progression of 81% and overall survival of 88%; the toxicity profile was similar to that of R-CHOP. Other phase II trials have reported similarly promising results with DA-EPOCH-R, and a randomized comparison of DA-EPOCH-R and R-CHOP, with analysis of outcome according to molecular subtype, has recently completed accrual.[23-25]
Treating GCB DLBCL
By far the most common molecular subtype of DLBCL, GCB DLBCL typically occurs in children and young adults. While it has a much better prognosis than ABC DLBCL, approximately 30% of patients with GCB DLBCL are not cured with R-CHOP chemotherapy. One of the most interesting targets in GCB DLBCL is BCL6; while it is typically highly expressed in the GCB subtype, it is rarely expressed in ABC DLBCL. BCL6 is a key transcription factor that represses many target genes involved in such processes as lymphocyte activation, apoptosis, and the DNA damage response. Chromosomal translocations can lead to the deregulation of BCL6, or it may be altered by multiple somatic mutations. These translocations/mutations enhance the inhibitory effect of BCL6 on the apoptotic stress response, leading to tumor proliferation and treatment failure.
It is challenging to target BCL6 directly, but specific inhibitors of BCL6 are in development. One of these, the 79-6 complex, is a small molecule inhibitor of BCL6 that binds to the corepressor binding groove of the BCL6 BTB domain and kills BCL6-positive cell lines. It is possible to target other BCL6 domains, and strategies such as inhibition of histone deacetylation to overcome the effects of BCL6 repression on p53 and cell cycle inhibitory proteins may also be useful.
In the context of targeting GCB DLBCL, the effect of topoisomerase II inhibition on BCL6 expression is potentially interesting. Inhibiting topoisomerase II with agents such as etoposide—through ubiquitin-mediated protein degradation and possibly transcriptional inhibition—results in downregulation of BCL6 expression. This may help explain the improvement in event-free survival in younger patients who received etoposide in addition to CHOP (CHOEP vs CHOP alone) in the DSHNHL study. Because the incidence of GCB DLBCL is higher in younger persons than in older ones, younger patients may benefit more from the addition of etoposide.[11,12] Although the benefit of adding etoposide was lost when rituximab was added to CHOEP (R-CHOEP), the results nonetheless suggest that topoisomerase inhibition may be an important strategy in GCB DLBCL.
This relationship between topoisomerase II inhibition and BCL6 suggests that regimens that can more effectively inhibit topoisomerase II may be more effective in patients with GCB DLBCL, even when considering regimens that include rituximab. The DA-EPOCH-R regimen incorporates two topoisomerase II inhibitors, etoposide and doxorubicin. Inhibition of topoisomerase II is optimized by both continuous delivery of the drugs over 96 hours and pharmacodynamic dose adjustment on successive cycles, which ensures adequate steady-state concentrations. In two studies of the regimen in patients with untreated DLBCL, the outcome for those with the GCB subtype was particularly good: event-free survivals ranged from 95% to 100% after 5 years of follow-up.[22,32]
Another promising therapeutic target in GCB DLBCL is EZH2. Gain-of-function mutations in EZH2 result in increased H3K27 methylation and are present in 25% of patients with GCB DLBCL, and inhibitors of EZH2 are toxic to GCB cell lines.