Introduction

Adenosine triphosphate (ATP)-competitive tyrosine kinase inhibitors (TKIs; first-generation imatinib; second-generation nilotinib, dasatinib, and bosutinib; and third-generation ponatinib) targeting the BCR::ABL1 ATP-binding site have extended life expectancy of patients with chronic phase chronic myeloid leukemia (CML-CP) to near that of the general population [1, 2]. However, up to 25% and 24% of patients in clinical trials and real-world studies (across all lines of therapy with varying follow-up), respectively, discontinue ATP-competitive TKIs due to lack of efficacy (i.e., resistance) [3,4,5,6,7,8,9,10,11,12,13]. Resistance to ATP-competitive TKIs can increase disease progression risk and may be conferred by treatment-emergent mutations (e.g., T315I) [1, 14, 15].

Multiple lines of therapy, particularly when required due to TKI resistance, are associated with lower probabilities of response, higher disease progression risk, and decreased survival [16,17,18]. Overall survival rates at 8 years were significantly decreased for patients receiving ≥3 TKIs (22%) versus those remaining on imatinib as their first therapy (83%) [17]. Use of a second-generation TKI after failure with a prior second-generation TKI may have limited clinical benefit [19]. Ponatinib use may be limited by its safety profile and patients’ comorbidities [1, 11, 20, 21]. Sequential use of TKIs can result in resistant mutations, including T315I [22]. Different treatment strategies offering stronger efficacy are needed for patients with ATP-competitive TKI resistance.

Asciminib, the first BCR::ABL1 inhibitor that Specifically Targets the ABL Myristoyl Pocket (STAMP) [23,24,25], is indicated for adults with previously treated Philadelphia chromosome-positive CML-CP in >60 countries and regions and in some countries for patients with the T315I mutation [26,27,28,29,30,31]. Asciminib recently received accelerated approval for the treatment of adults with newly diagnosed Philadelphia chromosome-positive CML-CP [26]. Asciminib maintains activity against most BCR::ABL1 ATP-binding site mutations [32], including T315I, and has improved specificity versus ATP-competitive TKIs [23,24,25]. This novel mechanism of action allows asciminib to be used in combination with ATP-competitive TKIs to overcome drug resistance, offering new treatment options for patients with CML not responding to ATP-competitive TKI monotherapy. Using agents with distinct mechanisms of action may synergistically inhibit BCR::ABL1 [33, 34]. Studies of asciminib combined with ATP-competitive TKIs showed suppression of resistant outgrowth in BCR::ABL1 mutant cell lines [33, 34], and asciminib combined with ponatinib demonstrated effectiveness against compound mutations at clinically relevant concentrations [33].

Asciminib has shown rapid, durable molecular responses with favorable safety and tolerability [27,28,29]. The first report of the phase 1 trial (median follow-up, ≈14 months) established asciminib’s favorable safety and tolerability profile in patients with CML-CP or accelerated phase (AP) without T315I with ≥2 prior TKIs or with T315I with ≥1 prior TKI [27]. Asciminib continued to demonstrate favorable efficacy, safety, and tolerability in patients with CML-CP without T315I (4-year follow-up) [28] and with T315I (2-year follow-up) [35]. The phase 3 ASCEMBL trial demonstrated asciminib’s superior efficacy versus bosutinib in patients with CML-CP with ≥2 prior TKIs [29, 36]. In the phase 3 ASC4FIRST trial, asciminib demonstrated superior efficacy and favorable safety and tolerability versus standard-of-care TKIs in frontline patients with CML-CP [37].

Preliminary analyses from the phase 1 study of asciminib in combination with nilotinib (ASC + NIL), imatinib (ASC + IMA), or dasatinib (ASC + DAS) (median follow-up, ≤88.3 weeks; cutoff: July 15, 2018) showed durable responses and adequate tolerability in patients without T315I with ≥2 prior TKIs or with T315I with ≥1 prior TKI [38, 39]. We report final results after ≈4.7 years' additional follow-up in these combination arms.

Methods

Study oversight

The study was designed collaboratively by the sponsor (Novartis Pharma AG) and study investigators. The protocol was approved by the sites’ institutional review boards or independent ethics committees (see supplementary appendix) and conducted in accordance with the Declaration of Helsinki and Good Clinical Practice. All patients provided written informed consent. The sponsor collected and analyzed the data. The sponsor and the authors interpreted the data. All authors contributed to the development and writing of the manuscript. All authors and representatives of the sponsor reviewed and amended the manuscript and vouch for the accuracy and completeness of the data and fidelity of the study to the protocol.

Study design

Methods were previously described [27, 28]. Briefly, this analysis focuses on the combination arms of the phase 1, multicenter, open-label study of asciminib monotherapy and combination therapy with imatinib, nilotinib, or dasatinib (NCT02081378) [40]. Monotherapy results are published separately [27, 28, 35, 41, 42]. Adults (aged ≥18 years) with Philadelphia chromosome-positive CML-CP/AP with an Eastern Cooperative Oncology Group performance status of 0–2, with ≥2 prior TKIs, or with T315I with ≥1 prior TKI, were eligible.

Patients were assigned to study arms by the sponsor. Doses administered were based on available asciminib monotherapy dose-escalation data and the anticipated risk when combined with nilotinib, imatinib, or dasatinib. The provisional doses of combination agents satisfied the escalation-with-overdose-control criteria [27]. An adaptive Bayesian logistic regression model (5 parameters) guided by the escalation-with-overdose-control principle was used to make dose recommendations and estimate the maximum tolerated dose (MTD) and recommended dose for expansion (RDE); separate Bayesian logistic regression models were used for each combination arm. Patients in the ASC + NIL arm received standard-dose nilotinib (300 mg twice daily [BID]) [21] plus asciminib 20 mg or 40 mg BID. Based on the asciminib monotherapy dose-escalation data, the provisional starting dose of asciminib was 20 mg BID; however, asciminib 40 mg BID starting dose was selected when the ASC + NIL cohort was opened, satisfying escalation-with-overdose-control criteria. In the ASC + IMA arm, patients received standard-dose imatinib (400 mg once daily [QD]) [21] plus asciminib 40 mg BID or 40, 60, or 80 mg QD. Patients in the ASC + DAS arm received standard-dose dasatinib (100 mg QD) [21] plus asciminib 40 mg BID or 80 or 160 mg QD. Dosing was administered in continuous 28-day cycles. Patients who experienced adverse events (AEs) resulting in treatment discontinuation could continue with asciminib monotherapy if the AEs were clearly not asciminib-related and patients were benefiting from treatment (Figure S1).

The primary objective was to determine the MTD and/or RDE via the incidence of dose-limiting toxicities during the first treatment cycle. Secondary objectives included safety, tolerability, preliminary efficacy, and pharmacokinetics (PK). End-of-treatment period was declared when all patients enrolled were followed for ≥64 weeks or discontinued from treatment, whichever occurred first, and had post-trial access options available.

Study assessments

Assessments were previously described [27, 28].

Statistical analyses

MTD and RDE were previously described [27]. Data for this report were based on an end-of-study (EOS) cutoff date of March 14, 2023. Safety and efficacy analyses included patients who received ≥1 dose of study treatment. The dose-determining analysis set included patients from the safety set who either received ≥75% of the planned doses of treatment in the first cycle and had sufficient safety evaluations during the first cycle of dosing or discontinued earlier due to dose-limiting toxicities.

Results

Patients

This analysis included 26, 25, and 32 patients with CML-CP/AP in the ASC + NIL, ASC + IMA, and ASC + DAS arms, respectively. Patient demographics and clinical characteristics are summarized in Table 1. Data for each dose cohort (described in the methods) are in Table S1. Patients were heavily pretreated; 61.5%, 60.0%, and 46.9% of patients receiving ASC + NIL, ASC + IMA, and ASC + DAS, respectively, received ≥3 prior TKIs. Of patients receiving ASC + NIL, 69.2% previously received nilotinib; with ASC + IMA, 68.0% previously received imatinib; and with ASC + DAS, 56.3% previously received dasatinib. At screening, most patients had BCR::ABL1IS >1% (69.2%, 60.0%, and 59.4% receiving ASC + NIL, ASC + IMA, and ASC + DAS, respectively), while some were in MMR (15.4%, 12.0%, and 12.5% receiving ASC + NIL, ASC + IMA, and ASC + DAS, respectively). Of patients receiving ASC + NIL, ASC + IMA, and ASC + DAS, 26.9%, 16.0%, and 6.3%, respectively, were not in complete hematologic response. The T315I mutation was detected at screening in 2 patients (6.3%) receiving ASC + DAS.

Table 1 Baseline patient demographics and clinical characteristics.
Full size table

At data cutoff, ≈40–66% of patients in each arm continued to receive post-trial asciminib (Table 2; Table S2). Throughout the study, with ASC + NIL, ASC + IMA, and ASC + DAS, 10, 6, and 12 patients discontinued nilotinib, imatinib, and dasatinib, respectively, and remained on asciminib monotherapy. With ASC + NIL, ASC + IMA, and ASC + DAS, 12 (46.2%), 10 (40.0%), and 21 (65.6%) patients received treatment until EOS and continued to receive post-trial asciminib: 7, 7, and 10 patients in combination with the ATP-competitive TKI, and 5, 3, and 11 patients with asciminib monotherapy, respectively.

Table 2 Patient disposition at end of study.
Full size table

By EOS, the median duration of exposure (range) with ASC + NIL, ASC + IMA, and ASC + DAS, respectively, was 3.0 (0.2–7.4), 5.2 (0.5–6.6), and 2.8 (0.2–6.3) years for asciminib and 1.6 (0.0–7.2), 1.6 (0.0–6.6), and 1.7 (0.2–5.8) years for nilotinib, imatinib, and dasatinib, respectively. While all patients began treatment with nilotinib, imatinib, or dasatinib simultaneously with asciminib, the end dates for the ATP-competitive TKI and asciminib could have been different, as patients could discontinue their ATP-competitive TKI due to AEs and remain on asciminib monotherapy.

Safety

The safety set included 26, 25, and 32 patients receiving ASC + NIL, ASC + IMA, and ASC + DAS, respectively. All-grade AEs (≥30% of patients) are shown in Fig. 1 and Table S3. Grade ≥3 AEs reported in ≥15% of patients were thrombocytopenia (26.9%) and increased lipase (19.2%) with ASC + NIL, hypertension (16.0%) with ASC + IMA, and thrombocytopenia (18.8%) and increased lipase (15.6%) with ASC + DAS.

Fig. 1: Treatment-emergent all-grade AEs occurring in ≥ 30% of patients in any arm.a
figure 1

ASC asciminib, DAS dasatinib, IMA imatinib, NIL nilotinib. aAEs were counted if they occurred after treatment initiation through 30 days after the end of treatment.

Full size image

AEs led to dose adjustments or interruptions of asciminib in 17 (65.4%), 15 (60.0%), and 22 (68.8%) patients with ASC + NIL, ASC + IMA, and ASC + DAS, respectively. Most frequent AEs (≥10% of patients) requiring dose adjustment or interruption were thrombocytopenia (15.4%), lipase increase (11.5%), and amylase increase (11.5%) with ASC + NIL; nausea and pancreatitis (12.0% each) with ASC + IMA; and pleural effusion (15.6%), fatigue, lipase increase, and thrombocytopenia (12.5% each) with ASC + DAS.

AEs requiring additional therapy (concomitant medications treating AEs and cancer-related symptoms) occurred in 21 (80.8%), 24 (96.0%), and 30 (93.8%) patients with ASC + NIL, ASC + IMA, and ASC + DAS, respectively. AEs requiring additional therapy (≥10% of patients) were required in all 3 arms. With ASC + NIL, these included arthralgia (19.2%), anemia, constipation, pneumonia, pain in extremity, nausea (15.4% each), atrial fibrillation, upper abdominal pain, hyperuricemia, bone pain, sleep disorder, thrombocytopenia, and vomiting (11.5% each). With ASC + IMA, these included peripheral edema (20.0%), dry eye (16.0%), arthralgia, constipation, hypertension, influenza, pyrexia, rash, anemia, and upper respiratory tract infection (12.0% each). With ASC + DAS, these included hypertension, pleural effusion (21.9% each), nausea, upper respiratory tract infection (18.8% each), cough, pneumonia, pyrexia (15.6% each), anemia, COVID-19, diarrhea, and headache (12.5% each).

AEs led to treatment discontinuation in 3 (11.5%), 5 (20.0%), and 4 (12.5%) patients with ASC + NIL, ASC + IMA, and ASC + DAS, respectively (Table S4). With ASC + NIL, 1 patient experienced cardiac dysfunction; 1 patient experienced arteriosclerosis, extremity necrosis, and peripheral arterial occlusive disease; 1 patient experienced neutropenia and myelodysplastic syndrome. With ASC + IMA, 1 patient experienced leukocytosis and blast crisis, and 1 patient each experienced dysphagia, anemia, myopathy, and pruritus. With ASC + DAS, 1 patient each experienced pregnancy, lung adenocarcinoma, and peripheral ischemia; 1 patient experienced peripheral sensory neuropathy and fluid retention.

All-grade AEs of special interest (≥ 30% of patients in any arm) were gastrointestinal toxicity (61.5%, 84.0%, 65.6%), hypersensitivity (46.2%, 48.0%, 40.6%), myelosuppression (42.3%, 48.0%, 43.8%), hepatotoxicity (including laboratory terms; 34.6%, 36.0%, 18.8%), thrombocytopenia (30.8%, 32.0%, 37.5%), edema and fluid retention (26.9%, 28.0%, 43.8%), and pancreatic events (including isolated pancreatic enzyme elevations; 42.3%, 36.0%, 34.4%) with ASC + NIL, ASC + IMA, and ASC + DAS, respectively; 3.8%, 12.0%, and 0% of patients, respectively, experienced pancreatitis (clinical events) (Fig. 2; Table S5). Gastrointestinal and hypersensitivity events were managed with additional therapy or dose adjustment/interruption. While most myelosuppression events were managed by dose adjustment/interruption, 1 patient each with ASC + NIL and ASC + IMA discontinued due to decreased neutrophil count and anemia, respectively.

Fig. 2: AEs of special interest in ≥ 30% in any arm.a
figure 2

ASC asciminib, DAS dasatinib, GI gastrointestinal, IMA imatinib, NIL nilotinib. aA patient with multiple severity grades for an AE was only counted under the maximum grade. bMyelosuppression includes anemia, leukopenia, thrombocytopenia, and cytopenias affecting >1 lineage.

Full size image

All-grade arterial occlusive events (AOEs) were reported in 3 (11.5%), 1 (4.0%), and 1 (3.1%) patients with ASC + NIL, ASC + IMA, and ASC + DAS, respectively (Table 3). In the asciminib 40 mg BID plus nilotinib 300 mg BID cohort, the patient who experienced grade 3 angina pectoris, grade 3 peripheral arterial occlusive disease, grade 1 cerebral arteriosclerosis, and grade 1 cerebrovascular accident was a 76-year-old man previously treated with imatinib, nilotinib, dasatinib, and bosutinib; active, controlled medical conditions included coronary artery disease, hypertension, and cardiac failure. A 73-year-old woman with previous nilotinib, dasatinib, and ponatinib treatment in the asciminib 20 mg BID plus nilotinib 300 mg BID cohort experienced grade 1 angina pectoris; no relevant prior or active cardiovascular medical conditions were noted. In the asciminib 20 mg BID plus nilotinib 300 mg BID cohort, a 41-year-old man previously treated with dasatinib and imatinib experienced grade 2 peripheral arterial occlusive disease; no relevant prior cardiovascular medical history was noted; active, controlled conditions included hypertension. A 75-year-old man previously treated with dasatinib, nilotinib, and bosutinib in the asciminib 40 mg BID plus imatinib 400 mg QD cohort experienced grade 1 arterial embolism; no relevant prior cardiovascular medical history was noted; active, controlled conditions included arterial hypertension, aortic valve disease, mitral valve disease, and pulmonary valve disease. A 71-year-old man in the asciminib 40 mg BID plus dasatinib 100 mg QD cohort with prior imatinib, bosutinib, and nilotinib treatment experienced grade 2 cerebrovascular accident; relevant prior medical history included cerebrovascular disease and peripheral artery bypass; active, controlled conditions included angina pectoris, type 2 diabetes mellitus, arterial hypertension, coronary artery disease, and peripheral vascular disease (Table S6).

Table 3 Arterial occlusive eventsa.
Full size table

Dose-limiting toxicities were reported in 1 (6.3%), 6 (24.0%), and 2 (9.1%) patients with ASC + NIL, ASC + IMA, and ASC + DAS, respectively (Table 4; Table S7) [43]. With ASC + NIL, 1 patient in the asciminib 20 mg BID cohort experienced maculopapular rash. With ASC + IMA, 1 patient in the asciminib 40 mg QD cohort experienced neutropenia, 1 patient each in the asciminib 60 mg QD cohort experienced abdominal pain and nausea, 1 patient each in the asciminib 80 mg QD cohort experienced increased lipase and pancreatitis, and 1 patient in the asciminib 40 mg BID cohort experienced pancreatitis. With ASC + DAS, 1 patient in the asciminib 40 mg BID cohort experienced increased lipase and 1 patient in the asciminib 160 mg QD cohort experienced thrombocytopenia.

Table 4 DLTs in patients eligible for dose-determining analysis set.
Full size table

No on-treatment (occurring during treatment or ≤30 days after last treatment) deaths were reported; 1 patient in the asciminib 60 mg QD plus imatinib 400 mg QD cohort died due to pneumonia and 1 patient in the asciminib 40 mg BID plus nilotinib 300 mg BID cohort died due to leukemia during the safety follow-up (death occurring >30 days after treatment discontinuation).

Overall, median dose intensity was similar to the starting doses in each treatment arm (Table 5).

Table 5 Dose intensity.
Full size table

Pharmacokinetics

PK assessment showed a moderate increase in exposure of asciminib plus imatinib or nilotinib. Dasatinib plus asciminib 80 mg QD showed a moderate increase in exposure of asciminib, whereas dasatinib plus asciminib 40 mg BID had no effect on asciminib’s PK (Table S8). The MTD for ASC + IMA was reached at asciminib 60 mg QD plus imatinib 400 mg QD; MTD for ASC + NIL and ASC + DAS was not reached. Based on the safety, tolerability, PK, and preliminary efficacy data observed, the following were the RDE: asciminib 40 or 60 mg QD plus imatinib 400 mg QD, asciminib 40 mg BID plus nilotinib 300 mg BID, and asciminib 80 mg QD plus dasatinib 100 mg QD.

Efficacy

Of 26, 25, and 32 patients receiving ASC + NIL, ASC + IMA, and ASC + DAS, respectively, 0, 2, and 2 patients were not evaluable for MMR due to atypical/unknown transcripts at screening; 4, 3, and 4, respectively, were excluded due to being in MMR at baseline. Overall (by any time point during the study), 8/22, 11/20, and 15/26 patients achieved MMR with ASC + NIL, ASC + IMA, and ASC + DAS, respectively. Responses were achieved rapidly: median time to first MMR was 20.1, 20.9, and 22.1 weeks with ASC + NIL, ASC + IMA, and ASC + DAS, respectively. By week 96, 31.8%, 45.0%, and 46.2% of MMR-evaluable patients with ASC + NIL, ASC + IMA, and ASC + DAS, respectively, achieved MMR. By week 432, 36.4%, 55.0%, and 57.7% of patients with ASC + NIL, ASC + IMA, and ASC + DAS, respectively, achieved MMR (Fig. 3).

Fig. 3: MMR by time point in patients not in MMR at screening.a
figure 3

ASC asciminib, DAS dasatinib, IMA imatinib, MMR major molecular response (BCR::ABL1IS ≤ 0.1%), NIL nilotinib. aExcludes patients with atypical transcripts and those in MMR at baseline.

Full size image

With ASC + NIL, ASC + IMA, and ASC + DAS, among 8/22, 11/20, and 15/26 patients who achieved MMR, 7, 7, and 13 patients, respectively, maintained or improved this response to a deeper level up to the cutoff. By week 444, 61.5%, 69.6%, and 80.0% of patients with ASC + NIL, ASC + IMA, and ASC + DAS, respectively, achieved BCR::ABL1IS ≤1% (Fig. 4).

Fig. 4: BCR::ABL1IS ≤ 1% by time point.a
figure 4

ASC asciminib, DAS dasatinib, IMA imatinib, NIL nilotinib. aExcludes patients with atypical/p190 (e1a2 or e1a3)/unknown transcripts and those with a missing evaluation at screening.

Full size image

Discussion

This study provides evidence of asciminib plus nilotinib, imatinib, or dasatinib as potential treatment strategies that may help some patients with limited therapeutic options after several lines of therapy achieve their goals (e.g. durability of response, deeper molecular response, and preventing resistance emergence) [18, 44, 45]. Combination therapy in patients with CML-CP/AP with ≥2 prior TKIs (n = 81) or with T315I with ≥1 prior TKI (n = 2) demonstrated rapid responses with adequate tolerability.

MMR, a well-established treatment goal [1, 21], can be associated with improved outcomes, including overall survival and progression-free survival [46, 47]. Within separate study comparison limits, MMR rates by week 96 with ASC + IMA (45.0%) and ASC + DAS (46.2%) were similar to those observed with asciminib monotherapy (42.7%, ASCEMBL study) with relatively similar durations of exposure ( ≈ 2.0–3.0 years) [36]. However, patients receiving combination treatment with asciminib plus nilotinib or imatinib were more heavily pretreated [36]. These data suggest that the efficacy of asciminib in combination with ATP-competitive TKIs is comparable with that of asciminib monotherapy.

ASC + NIL had the lowest MMR rate and the highest incidence of AOEs. While no hypothesis testing was done and comparisons were not powered to detect significant differences between arms, these results suggest that ASC + IMA or ASC + DAS may be safer combinations providing higher rates of responses.

In this analysis, 66–77% of patients experienced grade ≥3 AEs. Lower rates occurred with asciminib monotherapy in ASCEMBL (56.4%) with similar duration of exposure [36]. By EOS, 9.4–16% of patients discontinued treatment due to AEs. Rates were lower with asciminib monotherapy in ASCEMBL in patients with ≥2 prior TKIs (7.0%) [36], suggesting that combination therapy in pretreated patients with CML-CP may be associated with slightly higher AE burden versus asciminib monotherapy [27, 28, 36]. It remains to be seen if the possible increased benefit justifies the minor increase in AEs and treatment discontinuation. Future studies should compare monotherapy to combination therapy and define patient populations where combination therapy is preferable, such as those with limited therapeutic options due to tolerability or resistance.

Unique AEs are associated with ATP-competitive TKIs [48, 49]. Notably, nilotinib is associated with increased risk of cardiovascular events [48]. AOEs were higher with ASC + NIL than with asciminib monotherapy (11.5% with ASC + NIL vs 5.1% and 8.7% with asciminib in the 2.3-year ASCEMBL follow-up and 4-year phase 1 follow-up, respectively) [28, 36]. Accordingly, it is important to evaluate a patient’s risk factors for AOE development when considering treatment with ASC + NIL.

Although imatinib is generally well tolerated, it is associated with gastrointestinal disturbances (including nausea, vomiting, and diarrhea) [49, 50]. Grade ≥3 nausea has been reported in ≈1–3% of newly diagnosed patients with CML receiving imatinib [51, 52]. Grade ≥3 nausea was reported at 8.0% with ASC + IMA. With asciminib monotherapy, grade ≥3 nausea was experienced by 0.6% of patients in the 2.3-year ASCEMBL follow-up and 1.7% of patients in the phase 1 4-year follow-up [28, 36].

Pleural effusion is commonly associated with dasatinib treatment [49]. After 6 years’ follow-up in the phase 3 CA180-034 study (dasatinib in imatinib-resistant/intolerant patients), 5.3% of patients experienced grade ≥3 pleural effusion and 1.8% discontinued treatment due to grade ≥3 pleural effusion [53]. At 5 years' follow-up in the DASISION study (dasatinib vs imatinib in newly diagnosed patients), 3% of patients with dasatinib experienced grade 3/4 drug-related pleural effusion and 6% discontinued dasatinib due to pleural effusion [7]. Higher rates were seen with ASC + DAS, with 4 patients (12.5%) experiencing grade ≥3 pleural effusion. Three of the 4 patients with grade ≥3 pleural effusion permanently discontinued dasatinib and continued asciminib; 2 of these 3 patients had pleural effusion resolved by EOS. The fourth patient (in the asciminib 160 mg QD plus dasatinib 100 mg QD cohort) temporarily interrupted dasatinib treatment until the pleural effusion resolved to grade 1; dasatinib was restarted at a reduced dose of 70 mg QD and the pleural effusion was ongoing at cutoff. No pleural effusion events with asciminib were reported in the ASCEMBL 2.3-year follow-up or the phase 1 4-year follow-up [28, 36].

Myelosuppression events are often experienced by patients with CML receiving TKIs, sometimes resulting in treatment discontinuation [28, 36, 49]. Thrombocytopenia and neutropenia are hematologic toxicities associated with asciminib therapy [26, 28, 36]. Here, grade ≥3 thrombocytopenia and neutropenia were reported in 26.9%, 12.0%, and 18.8% and 15.4%, 16.0%, and 9.4% of patients receiving ASC + NIL, ASC + IMA, and ASC + DAS, respectively. In the 2.3-year ASCEMBL follow-up, 22.4% and 18.6% of patients experienced grade ≥3 thrombocytopenia and neutropenia with asciminib monotherapy; in the phase 1 4-year follow-up, 13.9% and 12.2% of patients experienced thrombocytopenia and neutropenia, respectively [28, 36]. Discontinuation due to thrombocytopenia and neutropenia events here (0% and 0–3.8%, respectively) was similar to that with asciminib monotherapy (3.2% and 2.6%, respectively, in the ASCEMBL 2.3-year follow-up; 1.7% and 0%, respectively, in the phase 1 4-year follow-up), and most events were managed by dose adjustment/interruption, or additional therapy [28, 36].

Elevated lipase and clinical pancreatitis are important to consider with ATP-competitive TKIs and asciminib [28, 36, 49, 50]. Levels of elevated lipase were relatively similar between the combination arms (38.5%, 31.0%, and 25.0% with ASC + NIL, ASC + IMA, and ASC + DAS, respectively) and asciminib monotherapy in the phase 1 4-year follow-up (37.4%) [28]; in the ASCEMBL 2.3-year follow-up, 5.1% of patients receiving asciminib experienced elevated lipase [36]. While all-grade clinical pancreatitis was slightly lower with ASC + NIL (3.8%) and ASC + DAS (0%) than with asciminib monotherapy in the phase 1 4-year follow-up (7.0%) [28], more patients with ASC + IMA experienced clinical pancreatitis (12.0%). Levels of clinical pancreatitis were not reported in the ASCEMBL 2.3-year follow-up [36].

Taken together, these data suggest patients treated with asciminib plus conventional TKIs may experience the unique AEs of ATP-competitive TKI monotherapy. Therefore, patient comorbidities and risk factors should be considered when selecting combination regimens. As asciminib’s safety and tolerability profile has been established at 80 mg daily [26, 36, 37], and to decrease the risk of developing ATP-competitive TKI-associated AEs, patients may benefit from combination therapy with the approved dose of asciminib and reduced doses of ATP-competitive TKIs. Combination strategies with asciminib may reduce the emergence of resistance mutations, but larger studies will need to be conducted.

In this heavily pretreated population, ≈40–66% of patients with combination therapy were receiving combination treatment at EOS; only 9–16% discontinued due to AEs. Treatment discontinuation due to progressive disease occurred in 1 patient each with ASC + NIL and ASC + DAS and in 2 patients with ASC + IMA. Therefore, patients with multiple prior treatments may benefit from the rapid responses afforded by combination therapy. However, the benefits and risks of combination therapy will need to be carefully considered for patients at higher risk of developing AEs [54]. Exploring reduced standard dose (escalating as necessary) of ATP-competitive TKIs in combination with asciminib 80 mg daily could be a safer approach.

Preclinical data suggest that asciminib plus ponatinib combination therapy may be more effective in patients with compound mutations (including T315I) than asciminib with other ATP-competitive TKIs, including nilotinib [33]. Asciminib paired with ponatinib could be a treatment option for patients with the T315I mutation [33], but its effectiveness and safety remain to be determined.

The MTD was reached only with ASC + IMA (asciminib 60 mg QD plus imatinib 400 mg QD). Based on the safety, tolerability, PK, and preliminary efficacy data, the RDE for each arm was as follows: asciminib 40 mg BID plus nilotinib 300 mg BID; asciminib 40 or 60 mg QD plus imatinib 400 mg QD; and asciminib 80 mg QD plus dasatinib 100 mg QD. The ASC + IMA dose combinations are being evaluated as recommended phase 2 doses in ASC4MORE (NCT03578367), a phase 2 study of asciminib add-on to imatinib, versus continued imatinib, versus switch to nilotinib in patients with CML-CP without deep molecular response with ≥1 year of imatinib as their first TKI [55, 56]. At week 96 in ASC4MORE, more patients with asciminib add-on to imatinib achieved BCR::ABL1IS ≤ 0.0032% than with continued imatinib or switch to nilotinib [56]. At early time points, more patients with asciminib add-on to imatinib achieved MMR than with continued imatinib or switch to nilotinib [57]. No AOEs were reported with asciminib add-on, which may be safer than switching to nilotinib [56].

Patients with resistance to earlier lines of treatment (with imatinib or second-generation TKIs) may have limited subsequent treatment options, particularly if they have contraindicated mutations, comorbidities, or specific toxicity profiles [1, 18, 45, 58, 59]. These patients may benefit from asciminib plus imatinib or dasatinib to achieve treatment goals (durability of response, deeper molecular response to increase the likelihood of treatment-free remission eligibility, and preventing emergence of resistance) [1, 18, 21, 44, 45]. This population of patients may respond to combination therapy offering stronger efficacy versus monotherapy, thus preventing TKI switching, which may result in treatment-resistant mutations [22, 60]. The results presented in this final analysis of the combination arms of the phase 1 study, supported by ASC4MORE results, provide evidence for asciminib combination therapy as a potential strategy that may help patients with limited therapeutic options achieve their treatment goals.