Asciminib Elicits Safe and Durable Responses in Previously Treated CML

Treatment with asciminib elicited greater efficacy and had a more tolerable safety profile compared with bosutinib for patients chronic myeloid leukemia in chronic phase following treatment with 2 or more TKIs.

Treatment with asciminib (Scemblix) elicited greater efficacy and had a more tolerable safety profile compared with bosutinib (Bosulif) for patients with chronic myeloid leukemia in chronic phase (CML-CP) following treatment with 2 or more tyrosine kinase inhibitors (TKIs), according to data from the phase 3 ASCEMBL trial (NCT03106779) presented at the 2023 ASH Annual Meeting.

“Most patients with chronic phase CML on long-term TKI therapy do have the potential for long-term chronic health issues,” lead study author Michael Mauro, MD, leader of the Myeloproliferative Neoplasms Program and hematologist at Memorial Sloan Kettering Cancer Center, said in a presentation of the data. “And those with treatment failure of at least 2 prior TKIs face additional challenges, such as high frequency of resistance causing mutations as well as toxicity.”

“Now, results from the ASCEMBL trial demonstrated superior efficacy and favorable safety at 24 and 96 weeks with asciminib versus bosutinib in these patients with chronic phase CML, previously treated with at least 2 TKIs,” said Mauro.

The major molecular response (MMR) rate for asciminib (33.8%) at week 156 (end of study) continued to be higher than with bosutinib (10.5%). After adjusting for baseline, the difference for major cytogenic response was 23.2% (95% CI, 13.1%-33.2%; 2-sided P<.001). The BCL-ABL1 ≤1% rate at week 156 in patients without the major cytogenic response of 23.2% at baseline continued to be higher when treated with asciminib than bosutinib (43.0% vs 11.1%). 

In terms of progression-free survival (PFS) and overall survival (OS), the investigators found that the PFS rate at 3 years was 85.2% (95% CI, 76.8%-90.7%) with asciminib and was 84.0% 95% CI; 67.5%-92.6%) with bosutinib. The 5-year OS rate was 87.8% (95% CI, 78.7%-93.1%) with asciminib and was 89.7% (95% CI, 76.3%-95.7%) with bosutinib.

A total of 233 patients in the study with CML-CP following ≥2 prior TKIs, experiencing intolerance or lack of efficacy according to 2013 ELN recommendations were randomized 2:1 into the asciminib cohort (n = 156) or bosutinib cohort (n = 76). Patients either received 40 mg of asciminib twice daily or received 500 mg of bosutinib once daily.

The investigators noted that if patients receiving bosutinib did not meet the treatment criteria according to 2013 ELN recommendations, they could switch to asciminib and were analyzed separately. Patients who were intolerant to bosutinib and discontinued the treatment could not switch to asciminib.

Of the 28 patients who discontinued bosutinib because of lack of efficacy, 25 switched to asciminib. Almost every patient who switched (96%) had previously received a BCL-ABL1 >10% before switching. None of the switch patients achieved an MMR at or by week 48 after switching. At week 48, 24% of patients achieved BCL-ABL1 ≤10% and 8% achieved BCL-ABL1 ≤1%.

At the end of study treatment cutoff date, there were 77 (49.4%) and 8 (10.5%) patients who were still receiving asciminib and bosutinib, respectively. Patients who experienced beneficial activity after the end of the study, after investigator assessment, continued receiving the treatment posttrial. 

Common reasons for discontinuation of treatments were because of the lack of efficacy in 40 patients (25.5%) receiving asciminib and 28 patients (36.8%) receiving bosutinib.

Although asciminib had a longer median duration of exposure (156.0 [0.1-256.3] weeks) compared with bosutinib (30.5 [1.0-239.3] weeks) the safety profile and tolerability of asciminib demonstrated better results vs bosutinib, remaining consistent with prior analyses.

Two patients in the study discontinued treatments because of adverse events (AEs) following the week 96 cutoff. One patient who had received asciminib reported pregnancy and the other patient who had received bosutinib reported diarrhea. Rates of discontinuation were lower for asciminib (8.3%) vs bosutinib (27.6%).

The most common (≥10%) grade ≥3 AEs for asciminib vs bosutinib were thrombocytopenia (22.4% vs 9.2%), neutropenia (18.6% vs 14.5%), diarrhea (0% vs 10.5%), and increased alanine aminotransferase (0.6% vs 14.5%). Most of the AEs occurred within the first 6 months of treatment.

“During the survival follow-up, 6 additional deaths occurred with asciminib due to CML in 3 patients and hemorrhagic stroke, multiple organ dysfunction syndrome, and COVID-19 and 1 patient each. And 3 with bosutinib treatment, 1 due to CML, 1 respiratory distress, and 1 due to COVID-19,” Mauro reported.

The safety profile for ascliminib in patients who switched treatments remained consistent when compared with patients receiving ascliminib during the randomized period. The most frequent (≥10%) grade ≥3 AEs were neutropenia (32.0%) and thrombocytopenia (24.0%). Eight percent of switch patients experienced AEs and discontinued treatment.

Since the week 96 cutoff, exposure-adjusted incidence rates of arterial occlusive events (AOEs) with asciminib decreased from 3.0 to 2.2 per 100 patient years, and no new AOEs occurred with the respective drug, which emphasized that the risk of AOEs did not increase over time. No new mutations occurred for patients since the week 96 cutoff who discontinued treatment due to lack of efficacy or disease progression.

“By the end of study, newly emerging mutations did not change since week 96, occurring in 7.6% of patients with asciminib and 2.6% with bosutinib,” Mauro noted.

References

  1. Mauro M, Minami Y, Hochhaus A, et al. Sustained efficacy and safety with asciminib after almost 4 years of median follow-up from Ascembl, a phase 3 study of ASC vs bosutinib in patients with chronic myeloid leukemia in chronic phase after ≥2 prior tyrosine kinase inhibitors: an end of study treatment update, including results from switch population. Blood. 2023; 142(1): 4536. doi: https://doi.org/10.1182/blood-2023-186854