Maintenance rituximab every 2 months is more toxic than every 3 months in patients with non-Hodgkin lymphoma

Armin Rashidi, Eunhye Oak and Nancy L. Bartlett

To the editor:

The phase III PRIMA study of patients with previously untreated follicular lymphoma randomly assigned to maintenance rituximab (MR; every 8 weeks for 2 years) vs observation following rituximab-based immunochemotherapy (rituximab plus cyclophosphamide, vincristine, and prednisone [R-CVP], rituximab plus cyclophosphamide, doxorubicin, vincristine, and prednisone [R-CHOP], or rituximab plus fludarabine, cyclophosphamide, and mitoxantrone [R-FCM]) demonstrated an improvement of 16.5% in 6-year progression-free survival (PFS) with MR (59.2% vs 42.7%; P < .001) as well as increased frequency of grade 3 to 4 adverse events (24% vs 17%; most commonly infections).1,2 These results established MR as a standard-of-care option in such patients, and MR is now empirically generalized to accompany other rituximab-based induction regimens in treating indolent non-Hodgkin lymphomas and mantle cell lymphoma. For example, the increasingly popular regimen of bendamustine and rituximab (BR) is often followed by MR every 2 months for 2 years. In our recent experience with BR, we noticed an apparently increased incidence of toxicities during maintenance that required delay or discontinuation of MR. We conducted a retrospective study to determine the predictors of MR toxicity. We hypothesized that the use of BR (vs R-CHOP/R-CVP) for induction and/or MR administered every 2 months (vs every 3 months) are associated with increased MR toxicity.

We reviewed the medical records of all patients with B-cell indolent non-Hodgkin lymphoma or mantle cell lymphoma treated at our institution with R-CHOP/R-CVP or BR as first-line induction followed by MR (375 mg/m2) every 2 months or every 3 months for a maximum of 2 years. A total of 148 patients who received their first dose of induction after January 1, 2006, and either completed or discontinued the planned maintenance were included (Table 1). The most common diagnosis was follicular lymphoma (82%), and approximately half the patients (44%) received BR. The planned maintenance schedule was every 2 months in 28% of patients and every 3 months in the remaining patients. The decision to offer one schedule of MR vs another was largely based on whether patients were treated before or after the PRIMA study.

Table 1

Patient characteristics in groups with and without MR toxicity

Toxicities attributed to MR occurred in 33 patients (22%) (76% grade 3-4, 18% grade 1-2, 6% grade unknown). Neutropenia (7%), infection (6%), or both (7%) accounted for 88% of all toxicities (supplemental Table 1). A delay/omission of at least 1 dose of MR (but not discontinuation of treatment) was required in 18 patients (12%), with MR toxicity being the reason for delay/omission in 16 (89%). MR was prematurely discontinued in 40 patients (58% of those with and 18% of those without MR toxicity; P < .001), most commonly as a result of toxicity (45%; n = 18) or disease progression (30%; n = 12). A variety of other uncommon reasons (eg, intravenous access, transportation difficulties, fatigue) led to discontinuation in the remaining minority.

In univariate analysis, both the induction regimen and maintenance schedule were significantly associated with MR toxicity. Of the patients treated with BR, 31% experienced toxicity during MR compared with 16% of those who received R-CHOP/R-CVP (P = .046). Similarly, toxicity was observed in 41% of patients in the every-2-months cohort vs 15% in the every-3-months cohort (P = .002). In multivariate binary logistic regression using the maintenance schedule and induction regimen as predictors, only the every-2-months schedule remained statistically significant (odds ratio, 3.42; 95% confidence interval, 1.36-8.60; P = .009). The univariate association between BR and MR toxicity likely reflects an increase in BR use that coincided with adoption of every-2-months MR. With a median follow-up of 53 months, the median PFS has not been reached. PFS was not significantly associated with MR toxicity (P = .54) or induction regimen (P = .34). However, there was a trend toward higher 3-year PFS in the every-3-months compared with the every-2-months cohort (91% vs 83%, respectively; P = .09) and in the group who completed MR compared with those who discontinued MR for reasons other than progression (97% vs 86%, respectively; P = .09). There were no significant differences between the every-2-months and every-3-months cohorts other than the higher frequency of BR in the former (80% vs 30%; P < .001).

This study is the first comparison, albeit retrospective, of every-2-months vs every-3-months MR. Patients in the every-2-months cohort were 3.4 times more likely to experience toxicity and showed a trend toward shorter PFS, which may be the result of more dose delays/omissions and, ultimately, early treatment discontinuation. A randomized prospective study comparing the 2 schedules in terms of toxicity and PFS is warranted.


Acknowledgments: N.L.B. is supported in part by the Barnes-Jewish Hospital Foundation.

Contribution: A.R. and E.O. collected and analyzed the data; A.R., E.O., and N.L.B. wrote the letter.

Conflict-of-interest disclosure: The authors declare no competing financial interests.

Correspondence: Nancy L. Bartlett, Division of Oncology, Washington University Medical School, 660 South Euclid Avenue, Campus Box 8056, St. Louis, MO 63110; e-mail: nbartlet{at}


  • The online version of this article contains a data supplement.