A Comprehensive Geriatric Assessment Is More Effective Than Clinical Judgement to Prospectively Identify Elderly Patients with Diffuse Large Cell Lymphoma Who Can Tolerate Full-Dose Anthracyclin-Based Treatment and Achieve the Same Outcome as Young Patients.

Introduction: Defining the optimal treatment strategy for elderly patients (pts) with diffuse large cell lymphoma (DLCL) represents a major challenge, since age, end-organ damage and comorbidities often preclude the delivery of full-dose intensive therapy. Objective means to prospectively identify those pts who can be safely treated with a curative intent are still lacking. We have analysed the performance of a comprehensive geriatric assessment (CGA) in defining those elderly DLCL pts which could tolerate an intensive and potentially curative treatment approach.

Methods: In addition to staging procedures, in all consecutive pts aged >65 with newly diagnosed DLCL a CGA was prospectively performed, including assessment of activity of daily living (ADL), instrumental ADL (IADL), comorbidity score, and geriatric syndrome, and pts were classified in the category of “fit” vs “frail”, i.e. potentially able vs unable to tolerate intensive treatment. However the decision to treat pts with intensive, anthracyclin-based, chemotherapy (CT) (CHOP/CHOP-like regimens +/− Rituximab) vs palliation (radiotherapy, low-dose CT or corticosteroids) was based on staging and clinical judgement only, irrespective of the results of CGA.

Results: From January 2003 to December 2006, 88 pts aged >65 were consecutively diagnosed with DLCL at our Institution and 84 had fully evaluable data. Their median age was 73 (range 66–89), 66% were in stage III-IV, 32% had B symptoms and 63% had an IPI score int-high or high. Based on clinical judgement, 62 pts (74%) received full-dose therapy and 22 received palliation. The proportion of pts treated aggressively was identical to that recorded in elderly DLCL pts diagnosed between 1995 and 2002. Their response rate (RR) (79,7% vs 55%; P=0.042), 2-year progression-free survival (PFS) (55,9% vs 22,2%; P=0.0002) and overall survival (OS) (57,7% vs 26,1%; P=0.0014) were significantly better compared to pts receiving palliation. According to CGA, 42 pts were classified as “fit” (50%) and 42 as “frail” (50%). The two subgroups significantly differed in mean age (70,8 vs 76,3; P<0.001) but not in lymphoma-related prognostic variables, including IPI and stage.”Fit” pts obtained significantly better RR (92,5% vs 48,8%; P<0.0001), 2-year PFS (73,4% vs 21,7%; P<0.0001) and OS (77,6% vs 23,8%; P<0.0001) compared to “frail” pts. Based on clinical judgement, all pts classified as “fit” as well as 20 of 42 pts classified as “frail” by CGA were treated aggressively, with curative intent. Remarkably, the 20 “frail” pts actually receiving aggressive treatment fared as poor as those given palliation only (2-year OS: 19,8% vs 26,1%; P= 0.85), lymphoma rather than toxicity being the main cause of failure also in this subgroup. Overall, the survival of patients identified by CGA as “fit” was significantly better compared to that of pts treated intensively based on clinical judgement only, both after 2003 (P=0.049) and between 1995 and 2002 (P=0.027).

Conclusion: By CGA approximately one half of unselected elderly pts with DLCL are classified as “fit” and one half as “frail”. Compared to clinical judgement, performing a CGA seems a more effective and objective tool to prospectively identify those pts which can be safely treated with full-dose immuno-CT and can achieve an outcome similar to that of younger DLCL pts. Alternative approaches to the category of pts identified as “frail” are warranted.