From 9/98 to 11/99, 126 patients with symptomatic previously untreated or first relapse (< 6 months of chlorambucil and/or local radiotherapy) CD20+ low-grade lymphoma, were included in a multicenter randomised phase II study. The treatment consisted of a first cycle of rituximab 375 mg/sqm q wk x 4. Pts in CR at week 14 were observed with no further treatment until symptomatic relapse, while pts with SD or PD went off study. Pts with PR or minor response were randomised to receive either a second cycle of rituximab 375 mg/sqm q wk x 4 or interferon-alpha-2a (IFN) 3 MIU/day sc (wk 1), 4,5 MIU/day (wk 2–5) in combination with rituximab 375 mg/sqm q wk (w 3–6). The clinical data from this study has previously been reported (

Kimby E, et al.
Ann Oncol
(Suppl 2):
). 38 patients (30%) fulfilled the criteria for CR, and were eligible for analysis of minimal residual disease (MRD). 14 more patients achieved CR at a time point later than first follow up after end of treatment. Per protocol, these patients are not included in the present analysis. By standard DNA-based PCR, presence of either a t(14;18) fusion transcript (MCR/mbr) or a clonal rearrangement of the Ig heavy chain (CDR3) could be detected in the diagnostic bone marrow and blood sample from 23 patients. These patients have now been studied for MRD, with a median follow-up time of 62 months. In dilution experiments the sensitivity of the assays was between 1:10−3 and 1:10−4. A given sample was considered negative if the PCR reaction was negative in three independent experiments, using up to 2 μg of template DNA. Patients were tested in blood and bone marrow at 10–16 weeks, 38–40 weeks and 52 weeks following treatment. A total of 175 samples, including 49 samples from patients in continued CR up to 5 years after treatment, have been analysed. Of 72 paired blood and bone marrow samples, only three showed inconsistency between blood and bone marrow, all three being positive in bone marrow and negative in blood. The frequency of MRD negativity 10–16 weeks after treatment was 4/9 (44%) in patients who received 1 cycle of rituximab, 3/5 (66%) in patients who received two cycles of rituximab and 7/9 (77%) in patients who received two cycles of rituximab with IFN priming. This trend towards a dose-response relation was however not significant, due to the small number of patients in each treatment group. The median duration of CR in patients who were negative at all three timepoints during the first year (n=14) was 62 months, compared to 21 months in patients (n=9) with one or more positive samples (p<0,005). At a median follow up of 62 months 9/14 patients who were MRD negative through the first year remain in complete molecular remission, compared to 1/9 patients who had one or more positive blood or bone marrow samples during the first year (p<0,03). Thus, sustained long-term complete molecular remissions are achievable with rituximab alone or in combination with IFN, and predictable by MRD status during the first year post treatment. Whether the quality of response is related to the dose of rituximab or the combination with IFN, and whether the response can be predicted using blood samples alone, must await the results of the ongoing ML16865 randomised phase III trial of rituximab vs IFN/rituximab in the same group of patients.

Author notes

Corresponding author