- © 2009 by American Society of Clinical Oncology
Pre-Emptive Treatment With Rituximab of Molecular Relapse After Autologous Stem Cell Transplantation in Mantle Cell Lymphoma
- Niels S. Andersen,
- Lone B. Pedersen,
- Anna Laurell,
- Erkki Elonen,
- Arne Kolstad,
- Anne Marie Boesen,
- Lars M. Pedersen,
- Grete F. Lauritzsen,
- Roald Ekanger,
- Herman Nilsson-Ehle,
- Marie Nordström,
- Susanne Fredén,
- Mats Jerkeman,
- Mikael Eriksson,
- Jaan Väärt,
- Beatrice Malmer and
- Christian H. Geisler
- From the Department of Hematology, Rigshospitalet, Copenhagen; Department of Hematology, Aarhus University Hospital, Aarhus; Department of Hematology, Herlev University Hospital, Herlev, Denmark; Department of Hematology, Uppsala University Hospital, Uppsala; Department of Hematology, Sahlgrenska University Hospital, Gothenburg; Department of Hematology, Karolinska Institute, Stockholm; Department of Oncology, Jönköping Hospital, Jönköping; Department of Oncology, Lund University Hospital, Lund; Department of Medicine, Skövde Hospital, Skövde, Sweden; Department of Hematology, Helsinki University Central Hospital, Helsinki; Department of Radiation Sciences, Oncology, Umeå University, Umeå, Finland; Department of Oncology, Norwegian Radium Hospital, Oslo; and Department of Oncology, Haukeland University Hospital, Bergen, Norway.
- Corresponding author: Niels S. Andersen, MD, PhD, Department of Hematology L-4041, Rigshospitalet, Blegdamsvej 9, 2100 Copenhagen, Denmark; e-mail: ns_andersen{at}msn.com.
-
Presented in part at the 49th Annual Meeting of the American Society of Hematology December 8-11, 2007, Atlanta, GA.
Abstract
Purpose Minimal residual disease (MRD) is predictive of clinical progression in mantle-cell lymphoma (MCL). According to the Nordic MCL-2 protocol we prospectively analyzed the efficacy of pre-emptive treatment using rituximab to MCL patients in molecular relapse after autologous stem cell transplantation (ASCT).
Patients and Materials MCL patients enrolled onto the study, who had polymerase chain reaction (PCR) detectable molecular markers and underwent ASCT, were followed with serial PCR assessments of MRD in consecutive bone marrow and peripheral blood samples after ASCT. In case of molecular relapse with increasing MRD levels, patients were offered pre-emptive treatment with rituximab 375 mg/m2 weekly for 4 weeks.
Results Of 160 MCL patients enrolled, 145 underwent ASCT, of whom 78 had a molecular marker. Of these, 74 were in complete remission (CR) and four had progressive disease after ASCT. Of the CR patients, 36 underwent a molecular relapse up to 6 years (mean, 18.5 months) after ASCT. Ten patients did not receive pre-emptive treatment mainly due to a simultaneous molecular and clinical relapse, while 26 patients underwent pre-emptive treatment leading to reinduction of molecular remission in 92%. Median molecular and clinical relapse-free survival after pre-emptive treatment were 1.5 and 3.7 years, respectively. Of the 38 patients who remain in molecular remission for now for a median of 3.3 years (range, 0.4 to 6.6 years), 33 are still in clinical CR.
Conclusion Molecular relapse may occur many years after ASCT in MCL, and PCR based pre-emptive treatment using rituximab is feasible, reinduce molecular remission, and may prevent clinical relapse.
INTRODUCTION
Mantle-cell lymphoma (MCL) remains a B-cell lymphoma entity with a poor prognosis.1–3 Despite prolonged disease-free survival after high-dose regimens with stem cell support the majority of patients relapse.4–6 Anti-CD20 antibody therapy has proved important for response rates and disease-free survival in phase III studies of MCL as well as for successful in vivo stem cell purging in MCL cases undergoing autologous stem cell transplantation (ASCT).7–9 Induction of a clinical complete remission (CR) at time of transplant has been found significantly important for survival in previous studies.5,6,10 Therefore, most studies, including the present second Nordic MCL-2 study, have focused on strategies for intensification of induction.11–14
We and others have shown superior outcome in MCL patients obtaining continuous polymerase chain reaction (PCR) negative bone marrow (BM) and peripheral blood (PB) samples after ASCT.14–19 In contrast, poor prognosis was observed in cases with PCR conversion from negative to positive or rising quantitative PCR detectable levels of MRD in consecutive BM/PB samples after ASCT.17,18 MCL patients who remained standard nested PCR positive, but with stable or declining quantitative PCR detectable MRD levels had a better outcome.16,17 These findings suggested that PCR monitoring of MRD could define high-risk MCL patients and direct pre-emptive treatment with rituximab after ASCT in order to prevent clinical relapse. Whether the optimal post-transplant strategy should be standard maintenance to all patients, including long-term survivors without PCR detectable MRD, or PCR based pre-emptive treatment, excluding patients without molecular markers, remains to be studied. In the second Nordic (Denmark, Finland, Norway, and Sweden) multicenter single arm phase II trial, we aimed at assessing the feasibility and efficacy of PCR based pre-emptive treatment with four weekly doses of rituximab to patients in molecular relapse after ASCT.
PATIENTS AND METHODS
Patients
One hundred sixty patients were included in the second Nordic MCL protocol. Patient BM/PB samples were obtained at defined time points, see below, and shipped overnight to Copenhagen for centralized PCR analysis. The diagnosis of MCL was confirmed by central expert pathology review and based on morphology, immunophenotype, and genotype. All samples were procured only after patient informed consent and after regional review board and ethical institution approval of the protocol.
Treatment
The immunochemotherapy, conditioning, and ASCT treatment scheme is outlined in Figure 1.
After ASCT four weekly doses of anti-CD20 antibody treatment (rituximab: 375 mg/m2) were given to patients with a PCR detectable molecular relapse in BM and/or PB samples. In the event of a further molecular relapse repeated treatment with rituximab was allowed at the discretion of the investigator.
Clinical Response Evaluation and Follow-Up
All patients underwent initial and post-transplant staging according to standard procedure, including physical examination, blood tests, computed tomography scans, and BM aspiration and biopsy. Clinical and molecular response evaluation was done 2 months after ASCT and then every 4 to 6 months for the next 5 years. Response was defined according to National Cancer Institute criteria.20
No screening of infections, cytomegalovirus/Epstein-Barr virus reactivation, or immunoglobulin level assessment was planned during follow-up after ASCT or before and after pre-emptive treatment.
Molecular Follow-Up
Detection of a molecular relapse after ASCT followed two lines of direction depending on the standard nested PCR status of the first follow-up samples after ASCT. In patients where the initial post-transplant BM/PB sample was standard nested PCR negative, a conversion to standard nested PCR positive in any subsequent BM/PB sample defined a molecular relapse. In patients where the initial BM/PB sample was standard nested PCR positive after ASCT (or after pre-emptive treatment) we awaited the subsequent BM/PB sample. If this BM/PB sample was also standard nested PCR positive we proceeded to quantitative PCR analysis of these two consecutive samples. A significant (> 5-fold) increase of the real time quantitative PCR detectable MRD level in two consecutive BM or PB samples defined a molecular relapse.
End Points
The end points were relapse-free survival and duration of molecular remission after ASCT or pre-emptive treatment, respectively.
Relapse-free survival was calculated from the date of ASCT or pre-emptive treatment, respectively, to the date of clinical relapse, progression, or death of lymphoma, or to the date of last follow-up. Duration of molecular remission and molecular relapse-free survival was calculated from ASCT or pre-emptive treatment, respectively, to the date of molecular relapse, clinical relapse, death, or to the date of last molecular follow-up.
PCR Analysis
DNA was extracted from patient specimens using minipreps (Qiagen Inc, Valencia, CA). BM samples were mandatory in the protocol. This is based on previous findings where BM samples were more frequently PCR positive compared to PB samples.16 However, in most cases we also received an additional PB sample for PCR analysis. Diagnostic BM/PB samples were used for PCR primer design. The DNA content was determined by spectrophotometry.
A combined standard nested and quantitative real-time PCR assay for Bcl-1/Immunoglobulin heavy chain gene (IgH) and clonal IgH rearrangements was used to estimate MRD involvement in consecutive post-transplant BM/PB samples as previously described.16,17 The sensitivity of the standard nested PCR analysis remained more sensitive than the quantitative real time PCR analysis in our hands. In this study our focus was on the MRD kinetics and the MRD level difference between two consecutive follow-up samples rather than the exact tumor cell concentration. We, therefore, substituted patient specific standard curves based on serially diluted recombinant plasmids with a patient specific standard. The patient specific standard derived from the particular diagnostic BM sample of each patient. First, the relative tumor cell content of each post-transplant sample was estimated by comparison to the individual patient specific standard, according to the Applied Biosystem user manual.21 Subsequently, the MRD kinetic between to consecutive post-transplant samples was compared individually.
Statistics
Survival-rates were calculated according to the Kaplan-Meier method and differences between subgroups analyzed by the log-rank test. Statistical analyses were performed with GraphPad prism, version 2.01 (GraphPad Software, San Diego, CA).
RESULTS
A total of 160 patients (100%) with MCL were included in the second Nordic MCL trial (Fig 1; Outline of treatment plan). One hundred forty-five patients underwent ASCT.
PCR Markers
Since only acquisition of diagnostic BM/PB and not frozen lymph node tissue was mandatory in the protocol, molecular markers were only found in patients carrying t11,14 major translocation cluster breakpoints or with a significant tumor cell infiltration in the diagnostic BM or PB samples. In 92 (58%) of the 160 patients included, 34 patients (37%) carried PCR detectable BCL-1/JH and 58 (63%) clonal IGH rearrangements. In the remaining 68 of 160 included patients a diagnostic sample was either not received (13 cases) or no PCR marker could be detected in the diagnostic sample probably due to absent or low BM/PB tumor cell infiltration (55 cases). The survival of patients with a molecular marker detectable in diagnostic BM or PB samples was not significantly different from patients without a marker. However, patients with a PCR detectable BCL-1/IgH rearrangement had a significantly shorter progression-free survival rate, compared to patients without a marker (P = .005, data not shown), whereas for patients with only a PCR detectable clonal IgH rearrangement, this difference was not significant (P = .09).
PCR Detection of MRD in BM Versus PB After ASCT
A total of 875 post-transplant BM/PB samples were monitored for MRD. In 18 (5%) of 325 BM/PB pairs only the BM was standard nested PCR positive, while no PB samples were positive without a positive BM. The time points of these discordant BM/PB pairs were neither related to ASCT nor to pre-emptive rituximab therapy.
Clinical and Molecular Status After ASCT
Of the 145 patients (91%) who received transplants, 78 patients (49%) had a PCR marker available. In one additional case only one late PCR-positive post-transplant sample was available, and this case was excluded. Seventy-four of 78 patients were in clinical complete remission (CR) and four had progressive disease after ASCT (Fig 2; summary of clinical and molecular status).
Of the 74 patients in continuous clinical CR after ASCT, 38 also remained in continuous molecular remission for a median follow-up time of 3.3 years (range, 0.4 to 6.6 years). In four of these a clinical relapse was observed at day +180, +390, +630, and +990 after ASCT, without any PCR detectable MRD present in BM or PB before or at time of clinical relapse.
The remaining 36 of the 74 patients in continuous clinical CR after ASCT underwent a molecular relapse and were pre-emptive treatment candidates (Fig 2). Time to PCR detectable molecular relapse after ASCT ranged from 6 months up to 6 years, in six patients more than 2 years and in seven patients more than 3 years post-transplant (mean, 18.5 months).
Ten of the 36 patients who underwent molecular relapse did not receive pre-emptive treatment for the following reasons: six patients underwent simultaneous molecular and clinical relapse leaving no therapeutic window for pre-emptive treatment, four patients did not receive pre-emptive treatment after molecular relapse due to three delayed responses between the PCR laboratory and the treating physicians and one patient refusal, of whom three relapsed clinically and one remained PCR positive but quantitative PCR low (Fig 2).
Of note, only three of these nine clinically relapsed patients survived more than 6 months after relapse. No MRD monitoring was done after clinical relapse or salvage therapy.
Twenty-six of the 36 patients who underwent molecular relapse received pre-emptive treatment leading to reconversion to standard PCR negativity in 20 patients, four rounds led to a state of quantitative PCR negative/low (unique patient number [UPN] 3, 96, 124, 156) equaling molecular remission, and two patients (UPN 43, 6) relapsed clinically shortly after pre-emptive treatment (Fig 3A, 3B). Thus, 24 rounds of pre-emptive treatment (92%) led to reinduction of molecular remission for a median of 1.5 years (range, 0.2 to 3.2 years) with no long-term molecular remissions observed (Fig 4A). Clinically, 16 (62%) of the 26 assessable patients remain in CR (Fig 3A) while 10 (38%) have relapsed (Fig 3B) after pre-emptive treatment. The median clinical relapse-free survival was 3.7 years (range, 0.2 to 4.6 years; Fig 4B). Of note, the clinical relapse-free survival curve includes five patients (UPN 13, 75, 142, 78, 148; Fig 3A), who after a secondary molecular relapse, at the discretion of the treating physician, underwent an additional round of pre-emptive treatment, leading to reinduction of molecular remission and continuous clinical CR. Another five patients did not undergo pre-emptive treatment after a secondary molecular relapse and underwent clinical relapse either simultaneous (UPN 77, 132, 110) or shortly after (UPN 113, 117; Fig 3B).
Complications to Pre-Emptive Treatment
Only few, mild, and transcient adverse effects to the pre-emptive rituximab treatment were reported. In 20 evaluated patients, one suffered from upper respiratory tract adeno virus infection, one had transient neutropenia, and one had a herpes zoster reactivation.
Patient- and Disease-Specific Characteristics
In the majority of the 78 patients with a molecular marker, we were able to compare International Prognostic Index (IPI) and the Ki-67 proliferation index at the time of diagnosis with molecular results, but neither IPI (0 to 2 v 3 to 5) nor Ki-67 expression (< 30% v ≥ 30%) were significantly related to the length of molecular remission after ASCT (P = .99 and P = .29, respectively; data not shown). In 12 patients with early (< 1 year) molecular and clinical relapse after ASCT, the mean Ki-67 expression was relatively high (46%; range, 20% to 91%), compared to four patients (UPN 13, 75, 135, 77) with early molecular relapse but no or delayed clinical relapse (mean, 12.5%; range, 5% to 20%).
DISCUSSION
We here report the results of the first larger study of pre-emptive treatment in lymphoma. We found PCR-based pre-emptive treatment feasible in MCL patients in a large, prospective, multicenter trial. However, based on diagnostic BM/PB samples, and not frozen lymph node material, only 49% of the included patients had molecular markers available. In contrast to the typical outcome of MCL patients after molecular relapse described in the literature,16–19,22 the reinduction of molecular remission in 92% of the patients was paralleled by a median relapse-free survival of 3.7 years after pre-emptive treatment.
Patients with a molecular marker all have BM/PB lymphoma cell infiltration or PCR detectable BCL-1/IgH rearrangements with breakpoints in the major translocation cluster only, which is related to a more aggressive MCL subtype.23 This may conceal, in part, the benefit from interventions carried out in this group.
Reinduction of a molecular remission in our study most likely is accomplished by the pre-emptive treatment. This is supported by several previous studies where spontaneous molecular remission is rarely observed and where most cases show increasing MRD levels in consecutive samples, leading to subsequent clinical relapse. In our previous studies of the clinical significance of MRD in MCL patients and in studies of translocation (14,18) positive lymphoma, PCR conversion from negative to positive, as well as rising MRD levels, remained a poor prognostic factor.16,17,24 Moreover, a PCR positive status the first year after ASCT has proven highly related to poor outcome.14,19 In the 27 patients studied by Pott et al19 only one of 13 PCR-positive patients spontaneously became PCR negative again. Ladetto et al22 recently observed an increasing MRD level by surveillance of consecutive BM samples which became PCR positive during follow-up after transplant in three of three MCL patients studied. Like in our study, reinduction of molecular remission was also observed in three MCL patients undergoing pre-emptive treatment using rituximab and in two MCL patients undergoing rituximab consolidation at weeks 8 to 11 after ASCT.22,25 In this study, none of six patients with molecular relapse, who did not receive pre-emptive treatment became PCR negative again and five patients subsequently relapsed clinically and one remained in clinical CR with low, but detectable, levels of MRD.
We found no indication of a molecular relapse-free survival plateau after pre-emptive treatment. However, the pre-emptive treatment given, so far, to five of 10 patients with a second molecular relapse, led to PCR negativity again and continuous clinical remission. Although, only the first molecular relapse after ASCT should be treated per protocol, the good effect of this treatment made it natural to allow re-treatment of secondary molecular relapses at the discretion of the investigators.
The reported finding19 of a slow increase of MRD levels over 12 to 19 months before clinical relapse in patients with very late relapse after ASCT, as compared to a more rapid increase of MRD levels over a median of 7 months in patients with early relapse, suggests that the tempo of the evolving subclinical disease on the molecular level mirrors the tempo of the clinical disease. Accordingly, in our series, early molecular relapse (ie, within the first 2 years after ASCT) was associated with a short interval between molecular and clinical relapse, so short that it precluded pre-emptive therapy in a number of patients. In theory, the lack of a therapeutical window might be solved using more frequent sampling. In practice, though, BM aspirations at shorter intervals than carried out in our series are unrealistic, and, at least in our study, PB samples would not suffice. However, this may be a question of the sensitivity of the laboratory set-up. Pott et al18 found no significant difference between MRD levels in BM and PB, and Bahloul et al19 did show a correlation between rising MRD levels in consecutive PB samples after ASCT and subsequent clinical relapse. Late molecular relapses, here up to 6 years after ASCT, indicate, similar to acute lymphoblastic leukemia patients, that MCL patients may harbor dormant or slowly accumulating lymphoma cells surviving initial therapy but with malignant potential.26
Standard maintenance and consolidation therapy using rituximab treatment at fixed time points after ASCT has been studied in smaller MCL series with improved outcome—varying from one dose every 3 months for a period of 2 years to four weekly doses 2 or 3 and 6 months after ASCT.8,25,27,28 A needed randomized trial of rituximab maintenance after first-line therapy is under way.29
Thus, the optimal schedule, duration, and dosage for rituximab administration after ASCT remains to be clarified. A PCR-based pre-emptive therapy will, on one hand evidently mean undertreatment of patients without molecular markers available, while rituximab maintenance of all patients will mean overtreatment of a significant number of patients, who would have remained PCR negative (approximately 50% of the patients who received transplants) and in clinical CR for a prolonged period after ASCT without need of such maintenance. Although we only observed few and mild adverse effects after rituximab treatment, previously observed serious adverse effects and immunoglobulin level reduction must be taken into consideration, as well as negative effects of rituximab may potentially hamper further treatment options.
Regarding duration, most studies offer consolidation and maintenance strategies for 2years after transplant, but we and others have documented that molecular relapse may take place much later.19 Since the molecular relapse-free survival curve showed no plateau phase after pre-emptive treatment, additional rounds of rituximab or other therapies are needed. Finally, the exact dosage of rituximab needed remains to be studied.
Although Ki-67 and IPI did not directly predict the duration of molecular remission, they were strongly associated with outcome in the clinical part of this trial,14 and may still direct intervention after transplant. In conclusion, a combined strategy of securing circulating rituximab levels at time of high-dose therapy reported by Dreger et al30 and in our series14 as well as giving maintenance therapy every 3 to 6 months the first 2 years after ASCT, followed by PCR based pre-emptive treatment strategy addressing later molecular relapses, might prove successful.
Despite PCR analysis is rather laborious and expensive the strong clinical significance of MRD detection now established by the studies of Pott et al19 and us favors application of PCR analysis in future MCL studies.14 Mandatory lymph node biopsy and/or extended Bcl-1 and IgH primer design, as recently suggested by the BIOMED-2 report would significantly increase the number of cases with a molecular marker available and should be a prerequisite for future prospective studies with PCR-based treatment decisions.31
In conclusion, we present here data which clearly indicate a role of rituximab after ASCT in MCL, and our data suggest a necessity of achieving and maintaining molecular remission in order to prolong survival of MCL patients.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The author(s) indicated no potential conflicts of interest.
AUTHOR CONTRIBUTIONS
Conception and design: Niels S. Andersen, Anna Laurell, Erkki Elonen, Arne Kolstad, Anne Marie Boesen, Lars M. Pedersen, Grete F. Lauritzsen, Roald Ekanger, Herman Nilsson-Ehle, Marie Nordström, Susanne Fredén, Mats Jerkeman, Mikael Eriksson, Jaan Väärt, Beatrice Malmer, Christian H. Geisler
Collection and assembly of data: Niels S. Andersen, Lone B. Pedersen, Anna Laurell, Erkki Elonen, Arne Kolstad, Anne Marie Boesen, Lars M. Pedersen, Grete F. Lauritzsen, Roald Ekanger, Herman Nilsson-Ehle, Marie Nordström, Susanne Fredén, Mats Jerkeman, Mikael Eriksson, Jaan Väärt, Beatrice Malmer, Christian H. Geisler
Data analysis and interpretation: Niels S. Andersen, Lone B. Pedersen, Christian H. Geisler
Manuscript writing: Niels S. Andersen, Anna Laurell, Erkki Elonen, Arne Kolstad, Christian H. Geisler
Final approval of manuscript: Niels S. Andersen, Lone B. Pedersen, Anna Laurell, Erkki Elonen, Arne Kolstad, Anne Marie Boesen, Lars M. Pedersen, Grete F. Lauritzsen, Roald Ekanger, Herman Nilsson-Ehle, Marie Nordström, Susanne Fredén, Mats Jerkeman, Mikael Eriksson, Jaan Väärt, Beatrice Malmer, Christian H. Geisler
Acknowledgment
We thank the medical and nursing staffs for their collaboration, and the patients for their participation.
Footnotes
-
Supported by grants from the Danish Cancer Society, The Nordic Cancer Union, The John and Birthe Meyer Foundation, and the Novo Nordisk Foundation.
-
Written on behalf of the Nordic Lymphoma Group.
-
Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.
-
Clinical Trials repository link available on JCO.org.
-
Clinical trial information can be found for the following: ISRCTN87866680.
- Received November 27, 2008.
- Accepted March 19, 2009.