Outcome of Observation As Initial Strategy in Patients with Mantle Cell Lymphoma

Background

Patients with mantle cell lymphoma (MCL) follow a heterogeneous clinical course ranging from very indolent to very aggressive. While patients with MCL generally require treatment initiation shortly after diagnosis, it is unclear whether deferring treatment in patients with "indolent" MCL affects their overall outcome. Because it is difficult to identify such patients at the time of diagnosis, their course can only be retrospectively described as indolent after a prolonged period of observation. The aim of this study was to describe the subgroup of patients with MCL who underwent observation as their initial management, including their clinical and biological characteristics and outcomes.

Methods

Patients diagnosed with MCL from 1998-2015 who were initially observed for ≥3 months from the date of definitive diagnosis were identified in the BCCA Lymphoid Cancer and Pharmacy Databases. Pathology was centrally reviewed at the time of diagnosis, and only cases positive for CCND1 by immunohistochemistry and/or t(11;14) by FISH were included. During the study period, there were no predefined criteria guiding observation or active treatment. Eventual treatment indications included high tumor burden, disease associated symptoms or peripheral blood cytopenias. Clinical-biological features at diagnosis, treatment and outcomes, were analyzed.

Results

A total of 725 patients with MCL were initially identified, but 286 were excluded: missing data (n=179), treatment refusal (n=7), no treatment due to frailty (n=16), or absence of CCDN1 or t(11;14) confirmation (n=84). 365 (83%) patients received early treatment (ET) and 74 (17%) were observed >3 months (OBS), as shown in Table 1. In the OBS group, 52 (71%) patients had measurable lymph nodes at presentation, 16 (22%) a non-nodal presentation (defined as peripheral blood, bone marrow, and/or spleen only), and 5 (7%) only had gastro-intestinal involvement. Patients in the OBS group were older, with favorable presenting features including good performance status, less frequent B symptoms or increased LDH, and lower Ki67 (<30% vs ≥ 30%) than the ET group. However, MIPI scores were similar between both groups. The majority of patients received rituximab-containing chemotherapy (most commonly R-CHOP or R-bendamustine) at the time of initial treatment in both the ET group (70%) and the OBS group (72%).

In the OBS cohort, with a median follow-up of 47 months (range 3.4 - 158 months) in living patients, the median time to treatment (TtT) was 35.5 months (range 5 - 79 months). 10 patients (14%) were observed for > 5 years without requiring treatment. Factors associated with longer TtT included clinical presentation (non-nodal vs nodal, median not reached vs 29 months; P=.005) and Ki-67 (<30% vs ≥ 30%, median 59 vs 20 months, P=.033). Median OS was significantly longer in the OBS group than in the ET group (66 vs 50 months, respectively, P=.024) reflecting the more favorable disease characteristics of the OBS group. Clinical presentation (ie, non-nodal vs nodal) was the only factor associated with OS (median 123 vs 47 months, P=.003) in the OBS group. Finally, the median OS from date of treatment initiation for patients eventually requiring therapy in the OBS group was 34.4 months. With a median age at treatment initiation of 71 yrs (range 40 - 91 yrs) in the OBS group, OS was not significantly different in comparison with the ET group when the analysis was adjusted by age at treatment.

Conclusions

A subgroup of patients with MCL may be safely observed at diagnosis of the disease without negatively impacting their outcomes, including not only those patients with non-nodal presentation but also asymptomatic patients with low burden nodal presentation, particularly those with a low proliferative rate.