Effect of Single-Agent Rituximab Given at the Standard Schedule or As Prolonged Treatment in Patients With Mantle Cell Lymphoma: A Study of
http://www.100md.com
《临床肿瘤学》
the Oncology Institute of Southern Switzerland, Bellinzona
Swiss Institute of Applied Cancer Research Coordinating Centre, Bern
Institute of Medical Oncology, Inselspital, University of Bern
Swiss Reference Centre for Lymphoma Pathology
Kantonsspital, Aarau
Kantonsspital St Gallen, St Gallen
Kantonsspital Luzern, Luzern
Universittsspital Zürich, CHUV, Lausanne, Switzerland
ABSTRACT
PATIENTS AND METHODS: After induction treatment with the standard schedule (375 mg/m2 weekly x 4), patients who were responding or who had stable disease at week 12 from the start of treatment were randomly assigned to no further treatment (arm A) or prolonged rituximab administration (375 mg/m2) every 8 weeks for four times (arm B).
RESULTS: The trial enrolled 104 patients. After induction, clinical response was 27% with 2% complete responses. Among patients with detectable t(11;14)-positive cells in blood and bone marrow at baseline, four of 20, and one of 14, respectively, became polymerase chain-reaction–negative after induction. Anemia was the only adverse predictor of response in the multivariate analysis. After a median follow-up of 29 months, response rate and duration of response were not significantly different between the two schedules in 61 randomly assigned patients. Median event-free survival (EFS) was 6 months in arm A versus 12 months in arm B; the difference was not significant (P = .1). Prolonged treatment seemed to improve EFS in the subgroup of pretreated patients (5 months in arm A v 11 months in arm B; P = .04). Thirteen percent of patients in arm A and 9% in arm B presented with grade 3 to 4 hematologic toxicity.
CONCLUSION: Single-agent rituximab is active in MCL, but the addition of four single doses at 8-week intervals does not seem to significantly improve response rate, duration of response, or EFS after treatment with the standard schedule.
INTRODUCTION
Rituximab, a human/mouse chimeric monoclonal antibody with specificity for the CD20 B-cell–surface antigen, is active as a single-agent treatment for non-Hodgkin's lymphoma. Response rates (RRs) of 45% to 70% have been observed in several types of indolent lymphoma,3,4 while in more aggressive non-Hodgkin's lymphoma, RRs are approximately 30%.5 When added to chemotherapy, rituximab has been shown to significantly improve survival compared with chemotherapy alone.6
The majority of clinical experience with rituximab in MCL has been collected in patients receiving the standard schedule (375 mg/m2 weekly x 4). In a study of 34 previously untreated and 40 previously treated patients, the RRs were 38% and 37%, respectively, with 16% and 14% complete response (CR). The time to progression was 7 months, and the response duration was 1 year.7 Other smaller studies reported similar results.5,8
Because pharmacokinetics data suggested that persistence of rituximab in patients' blood is associated with a higher RR,9 we speculated that a longer exposure to rituximab might augment its biologic effect. Two parallel trials (one in follicular lymphoma and one in MCL) were designed to test this hypothesis. We compared the standard rituximab schedule with a prolonged rituximab schedule in which the standard treatment was followed by a single 375-mg/m2 infusion every 2 months at four intervals, in order to maintain a biologically active serum concentration for approximately 1 year. The 2-month interval was chosen based on preliminary pharmacokinetics data.9 This article reports the results of the trial in MCL.
PATIENTS AND METHODS
Patients were initially treated with rituximab 375 mg/m2 per week for 4 weeks ("induction" phase). Patients with stable disease (SD), or in partial response (PR) or CR at week 12 (from treatment start) were randomly assigned in a 1:1 ratio to no further treatment (arm A, "standard treatment") or to treatment with a single infusion of rituximab 375 mg/m2 at week 12, and again at months 5, 7, and 9 (arm B, "prolonged treatment"). The randomization was stratified, using the minimization method,10 according to status of disease at trial entry (first presentation v refractory or relapsed), response to induction treatment (SD v response), and center. Patients were centrally randomized by fax via the Swiss Group for Clinical Cancer Research (SAKK) Trials Office in Bern, Switzerland. On disease progression or relapse, further treatment was at the treating physician's discretion.
Event-free survival (EFS) time was the primary end point and was calculated as the time from first induction infusion to progression, relapse, second tumor, or death from any cause. Remission duration was defined in the same way, but was restricted to responding patients. For the randomized phase of the trial, a group sequential design with two interim analyses and one final analysis using EFS as the primary end point was adopted.
Patients
Inclusion criteria were biopsy-proven MCL, age ≥ 18 years, and measurable disease, defined as the presence of at least one previously unirradiated lesion with two measurable perpendicular diameters, of which at least one should be 2 cm. The interval between last systemic anticancer treatment and trial entry should not be less than 28 days. Other inclusion criteria were an Eastern Cooperative Oncology Group performance status ≤ 2 and, after observing unexpected cardiac events in the first 10 months of trial,11 a cardiac ejection fraction ≥ 50% by echocardiography. Exclusion criteria included symptomatic CNS disease, a history of significant medical conditions including previous malignancies within 5 years, a reduced renal function (creatinine > 2x the upper limit of normal [ULN]) or liver function (bilirubin > 2x ULN). We also excluded pregnant or lactating females; patients with active opportunistic infections; or patients with known HIV, or hepatitis B or C infections. Previous treatment with rituximab was not allowed.
Trial Assessments
For all patients, a complete staging, including clinical examination, chest and abdomen computed tomography scan, bone marrow (BM) aspirate, and trephine biopsy was required not more than 2 weeks before trial entry. Any other clinically indicated tumor evaluations were at the discretion of the investigator. A central histology review was performed between registration and randomization to confirm the diagnosis. A formal evaluation of all the involved sites was performed at 12 weeks; at 7, 12, 18, and 24 months; then yearly or when clinically required. Patients enrolled during the first 7 months had an additional staging procedure at week 8. The first interim analysis showed similar RRs at weeks 8 and 12; hence, it was decided to limit the first restaging at week 12 for subsequent patients. Re-evaluation of the BM was required only at week 12 and month 12 if involved at trial entry. Routine blood counts and chemistries were assessed at baseline, before each rituximab administration, and at months 2, 3, 5, 7, 9, and 12. Serum immunoglobulins (IgG, IgA, and IgM) were measured at baseline and again at months 3, 7, and 12, while blood samples for immunophenotyping were taken at baseline, week 12, and at months 9, 12, 18, and 24. Analysis of lymphocyte subsets was performed as described previously.12 Polymerase chain reaction amplification of the t(11;14) (bcl-1 rearrangement) chromosomal translocation was used to detect and monitor minimal residual disease. The presence of cells carrying the t(11;14) (q13;q32) chromosomal translocation in peripheral blood and BM was assessed using a seminested polymerase chain reaction assay.13
Statistical Methods
The randomized phase of the trial utilized a group sequential design, with two interim analyses and one final analysis using EFS as the primary end point. The overall type I error probability was 5%, and the statistical power was 80% for the two-sided log-rank test to detect an increase of EFS from 6 months in the standard treatment arm to 13 months in the prolonged treatment arm. A total of 54 events in about 65 patients were required for the final analysis.
The comparisons between treatment arms or groups defined by other factors (ie, previously treated v chemotherapy-nave, or responders v nonresponders) were carried out by Wilcoxon rank sum test for continuous variables and by {chi}2 or Fisher's exact test for categorical variables. Odds ratios (ORs) were also calculated for associations between a binary response variable and a binary factor. Remission duration and EFS time were estimated by the Kaplan-Meier method and were compared between groups by log-rank test. A Renyi-type test using log-rank weights was also performed to compare EFS, because the assumption of proportional hazards was violated; hence, the usual log-rank test was inadequate.14 Hazard ratios (HRs) were calculated as the ratio of hazard of experiencing an event in the nonreference group with respect to the reference group.
The simultaneous impact of some potential predictive variables on clinical response and EFS was analyzed by multiple logistic regression and multiple Cox regression, respectively, employing a stepwise model selection procedure. All tests were two-sided. No adjustment for multiple comparisons was performed. CIs for RR, OR, HR, and median EFS time are provided where appropriate. For randomly assigned patients, all between-arm comparisons were carried out according to the intention-to-treat principle.
RESULTS
Of the 88 eligible and assessable patients, 27 were not randomly assigned to the second phase of the study: 24 because of disease progression and three because of major toxicity during the induction phase. The clinical and pathological characteristics of the 61 randomly assigned patients (27 in arm A and 34 in arm B) were similar in the two arms (Table 1) and had, as expected, more favorable characteristics at randomization than at baseline—the levels of hemoglobin, platelets, and lactate dehydrogenase were nearer to normal, and the incidence of bulky disease and BM infiltration was lower.
The proportion of patients having responded to previous treatments was similar in the two arms (44% v 47%, respectively), while the response to induction treatment (37% v 47%, respectively) tended to be higher in arm B.
Description of Treatment
Induction phase. Ten patients received fewer than four infusions. One patient died before the treatment could start, two did not receive the fourth infusion due to excessive infusion-related toxicity of the three previous treatments, and seven did not complete treatment due to progression. The intervals between infusions in the induction phase were maintained in the planned range, except in six patients in whom infusions were delayed for 5 to 10 days, owing to toxicity in five and logistic reasons in one. The full planned dose of the drug was given in all patients but one, who had an interruption of the first infusion due to toxicity.
Prolonged treatment phase. The majority of patients in the prolonged treatment arm received the treatment according to the protocol. Fifteen patients had an interruption of the treatment before the fourth administration: two because of toxicity, 11 because of disease progression or relapse, and two because of refusal. The timing of the infusions followed the protocol for the majority of patients, but with a rather wide range: the planned week-20 infusion was administered at median week 20 (range, 15 to 26 weeks), week-28 infusion at median week 28 (range, 25 to 36 weeks), week-36 infusion at median week 36 (range, 30 to 47 weeks), and week-52 infusion at week 51 (range, 47 to 63 weeks).
Objective Response
Induction phase. At week 12, the RR for the 88 assessable patients was 27% (95% CI, 18% to 38%; 2.3% CR). There was no difference between the RR of the 34 chemotherapy-naive patients (RR, 27%; 95% CI, 13% to 44%; 3% CR) and the 54 previously treated patients (RR, 28%; 95% CI, 17% to 42%; 2% CR). When including ineligible and nonassessable patients in an intent-to-treat analysis, the RRs were 26% and 27%, respectively. Favorable predictive factors identified by univariate analyses in eligible and assessable patients included a normal hemoglobin level (P = .007), absence of "B" symptoms (P = .04), and absence of bulky disease (P = .05). In the multivariate analysis employing a stepwise model selection procedure, only hemoglobin level remained significant (OR, 5.3; 95% CI, 1.6 to 17.4; P = .006). The results were similar if ineligible and nonassessable patients were included. Of the 71 patients with BM involvement at baseline, 47 had BM re-evaluation during the induction phase, with 32% having become morphologically negative (37% if only eligible patients are considered). It is likely that the lack of BM re-evaluations for 24 patients had resulted in an underestimation of the CR rate, since patients assessed as achieving CR on the basis of clinical and radiologic evaluation, but who did not have a re-evaluation of a previously positive BM, were considered as achieving a PR. Among the 20 peripheral blood and 14 BM samples with detectable t(11;14)-positive cells at baseline in eligible patients, four peripheral blood samples, and one BM sample had become negative after induction.
Prolonged treatment/observation phase. The response assessments at 3, 7, 12, 18, and 24 months, and the overall best response for randomly assigned patients are summarized in Figure 1. In arm A, the proportion of patients with response decreased steadily from 33% (95% CI, 17% to 54%) at week 12, to 26% (95% CI, 11% to 46%) at month 12 and 15% (95% CI, 4% to 34%) at 24 months. In arm B, the proportion of patients remained initially stable, with 39% (95% CI, 23% to 58%) at week 12 and 36% (95% CI, 20% to 55%) at month 12, declining then to 9% (95% CI, 2% to 24%) at 24 months from the start of treatment. The proportion of patients in CR remained low in both arms: overall, 11% of patients in arm A and 12% in arm B experienced a CR at some time during the study. The number of patients improving their response in the first 2 years (from PR to CR, from SD to CR, or from SD to PR) was low in both arms—three cases in arm A and six cases in arm B. The difference in RR between arms was not statistically significant at any time point. Among patients with BM involvement at baseline and with BM re-evaluation after random assignment, six of seven in arm A and seven of 11 in arm B became negative. Among responders at week 12 (n = 22), the proportion of CR and PR at month 12 was surprisingly higher in arm A compared with arm B (67% v 54%, respectively), but the difference was not significant (P = .67).
In the selected population of randomized patients (ie, eligible, assessable, and not progressing under induction treatment with rituximab), the overall best response was 41% (95% CI, 22% to 61%) for arm A and 55% (95% CI, 36% to 72%) for arm B (P = .31). The overall best response for chemotherapy-naive patients (n = 27) was 39% (95% CI, 14% to 68%) for arm A and 50% (95% CI, 23% to 77%) for arm B (P = .70). The median time to response was 9 weeks in both arms, and the median remission duration among week-12 responders was 15 months in both arms.
EFS
The median follow-up time in 41 living patients was 29 months, with a total of 57 events observed. The median EFS time among the 61 randomly assigned patients was 6 months (95% CI, 4 to 14 months) in arm A and 12 months (95% CI, 8 to 17 months) in arm B. The difference was noticeable until 17 months, but it became undistinguishable afterwards (P = .45 by the initially planned two-sided log-rank test; Fig 2). Using the more appropriate Renyi-type test with log-rank weights, the difference between the two arms was not significant (P = .1). When EFS was analyzed according to response to the induction phase, responders had the same EFS in both arms (15 months; Fig 3), and patients with SD had a shorter EFS (6 months in arm A and 9 months in arm B). The multivariate analysis confirmed that treatment arm was not predictive for EFS, while thrombocytopenia grade at study entry (grade ≥ 1 v 0; HR, 2.4; P = .04) and number of previous chemotherapy regimens (≥ 2 v ≤ 1 HR, 2.0; P = .03) were independently predictive.
Toxicity
During the induction phase, the majority of toxicities were mild infusion-related symptoms during the first infusion. Additionally, 13% and 11% of patients had grade 3 and 4 toxicity, respectively, including two episodes (one patient) of pneumonia, three episodes (two patients) of pain, three cases (two patients) of dyspnea, three reports (one patient) of thrombocytopenia, one major cardiac event, one allergic reaction, and one case of nausea. Cardiac ejection fraction was reassessed in 40 patients at week 12 and remained stable (from 63% [range, 50% to 78%] at baseline to 64% [range, 44% to 79%]). Seventeen cases of serious adverse events were documented during induction treatment. Two patients died of causes other than progression—one of a probable myocardial infarction and one of Pneumocystis carinii pneumonia.
After random assignment, nonhematological grade 3 or 4 toxicity included two cases of pain, one diarrhea episode, and one case of cholecystitis in arm A; and one patient experienced three episodes of pneumonia, one case of hepatitis, and one renal failure case in arm B. Grade 3 or 4 hematological toxicity was observed in 13% of patients in arm A and 9% in arm B. Among 19 patients with cardiac function reassessed after random assignment (seven in arm A and 12 in arm B), the median ejection fraction was 62% (range, 60% to 66%) in arm A and 59% (range, 40% to 75%) in arm B. The number of serious adverse events after random assignment (including follow-up period) was six in arm A (including two second tumors) and eight in arm B (including one second tumor, one septic shock, and one cardiac death).
The lymphocytes subset analysis (performed excluding patients with baseline WBC > 10 x 109/L who were suspected of having leukemic disease) showed a very early and important reduction of circulating B cells, while T-helper, T-suppressor, and natural killer cells remained approximately stable during the induction phase. The evolution of B lymphocytes for randomly assigned patients throughout the first 12 months of follow-up is shown in Figure 4. Median B-cell levels returned to levels similar to baseline values after 9 months in arm A, while in arm B, the B-cell recovery had still not taken place after 12 months. The difference at 9 months was significant (median, 90.5 x 106/L v 13.9 x 106/L; P = .03).
DISCUSSION
In phase II trials, a minimum RR of 30% is required for a drug to be considered a potentially active new anticancer agent. Although rituximab reaches this threshold, the relatively short duration of response (compared with what was seen in other types of lymphomas), and the fact that this RR is surprisingly not higher in chemotherapy-nave patients relegates the antibody in the group of modestly active drugs for this disease. It is therefore not astonishing that the effect of a prolonged administration is similarly modest.
Compared with the only other sufficiently sized study of single-agent rituximab in MCL,7 our trial has the following advantages: larger sample size, histologic diagnosis reviewed centrally according to strict criteria, and evaluation of the evolution of lymphocyte subsets and molecular remission (at least in a subset of patients). Unfortunately, although this latter analysis was performed in the majority of patients at trial entry, it was repeated in only a minority at restaging, reducing the reliability of the molecular response results. Compared with the other series, we observed a slightly lower RR (possibly due to our more stringent response criteria), a similar response duration, and concordant prognostic factors.7
The parallel trial, which was conducted with exactly the same design by our group in patients with follicular lymphoma, showed that (in rituximab responders only) the prolonged schedule significantly doubles the median EFS (from 12 to 23 months, P = .02), and we concluded that single-agent rituximab, particularly if given with a prolonged schedule, is a valid treatment option for follicular lymphoma.12 The same conclusion cannot be made for MCL. In this disease, single-agent rituximab is less active at either schedule; therefore, the drug should instead be incorporated into combination schemes with chemotherapy. Several phase II trials and preliminary data from randomized studies confirm this assumption.15,16 Because rituximab can induce molecular remissions, its use in combination with chemotherapy as an in vivo purging agent for peripheral stem collection is theoretically sound; this application in MCL has already shown promising results.17 The use of rituximab maintenance therapy following autologous stem-cell transplantation also seems to improve outcomes in patients with advanced MCL.18
It is not surprising that MCL responds less well to single-agent rituximab than indolent lymphomas (the same is true for chemotherapy), but it is surprising that the agent shows a similar effect in chemotherapy-nave and pretreated patients: in oncology practice, RR and duration are usually higher in newly diagnosed patients. Even though the same observation was made by Foran et al,7 in our trial, the number of previous regimens in pretreated patients is a prognostic factor, raising the suspicion that the amount of previous treatment is indeed important. The similar outcomes in chemotherapy-nave and pretreated patients observed in our trial could be explained by the small group sizes (38 and 66 patients, respectively) and/or an unfavorable selection of chemotherapy-nave patients proposed for rituximab monotherapy as a first-line treatment.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Patricia Katz and Corinne Friedly for central data management, Dina Delle Pezze for the fluorescence-activated cell sorting analysis of lymphocyte subsets, and Michela Gisi for assistance in molecular biology analysis.
NOTES
Supported in part by research funding from Roche Pharma Schweiz AG to the Swiss Group for Clinical Cancer Research (SAKK).
Presented orally (in summary) at the 2003 Annual Meeting of the European Haematology Association (EHA), Lyon, France, June 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Zucca E, Roggero E, Pinotti G, et al: Patterns of survival in mantle cell lymphoma. Ann Oncol 6:257-262, 1995
2. Weisenburger DD, Armitage JO: Mantle cell lymphoma: An entity comes of age. Blood 87:4483-4494, 1996
3. McLaughlin P, Grillo-López AJ, Link BK, et al: Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: Half of patients respond to a four-dose treatment program. J Clin Oncol 16:2825-2833, 1998
4. Conconi A, Martinelli G, Thieblemont C, et al: Clinical activity of rituximab in extranodal marginal B-cell lymphoma of MALT type. Blood 102:2741-2745, 2003
5. Coiffier B, Haioun C, Ketterer N, et al: Rituximab (anti-CD20 monoclonal antibody) for the treatment of patients with relapsing or refractory aggressive lymphoma: A multicenter phase II study. Blood 92:1927-1932, 1998
6. Coiffier B, Lepage E, Brière J, et al: CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 346:235-242, 2002
7. Foran JM, Cunningham D, Coiffier B, et al: Treatment of mantle-cell lymphoma with rituximab (chimeric monoclonal anti-CD20 antibody): analysis of factors associated with response. Ann Oncol 11:117-121, 2000
8. Nguyen DT, Amess JA, Doughty H, et al: IDEC-C2B8 anti-CD20 (rituximab) immunotherapy in patients with low-grade non-Hodgkin's lymphoma and lymphoproliferative disorders: Evaluation of response on 48 patients. Eur J Haematol 62:76-82, 1999
9. Berinstein NL, Grillo-López AJ, White CA, et al: Association of serum rituximab (IDEC-C2B8) concentration and anti-tumor response in the treatment of recurrent low-grade or follicular non-Hodgkin's lymphoma. Ann Oncol 9:995-1001, 1998
10. Pocock SJ: Clinical Trials: A Practical Approach. Chichester, UK, John Wiley and Sons, 1983, pp 84-86
11. Ghielmini M, Hsu Schmitz SF, Bürki K, et al: The effect of rituximab on patients with follicular and mantle-cell lymphoma: Swiss Group for Clinical Research (SAKK). Ann Oncol 11:123-126, 2000
12. Ghielmini M, Hsu Schmitz SF, Cogliatti S, et al: Prolonged treatment with rituximab in patients with follicular lymphoma significantly increases event-free survival and response duration compared with the standard weekly x 4 schedule. Blood: Prepublished online February 24, 2004:DOI 10.1182/blood-2003-10-3411
13. Bertoni F, Roggero E, Luscieti P, et al: Clonality assessment in blood of patients with mantle cell lymphoma. Leuk Lymphoma 32:372-379, 1999
14. Klein JP, Moeschberger ML. Survival Analysis: Techniques for Censored and Truncated Data. New York, NY, Springer-Verlag Inc, 2003, section 7.6
15. Dreyling M H., Forstpointer R. Repp R., et al: Combined immuno-chemotherapy (R-FCM) results in superior remission and survival rates in recurrent follicular and mantle cell lymphoma - final results of a prospective randomized trial of the German Low Grade Lymphoma Study Group (GLSG). Blood 102: 2003 (abstr 351; suppl)
16. Hiddemann W, Dreyling MH, Forstpointner R, et al: Combined immuno-chemotherapy (R-CHOP) significantly improves time to treatment failure in first line therapy of follicular lymphoma: Results of a prospective randomized trial of the German Low Grade Lymphoma Study Group (GLSG). Blood 102: 2003 (abstr 352; suppl)
17. Romaguera J, Cabanillas F, Dang NH, et al: Mantle cell lymphoma (MCL): High rates of complete remission (CR) and prolonged failure-free survival (FFS) with Rituxan-Hyper CVAD (R-HCVAD) without stem cell transplant (SCT). Blood 98: 2001 (abstr 3030; suppl)
18. Mangel J, Leitch HA, Connors JM, et al: Intensive chemotherapy and autologous stem cell transplantation plus rituximab is superior to conventional chemotherapy for newly diagnosed advanced stage mantle-cell lymphoma: A matched pair analysis. Ann Oncol 15:283-289, 2004(Michele Ghielmini, Shu-Fa)
Swiss Institute of Applied Cancer Research Coordinating Centre, Bern
Institute of Medical Oncology, Inselspital, University of Bern
Swiss Reference Centre for Lymphoma Pathology
Kantonsspital, Aarau
Kantonsspital St Gallen, St Gallen
Kantonsspital Luzern, Luzern
Universittsspital Zürich, CHUV, Lausanne, Switzerland
ABSTRACT
PATIENTS AND METHODS: After induction treatment with the standard schedule (375 mg/m2 weekly x 4), patients who were responding or who had stable disease at week 12 from the start of treatment were randomly assigned to no further treatment (arm A) or prolonged rituximab administration (375 mg/m2) every 8 weeks for four times (arm B).
RESULTS: The trial enrolled 104 patients. After induction, clinical response was 27% with 2% complete responses. Among patients with detectable t(11;14)-positive cells in blood and bone marrow at baseline, four of 20, and one of 14, respectively, became polymerase chain-reaction–negative after induction. Anemia was the only adverse predictor of response in the multivariate analysis. After a median follow-up of 29 months, response rate and duration of response were not significantly different between the two schedules in 61 randomly assigned patients. Median event-free survival (EFS) was 6 months in arm A versus 12 months in arm B; the difference was not significant (P = .1). Prolonged treatment seemed to improve EFS in the subgroup of pretreated patients (5 months in arm A v 11 months in arm B; P = .04). Thirteen percent of patients in arm A and 9% in arm B presented with grade 3 to 4 hematologic toxicity.
CONCLUSION: Single-agent rituximab is active in MCL, but the addition of four single doses at 8-week intervals does not seem to significantly improve response rate, duration of response, or EFS after treatment with the standard schedule.
INTRODUCTION
Rituximab, a human/mouse chimeric monoclonal antibody with specificity for the CD20 B-cell–surface antigen, is active as a single-agent treatment for non-Hodgkin's lymphoma. Response rates (RRs) of 45% to 70% have been observed in several types of indolent lymphoma,3,4 while in more aggressive non-Hodgkin's lymphoma, RRs are approximately 30%.5 When added to chemotherapy, rituximab has been shown to significantly improve survival compared with chemotherapy alone.6
The majority of clinical experience with rituximab in MCL has been collected in patients receiving the standard schedule (375 mg/m2 weekly x 4). In a study of 34 previously untreated and 40 previously treated patients, the RRs were 38% and 37%, respectively, with 16% and 14% complete response (CR). The time to progression was 7 months, and the response duration was 1 year.7 Other smaller studies reported similar results.5,8
Because pharmacokinetics data suggested that persistence of rituximab in patients' blood is associated with a higher RR,9 we speculated that a longer exposure to rituximab might augment its biologic effect. Two parallel trials (one in follicular lymphoma and one in MCL) were designed to test this hypothesis. We compared the standard rituximab schedule with a prolonged rituximab schedule in which the standard treatment was followed by a single 375-mg/m2 infusion every 2 months at four intervals, in order to maintain a biologically active serum concentration for approximately 1 year. The 2-month interval was chosen based on preliminary pharmacokinetics data.9 This article reports the results of the trial in MCL.
PATIENTS AND METHODS
Patients were initially treated with rituximab 375 mg/m2 per week for 4 weeks ("induction" phase). Patients with stable disease (SD), or in partial response (PR) or CR at week 12 (from treatment start) were randomly assigned in a 1:1 ratio to no further treatment (arm A, "standard treatment") or to treatment with a single infusion of rituximab 375 mg/m2 at week 12, and again at months 5, 7, and 9 (arm B, "prolonged treatment"). The randomization was stratified, using the minimization method,10 according to status of disease at trial entry (first presentation v refractory or relapsed), response to induction treatment (SD v response), and center. Patients were centrally randomized by fax via the Swiss Group for Clinical Cancer Research (SAKK) Trials Office in Bern, Switzerland. On disease progression or relapse, further treatment was at the treating physician's discretion.
Event-free survival (EFS) time was the primary end point and was calculated as the time from first induction infusion to progression, relapse, second tumor, or death from any cause. Remission duration was defined in the same way, but was restricted to responding patients. For the randomized phase of the trial, a group sequential design with two interim analyses and one final analysis using EFS as the primary end point was adopted.
Patients
Inclusion criteria were biopsy-proven MCL, age ≥ 18 years, and measurable disease, defined as the presence of at least one previously unirradiated lesion with two measurable perpendicular diameters, of which at least one should be 2 cm. The interval between last systemic anticancer treatment and trial entry should not be less than 28 days. Other inclusion criteria were an Eastern Cooperative Oncology Group performance status ≤ 2 and, after observing unexpected cardiac events in the first 10 months of trial,11 a cardiac ejection fraction ≥ 50% by echocardiography. Exclusion criteria included symptomatic CNS disease, a history of significant medical conditions including previous malignancies within 5 years, a reduced renal function (creatinine > 2x the upper limit of normal [ULN]) or liver function (bilirubin > 2x ULN). We also excluded pregnant or lactating females; patients with active opportunistic infections; or patients with known HIV, or hepatitis B or C infections. Previous treatment with rituximab was not allowed.
Trial Assessments
For all patients, a complete staging, including clinical examination, chest and abdomen computed tomography scan, bone marrow (BM) aspirate, and trephine biopsy was required not more than 2 weeks before trial entry. Any other clinically indicated tumor evaluations were at the discretion of the investigator. A central histology review was performed between registration and randomization to confirm the diagnosis. A formal evaluation of all the involved sites was performed at 12 weeks; at 7, 12, 18, and 24 months; then yearly or when clinically required. Patients enrolled during the first 7 months had an additional staging procedure at week 8. The first interim analysis showed similar RRs at weeks 8 and 12; hence, it was decided to limit the first restaging at week 12 for subsequent patients. Re-evaluation of the BM was required only at week 12 and month 12 if involved at trial entry. Routine blood counts and chemistries were assessed at baseline, before each rituximab administration, and at months 2, 3, 5, 7, 9, and 12. Serum immunoglobulins (IgG, IgA, and IgM) were measured at baseline and again at months 3, 7, and 12, while blood samples for immunophenotyping were taken at baseline, week 12, and at months 9, 12, 18, and 24. Analysis of lymphocyte subsets was performed as described previously.12 Polymerase chain reaction amplification of the t(11;14) (bcl-1 rearrangement) chromosomal translocation was used to detect and monitor minimal residual disease. The presence of cells carrying the t(11;14) (q13;q32) chromosomal translocation in peripheral blood and BM was assessed using a seminested polymerase chain reaction assay.13
Statistical Methods
The randomized phase of the trial utilized a group sequential design, with two interim analyses and one final analysis using EFS as the primary end point. The overall type I error probability was 5%, and the statistical power was 80% for the two-sided log-rank test to detect an increase of EFS from 6 months in the standard treatment arm to 13 months in the prolonged treatment arm. A total of 54 events in about 65 patients were required for the final analysis.
The comparisons between treatment arms or groups defined by other factors (ie, previously treated v chemotherapy-nave, or responders v nonresponders) were carried out by Wilcoxon rank sum test for continuous variables and by {chi}2 or Fisher's exact test for categorical variables. Odds ratios (ORs) were also calculated for associations between a binary response variable and a binary factor. Remission duration and EFS time were estimated by the Kaplan-Meier method and were compared between groups by log-rank test. A Renyi-type test using log-rank weights was also performed to compare EFS, because the assumption of proportional hazards was violated; hence, the usual log-rank test was inadequate.14 Hazard ratios (HRs) were calculated as the ratio of hazard of experiencing an event in the nonreference group with respect to the reference group.
The simultaneous impact of some potential predictive variables on clinical response and EFS was analyzed by multiple logistic regression and multiple Cox regression, respectively, employing a stepwise model selection procedure. All tests were two-sided. No adjustment for multiple comparisons was performed. CIs for RR, OR, HR, and median EFS time are provided where appropriate. For randomly assigned patients, all between-arm comparisons were carried out according to the intention-to-treat principle.
RESULTS
Of the 88 eligible and assessable patients, 27 were not randomly assigned to the second phase of the study: 24 because of disease progression and three because of major toxicity during the induction phase. The clinical and pathological characteristics of the 61 randomly assigned patients (27 in arm A and 34 in arm B) were similar in the two arms (Table 1) and had, as expected, more favorable characteristics at randomization than at baseline—the levels of hemoglobin, platelets, and lactate dehydrogenase were nearer to normal, and the incidence of bulky disease and BM infiltration was lower.
The proportion of patients having responded to previous treatments was similar in the two arms (44% v 47%, respectively), while the response to induction treatment (37% v 47%, respectively) tended to be higher in arm B.
Description of Treatment
Induction phase. Ten patients received fewer than four infusions. One patient died before the treatment could start, two did not receive the fourth infusion due to excessive infusion-related toxicity of the three previous treatments, and seven did not complete treatment due to progression. The intervals between infusions in the induction phase were maintained in the planned range, except in six patients in whom infusions were delayed for 5 to 10 days, owing to toxicity in five and logistic reasons in one. The full planned dose of the drug was given in all patients but one, who had an interruption of the first infusion due to toxicity.
Prolonged treatment phase. The majority of patients in the prolonged treatment arm received the treatment according to the protocol. Fifteen patients had an interruption of the treatment before the fourth administration: two because of toxicity, 11 because of disease progression or relapse, and two because of refusal. The timing of the infusions followed the protocol for the majority of patients, but with a rather wide range: the planned week-20 infusion was administered at median week 20 (range, 15 to 26 weeks), week-28 infusion at median week 28 (range, 25 to 36 weeks), week-36 infusion at median week 36 (range, 30 to 47 weeks), and week-52 infusion at week 51 (range, 47 to 63 weeks).
Objective Response
Induction phase. At week 12, the RR for the 88 assessable patients was 27% (95% CI, 18% to 38%; 2.3% CR). There was no difference between the RR of the 34 chemotherapy-naive patients (RR, 27%; 95% CI, 13% to 44%; 3% CR) and the 54 previously treated patients (RR, 28%; 95% CI, 17% to 42%; 2% CR). When including ineligible and nonassessable patients in an intent-to-treat analysis, the RRs were 26% and 27%, respectively. Favorable predictive factors identified by univariate analyses in eligible and assessable patients included a normal hemoglobin level (P = .007), absence of "B" symptoms (P = .04), and absence of bulky disease (P = .05). In the multivariate analysis employing a stepwise model selection procedure, only hemoglobin level remained significant (OR, 5.3; 95% CI, 1.6 to 17.4; P = .006). The results were similar if ineligible and nonassessable patients were included. Of the 71 patients with BM involvement at baseline, 47 had BM re-evaluation during the induction phase, with 32% having become morphologically negative (37% if only eligible patients are considered). It is likely that the lack of BM re-evaluations for 24 patients had resulted in an underestimation of the CR rate, since patients assessed as achieving CR on the basis of clinical and radiologic evaluation, but who did not have a re-evaluation of a previously positive BM, were considered as achieving a PR. Among the 20 peripheral blood and 14 BM samples with detectable t(11;14)-positive cells at baseline in eligible patients, four peripheral blood samples, and one BM sample had become negative after induction.
Prolonged treatment/observation phase. The response assessments at 3, 7, 12, 18, and 24 months, and the overall best response for randomly assigned patients are summarized in Figure 1. In arm A, the proportion of patients with response decreased steadily from 33% (95% CI, 17% to 54%) at week 12, to 26% (95% CI, 11% to 46%) at month 12 and 15% (95% CI, 4% to 34%) at 24 months. In arm B, the proportion of patients remained initially stable, with 39% (95% CI, 23% to 58%) at week 12 and 36% (95% CI, 20% to 55%) at month 12, declining then to 9% (95% CI, 2% to 24%) at 24 months from the start of treatment. The proportion of patients in CR remained low in both arms: overall, 11% of patients in arm A and 12% in arm B experienced a CR at some time during the study. The number of patients improving their response in the first 2 years (from PR to CR, from SD to CR, or from SD to PR) was low in both arms—three cases in arm A and six cases in arm B. The difference in RR between arms was not statistically significant at any time point. Among patients with BM involvement at baseline and with BM re-evaluation after random assignment, six of seven in arm A and seven of 11 in arm B became negative. Among responders at week 12 (n = 22), the proportion of CR and PR at month 12 was surprisingly higher in arm A compared with arm B (67% v 54%, respectively), but the difference was not significant (P = .67).
In the selected population of randomized patients (ie, eligible, assessable, and not progressing under induction treatment with rituximab), the overall best response was 41% (95% CI, 22% to 61%) for arm A and 55% (95% CI, 36% to 72%) for arm B (P = .31). The overall best response for chemotherapy-naive patients (n = 27) was 39% (95% CI, 14% to 68%) for arm A and 50% (95% CI, 23% to 77%) for arm B (P = .70). The median time to response was 9 weeks in both arms, and the median remission duration among week-12 responders was 15 months in both arms.
EFS
The median follow-up time in 41 living patients was 29 months, with a total of 57 events observed. The median EFS time among the 61 randomly assigned patients was 6 months (95% CI, 4 to 14 months) in arm A and 12 months (95% CI, 8 to 17 months) in arm B. The difference was noticeable until 17 months, but it became undistinguishable afterwards (P = .45 by the initially planned two-sided log-rank test; Fig 2). Using the more appropriate Renyi-type test with log-rank weights, the difference between the two arms was not significant (P = .1). When EFS was analyzed according to response to the induction phase, responders had the same EFS in both arms (15 months; Fig 3), and patients with SD had a shorter EFS (6 months in arm A and 9 months in arm B). The multivariate analysis confirmed that treatment arm was not predictive for EFS, while thrombocytopenia grade at study entry (grade ≥ 1 v 0; HR, 2.4; P = .04) and number of previous chemotherapy regimens (≥ 2 v ≤ 1 HR, 2.0; P = .03) were independently predictive.
Toxicity
During the induction phase, the majority of toxicities were mild infusion-related symptoms during the first infusion. Additionally, 13% and 11% of patients had grade 3 and 4 toxicity, respectively, including two episodes (one patient) of pneumonia, three episodes (two patients) of pain, three cases (two patients) of dyspnea, three reports (one patient) of thrombocytopenia, one major cardiac event, one allergic reaction, and one case of nausea. Cardiac ejection fraction was reassessed in 40 patients at week 12 and remained stable (from 63% [range, 50% to 78%] at baseline to 64% [range, 44% to 79%]). Seventeen cases of serious adverse events were documented during induction treatment. Two patients died of causes other than progression—one of a probable myocardial infarction and one of Pneumocystis carinii pneumonia.
After random assignment, nonhematological grade 3 or 4 toxicity included two cases of pain, one diarrhea episode, and one case of cholecystitis in arm A; and one patient experienced three episodes of pneumonia, one case of hepatitis, and one renal failure case in arm B. Grade 3 or 4 hematological toxicity was observed in 13% of patients in arm A and 9% in arm B. Among 19 patients with cardiac function reassessed after random assignment (seven in arm A and 12 in arm B), the median ejection fraction was 62% (range, 60% to 66%) in arm A and 59% (range, 40% to 75%) in arm B. The number of serious adverse events after random assignment (including follow-up period) was six in arm A (including two second tumors) and eight in arm B (including one second tumor, one septic shock, and one cardiac death).
The lymphocytes subset analysis (performed excluding patients with baseline WBC > 10 x 109/L who were suspected of having leukemic disease) showed a very early and important reduction of circulating B cells, while T-helper, T-suppressor, and natural killer cells remained approximately stable during the induction phase. The evolution of B lymphocytes for randomly assigned patients throughout the first 12 months of follow-up is shown in Figure 4. Median B-cell levels returned to levels similar to baseline values after 9 months in arm A, while in arm B, the B-cell recovery had still not taken place after 12 months. The difference at 9 months was significant (median, 90.5 x 106/L v 13.9 x 106/L; P = .03).
DISCUSSION
In phase II trials, a minimum RR of 30% is required for a drug to be considered a potentially active new anticancer agent. Although rituximab reaches this threshold, the relatively short duration of response (compared with what was seen in other types of lymphomas), and the fact that this RR is surprisingly not higher in chemotherapy-nave patients relegates the antibody in the group of modestly active drugs for this disease. It is therefore not astonishing that the effect of a prolonged administration is similarly modest.
Compared with the only other sufficiently sized study of single-agent rituximab in MCL,7 our trial has the following advantages: larger sample size, histologic diagnosis reviewed centrally according to strict criteria, and evaluation of the evolution of lymphocyte subsets and molecular remission (at least in a subset of patients). Unfortunately, although this latter analysis was performed in the majority of patients at trial entry, it was repeated in only a minority at restaging, reducing the reliability of the molecular response results. Compared with the other series, we observed a slightly lower RR (possibly due to our more stringent response criteria), a similar response duration, and concordant prognostic factors.7
The parallel trial, which was conducted with exactly the same design by our group in patients with follicular lymphoma, showed that (in rituximab responders only) the prolonged schedule significantly doubles the median EFS (from 12 to 23 months, P = .02), and we concluded that single-agent rituximab, particularly if given with a prolonged schedule, is a valid treatment option for follicular lymphoma.12 The same conclusion cannot be made for MCL. In this disease, single-agent rituximab is less active at either schedule; therefore, the drug should instead be incorporated into combination schemes with chemotherapy. Several phase II trials and preliminary data from randomized studies confirm this assumption.15,16 Because rituximab can induce molecular remissions, its use in combination with chemotherapy as an in vivo purging agent for peripheral stem collection is theoretically sound; this application in MCL has already shown promising results.17 The use of rituximab maintenance therapy following autologous stem-cell transplantation also seems to improve outcomes in patients with advanced MCL.18
It is not surprising that MCL responds less well to single-agent rituximab than indolent lymphomas (the same is true for chemotherapy), but it is surprising that the agent shows a similar effect in chemotherapy-nave and pretreated patients: in oncology practice, RR and duration are usually higher in newly diagnosed patients. Even though the same observation was made by Foran et al,7 in our trial, the number of previous regimens in pretreated patients is a prognostic factor, raising the suspicion that the amount of previous treatment is indeed important. The similar outcomes in chemotherapy-nave and pretreated patients observed in our trial could be explained by the small group sizes (38 and 66 patients, respectively) and/or an unfavorable selection of chemotherapy-nave patients proposed for rituximab monotherapy as a first-line treatment.
Authors' Disclosures of Potential Conflicts of Interest
Acknowledgment
We thank Patricia Katz and Corinne Friedly for central data management, Dina Delle Pezze for the fluorescence-activated cell sorting analysis of lymphocyte subsets, and Michela Gisi for assistance in molecular biology analysis.
NOTES
Supported in part by research funding from Roche Pharma Schweiz AG to the Swiss Group for Clinical Cancer Research (SAKK).
Presented orally (in summary) at the 2003 Annual Meeting of the European Haematology Association (EHA), Lyon, France, June 2003.
Authors' disclosures of potential conflicts of interest are found at the end of this article.
REFERENCES
1. Zucca E, Roggero E, Pinotti G, et al: Patterns of survival in mantle cell lymphoma. Ann Oncol 6:257-262, 1995
2. Weisenburger DD, Armitage JO: Mantle cell lymphoma: An entity comes of age. Blood 87:4483-4494, 1996
3. McLaughlin P, Grillo-López AJ, Link BK, et al: Rituximab chimeric anti-CD20 monoclonal antibody therapy for relapsed indolent lymphoma: Half of patients respond to a four-dose treatment program. J Clin Oncol 16:2825-2833, 1998
4. Conconi A, Martinelli G, Thieblemont C, et al: Clinical activity of rituximab in extranodal marginal B-cell lymphoma of MALT type. Blood 102:2741-2745, 2003
5. Coiffier B, Haioun C, Ketterer N, et al: Rituximab (anti-CD20 monoclonal antibody) for the treatment of patients with relapsing or refractory aggressive lymphoma: A multicenter phase II study. Blood 92:1927-1932, 1998
6. Coiffier B, Lepage E, Brière J, et al: CHOP chemotherapy plus rituximab compared with CHOP alone in elderly patients with diffuse large-B-cell lymphoma. N Engl J Med 346:235-242, 2002
7. Foran JM, Cunningham D, Coiffier B, et al: Treatment of mantle-cell lymphoma with rituximab (chimeric monoclonal anti-CD20 antibody): analysis of factors associated with response. Ann Oncol 11:117-121, 2000
8. Nguyen DT, Amess JA, Doughty H, et al: IDEC-C2B8 anti-CD20 (rituximab) immunotherapy in patients with low-grade non-Hodgkin's lymphoma and lymphoproliferative disorders: Evaluation of response on 48 patients. Eur J Haematol 62:76-82, 1999
9. Berinstein NL, Grillo-López AJ, White CA, et al: Association of serum rituximab (IDEC-C2B8) concentration and anti-tumor response in the treatment of recurrent low-grade or follicular non-Hodgkin's lymphoma. Ann Oncol 9:995-1001, 1998
10. Pocock SJ: Clinical Trials: A Practical Approach. Chichester, UK, John Wiley and Sons, 1983, pp 84-86
11. Ghielmini M, Hsu Schmitz SF, Bürki K, et al: The effect of rituximab on patients with follicular and mantle-cell lymphoma: Swiss Group for Clinical Research (SAKK). Ann Oncol 11:123-126, 2000
12. Ghielmini M, Hsu Schmitz SF, Cogliatti S, et al: Prolonged treatment with rituximab in patients with follicular lymphoma significantly increases event-free survival and response duration compared with the standard weekly x 4 schedule. Blood: Prepublished online February 24, 2004:DOI 10.1182/blood-2003-10-3411
13. Bertoni F, Roggero E, Luscieti P, et al: Clonality assessment in blood of patients with mantle cell lymphoma. Leuk Lymphoma 32:372-379, 1999
14. Klein JP, Moeschberger ML. Survival Analysis: Techniques for Censored and Truncated Data. New York, NY, Springer-Verlag Inc, 2003, section 7.6
15. Dreyling M H., Forstpointer R. Repp R., et al: Combined immuno-chemotherapy (R-FCM) results in superior remission and survival rates in recurrent follicular and mantle cell lymphoma - final results of a prospective randomized trial of the German Low Grade Lymphoma Study Group (GLSG). Blood 102: 2003 (abstr 351; suppl)
16. Hiddemann W, Dreyling MH, Forstpointner R, et al: Combined immuno-chemotherapy (R-CHOP) significantly improves time to treatment failure in first line therapy of follicular lymphoma: Results of a prospective randomized trial of the German Low Grade Lymphoma Study Group (GLSG). Blood 102: 2003 (abstr 352; suppl)
17. Romaguera J, Cabanillas F, Dang NH, et al: Mantle cell lymphoma (MCL): High rates of complete remission (CR) and prolonged failure-free survival (FFS) with Rituxan-Hyper CVAD (R-HCVAD) without stem cell transplant (SCT). Blood 98: 2001 (abstr 3030; suppl)
18. Mangel J, Leitch HA, Connors JM, et al: Intensive chemotherapy and autologous stem cell transplantation plus rituximab is superior to conventional chemotherapy for newly diagnosed advanced stage mantle-cell lymphoma: A matched pair analysis. Ann Oncol 15:283-289, 2004(Michele Ghielmini, Shu-Fa)