- © 2008 by American Society of Clinical Oncology
Impact of Rituximab and/or High-Dose Therapy With Autotransplant at Time of Relapse in Patients With Follicular Lymphoma: A GELA Study
- Catherine Sebban,
- Pauline Brice,
- Richard Delarue,
- Corinne Haioun,
- Bertrand Souleau,
- Nicolas Mounier,
- Nicole Brousse,
- Pierre Feugier,
- Hervé Tilly,
- Philippe Solal-Céligny and
- Bertrand Coiffier
- From the Centre Léon Bérard, Lyon; Hôpital Saint-Louis, Assistance Publique, Paris; Hôpital Necker, Assistance Publique, Paris; Hôpital Henri Mondor, Assistance Publique, Créteil; HIA Percy, Clamart; l'Archet, Centre Hospitalier Universitaire, Nice; Hôpital Necker, Paris; Hôpital de Brabois, Vandoeuvre les Nancy; Centre Henri Becquerel, Rouen; Clinique Jean Bernard Le Mans; and the Centre Hospitalier Lyon-Sud, Hospices Civils de Lyon, Lyon, France
- Corresponding author: Bertrand Coiffier, MD, Centre Hospitalier Lyon-Sud, Service d'Hématologie, Pierre Benite, Ce, France 69495; e-mail: bertrand.coiffier{at}chu-lyon.fr
Abstract
Purpose The treatment of patients with follicular lymphoma has changed with the introduction of high-dose therapy (HDT) with autologous stem-cell transplant then with rituximab. The effect of these two strategies on the outcome of relapsing patients with follicular lymphoma has never been compared.
Patients and Methods We analyzed two cohorts of patients treated in two successive randomized studies with the same treatment, cyclophosphamide, doxorubicin, teniposide, and prednisolone plus interferon, to evaluate the role of rituximab and HDT in salvage therapy after first disease progression or relapse.
Results Of the 364 patients included in these two studies, 254 progressed or relapsed and constitute the population of this analysis. Among them, 98 had been treated with HDT, including 33 of them after rituximab-containing salvage regimen, and 69 with rituximab alone or combined with chemotherapy but without HDT. Patients’ characteristics at diagnosis were similar in all subgroups. If event-free survival was identical for patients treated within Groupe d'Etude des Lymphomes Folliculaires (GELF) -86 or GELF-94 studies, overall survival was longer in GELF-94 study. HDT was associated with a statistically significant benefit in terms of event-free survival from relapse and survival after relapse (SAR). Rituximab was associated with a greater benefit than HDT for these two end points. When both treatments were combined, patients treated with rituximab-containing salvage regimen followed by HDT had 5-year SAR more than 90%.
Conclusion In follicular lymphoma, for patients treated with first-line chemotherapy the combination of a salvage regimen containing rituximab with or without HDT leads to a dramatic improvement of long-term outcome.
INTRODUCTION
Follicular lymphoma accounts for 20% to 25% of all lymphomas with a median overall survival of 9 to 10 years.1,2 Most of the patients relapsed and until recently there was no curative option. However, new treatment modalities, such as monoclonal antibodies or autologous and allogeneic stem-cell transplantation, have been developed and tested in first-line therapy or in relapse. The real long-term impact of these therapies on survival is difficult to establish. Some data are consistent with the absence of significant improvement in survival during the past 30 years.3 However, in a large population-based registry compiled in the United States between 1978 and 1999, an improvement of median survival from 84 to 93 months was observed as the result of the sequential application of effective therapies and the improvement of supportive care.4 A significant improvement in failure-free survival and overall survival (OS) has also been reported for stage IV follicular lymphoma treated in the M.D. Anderson Cancer Center (Houston, TX) between 1972 and 2002 by analyzing successive prospective studies.5 Recently, a retrospective analysis of Southwest Oncology Group trials for patients treated for follicular lymphoma found better OS rates in recent studies compared with older studies.6
In this study, we have analyzed the outcome of two successive cohorts of patients with high tumor burden follicular lymphoma treated by the same induction and consolidation regimen within two randomized controlled trials run between 1986 and 1994 Groupe d'Etude des Lymphomes Folliculaires (GELF-86 study) and 1994 to 2001 (GELF-94 study) by the Groupe d'Etude des Lymphomes de l'Adulte (GELA).7,8 The objective of this analysis was to demonstrate if these salvage treatments provide any benefit and, if so, its magnitude in terms of OS and event-free survival (EFS).
PATIENTS AND METHODS
Patients
The patients included in this analysis were those included in the two prospective studies conducted by the GELA: GELF-86 and GELF-94. The GELF-86 group 2 study was a randomized controlled study including patients age 18 to 70 years with follicular lymphoma and criteria of high tumor burden.7,9,10 High tumor burden according to the GELF criteria was defined by at least one of the following parameters: systemic symptoms (> 10% weight loss, > = 38 degrees for > 5 days, abundant night sweats); a single lymph node site larger than 7 cm; at least three nodal sites each of which with a diameter of more than 3 cm; a marked splenomegaly; organ failure; pleural effusion or ascitis; orbital or epidural involvement; blood infiltration or cytopenia. Patients were randomly assigned to receive CHVP (once per month for six cycles with cyclophosphamide 600 mg/m2 day 1, doxorubicin 25 mg/m2 day 1, teniposide 60 mg/m2 day 1, and prednisolone 40 mg/m2 from days 1 to 5, then in responders once every other month for six cycles of CHVP for a total of 12 cycles) or CHVP plus interferon. Interferon alfa (Intron A; Schering, Kenilworth, NJ) was given at the dosage of 5 MU subcutaneously 3 times per week for 18 months. CHVP plus interferon allowed significantly longer EFS and OS compared with CHVP alone and was considered to be the best standard regimen when designing the GELF-94 trial treatment.7
The GELF-94 group 2 study was a randomized controlled study including patients with follicular lymphoma and high tumor burden age 18 to 60 years.8 For definition of high tumor burden, high levels of lactate dehydrogenase or β2-microglobuline were added to the definition used in 1986. Patients were randomly assigned to CHVP plus interferon as described in the GELF-86 study or to four cycles of cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) followed by autologous stem-cell transplantation (ASCT) with a conditioning regimen including cyclophosphamide, etoposide, and total-body irradiation (TBI) of 10 Gy.8 With a median follow-up of 7.5 years, no benefit in terms of EFS or OS was observed in the intensive arm. All patients gave written informed consent for both studies.
Those patients treated in the CHVP plus interferon arm in both studies and younger than 61 years (ie, patients able to be treated with high-dose therapy [HDT] at time of relapse) were retained for this analysis. There were 155 such patients in the GELF-86 study and 209 patients in the GELF-94 study. Their median follow-up was 153 months and 91 months, respectively.
Methods
First we compared the outcome of these two patient cohorts with intent-to-treat analysis. Then, for patients who progressed during treatment or who relapsed, we collected data on the second-line treatment, particularly the type of salvage regimen, the use of rituximab, and the use of HDT with ASCT as consolidation regimen. There were not any recommendations in either study for any specific treatment at the time of progression or relapse. We compared the effect of these different regimens on the outcome of these patients, mainly EFS and OS from the date of disease progression.
Patient outcome was censored at the last contact date. OS was calculated from the date of random assignment until death or last follow-up. EFS was calculated as the time period from the random assignment to induction failure (stable disease at the end of treatment or progression during treatment), death irrespective of the cause, progression after partial response, relapse after CR, or last follow-up. Survival after first relapse (SAR) or progression was measured from the date of relapse or progression until last follow-up or death from any cause. Event-free survival after first relapse (EFSR) or progression was measured from the date of relapse or progression until last follow-up or next progression, change of salvage treatment, or death from any cause. OS, EFS, SAR, and EFSR were calculated with their product-limit method and the differences between the two survival curves were tested for significance with the log-rank test. The estimations were calculated with 95% CIs. Differences in patient characteristics were tested for significance by χ2 test. All tests for significance were at the 5% significance level and not adjusted for multiple comparisons. Two-sided P values are reported. All statistics were calculated in SAS, version 9 (SAS Institute, Cary, NC).
RESULTS
Comparison of GELF-86 and GELF-94
Appendix Table A1 (online only) displays the main clinical and biologic characteristics at diagnosis of the 364 patients who were retained from the two studies. There was no difference between the two cohorts. More than 75% of patients had stage IV disease and more than two thirds of them had a bone marrow involvement. The response after six cycles and at the end of therapy was similar in both studies (Table A1).
EFS was similar in the two studies (Fig 1): for GELF-86 and GELF-94 patients, 5-year and 10-year EFS were 34% (95% CI, 26% to 41%) and 20% (95% CI, 14% to 26%) versus 35% (95% CI, 29% to 42%) and 24% (95% CI, 18% to 31%), respectively (P = .71). However, OS was significant longer in the GELF-94 study compared with the GELF-86 study (Fig 1): 10-year survival rates were 66% (95% CI, 57% to 73%) compared with 48% (95% CI, 40% to 56%), respectively (P = .0128).
(A) Event-free survival and (B) overall survival for patients treated within the Groupe d'Etude des Lymphomes Folliculaires (GELF) -86 and GELF-94 studies. NA, not applicable.
Among these 364 patients, 273 had an event but only 254 met the criteria for inclusion in this retrospective analysis and had available data. Nineteen patients died without progression and were excluded from this analysis. Among these 254 patients, 83 had disease progression during treatment and 171 relapsed. Appendix Table A2 (online only) reports the therapeutic modalities performed for first relapse/progression in both cohorts. Salvage chemotherapy was given in 94% of GELF-86 patients but in only 83% of the GELF-94 patients. The most used regimens were dexamethasone, high-dose cytarabine, and cisplatin10a; ifosfamide, carboplatin, and etoposide10b; mesna, ifosfamide, mitoxantrone, and etoposide10c; and fludarabine-based regimens.10d,10e Because of drug availability, rituximab was given as treatment of first relapse/progression, associated with chemotherapy or alone, in only 2% of GELF-86 patients compared with 47% of GELF-94 patients. HDT with ASCT was performed as consolidation, with TBI or BEAM conditioning regimen, in 38% of GELF-86 patients and in 39% of GELF-94 patients. Allogeneic transplantation as treatment of first relapse/progression was uncommon in our group and was performed in only 10 of these patients. These 10 patients were included in the group of patients treated without HDT and ASCT. No rituximab maintenance was done in any of the patients.
Outcome of Patients With Progressive Disease or Relapse
For the following analyses, patients from both studies were grouped according to the treatment they received: chemotherapy alone, rituximab, and/or HDT. Table 1 presents the characteristics of the patients included in this analysis according to the salvage therapy. Few differences were observed between the different subgroups: patients treated with HDT were younger and had more frequently relapse than disease progression; patients treated with rituximab had a lower occurrence of B symptoms and a high Follicular Lymphoma International Prognostic Index (FLIPI) score at diagnosis. Patients treated with rituximab relapsed later than those not treated with rituximab. This last difference is likely related to the registration of rituximab in 1998 in France; most patients who progressed or relapsed early after the end of treatment had not the opportunity to receive the drug. As reported in Table A2, a limited number of patients received rituximab once per week for 4 weeks due to low tumor burden relapses. The other patients received induction salvage chemotherapy alone or with rituximab at each cycle of chemotherapy.
Characteristics of the Patients at Diagnosis in the Different Therapeutic Subgroups
Table 2 reports the main outcome according to salvage treatment at relapse. SAR was significantly longer in the GELF-94 study with 5-year SAR of 61% (95% CI, 52% to 69%) versus 45% (95% CI, 35% to 54%) for GELF-86 patients. Ten-year SAR was 30% (95% CI, 21% to 40%) for GELF-86 patients and 53% (95% CI, 42% to 63%) for GELF-94 patients (P = .0172). EFSR was also significantly longer in the GELF-94 cohort (Table 2).
Outcome of the Patients in First Relapse According to the Cohort and Therapeutic Modalities
Effect of High-Dose Therapy With ASCT
The percentage of patients who went to ASCT was nearly identical in both studies. The conditioning regimen did not change over time with cyclophosphamide, etoposide, and TBI, or BEAM regimen. The outcome of patients treated with HDT and ASCT was statistically better than those of patients who did not receive transplant: 70% of 5-year SAR (95% CI, 59% to 79%) and 51% of 5-year EFSR (95% CI, 40% to 61%) versus 42% (95% CI, 33% to 50%) and 24% (95% CI, 17% to 31%), respectively (P < .0001 in both cases; Table 2, Fig 2). When only patients responding to salvage therapy were included in the analysis, the difference remains statistically significant (data not shown). In the group of patients treated with rituximab at time of relapse, only one patient in each group, who did and did not received transplant, did not respond to salvage treatment.
(A) Event-free survival after relapse and (B) overall survival after relapse for patients in first relapse or progression after salvage therapy followed or not by high-dose therapy (HDT) with autologous stem-cell transplantation. NA, not applicable.
Effect of Rituximab Therapy
Patients receiving rituximab as part of salvage therapy or as unique salvage treatment of first relapse or progression had a statistically better outcome compared with patients who did not receive rituximab: 5-year SAR of 80% (95% CI, 65% to 90%) versus 44% (95% CI, 37% to 51%; P < .001), respectively, and 5-year EFSR of 52% (95% CI, 37% to 65%) versus 29% (95% CI, 22% to 35%; P < .0001), respectively (Table 2, Fig 3).
(A) Event-free survival after relapse and (B) overall survival after relapse for patients in first relapse or progression after salvage therapy containing or not rituximab. NA, not applicable.
Effect of Rituximab and HDT
Patients who underwent HDT with ASCT after salvage chemotherapy including rituximab had the best SAR and EFSR of all the different subgroups (Table 2, Fig 4). Similarly, patients who underwent salvage chemotherapy with rituximab but without ASCT had a better outcome than patients undergoing chemotherapy with neither ASCT nor rituximab (Table 2, Fig 4).
Combined effect of rituximab and high-dose therapy (HDT) with autologous stem-cell transplantation (ASCT): (A) event-free survival after relapse (EFSR) and (B) survival after relapse (SAR) for patients treated with or without rituximab-containing salvage therapy plus HDT with ASCT; (C) EFSR and (D) SAR for patients treated with or without rituximab-containing salvage therapy without HDT. NA, not applicable.
DISCUSSION
This analysis reports the long-term outcome of two groups of patients with advanced follicular lymphoma receiving the same first-line therapy (CHVP plus interferon) in two successive randomized controlled studies, GELF-86 and GELF-94 studies. The eligibility criteria for inclusion and exclusion and therapy were similar for both cohorts. As expected, EFS did not differ between the two cohorts but there was a significant gain in OS for patients who had been treated more recently. This gain in survival may only be related to a better treatment of progressive or relapsed patients. In both studies, salvage therapy was not planned and the patients received various second-line therapies according to the local physician decision. HDT with ASCT was performed in a substantial proportion of patients responding to salvage therapy: 38% of relapsed patients received transplants in the GELF-86 study and 39% in the GELF-94 study. As expected in such a retrospective analysis, there is a bias because patients who were not responding to salvage therapy were not offered HDT but the same results were obtained if only patients responding to the salvage therapy were analyzed, thus eliminating this bias. Many series of patients with relapsed follicular lymphoma treated with HDT and different conditioning regimens reported encouraging results.11-15 Even though, there is only one published randomized controlled study testing this strategy, until recently most of the physicians were willing to recommend it as standard therapy for relapsing patients.16 The outcome of our patients treated with HDT without prior salvage with a rituximab-containing regimen is consistent with the outcome of patients treated in other series, particularly the recent update of Rohatiner et al (ie, 50% SAR at 10 years).15
Rituximab has been demonstrated to be a highly active agent in lymphoma and was associated with a short-term favorable outcome in relapsed follicular lymphoma.17-19 The only difference in the treatment of our relapsing patients was the introduction of rituximab after 1998. Indeed, 47% of patients in the GELF-94 cohort versus 2% in the GELF-86 cohort received rituximab alone or associated with chemotherapy. The introduction of this therapy translated into a dramatic improvement of outcome in terms of EFS and OS from first relapse or progression, even if there might be a bias in favor of rituximab because more patients with late relapse had the possibility to receive it. This improvement remained significant whether patients received HDT with ASCT or not. For these rituximab-naive patients, a gain of approximately 30% in 5-year survival after relapse was observed for patients treated with HDT or not if they received rituximab as part of the salvage regimen. Our data are consistent with those reported by three other studies, even if these studies looked at the effect of rituximab in first-line therapy.5,6,20 In the Southwest Oncology Group experience, there was an improvement of OS over time due to salvage therapies but failure-free survival was only improved when rituximab was associated with first-line chemotherapy.6 Similarly in the successive cohorts reported by the M.D. Anderson Cancer Center,5 before the introduction of rituximab, only a minimal gain was observed in terms of failure-free survival due to the association of interferon with chemotherapy as reported in a recent meta-analysis.21 The only cohort associated with longer EFS was for patients receiving chemotherapy with rituximab. The same observation was made in the Italian study where only rituximab in first line or in relapse lead to an improved outcome.20
More recently, rituximab has been widely tested with chemotherapy in first-line advanced follicular lymphoma and has demonstrated an advantage in terms of OS and EFS22-24: two recent randomized controlled trials have evaluated the impact of rituximab when associated with chemotherapy (CHOP or fludarabine-based regimens) in relapsed patients naive for rituximab.25,26 Both are consistent with a better complete response rate, and longer EFS and OS when rituximab was added to chemotherapy. A meta-analysis including seven randomized controlled studies testing rituximab in first line or in relapse concluded on the superiority of rituximab-containing chemotherapy compared with chemotherapy alone in terms of OS.27 Our results may not apply to patients treated with rituximab in first line but there is not any data supporting this. In fact, all reports (mostly retrospective analyses) showed a benefit of rituximab at time of progression for patients already treated with rituximab. The potential benefit of maintenance therapy with rituximab was also established for relapsed follicular lymphoma and is currently under investigation in first-line follicular lymphoma.25,27 The magnitude of the benefit observed in the rituximab arm in terms of EFSR and SAR is superior to the benefit observed with rituximab plus CHOP or rituximab plus FCM in previous studies.26,28 However, the rituximab plus CHOP study included patients in first and second relapse and the rituximab plus FCM study included patients with follicular and mantle cell lymphomas, both situations where such a benefit may be reduced compared to follicular lymphoma in first relapse.
In our analysis, the potential of rituximab for improving survival was greater than HDT and the benefit of the later may be questioned. As presented in Table 2 and Figure 4, the improvement obtained with the addition of rituximab to salvage chemotherapy was higher than the improvement obtained by HDT alone. However, patients treated with rituximab-containing salvage chemotherapy plus HDT have the longest SAR and OS of all these relapsing patients. Even, if we consider that these patients were selected because they had to respond to salvage therapy in order to be transplanted, SAR is particularly high, more than 90% at 5 years with only two events. Even if some events might occur with a longer follow-up and late deaths might be seen, such results have never been described in relapse patients with follicular lymphoma. A randomized study comparing the benefit of HDT with autotransplant to rituximab maintenance after salvage with a rituximab-containing regimen will certainly be the only way to confirm our data.
Major changes are occurring in the optimal management of follicular lymphoma and many questions have arisen regarding the optimal treatment of relapsed patients not naive to rituximab, the place of ASCT, and the optimal schedule of maintenance therapy. All these evolving modalities of therapy in first line and in relapse will probably lead to a major improvement of survival for patients with follicular lymphoma. The best strategies may only be described within a randomized prospective study but analyses like the one reported here help to define the hypotheses to be tested.
AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTSOF INTEREST
Although all authors completed the disclosure declaration, the following author(s) indicated a financial or other interest that is relevant to the subject matter under consideration in this article. Certain relationships marked with a “U” are those for which no compensation was received; those relationships marked with a “C” were compensated. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.
Employment or Leadership Position: None Consultant or Advisory Role: Bertrand Coiffier, Roche (C) Stock Ownership: None Honoraria: Pauline Brice, Roche; Nicole Brousse, Roche; Hervé Tilly, Roche; Philippe Solal-Céligny, Roche; Bertrand Coiffier, Roche, Genentech Research Funding: None Expert Testimony: None Other Remuneration: None
AUTHOR CONTRIBUTIONS
Conception and design: Catherine Sebban, Nicolas Mounier, Bertrand Coiffier
Financial support: Bertrand Coiffier
Administrative support: Bertrand Coiffier
Provision of study materials or patients: Catherine Sebban, Pauline Brice, Richard Delarue, Corinne Haioun, Bertrand Souleau, Nicole Brousse, Pierre Feugier, Hervé Tilly, Philippe Solal-Celigny, Bertrand Coiffier
Collection and assembly of data: Catherine Sebban, Pauline Brice, Bertrand Coiffier
Data analysis and interpretation: Catherine Sebban, Corinne Haioun, Nicolas Mounier, Bertrand Coiffier
Manuscript writing: Catherine Sebban, Bertrand Coiffier
Final approval of manuscript: Catherine Sebban, Pauline Brice, Richard Delarue, Corinne Haioun, Bertrand Souleau, Nicolas Mounier, Nicole Brousse, Pierre Feugier, Hervé Tilly, Philippe Solal-Celigny, Bertrand Coiffier
Appendix
Characteristics of the Patients Included in the Study
Treatment at First Progression or Relapse
Acknowledgments
We thank Laurence Girard for data management and Marion Fournier for statistical analyses—both from the GELA recherche clinique.
Footnotes
-
published online ahead of print at www.jco.org on June 16, 2008.
-
Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
- Received November 29, 2007.
- Accepted April 8, 2008.
