Recent data from phase III clinical trials showed that in previously untreated patients (pts) with indolent (low-grade) non-Hodgkin’s lymphoma (iNHL) bendamustine plus rituximab (BR) resulted in superior progression-free survival (STiL NHL 1-2003) and non-inferior response rates (STiL NHL 1-2003 and BRIGHT) compared to R-CHOP. Since clinical trials are restricted to highly selected pts, we here investigated effectiveness of BR and R-CHOP in unselected pts treated in routine practice by German office-based haematologists.


The open, longitudinal, multicentre, clinical registry on lymphoid neoplasms (TLN Registry, registry NCT00889798) prospectively collects data on the treatment of pts with lymphoid B-cell neoplasms as administered by a network of German office-based haematologists. Pts are followed for 5 years. A broad set of data regarding patient and tumour characteristics, comorbidities, all systemic treatments and response rates, progression-free survival and overall survival are recorded. Automated plausibility and completeness checks with subsequently generated queries by the electronic data capture system ensure data reliability. In addition, data managers regularly check for plausibility and issue queries. Since May 2009, 111 sites have recruited a total of 2897 pts.


633 pts with iNHL (52% follicular, 13% mantle cell lymphoma), recruited at the onset of their 1st-line therapy and treated with BR (82%) or R-CHOP (18%), were included in this analysis. The choice of the regimen was upon the decision of the treating physician in accordance with the patient´s informed consent.

Pts were median 69 years (yrs) old (range 24-93 yrs), 54% were male, 55% had tumour stage IV (Ann Arbor), 24% presented with B symptoms, 25% with bulky disease, and 61% with at least one comorbidity.

Clinical and tumour characteristics differed between pts receiving BR or R-CHOP: Pts treated with BR were older (median 70 vs. 61 yrs; p<0.0001), presented more often with stage IV disease (59% vs. 40%; p=0.0002) or comorbidities (63% vs. 50%; p=0.009), whereas pts treated with R-CHOP were more often diagnosed with follicular lymphoma (72% vs. 48%; p<0.0001) and presented more frequently with bulky disease (39% vs. 21%; p=0.0003).

Objective response rate (ORR) as assessed by the local site was comparable between the two regimens: 91% of pts receiving BR (39% complete response (CR)) and 94% receiving R-CHOP (43% CR) responded to 1st-line therapy. Both regimens were applied with median 6 cycles.

In univariate analyses young age, male sex, follicular subtype and absence of comorbidities were significantly associated with an objective clinical response to the 1st-line regimen. In a multiple logistic regression analysis adjusted for type of 1st-line regimen (BR vs. R-CHOP) and age at the onset of therapy, the likelihood for response was lower for older pts (OR=0.96; p=0.017), while the type of 1st-line regimen had no effect (OR=1.19; p=0.738). At this point, the small number of non-responders (n=36) precluded analyses of more than two potential confounders.

After a median follow-up of 15 months (maximum 39 mth), 94% of pts receiving BR are alive and 4% received 2nd-line therapy. In pts receiving R-CHOP 91% are alive and 9% pts received 2nd-line therapy. Overall, 4% are lost to follow-up.


Our data show that previously untreated pts with iNHL receiving BR or R-CHOP in routine practice differ, with BR preferentially given to pts with a less favourable prognostic profile. Nevertheless, response rates to 1st-line treatment with BR or R-CHOP appeared to be comparable. These results support response data from the NHL 1-2003 (StiL) and the BRIGHT study.

BR: bendamustine + rituximab ± prednisone │ R-CHOP: cyclophosphamide + doxorubicin + vincristine ± prednisone + rituximab


Knauf:Mundipharma, Janssen, Roche Pharma: Consultancy, Honoraria.

Author notes


Asterisk with author names denotes non-ASH members.

Sign in via your Institution