The end(point) game in indolent lymphoma: After years-long search, data suggest genuine progress
Evidence from several sources suggests that OS for patients with follicular lymphoma has improved in recent years.
The reasons for this improvement are not well defined, although some evidence suggests the introduction of monoclonal antibodies is an important factor, as is the continued development of treatments for relapsed and refractory disease.
As the number of available therapies rapidly increases, the value of OS as an endpoint in trials of first-line treatment has been reduced to almost no value at all. The effects of subsequent treatments, in a disease with a very long natural history, tend to cancel out the benefits observed in the first-line setting.
As a result, PFS has been widely adopted as an acceptable endpoint for front-line studies, but this has its own limitations.
First-line therapies such as R-CHOP — which consists of rituximab (Rituxan; Genentech, Biogen Idec), cyclophosphamide, doxorubicin, vincristine and prednisone — have been associated with median PFS rates of 5 to 8 years in several studies, but that is a long time to wait to test the effect of an early intervention.
This is a source of frustration for patients — who are eager to see new treatments introduced as quickly as possible — and for clinical trial groups, which often are forced to second-guess the outcome of a study, with incomplete data, when trying to figure out an appropriate design for their next trial.
Signs of progress
The search for surrogate endpoints that will allow an early readout from trials in follicular lymphoma and improve outcomes for patients with this condition has been on for many years. Data from some recent studies show signs of genuine progress.
At this year’s ASCO Annual Meeting, an international collaborative group presented results from the Follicular Lymphoma Analysis of Surrogacy Hypothesis (FLASH). The investigators assessed the potential utility of a complete remission at 30 months from the beginning of therapy (CR30) as a surrogate for long-term PFS.
The study essentially was a meta-analysis of individual patient data from 13 qualifying prospective randomized trials in follicular lymphoma, some of which included rituximab, and some of which also had a maintenance component. The analysis included data from more than 3,000 patients.
The details of the statistical analysis are difficult for me to grasp; however, the presentation at ASCO by Daniel J. Sargent, PhD, was elegant and the conclusions from this study were very clear: CR30 appears to be a highly reliable surrogate for PFS and is a candidate to become a widely accepted primary endpoint in new front-line studies. If confirmed, it certainly would fulfill the need for an earlier readout.
A second report — recently published online in Journal of Clinical Oncology — came from the National LymphoCare Study. This industry-sponsored observational study of care patterns for lymphoma in the United States, led by Jonathan W. Friedberg, MD, MMSc, of University of Rochester, has yielded important data so far.
This study looked specifically at patients with follicular lymphoma who received R-CHOP as first-line therapy and investigated the effect of early disease progression on subsequent outcome. Disease progression within 2 years of diagnosis served as the primary endpoint, based on the observation that about 20% of patients relapse within 2 years and the fact maintenance therapy does not appear to have altered this pattern of relapse. This may indicate that affected patients have an intrinsically distinct disease that may need a different intervention.
The unsurprising conclusion from this study is that patients who experience early disease progression had inferior OS compared with those who did not, a finding that was confirmed in an independent validation data set.
These results are consistent with the FLASH analysis and — as the researchers point out — will be highly relevant to the design of clinical trials in the relapsed and refractory setting, suggesting that early vs. late disease progression may be a highly relevant factor for stratification and/or treatment selection.
Other promising approaches
Both of these studies offer hope for new, meaningful endpoints to accelerate progress in follicular lymphoma. They also have problems, freely acknowledged by the authors. Maybe the most notable is the lack of data for functional imaging, which were not included in either study.
Several recent studies, including a meta-analysis, have evaluated the role of functional imaging with FDG-PET to evaluate response in follicular lymphoma. Perhaps surprisingly, they have demonstrated that post-treatment PET scanning — particularly in patients with high tumor burden — has prognostic value and may serve as another potential early surrogate for PFS in follicular lymphoma.
If this result is confirmed in subsequent studies, and it compares favorably with CR30, it likely would “win out” based on the earlier time point for evaluation. That said, consistency in interpretation of PET scans remains problematic, and although much of the interobserver variability can be reduced by central review of images, the feasibility of central review outside the trial setting may limit the use of this approach in practice.
Encouraging data also have been published for the use of minimal residual disease measurement by polymerase chain reaction — using BCL2/IgH rearrangement — as a predictor of PFS. More recent data in B-cell lymphoma have suggested a possible role for measurement of circulating tumor DNA in this context. These approaches are promising, but their applicability in a multicenter or community-based setting has not been fully evaluated.
My own conclusion from these studies is that — as we wait to see whether functional imaging or molecular studies will prove to be reliable primary endpoints in follicular lymphoma studies — the data from the CR30 study are compelling, and this likely will be adopted as a primary endpoint in future studies.
Another issue, on which I briefly touched in a previous editorial, is whether PFS in isolation should be the only primary endpoint in these studies.
As we move toward more maintenance strategies and the use of prolonged courses of oral targeted therapy — and as we continue to struggle with the role of observation only and watch the evolution of follicular lymphoma toward a “manageable” chronic condition — patient-reported outcomes, quality of life and value endpoints will become highly relevant. They, too, will need to be factored into trial design in the front line and later.
Casulo C, et al. J Clin Oncol. 2015;doi:10.1200/JCO.2014.59.7534.
Fisher RI, et al. J Clin Oncol. 2005;23:8447-8452.
Sargent DJ, et al. Abstract 8504. Presented at: ASCO Annual Meeting; May 29-June 2, 2015; Chicago.
Trotman J, et al. Lancet Hematol. 2014;doi:10.1016/S2352-3026(14)70008-0.
For more information:
John Sweetenham, MD, FACP, FRCP, is HemOnc Today’s Chief Medical Editor for Hematology. He also is senior director of clinical affairs and executive medical director at Huntsman Cancer Institute at the University of Utah. He can be reached at firstname.lastname@example.org.