Efficacy and Safety of Taletrectinib in Chinese Patients With ROS1+ Non–Small Cell Lung Cancer: The Phase II TRUST-I Study

Abstract

PURPOSE

Taletrectinib, a highly potent, CNS-active, ROS1 tyrosine kinase inhibitor (TKI), has demonstrated high and durable response rates, high intracranial objective response rate (ORR), prolonged progression-free survival (PFS), and activity against G2032R with a favorable safety profile. We report outcomes from the pivotal TRUST-I study (ClinicalTrials.gov identifier: NCT04395677) of taletrectinib for ROS1+ non–small cell lung cancer in China.

METHODS

TRUST-I evaluated TKI-naїve and crizotinib-pretreated patients. The primary end point was confirmed ORR (cORR) by independent review committee; key secondary end points included duration of response (DOR), PFS, and safety.

RESULTS

As of November 2023, 173 patients were enrolled (median age, 55 years; 58% female; 73% never smoked; TKI naїve: n = 106; crizotinib pretreated: n = 67). In TKI-naїve patients, cORR and intracranial cORR were 91% and 88%, respectively, and 52% and 73% in crizotinib-pretreated patients. In TKI-naїve patients, median DOR and median PFS were not reached (NR) with 22.1-month and 23.5-month follow-up, respectively. In crizotinib-pretreated patients, the median DOR was 10.6 months (95% CI, 6.3 months to NR; 8.4-month follow-up), and the median PFS was 7.6 months (95% CI, 5.5 to 12.0 months; 9.7-month follow-up). Eight of 12 patients (67%) with G2032R mutations responded. The most frequent treatment-emergent adverse events (TEAEs) were increased AST (76%), diarrhea (70%), and increased ALT (68%), most of which were grade 1-2. Incidences of neurologic TEAEs were low (dizziness: 23%; dysgeusia: 10%) and mostly grade 1. Discontinuations (5%) and dose reductions (19%) due to TEAEs were low.

CONCLUSION

Taletrectinib continues to show high and durable overall responses, prolonged PFS, robust activity against intracranial lesions and acquired resistance mutations including G2032R, and a favorable safety profile with a low incidence of neurologic TEAEs.

The efficacy and safety profile of taletrectinib adds an emerging option to the NSCLC ROS1 TKI therapeutic landscape

INTRODUCTION

Tumors positive for proto-oncogene tyrosine protein kinase-1 (ROS1) gene fusions account for 0.9%-2.6% of non–small cell lung cancers (NSCLCs); most are lung adenocarcinomas.¹ In ROS1-rearranged (ROS1+) NSCLC, gene rearrangements result in constitutive activation of the ROS1 kinase domain and downstream activation of various signaling pathways related to cell differentiation, proliferation, and survival.¹ ROS1 rearrangements in NSCLC are generally mutually exclusive, as newly diagnosed patients typically test negative for other known oncogenic alterations such as ALK rearrangements and EGFR mutations.¹ Patients with ROS1+ NSCLC tend to be female, younger than patients with NSCLC without driver mutations, and most are never smokers.¹

CONTEXT

• Key Objective

• To examine outcomes in ROS1+ non–small cell lung cancer (NSCLC) from the pivotal TRUST-I study (ClinicalTrials.gov identifier: NCT04395677) of taletrectinib, a highly potent, CNS-active, ROS1 tyrosine kinase inhibitor (TKI).

• Knowledge Generated

• Taletrectinib showed high and durable overall responses, prolonged progression-free survival, robust activity against intracranial lesions and acquired resistance mutations including G2032R, and a favorable safety profile with low incidence of neurologic treatment-emergent adverse events. The efficacy and safety profile of taletrectinib observed in this study adds a new emerging option to the current ROS1 TKI therapeutic landscape.

• Relevance (T.E. Stinchcombe)

• Taletrectinib demonstrated activity in ROS1-rearranged NSCLC, and is a potential new therapy for this patient population.*

• *Relevance section written by JCO Associate Editor Thomas E. Stinchcombe, MD.

ROS1 tyrosine kinase inhibitors (TKIs) are current standard of care for ROS1+ NSCLC.² Crizotinib was the first TKI approved in multiple countries for treatment of metastatic ROS1+ NSCLC.³ Despite initial responses (objective response rate [ORR], 72%)⁴ to first-line crizotinib, most patients eventually relapse within 2 years because of emergence of acquired resistance mutations (ie, G2032R) or brain metastases.¹ Crizotinib has limited CNS penetration. As high as 50% of crizotinib-treated patients with ROS1+ NSCLC develop brain metastases within 24 months of therapy.^5,6 Entrectinib, the second ROS1 inhibitor approved for ROS1+ NSCLC in the United States,⁷ Europe, and Japan has intracranial activity⁸ but is ineffective against G2032R resistance mutations.^5,9-11 As such, entrectinib provides a first-line option for patients with ROS1+ NSCLC who have or are likely to develop brain metastases. However, improved intracranial activity came with the cost of higher neurologic adverse events (AEs), with patients experiencing dysgeusia (44%) and dizziness (38%). Repotrectinib, recently US Food and Drug Administration (FDA)–approved for locally advanced or metastatic ROS1+ NSCLC,¹² is CNS-active and designed to inhibit G2032R mutations.^8,13 Although repotrectinib showed activity against G2032R, high incidence of neurologic AEs such as dizziness (any grade, 62%; grade ≥3, 3%), dysgeusia (any grade, 53%; grade ≥3, 0%), peripheral neuropathy (any grade, 34%; grade ≥3, 1%), dyspnea (any grade, 27%; grade ≥3, 6%), ataxia (any grade, 21%; grade ≥3, <1%), fatigue (any grade, 22%; grade ≥3, 1%), muscular weakness (any grade, 20%; grade ≥3, 2%), and memory impairment (any grade, 15%; grade ≥3, <1%) likely affect patient quality of life.^12,14 Overall, unmet treatment needs remain for ROS1+ NSCLC to optimize disease control and tolerability.

Taletrectinib (AB106/DS-6051b) is a highly potent, CNS-active, ROS1 TKI.¹⁵ In cell-based studies, taletrectinib has low nanomolar potency against ROS1 fusions including G2032R and other acquired resistance mutations.¹⁶ In vitro, taletrectinib shows enzymatic selectivity for ROS1 (wild type and resistant mutations) over TRKB,^17,18 demonstrating a 20-fold decrease in half-maximal inhibitory concentration (IC₅₀) for ROS1 over TRKA and TRKB (0.07 nM v 1.26 nM and 1.47 nM, respectively) and a two-fold decrease over TRKC (0.18 nM).¹⁵ Taletrectinib has excellent blood-brain barrier penetration and promising preclinical intracranial activity.^15,17 In an orthotopic CNS model of ROS1+ NSCLC, taletrectinib showed sustained brain penetration and improved survival versus vehicle or repotrectinib in rodents with intracranial patient-derived xenograft tumors.^15,17,18 In two phase I studies and interim analysis of TRUST-I, taletrectinib demonstrated clinically meaningful efficacy and a favorable safety profile.^17,18

Here, we report outcomes from the pivotal phase II TRUST-I study of taletrectinib monotherapy in Chinese patients with ROS1+ NSCLC.

METHODS

Study Design and Treatment

TRUST-I (ClinicalTrials.gov identifier: NCT04395677) is a phase II, multicenter, single-arm study of open-label taletrectinib treatment for advanced ROS1+ NSCLC who were either TKI-naїve or crizotinib-pretreated. A safety lead-in stage evaluated the safety, tolerability, and pharmacokinetics of taletrectinib in six patients at two dose levels (400 mg once daily and 600 mg once daily) to determine the clinical optimal dose. Three patients were treated at 400 mg once daily, two of whom escalated to 600 mg once daily. The tolerability of the recommended phase II dose established during phase I studies (600 mg once daily) was confirmed during the safety lead-in.^18,19 In stage II, the safety and efficacy was evaluated at 600 mg once daily. Taletrectinib was administered orally once daily in 21-day treatment cycles.

Patients

Eligible patients were ≥18 years old with histologically/cytologically confirmed locally advanced or metastatic NSCLC, measurable disease per RECIST version 1.1, and locally documented evidence of ROS1 fusion in tumor tissue (detected by polymerase chain reaction [PCR], next-generation sequencing [NGS], fluorescence in situ hybridization, or immunohistochemistry). Crizotinib-pretreated patients must have experienced treatment failure (disease progression or intolerable side effects) on or after treatment. Up to two lines of prior anticancer therapy for advanced disease were allowed. Brain metastases were allowed if stable without requiring an increased glucocorticoid dose to control CNS symptoms within 2 weeks before screening. No prior therapy, including radiotherapy, was required. Other inclusion criteria included Eastern Cooperative Oncology Group performance status (ECOG PS) of 0 or 1, adequate renal and hepatic function, adequate bone marrow function, and normal QT interval by electrocardiogram.

Patients must not have received: immune checkpoint inhibitors (including PD-1 or PD-L1 monoclonal antibodies) within 12 weeks before starting taletrectinib, expected use of any other antitumor drugs within 14 days before the first dose of taletrectinib and/or during the study, and prior treatment in another interventional clinical trial within 4 weeks before the first dose of taletrectinib. Patients who had major surgery or radiotherapy within 1 month before enrollment (except palliative radiotherapy for bone metastasis pain or radiotherapy for brain metastases 2 weeks before enrollment) or were scheduled for major surgery during the study were also excluded. Other exclusions included ongoing or history of interstitial lung disease or interstitial fibrosis; history of other malignancy (except radically treated nonmelanoma skin cancer, in situ cervical cancer, and prostatic intraepithelial neoplasia); history of stem cell or organ transplantation; and systemic use of potent inhibitors of CYP3A4 within 28 days before starting taletrectinib and/or during the study.

Assessments

Disease response and progression were evaluated per RECIST version 1.1.²⁰ Contrast-enhanced computed tomography or contrast-enhanced magnetic resonance imaging of chest, head, abdomen, and pelvis was performed at baseline, during treatment, and for patients with baseline brain metastases. After treatment, radiologic tumor assessments were performed every 6 weeks until week 24, every 9 weeks until week 78, and then every 12 weeks until progression, loss of clinical benefit, initiation of new anticancer therapy, death, withdrawal, or end of follow-up, whichever comes first. Patients with baseline brain metastases were evaluated for intracranial metastases at the above time points.²¹

Safety assessments included physical and laboratory examinations, vital signs, electrocardiograms, and AEs (coded using Medical Dictionary for Regulatory Activities version 26.1 and graded per National Cancer Institute Common Terminology Criteria for Adverse Events, version 5.0).

Archived or freshly collected tumor tissue samples were analyzed retrospectively at a central laboratory by AmoyDx Essential NGS Panel or PCR-based AmoyDx ROS1 Gene Fusion Detection Kit to confirm ROS1 status, fusion partners, and other biomarkers. Circulating cell-free DNA isolated from anticoagulated peripheral whole-blood specimens was collected at baseline and during treatment and analyzed by Haplox HapOncoCDx-107 panel to detect tumor mutations and rearrangements. Fresh tumor tissue samples after treatment failure were preferred in crizotinib-pretreated patients.

End Points

The primary endpoint was independent review committee (IRC)–assessed confirmed ORR (cORR) per RECIST 1.1. Secondary efficacy end points included investigator-assessed ORR, progression-free survival (PFS), disease control rate (DCR), duration of response (DOR), time to response (TTR), and time to progression per RECIST 1.1. For patients with measurable baseline intracranial metastases, IRC-assessed intracranial ORR and DCR of their brain lesions were also assessed. Safety was a secondary end point and included incidence, severity, duration, and study drug relatedness of AEs, serious AEs, and AEs of special interest, defined as combined elevation of ALT, AST, and total bilirubin.

Statistical Analysis

The primary end point (cORR) was defined as the proportion of IRC-assessed confirmed (≥4 weeks from initial documentation) complete response (CR) or partial response (PR) per RECIST v1.1; 95% CIs were calculated by the Clopper-Pearson method. DOR was defined as time from first documented CR/PR to first documented PD for patients with confirmed objective responses; patients not experiencing PD were censored at the time of last tumor assessment or death. PFS was defined as time from first taletrectinib dose to first PD or death due to any cause. Patients not experiencing PD or death who started new anticancer therapy were censored at last tumor assessment; PFS was considered to have occurred on the date of death in patients who died before postbaseline assessment. The Kaplan-Meier method estimated PFS and DOR time curves; median duration of PFS and DOR were reported with corresponding Clopper-Pearson 95% CIs.

For TKI-naїve patients, ORR >70% by the exact probability method of binomial distribution was required to declare a positive result of taletrectinib treatment; for crizotinib-pretreated patients, an ORR >20% was required. Enrollment of approximately 60 TKI-naїve patients was planned to ensure enrollment of 54 patients to control for type I errors at α = .025 (one-sided); there were no formal statistical hypotheses for sample size calculation of crizotinib-pretreated patients.

Safety was analyzed in participants who received ≥one dose of study drug during the on-treatment period, defined as first dose to 30 days after the last dose of taletrectinib or start date of new anticancer therapy minus 1 day, whichever occurred first. Treatment-emergent adverse events (TEAEs) were considered related to treatment if investigators judged them definitely related, probably related, or possibly related.

Study Oversight

This study was conducted in accordance with international consensus ethical principles, including the Declaration of Helsinki and Council for International Organizations of Medical Sciences and applicable ICH Good Clinical Practice and local regulations. The protocol was approved by institutional review boards and independent ethics committees, according to the practice at each participating trial site. All patients provided written informed consent.

RESULTS

Patient Characteristics

As of November 29, 2023, 173 Chinese patients (106 TKI naїve; 67 crizotinib-pretreated) received taletrectinib (Fig 1). At baseline, the median age was 55.0 years, and 57.8% of patients were female (Table 1). Most patients (73.4%) had never smoked. In TKI-naїve patients, 18.9% of patients had an ECOG PS of 0, 91.5% had stage IV disease, and 18 (17.0%) had brain metastases of which eight (7.5%) were measurable. Twenty two (20.8%) patients had prior chemotherapy. In crizotinib-pretreated patients, 28.4% had an ECOG PS of 0, 65 (97.0%) had stage IV disease, and 28 (41.8%) had brain metastases of which 15 (22.4%) were measurable. Twenty-three (34.3%) patients received prior chemotherapy.

FIG 1.

Treatment disposition flow diagram. TKI, tyrosine kinase inhibitor.

TABLE 1.

Baseline Demographic and Disease Characteristics

Category	TKI Naїve (n = 106)	Crizotinib Pretreated (n = 67)	Overall (N = 173)
Age, years, median (min-max)	56.0 (26-78)	51.0 (31-77)	55.0 (26-78)
Sex, No. (%)
Male	47 (44.3)	26 (38.8)	73 (42.2)
Female	59 (55.7)	41 (61.2)	100 (57.8)
Smoking status, No. (%)
Never smoker	78 (73.6)	49 (73.1)	127 (73.4)
Former smoker	22 (20.8)	17 (25.4)	39 (22.5)
Current smoker	6 (5.7)	1 (1.5)	7 (4.0)
ECOG PS, No. (%)
0	20 (18.9)	19 (28.4)	39 (22.5)
1	86 (81.1)	48 (71.6)	134 (77.5)
Disease stage at enrollment, No. (%)
III	9 (8.5)	2 (3.0)	11 (6.4)
IV	97 (91.5)	65 (97.0)	162 (93.6)
Histology, No. (%)
Adenocarcinoma	102 (96.2)	62 (92.5)	164 (94.8)
Lines of prior anticancer therapy, No. (%)
0	83 (78.3)	0	83 (48.0)
1-2	23 (21.7)	66 (98.5)	89 (51.4)
≥3	0	1 (1.5)	1 (0.6)
Prior anticancer chemotherapy, No. (%)
Yes	22 (20.8)	23 (34.3)	45 (26.0)
No	84 (79.2)	44 (65.7)	128 (74.0)
Brain metastasis status by IRC (RANO-BM), No. (%)
No	88 (83.0)	39 (58.2)	127 (73.4)
Yes	18 (17.0)	28 (41.8)	46 (26.6)
Measurable	8 (7.5)	15 (22.4)	23 (13.3)
Nonmeasurable	10 (9.4)	13 (19.4)	23 (13.3)
Method of ROS1 fusion detection, No. (%)
PCR	48 (45.3)	28 (41.8)	76 (43.9)
NGS	57 (53.8)	39 (58.2)	96 (55.5)
Immunohistochemistry	1 (0.9)	0	1 (0.6)

Abbreviations: ECOG PS, Eastern Cooperative Oncology Group performance status; IRC, independent review committee; NGS, next-generation sequencing; PCR, polymerase chain reaction; RANO-BM, response assessment in neuro-oncology brain metastases; TKI, tyrosine kinase inhibitor.

Efficacy in Response-Evaluable Patients

Among 106 TKI-naїve patients, IRC-assessed cORR was 90.6% (95% CI, 83.33 to 95.38) with four (3.8%) patients achieving CR and 92 (86.8%) achieving PR (Table 2; Figs 2A and 2C). The DCR was 95.3% (95% CI, 89.33 to 98.45), and the median TTR was 1.4 months (95% CI, 1.38 to 1.41). With a median follow-up of 22.1 months and 23.5 months, neither median DOR (Fig 2E) or median PFS (Fig 2F) were reached; the 24-month DOR and PFS were 78.6% (95% CI, 66.86 to 86.56) and 70.5% (95% CI, 59.17 to 79.16), respectively.

TABLE 2.

Summary of IRC-Based Confirmed Best Overall Tumor Response Among Response-Evaluable Patients

Category	TKI Naїve (n = 106)	Crizotinib Pretreated (n = 66)a
Best overall response, No. (%)
CR	4 (3.8)	0
PR	92 (86.8)	34 (51.5)
SD	5 (4.7)	21 (31.8)
PD	3 (2.8)	5 (7.6)
Not evaluable	2 (1.9)	6 (9.1)
ORR (CR + PR), No. (%)	96 (90.6)	34 (51.5)
Two sided 95% CI	83.33 to 95.38	38.88 to 64.01
DCR (CR + PR + SD), No. (%)	101 (95.3)	55 (83.3)
Two-sided 95% CI	89.33 to 98.45	72.13 to 91.38
Patients with measurable baseline brain metastases (RANO-BM)	n = 8	n = 15
Best overall response, No. (%)
CR	0	0
PR	7 (87.5)	11 (73.3)
SD	1 (12.5)	3 (20.0)
PD	0	0
Not evaluable	0	1 (6.7)
ORR (CR + PR), No. (%)	7 (87.5)	11 (73.3)
Two-sided 95% CIa	47.35 to 99.68	44.90 to 92.21
DCR (CR + PR + SD), no. (%)	8 (100)	14 (93.3)
Two-sided 95% CIb	63.06 to 100.00	68.05 to 99.83

Abbreviations: CR, complete response; DCR, disease control rate; IRC, independent review committee; ORR, objective response rate; PD, progressive disease; PR, partial response; RANO-BM, response assessment in neuro-oncology brain metastases; SD, stable disease; TKI, tyrosine kinase inhibitor.

^a One patient was excluded from the response-evaluable population in the crizotinib-pretreated group because of the presence of secondary cancer.

^b Clopper and Pearson method.

FIG 2.

Responses by IRC assessment of taletrectinib in TKI-naїve and crizotinib-pretreated patients with ROS1+ NSCLC. Best percent change in the sum of diameters from baseline in (A) TKI-naїve (n = 104) and (B) crizotinib-pretreated (n = 60) patients, (C and D) percent change in response over time, DOR and PFS in (E and F) TKI-naїve and (G and H) crizotinib-pretreated patients, and (I) best percent change in the sum of diameters from baseline in patients with baseline metastases. CR, complete response; DOR, duration of response; IRC, independent review committee; NSCLC, non–small cell lung cancer; PD, progressive disease; PFS, progression-free survival; PR, partial response; SD, stable disease; TKI, tyrosine kinase inhibitor.

Among 66 crizotinib-pretreated patients, IRC-assessed cORR was 51.5% (95% CI, 38.88 to 64.01) with 34 (51.5%) patients achieving PR (Table 2; Figs 2B and 2D). The DCR was 83.3% (95% CI, 72.13 to 91.38), and the median TTR was 1.4 months (95% CI, 1.38 to 1.41). With 8.4 months and 9.7 months of follow-up, the median DOR was 10.6 months (95% CI, 6.31 to not reached [NR]; Fig 2G), and the median PFS was 7.6 months (95% CI, 5.52 to 11.96; Fig 2H). The 9-month DOR and PFS rates were 69.8% (95% CI, 47.94 to 83.84) and 47.4% (95% CI, 33.50 to 60.02), respectively. Similar results were seen when assessed by investigators (Fig 1; Appendix Tables A1 and A2, online only).

Efficacy in Measurable Baseline Brain Metastases

In eight TKI-naїve patients with measurable baseline brain metastases (Table 2; Fig 2I), IRC-assessed intracranial cORR was 87.5% (95% CI, 47.35 to 99.68) with seven (87.5%) patients achieving PR, and the intracranial DCR was 100.0% (95% CI, 63.06 to 100.00). In 15 crizotinib-pretreated patients with measurable baseline brain metastases (Table 2; Fig 2I), intracranial cORR was 73.3% (95% CI, 44.90 to 92.21) with 11 (73.3%) patients achieving PR, and the intracranial DCR was 93.3% (95% CI, 68.05 to 99.83).

Subgroup and Translational Analyses

In both cohorts, patients had a similar, high ORR regardless of age younger than or ≥65 years, sex, presence or absence of brain metastases at baseline, presence or absence of prior anticancer chemotherapy, and smoking status (Appendix Figs A1 and A2). TKI-naїve patients (n = 106) had confirmation from a central laboratory for ROS1 fusion through tissue NGS or PCR. Among 64 patients where a ROS1 fusion partner was detected by tissue or liquid NGS, CD74 was the most common (n = 28 [44%]; Table 3). Among crizotinib-pretreated patients, 28 patients had a fresh tumor biopsy collected after treatment failure with crizotinib and before treatment with taletrectinib. Tissue-based NGS detected ROS1 rearrangements for all 28 rebiopsy samples; 21 patients were concurrently tested by liquid-based NGS. ROS1 resistance mutations were detected by either tissue or liquid NGS in 53.6% (15/28) of crizotinib-pretreated patients. Twelve patients had G2032R, one patient had L2026M, one patient had S1986F, and one patient had G2032R and G2101A. The ORR in patients with G2032R mutations was 66.7% (8/12 patients; 95% CI, 34.89 to 90.08), 60.0% (9/15) for ROS1-acquired resistance mutations and 38.5% (5/13; 95% CI, 13.86 to 68.42) for those without ROS1-acquired resistance mutations (Table 3).

TABLE 3.

Efficacy Outcomes Based on Molecular Status

ROS1 Fusion Partner	TKI-Naїve Patients With Detected ROS1 Fusion (n = 64), No. (%)
CD74	28 (43.8)
SDC4	12 (18.8)
EZR	11 (17.2)
SLC34A2	4 (6.3)
TPM3	2 (3.1)
CD83	1 (1.6)
LRIG1	1 (1.6)
MYH9	1 (1.6)
SDC4 and KLHDC2	1 (1.6)
SLC34A2 and ZNF608 CSNK1G3_Intergenic_Region	1 (1.6)
CCNH-TMEM161B_Intergenic_Region	1 (1.6)
TPR	1 (1.6)
ORR	Crizotinib Pretreated (n = 28), % (95% CI)
Acquired ROS1 resistance mutation (n = 15)a	60.0 (32.29 to 83.66)
No acquired ROS1 mutation (n = 13)	38.5 (13.86 to 68.42)
G2032R mutation (n = 12)	66.7 (34.89 to 90.08)

NOTE. ROS1 fusion partners at baseline in TKI-naїve patients and efficacy on the basis of freshly collected tumor tissue biomarker in crizotinib-pretreated patients.

Abbreviations: ORR, objective response rate; TKI, tyrosine kinase inhibitor.

^a Included three patients harboring G2101A and G2032R comutations, L2026M, and S1986F mutation, n = 1 each.

Safety

Among all patients, the median duration of taletrectinib exposure was 12.2 months (range, 0.2-40.0). The most common TEAEs were increased AST (n = 132, 76.3%), diarrhea (n = 121, 69.9%), increased ALT (n = 117, 67.6%), vomiting (n = 92, 53.2%), and anemia (n = 85, 49.1%; Table 4); most of these events were grade 1 and 2. Neurologic TEAEs reported in ≥5% of patients were dizziness (23.1%), dysgeusia (10.4%), headache (8.7%), and hypesthesia (7.5%), and most were grade 1 (88%). Two crizotinib-pretreated patients experienced interstitial lung disease (one grade 3 and one grade 4). TEAEs leading to dose reduction were reported in 33 (19.1%) patients, with the most common being abnormal hepatic function (4.0%), increased AST (2.9%), and diarrhea (2.9%). TEAEs leading to treatment discontinuation were reported in nine (5.2%) patients, with the most common being abnormal hepatic function (1.2%) and interstitial lung disease (1.2%). Grade 3-4 TEAEs occurred in 76 (43.9%) patients; the most common (≥5%) were increased AST (8.1%), decreased neutrophil count (5.8%), and increased ALT (5.2%). Three patients experienced a treatment-related AE per investigator leading to death: two TKI-naїve patients (one hepatic failure, one pneumonia) and one crizotinib-pretreated patient (hepatic function abnormal). The hepatic failure-related death was not considered drug-related by sponsor because more than five half-lives elapsed before the occurrence and alternative potential etiologies involving concomitant medications were deemed more likely to be the cause.

TABLE 4.

TEAEs Reported in ≥15% of Patients Treated With Taletrectinib 600 mg Once Daily (n = 173) Shown by Severity

Category	Grade 1, No. (%)	Grade 2, No. (%)	Grade 3, No. (%)	Grade 4, No. (%)	Grade 5, No. (%)	Any Grade, No. (%)
AST increased	92 (53.2)	26 (15.0)	14 (8.1)	0	0	132 (76.3)
Diarrhea	99 (57.2)	16 (9.2)	6 (3.5)	0	0	121 (69.9)
ALT increased	79 (45.7)	29 (16.8)	8 (4.6)	1 (0.6)	0	117 (67.6)
Vomiting	75 (43.4)	16 (9.2)	1 (0.6)	0	0	92 (53.2)
Anemia	52 (30.1)	30 (17.3)	3 (1.7)	0	0	85 (49.1)
Nausea	64 (37.0)	8 (4.6)	1 (0.6)	0	0	73 (42.2)
Neutrophil count decreased	25 (14.5)	10 (5.8)	6 (3.5)	4 (2.3)	0	45 (26.0)
Hepatic function abnormal	21 (12.1)	8 (4.6)	14 (8.1)	0	1 (0.6)	44 (25.4)
WBC count decreased	27 (15.6)	14 (8.1)	3 (1.7)	0	0	44 (25.4)
Blood bilirubin increased	34 (19.7)	6 (3.5)	2 (1.2)	1 (0.6)	0	43 (24.9)
Dizziness	36 (20.8)	3 (1.7)	1 (0.6)	0	0	40 (23.1)
Proteinuria	34 (19.7)	5 (2.9)	0	0	0	39 (22.5)
Weight increased	17 (9.8)	16 (9.2)	3 (1.7)	0	0	36 (20.8)
Blood creatinine increased	33 (19.1)	2 (1.2)	0	0	0	35 (20.2)
Electrocardiogram QT prolonged	26 (15.0)	4 (2.3)	5 (2.9)	0	0	35 (20.2)
Hypercholesterolemia	29 (16.8)	4 (2.3)	0	0	0	33 (19.1)
Hyperuricemia	30 (17.3)	2 (1.2)	0	0	0	32 (18.5)
Weight decreased	23 (13.3)	8 (4.6)	0	0	0	31 (17.9)
Constipation	28 (16.2)	2 (1.2)	0	0	0	30 (17.3)
Decreased appetite	26 (15.0)	3 (1.7)	0	0	0	29 (16.8)
Bilirubin conjugated increased	22 (12.7)	3 (1.7)	2 (1.2)	1 (0.6)	0	28 (16.2)
COVID-19	10 (5.8)	15 (8.7)	3 (1.7)	0	0	28 (16.2)
Pyrexia	23 (13.3)	3 (1.7)	1 (0.6)	0	0	27 (15.6)
Blood creatinine phosphokinase increased	21 (12.1)	5 (2.9)	0	0	0	26 (15.0)
Hypertriglyceridemia	24 (13.9)	2 (1.2)	0	0	0	26 (15.0)

NOTE. Worst grade per subject/systemic organ class or preferred term is reported. TEAE is defined as an AE with onset date or increase in severity level after the first dose of study drug and within 30 days after the last dose of study drug (permanent discontinuation of study treatment) or before initiation of new anticancer therapy, whichever occurs first. If an AE occurred after initiation of new anticancer therapy and assessed as related by investigator, it is considered a TEAE.

Abbreviations: AE, adverse event; TEAE, treatment emergent adverse event.

DISCUSSION

Taletrectinib, a potent, next-generation ROS1 TKI, demonstrated meaningful efficacy outcomes in both TKI-naїve and crizotinib-pretreated ROS1+ NSCLC. High and durable ORR was observed in both TKI-naїve and crizotinib-pretreated patients, high intracranial ORR regardless of line of therapy, prolonged PFS in TKI-naїve and crizotinib-pretreated patients, and activity against G2032R were also of high clinical significance.

The observed efficacy profile of taletrectinib adds a new emerging option to the ROS1 TKI therapeutic landscape. While geographically limited, to our knowledge, TRUST-I represents the largest prospective clinical trial to date conducted in patients living with ROS1+ NSCLC. The high response rate (91%) and extended DOR (24-month rate 78.6%) in TKI-naїve patients suggest that most patients responded with durable outcomes. Supporting this, IRC-assessed median PFS was NR, and 71% of patients remained progression-free at 24 months. This strong efficacy profile suggests taletrectinib may be a suitable treatment option for TKI-naїve ROS1+ NSCLC.

While crizotinib has become a transformative standard of care for ROS1+ NSCLC, resistance often develops within 2 years of initiating crizotinib because of acquired ROS1 resistance mutations or insufficient CNS penetration to treat brain metastases.^5,6,22 In crizotinib-pretreated patients, taletrectinib reported a strong ORR of 51.5% with durable outcomes (median DOR, 10.6 months; median PFS, 7.6 months) and an ORR of 66.7% in patients with G2032R mutations and 60.0% for acquired resistance overall. These findings suggest taletrectinib can overcome common mechanisms of resistance to crizotinib and provide benefit in the crizotinib-pretreated setting. Similar resistance mutation mechanisms were observed after entrectinib treatment, which is ineffective against G2032R mutations.^5,9-11 The efficacy of taletrectinib in patients receiving prior entrectinib or repotrectinib, two other FDA-approved ROS1 TKIs for metastatic ROS1+ NSCLC, was not assessed in TRUST-1. However, after 11.7 months of follow-up, all four patients receiving entrectinib in the ongoing TRUST-II trial (ClinicalTrials.gov identifier: NCT04919811) responded to taletrectinib,²³ including one CR; longer follow-up with additional patients will further examine the efficacy of taletrectinib in patients experiencing treatment failure after entrectinib or repotrectinib.

Brain metastases are a common treatment challenge for ROS1+ NSCLC. The rate of baseline brain metastases in TKI-naїve patients was lower (17.0%) than the 25%-30% observed in other trials of taletrectinib¹⁸ or other ROS1 TKIs.¹⁴ This was unexpected, as TRUST-I did not limit enrollment of these patients. However, taletrectinib induced robust intracranial responses for measurable brain metastases (TKI naive, 87.5%; crizotinib pretreated, 73.3%) regardless of prior crizotinib treatment, further indicating that taletrectinib can address disease in a common metastatic site in TKI-naїve and crizotinib-pretreated settings. ROS1 resistance mutations accounted for about 54% of evaluable patients who progressed after prior crizotinib treatment in this study. While higher than reported rates of acquired on-target resistance mutations after crizotinib treatment,^5,6 this may be because only 28 patients had fresh tumor biopsies collected for testing. Furthermore, the numerically lower ORR in patients without detectable ROS1 resistance mutations may suggest some tumors develop alternate resistance mechanisms and become ROS1 independent over time.

Taletrectinib demonstrated favorable safety for ROS1+ NSCLC. TEAEs leading to treatment discontinuations (5.2%) and dose reductions (19.1%) were low. The most common side effects were low-grade elevated transaminases, diarrhea, vomiting, and anemia, which were transient and managed by dose interruptions or dose reductions. Despite strong activity against intracranial lesions, there was a low incidence of neurologic AEs, which were mostly grade 1 and comparable with crizotinib, which has low brain penetration.⁵ In TKI-naїve patients, entrectinib had an ORR of 68% and an IC-ORR of 80% but is associated with neurologic side effects related to inhibition of TRK proteins (dysgeusia, dizziness, paresthesia) and has frequent dose reductions (34%) or dose interruptions (37%) because of cardiac events.⁸ Although repotrectinib had promising activity (TKI-naїve ORR: 79%; TKI-pretreated ORR: 38%) and was active against brain metastases and ROS1 resistance mutations, neurologic AEs such as dizziness (58%), dysgeusia (50%), and paresthesia (30%) were high.¹⁴ Previous evidence suggests TRK inhibition could, in part, be responsible for development of neurologic AEs.²⁴ The selectivity profile of taletrectinib for ROS1 over TRKB may explain the relatively lower rates of AEs such as dizziness (23.1%), dysgeusia (10.4%), and hypesthesia (7.5%) versus other ROS1 TKIs, although the relationship between these observed neurologic AEs and potential TRKC inhibition cannot be excluded. Patients receiving crizotinib, which has limited CNS penetration, reported similar rates of dizziness (16.0%) and dysgeusia (16.0%).⁴ Notably, ongoing phase III clinical trials are evaluating the safety and efficacy of crizotinib versus repotrectinib (TRIDENT 3; ClinicalTrials.gov identifier: NCT06140836) and crizotinib versus entrectinib (ClinicalTrials.gov identifier: NCT04603807) for metastatic ROS1+ NSCLC. A compound like taletrectinib, which offers robust efficacy and a favorable safety profile, could fulfill unmet medical needs for patients with ROS1+ NSCLC regardless of treatment history.

Limitations of TRUST-I include the lack of placebo or standard-of-care comparator arm and that patients were from a single geographic region. Importantly, since entrectinib is not approved in China, this trial did not have the opportunity to address the CNS activity of taletrectinib in entrectinib-pretreated patients. Taken together, the efficacy and safety findings from this study present a favorable benefit-risk profile for taletrectinib in ROS1+ NSCLC. Taletrectinib demonstrated high overall and intracranial response rates, prolonged PFS, and activity against tumors with G2032R mutations in the context of a favorable safety profile with minimal CNS-related side effects. Additional follow-up is ongoing. The ongoing pivotal, global, phase II study (TRUST-II; ClinicalTrials.gov identifier: NCT04919811) will further shed light on the generality of taletrectinib efficacy and safety profile across regions and ethnicities.

APPENDIX

TABLE A1.

Summary of Best Overall Tumor Response by Investigator Assessment Among Response-Evaluable Patients

Response	TKI Naїve (n = 106)	Crizotinib Pretreated (n = 66)
Best overall response, No. (%)
CR	3 (2.8)	1 (1.5)
PR	87 (82.1)	28 (42.4)
SD	11 (10.4)	25 (37.9)
PD	3 (2.8)	7 (10.6)
Not evaluable	2 (1.9)	5 (7.6)
Objective response rate (CR + PR), No. (%)	90 (84.9)	29 (43.9)
Two-sided 95% CIa	76.65 to 91.12	31.74 to 56.70
Disease control rate (CR + PR + SD), No. (%)	101 (95.3)	54 (81.8)
Two-sided 95% CI	89.33 to 98.45	70.39 to 90.24

Abbreviations: CR, complete response; PD, progressive disease; PR, partial response; SD, stable disease; TKI, tyrosine kinase inhibitor.

^a Clopper and Pearson method.

TABLE A2.

DOR and PFS by Investigator Assessment Among Response-Evaluable Patients

Efficacy	TKI Naїve (n = 106)	Crizotinib Pretreated (n = 66)
DOR, months, median (95% CI)	NR (28.98 to NR)	10.4 (6.24 to NR)
PFS response, months, median (95% CI)	31.8 (26.25 to NR)	7.6 (5.55 to 11.76)

Abbreviations: DOR, duration of response; NR, not reached; PFS, progression-free survival; TKI, tyrosine kinase inhibitor.

FIG A1.

Forest plots of IRC-assessed overall response rates by subgroup in (A) TKI-naїve patients and (B) crizotinib-treated patients. Dotted vertical lines represent benchmarks used to declare positive results for each patient cohort. IRC, independent review committee; ORR, objective response rate; RANO-BM, response assessment in neuro-oncology brain metastases; TKI, tyrosine kinase inhibitor.

FIG A2.

Forest plots of investigator-assessed overall response rate by subgroup in (A) TKI-naїve patients and (B) crizotinib-treated patients. Dotted vertical lines represent benchmarks used to declare positive results for each patient cohort. IRC, independent review committee; ORR, objective response rate; RANO-BM, response assessment in neuro-oncology brain metastases; TKI, tyrosine kinase inhibitor.

DATA SHARING STATEMENT

Clinical trial data can be requested by qualified researchers for use in rigorous, independent scientific research as long as the trials are not part of an ongoing or planned regulatory submission. Sharing of data is subject to protection of patient privacy and respect for the patient's informed consent. The data will be provided following review and approval of a research proposal and Statistical Analysis Plan and execution of a Data Sharing Agreement. For approved requests the data will be accessible for 12 months, with possible extensions considered. For more information on the process or to submit a request, contact trials@anhearttherapeutics.com.

AUTHOR CONTRIBUTIONS

Conception and design: Xue Meng, Jingxun Wu, Kaihua Lu, Wu Zhuang, Bing Yan, Caicun Zhou

Financial support: Jingxun Wu, Kaihua Lu, Bing Yan

Administrative support: Jingxun Wu, Xicheng Wang, Kaihua Lu, Bing Yan

Provision of study materials or patients: Wei Li, Anwen Xiong, Huijie Fan, Yanqiu Zhao, Yongsheng Wang, Jingxun Wu, Ziping Wang, Yunpeng Liu, Xicheng Wang, Xintian Qin, Kaihua Lu, Wu Zhuang, Bing Yan

Collection and assembly of data: Wei Li, Anwen Xiong, Huijie Fan, Qitao Yu, Yanqiu Zhao, Yongsheng Wang, Xue Meng, Jingxun Wu, Ziping Wang, Yunpeng Liu, Xicheng Wang, Xintian Qin, Kaihua Lu, Yizhong Ren, Bing Yan, Caicun Zhou

Data analysis and interpretation: Huijie Fan, Xue Meng, Jingxun Wu, Xicheng Wang, Kaihua Lu, Yizhong Ren, Xianyu Zhang, Bing Yan, Christine M. Lovly, Caicun Zhou

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

Efficacy and Safety of Taletrectinib in Chinese Patients With ROS1+ Non–Small Cell Lung Cancer: The Phase II TRUST-I Study

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/jco/authors/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).