Pfreundschuh et al recently presented the results of a German High-Grade Non-Hodgkin Lymphoma Study Group (DSHNHL) trial detailing outcomes with 2- or 3-weekly cyclophosphamide, doxorubicin, vincristine, and prednisone [CHOP] chemotherapy, with or without etoposide, in elderly patients with aggressive non-Hodgkin lymphoma (NHL).1 The predefined analyses failed to demonstrate a significant superiority of any of the regimens in terms of the primary end point of time to treatment failure (TTF). However, unplanned multivariate analyses showed a relative risk reduction for event-free and overall survivals after CHOP-14 (biweekly) compared with CHOP-21 (every 3 weeks). Another study by the Japan Clinical Oncology Group (JCOG) was terminated early by the Data and Safety Monitoring Committee (DSMC)2 because the first interim analysis revealed shorter progression-free survival with CHOP-14 compared with CHOP-21 (24 vs 34 months, respectively). Furthermore, CHOP-14 was only marginally superior to CHOP-21 in a group of younger patients with good-risk aggressive lymphoma in another DSHNHL study.3 Thus, it seems premature to conclude that CHOP-14 should be considered the new standard for elderly patients with diffuse large B-cell lymphoma (DLBCL), and other regimens may challenge this title.
The DSHNHL trial randomized 831 patients, but efficacy and safety analyses were performed on only 689 patients with confirmed histology of aggressive lymphoma.1 The majority of the patients had low-risk disease defined as an age-adjusted international prognostic index (aaIPI) of 1 or less, and fewer patients had adverse characteristics than in our study comparing CHOP plus rituximab (R-CHOP) with CHOP alone in patients ages 60 to 80 years.4,5 Of our patients, 60% had high-risk disease (aaIPI ≥ 2; Table 1). Patients were analyzed as intent-to-treat (all patients) compared with trial-defined population in the DSHNHL study. After a median follow-up of 4 years, event-free and overall survivals are significantly longer for R-CHOP patients (P = .00008 and P = .008, respectively).6 Survival benefit with R-CHOP was found independent of age or comorbidity scores.7 Importantly, these improved outcomes were achieved with minimal additional toxicity.2,6
The benefit of adding rituximab to CHOP has been confirmed by several large randomized trials. A US Intergroup study randomized DLBCL patients aged 60 years or older to receive either 6 to 8 cycles of CHOP or R-CHOP chemotherapy.8 Responding patients were randomized to receive either rituximab maintenance or no further treatment. If this study was not designed to directly compare R-CHOP with CHOP, additional analyses reveal that R-CHOP induction significantly prolongs TTF and overall survival.8 A large historical population-based analysis was conducted in British Columbia comparing CHOP or R-CHOP in DLBCL patients and confirmed the Groupe d'Etude des Lymphomes de l'Adulte (GELA) results.9 The MabThera (rituximab) International Trial (MInT) determined whether the benefits seen with R-CHOP in elderly patients could be extended to younger patients with DLBCL. A preplanned interim analysis has demonstrated a statistically significant improvement in TTF for patients receiving rituximab.10 Accordingly, further randomization has been stopped by the DSMC.
These results indicate that the addition of rituximab to CHOP is associated with substantial and repeatedly demonstrated clinical benefit and that immunochemotherapy should be considered as the new standard of care in DLBCL patients.
A different view of “standards” in the treatment of aggressive lymphomas
Several points discussed in the letter by Coiffier and Salles need clarification:
1. Contrary to the statement of Coiffier and Salles and as described in our paper,1 the analysis of the NHL-B2 trial was preplanned and performed according to the protocol that provided exact biometrical guidelines for the case that a 2 × 2 factorial design would be impossible due to a relevant interaction term (which was indeed the case in the NHL-B2 trial). Contrary to the Coiffier's and Salles' statement, there was a highly significant advantage of the 2-weekly CHOP-14 over the classical 3-weekly CHOP with respect to both primary (event-free survival) and secondary (rates of complete remission, progression under therapy, and overall survival) end points.
2. Contrary to the statement of Coiffier and Salles, the Japan Clinical Oncology Group (JCOG) trial was not stopped because the first interim analysis revealed shorter progression-free survival, but rather because an advantage of the 2-weekly CHOP over 3-weekly CHOP was calculated to be unlikely. The JCOG trial can hardly be compared with the NHL-B2 trial in elderly patients (ages 61-75 years; median, 68 years) because the study population (ages 15-69 years; 75% low and low-intermediate risk according to the international prognostic index [IPI]) resembles more the study population of the NHL-B12 trial for young low-risk (low lactate dehydrogenase [LDH]) patients (ages 18-60 years). Besides several other major flaws (< 70% aggressive lymphomas; 40% without restaging), the most important difference between the Japanese and the German trials is the adherence to the protocol. While the 2-weekly schedule was well adhered to in the Japanese trial, there were significant dose reductions: for example, 27% of the patients in the Japanese trial received less than 90% of the planned doxorubicin dose and 20% received even less than 80%; the figures in the German trials are 9% and 2% (NHL-B1) and 11% and 9% (NHL-B2), respectively. This demonstrates that the 2-weekly CHOP-14 must be given at both the planned dose and time intervals to maintain its superiority to the 3-weekly regimen.
3. Interval reduction from 3 to 2 weeks works best in high-risk patients and in particular in those with an elevated LDH level; hence, it is not surprising that the benefit of the 2-weekly regimens was smaller in the NHL-B1 trial (in which only low-risk patients with normal LDH level were included) than in the NHL-B2 trial (which included both low-risk and high-risk patients).
4. If the authors had read our paper more carefully they would have realized that we never claimed that CHOP-14 should be considered “the new standard” of care for elderly patients; rather, we suggested CHOP-14 as the new standard chemotherapy regimen for this population. Since CHOP-14 is the first chemotherapy regimen that proved to be superior to classical CHOP-21 for elderly patients with aggressive lymphoma without additional clinically relevant toxicity, we think that this statement is valid.
5. There are indeed differences between the elderly population in the Groupe d'Etudes des Lymphomes de l'Adulte (GELA)3 trial and the NHL-B2 trial. However, since the interval reduction from the 3-weekly to the 2-weekly regimen is of particular benefit for high-risk patients, the differences in favor of CHOP-14 over classical CHOP-21 (which are similar to the ones obtained by adding rituximab to the 3-weekly CHOP-21) would have been even more pronounced in a study population with a poorer prognostic risk profile.
6. While rituximab might be of benefit for elderly patients in combination with CHOP-21 for bcl-2–positive patients, the value of rituximab in bcl-2–negative patients remains to be shown.4
7. The authors' interpretation of the US Intergroup study is surprising. In our opinion, this trial neither confirms nor disproves the results of the GELA trial.
8. The positive effect of the MabThera (rituximab) International Trial (MInT) trial in young patients with low-risk diffuse large B-cell lymphoma (DLBCL) confirms the findings of the GELA trial3 that rituximab is most beneficial in low-risk disease.
9. A benefit of rituximab in combination with CHOP-14 remains to be shown. A planned interim analysis of the first 500 patients (ages 61-80 years) treated within RICOVER-60 trial of the German High-Grade Non-Hodgkin Lymphoma Study Group (DSHNHL; which compares 6 and 8 cycles of CHOP-14, both with and without rituximab in a randomized fashion) did not reveal any differences that would have allowed for the early termination of the trial. This indicates that if differences between CHOP-14 and R-CHOP-14 exist at all, these differences must be considerably smaller than the ones observed in the GELA trial (with only 399 patients).
10. The question of whether rituximab can improve results of CHOP-14 will not be answered until the analysis of the completed RICOVER-60 trial (1220 patients), which should be available in summer 2005. Moreover, there are no data on the addition of rituximab to intensified chemotherapy regimens in young high-risk patients, and the respective Mega-CHOEP (CHOP + etoposide) trial of the DSHNHL has just started recently.
11. Before the results of the RICOVER-60 and Mega-CHOEP trials are available, there is no scientifically justified claim of a “standard care” for elderly patients and young high-risk patients with DLBCL.
In summary, treatment of aggressive lymphoma is more complex than Coiffier and Salles try to make us believe. For the time being, the conclusion of Coiffier's and Salles' letter with the generalizing claim of a “standard of care in DLBCL” lacks scientific evidence and must be rejected.
Correspondence: Michael Pfreundschuh, Med Klinik I, Universität des Saarlandes, D-66421 Homburg, Germany; e-mail: michael.pfreundschuh@uniklinik-saarland.de
B.C. is part of the speaker bureau of Roche and Genentech. G.S. is part of the speaker bureau of Roche. The GELA has received unrestricted grants from Roche and Genentech.
