New data from the ARROS-1 trial show substantial clinical activity with zidesamtinib (Jideytro®; GSK) in patients with advanced or metastatic ROS1-positive non-small cell lung cancer (NSCLC) who had not previously received a ROS1 tyrosine kinase inhibitor (TKI). At a median follow-up of 15.2 months, the objective response rate (ORR) was 94% among 94 efficacy-evaluable patients, and 90% remained progression-free at 12 months [1].
The findings, presented at the 2026 World Conference on Lung Cancer (WCLC) held September 12–15, 2026, in Seoul, South Korea, provide the clinical data GSK plans to use in seeking a U.S. label expansion for first-line treatment [1]. Zidesamtinib is currently FDA-approved for adults with locally advanced or metastatic ROS1-positive NSCLC who have previously received a ROS1 TKI [2].
The first-line indication remains investigational.
Key Takeaways
- In TKI-naive ROS1-positive NSCLC, zidesamtinib produced a 94% ORR, including a 15% complete response rate, among 94 efficacy-evaluable patients [1].
- The 12-month progression-free survival rate was 90%; median progression-free survival and median duration of response had not been reached at the time of analysis [1].
- Among 10 patients with measurable brain metastases at baseline, the intracranial ORR was 100%, with a 70% intracranial complete response rate [1].
- Zidesamtinib is currently FDA-approved only for patients with locally advanced or metastatic ROS1-positive NSCLC who have received a prior ROS1 TKI [2,3].
- GSK plans to submit a supplemental New Drug Application seeking a first-line indication in the United States [1].
A Kinase That Drives a Distinct Form of Lung Cancer
ROS1 rearrangements are found in a small subset of NSCLC, generally estimated at approximately 1% to 2% of cases. ROS1-positive disease is more commonly identified in younger patients and people who have never smoked [4]. The clinical challenge extends beyond controlling disease outside the brain. Central nervous system involvement is common in ROS1-positive NSCLC, both at diagnosis and during treatment [5][6]. As a result, CNS penetration has become an important consideration when evaluating ROS1-targeted therapies.
Another challenge is acquired resistance. ROS1 kinase-domain alterations can emerge during treatment with ROS1 TKIs, with the solvent-front substitution G2032R representing an important resistance mutation [7]. Zidesamtinib was designed as a selective, brain-penetrant ROS1 inhibitor with activity against ROS1 fusions and a range of resistance mutations, including G2032R. Its structure was developed to preserve ROS1 activity while minimizing inhibition of the tropomyosin receptor kinase (TRK) family. This distinction is clinically relevant because TRK inhibition has been associated with neurologic adverse effects with some ROS1 inhibitors [7][8].
Preclinical structural studies have provided a mechanistic basis for zidesamtinib’s activity against G2032R-mutant ROS1. A 2.2-Å cocrystal structure demonstrated that zidesamtinib can accommodate the substituted residue without the steric interaction that reduces binding of several other ROS1 inhibitors [8].
What Is New: First-Line Data From ARROS-1
The new findings come from the global, single-arm, first-in-human phase I/II ARROS-1 study (NCT05118789). The treatment-naive cohort included patients with locally advanced or metastatic ROS1-positive NSCLC who had not previously received a ROS1 TKI. Up to one prior line of chemotherapy and/or immunotherapy was permitted [1].
Of 532 patients with ROS1-positive NSCLC who received zidesamtinib at 100 mg once daily across the ARROS-1 program, 183 were receiving their first ROS1-targeted therapy. For the efficacy analysis presented at WCLC, 94 patients with measurable disease were evaluable after sufficient follow-up [1]. At a median follow-up of 15.2 months, 88 of 94 patients responded, producing an ORR of 94% (95% CI, 87%-98%). Fourteen patients, or 15%, achieved a complete response [1].
The responses also appeared durable. Among responders, 94% remained in response at nine months, and 86% remained in response at 12 months. The 12-month progression-free survival rate was 90%. Median duration of response and median progression-free survival had not been reached at the time of the analysis [1]. These findings are important, but the single-arm design matters. Without a randomized comparator, the results establish zidesamtinib activity in this population but do not establish superiority over another first-line ROS1 inhibitor.
Intracranial Activity
The CNS findings are particularly relevant in ROS1-positive disease. Among the 10 efficacy-evaluable patients with measurable brain metastases at baseline, all 10 achieved an intracranial response. Seven patients, or 70%, achieved an intracranial complete response, meaning complete disappearance of detectable intracranial disease according to the study assessment [1].
At 12 months, 78% of these patients maintained an intracranial response. No patients without brain metastases at baseline experienced CNS progression during the reported follow-up period [1]. The small number of patients with measurable CNS disease—10 in this analysis—limits how precisely these findings can be interpreted. Nevertheless, the results add to the evidence that zidesamtinib has clinically relevant CNS activity.
Safety in the Treatment-Naive Population
The safety findings in the treatment-naive cohort were consistent with previous reports of zidesamtinib. The most common treatment-related adverse events occurring in at least 15% of patients were peripheral edema (34%), increased weight (18%), increased blood creatine phosphokinase (CPK) (18%), dysgeusia (17%), and increased aspartate aminotransferase (AST) (15%). Most were low grade [1].
Treatment-related adverse events resulted in dose reductions in 11% of patients and treatment discontinuation in 1% [1]. The low discontinuation rate is notable, but it should not be interpreted as an absence of clinically important toxicity. The FDA prescribing information includes warnings and precautions for central nervous system adverse reactions, QTc interval prolongation, interstitial lung disease/pneumonitis, skeletal fractures, myalgia with CPK elevation, pancreatic toxicity, and embryo-fetal toxicity [3].
In the pooled safety population supporting the FDA label, CNS adverse reactions occurred in 25% of patients, including dizziness, ataxia, cognitive impairment, and psychiatric disorders. Grade 3 or 4 CNS events occurred in 2.5% [3]. The prescribing information therefore remains important when managing patients receiving zidesamtinib.
The Evidence Behind the Existing Approval
The FDA’s July 2026 approval was based on the previously treated population in ARROS-1 [2]. The efficacy population included 117 patients with locally advanced or metastatic ROS1-positive NSCLC who had received at least one prior ROS1 TKI. The confirmed ORR was 44% (95% CI, 34%-53%), with 82% of responders maintaining a response for at least six months and 69% maintaining a response for at least 12 months [2].
The ORR was 49% among patients who had received one prior ROS1 TKI and 38% among those who had received two or more prior ROS1 TKIs [2]. Among patients with measurable CNS metastases, the intracranial response rate was 48%, including intracranial complete responses. Activity was also observed in patients whose tumors carried the ROS1 G2032R resistance mutation [8,9]. These data established zidesamtinib as a treatment option for patients whose disease had progressed after prior ROS1-directed therapy and provided the regulatory foundation for the current U.S. indication.
From Molecular Design to Clinical Activity
The development of zidesamtinib reflects a broader effort to address three related problems in ROS1-positive NSCLC: acquired resistance, CNS disease, and toxicity associated with off-target kinase inhibition. Preclinical studies demonstrated activity against a broad range of ROS1 kinase-domain mutations. In particular, zidesamtinib retained activity against G2032R, the most frequently identified ROS1 resistance mutation in several clinical settings [8].
The drug’s macrocyclic structure also contributes to its selectivity. Biochemical and structural studies showed potent inhibition of ROS1, including G2032R-mutant ROS1, while demonstrating substantial selectivity over TRK family kinases [8]. These laboratory findings help explain the rationale for the clinical development program. They do not, however, establish that the molecular design will translate into superior long-term outcomes compared with other modern ROS1 inhibitors. That question requires clinical comparisons.
A Changing ROS1 Treatment Landscape
Zidesamtinib is entering an increasingly competitive ROS1 treatment landscape. Other next-generation ROS1 inhibitors have also been developed to address CNS disease, acquired resistance, and TRK-associated toxicity. One example is JYP0322, a selective, brain-penetrant ROS1 inhibitor evaluated in a first-in-human phase I study. In the published phase I study, JYP0322 produced an ORR of 95.7% among TKI-naive patients. The ORR among patients with ROS1 G2032R-mutant disease was 77.8%, while the intracranial ORR among patients with brain metastases was 55.6% [10].
These data demonstrate the level of activity now being reported across the next generation of ROS1 inhibitors. They should not, however, be interpreted as evidence that one agent is superior to another. Differences in study populations, eligibility criteria, follow-up, assessment methods, prior therapies, and CNS disease burden make cross-trial comparisons unreliable [11]. The more clinically important question is likely to be how these agents perform over time, including their durability of CNS control, resistance patterns, toxicity profiles, and sequencing after disease progression.
A Regulatory Path Forward
GSK has stated that the ARROS-1 first-line findings will support a supplemental New Drug Application to the FDA later in 2026 [1]. For now, the regulatory position is clear: zidesamtinib is approved in the United States for adults with locally advanced or metastatic ROS1-positive NSCLC who have received a prior ROS1 TKI. It is not currently FDA-approved for first-line treatment. [2][3]
The first-line data are also not yet supported by a randomized comparison. The efficacy population reported at WCLC consisted of 94 patients, and the data were presented at a medical meeting and released by the sponsor rather than published as a full peer-reviewed clinical report. Those limitations do not diminish the observed response rate. They do, however, define what can reasonably be concluded from the current evidence.
What Remains Unknown
The first-line ARROS-1 results are notable for the depth of response and the early durability of disease control. Several clinically important questions remain unanswered.
- First, the treatment-naive cohort is a single-arm, non-randomized population. The 94% ORR cannot be used to establish superiority over other ROS1 inhibitors without a direct comparison.
- Second, median progression-free survival and median duration of response had not been reached at the time of the analysis. Longer follow-up will therefore be necessary to define the durability of first-line treatment.
- Third, the resistance mechanisms that emerge after first-line zidesamtinib remain incompletely characterized. Understanding how tumors evolve after prolonged exposure will be important for determining subsequent treatment strategies.
Finally, the role of zidesamtinib within the broader sequence of ROS1-directed therapies remains to be established. As more selective, brain-penetrant ROS1 inhibitors become available, treatment selection may increasingly depend on the molecular characteristics of resistance, CNS disease, previous TKI exposure, and individual tolerability.
For now, the ARROS-1 data establish a clear signal of activity in TKI-naive ROS1-positive NSCLC. Whether that signal ultimately changes the first-line treatment landscape will depend on regulatory review, longer follow-up, and how zidesamtinib compares with the other modern ROS1 inhibitors now being used or developed.
Clinical trials
A Study of Zidesamtinib (NVL-520) in Patients With Advanced NSCLC and Other Solid Tumors Harboring ROS1 Rearrangement (ARROS-1)
ClinicalTrials.gov ID NCT05118789
Expanded Access Program of Zidesamtinib (NVL-520) for Patients With Advanced ROS1+ NSCLC or Other ROS1+ Solid Tumors – ClinicalTrials.gov ID NCT06797362
Highlights of Prescribing Information
Zidesamtinib (Jideytro®; GSK) [Prescribing Information]
- Current FDA-approved indication: Adult patients with locally advanced or metastatic ROS1-positive NSCLC who have received a prior ROS1 kinase inhibitor [3].
- Recommended dosage: 100 mg orally once daily, with or without food, until disease progression or unacceptable toxicity [3].
- Available strengths: 25-mg and 100-mg tablets [3].
- Important warnings and precautions: CNS adverse reactions, QTc interval prolongation, interstitial lung disease/pneumonitis, skeletal fractures, myalgia with CPK elevation, pancreatic toxicity, and embryo-fetal toxicity [3].
- Common adverse reactions: Edema, peripheral neuropathy, constipation, fatigue, and dyspnea [3].
- Drug interactions: Avoid strong and moderate CYP3A inhibitors and inducers, per the prescribing information [3].
- This is not a complete summary of the prescribing information. Clinicians should consult the current FDA-approved label before prescribing zidesamtinib.
References
[1] GSK plc. GSK presents positive registrational data to potentially expand Jideytro (zidesamtinib) to first-line ROS1-positive non-small cell lung cancer. Presented at: 2026 World Conference on Lung Cancer; September 12-15, 2026; Seoul, South Korea. September 14, 2026.
[2] U.S. Food and Drug Administration. FDA approves zidesamtinib for ROS1-positive non-small cell lung cancer. July 22, 2026.
[3] JIDEYTRO (zidesamtinib) tablets, for oral use. U.S. Prescribing Information. Nuvalent, Inc.; revised July 2026.
[4] Ou SI, Zhu VW. CNS metastasis in ROS1+ NSCLC: An urgent call to action, to understand, and to overcome. Lung Cancer. 2019 Apr;130:201-207. doi: 10.1016/j.lungcan.2019.02.025. Epub 2019 Feb 22. PMID: 30885345.
[5] Patil T, Smith DE, Bunn PA, Aisner DL, Le AT, Hancock M, Purcell WT, Bowles DW, Camidge DR, Doebele RC. The Incidence of Brain Metastases in Stage IV ROS1-Rearranged Non-Small Cell Lung Cancer and Rate of Central Nervous System Progression on Crizotinib. J Thorac Oncol. 2018 Nov;13(11):1717-1726. doi: 10.1016/j.jtho.2018.07.001. Epub 2018 Jul 5. PMID: 29981925; PMCID: PMC6204290.
[6] Gainor JF, Tseng D, Yoda S, Dagogo-Jack I, Friboulet L, Lin JJ, Hubbeling HG, Dardaei L, Farago AF, Schultz KR, Ferris LA, Piotrowska Z, Hardwick J, Huang D, Mino-Kenudson M, Iafrate AJ, Hata AN, Yeap BY, Shaw AT. Patterns of Metastatic Spread and Mechanisms of Resistance to Crizotinib in ROS1-Positive Non-Small-Cell Lung Cancer. JCO Precis Oncol. 2017;2017:PO.17.00063. doi: 10.1200/PO.17.00063. Epub 2017 Aug 16. PMID: 29333528; PMCID: PMC5766287.
[7] Drilon A, Horan JC, Tangpeerachaikul A, Besse B, Ou SI, Gadgeel SM, Camidge DR, van der Wekken AJ, Nguyen-Phuong L, Acker A, Keddy C, Nicholson KS, Yoda S, Mente S, Sun Y, Soglia JR, Kohl NE, Porter JR, Shair MD, Zhu V, Davare MA, Hata AN, Pelish HE, Lin JJ. NVL-520 Is a Selective, TRK-Sparing, and Brain-Penetrant Inhibitor of ROS1 Fusions and Secondary Resistance Mutations. Cancer Discov. 2023 Mar 1;13(3):598-615. doi: 10.1158/2159-8290.CD-22-0968. PMID: 36511802; PMCID: PMC9975673.
[8] Tangpeerachaikul A, Mente S, Magrino J, Gu F, Horan JC, Pelish HE. Zidesamtinib Selective Targeting of Diverse ROS1 Drug-Resistant Mutations. Mol Cancer Ther. 2025 Jul 2;24(7):1005-1019. doi: 10.1158/1535-7163.MCT-25-0025. PMID: 40299789; PMCID: PMC12214885.
[9] Nuvalent, Inc. Pivotal ARROS-1 efficacy and safety data: zidesamtinib in TKI-pretreated patients with advanced/metastatic ROS1-positive NSCLC. Press release. Online/Presented data, 2025 World Conference on Lung Cancer (WCLC 2025; held September 6-9, 2025, in Barcelona, Spain.
[10] Ma Y, Xue J, Gao F, Yang Y, Zhao Y, Wang L, Huang Z, Wu C, Luo Y, Li J, Fang J, Song Z, Li Y, Ma H, Guo Q, Yi T, Liu X, Liu Z, Zhou J, Dong X, Yu Q, He J, Yang L, Fang W, Sun Y, Qi Y, Lv J, Ge Z, Xia B, Li L, Wang W, Huang Y, Zhang L, Zhao H. JYP0322, a highly selective and brain-penetrant ROS1 inhibitor, overcomes ROS1G2032R resistance mutation in NSCLC: The first-in-human phase 1 trial. Cancer Cell. 2026 May 11;44(5):983-994.e5. doi: 10.1016/j.ccell.2026.03.018. Epub 2026 Apr 23. PMID: 42030931.
[11] Desilets A, Repetto M, Yang SR, Drilon A. Targeting ROS1 rearrangements in non-small cell lung cancer: Current insights and future directions. Cancer. 2025 Mar 15;131 Suppl 1(Suppl 1):e35784. doi: 10.1002/cncr.35784. PMID: 40171848; PMCID: PMC12551656.
This article is intended for informational purposes for healthcare professionals and does not constitute medical advice. Prescribing information referenced here is not a complete summary of safety data. Clinicians should consult the full FDA-approved prescribing information before making treatment decisions. First-line zidesamtinib remains investigational in the United States. The 2026 WCLC data discussed in this article were presented by the sponsor and have not yet undergone full peer review; therefore, interpret reported results in the context of the available conference data and subsequent regulatory review.
Featured image: © 2019 – 2026 Robina Weermeijer. Used with permission under the Unsplash License
DOI
){kind=link}