In this issue of Blood Advances, Mauro et al1 report the 4-year results of the ASCEMBL trial, which compared asciminib vs bosutinib for patients with chronic myeloid leukemia in chronic phase (CP-CML) after ≥2 previous tyrosine kinase inhibitors (TKIs). The primary end point of ASCEMBL was the rate of major molecular response (MMR) at 24 weeks, and a key secondary end point was the rate of MMR at 96 weeks. This study met these end points, with more asciminib-treated patients achieving this response at both time points (25.5% vs 13.2%, and 37.6% vs 15.8%, respectively).2,3 The earlier results from this study also report fewer grade ≥3 adverse events (AEs) and AEs leading to treatment discontinuation with asciminib compared with bosutinib.2,3 This article reports the end of study (EOS) analysis after a median of 3.7 years on treatment and 5 years from the time the last patient was enrolled. Crossover was allowed for bosutinib-treated patients who met the European Leukemia Network 2013 definition of treatment failure at ≤96 weeks after the last patient was randomized.
This study enrolled 233 patients with a 2:1 randomization to asciminib vs bosutinib. At the EOS, 36.5% completed protocol-driven treatment. The most common reasons for treatment discontinuation were lack of efficacy (25.5% with asciminib, 36.8% with bosutinib) and AEs (7% with asciminib, 27.6% with bosutinib). The increased MMR rate with asciminib vs bosutinib was maintained at the 156-week analysis (33.8% vs 10.5%, respectively). When looking specifically at patients with baseline International Standardization (IS) of BCR::ABL1IS of >1%, 43% of asciminib-treated patients and 11.1% of bosutinib-treated patients had achieved BCR::ABL1IS of ≤1% at 156 weeks. The durability of these responses was similar with both drugs. Importantly, no significant difference in progression-free survival or overall survival was seen between the 2 groups, although this included 25 patients who switched from bosutinib to asciminib. Notably, of the patients who discontinued asciminib owing to lack of efficacy, 38.1% had acquired new BCR::ABL1 mutations.
Responses in the 25 patients who switched from bosutinib to asciminib due to lack of bosutinib efficacy were quite poor. It is unclear whether these patients were enriched by poor genomic features such as ASXL1 mutations. Nearly half of these patients either had mutations at the time of switching or acquired mutations while on asciminib. The M244V mutation was noted in 25% of these patients, which has recently been reported as resistant to asciminib.4 At EOS, 12% of patients had BCR::ABL1IS of ≤1%, and 8% had achieved MMR. These were heavily pretreated patients, with 84% having received ≥4 previous TKIs before asciminib, thus suggesting the benefit of asciminib is more notable when used in earlier lines of therapy.
Perhaps the most commonly discussed benefit of asciminib is the tolerability compared with the adenosine triphosphate-binding site TKI. Mauro et al have continued to show a lower incidence of grade ≥3 AEs with asciminib, along with fewer patients requiring ≥1 dose interruption or reduction because of AEs. Given the known cardiac toxicity with nilotinib and ponatinib, the possibility of late cardiovascular AEs is often brought up as a question with asciminib.5-7 This report suggests that the frequency of arterial occlusive events was 5.1% with asciminib, and 1.3% with bosutinib. The exposure-adjusted incident rate of arterial occlusive events was 2.2 per 100 patient-years with asciminib, and 1.2 with bosutinib. In spite of this, the ASCEMBL trial did not report any new safety signals in the longer-term follow-up, suggesting most AEs with asciminib occur early in the treatment period, and the safety profile of the drug is relatively well established at this time. However, further follow-up on all asciminib trials should clarify whether there is, indeed, a cardiovascular signal in the long-term follow-up, similar to what took place with nilotinib and ponatinib.
The most prominent question that was not answered by the ASCEMBL trial is how asciminib compares with ponatinib in heavily pretreated patients. Although the authors touch on this topic in their discussion, it is crucial to point out that this is a cross-trial comparison of different patient populations. Given the MMR rates of 24% to 45% in the PACE and OPTIC trials, it remains unclear whether asciminib would out-perform ponatinib in a head-to-head trial.8,9 From a treatment decision-making perspective, this question remains very clinically relevant, and unfortunately we are left with cross-trial comparisons and retrospective studies using matched-adjusted indirect comparisons to piece together the answer. Furthermore, the results of these analyses have not been consistent and are likely biased by the sponsor of the 2 analyses, further highlighting the importance of a head-to-head comparison.10,11
Although the response to asciminib after ponatinib failure seen in this trial is low (3/12 [25%]) it is unclear what the outcomes of ponatinib after asciminib failure would be. Possibly a later analysis of this trial could address this question as some patients may have received ponatinib after coming off the study. A small real-world evidence study suggests limited efficacy in this setting, similar to what is seen with asciminib after ponatinib.12
With regard to mutations, it is interesting to see the acquisition of ABL mutations outside the myristoyl pocket while on asciminib. The genomic analysis of these mutations may be important for selecting the next drug in resistant CML.13
Finally, another remaining question surrounds appropriate treatment for patients who have failed asciminib and ponatinib. In this regard, a recent study suggests that olverembatinib may be an effective option.14
Despite these significant, unanswered questions, the data reported by Mauro et al support the use of asciminib in patients with CP-CML who have received ≥2 previous lines of treatment. The safety and tolerability of asciminib sets it apart from other TKIs, and it is clear that asciminib has a well-established place in the treatment armamentarium for CML. As other asciminib trials have shown, the efficacy of the drug is superior even in earlier lines of treatment, and it is likely becoming the drug with which all future CML-directed therapies will need to be compared.
Conflict-of-interest disclosure: The authors declare no competing financial interests.
