Comparative Efficacy of Asciminib vs. Second-Generation TKIs in Newly Diagnosed PhCML Using Unanchored Matching-Adjusted Indirect Comparisons
Author(s)
Valentine Laizet, MSc1, Katerina Papadimitropoulou, PhD2, Shaun Walsh, MSc3, Isabelle Lundqvist, MSc4, Vladimir Babiy, PharmD, PhD5.
1Amaris, Toronto, ON, Canada, 2Amaris, Lyon, France, 3Novartis, Dublin, Ireland, 4Novartis, Kista, Sweden, 5Novartis, London, United Kingdom.
1Amaris, Toronto, ON, Canada, 2Amaris, Lyon, France, 3Novartis, Dublin, Ireland, 4Novartis, Kista, Sweden, 5Novartis, London, United Kingdom.
OBJECTIVES: Asciminib 80mg (ASC) demonstrated favourable efficacy and safety versus standard-of-care tyrosine kinase inhibitors (TKIs) in newly diagnosed adults with Philadelphia chromosome-positive chronic myeloid leukaemia (Ph+CML) in the phase-III ASC4FIRST trial. Indirect comparative efficacy of ASC versus each individual second-generation (2G)-TKIs including nilotinib 300mg (NIL), bosutinib 400mg (BOS) and dasatinib 100mg (DAS) was assessed in this work.
METHODS: Individual patient data (IPD) from ASC4FIRST were restricted to ASC-treated patients randomized within the 2G-TKI stratum (n=100) and weighted to match aggregate-level comparator populations on key treatment effect modifiers (TEMs) and prognostic factors (PFs) (age, sex, race, white blood cell and platelet count). Risk score adjustment was not feasible due to differing systems across trials (European Treatment and Outcome Study long‐term survival (ELTS) versus Sokal or Hasford). Unanchored matching-adjusted indirect comparisons estimated weighted odds ratios and 95% confidence intervals, using logistic regression and aggregate data from NIL (ENESTnd; n=282), DAS (DASISION; n=259) and BOS (BFORE; n=268) for major molecular response (MMR) and deep molecular response (DMR/MR4.5) by 96 weeks.
RESULTS: Matching successfully led to balanced available characteristics and sufficient effective sample sizes (ASC versus NIL: 72.61; DAS: 84.82; BOS: 61.57). ASC demonstrated significant improvements in MMR versus DAS (OR, 2.58 [1.39-4.81]) and BOS (2.46 [1.20-5.07]) and favourable trends versus NIL (1.94 [1.00-3.75]). ASC also showed significant benefits in DMR/MR4.5 versus DAS (2.09 [1.20-3.64]), with favourable trends versus NIL (1.38 [0.77-2.44]) and BOS (1.64 [0.87-3.11]). Increase in high-risk ELTS patients post-adjustment (14% pre-adjustment, 20% and 17% post-adjustment in ASC versus BOS and NIL respectively) may explain the non-significant trends.
CONCLUSIONS: Although the adjustment was not feasible on risk scores, results support ASC as a first-line therapy for newly diagnosed Ph+CML patients, with consistent numerical or significant improvements in MMR and DMR/MR4.5 compared with 2G-TKIs.
METHODS: Individual patient data (IPD) from ASC4FIRST were restricted to ASC-treated patients randomized within the 2G-TKI stratum (n=100) and weighted to match aggregate-level comparator populations on key treatment effect modifiers (TEMs) and prognostic factors (PFs) (age, sex, race, white blood cell and platelet count). Risk score adjustment was not feasible due to differing systems across trials (European Treatment and Outcome Study long‐term survival (ELTS) versus Sokal or Hasford). Unanchored matching-adjusted indirect comparisons estimated weighted odds ratios and 95% confidence intervals, using logistic regression and aggregate data from NIL (ENESTnd; n=282), DAS (DASISION; n=259) and BOS (BFORE; n=268) for major molecular response (MMR) and deep molecular response (DMR/MR4.5) by 96 weeks.
RESULTS: Matching successfully led to balanced available characteristics and sufficient effective sample sizes (ASC versus NIL: 72.61; DAS: 84.82; BOS: 61.57). ASC demonstrated significant improvements in MMR versus DAS (OR, 2.58 [1.39-4.81]) and BOS (2.46 [1.20-5.07]) and favourable trends versus NIL (1.94 [1.00-3.75]). ASC also showed significant benefits in DMR/MR4.5 versus DAS (2.09 [1.20-3.64]), with favourable trends versus NIL (1.38 [0.77-2.44]) and BOS (1.64 [0.87-3.11]). Increase in high-risk ELTS patients post-adjustment (14% pre-adjustment, 20% and 17% post-adjustment in ASC versus BOS and NIL respectively) may explain the non-significant trends.
CONCLUSIONS: Although the adjustment was not feasible on risk scores, results support ASC as a first-line therapy for newly diagnosed Ph+CML patients, with consistent numerical or significant improvements in MMR and DMR/MR4.5 compared with 2G-TKIs.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
CO56
Topic
Clinical Outcomes
Topic Subcategory
Comparative Effectiveness or Efficacy
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, Systemic Disorders/Conditions (Anesthesia, Auto-Immune Disorders (n.e.c.), Hematological Disorders (non-oncologic), Pain)