Abstract
High-dose cytarabine containing immuno-chemotherapy followed by autologous stem cell transplantation (ASCT) has been established as standard first-line treatment in younger mantle cell lymphoma (MCL) patients (ESMO guideline, Ann Oncol 2013). However, the role of total body irradiation (TBI) as conditioning regimen for ASCT has not been clarified so far. A retrospective EBMT survey of 418 MCL patients suggested a benefit of TBI with respect to the rate of relapse in patients who achieved a partial remission before ASCT (Rubio et al., ASH 2010). We combined the individual patient data of the Nordic MCL2 trial (Geisler et al., Blood 2008), the HOVON 45 trial (van’t Veer et al., BJH 2008, both without TBI), and the high-dose cytarabine containing study arm of the European MCL Younger trial (Hermine et al., ASH 2012, with TBI) to evaluate potential differences in long term outcome.
All trials included patients with advanced stage MCL up to 66 years and a central review of histology. Induction treatment of Nordic MCL2 consisted of 6 alternating courses of Maxi-CHOP or high-dose cytarabine (12g/m2 per cycle, 8g/m2 in patients > 60 years, total cumulative dose 24-36g/m2) combined with 4-6 cycles rituximab, in HOVON 45 three courses R-CHOP were followed by one cycle of high-dose cytarabine (16g/m2), and in MCL Younger patients received 6 alternating courses of R-CHOP or R-DHAP (4g/m2 cytarabine per cycle, total cumulative dose 12g/m2). Conditioning before ASCT consisted of BEAM or BEAC in Nordic MCL2, and BEAM in HOVON 45, whereas a TBI-containing regimen (TAM: 10 Gray TBI, 6g/m2 cytarabine, 140mg/m2 melphalan) was applied in MCL Younger. We compared baseline characteristics, progression free survival (PFS) overall survival (OS) after ASCT from the three patient cohorts with adjustment for MIPI (Hoster et al., Blood 2008) and quality of remission before ASCT.
In Nordic MCL2, HOVON 45, and MCL Younger 145, 61, and 171 patients were transplanted in complete (CR) or partial (PR) remission (median age 57, 55, and 56 years, p=0.24). According to MIPI, 46%, 56%, and 67% were low risk, whereas 22%, 13%, and 11% were high risk patients (p=0.006). Five-year PFS after ASCT was 61%, 50%, and 78%, respectively (p<0.001) after a median observation time of 6.6, 6.5, and 4.2 years). After adjustment for MIPI score and quality of response (CR/unconfirmed CR vs. PR) before ASCT, the hazard ratio for PFS after ASCT of MCL Younger vs. Nordic MCL2 was 0.75 (p=0.17). Among the patients with CR or unconfirmed CR before ASCT, the adjusted hazard ratio was 0.91 (p=0.76) in comparison to 0.66 (p=0.17) among the PR patients. On the other hand, the hazard ratio of MCL Younger in comparison to HOVON 45 was 0.45 (p=0.001) after adjustment for MIPI score. OS after ASCT was not different between the three cohorts with 5-year OS rates of 75%, 71%, and 80%, respectively (p=0.61).
The longer PFS after ASCT of MCL Younger compared to HOVON 45 patients may be due to the different time schedule of cytarabine during induction or to a higher efficacy of TBI vs. BEAM, as the total chemotherapy doses were comparable and we adjusted for MIPI. We could not see an improved PFS after ASCT of MCL Younger compared to Nordic MCL2 patients transplanted in CR or unconfirmed CR and, after adjustment for the different risk profiles, this was also not clearly seen for patients transplanted in PR. Taking into account the substantially higher cytarabine dose in Nordic MCL2 during induction one might speculate that this fact at least partially compensated for a higher efficacy of TBI. Multivariable analyses incorporating additional prognostic factors (Ki-67 and MRD) will help to clarify the impact of the individual parts of the current multimodal approach in younger MCL patients.
On behalf of the European Mantle Cell Lymphoma Network
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.
