Safety, pharmacokinetic, pharmacodynamic, and efficacy properties of orally administered APG-2449 in patients with advanced ALK⁺ and ROS1⁺ non-small-cell lung cancer: a multicentre, open-label, single-arm phase 1 trial

Summary

Background

APG-2449 is a focal adhesion kinase (FAK) inhibitor and a third-generation anaplastic lymphoma kinase (ALK)-proto-oncogene receptor tyrosine kinase ROS (ROS1) tyrosine kinase inhibitor (TKI). The aim of this first-in-human study was to evaluate safety and efficacy of APG-2449 in patients with advanced non-small-cell lung cancer (NSCLC).

Methods

This single-arm, multicentre, phase 1 clinical trial included a dose escalation to determine maximum tolerated dose (MTD) and recommended phase 2 dose (RP2D), followed by a dose-expansion stage at RP2D to evaluate safety. Secondary endpoints included pharmacokinetics, pharmacodynamics, and preliminary efficacy. APG-2449 was administered once daily in 28-day cycles through a 3 + 3 dose escalation. Eligible patients were aged 18 years or older and had advanced ALK or ROS1 fusion gene–positive NSCLC and other solid tumours (during the dose escalation). Patients with ALK-positive or ROS1-positive NSCLC regardless of previous TKI treatment were enrolled during the dose escalation. After RP2D was determined, patients with second-generation TKI-treated ALK⁺ NSCLC, TKI-naïve ALK⁺, and TKI-naïve or TKI-treated ROS1⁺ NSCLC were enrolled in dose expansion. Eligible patients also had: (1) an Eastern Cooperative Oncology Group performance status (ECOG PS) score of 1 or less; (2) adequate bone marrow reserve and organ function; and (3) no or stable brain metastases. This study was registered with ClinicalTrials.gov (NCT03917043). Recruitment is ongoing.

Findings

144 patients were enrolled from May 27, 2019, to April 2, 2024, and received APG-2449 (150–1500 mg/day). MTD was not reached, and RP2D was 1200 mg. The most common grade ≥3 treatment-emergent adverse events (TEAE) were anaemia [6 (4.2%)], alanine aminotransferase (ALT) increased [6 (4.2%)], electrocardiogram QT prolonged [5 (3.5%)], and pneumonia [5 (3.5%)]. The most common grade ≥3 treatment-related adverse events (TRAEs) were ALT increased [5 (3.5%)], QT prolonged [5 (3.5%)], and vomiting [2 (1.4%)]. There were no treatment-related deaths. Pharmacokinetic analyses indicated a dose-proportional increase in plasma exposure, and that APG-2449 penetrated the blood-brain barrier (cerebrospinal fluid [CSF]-to-free plasma ratio 0.65–1.66). Of patients with ALK-positive, TKI-naïve NSCLC (n = 14) treated at RP2D, the objective response rate (ORR) was 78.6% (11/14), and median progression-free survival (mPFS) was not reached. Of 22 patients with NSCLC resistant to second-generation ALK inhibitors (without ALK compound mutations or activation of bypass or downstream signalling pathways), 10 (45.5%) achieved partial response, with a mPFS of 13.6 months. Intracranial partial response occurred in 9/12 patients at the RP2D (intracranial ORR: 75.0%).

Interpretation

APG-2449 demonstrated favourable preliminary safety, pharmacokinetics, and efficacy in TKI-untreated or second-generation ALK-resistant NSCLC. Higher baseline tumour phosphorylated FAK levels were associated with greater APG-2449 treatment benefit. Targeting FAK signalling may provide a feasible strategy for overcoming second-generation ALK TKI resistance.

Funding

Ascentage Pharma Group Corp Ltd. (Hong Kong).

Research in context.

Evidence before this study

We searched for articles and abstracts in PubMed and from major oncology congresses (i.e., American Society of Clinical Oncology, World Conference on Lung Cancer, American Association for Cancer Research, and European Society for Medical Oncology) for studies on non-small-cell lung cancer (NSCLC) and targeted therapies published from database inception to July 21, 2025, using search terms “non-small-cell lung cancer,” “anaplastic lymphoma kinase (ALK) tyrosine kinase inhibitor (TKI),” “resistance,” and “focal adhesion kinase (FAK)” (without language restriction). Although second-generation ALK tyrosine kinase inhibitors such as ceritinib, alectinib, and brigatinib have both systemic and intracranial antitumour activity with acceptable safety profiles in patients previously treated with crizotinib, treatment options after failure of a second-generation ALK inhibitor are limited. The only third-generation TKI, lorlatinib, has an objective response rate of 40%, but it is associated with metabolic and central nervous system toxicities. To develop novel, targeted therapies in this context, it is important to understand mechanisms of resistance. Preclinical studies provided evidence that targeting FAK could overcome resistance to crizotinib through the phosphoinositide-dependent protein kinase-1-AKT serine–threonine kinase 1 (PDPK1-AKT1) pathway, but no clinical trials have yet targeted FAK in this setting. APG-2449 is a novel third-generation ALK-proto-oncogene receptor tyrosine kinase ROS (ROS1) TKI and active FAK inhibitor.

Added value of this study

This study is the first to examine the safety and efficacy of APG-2449, an oral third-generation ALK-ROS1 TKI and FAK inhibitor, in patients with ALK-positive NSCLC, including those with TKI-naïve and second-generation TKI-resistant disease. APG-2449 was administered at a dose of 1200 mg (recommended phase 2 dose) and achieved response rates of 78.6% in TKI-naïve patients and 45.5% in those with resistant cases, with a median progression-free survival of 13.6 months. Notably, APG-2449 crossed the blood-brain barrier, resulting in an intracranial response rate of 75.0% in patients with brain metastases, meeting a critical treatment need. Biomarker analyses indicated that targeting FAK signalling may help to overcome resistance to previous generations of ALK TKIs.

Implications of all the available evidence

Together with prior research, our results suggest that APG-2449 may be a viable treatment option for patients with NSCLC resistant to current ALK TKIs, particularly those with brain metastases. By targeting both ALK-ROS1 and FAK, APG-2449 offers a novel approach to overcome resistance and improve outcomes for patients with limited treatment options. Further trials are warranted to confirm these results and investigate FAK-targeted therapies, which could redefine treatment strategies for ALK-positive NSCLC, particularly for brain-involved cases.

Introduction

To date, multiple anaplastic lymphoma kinase (ALK) tyrosine kinase inhibitors (TKIs) have been approved for first- or second-line treatment of patients with ALK-positive (ALK⁺) non-small-cell lung cancer (NSCLC). With approvals of second-generation ALK TKIs, these agents have gradually replaced the first-generation TKI crizotinib as the standard initial treatment for patients with newly diagnosed advanced ALK⁺ NSCLC.

Despite initial clinical benefits of second-generation ALK TKIs, almost all patients eventually develop resistance and experience disease relapse. After disease progression with second-generation TKIs, the standard of care for ALK resistance comprises third-generation ALK inhibitors such as lorlatinib. However, the challenge of patients with other or unclear resistance mechanisms opens avenues for pharmacologic innovation. Studies suggest that alteration of the focal adhesion kinase (FAK) pathway is involved in early adaptive tolerance induced by ALK inhibitors in patients with ALK⁺ NSCLC.¹ Therefore, there is still an unmet clinical need in this patient population, and further research and development of other next-generation ALK TKIs with improved or equivalent efficacy and safety profiles as compared with lorlatinib are needed.

Investigational APG-2449 is an orally active small-molecule FAK-ALK-ROS1 multitarget kinase inhibitor. Binding constants (K_d) for FAK-ALK-ROS1 kinase are 5.4, 1.6, and 0.81 nM, respectively. Preclinical data showed that half-maximal inhibitory concentration (IC₅₀) values of APG-2449 against wild-type ALK and five ALK mutations (i.e., L1196M, F1197M, G1269A, S1206Y, G1202R) were 0.85, 0.73, 3.3, 4.3, 1.3, and 9.2 nM, respectively.²

G1202R is the most important resistant/refractory ALK mutation, and APG-2449 is effective against this mutation in vitro, suggesting that it might be a clinically active third-generation ALK inhibitor. In animal models, APG-2449 also has significant antitumour effects in tumours harbouring these mutations (EML4-ALK, G1202R) and appears to be more potent than approved first-generation and second-generation ALK inhibitors.³

Therefore, we conducted this study to evaluate the tolerability, safety, preliminary antitumour activity, and pharmacologic profile of APG-2449 monotherapy in subjects with advanced solid tumours, especially NSCLC.

Methods

Study design and participants

This single-arm, multicentre, phase 1 clinical trial included a dose-escalation stage to determine maximum tolerated dose (MTD) and recommended phase 2 dose (RP2D), followed by a dose-expansion stage at the RP2D to evaluate safety. Secondary endpoints included evaluating the pharmacokinetic (including the effects of fasting and fed administration on pharmacokinetics) and pharmacodynamic profiles, as well as the preliminary efficacy of APG-2449.

The dose-escalation stage was conducted using a standard 3 + 3 design. APG-2449 was initially administered at a dose of 150 mg. Subsequent doses were adjusted from 300 to 1500 mg across seven dose-escalation levels (300, 450, 600, 750, 900, 1200, and 1500 mg), according to observed dose-limiting toxicities (DLTs). In the dose-expansion stage, patients received APG-2449 at the RP2D established during the dose-escalation stage.

APG-2449 was administered orally once daily during 28 days in one cycle. A patient could be discontinued from the study for any of the following reasons (whichever came first): disease progression, permanent drug discontinuation because of intolerable drug-related adverse events (AEs), or an investigator's decision that a patient would not continue to benefit from the study.

Eligible patients were ≥18 years of age and had histologically and/or cytologically confirmed ALK or ROS1 fusion gene–positive NSCLC, malignant pleural mesothelioma, oesophageal carcinoma, or ovarian carcinoma (during the dose-escalation stage).

Patients with ALK-positive or ROS1-positive NSCLC regardless of previous TKI treatment (including patients with TKI-naïve disease and those who had experienced failure with first-, second-, or third-generation TKIs) were enrolled during the dose-escalation stage. After the RP2D of 1200 mg was determined, patients with second-generation TKI-treated ALK⁺ NSCLC (second-generation TKI failed ALK⁺ NSCLC), TKI-naïve ALK⁺, and TKI-naïve or TKI-treated ROS1⁺ NSCLC were enrolled in the dose-expansion stage. For the efficacy report, all NSCLC patients were divided into four cohorts: Cohort 1 included patients previously untreated with ALK TKIs; Cohort 2 included those with ALK TKI–failed disease; Cohort 3 included patients without previous ROS1 TKI treatment; and Cohort 4 included those with ROS1 TKI–resistant disease.

The molecular diagnosis of these patients was confirmed by the investigator. ALK and ROS1 were detected by immunohistochemistry (IHC), reverse transcription-polymerase chain reaction, fluorescence in situ hybridisation, or next-generation sequencing (NGS) according to clinical guidelines in China (Supplementary Table S1). Eligible patients also had: (1) an Eastern Cooperative Oncology Group performance status (ECOG PS) score of 1 or lower; (2) expected survival of at least 3 months; (3) at least one measurable lesion according to Response Evaluation Criteria in Solid Tumours (RECIST) version 1.1; (4) adequate bone marrow reserve and organ function; and (5) no brain metastases or brain metastasis in stable condition for 28 days after treatment. A further requirement was to provide either fresh (for recurrent subjects only) or archived tumour tissue samples within 28 days before treatment. If none of these specimens were available, enrolment was permitted after consultation with the sponsor (Ascentage Pharma Group Inc.).

Ethics

This report presents results of APG-2449 in subjects with advanced solid tumours (ClinicalTrials.gov identifier: NCT03917043). The study was conducted in accordance with the Declaration of Helsinki, Good Clinical Practice guidelines, and applicable local regulations. The study protocol was approved by the ethics committee of Sun Yat-sen University Cancer Center (approved number: A2019-006-01) and each participating institution. Written informed consent was obtained from all subjects before enrolment.

Procedures

APG-2449 was administered orally as a single dose in a 3-day pharmacokinetic run-in period, followed by once-daily dosing in continuous 28-day cycles. Blood samples were drawn for serial pharmacokinetic profiling of APG-2449 up to 72 h postdose on Cycle 1 Day 1 (C1D1) and up to 24 h postdose on C1D28 for all patients.

To evaluate the potential effect of food on pharmacokinetics, we administered APG-2449 at doses ranging from 150 to 900 mg to a subset of patients under fasting conditions, while the other patients received APG-2449 at doses ranging from 450 to 1500 mg with meals.

To explore the relationship between concentrations of APG-2449 in cerebrospinal fluid (CSF) and plasma, we collected CSF with time-matched plasma samples from six patients who had ALK-positive NSCLC with brain metastases and underwent lumbar puncture. Concentrations of APG-2449 in plasma and CSF samples were determined using a validated liquid chromatography with tandem mass spectrometry method. Pharmacokinetic parameters of APG-2449 were calculated for each patient by employing standard noncompartmental analysis using WinNonlin software version 8.3 (Certara USA, Inc., Princeton, NJ).

For pharmacodynamic analyses, peripheral blood samples were collected before and 4 h after dosing on C1D1 and 4 h after dosing on C1D28. Phosphorylated FAK (pFAK) levels in peripheral blood mononuclear cells (PBMCs) were evaluated using the FAK [pY397] human enzyme-linked immunosorbent assay kit (Cat. # KHO0441; ThermoFisher Scientific, Waltham, MA).

Baseline tumour tissue samples were collected to evaluate protein expression of pFAK by IHC with anti-pFAK Y397 rabbit antibody (Cat. # 341292; Merck Millipore, Burlington, MA). pFAK level was expressed as the H-score, which was in turn defined as the percentage of pFAK-stained tumour area multiplied by pFAK mean intensity⁴ (Supplementary Fig. S1).

$$H-score=\left(\%pFAKstainedareaoflowintensity\times 1\right)+\left(\%pFAKstainedareaofmediumintensity\times 2\right)+\left(\%pFAKstainedareaofhighintensity\times 3\right)$$

Formalin-fixed paraffin-embedded (FFPE) tissues were collected for sequencing. Peripheral blood was collected to separate circulating tumour DNA. DNA was extracted using the QIAamp DNA FFPE Tissue Kit (Qiagen). Targeted NGS was performed using a panel covering 139 cancer-relevant genes (Supplementary Table S2) on the Illumina HiSeq4000 platform. The entire NGS process and analyses were conducted in a clinical laboratory improvement amendments (CLIA)-certified and College of American Pathologists (CAP)-accredited central laboratory (Nanjing Geneseeq Technology Inc.).

Outcomes

Safety assessments included vital signs, physical examination, haematology tests, blood biochemistry, coagulation function, electrocardiogram, pregnancy tests for women, and clinical symptoms. AEs were graded according to Common Terminology Criteria for Adverse Events version 5.0 at each clinic visit from the time of informed consent until 30 days after the last dose of study drug or when the participant started treatment with a new clinical regimen, died, or was lost to follow-up, whichever occurred first.

MTD was defined as the level below the dose at which DLTs occurred in at least two of six subjects in treatment Cycle 1. Tumours were evaluated by computed tomography or magnetic resonance imaging according to RECIST version 1.1 during the screening period (within 4 weeks before the first dose), Cycle 2, Cycle 4, and Day 1 of even-numbered cycles.

Antitumour activity parameters included complete response, partial response, stable disease, progressive disease, objective response rate (ORR; defined as the proportion of study participants who achieved complete response or partial response), disease control rate (DCR; defined as the proportion of subjects with complete response, partial response, or stable disease), and duration of response (DOR; defined as the time from first documented complete response or partial response to the first assessment of progressive disease or death, whichever occurred first). Progression-free survival (PFS) was defined as the time from first administration of APG-2449 to the first record of disease progression (according to RECIST version 1.1) or the date of death due to any reason.

Statistical analysis

No formal hypotheses were tested in this study. Sample sizes were determined by discussion between the sponsor (Ascentage Pharma Group Inc.) and investigators according to observed toxicities and the pharmacokinetic profiles of all subjects who received at least one dose of APG-2449. Ninety-five percent confidence intervals (CIs) of the DCR and ORR were assessed by the Clopper-Pearson method. The Kaplan–Meier method was used to assess PFS and DOR. For categorical data, numbers and percentages of individuals are presented. Pearson's correlation test was used to evaluate the relationship between pFAK expression and PFS. All statistical analyses were performed using SAS version 9.4 (SAS Institute Inc., Cary, NC).

Role of the funding source

The funder of the study was involved in the study design, data collection and analysis, and writing of the report.

Results

The study is ongoing and for this report, patients were enrolled from May 27, 2019, up to the cutoff date of April 2, 2024. A total of 144 patients with advanced solid tumours were enrolled, including 130 (90.3%) with NSCLC, five (3.5%) with malignant pleural mesothelioma, and nine (6.3%) with ovarian cancer. Among them, 67 (46.5%) of 144 patients were men, and the median (interquartile range [IQR]) age was 53.0 (46–59) years. Data on race/ethnicity were not collected. Baseline characteristics are summarised in Table 1.

Table 1.

Baseline patient characteristics.

	Total	TKI-naive ALK+ NSCLC	TKI-failed ALK+ NSCLC	TKI-naive ROS1+ NSCLC	TKI-failed ROS1+ NSCLC	Othersa
Safety population	144	16	75	26	13	14
Age, yr
Mean (SD)	52.2 (10.6)	49.8 (12.5)	52.6 (10.4)	52.4 (10.8)	50.4 (12.0)	53.9 (7.9)
Median (IQR)	53.0 (46–59)	52.5 (42.5–59)	54.0 (48–60)	52.5 (46–58)	50.0 (45–55)	53.0 (47–61)
Sex, n (%)
Female	77 (53.5)	9 (56.3)	36 (48.0)	12 (46.2)	8 (61.5)	12 (85.7)
Male	67 (46.5)	7 (43.8)	39 (52.0)	14 (53.8)	5 (38.5)	2 (14.3)
ECOG PS, n (%)
0	46 (31.9)	9 (56.3)	24 (32.0)	8 (30.8)	3 (23.1)	2 (14.3)
1	98 (68.1)	7 (43.8)	51 (68.0)	18 (69.2)	10 (76.9)	12 (85.7)
Time from diagnosis to APG-2449 treatment, yr
Mean (SD)	2.3 (2.3)	0.8 (2.3)	2.7 (1.5)	0.6 (1.0)	3.7 (3.4)	4.3 (3.1)
Median (IQR)	1.9 (0.6–3.4)	0.1 (0.06–0.09)	2.3 (1.6–3.7)	0.1 (0.04–0.3)	2.2 (1.3–5.2)	3.3 (1.7–5.9)
Prior TKIs, n (%)
Alectinib	37 (25.7)	0	37 (49.3)	0	0	0
Crizotinib	45 (31.3)	0	35 (46.7)	0	12 (92.3)	0
Lorlatinib	4 (2.8)	0	4 (5.3)	0	0	0
Others	22 (15.3)	0	19 (25.3)	0	3 (23.1)	0
Lines of prior TKI treatment, n (%)
0	50 (34.7)	16 (100.0)	0	26 (100.0)	0	8 (57.1)
1	42 (29.2)	0	30 (40.0)	0	8 (61.5)	4 (28.6)
2	33 (22.9)	0	29 (38.7)	0	2 (15.4)	2 (14.3)
≥3	19 (13.2)	0	16 (21.3)	0	3 (23.1)	0
Prior chemotherapy, n (%)	56 (38.9)	2 (12.5)	27 (36.0)	6 (23.1)	8 (61.5)	13 (92.9)
TKI-failed mutation at baseline
ALK-ROS1-resistant mutation	–	–	19 (25.3)	–	0	–
Activation of bypass and/or downstream signalling pathwaysb	–	–	10 (13.3)	–	1 (7.7)	–
TP53	–	–	23 (30.7)	–	5 (38.5)	–
Brain metastasis, n (%)	77 (53.5)	5 (31.3)	51 (68.0)	10 (38.5)	11 (84.6)	0

Abbreviations: ALK, anaplastic lymphoma kinase; ECOG PS, Eastern Cooperative Oncology Group performance status; IQR, interquartile range; NSCLC, non-small-cell lung cancer; TKI, tyrosine kinase inhibitor.

Cutoff date: April 2, 2024.

^aOthers included malignant pleural mesothelioma and ovarian cancer.

^bActivation of bypass and/or downstream signalling pathways including FGFR3, FGFR4, ERBB2, RET, KRAS, BRAF V600E, and PIK3CA.

Patients were enrolled in eight dose-level cohorts receiving APG-2449 at doses ranging from 150 to 1500 mg (Fig. 1). According to baseline TKI treatment, four cohorts were defined within the NSCLC group. In Cohort 1 (ALK TKI-naïve disease), 16 patients received APG-2449 at doses of 150–1200 mg. In Cohort 2 (ALK TKI-failed disease, n = 75), 67 patients with ALK⁺ NSCLC that had failed second-generation TKIs (excluding three with prior failure on only first-generation TKIs and five with prior failure on third-generation TKIs) received APG-2449 at doses of 300 to 1500 mg; of these, 45 (67.2%) of 67 had brain metastases. In Cohort 3 (ROS1⁺ TKI-naïve disease), 26 patients received APG-2449 at 1200 mg; of these, 10 (38.5%) had brain metastases. In Cohort 4 (ROS1⁺ TKI-failed disease), 13 patients with ROS1⁺ NSCLC and prior TKI failure received APG-2449 at doses of 300–1500 mg; of these, 11 (84.6%) had brain metastases. Patient disposition is presented in Fig. 1.

Fig. 1.

Patient disposition. Cutoff date: April 2, 2024. Abbreviations: AE, adverse event; ALK⁺, anaplastic lymphoma kinase positive; EAS, efficacy analysis set; NSCLC, non-small-cell lung cancer; PD, progressive disease; PI, principal investigator; ROS1⁺, ROS1 positive; SAS, safety analysis set; TKI, tyrosine kinase inhibitor.

A total of 132 (91.7%) of 144 patients had at least one treatment-related adverse event (TRAE; Table 2). Common (≥10%) TRAEs included elevated blood creatinine (51.4%), elevated alanine aminotransferase (ALT; 45.1%) or aspartate aminotransferase (AST; 38.2%), nausea (28.5%), vomiting (23.6%), decreased leucocyte count (22.9%), diarrhoea (22.2%), decreased neutrophil count (18.8%), and rash (12.5%). The incidence of grade ≥ 3 TRAEs was 15.3% (22/144). Grade ≥ 3 TRAEs were not greater than 33.3% in any of the eight dose cohorts (Table 2). Dose reductions and dose interruptions attributable to TRAEs occurred in nine (6.3%) and 28 (19.4%) patients, respectively. APG-2449 was permanently discontinued in one (0.7%) patient due to TRAEs. No DLTs occurred, and the MTD was not reached. Of 144 patients, 37 (25.7%) experienced serious AEs (SAEs). Among them, six (4.2% of the total) were related to APG-2449, including vomiting (n = 2), elevated fibrin D-dimer (n = 1), pulmonary embolism (n = 1), interstitial lung disease (n = 1), and diabetes mellitus (n = 1). Except for interstitial lung disease and elevated fibrin D-dimer, all events fully resolved. There were no treatment-related deaths; one patient died with interstitial lung disease. Treatment-emergent adverse events (TEAEs) were tabulated as Supplementary Table S3. Common (≥3.5%) grade ≥ 3 TEAEs included ALT increased, QT interval prolonged, pneumonia, and disease progression.

Table 2.

Treatment-related adverse events in all enrolled patients.

Category, n (%)	150 mg (n = 3)		300 mg (n = 3)		450 mg (n = 9)		600 mg (n = 8)		750 mg (n = 13)		900 mg (n = 14)		1200 mg (n = 76)		1500 mg (n = 18)		Total (N = 144)
	Any grade	Grade ≥ 3	Any grade	Grade ≥ 3	Any grade	Grade ≥ 3	Any grade	Grade ≥ 3	Any grade	Grade ≥ 3	Any grade	Grade ≥ 3	Any grade	Grade ≥ 3	Any grade	Grade ≥ 3	Any grade	Grade ≥ 3
Any TRAE	2 (66.7)	0	2 (66.7)	1 (33.3)	6 (66.7)	0	8 (100.0)	2 (25.0)	11 (84.6)	1 (7.7)	14 (100.0)	1 (7.1)	72 (94.7)	13 (17.1)	17 (94.4)	4 (22.2)	132 (91.7)	22 (15.3)
Serious TRAE	0	–	1 (33.3)	–	0	–	1 (12.5)	–	1 (7.7)	–	0	–	2 (2.6)	–	1 (5.6)	–	6 (4.2)	–
Leading to dose interruption	0	–	1 (33.3)	–	0	–	1 (12.5)	–	1 (7.7)	–	1 (7.1)	–	18 (23.7)	–	6 (33.3)	–	28 (19.4)	–
Leading to dose reduction	0	–	0	–	0	–	0	–	1 (7.7)	–	1 (7.1)	–	6 (7.9)	–	1 (5.6)	–	9 (6.3)	–
Leading to discontinued treatment	0	–	0	–	0	–	1 (12.5)a	–	0	–	0	–	0	–	0	–	1 (0.7)	–
Leading to death	0	–	0	–	0	–	1 (12.5)a	–	0	–	0	–	0	–	0	–	1 (0.7)	–
Grade ≥ 3 TRAE or grade 1 or 2 TRAE with ≥5% occurrence
Blood creatinine increased	1 (33.3)	0	0	0	4 (44.4)	0	4 (50.0)	0	3 (23.1)	0	7 (50.0)	0	47 (61.8)	0	8 (44.4)	0	74 (51.4)	0
ALT increased	1 (33.3)	0	0	0	4 (44.4)	0	3 (37.5)	1 (12.5)	2 (15.4)	0	3 (21.4)	0	47 (61.8)	4 (5.3)	5 (27.8)	0	65 (45.1)	5 (3.5)
AST increased	1 (33.3)	0	0	0	3 (33.3)	0	2 (25.0)	0	3 (23.1)	0	2 (14.3)	0	40 (52.6)	1 (1.3)	4 (22.2)	0	55 (38.2)	1 (0.7)
Nausea	1 (33.3)	0	1 (33.3)	0	1 (11.1)	0	2 (25.0)	0	2 (15.4)	0	8 (57.1)	0	20 (26.3)	0	6 (33.3)	0	41 (28.5)	0
Vomiting	0	0	1 (33.3)	1 (33.3)	0	0	3 (37.5)	0	2 (15.4)	0	4 (28.6)	0	13 (17.1)	0	11 (61.1)	1 (5.6)	34 (23.6)	2 (1.4)
WBC count decreased	0	0	0	0	1 (11.1)	0	1 (12.5)	0	2 (15.4)	0	1 (7.1)	0	21 (27.6)	0	7 (38.9)	1 (5.6)	33 (22.9)	1 (0.7)
Diarrhoea	0	0	0	0	0	0	0	0	4 (30.8)	0	3 (21.4)	0	20 (26.3)	0	5 (27.8)	0	32 (22.2)	0
Neutrophil count decreased	0	0	0	0	0	0	1 (12.5)	0	1 (7.7)	0	3 (21.4)	1 (7.1)	17 (22.4)	0	5 (27.8)	0	27 (18.8)	1 (0.7)
Rash	1 (33.3)	0	0	0	2 (22.2)	0	0	0	2 (15.4)	0	2 (14.3)	0	9 (11.8)	0	2 (11.1)	0	18 (12.5)	0
Decreased appetite	0	0	0	0	0	0	0	0	0	0	3 (21.4)	0	9 (11.8)	0	2 (11.1)	0	14 (9.7)	0
Hyperuricaemia	0	0	0	0	0	0	0	0	1 (7.7)	0	1 (7.1)	0	11 (14.5)	0	0	0	13 (9.0)	0
Hypertriglyceridaemia	0	0	0	0	0	0	0	0	0	0	2 (14.3)	0	8 (10.5)	0	2 (11.1)	0	12 (8.3)	0
Hyponatraemia	0	0	1 (33.3)	0	0	0	0	0	1 (7.7)	0	1 (7.1)	0	6 (7.9)	0	3 (16.7)	0	12 (8.3)	0
Anaemia	0	0	0	0	0	0	0	0	1 (7.7)	0	1 (7.1)	0	6 (7.9)	1 (1.3)	4 (22.2)	0	12 (8.3)	1 (0.7)
Hyperglycaemia	0	0	0	0	0	0	0	0	2 (15.4)	0	0	0	4 (5.3)	0	5 (27.8)	1 (5.6)	11 (7.6)	1 (0.7)
Pruritus	0	0	0	0	1 (11.1)	0	1 (12.5)	0	1 (7.7)	0	2 (14.3)	0	5 (6.6)	0	1 (5.6)	0	11 (7.6)	0
Electrocardiogram QT prolonged	0	0	0	0	0	0	0	0	0	0	0	0	9 (11.8)	4 (5.3)	2 (11.1)	1 (5.6)	11 (7.6)	5 (3.5)
Amylase increased	0	0	0	0	1 (11.1)	0	0	0	1 (7.7)	0	0	0	7 (9.2)	1 (1.3)	1 (5.6)	0	10 (6.9)	1 (0.7)
Hypoalbuminaemia	0	0	0	0	0	0	0	0	0	0	0	0	5 (6.6)	0	5 (27.8)	0	10 (6.9)	0
Blood alkaline phosphatase increased	0	0	0	0	0	0	0	0	0	0	2 (14.3)	0	4 (5.3)	0	3 (16.7)	0	9 (6.3)	0
Blood glucose increased	0	0	0	0	0	0	0	0	0	0	0	0	9 (11.8)	1 (1.3)	0	0	9 (6.3)	1 (0.7)
Blood bilirubin increased	0	0	0	0	1 (11.1)	0	0	0	1 (7.7)	0	0	0	4 (5.3)	0	2 (11.1)	0	8 (5.6)	0
γ-glutamyltransferase increased	0	0	0	0	0	0	0	0	1 (7.7)	0	1 (7.1)	0	4 (5.3)	0	2 (11.1)	0	8 (5.6)	0
Lipase increased	0	0	0	0	0	0	0	0	0	0	0	0	8 (10.5)	1 (1.3)	0	0	8 (5.6)	0
Dizziness	0	0	0	0	0	0	0	0	0	0	2 (14.3)	0	5 (6.6)	0	1 (5.6)	0	8 (5.6)	0
Hypokalaemia	0	0	1 (33.3)	1 (33.3)	0	0	0	0	1 (7.7)	0	3 (21.4)	0	2 (2.6)	0	0	0	7 (4.9)	1 (0.7)
Diabetes mellitus	0	0	0	0	0	0	0	0	0	0	0	0	2 (2.6)	1 (1.3)	0	0	2 (1.4)	1 (0.7)
Interstitial lung disease	0	0	0	0	0	0	1 (12.5)	1 (12.5)	0	0	0	0	0	0	0	0	1 (0.7)	1 (0.7)
Pulmonary embolism	0	0	0	0	0	0	0	0	1 (7.7)	1 (7.7)	0	0	0	0	0	0	1 (0.7)	1 (0.7)
Ejection fraction decreased	0	0	0	0	0	0	0	0	0	0	0	0	1 (1.3)	1 (1.3)	0	0	1 (0.7)	1 (0.7)

Abbreviations: ALT, alanine aminotransferase; AST, aspartate aminotransferase; TRAE, treatment-related adverse event; WBC, white blood cell.

^aOne patient with interstitial lung disease who received 600 mg.

After single oral administration of APG-2449 under fasting conditions, systemic exposure (maximum concentration [C_max] and area under the concentration time curve [AUC]) increased in an approximately dose-proportional manner from 150 to 600 mg but appeared to plateau from 600 to 900 mg (Fig. 2A and B, Supplementary Tables S4 and S5). On the other hand, the systemic exposure of APG-2449 under fed conditions increased in an approximately dose-proportional manner from 450 to 1500 mg, showing linear pharmacokinetics (Fig. 2C and D, Supplementary Tables S6 and S7). Compared to fasting conditions, the steady-state systemic exposure of APG-2449 was increased when it was administered with meals (Supplementary Table S8). The mean CSF concentration of APG-2449 reached 65%–166% of free plasma concentrations. An apparent linear relationship (r² = 0.79) was observed between CSF APG-2449 concentration and paired free plasma concentrations (assuming an unbound plasma fraction of 1%, based on unpublished in vitro protein binding data) (Fig. 2E).

Fig. 2.

Pharmacokinetics of APG-2449 under fasting or fed conditions. A, B Mean plasma APG-2449 concentration-time profiles following (A) single oral doses (Cycle 1 Day 1) and (B) multiple oral doses (Cycle 1 Day 28) under fasting conditions. C, D Mean plasma APG-2449 concentration-time profiles following (C) single oral doses (Cycle 1 Day 1) and (D) multiple oral doses (Cycle 1 Day 28) under fed conditions. E Correlation between APG-2449 concentration in cerebrospinal fluid (CSF) and paired systemic unbound APG-2449.

To determine the RP2D of APG-2449, we enrolled 144 patients from the dose-escalation and dose-expansion stages in the safety analysis. A total of 76 patients in the 1200-mg group and 18 in the 1500-mg group were available for analysis. At these APG-2449 doses, no DLTs or treatment-related deaths were observed, and no AEs led to treatment discontinuation. Grade ≥ 3 TRAEs, serious AEs (SAEs), and dose interruptions were less frequent in the 1200- than the 1500-mg group (Table 2).

The exposure-response analysis showed a positive trend, with a higher systemic exposure being significantly associated with greater probabilities of ORR and grade ≥ 3 TRAEs. However, respective incidences of grade ≥ 3 TRAEs and SAEs at 1500 mg were 22.2% and 5.6%, which were markedly greater than that at 1200 mg (17.1% and 2.6%, respectively). Based on the safety exposure-response analysis, APG-2449 1200 mg once daily was well tolerated, safe, and defined as the RP2D.

A total of 16 patients with ALK⁺ NSCLC previously untreated with TKIs received APG-2449, including 14 at the 1200-, one at the 150-, and one at the 600-mg dose. The ORR was 75% (12/16; 95% CI, 47.6%‒92.7%). The median (IQR) follow-up period was 20.2 (8.2–27.7) months. The median PFS was not reached. The probability of PFS at 24 months was 64.3% (95% CI, 34.3%–83.3%) (Fig. 3A). Two patients had measurable brain metastases during the screening period; intracranial PR was observed in two cases, and the intracranial ORR was 100%.

Fig. 3.

Efficacy of APG-2449 in anaplastic lymphoma kinase–positive non-small-cell lung cancer. A Efficacy for the anaplastic lymphoma kinase (ALK)-naïve group. B Waterfall plot of second-generation tyrosine kinase inhibitor (TKI)-failed ALK-positive (ALK⁺) non-small-cell lung cancer (NSCLC). Only 62 columns are shown because one patient has no target lesion. C Waterfall plot of TKI-failed ALK⁺ NSCLC in brain metastasis. Cutoff date was April 2, 2024.

Of 75 patients with ALK⁺ NSCLC with prior TKI failure, 70 had tumour assessments. The ORR was 20% (14/70; 95% CI, 11.4%–31.3%), DCR was 82.9% (58/70; 95% CI, 72.0%–90.8%), median DOR was not reached (NR) (95% CI, 3.71 months‒NR), and median PFS was 4.6 months (95% CI, 2.8–6.4).

Among 70 efficacy evaluable patients, three patients had prior failure on first-generation TKIs only, 63 patients had failure on second-generation TKIs but without third-generation and four had failure on third-generation TKIs. Tumour responses were observed only in patients with treatment failure on second-generation TKIs, in which the ORR was 22.2% (14/63; 95% CI, 12.7%–34.5%) (Fig. 3B), DCR 84.1% (53/63; 95% CI, 72.7%–92.1%), and median PFS 5.5 months (95% CI, 3.0–6.4). The median PFS was 2.6 months and 1.8 months in patients with prior treatment failure on first-generation TKIs only and on third-generation TKIs, respectively.

The ORR of 29 patients with ALK⁺ NSCLC treated with APG-2449 at RP2D and failed on second-generation ALK TKIs was 34.5%, which was higher than other dose levels (11.8%). Twenty-nine patients with PD on second-generation ALK TKI treatment had measurable brain metastases during the screening period. Intracranial PR was observed in 11 cases, and the intracranial ORR was 37.9% (Fig. 3C). Of these 29 patients, 12 were in the 1200-mg group; of them, nine achieved intracranial PR, for an intracranial ORR of 75.0%.

A total of 26 patients with ROS1⁺ NSCLC without prior TKI treatment received APG-2449 1200 mg, of whom 24 were included in the efficacy analysis. The ORR, DOR, and median PFS values in these patients were 75.0%, 9.46 months, and 8.28 months, respectively. Best tumour changes of target lesions are depicted in Supplementary Fig. S2. Three patients had measurable brain metastases during the screening period, and no patient had intracranial tumour response. Thirteen patients with previously TKI-treated ROS1⁺ NSCLC received APG-2449, of whom 12 were included in the efficacy analysis. The ORR and median PFS values in these patients were 0 and 4.57 months, respectively. Individual case summaries of patients with TKI-failed ROS1⁺ NSCLC are tabulated (Supplementary Table S9). Four patients had measurable brain metastases during the screening period, and no intracranial tumour response was observed.

Phosphorylated FAK (pFAK) protein expression was tested in baseline tumour tissues from 43 patients with available samples, of which data from 17 patients with treatment failure on second-generation TKIs, who were dosed with APG-2449 at the RP2D and had efficacy assessments, were used to evaluate the relationship between pFAK and efficacy. Although no statistically significant relationship was found with ORR (50% in patients with pFAK > median and 18% in patients without, P = 0.280), a significant correlation did emerge between PFS and pFAK expression level (P = 0.026 based on Pearson's coefficient); patients with higher pFAK expression were more likely to benefit from APG-2449 treatment (Supplementary Fig. S3). Interestingly, one patient (ID 01075) with ALK fusion loss and high pFAK expression experienced both a PR and sustained PFS of 20.9 months. Another patient (ID 01060) with ALK L1196Q also had high expression of pFAK and experienced a PR with a prolonged PFS of 22.1 months (Fig. 4A and B). A total of 38 patients (37 with efficacy data) with treatment failure on second-generation ALK inhibitors were assessed for pFAK levels in PBMCs. Four hours after APG-2449 administration on Days 1 and 28, pFAK levels in PBMCs showed a trend toward an overall decrease. Of note, in high-dose cohorts (750, 900, 1200, and 1500 mg), pFAK levels decreased more markedly, to lower than 50% of the baseline level (Fig. 4C). Moreover, the extent of decrease in pFAK levels in PBMCs was more marked in responders (Fig. 4D).

Fig. 4.

Pharmacodynamics of APG-2449. A, B Immunohistochemical staining of pFAK in baseline tumour tissue (A, #01060; B, #01075). C, D Change of pFAK levels in peripheral blood mononuclear cells (PBMCs) from patients with second-generation anaplastic lymphoma kinase (ALK) tyrosine kinase inhibitor (TKI)–resistant non-small-cell lung cancer. Patients were grouped per dose (C) and treatment response (D). Data are shown as mean with SEM. E Oncoplot of TKI-failed ALK-positive NSCLC in the 1200-mg group. Five patients had single ALK mutations (#01060: L1196Q; #03003: C1156F; #09002: G1202R; #02020: L1196M; #01081: L1196M). Patient #01062 had compound ALK mutations (F1174L, G1269A, D1203N, G1202R, L1122V, and C1156Y). Abbreviation: C1D1, Cycle 1 Day 1.

In 29 patients with TKI-failed ALK⁺ NSCLC treated with APG-2449 at RP2D, best responses and gene alterations are shown in Fig. 4E. Of them, one case (subject ID 01062) with ALK compound mutations (p.F1174L; p.G1269A; p.D1203N; p.G1202R; p.L1122V; p.C1156Y), and six cases (subject IDs 02015, 03006, 01064, 02023, 01090, and 02020) with activation of bypass and/or downstream signalling pathways (FGFR3/FGFR4/ERBB2/KRAS/BRAF/PIK3CA)⁵ showed poor responses to APG-2449 (ORR was 0%; Supplementary Table S10). Of the remaining 22 patients, 10 achieved partial response; 11 had stable disease; and the ORR was 45.5% (95% CI, 24.4%–67.8%). The median DOR was not reached, and the median PFS was 13.6 months. A total of five of 14 patients who had previously received alectinib achieved partial response and eight had stable disease, with a median PFS of 8.3 months. Of 14 patients with brain metastases, seven achieved partial response and seven had stable disease.

In patients with ROS1⁺ NSCLC previously untreated with TKIs, six had TP53 mutations, of whom four achieved partial response, for an ORR of 66.7% (Supplementary Fig. S2). In patients with TKI-treated ROS1⁺ NSCLC, eight (61.5%) of 13 had loss of ROS1 gene fusion.

Discussion

Patients with ALK-ROS1 fusion–positive NSCLC, especially after progression on TKIs, have urgent unmet needs, resulting in rigorous clinical and translational research in recent years. We report favourable safety and pharmacokinetic profiles, along with promising antitumour activity at the RP2D, for APG-2449, a novel small-molecule ALK-ROS1-FAK multitarget kinase inhibitor, in a first-in-human phase 1 trial with dose-escalation and dose-expansion stages, involving patients with ALK-ROS1 fusion–positive NSCLC with or without prior TKI therapy.

The study achieved its primary objectives, including establishing the APG-2449 RP2D (1200 mg) based on combined safety and pharmacokinetic analyses. In addition, APG-2449 showed robust intracranial antitumour activity and a relatively low risk of central nervous system (CNS) toxicity. A putative mechanism of action for APG-2449 via downregulation of the FAK signalling pathway emerges from the pharmacodynamic finding that TKI-treated patients with higher baseline pFAK expression were more likely to derive clinical benefit, as evidenced by a longer median PFS. The ORR of 75% in patients with previously TKI-untreated disease in our study is slightly lower than that reported with lorlatinib (81%).⁶ However, in the second-generation TKI–resistant treatment group, the ORR of APG-2449 (at RP2D) was 34.5%, which is similar to that of lorlatinib in a similar population (32.1%).⁷ Impenetrability of the blood-brain barrier renders CNS metastases unaffected by most systemic anticancer treatments, and even small-molecule drugs⁷ have a low rate of diffusion across the barrier. Fortunately, second-generation and third-generation ALK inhibitors have demonstrated improved outcomes in controlling intracranial diseases. For context, intracranial response rates of lorlatinib in first-line patients and those with second-generation ALK resistance and measurable CNS lesions were 82% and 38.7%, respectively.^6,7

The CSF/free plasma ratios of APG-2449 indicated that APG-2449 penetrates the blood-brain barrier. In our study, patients who had NSCLC with second-generation ALK TKI resistance and measurable intracranial lesions had an intracranial ORR of 75.0% with APG-2449 at RP2D, indicating that the agent exerted significant intracranial activity in these patients. APG-2449 was generally well tolerated, with no DLTs observed in 144 patients. As with other ALK TKIs, the most common AEs caused by APG-2449 were increased transaminase levels and gastrointestinal untoward effects (nausea, vomiting, diarrhoea).⁸ Most of these cases were mild and improved after a short treatment time. The incidence of dose reduction and discontinuation due to AEs was lower than with ceritinib.⁹ Although APG-2449 resulted in increased serum creatinine, this effect—similar to that observed with some other second-generation ALK TKIs—was only slightly higher than that seen with ceritinib (10%–18%).^10,11 All of the increased serum creatinine cases were grade 1 or 2, without worsening or apparent clinical implications. We considered that this effect was not due to direct nephrotoxicity, but perhaps an effect on renal transporters similar to other ALK TKIs.^12,13 Of particular interest, common CNS AEs such as peripheral neuropathy as well as cognitive and emotional effects were infrequent or rare in patients treated with APG-2449. In contrast, 35%–42% of patients in the lorlatinib group experienced all-cause CNS AEs, including principally cognitive and emotional effects.^6,14

For patients with previously TKI-untreated ROS1-positive NSCLC, APG-2449 resulted in a promising ORR of 75%, which is similar to data reported in a phase 1/2 trial of ROS1 TKI repotrectinib in patients with advanced, ROS1 mutation–positive NSCLC (79%).¹⁵ However, the suboptimal PFS with APG-2449 in this small patient subset may be partly explained by a high proportion of TP53 mutations.¹⁶ APG-2449 had no effect in this small sample of patients with previously TKI-treated ROS1⁺ NSCLC, probably because most of these patients had lost the target-ROS1 fusion gene. In addition, APG-2449 demonstrated similar efficacy to ceritinib in ROS1-positive drug-resistant models in preclinical research and showed poor activity in secondary mutations of ROS1 acquired resistance.²

Our pharmacodynamic data suggest a putative mechanism for APG-2449. FAK is a nonreceptor protein tyrosine kinase, and its autophosphorylation at tyrosine residue 397 culminates in activation and initiation of downstream signalling cascades. These signal transduction mechanisms in turn regulate fundamental cellular processes such as adhesion, migration, proliferation, and survival.^17,18 FAK overexpression is common in many human malignancies and is associated with tumour progression, metastasis, and poor patient outcomes.19, 20, 21 Moreover, FAK activation is also linked to drug resistance in some cancers, especially lung cancer.^1,22,23

Several FAK inhibitors have shown promising results in animal models and are now under clinical development.^24,25 In the present study, with small sample size detected for patients with NSCLC resistant to second-generation ALK inhibitors, both PFS and ORR seem to be improved in patients with high pFAK expression, with PFS showing statistical significance and ORR exhibiting a positive, albeit nonsignificant trend. The results indicated that patients with high pFAK expression were more likely to benefit from treatment with APG-2449. Furthermore, the decrease in pFAK expression level in PBMCs after APG-2449 administration was more pronounced in patients with partial response compared to progressive disease. Taken together, these data suggest that reducing pFAK levels may be one mechanism by which APG-2449 overcomes second-generation ALK inhibitor resistance in NSCLC. Similar research found that FAK signalling can promote residual disease in lung cancer, and inhibition of FAK combined with the primary targeted therapy suppressed residual drug-tolerant cells and enhanced tumour responses.¹

This study should be considered in light of potential limitations. First, this was a single-arm phase 1 trial. Study findings should be validated in larger, prospective, randomised controlled trials. Second, the analysis of biomarkers was retrospective, and the sample size was not large. Prospective trials are needed to examine the relationship between pFAK expression and changes in the efficacy of APG-2449. Third and finally, all study subjects were Chinese. Consequently, it is not possible to assess the effects of race or ethnicity on the findings.

In summary, this phase 1 study demonstrated that APG-2449 was generally well tolerated (with a benign CNS safety profile), had a favourable pharmacokinetic profile, and exhibited encouraging antitumour effects in patients with NSCLC either untreated with TKIs or resistant to second-generation ALK inhibitors, especially individuals with brain metastases. These promising data warrant further phase 2 and 3 clinical trials. Biomarker results revealed that pFAK expression levels may be correlated with improved APG-2449 treatment responses in patients with NSCLC resistant to second-generation ALK inhibitors.

Contributors

Y. Ma: Investigation, resources, project administration, writing and reviewing manuscript.

Z. Song: Investigation, resources, writing and reviewing manuscript.

J. Chen: Investigation, resources, writing and reviewing manuscript.

Y. Zhao: Investigation, resources, writing and reviewing manuscript.

W. Fang: Investigation, resources, writing and reviewing manuscript.

Y. Guo: Investigation, resources, writing and reviewing manuscript.

Y. Dong: Investigation, resources, writing and reviewing manuscript.

Y. Yang: Investigation, resources, writing and reviewing manuscript.

G. Wu: Investigation, resources, writing and reviewing manuscript.

J. Fang: Investigation, resources, writing and reviewing manuscript.

X. Lin: Investigation, resources, writing and reviewing manuscript.

J. Li: Investigation, resources, writing and reviewing manuscript.

Y. Huang: Investigation, resources, writing and reviewing manuscript.

Yuanyuan Zhao: Investigation, resources, writing and reviewing manuscript.

S. Hong: Investigation, resources, writing and reviewing manuscript.

J. Xue: Investigation, resources, writing and reviewing manuscript.

Y. Zhang: Investigation, resources, writing and reviewing manuscript.

Q. Liu: Investigation, resources, writing and reviewing manuscript.

C. Yang: Medical monitor, writing and reviewing manuscript.

L. Xu: Clinical data analyses, writing and reviewing manuscript.

Yun Yang: Pharmacokinetic analyses, writing and reviewing manuscript.

D. Xiong: Biomarker analyses, writing and reviewing manuscript.

D. Yang: Conceptualisation, methodology, project administration, data curation, supervision, validation, writing and reviewing manuscript.

Y. Zhai: Biomarker analyses, pharmacokinetic analyses, conceptualisation, supervision, study design, PI and site selection, writing and reviewing manuscript.

L. Zhang: Conceptualisation, methodology, data curation, supervision, validation, writing and reviewing manuscript.

H. Zhao: Conceptualisation, methodology, data curation, supervision, validation, writing and reviewing manuscript.

All authors have directly accessed and verified the underlying data reported, have read and approved the final manuscript, and accept responsibility to submit it for publication.

Data sharing statement

Individual-level data used in this study cannot be publicly shared because of ethical approval considerations. Data may be shared after approval of a proposal by an Ascentage Pharma Group International independent review committee. Sharing is subject to a confidentiality agreement and intellectual property review. Data will be available 9–36 months after publication. To request data related to this study, please contact the corresponding authors of this report.

Declaration of interests

Drs. C. Yang, L. Xu, Y. Yang, and D. Xiong are employed by Ascentage Pharma (Suzhou) Co., Ltd and are stockholders of its parent company, Ascentage Pharma Group International. Dr. Y. Zhai is employed by Ascentage Pharma (Suzhou) Co., Ltd. and Ascentage Pharma Group Inc. and holds a leadership position in these companies. Dr D. Yang is employed by Ascentage Pharma (Suzhou) Co., Ltd. and Ascentage Pharma Group International and holds a leadership position in these companies. Drs Y. Zhai and D. Yang are stockholders in Ascentage Pharma Group International. Dr. L. Zhang reports institutional and personal research grants from AstraZeneca PLC and F. Hoffmann-La Roche Ltd.; payments for speaking engagements from Akeso Biopharma, Inc. and Sichuan Biokin Pharmaceutical Co., Ltd.; and paid board positions as Trial Chair for Akeso Biopharma, Inc., CSPC Pharmaceutical Group Ltd., Jiangsu Hengrui Medicine Co., Ltd., Sichuan Kelun Pharmaceutical Co., Ltd., Novartis AG, Pierre Fabre S.A., Pfizer Inc., and Qilu Pharmaceutical Co., Ltd. All other authors have no conflicts of interest to declare.