The article “Diffuse Large B-Cell Non-Hodgkin Lymphoma in the Very Elderly: Challenges and Solutions” by Latta and colleagues in this issue of ONCOLOGY does a nice job of focusing attention on a very substantial and probably underappreciated problem in the field of malignant hematology. Indeed, it is refreshing to see a review article on a highly treatable disease in which the authors do not focus on how drug X is 5% or 10% better than drug Y. Rather, they have highlighted the concern that lessons learned about small outcomes differences from large clinical trials in diffuse large B-cell lymphoma (DLBCL) don’t apply well to a substantial fraction of DLBCL patients—those with very advanced age. While there is no consensus on the definitions of “elderly” or “very elderly,” perhaps, as these authors suggest, the difficulties involved in treating patients with advanced age/frailty are not best measured by age. There are bigger fish to fry in this arena, and it is probably wise not to fret over the definitions and instead to think carefully about the challenges.
Delivery of R-CHOP (rituximab [Rituxan] plus cyclophosphamide(Drug information on cyclophosphamide), doxorubicin(Drug information on doxorubicin), vincristine [Oncovin], and prednisone(Drug information on prednisone)) in elderly patients with DLBCL is a significant public health issue. Prospective studies from research centers have left little doubt that R-CHOP is the best known therapy for the disease in fit elderly patients.[1,2] There is a long-held presumption, and some supporting evidence from the CHOP literature, that more intense chemotherapy is better than less.[3,4] However, studies of Medicare billing records and population registries suggest that a troubling fraction of older people (presumably those who are not “fit”) do not receive R-CHOP—or any chemotherapy at all.[5-7] When R-CHOP is delivered, it is given at a lower dose intensity than the densities used in published trials, and advanced age correlates strongly with lower dose intensity—presumably due to patients’ or physicians’ toxicity concerns. Data presented in the Latta review and elsewhere argue that there is potential for some meaningful benefit to the large untreated fraction if we can overcome barriers to treatment.
While new therapeutic options with less toxicity certainly hold an important key to mitigating the plight of elderly patients with DLBCL, at least three important obstacles are hindering progress in treating these patients with currently available options: 1) difficulties in interpreting the available literature, given the imperfect choice of endpoints such as overall survival (OS) and event-free survival (EFS); 2) an incompletely justified emphasis on the importance of dose intensity; and 3) an unfortunate terminology tradition in which references to management goals are limited to cure or palliation—which suggests that treatment of DLBCL is an all-or-nothing phenomenon.
In their article, Latta et al review published retrospective experiences in very elderly patients with DLBCL, and they make the summary observation that these patients are treated less frequently and have less favorable outcomes, with worse OS and progression-free survival (PFS) compared with younger patients. PFS and OS, while conventionally the gold-standard endpoints in DLBCL, are not ideal for measuring the value of therapy in the very elderly, due to competing risks that affect each of them. The cited article from Lyon reported a median OS of 1.3 years, but 30% of deaths were unrelated to lymphoma. For these patients we need more precise information on the impact of the lymphoma on outcomes; thus, for measuring survival it is preferable to look at relative survival or perhaps lymphoma-specific survival.[11,12] The relatively poor outcomes as measured by PFS ond OS in series of older patients dampen the enthusiasm of clinical decision makers for considering intervention.
Significant attention has been focused on the importance of dose intensity in DLBCL, particularly in elderly patients. Nearly all retrospective studies to date have shown a correlation between reduced dose intensity of cyclophosphamide and/or anthracycline and inferior outcomes as measured by OS.[13-17] Unfortunately, reduced dose intensity is also correlated with advanced age, worse performance status, and multiple comorbidities—all of which have confounding effects on OS. Furthermore, many of the studies historically cited were performed prior to the chemoimmunotherapy era. The best studies assessing the importance of dose intensity will include lymphoma-specific survival as an outcome. Prospective studies that formally assess the importance of dose intensity are generally lacking, although the report by Meguro et al (cited by Latta et al) is not supportive of a critical minimum dose intensity. Population registry–based reports of patterns of DLBCL treatment in the elderly describe survival outcomes for anthracycline omission in the immunochemotherapy era that are essentially as good as the outcomes for anthracycline-based regimens in the pre-immunochemotherapy era.[6,19] Clinicians and patients might be more amenable to some degree of anti-lymphoma therapy if they did not feel that strategies other than full doses of anthracycline were futile.
Finally, treating the elderly requires redefining goals. For most clinicians, DLBCL is thought of as an aggressive malignancy that rapidly threatens a patient’s life if not brought to complete and durable remission. There is little tolerance for the notion of chronic coexistence with partially treated disease, or prolonged survival after relapse in the absence of aggressive salvage therapy. This leads to an implicit expectation that in the absence of “cure,” the only alternative is “palliation”—which is highlighted in the terminology used throughout the Latta review. While this is certainly true for some patients, the heterogeneity of DLBCL that we all appreciate pathologically also exists clinically, and some patients will have disease that is controlled for a few years but not eradicated. In the immunochemotherapy era, the best long-term experience with elderly patients in DLBCL is the excellent series of follow-up reports of the Groupe d’Etudes des Lymphomes de l’Adulte (GELA) LNH-985 trial. With over 10 years of follow-up on all living patients, a recent report notes that only 6 of 14 patients who relapsed after 5 years died of lymphoma.
The most important messages emphasized by Latta and colleagues are easy to support and form the basis of a consensus. Very elderly patients with DLBCL represent a special challenge that is probably best addressed by extra attention rather than by discouraging therapy. Patients should ideally undergo very thoughtful evaluation with tools such as the comprehensive geriatric assessment or something similar that will give the treating physician the best estimate of therapy tolerance. Once therapy is initiated, very frequent monitoring is useful for providing early support of toxicity and for most promptly and effectively informing appropriate dose adjustments to ensure optimal dosing. It is important to keep in mind, however, that there are not yet strong enough data to suggest that any level of dose intensity is futile, and all involved in treatment decisions should be aware that goals other than cure can still add meaningful quality and quantity of life for these patients.
Financial Disclosure: Dr. Link serves on the scientific advisory board of Genentech, and he is a compensated consultant for Genentech, Millenium, and Seattle Genetics.