Landmark and longitudinal exposure-response analyses for multiple efficacy and safety endpoints to justify the clinical dose of valemetostat for adult T-cell leukemia/lymphoma
Objectives: Valemetostat (DS-3201) is an orally administered dual inhibitor of enhancer of zeste homolog (EZH) 1 and EZH2, which is under development for the treatment of patients with relapsed/refractory (R/R) adult T-cell leukemia/lymphoma (ATL) and other cancers. The objectives of this work were to determine (i) the valemetostat exposure associated with acceptable efficacy and safety or tolerability in R/R ATL patients and (ii) whether there were any clinically identifiable subpopulations that would warrant a dosing regimen other than the recommended dose (200 mg QD).
Methods: The analysis was conducted using data from Studies DS3201-A-J101 and DS3201-A-J201 with non-Hodgkin’s lymphoma (NHL), including ATL. The efficacy endpoints included central objective response rate (ORR) and investigator ORR. The safety endpoints included three lab-based assessments (Grade 3 or higher), anemia, neutrophil count decreased and platelet count decreased, plus dose reduction due to adverse events (AEs), dose interruption due to AEs and any AE Grade 3 or higher. Each of these outcome variables were dichotomous, therefore logistic regression was performed in a Bayesian framework. Models of the efficacy endpoints included ATL patients only (n=39) while models of safety endpoints included all NHL patients (n = 102). Unbound AUCss was utilized as an exposure metric for all endpoints. The models were also used to establish a region of practical equivalence (ROPE). The ROPE was defined to achieve satisfactory efficacy (probability of objective response greater than 30%) for typical patients, while ensuring acceptable safety (probability of dose reduction due to AE of less than 50%) for 90% of patients. In addition, longitudinal platelet count data were analyzed by nonlinear mixed effects modeling and used for simulations to characterize platelet dynamics in different scenarios. These analyses were conducted using NONMEM® 7.5 and R 4.0.3.
Results: Positive associations with exposure were estimated for all efficacy and safety endpoints. Applying working definitions described above, a ROPE of (0 to 1255 ng*hr/mL) and modified ROPE with direct empirical support of (184 to 887 ng*hr/mL) were estimated as target exposure ranges. Simulations at each dose level suggested that 200 mg QD dosing is most likely to achieve exposures within the modified ROPE for subpopulations of interest. The longitudinal platelet model included two proliferation compartments, both representing the stem cells and the proliferating precursors cells in the bone marrow with different sensitivity to valemetostat. The longitudinal model predicted the incidence of treatment termination due to Grade 4 platelet count decreased with or without dose adjustment based on platelet dynamics (2.36% with dose adjustment vs 8.45% dose held), which justified the proposed dose adjustment guidance.
Conclusions: Positive exposure-response relationship for all endpoints were confirmed. Target exposure range, which provides satisfactory efficacy and acceptable safety was established. The dose adjustment guidance based on platelet count was justified. These analyses justified the recommended dose of 200 mg and dose adjustment for patients with R/R ATL.
Author(s)
- Masato Fukae, Daiichi Sankyo Co., Ltd. (Presenting Author)
- Kyle Baron, Metrum Research Group (CoAuthor)
- James Rogers, Metrum Research Group (CoAuthor)
- Ramon Garcia, Metrum Research Group (CoAuthor)
- Masaya Tachibana, Daiichi Sankyo Co., Ltd. (CoAuthor)
- John Mondick, Metrum Research Group (CoAuthor)
- Takako Shimizu, Daiichi Sankyo Co., Ltd. (CoAuthor)