Olverembatinib (HQP1351), a well-tolerated and effective tyrosine kinase inhibitor for patients with T315I-mutated chronic myeloid leukemia: results of an open-label, multicenter phase 1/2 trial

Patient characteristics

Between October 26, 2016, and October 8, 2019, 165 patients were enrolled: 127 with CML-CP and 38 with CML-AP. A total of 110 (66.7%) patients were men. The median age was 42 (range, 20–74) years (Table 1) and median interval from CML diagnosis to first olverembatinib dose.

Table 1 Patient characteristics

5.7 (range, 0.3–23.2) years. A total of 30 (18.2%) patients had received 1 prior TKI (Additional file 1: Table 5), 90 (54.5%) had 2 prior TKIs, and 45 (27.3%) at least 3. These second-line patients were predominantly male and in chronic phase (70% each), with a median time from diagnosis to olverembatinib treatment of 1.6 years and of 28 of them had T315I mutation identified by Sanger sequencing.

Sanger sequencing identified 102 (61.8%) patients with a single T315I mutation, 25 (15.2%) with T315I and additional mutations (Additional file 1: Table 6), 14 (8.5%) with other mutations, and 24 (14.5%) with no BCR-ABL1 mutation. Corresponding data in 118 patients, (94 CML-CP and 24 CML-AP) receiving NGS were 53 (44.9%; single T315I mutation), 19 (16.1%; T315I and additional noncompound mutations), 14 (11.9%; other noncompound mutations), 12 (10.2%; compound mutations), and 20 (16.9%; no BCR-ABL1 mutation). Additional mutations observed in conjunction with T315I included phosphate-binding loop (P-loop) residue E255 (which confers resistance against imatinib, nilotinib, and bosutinib), C-lobe residue F359 (nilotinib), and gatekeeper F317 (dasatinib).

Phase 1 study

A total of 101 patients with TKI-resistant CML were enrolled in the phase 1 study between October 26, 2016, and December 12, 2018. Among these 101 patients, 86 had CML-CP and 15, CML-AP. Demographic and baseline disease characteristics are summarized in Additional file 1: Table 7. Sanger sequencing identified 46 (45.5%) patients with a single T315I mutation, 17 (16.8%) with T315I and additional mutations, 14 (13.9%) with other mutations, and 24 (23.8%) with no BCR-ABL1 mutation. Corresponding data in 94 patients (81 CML-CP and 13 CML-AP) receiving NGS were 33 (35.1%; single T315I mutation), 16 (17.0%; T315I and additional noncompound mutations), 13 (13.8%; other noncompound mutations), 12 (12.8%; compound mutations), and 20 (21.3%; no BCR-ABL1 mutation). Across 11 dose cohorts (1 to 60 mg QOD) in 28-day cycles, no DLT was observed at doses below 60 mg. Two of three patients in the 60 mg cohort experienced DLTs, including 1 G4 thrombocytopenia and 1 myocardial injury, which resulted in dose interruption. Among 33 patients in dose escalation, the MTD was 50 mg. Based on preliminary safety and efficacy results, we expanded the 30, 40, and 50 mg QOD dose cohorts. Finally, the RP2D was established as 40 mg QOD.

Phase 2 study

After determination of RP2D, two pivotal studies were initiated at 10 sites in China from April 26, 2019, to October 8, 2019, that enrolled 41 patients with T315I-mutated CML-CP and 23 with T315I-mutated CML-AP. Demographic and baseline disease characteristics are summarized in Additional file 1: Table 7. More than 60% of patients were treated in second-line therapy. Sanger sequencing identified 37 (90.2%) patients with a single T315I mutation and 4 (9.8%) with T315I and additional mutations in CML-CP; while patients with CML-AP were 19 (82.6%) with a single T315I mutation and 4 (17.4%) with T315I and additional mutations. Corresponding data in 24 patients (13 CML-CP and 11 CML-AP) receiving NGS were 11 (84.6%) with a single T315I mutation, 1 (7.7%) with T315I and additional noncompound mutations, and 1 (7.7%) with other noncompound mutations in patients with CML-CP, as well as 9 (81.8%; single T315I mutation) and 2 (18.2%; T315I and additional noncompound mutations) in those with CML-AP.

Patient disposition

By September 30, 2021, the median follow-up was 34.3 (range, 4.8–58.6) months. A total of 114 (69.0%) patients remained on treatment at doses of 20 (n = 11), 30 (n = 44), 40 (n = 48), or 50 mg (n = 11) QOD (Table 2). Treatment interruptions due to AEs occurred in 86 (52.1%) patients, including 58 (45.7%) with CML-CP and 28 (73.7%) with CML-AP. Median number and duration of treatment interruptions were 2 (range, 1–11) and 56 (range, 1–421) days.

Table 2 Patient disposition

Doses were reduced because of AEs in 50 (30.3%) patients, including 36 (28.3%) with CML-CP and 14 (36.8%) with CML-AP. The most common AE leading to dose reduction was severe thrombocytopenia (n = 26; 15.8%) (Additional file 1: Table 2). Fifty-one patients permanently discontinued treatment because of either CML progression (n = 15; 9.1%), AEs (n = 13; 7.9%), treatment failure (n = 12; 7.0%), consent withdrawal (n = 8; 4.8%), or death (n = 3; 1.8%) (Table 2).

Safety

The median treatment duration was 30.7 (range, 1.2–58.6) months. All 165 patients experienced at least 1 treatment-related AE (TRAE), of which 131 (79.4%) were G3/4 (Table 3). The most frequent nonhematologic TRAE was skin hyperpigmentation in 139 (84.2%) patients with pathologically confirmed lentiginous nevus in 2 patients, followed by hypertriglyceridemia (57.6%), proteinuria (50.9%), hyperbilirubinemia (41.8%), hypocalcemia (38.8%), and elevated liver transaminases (35.8%). Median time to onset of these TRAEs was 70 (range, 1–1,315) days.

Table 3 Treatment-related adverse events (≥ 10%, all grade)

Cardiovascular events (CVEs) possibly related to olverembatinib were observed in 53 (32.1%) patients at a median of 11 (range, 0.03–53) months on treatment, including hypertension (13.3%), pericardial effusion (8.5%), ventricular extrasystoles (4.2%), supraventricular extrasystoles or atrial fibrillation (3.0% each), retinal-vein occlusion (1.8%) or palpitations (1.2%); as well as angina pectoris, arrhythmia, atrial tachycardia, cardiomegaly, cerebral ischemia, and/or cerebral infarction in < 1%; of which 11.5% were G3/4 (Additional file 1: Table 3).

The median age of patients with CVEs was 43 (range, 20–74) years, including two patients with a history of hypertension and three with prior diabetes. All patients with CVEs were required to have temporary olverembatinib treatment suspension and received disease-specific treatment. Most recovered or improved and received olverembatinib treatment at a reduced dose except for one patient who discontinued because of acute myocardial infarction (MI) and one patient who died of pericardial effusion. None of these patients had a Fridericia-corrected QT interval exceeding 500 ms on treatment. Grade 3/4 hematologic TRAEs included thrombocytopenia (51.5%), neutropenia (11.5%), and anemia (23.0%) (Table 3). Myelosuppression tended to occur early, with a median (range) onset time of 28 (4–676) days and a median (range) duration of 36 (7–718) days. Most resolved after temporary treatment suspension or supportive care, including platelet or erythrocyte transfusion or dose adjustment. A total of 11.5% patients received platelet transfusions and 9.7% patients received red blood cell transfusions. Except for skin hyperpigmentation and proteinuria, incidences of TRAEs decreased over time during the follow-up period (Fig. 1). Patients with persistent proteinuria did not observe decreased renal function according to estimated glomerular filtration rate. Serious AEs (SAEs; in ≥ 1% of patients) included thrombocytopenia (9.0%), anemia (6.0%), pneumonia (3.0%), pyrexia or atrial fibrillation (2.0% each); as well as acute MI, cholelithiasis, pericardial effusion, upper-respiratory-tract infection, and urinary-tract infection (1% each). Most SAEs resolved after temporary treatment suspension or dose reduction, which was required in 75 (46.0%) patients at a median of 5 (range 1–36) months.

Fig. 1figure 1

Prevalence of treatment-related adverse events over time

EfficacyCML-CP

The median follow-up period of 126 evaluable patients with CML-CP since the start of an effective dose (≥ 30 mg QOD) was 37 (range, 7–58) months. All 84 patients without baseline CHR achieved this endpoint. Of 121 patients without MCyR at baseline, 96 (79.3%) and 84 (69.4%) achieved MCyR and CCyR at a median of 3 (range, 3–36) and 3 (range, 3–37) months, respectively. Among 126 patients, 70 (55.6%) achieved MMR, 56 (44.4%) MR4.0, and 49 (38.9%) MR4.5 on olverembatinib therapy. Cytogenetic and molecular response rates increased over time (Fig. 2A). The cumulative 3-year incidences of MCyR, CCyR, MMR, MR4.0, and MR4.5 were 78.6% (95% CI: 70.0%, 85.0%), 69.0% (59·7%, 76·5%), 55.9% (46.5%, 64.4%), 43.5% (34.6%, 52.1%), and 38.6% (30.0%, 47.1%), respectively. The probabilities of sustained MCyR, CCyR, and MMR at 3 years were 77.3% (66.8%, 84.8%), 72.2% (60.4%, 81.1%), and 76.0% (62.1%, 85.3%), respectively. A total of 5 patients progressed to CML-AP (n = 4) or CML-BP (n = 1). Seven patients died of disease progression (n = 3) and one each of either pericardial effusion, gastric cancer, hepatitis E virus infection (with prior autoimmune hepatitis), or an unknown reason. Probabilities of PFS and OS at 3 years were 92.0% (86.0%, 96.0%) and 94.0% (89.0%, 97.0%), respectively (Fig. 3A).

Fig. 2figure 2

Cumulative incidence of responses in the chronic phase (A) and accelerated phase (B) MCyR, major cytogenetic response; CCyR, complete cytogenetic response; MMR, major molecular response; MR4.0, molecular response 4; MR4.5, molecular response 4.5.

Fig. 3figure 3

Progression-free survival (PFS) and overall survival (OS) in the chronic phase (A) or accelerated phase (B)

CML-AP

The median follow-up was 27 (range, 5–56) months since the onset of an effective dose. Among 37 patients without baseline MaHR, 29 (78.4%) met this endpoint at a median of 3 (range, 1–7) months; and 27 (73%) experienced CHR at a median of 3 (range, 1–14) months. Of the 38 patients without baseline MCyR, 18 (47.4%) achieved MCyR and CCyR at a median of 3 (range, 1–9) months and 4 (range, 1–15) months. MMR and MR4.0 were achieved by 17 (44.7%) and 14 (36.8%) patients, respectively, and MR4.5 by 13 (34.2%) patients. Cytogenetic and molecular response rates increased over time (Fig. 2B). The 3-year cumulative incidences of achieving MCyR, CCyR, MMR, MR4.0, and MR4.5 were 47.4% (30.7%, 62.4%), 47.4% (30.6%, 62.4%), 44.7% (28.2%, 60.0%), 39.3% (22.3%, 56.0%), and 32.1% (17.6%, 47.6%), respectively. The probabilities of sustained MCyR and CCyR at 3 years were 86.0% (55.0%, 97.0%) and 71% (44.0%, 87.0%). A total of 11 patients had CML that progressed to blast phase, and 4 died. Probabilities of PFS and OS at 3 years were 60.0% (41.0%, 74.0%) and 71% (54.0%, 83.0%), respectively (Fig. 3B).

Responses according to BCR-ABL1 mutation status via Sanger sequencing

Among the four subgroups with CML-CP evaluated by Sanger sequencing, patients with a single T315I mutation had the highest 3-year cumulative incidences of achieving MCyR (85.3%), CCyR (76.0%), MMR (68.7%), MR4.0 (59.3%), and MR4.5 (54.7%); those with no BCR-ABL1 mutation had the lowest cumulative incidences of MCyR (59.1%), CCyR (50.0%), MMR (9.1%), MR4.0 (4.5%), and MR4.5 (0) (all P values among the four subgroups < 0.0001; Fig. 4A). Among patients with CML-AP, those with a single T315I mutation also had the highest 2-year cumulative incidences of MCyR or CCyR (60.0% each), MMR (52.0%), and MR4.0 or MR4.5 (40.0% each); followed by those with T315I and an additional mutation, who had MCyR, CCyR, or MMR (33.3% each), MR4.0 (22.2%), and MR4.5 (11.1%). No cytogenetic or molecular response was observed in patients with no BCR-ABL1 mutation or with other mutations at enrollment (Fig. 4B). We interrogated baseline covariates to evaluate associations with cytogenetic and molecular responses (Table 4).

Fig. 4figure 4

Responses by baseline BCR-ABL1 mutation status in the chronic phase (A) or accelerated phase (B) using Sanger sequencing. CCyR, complete cytogenetic response; MCyR, major cytogenetic response; MMR, major molecular response; MR4.0, molecular response 4; MR4.5, molecular response 4.5

Table 4 Multivariate analysis results of variables associated with treatment responses

In multivariate analyses, BCR-ABL1 mutation status before study entry was independently associated with higher cumulative incidences of achieving CCyR (P = 0.02), MMR (P = 0.006), MR4.0 (P = 0.006), and MR4.5 (P < 0.0001). Compared to no mutation or other mutation status, a single T315I mutation showed higher responses (or a trend in this direction). In addition, a longer interval from CML diagnosis to olverembatinib treatment onset, CML-AP rather than CML-CP, and more prior TKIs were significantly associated with lower rates of cytogenetic and molecular responses.

Responses according to BCR-ABL1 mutation status via NGS

Findings on treatment responses stratified by BCR-ABL1 mutation status with NGS paralleled those with Sanger sequencing (Fig. 5). Among 12 patients (7 with CML-CP and 5 with CML-AP) with compound mutations (Table 5), 8 (67.0%) had T315I-inclusive compound mutations; and 9 (75.0%), 2 (17.0%), and 1 (8.0%) had 1, 2, and 3 compound mutations, respectively. A total of 7 (58.0%) achieved MMR and 3 (25.0%) MR.4.5 By the last follow-up date, 3 patients had progressed to CML-AP or CML-BP and died, and 7 remained on treatment with olverembatinib, of whom one had CCyR and two each had CHR, MMR, or MR.4.5

Fig. 5figure 5

Responses by baseline BCR-ABL1 mutation status in the chronic phase (A) or accelerated phase (B) using next-generation sequencing. CCyR, complete cytogenetic response; MCyR, major cytogenetic response; MMR, major molecular response; MR4.0, molecular response 4; MR4.5, molecular response 4.5

Table 5 Characteristics of 12 patients harboring compound mutations at baseline

In multivariate analyses, BCR-ABL1 mutation status before study entry was independently associated with cumulative incidences of achieving MMR (P = 0.002), MR4.0 (P < 0.0001), and MR4.5 (P < 0.0001; Table 4). Univariate analysis results of variables associated with treatment response are in Additional file 1: Table S4. A single T315I mutation was also significantly associated with higher molecular response rates; compound mutation was comparable to other mutation status except no mutations; and no mutation was associated with lower response rates. In addition, longer intervals from CML diagnosis to olverembatinib treatment onset, CML-AP (rather than CML-CP), more prior TKIs, and increasing age were significantly associated with lower rates of cytogenetic and/or molecular responses.

Pharmacokinetics and pharmacodynamics

Based on plasma concentration–time curves on treatment Days 1 and 27, olverembatinib pharmacokinetics were linear, with mean terminal elimination half-life values (17.5–36.5 h) that are well suited to QOD administration (Fig. 6). Systemic exposure and maximum concentration were approximately dose proportional across olverembatinib doses (1–60 mg QOD). With multiple dosing, slight to moderate accumulation of olverembatinib was observed on C1D27. Significant dose- and time-dependent reductions in pCRKL levels (indicative of BCR-ABL1 inhibition) were observed within 8 h after olverembatinib dosing (30–50 mg QOD) on C1D1 and maintained at steady state on C1D15 and C1D27 (Fig. 7).

Fig. 6figure 6

Pharmacokinetics, mean plasma concentration–time curves on treatment Days 1 (A) and 27 (B)

Fig. 7figure 7

Pharmacodynamics by dose cohorts on Cycle 1 (A) and on Day 1 of Cycle 1 (B)

留言 (0)

沒有登入
gif