The 2013 American Society of Hematology Annual Meeting was relatively quiet in terms of lymphoma, in contrast to ongoing excitement in the other lymphoid malignancies, chronic lymphocytic, and myeloma. Still, a number of important issues were addressed.
Can we improve on the standard therapy for diffuse large B cell lymphoma (DLBCL), R-CHOP, by adding biologically targeted agents, either for all patients or in defined subgroups? At the 2012 meeting promising data from two Phase II trials of lenalidomide + R-CHOP were presented. The results of one of these, the Italian REAL07 trial (Chiappella A et al, Abstract 850), were updated for the 2013 meeting.
In this study, 49 fit patients age 60 to 80 (median age of 69) with untreated DLBCL received R-CHOP along with lenalidomide at 15 mg on days 1-14 of each of six 21-day cycles. With a median follow-up of 28 months, the overall response rate is 92 percent (86% complete remission [CR]).
All drugs were given at a dose intensity of more than 90 percent, generally on schedule, without undue toxicity. Cell-of-origin information was available on 32 patients (16 germinal center B-cell like [GCB] and 16 non-GCB or activated B cell type). There were no differences in outcomes between these groups: an 88 percent response rate in both, with a two-year progression-free survival (PFS) rate of 71 percent (95% CI 40%-88%) for GCB and 81 percent (95% CI 51%-93%) for patients with non-GCB.
This analysis supports the Mayo Clinic experience reported at ASH 2012 that the poor outcomes observed for patients with non-GCB DLBCL treated with R-CHOP are overcome by the addition of lenalidomide. There may be, however, increased myelosuppression.
R-CHOP with and without Lenalidomide
ECOG has opened a randomized trial of R-CHOP with and without lenalidomide (at the Mayo Clinic dose of 20 mg on days 1-10). Based on the REAL07 and Mayo Clinic Phase II trials, it is likely that lenalidomide will be beneficial in non-GCB DLBCL, while any effect on GCB outcomes remains unclear. While currently cell of origin is a prognostic factor, these data may eventually lead to different treatment regimens for these subtypes.
Other targeted agents, such as proteasome inhibitors and B cell receptor signal inhibitors, are being combined with R-CHOP as well. As an example, data were presented (Younes A et al, Abstract 852) from a Phase IB study that ibrutinib can safely be added at 560 mg daily to R-CHOP, with the expected high response rates for R-CHOP. A randomized Phase III trial of R-CHOP with and without ibrutinib is now ongoing.
An alternative strategy to improved outcomes for high-risk DLBCL is to utilize maintenance therapy. If such therapy were effective, it could suppress residual DLBCL cells to prolong PFS, or even eliminate them to increase “cure” rates. This approach was tested in the PRELUDE study (Crump M et al, Abstract 371), a randomized Phase III study of the PKC-b inhibitor enzastaurin (500 mg daily for up to three years) compared with placebo in patients with high-risk DLBCL, defined as initial IPI 3-5.
Eligible patients, however, had to achieve after six to eight cycles of R-CHOP CR/complete response unconfirmed (CRu) by CT or PET-negative CR, thus selecting a somewhat more favorable cohort as evidenced by the two-year disease-free survival (DFS) for the study overall being more than 75 percent.
This strategy required a large (758-patient) trial that began accrual in 2006, took four years to accrue, and now has three years of follow-up. The study showed no benefit of enzastaurin for any endpoint—DFS (79% vs. 75% at two years), event-free survival (EFS), or overall survival (81% vs. 82% at four years).
Subset analyses may generate hypotheses about GCB, or non-GCB DLBCL subtypes might benefit. While these negative results may reflect limited activity of this particular drug in DLBCL, they may also indicate that residual viable DLBCL cells are difficult to target, and this study certainly demonstrates the perils of this design.
The benefits and risks of maintenance strategies in follicular lymphoma require long-term follow-up. Two such studies were updated at this meeting. The PRIMA trial tested the value of two years of rituximab maintenance versus observation in patients who had responded to investigators' choice of three immuno-chemotherapy regimens (75% of patients received R-CHOP). Salles (Abstract 509) presented six-year follow-up data showing that the previously reported 16% absolute benefit in PFS for maintenance rituximab persists (59% for rituximab vs. 43% for the observation arm), for a hazard ratio (HR) of 0.58.
Deaths did not differ between the arms, with an overall survival rate at six years of 88 percent, which is in itself quite remarkable since in the pre-rituximab era median overall survival was eight to 10 years. Investigator-reported response to next therapy did not differ, although this was not broken down by use of subsequent rituximab, which might well differ depending on whether patients had maintenance rituximab.
In the SAKK35/03 trial (Taverna CJ et al, Abstract 508), a smaller trial with a more heterogeneous FL population including both previously treated and treatment-naïve patients, patients responding to treatment with rituximab weekly x 4 were then randomized to receive rituximab once every two months for four infusions (Arm A) or for five years (Arm B).
Although the primary endpoint of improved EFS was not met, interpretation is complicated since there were, presumably by chance, more events in Arm B during the initial eight months when treatments did not differ.
Using a landmark analysis for patients at risk at eight months did show longer PFS with prolonged maintenance (7.4 vs. 3.5 years). To decide on maintenance strategies, we now must consider these somewhat equivocal results that certainly suggest prolonged PFS with a five-year maintenance in a heterogeneous FL population, the two-year maintenance data in first remission FL from PRIMA, the previously reported RESORT data that in a low tumor burden FL population responding to single-agent rituximab re-treatment at relapse is as beneficial as prolonged maintenance, as well as the toxicity (hypogammaglobulinemia and increased infections) and financial costs of prolonged rituximab.
Clearly, the optimal strategy for each clinical situation remains unclear and subject to personal interpretation.
Many promising targeted agents are under investigation in lymphoma. One abstract of interest (Treon S et al, Abstract 251) begins to look at biomarkers of activity and resistance. The BTK inhibitor ibrutinib yielded a 57 percent major response rate in relapsed Waldenström's macroglobulinemia (WM), using the CLL dose of 420 mg, not the mantle cell lymphoma dose of 560 mg daily.
The activating L265P mutation in the MYD88 gene is found in more than 90 percent of cases of WM, and MYD88 interacts with the BTK signaling pathway, although it has other signaling effects as well. Here they showed that a mutation in the CXCR4 gene that is present in about 30 percent of WM patients induces BTK activity and confers relative resistance to this dose of ibrutinib.
While mutations in the BTK site covalently bound by ibrutinib have been identified as a mechanism of resistance in CLL, we do not yet know the range of resistance mechanisms that will certainly evolve to these signal inhibitors, and this report suggests that such resistance may involve complex regulation.
This is an exciting time as we look forward to practice-changing incorporation of targeted agents into standard cytotoxic regimens, and perhaps for targeted therapy to even replace cytotoxic chemotherapy regimens in indolent lymphomas.
Careful evaluation will be necessary to assess whether patients with prolonged disease courses, as in FL, really require treatment until progression, potentially lifelong, taking into account toxicity and financial concerns.