FDA Approval Summary: Asciminib for Ph+ CML in Chronic Phase Treated with Two or More Tyrosine Kinase Inhibitors and for the T315I Mutation

Abstract

On October 29, 2021, FDA granted accelerated approval to asciminib (Scemblix; Novartis), a tyrosine kinase inhibitor (TKI), for the treatment of adult patients with Philadelphia chromosome positive chronic myeloid leukemia (Ph+ CML) in chronic phase (CP), previously treated with two or more TKIs, and granted traditional approval to asciminib for adult patients with Ph+ CML in CP with the T315I mutation. The first indication was approved based on major molecular response (MMR) at 24 weeks in the ASCEMBL Study, a randomized trial comparing asciminib with bosutinib in patients who had failed two or more TKIs. This indication was ultimately granted traditional approval on October 12, 2022, based on safety data and MMR rate at 96 weeks of 38% (95% CI: 30, 46) in the asciminib arm vs. 16% (95% CI: 8, 26) in the bosutinib arm (p-value: 0.001). The second indication was approved based on MMR rate by 96 weeks of 49% (95% CI: 34, 64) in the single-arm CABL001X2101 Study. The most common (≥20%) adverse reactions included upper respiratory tract infections, musculoskeletal pain, headache, fatigue, nausea, rash, and diarrhea. The most common (≥20%) laboratory abnormalities were thrombocytopenia, neutropenia, anemia, lymphopenia, hypertriglyceridemia, hyperuricemia, increases in creatine kinase, ALT, AST, lipase, and amylase. This manuscript describes the basis for approval of these indications.

INTRODUCTION

Chronic myeloid leukemia (CML) is characterized by the Philadelphia chromosome (Ph+), which is a balanced translocation between chromosomes 9 and 22 and leads to the BCR-ABL1 fusion gene and BCR-ABL1 oncoprotein (1, 2). At the time of submission of the new drug application for asciminib, there were five approved tyrosine kinase inhibitors (TKIs) for the treatment of chronic phase (CP)-CML: imatinib, dasatinib, nilotinib, and bosutinib in patients with CP-CML that is newly diagnosed or resistant or intolerant to prior therapy, and ponatinib for patients with CP-CML that is resistant or intolerant to at least 2 prior TKIs or with the T315I mutation (3, 4).

Treatment options beyond second-line therapy and for patients with T315I mutation are limited. Available treatments for third-line therapy are a TKI that patients have not previously received, omacetaxine, and allogeneic stem cell transplantation, all of which have specific toxicities (5). Ponatinib, the first TKI for patients with the T315I mutation, poses the risk of arterial occlusive events, venous thromboembolic events, and cardiac arrhythmias (6). There is an unmet need for alternative drugs in CML patients who need a third-line treatment or for those with the T315I mutation.

Herein, we provide a summary of FDA’s review of the marketing applications that led to approval of asciminib for the treatment of adult patients with Ph+ CP-CML, previously treated with two or more TKIs, and Ph+ CP-CML with the T315I mutation.

NONCLINICAL PHARMACOLOGY AND TOXICOLOGY

Asciminib is a TKI that inhibits the ABL1 kinase activity of the BCR-ABL1 fusion protein. It differs from the other TKIs that target the ATP binding site of ABL1, as asciminib targets the myristoyl pocket. Normally, a myristoyl group binds to the myristoyl pocket to induce an inactive state. However, in CML this autoregulation is lost because of fusion with BCR which renders the ABL1 kinase constitutively active. Asciminib also has activity against mutant forms of the kinase including the T315I mutation (4).

Asciminib was evaluated in general toxicology studies in rats, dogs, and monkeys and the studies identified toxicities in the hematopoietic system, liver, adrenal gland, gastro-intestinal tract, and pancreas (dog only). Asciminib treatment consistently suppressed erythroid parameters including red blood cell count, hemoglobin, and hematocrit, and the suppression was associated with regenerative responses. Increased white blood cell counts were likely related to multi-organ inflammation. The nonclinical pancreatic and hepatobiliary toxicities supported the adverse effects observed in the patients.

Asciminib was not genotoxic in an in vitro bacterial mutagenicity (Ames) assay, and not aneugenic or clastogenic in the in vitro or in vivo micronucleus assays. In animal reproduction studies, administration of asciminib to pregnant rats and rabbits during the period of organogenesis caused adverse developmental outcomes including embryo-fetal mortality and malformations at maternal exposures equivalent to or less than those in patients at the recommended doses. In a combined male and female fertility and early embryonic development study in rats, asciminib treatment was associated with decreased mean spermatozoa count (by 10%) and percent motility (by 10%).

CLINICAL PHARMACOLOGY

FDA reviewed subject-level data collected from healthy participants and patients with cancer to characterize the pharmacokinetics (PK), pharmacodynamics, food effect, hepatic and renal impairment, and drug–drug interactions of asciminib.

Dose Selection

The dose selection of asciminib at 40 mg twice daily (BID) or 80 mg once daily (QD) for patients with Ph+ CP-CML and 200 mg BID for patients with Ph+ CP-CML harboring the T315I mutation was supported by nonclinical results, clinical data from Study CABL001A2301 (ASCEMBL), and CABL001X2101 and pharmacometrics analyses (7–11). Exposure-response (E-R) analyses for efficacy showed that a lower exposure of asciminib was associated with a smaller decrease in the BCR-ABL1 level and a lower major molecular response (MMR) rate at Week 24 over the asciminib dosages of 10 mg to 200 mg BID. E-R analyses for safety showed that a higher asciminib exposure was associated with slightly higher incidences of certain adverse reactions (AR) (e.g., Grade ≥3 lipase increase, Grade ≥3 hemoglobin decrease, Grade ≥2 ALT increase, Grade ≥2 AST increase, Grade ≥2 bilirubin increase, and any grade lipase increase) over the asciminib dosages of 10 mg to 280 mg BID. Pharmacometrics analysis showed comparable PK, efficacy, and safety between 40 mg BID and 80 mg QD asciminib dosages. There were no clinically meaningful differences in safety and efficacy between patients treated with asciminib 80 mg QD in X2101 trial and patients treated with asciminib 40 mg BID in A2301 trial. No food should be taken at least 2 hours before and 1 hour after administration of asciminib tablets since food significantly decreased the PK exposure of asciminib. No dose adjustment of asciminib is required for patients with mild to severe hepatic (NCI-ODWG criteria) or mild to severe renal impairment as the changes in asciminib PK were not considered clinically meaningful.

Pharmacokinetics

Asciminib steady-state exposures (AUC and C_max) increased at slightly more than dose proportionally across the dose range of 10 to 200 mg administered QD or BID.

Drug Interactions

Asciminib is subject to multiple cytochrome P-450 based drug-drug interactions. The product label recommends specific dose modification strategies for concomitant use of asciminib with other drug products (Supplementary Table S1).

ASSESSMENT OF EFFICACY

Clinical Trials Overview

The FDA review focused on two main clinical trials, ASCEMBL and CABL001X2101, in support of the approved indications.

ASCEMBL (NCT03106779) was a multicenter, randomized, open-label, active controlled phase 3 clinical trial in CP-CML patients previously treated with at least 2 prior TKIs. Patients were randomized 2:1 to asciminib 40 mg orally BID and bosutinib 500 mg orally QD and were treated until treatment failure or unacceptable toxicity. Randomization was stratified by cytogenetic response at baseline. For inclusion, patients had treatment failure as per 2013 ELN recommendations or were intolerant to their most recent TKI. Patients with T315I or V299L mutations were excluded as they were not expected to respond to bosutinib. The primary endpoint was the MMR rate at 24 weeks and the key secondary endpoint was the MMR rate at 96 weeks (7–9). A patient was considered to have achieved MMR at week 24 if meeting the MMR criterion (BCR-ABL1 ratio ≤0.1%) at week 24 while on study treatment, unless the patient met any treatment failure criteria prior to week 24.

CABL001X2101 (NCT02081378; hereafter called the X2101 trial) was a multicenter, open-label, first-in-human clinical trial of asciminib alone or in combination with other TKIs in CML and Ph+ ALL (10). For approval of the second indication, CP-CML patients with the T315I mutation treated with single-agent asciminib 200 mg orally BID were evaluated. Treatment continued until treatment failure or unacceptable toxicity. The main efficacy endpoint of interest was MMR by 24 weeks which was defined as the proportion of patients who achieved MMR at or before 24 weeks.

Demographics and Baseline Characteristics

ASCEMBL included 157 patients who received asciminib (40 mg BID) and 76 patients who received bosutinib (500 mg QD). The X2101 trial included 48 patients with CP-CML with the T315I mutation who received asciminib 200 mg BID; three of 48 patients had atypical/unknown transcripts and were not included in the efficacy analysis. Table 1 shows the demographics and baseline characteristics of the efficacy populations in the ASCEMBL and X2101 trials.

Table 1:

Demographics and baseline characteristics of patients in ASCEMBL and Study X2101.

	Study ASCEMBL (CABL001A2301)		Study CABL001X2101
	Asciminib 40 mg BID N=157 n (%)	Bosutinib 500 mg QD N=76 n (%)	Asciminib 200 mg BID N=45 n (%)
Age, median (range)	52 (24 – 83)	52 (19 – 77)	54 (26 – 86)
Gender
Male	82 (52)	31 (41)	36 (80)
Female	75 (48)	45 (59)	9 (20)
Race
White	118 (75)	56 (74)	21 (47)
Asian	22 (14)	11 (14)	12 (27)
Black or African American	8 (5)	2 (3)	1 (2)
Other	5 (3)	7 (9)	-
Unknown	3 (2)	0	11 (24)
American Indian or Alaska native	1 (0.6)	-	-
Number of lines of prior TKI therapy – n (%)
1	-	-	10 (22)
2	82 (52)	30 (39)	12 (27)
3	44 (28)	29 (38)	16 (36)
4	24 (15)	10 (13)	6 (13)
≥ 5	7 (4)	7 (9)	1 (2)
Reason to discontinue last TKI therapy – n (%)
Lack of efficacy	95 (61)	54 (71)	aNA
Lack of tolerability	59 (38)	22 (29)	aNA
Other	3 (1.9)	0	aNA
ECOG, 0-1	154 (98)	76 (100)	45 (100)

^aNA: This data was not captured.

Source: U.S. Food and Drug Administration.

In ASCEMBL, the median duration of treatment was 103 weeks (range: 0.1 to 201 weeks) with asciminib and 31 weeks (range: 1 to 188 weeks) with bosutinib. The median duration of follow-up for the asciminib arm was 28 months (range: 1 day to 45 months) and the median duration of response had not been reached for patients with MMR. The median duration of treatment in the X2101 trial was 108 weeks (range: 2 to 215 weeks).

Efficacy Evaluation

Table 2 shows a summary of the efficacy results from ASCEMBL. MMR at 24 weeks was significantly better on the asciminib arm at 25% (95% CI: 19, 33) compared to that in the bosutinib arm at 13% (95% CI: 6.5, 23). The treatment difference in the MMR rate at Week 24 was 12% (95% CI: 2.2, 22; two-sided p-value of 0.029, stratified by the MCyR status at baseline). At 96 weeks, MMR was significantly better on the asciminib arm at 38% (95% CI: 30, 46) compared to that in the bosutinib arm at 16% (95% CI: 8, 26). The treatment difference in the MMR rate at Week 96 was 22% (95% CI: 11, 33; two-sided p-value of 0.001).

Table 2:

Efficacy Results in Patients with Ph+ CP-CML, Previously Treated with Two or More TKIs (ASCEMBL)

	SCEMBLIX 40 mg twice daily	Bosutinib 500 mg once daily	Difference (95% CI)	p-value
MMR rate, % (95% CI) at 24 weeks	N = 157 25 (19, 33)	N = 76 13 (6.5, 23)	12a (2.2, 22)	0.029b
MMR rate, % (95% CI) at 96 weeks	N = 157 38 (30, 46)	N = 76 16 (8, 26)	22a (11, 33)	0.001b
CCyR rate, % (95% CI) at 24 weeks	N = 103c 41 (31, 51)	N = 62c 24 (14, 37)	17 (3.6, 31)
CCyR rate, % (95% CI) at 96 weeks	N = 103c 40 (30, 50)	N = 62c 16 (8, 28)	24a (10, 37)

Abbreviations: MMR, major molecular response (BCR-ABL1^IS ≤ 0.1%); CCyR, complete cytogenetic response (0% of Philadelphia-positive metaphases in bone marrow aspirate with at least 20 examined).

^aEstimated using a common risk difference stratified by baseline major cytogenetic response status.

^bEstimated using a Cochran-Mantel-Haenszel two-sided test stratified by baseline major cytogenetic response status.

^cCCyR analysis based on patients who were not in CCyR at baseline.

Source: U.S. Food and Drug Administration

The results from the X2101 trial showed that the MMR rate by 24 weeks and by 96 weeks was 42% (95% CI: 28, 58) and 49% (95% CI: 34 to 64), respectively, for the 45 patients with Ph+ CP-CML with the T315I mutation treated with asciminib.

ASSESSMENT OF SAFETY

For patients with Ph+ CP-CML previously treated with two or more TKIs, safety analysis results were based on the ASCEMBL trial. The primary safety population included patients who received asciminib 40 mg BID (n=156) or bosutinib 500 mg QD (n=76). One patient in the asciminib arm did not receive asciminib and was excluded from the safety set. Asciminib was permanently discontinued in 8% of patients, dosage was interrupted in 41%, and dosage was reduced in 6% due to an AR. Thrombocytopenia and neutropenia were the most frequent ARs leading to permanent discontinuation, dosage interruption, and dose reduction. Two deaths due to an AR occurred while on treatment with asciminib and were caused by arterial embolism and ischemic stroke. Common ARs were upper respiratory tract infection, musculoskeletal pain, headache, fatigue, rash, hypertension, arthralgia, diarrhea, nausea, abdominal pain (Table 3).

Table 3:

Treatment-emergent adverse reactions (≥ 10%) in the safety population^a in ASCEMBL.

	Asciminib N=156		Bosutinib N=76
Adverse Reactionb	All Grades (%)	Grade 3 or 4 (%)	All Grades (%)	Grade 3 or 4 (%)
Upper respiratory tract infection	26	0.6	12	1.3
Musculoskeletal pain	24	2.6	17	1.3
Headache	21	1.9	16	0
Fatigue	20	0.6	11	1.3
Rash	18	0.6	30	8
Hypertension	14	7	5	3.9
Arthralgia	13	0.6	3.9	0
Diarrhea	13	0	72	11
Nausea	12	0.6	46	0
Abdominal pain	14	0	24	2.6

^aN=232 adult Ph+ CP-CML patients treated with two or more TKIs

^bIncludes grouped terms. See Supplementary Table S2 for further information.

Source: U.S. Food and Drug Administration.

Hematologic laboratory abnormalities that were more frequent in the asciminib arm were decreased platelet count and decreased neutrophil count. Nonhematologic laboratory abnormalities that were more frequent included hypertriglyceridemia, elevated creatine kinase, hyperuricemia, elevated ALP, hyperbilirubinemia, hypercholesterolemia, and hypokalemia (Supplemental Table S3).

For patients with Ph+ CP-CML with the T315I mutation, safety analysis results were based on the X2101 trial. The primary safety population included patients who received asciminib 200 mg BID (n=48). Asciminib was permanently discontinued in 10% of patients, dosage was interrupted in 31%, and dosage was reduced in 23% due to an AR. An increase in pancreatic enzymes was the most frequent AR which led to permanent discontinuation (2.1%), dosage interruption (17%), and dose reduction (10%). Common ARs were musculoskeletal pain, fatigue, nausea, rash, diarrhea, vomiting, headache, arthralgia, abdominal pain, cough, hemorrhage, pruritus, hypertension, upper respiratory tract infection, and edema (Table 4).

Table 4:

Treatment-emergent adverse reactions (≥ 10%) in the safety population in the X2101 Study.

	Asciminib 200 mg twice daily N=48
Adverse Reactiona	All Grades (%)	Grade 3 or 4 (%)
Musculoskeletal pain	42	4.2
Fatigue	31	2.1
Nausea	27	0
Rash	27	0
Diarrhea	21	2.1
Vomiting	19	6
Headache	19	2.1
Arthralgia	17	0
Abdominal pain	17	8
Cough	15	0
Hemorrhage	15	2.1
Pruritus	13	0
Hypertension	13	8
Upper respiratory tract infection	13	0
Edema	10	4.2

^aIncludes grouped terms. See Supplementary Table S4 for further information.

Source: U.S. Food and Drug Administration.

Common hematologic laboratory abnormalities in the X2101 trial were decreased hemoglobin, decreased neutrophil count, decreased lymphocyte count, and decreased platelet count. Common nonhematologic laboratory abnormalities included elevated ALT and AST, hyperkalemia, hypertriglyceridemia, elevated lipase and amylase, hypophosphatemia, hyperuricemia, hypocalcemia, elevated creatinine, hyperbilirubinemia, hypercholesterolemia, and elevated ALP (Supplemental Table S5).

The Warnings and Precautions section of the Prescribing Information (PI) was based on safety data from the Asciminib All Patients Safety Pool (N=356), which consists of patients treated with asciminib in ASCEMBL (N=156) and patients with CP-/accelerated phase CML treated with asciminib in X2101 (N=200). Dosages in the latter trial ranged from 10 to 200 mg BID and 80 to 200 mg QD. Myelosuppression, especially thrombocytopenia and neutropenia, were common (≥20%). Pancreatic toxicity manifested as increases in lipase and/or amylase levels or as pancreatitis. Hypertension was diagnosed in 19% of patients. Hypersensitivity of mostly grades 1-2 was seen in 32% of patients and it usually manifested as a rash. Asciminib may lead to embryo-fetal toxicity and females of reproductive potential should use effective contraception.

The cardiovascular toxicity section of the Warnings & Precautions of the PI includes three ARs, which are cardiovascular toxicity, cardiac failure, and arrhythmia. Cardiovascular toxicity was diagnosed in 13% of patients, and it was grade 3 in 3.4% and grade 4 in 0.6%. Three of the cardiovascular toxicities (0.8%) were fatal and three patients (0.8%) permanently discontinued asciminib because of cardiovascular toxicity. Cardiac failure was seen in 2.2% of patients and was grade 3 in 1.1% of patients. One patient (0.3%) permanently discontinued asciminib because of cardiac failure. Arrhythmia, including QTc prolongation, was diagnosed in 7% of patients and was grade 3 in 2% of patients.

In the X2101 trial, CP-CML patients with the T315I mutation received a 5 times higher dose than patients in ASCEMBL. Adverse events of special interest of pancreatic toxicity, hypersensitivity, and cardiovascular toxicity were more frequent with the higher dose of asciminib. In addition, ARs like musculoskeletal pain, fatigue, gastrointestinal symptoms, and rash were more frequent with the higher dose of asciminib in X2101.

REGULATORY INSIGHTS

FDA bases regulatory approval decisions on substantial evidence of effectiveness which includes both the quality and the quantity of clinical evidence to establish effectiveness. For regular approval, substantial evidence of effectiveness is required which is based upon adequate and well-controlled trial(s), the endpoints should demonstrate clinical benefit, and the safety must support an overall favorable benefit-risk profile. Accelerated approval requires substantial evidence of a treatment effect on a surrogate endpoint or intermediate clinical endpoint that is reasonably likely to predict clinical benefit and provides a meaningful advantage taking into account the availability or lack of alternative therapies in serious and life-threatening diseases (12, 13). For both approval pathways, the quantity of clinical evidence to establish effectiveness can be provided by two adequate and well-controlled trials, one adequate and well-controlled trial plus confirmatory evidence, or one adequate and well-controlled large multicenter trial that can provide substantial evidence of effectiveness (12). When regulatory decisions are based upon substantial evidence of effectiveness from a single adequate and well-controlled clinical trial and confirmatory evidence, disease-specific considerations (e.g. unmet medical need, size of the patient population) are also considered (14). For the resistant or intolerant Ph+ CP-CML indication, the ASCEMBL trial, an adequate and well-controlled clinical trial, was considered as the primary trial; and confirmatory evidence of efficacy was provided by the X2101 trial. One interesting aspect of the asciminib application was that accelerated approval was granted for the resistant and intolerant CP-CML indication based on the randomized ASCEMBL trial; however, regular approval was granted for the T315I indication based on the X2101 trial which was a single-arm trial. For the T315I indication, the X2101 trial was considered as an adequate and well-controlled trial because it included a well-defined population in a disease setting with unmet medical need and the main efficacy endpoint was robust.

FDA’s decisions about regular approval are based on direct measure of clinical benefit or effect on an established surrogate. Accelerated approval decisions are based on a surrogate endpoint or intermediate clinical endpoint measured earlier than survival (13). According to the European LeukemiaNet recommendations, achievement of MMR at 12 months is considered to be an optimal milestone (15). MMR is associated with less progression to advanced phases of CML, a higher likelihood of deep and durable molecular responses, and hence treatment-free remission which may be one of the treatment goals in CML (16). Although long-term follow-up data from the IRIS and CML IV trials showed higher 10-year OS in patients who achieved MMR at 12 months, there is conflicting data about the timing of MMR achievement (17). The demonstration of a significantly higher MMR rate at 24 weeks with asciminib compared to bosutinib in ASCEMBL provided the primary basis of substantial evidence of efficacy for accelerated approval of asciminib in adult patients with Ph+ CP-CML, previously treated with two or more TKIs. Longer-term 96-week MMR and safety data supported conversion to regular approval. There were several regulatory considerations when making the decision to grant accelerated approval initially for the first indication and regular approval for the latter indication. The FDA approach toward approval in patients with resistant/intolerant CP-CML is to grant accelerated approval when there is at least 24 weeks of safety and efficacy data. For regular approval, FDA requires 96 weeks of safety and efficacy information. In the initial submission, safety and efficacy information from ASCEMBL was provided to 48 weeks; however, approximately 30% of patients had not reached the 96-week evaluation so this data could not support a regular approval. As one of the key secondary endpoints was MMR at 96 weeks, a post-marketing requirement was issued to obtain this information. In the X2101 trial, the majority of patients with the T315I mutation (90%) had reached the 96-week evaluation. MMR rate of 49% by 96 weeks supported by high MMR rates by 24 weeks in patients with or without prior ponatinib treatment and an acceptable safety profile of asciminib supported a regular approval.

In oncology, there are not many different dosing regimens for the same drug administered as monotherapy in different indications based on biological differences. Asciminib was approved at a higher dose for the T315I indication. FDA’s Project Optimus supports the development of a dosage(s) that is optimized to maximize efficacy as well as safety and tolerability for a specific indication. A comprehensive dose optimization approach of asciminib was utilized in the early clinical development where preliminary efficacy and safety data across a wide range of dosages or exposures were collected to understand exposure-response relationships (18). In addition, a robust nonclinical program allowed determination of target asciminib concentrations for optimal pharmacological activities (i.e. IC₅₀ – IC₉₀ values) in different tumour types and sub-mutations (7). With sufficient nonclinical and clinical data, pharmacometrics analyses was leveraged to select an optimized dosage for the specific tumour type or sub-mutation in late-stage clinical development (19). With the totality of data and a thorough understanding of exposure-response relationships, pharmacometrics analyses also helped explore alternative dosages for optimal risk-benefit assessment. The asciminib development program is a notable example where the dose optimization approach was used to select different dosages in different indications to minimize toxicity and maximize benefit to the patients (19).

For a chronic myeloid leukemia indication including relapsed/intolerant populations, FDA recommends use of randomized controlled trials to evaluate efficacy and safety. This is due to the availability of several approved TKIs for the treatment of CML. The use of randomized trials allows for the comparison of efficacy and safety between the investigational and control arms, which provides for better clinical context as compared to use of single-arm trials. This information is more useful for providers and patients when making appropriate treatment decisions.

CONCLUSIONS

Use of asciminib in patients with CP-CML previously treated with two or more TKIs led to higher MMR and CCyR rates at 96 weeks compared to bosutinib. In patients with CP-CML with the T315I mutation, asciminib led to an MMR rate of 49% by 96 weeks. Patients with CP-CML who were previously treated with two or more TKIs or have the T315I mutation have poor prognosis and there are limited treatment options. In general, the toxicities of asciminib were manageable, and adverse events of special interest, like thrombocytopenia, neutropenia, pancreatic enzyme elevations, are able to be mitigated with appropriate monitoring and dose modifications.