John Sweetenham, MD, FRCP
Professor of Medicine, Vice-Chair for Clinical Research, Taussig Cancer Institute, Cleveland Clinic
Dr. Sweetenham receives research funding from Novartis, Millennium, Seattle Genetics, Aegera, and Celgene.
(Editor’s Note: The original question was submitted to Dr. Sweetenham through Consult a Colleague. He expanded his answer for print.)
A previously healthy 72-year-old male presented with a new diagnosis of clinical stage IVB diffuse large B-cell lymphoma (DLBCL), with maxillary sinus involvement and involvement of multiple bony sites. His ECOG performance status was 0 and his International Prognostic Index (IPI) score was 3 based on his age, clinical stage, and involvement of multiple extranodal sites, all of which are adverse factors in the IPI. A biopsy from the maxillary sinus confirmed DLBCL by morphology and was positive for cyclin D1. Since cyclin D1 expression in DLBCL is reported to have a poor prognosis, should this patient be treated with CHOP-rituximab or should he receive a more intensive regimen such as dose-adjusted EPOCH-rituximab?
This clinical case raises several important questions regarding the management of advanced DLBCL, partly related to diagnostic criteria and to the optimal treatment regimen, but also regarding the broader issue of whether different immunophenotypic or molecular subtypes of DLBCL should be treated differently.
Cyclin D1 is a nuclear protein with a well-defined role as a regulator of cell cycle progression from G1. Its deregulation is central to the pathogenesis of some B-cell neoplasms, especially mantle cell lymphoma (MCL), where it is over-expressed in about 90 percent of cases, as well as in myeloma and hairy cell leukemia. In the vast majority of MCLs, cyclin D1 over-expression is associated with the t(11;14), whereas this association is much less common in other B-cell neoplasms.1
Reports of the expression of cyclin D1 in DLBCL have been conflicting, partly because different methodologies are used for detection of cyclin D1 and partly because the morphologic distinction between DLBCL and the blastoid variant of MCL can be problematic, particularly since other immunophenotypic characteristics of DLBCL and MCL can sometimes overlap. Despite these challenges, there are several case reports and some larger series that describe the presence of cyclin D1 in cases of DLBCL that are negative for CD5 but have other classical markers of DLBCL, such as bcl-6 and MUM-1.2
As indicated by the questioner, there is a suggestion from the limited literature that patients with cyclin D1-positive DLBCL have a worse prognosis. In the largest published series, nine of 10 patients had died from progressive disease after a median of 29 months from diagnosis.2 However, these data are anecdotal, and if this represents a true clinicopathologic entity, there are insufficient data to indicate whether it has a meaningfully different prognosis from other cases of DLBCL.
In contrast, other immunophenotypic and molecular features of DLBCL have been shown to have defined predictive and prognostic value. Examples include the adverse prognostic effect of bcl-2 expression and lack of bcl-6 expression, both of which can apparently be overcome by the addition of rituximab to standard chemotherapy regimens such as CHOP (cyclophosphamide, doxorucibin, vincristine, and prednisone). Gene expression profiling studies have identified two major molecular subtypes of DLBCL; one has a gene expression profile (GEP) consistent with germinal center B-cells (GCB-like), and one has a profile consistent with activated peripheral B-cells (ABC-like). Molecular subtyping by GEP has also been shown to have prognostic value, as patients with ABC-like GEPs have poorer progression-free and overall survival rates independent of the IPI. This difference in prognosis is seen for patients treated both with CHOP and with CHOP-rituximab.3
The clinical utility of these findings has been limited by the lack of routine availability of suitable tissue and resources for conducting GEP studies. As a result, several groups have explored the use of immunohistochemical (IHC) surrogates to assign GCB and ABC subtypes.4 Since the correlation between cell of origin identified by GEP and IHC is high, these algorithms are gaining widespread use as a method of risk stratification in clinical trials.
Although the prognostic significance of cell of origin (GCB versus ABC) in DLBCL is well established, the question of whether this or any other biologic predictor of outcome should be used to direct therapy for DLBCL is unknown. The dose-adjusted EPOCH-R regimen mentioned by the questioner uses infusional drug scheduling and pharmacodynamic dosing based on hematologic toxicity to exploit tumor proliferation as a target mechanism. Phase II studies of this regimen have shown impressive results and, interestingly, no difference in progression-free or overall survival according to GCB or non-GCB cell of origin.5 As a result, this regimen is now being compared directly with CHOP-rituximab in a randomized, prospective intergroup study led by the CALGB, which includes an analysis of cell of origin by GEP. The results of this study will not be available for several years, so for now, CHOP-rituximab remains the standard regimen for first line therapy of DLBCL, irrespective of cell of origin or expression of other biomarkers.
Early results from some recent studies suggest that this is likely to change. Two recent studies have suggested that the addition of bortezomib to standard chemotherapy-rituximab combinations for DLBCL produces improvements in response rates and progression-free survival rates, which may be restricted to patients with the ABC subtype.6,7 Additionally, data are emerging to suggest that some recently identified therapeutic targets such as Bruton’s tyrosine kinase (Btk) are differentially expressed in ABC- versus GCB-like DLBCL, providing a rationale for limiting trials of these agents to those patients with specific molecular subtypes.
Future progress in the treatment of DLBCL is likely to emerge from further characterization of specific molecular targets. In the meantime, CHOP-rituximab should be regarded as standard, first-line therapy for all patients with DLBCL outside clinical trials.
- Swerdlow SH, Zukerberg LR, Yang WI, et al. The morphologic spectrum of non- Hodgkin’s lymphomas with BCL1/cyclin D1 gene rearrangements. Am J Surg Pathol. 1996;20:627-640.
- Ehinger M, Linderoth J, Christensson B, et al. A subset of CD5- diffuse large B-cell lymphomas expresses nuclear cyclin D1 with aberrations at the CCND1 locus. Am J Clin Pathol. 2008;129:630-638.
- Lenz G, Wright G, Dave SS, et al. Stromal gene signatures in large B-cell lymphomas. N Engl J Med. 2008;359:2313-2323.
- Meyer PN, Fu K, Greiner TC, et al. Immunohistochemical methods for predicting cell of origin and survival in patients with diffuse large B-cell lymphoma treated with rituximab. J Clin Oncol. 2011;29:200-207.
- Wilson WH, Dunleavy K, Pittaluga S, et al. Phase II study of dose-adjusted EPOCH and rituximab in untreated diffuse large B-cell lymphoma with analysis of germinal center and post-germinal center biomarkers. J Clin Oncol. 2008;26:2717-2724.
- Dunleavy K, Pittaluga S, Czuczman MS, et al. Differential efficacy of bortezomib plus chemotherapy within molecular subtypes of diffuse large B-cell lymphoma. Blood. 2009;113:6069-6076.
- Ruan J, Martin P, Furman RR, et al. Bortezomib plus CHOP-rituximab for previously untreated diffuse large B-cell lymphoma and mantle cell lymphoma. J Clin Oncol. 2010. [Epub ahead of print]