Image by
Ashling Wahner & MJH Life Sciences Using AI
Preliminary efficacy and intracranial activity observed with zidesamtinib (NVL-520) in the phase 1/2 ARROS-1 trial (NCT05118789) indicates that this selective ROS1 inhibitor could overcome key limitations of available ROS1/TRK inhibitors—including suboptimal central nervous system (CNS) penetrance, a lack of activity against on-target resistance, and TRK-associated neurologic and gastrointestinal toxicities—for patients with ROS1-positive non–small cell lung cancer (NSCLC), according to Joshua K. Sabari, MD.1
Pooled data from the phase 1 and 2 portions of ARROS-1 showed that, among patients previously treated with at least 1 TKI (n = 117), treatment with zidesamtinib at the recommended phase 2 dose (RP2D) of 100 mg once daily generated an objective response rate (ORR) of 44% (95% CI, 34%-53%).1 Moreover, patients experienced durations of response lasting at least 6, 12, and 18 months at rates of 84% (95% CI, 71%-92%), 78% (95% CI, 62%-88%), and 62% (95% CI, 28%-84%), respectively. In patients who had received only 1 prior ROS1 TKI, including crizotinib (Xalkori) or entrectinib (Rozlytrek) with or without chemotherapy (n = 55), the ORR was 51% (95% CI, 37%-65%). The percentage of patients who remained in response for at least 6, 12, and 18 months was 93% (95% CI, 74%-98%).
“This was a notable phase 1 dose-escalation and -expansion [study] in an all-comer patient population with ROS1 fusion–positive NSCLC,” Sabari said in an interview with OncLive®. “We saw impressive [clinical activity with zidesamtinib] across all these patient subsets. [As the data mature], we’re only going to see better response rates with a better durability of response, and I hope to see this agent move into the frontline setting.”
In February 2024, the FDA granted breakthrough therapy designation tozidesamtinibfor patients with ROS1-positive metastatic NSCLC with prior exposure to 2 or more ROS1 TKIs.2
In the interview, Sabari discussed the limitations of the current arsenal of ROS1-targeted TKIs, the unique design and goal of the ongoing ARROS-1 trial, early-phase efficacy and safety data with zidesamtinib in ROS1 fusion–positive NSCLC, and the potential role for zidesamtinib in the ROS1 fusion–positive NSCLC treatment paradigm.
Sabari is the director of High Reliability Organization Initiatives at the Perlmutter Cancer Center and an assistant professor in the Department of Medicine at the NYU Grossman School of Medicine, both in New York, New York.
OncLive: Why is there an unmet need for more effective, tolerable targeted agents for patients with ROS1 fusion–positive lung cancer with and without acquired resistance mechanisms?
Sabari: ROS1 fusions are quite common. They occur in 1% to 2% of all NSCLC cases, predominantly in patients with adenocarcinoma. It is important to develop novel therapeutic strategies for this patient population. We have multiple agents that are FDA approved, including crizotinib, entrectinib, repotrectinib [Augtyro], and most recently, taletrectinib [Ibtrozi]. That being said, there remains a need for agents that address acquired resistance, particularly those targeting mutations such as G2032R. There is also a need for agents with better CNS penetration. Most importantly, there is a need for agents that are better tolerated—particularly those that spare TRK. It is known that TRK inhibition can cause dizziness, ataxia, and other neurocognitive adverse effects [AEs], so there is an unmet need in this patient population. Despite the availability of multiple approved agents, we still see patients who progress, and we still see patients who need novel therapies in this space.
Could you expand on the design and objectives of the ARROS-1 study?
The phase 1/2 ARROS-1 study evaluated zidesamtinib in [patients with] ROS1 fusion–positive solid tumors. This study was initially a dose-escalation trial, which was then followed by a dose-expansion phase. [The agent was assessed] in all-comers with NSCLC [harboring] ROS1 fusions, [regardless of whether they had exposure to] any ROS1 TKI therapy. This included patients who were naive to ROS1 TKIs, those who had received prior ROS1 TKIs including repotrectinib, and those with known ROS1 resistance mutations such as G2032R prior to repotrectinib. [The study also included] patients who had received [at least] 2 prior lines of ROS1-directed therapy, including lorlatinib [Lorbrena] and repotrectinib. Lastly, we looked at patients who had received only 1 prior ROS1 TKI, such as crizotinib.
What differentiates this trial from prior research with ROS1-targeted agents?
Most of the prior trials that we’ve had [in ROS1 fusion–positive NSCLC) have evaluated [patients with disease] progression on crizotinib, and that’s how we developed repotrectinib and taletrectinib. The ARROS-1 trial includes a patient population that has been previously treated with repotrectinib. This is a heavily pretreated patient population ranging from 1 to 4 prior lines of therapy. It is exciting to see that even in patients who are pretreated with repotrectinib, particularly those who have G2032R resistance mutations, we saw very high and durable response rates.
How do you see the broader inclusion criteria in ARROS-1 helping in the transition of zidesamtinib from bench to bedside?
Resistance mechanisms develop in all patients receiving targeted therapies. In clinical practice, we identify these mutations and then we can recapitulate them in the laboratory using patient-derived xenograft models. Bringing that back from the lab to the clinic, we’re then able to test these agents in patients who have these specific acquired resistance mutations [after treatment with] crizotinib or entrectinib. [In preclinical studies], we saw robust activity [with zidesamtinib] in repotrectinib-naive patients with the G2032R resistance mutation. It is exciting to see that the data generated in animal models and cell lines in the laboratory can be brought back into the clinic; to see those results [reproduced] in humans is quite impressive.
How would you characterize both the systemic and intracranial activity of zidesamtinibin the TKI-pretreated and -naive cohorts of ARROS-1 based on data presented at the 2024 ESMO Congress?
In the all-comer [patient population], a 44% ORR [was observed], including 2 complete responses. Notably, among patients who had received prior repotrectinib, a 38% response rate was reported, with 3 of 8 patients responding. [Notably, these patients] predominantly had the ROS1 G2032R resistance mutation. Interestingly, in patients who also harbored the G2032R mutation, but were also naive to repotrectinib, the ORR was 72%. In patients with 2 or more prior ROS1 TKIs in all-comers, the ORR was 41%. In those who were naive to repotrectinib, the ORR was 47%; in those who had received only 1 prior ROS1 TKI, the ORR was 73%.
Durable responses were observed across the entire cohort, with the median duration of response not reached. Given the importance of CNS activity, intracranial ORRs were reported in 50% of patients. Historically, agents such as crizotinib and entrectinib have shown limited CNS penetration; however, agents like taletrectinib, repotrectinib, and now zidesamtinib demonstrate significant intracranial activity.
What should be known about the tolerability of this agent relative to other ROS1-directed TKIs in NSCLC?
Safety remains an important issue in the treatment of patients with ROS1 fusion–positive NSCLC. A lot of our agents such as repotrectinib, inhibit TRK A, B, and C, leading to neurologic AEs like dizziness in up to 30% of patients. Taletrectinib, a newer FDA-approved agent, spares TRK B but continues to inhibit TRK A and C, offering a reduced—but still present—risk of neurologic toxicity. In contrast, zidesamtinib was specifically designed to avoid TRK inhibition altogether, so we don’t see neurologic toxicities in this population.
We did see grade 1 peripheral edema in 14% [of patients, although] 5% had grade 2 peripheral edema. We also saw grade 1 aspartate aminotransferase and alanine aminotransferase level elevations [in 11% of patients, respectively], and weight increase in 11% of all-comers. There were no treatment-related AEs that led to treatment discontinuation, and the dose reduction rate was 8%. This was significantly lower than what we’ve experienced in the past with repotrectinib or therapies like entrectinib.
How does zidesamtinib’s mechanism of action contribute to that low rate of dose reduction in this trial?
Because zidesamtinib spares TRK, we don’t see the CNS toxicity with this agent [that we see] with other agents like repotrectinib or potentially even taletrectinib. The very nature of this agent, the way that it was developed, truly removes this potential toxicity from the clinic.
Pending further positive data from the TKI-naive cohort, where could you see this agent being positioned in the ROS1 fusion–positive NSCLC paradigm?
Zidesamtinib is a best-in-class, ROS1-selective, TRK-sparing therapy. We are now enrolling newly diagnosed, untreated patients in the frontline. This is potentially practice-changing for the ROS1 fusion–positive NSCLC patient population.
References
- Nuvalent announces positive pivotal data from ARROS-1 clinical trial of zidesamtinib for TKI pre-treated patients with advanced ROS1-positive NSCLC. News release. Nuvalent. June 24, 2025. Accessed July 21, 2025. https://investors.nuvalent.com/2025-06-24-Nuvalent-Announces-Positive-Pivotal-Data-from-ARROS-1-Clinical-Trial-of-Zidesamtinib-for-TKI-Pre-treated-Patients-with-Advanced-ROS1-positive-NSCLC
- Nuvalent receives U.S. FDA breakthrough therapy designation for NVL-520. News release. Nuvalent. February 27, 2024. Accessed July 21, 2025. https://investors.nuvalent.com/2024-02-27-Nuvalent-Receives-U-S-FDA-Breakthrough-Therapy-Designation-for-NVL-520