ABSTRACT
Background: Peripheral T-cell lymphoma (PTCL) remains a therapeutic challenge. Due to the rarity and the heterogeneity of PTCL, no consensus has been achieved regarding even the type of first-line treatment. The benefit of autologous stem-cell transplantation (ASCT) is, therefore, still intensely debated. Patients and methods: In the absence of randomized trials addressing the role of ASCT, we performed a large multicentric retrospective study and used both a multivariate proportional hazard model and a propensity score matching approach to correct for sample selection bias between patients allocated or not to ASCT in intention-to-treat (ITT). Results: Among 527 patients screened from 14 centers in France, Belgium and Portugal, a final cohort of 269 patients ≤65 years old with PTCL-not otherwise specified (NOS) (N = 78, 29%), angioimmunoblastic T-cell lymphoma (AITL) (N = 123, 46%) and anaplastic lymphoma kinase-positive anaplastic large cell lymphoma (ALK-ALCL) (N = 68, 25%) with partial (N = 52, 19%) or complete responses (N = 217, 81%) after induction was identified and information about treatment allocation was carefully collected before therapy initiation from medical records. One hundred and thirty-four patients were allocated to ASCT in ITT and 135 were not. Neither the Cox multivariate model (HR = 1.02; 95% CI: 0.69-1.50 for PFS and HR = 1.08; 95% CI: 0.68-1.69 for OS) nor the propensity score analysis after stringent matching for potential confounding factors (logrank P = 0.90 and 0.66 for PFS and OS, respectively) found a survival advantage in favor of ASCT as a consolidation procedure for patients in response after induction. Subgroup analyses did not reveal any further difference for patients according to response status, stage disease or risk category. Conclusions: The present data do not support the use of ASCT for up-front consolidation for all patients with PTCL-NOS, AITL, or ALK-ALCL with partial or complete response after induction.
