Executive Director Certara USA, Inc, United States
Disclosure(s):
Michael G. Dodds, PhD: No financial relationships to disclose
Objectives: Lymphodepletion (LD) conditioning regimen prior to CAR T-cell therapy treatment has shown to play a critical role in CAR T-cell therapy success [1-4]. It creates an immunosuppressive environment for T-cell engraftment and expansion. Limited optimization of LD is typically performed while evaluating dose levels of CAR T-cells in Phase 1/2, setting the stage for underperformance of promising emerging interventions and regulatory scrutiny (i.e. Project Optimus) especially in the allogeneic CAR T-cell therapy space when adding alemtuzumab (ALEM) to the standard LD regimen of cylophosphamide (CY) and fludarabine (FL) to overcome potential alloimmune rejections.
Methods: Solutions to the problem were studied by simulation, selecting single CAR T-cell dose levels of 50, 100, 150 million cells, FL dose levels of 20, 30, 40 mg/m2 for three days, CY dose levels of 300, 400, 500 mg/m2 for three days, and ALEM dose levels of 0, 20, 30 mg for three days. A full factorial design would require 3^4=81 cohorts, which is impractical for a Phase 1/2 study. A Definitive Screening Design (DSD; varying all four factors over n=9 cohorts) was used to augment a traditional monotherapy design (Mono; varying only CAR T-cell dose levels over n=3 cohorts) for a total of n=11 cohorts (one cohort of DSD and Mono are identical) [5].
Results: Simulations of the probability of a patient receiving a favorable benefit-risk outcome, p(B-R), were conducted across plausible ranges of CAR T-cell + LD dose-response relationships to compare each approach to the true but unknown optimal p(B-R). The Mono, DSD and true optimal p(B-R) estimates were mean 58.2% (standard deviation (sd) 23%), 76.8% (14.6%), and 78.3% (13.6%), respectively. The odds ratio of Mono and DSD to true optimal was mean 0.419 (0.242) and 0.928 (0.137), respectively. Thus, the odds that patients would receive a favorable benefit-risk profile using a CART+LD regimen identified by the augmented DSD approach relative to the typical Mono approach was 2.37-fold higher in favor of the augmented DSD approach.
Conclusions: Optimizing CAR T-cell and LD dose regimens in Phase 1/2 clinical trials is paramount to the success and optimal benefit-risk outcome of CAR T-cell therapies. Beyond optimizing patient benefit-risk ratio and meeting Project Optimus guidance, drug development timelines can be accelerated: a confirmatory trial with one cohort is sufficient following a well-designed adaptive Phase 1/2 trial design.
Citations: 1. Turtle et.al. Sci Transl Med. 2016 Sep 7;8(355):355ra116. 2. Derippe et.al. Cancer Research Communications 2 November 2022; 2 (11): 1532–1544. 3. Fabrizio et.al. Optimal fludarabine lymphodepletion is associated with improved outcomes after CAR T-cell therapy. Blood Adv 2022; 6 (7): 1961–1968. 4. Haag et.al. Solid tumor patients require higher cylophosphamide dose than multiple myeloma patients to achieve adequate lymphodepletion necessary to enable allogeneic CAR-T expansion. Poster CT070, AACR 2024. 5. Jones and Nachtsheim. Journal of Quality Technology, 43(1), 1–15, 2017.
