Tafasitamab Plus Lenalidomide (TAFA+LEN) Versus Pola-BR, R2, and CAR-T in Non-Transplant Eligible (NTE) Relapsed/Refractory (R/R) DLBCL: A Post Hoc Analysis of the RE-MIND2 Study Comparing Multiple Cohort-Balancing Approaches
Author(s)
Cordoba R1, Zinzani PL2, Joffe E3, Yoon DH4, Peters A5, Waltl EE6, Alvero C7, Sabatelli L8, Salles G3, Nowakowski GS9
1Fundacion Jimenez Diaz University Hospital, Madrid, Spain, 2Institute of Hematology "Seràgnoli", University of Bologna, Bologna, Italy, 3Memorial Sloan Kettering Cancer Center, New York, NY, USA, 4Department of Oncology, Asan Medical Center, Seoul, Korea, Republic of (South), 5Department of Oncology, University of Alberta, Edmonton, AB, Canada, 6MorphoSys AG, Planegg, Germany, 7MorphoSys US Inc, Boston, MA, USA, 8Incyte Biosciences International Sàrl, Morges, VD, Switzerland, 9Division of Hematology, Mayo Clinic, Rochester, MN, USA
Presentation Documents
OBJECTIVES: In the single-arm, phase 2, L-MIND study (NCT02399085), TAFA+LEN demonstrated efficacy in NTE R/R diffuse large B-cell lymphoma (DLBCL) patients. The observational, retrospective cohort study, RE-MIND2 (NCT04697160), compared patient outcomes from L-MIND with matched patients treated with NCCN-/ESMO-recommended therapies. This post hoc analysis compared multiple cohort-balancing approaches, using 2 baseline prognostic factor and/or treatment effect-modifier sets (reported in Nowakowski et al. Blood 2021;abstract 183 and Nowakowski et al. DOI: 10.1158/1078-0432.CCR-21-3648), to estimate overall survival (OS) benefit of TAFA+LEN versus polatuzumab vedotin + bendamustine + rituximab (pola-BR), rituximab + LEN (R2), and CD19 CAR-T therapies (CAR-T) in RE‑MIND2.
METHODS: Patients eligible for analyses were 76, 92, 92, and 140, respectively, for TAFA+LEN, pola-BR, R2, and CAR-T. For each comparator, 5 cohort-balancing methods were used: 1:1 nearest-neighbor (NN) matching, 6 covariates (I) excluding patients with missing values or (II) with multiple imputations; inverse probability of treatment weighting, 9 covariates (III) excluding patients with missing values or (IV) with multiple imputations; and 1:1 NN matching, 9 covariates (V) with multiple imputations.
RESULTS: For TAFA+LEN versus pola-BR, 4/5 analyses showed significant improvement in OS; HR (95% CI) for methods I-V were 0.44 (0.20-0.96; P=0.0381), 0.35 (0.19-0.66; P=0.0012), 0.50 (0.20-1.25; P=0.1389), 0.28 (0.18-0.45; P<0.0001), 0.42 (0.23-0.78; P=0.0061). For TAFA+LEN versus R2, 5/5 analyses showed significant improvement in OS: HR (95% CI): 0.44 (0.22-0.85; P=0.0143), 0.42 (0.23-0.76; P=0.0042), 0.35 (0.17-0.73; P=0.0050), 0.34 (0.21-0.55; P<0.0001), 0.47 (0.26-0.83; P=0.0093). For TAFA+LEN versus CAR-T, no significant improvement in OS was seen: HR (95% CI): 0.95 (0.48-1.91; P=0.8915), 0.98 (0.48-2.03; P=0.9635), 0.52 (0.25-1.07; P=0.0735), 0.91 (0.55-1.48; P=0.6934), 0.88 (0.44-1.78; P=0.7302).
CONCLUSIONS: Multiple cohort-balancing approaches consistently showed TAFA+LEN is associated with OS benefit versus pola-BR and R2, and comparable OS versus CAR-T in NTE R/R DLBCL. Limitations due to sample size, retrospective comparator data, and known methodological features of indirect comparisons warrant further investigation.
Conference/Value in Health Info
Value in Health, Volume 25, Issue 12S (December 2022)
Code
CO67
Topic
Clinical Outcomes
Topic Subcategory
Comparative Effectiveness or Efficacy
Disease
STA: Drugs