Newer Generations, Resistance Profiling, and Toxicity Trade-offs Guide Modern TKI Selection/Use in Oncogene-Driven NSCLC

As the arsenal of targeted therapies continues to expand in non–small cell lung cancer (NSCLC) it is increasingly important to refine TKI selection and sequencing for patients displaying oncogenic driver mutations, according to Alexander Drilon, MD, who asserts that increased emphasis should be placed on selecting next-generation agents, performing resistance mutation profiling, and considering differences in toxicity profiles.1

In a presentation delivered during the 20th Annual New York Lung Cancers Symposium®, Drilon outlined a practical framework for choosing frontline TKIs in ALK-, ROS1 fusion–, RET-, NTRK-, and NRG1 fusion–positive NSCLC in both TKI-naive and -exposed patient populations; and discussed emerging strategies of interest post–TKI progression, including third-generation inhibitors and antibody-drug conjugates (ADCs).

“For ALK and ROS1 [fusions], you are generally looking at second- or third-generation TKIs in the frontline setting,” Drilon, chief of the Early Drug Development Service at Memorial Sloan Kettering Cancer Center in New York, explained. “There are no later-generation drugs approved for RET fusions, so we are still limited to first-generation RET-targeted TKIs. For NTRK fusions, although first- and second-generation agents exist, first-generation TRK-targeted TKIs remain a reasonable choice because of the safety penalties we see with later-generation therapies.”

Which TKI is preferred in the ALK-positive setting?

Drilon began by reviewing the current TKI landscape for ALK, ROS1, RET, and NTRK fusions in metastatic NSCLC. Across targets, Drilon framed the choice of frontline TKI around 3 core principles: depth and durability of response, resistance mutation coverage, and tolerability.

For ALK fusion–positive NSCLC, Drilon reviewed the 4 key randomized phase 3 trials comparing 3 next-generation ALK TKIs with crizotinib (Xalkori): lorlatinib (Lorbrena), alectinib (Alecensa), brigatinib (Alunbrig), and ensartinib (Ensacove).

“If you look across the 4 trials, we are seeing what appears to be the best data for lorlatinib,” Drilon noted, citing favorable HRs, progression-free survival (PFS) landmarks, and robust central nervous system (CNS) efficacy both in patients with and without brain metastases. “If we have a fit patient whom we think may tolerate lorlatinib, it would be my first choice in many cases.”

However, he underscored the neurocognitive adverse effects (AEs) associated with lorlatinib, including mood changes, cognitive slowing, and weight gain. “If you think you are going to run into issues with the cognitive changes that you see with lorlatinib, then the other drugs are very reasonable choices,” he said.

Among the alternatives are alectinib and brigatinib, both of which have more mature data. “If you look across those 2 drugs, alectinib does seem to be, based on the overall toxicity data, much more tolerable, and is something that I frequently choose in clinic,” he added.

How is TKI selection evolving for ROS1 fusion–positive NSCLC?

For ROS1 fusion–positive NSCLC, Drilon described a similar generational shift. “We are seeing a progression from first-generation TKIs such as crizotinib and entrectinib to second-generation agents, with improvements in objective response rate and progression-free survival,” he said.

He focused on taletrectinib (Ibtrozi) and repotrectinib (Augtyro), both second-generation ROS1 TKIs with activity against common resistance mutations.

As “the new kid on the block,” taletrectinib (Ibtrozi) demonstrated clinically meaningful efficacy in patients with ROS1-positive non–small cell lung cancer (NSCLC), according to updated findings from the phase 2 TRUST-II trial (NCT04919811).2

Results from this trial were presented during the International Association for the Study of Lung Cancer (IASLC) 2025 World Conference on Lung Cancer. At a median follow-up of 20.5 months (range, 8.3-34.5), taletrectinib achieved a confirmed overall response rate (cORR) of 85.2% (95% CI, 72.9%-93.4%) in the TKI-naive cohort (n = 54). In this cohort, the cORR was 90.0% in those who had previously received chemotherapy (n = 9) and 84.1% in those without prior chemotherapy exposure (n = 37). The median time to response (TTR) was 1.4 months (95% CI, 1.3-3.4), and the median duration of response (DOR) was not reached (NR; 95% CI, 20.6-NR).

In the TKI-pretreated cohort (n = 47), at a median follow-up of 20.4 months (range, 8.6-34.5), taletrectinib produced a cORR of 61.7% (95% CI, 46.4%-75.5%), with a median TTR of 1.4 months (95% CI, 1.4-1.6). The cORRs were 78.9% and 50.0% among patients who had received prior chemotherapy (n = 15) vs those who hadn ot (n = 14). The median DOR in this cohort was 19.4 months (95% CI, 10.7 months–not evaluable).

Drilon emphasized the importance of off-target neurologic toxicity when choosing between the 2 drugs, noting that repotrectinib appears to cause more neurologic toxicities, particularly dizziness, and at a higher frequency than taletrectinib.1

How should clinicians choose between TKIs in RET fusion–positive NSCLC?

In RET fusion–positive NSCLC, the current approved selective TKIs are selpercatinib (Retevmo) and pralsetinib (Gavreto). “We have registrational data for both agents, and when you look closely, you do see numerical differences between them,” Drilon said.

He noted that selpercatinib has the advantage of having randomized phase 3 data in the chemoimmunotherapy setting and appears to have performed better than pralsetinib according to data from a systematic matched analysis.1

Safety is another differentiator. “If you look at grade 3 or higher treatment-related adverse effects [AEs] and treatment-related discontinuations, these appear to be more frequent with pralsetinib,” Drilon explained. “For these reasons, I tend to choose selpercatinib as my first-line RET TKI,” he said, while acknowledging that both are active agents.

What is the preferred TRK-targeted TKI for NTRK fusion–positive NSCLC?

In NTRK-positive NSCLC, Drilon highlighted both first- and second-generation TRK TKIs but cautioned against reflexively moving to later-generation inhibitors in the frontline setting.

“Repotrectinib is a good drug, and it is already approved, but we still do not have mature longitudinal data on progression-free survival. We do not yet know if it is a slam dunk that next-generation TRK therapy will improve outcomes upfront.”

Given the more favorable safety profile of first-generation agents, Drilon stated that he would prefer to use larotrectinib (Vitrakvi) or entrectinib (Rozlytrek) in the clinic.

“If we look at the differential activity of larotrectinib and entrectinib in both the tumor-agnostic and lung cancer–specific [populations], it appears as if larotrectinib is [more effective],” he explained. “There is also a similar systematic analysis showing larotrectinib performing better than entrectinib. For that reason, I choose larotrectinib as my first-line TRK TKI.”

How should NRG1 fusion–positive NSCLC be approached?

NRG1 fusions represent a biologically distinct subset of oncogene-driven NSCLC, Drilon noted. Because the oncogenic driver is ligand-mediated rather than kinase activation alone, traditional small-molecule TKIs are less effective.

The bispecific HER3/HER2 monoclonal antibody zenocutuzumab-zbco (Bizengri) is one potential way to leverage large-molecule therapies for this tumor type, interrupting the interaction between NRG1 and HER3, he explained.

Tumor-agnostic data have demonstrated the agent’s efficacy across NRG1 fusion–positive disease, including lung and pancreatic cancer. In the phase 2 eNRGy trial (NCT02912949), patients with NSCLC (n = 64) achieved an ORR of 33% (95% CI, 22%-46%) with a median DOR of 7.4 months (95% CI, 4.0-16.6). Among patients with pancreatic adenocarcinoma (n = 30), the ORR was 40% (95% CI, 23%-59%), and the DOR ranged from 3.7 months to 16.6 months.3

Based on these data, the FDA granted accelerated approval to zenocutuzumab for adult patients with advanced, unresectable, or metastatic NSCLC harboring an NRG1 gene fusion with disease progression on or after prior systemic therapy; or advanced, unresectable, or metastatic pancreatic adenocarcinoma harboring an NRG1 gene fusion with disease progression on or after prior systemic therapy. Notably, this regulatory decision marked the first approval of a systemic therapy in these disease subtypes.

What is the optimal framework for managing NSCLC after progression on a TKI?

In patients with NSCLC who relapse after receiving a frontline TKI therapy and whose disease is not amenable to local therapy, a tumor and/or liquid biopsy should be performed when safe and feasible to determine whether the patient’s disease is displaying on- or off-target resistance, Drilon said.1

“For on-target resistance, if a TKI is available that is of an evolved generation relative to what the patient previously received, that is generally what I would consider next,” he explained. “If no appropriate TKI or clinical trial is available, platinum-based chemotherapy is reasonable.”

For off-target or mixed resistance, Drilon emphasized that systemic chemotherapy remains important. “If we have off-target or mixed resistance, we think about platinum doublet chemotherapy,” he said. “The reason to do a repeat biopsy is that the choice of the nonplatinum partner might be influenced by squamous, neuroendocrine, or small cell transformation.”

Why prioritize later-generation TKIs?

“The first step, in my mind, is to try to choose a more advanced TKI generation than the patient has already received,” Drilon said. Although TKI generations are not rigidly defined, he outlined several features common to newer agents:

• Greater potency against the primary target for improved systemic and CNS activity
• Clinical activity after prior TKI exposure, supporting their use in sequential paradigms
• Broader resistance mutation coverage

He went on to note the ongoing development of newer TKIs such as the investigational ALK-selective inhibitor neladalkib (NVL-655) and ROS1-selective TKI zidesamtinib (NVL-520).

Early data from the phase 1/2 ALKOVE-1 trial (NCT05384626) showed that, among evaluable TKI-pretreated patients with ALK-positive disease (n = 253), the ORR per blinded independent central review (BICR) was 31% (95% CI, 26%-37%).4 At a median follow-up of 11.3 months, the median DOR was not reached (NR). Responses lasted 12 months or longer in 64% (95% CI, 51%-75%) of patients, and 53% (95% CI, 34%-68%) of patients had responses lasting 18 months or more. Among TKI-pretreated patients who were naive to lorlatinib (n = 63), the ORR by BICR was 46% (95% CI, 33%-59%).

Of note, the FDA granted breakthrough therapy designation to NVL-655 as a potential therapeutic option in patients with locally advanced or metastatic ALK-positive NSCLC who previously received 2 or more ALK TKIs in May 2024.5

According to data from the phase 1/2 ARROS-1 trial (NCT05118789), which were presented at the 2025 IASLC World Conference on Lung Cancer, zidesamtinib elicited an ORR of 44% (95% CI, 34%-53%) and a complete response (CR) rate of 1% in patients with advanced ROS1-positive NSCLC.6 These patients had received between 1 and 4 prior lines of ROS1 TKI therapy with or without chemotherapy. In patients who had previously received either crizotinib (Xalkori) or entrectinib (Rozlytrek) with or without chemotherapy, the ORR was 51% (95% CI, 37%-65%), including a 2% CR rate. The ORRs for patients who had received only crizotinib vs only entrectinib with or without chemotherapy were 68% and 33%, respectively.

In February 2024, the FDA granted breakthrough therapy designation tozidesamtinib for patients with ROS1-positive metastatic NSCLC with prior exposure to 2 or more ROS1 TKIs.7

“These agents work very well after crizotinib, entrectinib, repotrectinib, and taletrectinib,” he said, emphasizing their design for improved CNS penetration and reduced neurologic toxicity, and expressing optimism that they may soon become available to patients.

Why is resistance mutation coverage critical when choosing a next-line TKI?

“The next step, when you think about TKIs, is to check the specific mutation you are dealing with when a kinase domain mutation is present,” Drilon said. “If you have a ROS1 solvent-front mutation, lorlatinib has a 0% response rate for that mutation,” he noted.1 “In contrast, repotrectinib and taletrectinib have substantial activity and work well in that context.”

He stressed that this principle extends across targets. “There are many other examples like this,” he said. “There are also some mutations that may confer pan-resistance to all available TKIs, and in those situations, you might need to pivot to something completely different.”

In ALK-positive NSCLC, he highlighted the emergence of compound ALK mutations after multiple sequential TKIs. “These can still be targetable with some of the newer agents,” Drilon said, adding that responses with neladalkib in this setting have been particularly encouraging.

How should toxicity profiles influence TKI choice?

Drilon underscored that AE profiles can be as important as efficacy when selecting among ROS1 TKIs, especially in pretreated patients.

“One example is in a ROS1-positive patient who has already received prior ROS1-directed therapy,” he said. “If you look at comparative data, the frequency of dizziness and disequilibrium appears to be higher with repotrectinib than with taletrectinib.”

“There are other toxicities that patients do not feel as directly, such as transaminitis, which you need to watch with taletrectinib,” he added. “But dizziness and neurologic effects are much more clinically bothersome. This often pushes me toward taletrectinib when I am weighing these 2 agents.”

References

1. Drilon A. Lung cancers with gene fusions. Presented at: 20th Annual New York Lung Cancers Symposium; November 15, 2025; New York, New York. Accessed November 15, 2025.
2. Liu G, Choi C, Sugawara S, et al. Updated efficacy and safety of taletrectinib in patients with ROS1+ non–small cell lung cancer: the global TRUST-II study. Presented at: International Association for the Study of Lung Cancer 2025 World Conference on Lung Cancer; September 6-9, 2025; Barcelona, Spain. Abstract OA14.05.
3. FDA grants accelerated approval to zenocutuzumab-zbco for non-small cell lung cancer and pancreatic adenocarcinoma. FDA. December 4, 2024. Accessed November 17, 2025. https://www.fda.gov/drugs/resources-information-approved-drugs/fda-grants-accelerated-approval-zenocutuzumab-zbco-non-small-cell-lung-cancer-and-pancreatic
4. Nuvalent announces positive topline pivotal data from ALKOVE-1 clinical trial of neladalkib for TKI pre-treated patients with advanced ALK-positive NSCLC. News release. Nuvalent, Inc. November 17, 2025. Accessed November 17, 2025. https://investors.nuvalent.com/2025-11-17-Nuvalent-Announces-Positive-Topline-Pivotal-Data-from-ALKOVE-1-Clinical-Trial-of-Neladalkib-for-TKI-Pre-treated-Patients-with-Advanced-ALK-positive-NSCLC
5. Nuvalent receives US FDA breakthrough therapy designation for NVL-655. News release. Nuvalent. May 16, 2024. Accessed November 17, 2025. https://investors.nuvalent.com/2024-05-16-Nuvalent-Receives-U-S-FDA-Breakthrough-Therapy-Designation-for-NVL-655
6. Drilon AE, Cho BC, Lin JJ, et al. Pivotal ARROS-1 efficacy and safety data: zidesamtinib in TKI-pretreated patients with advanced/metastatic ROS1+ NSCLC. Presented at: 2025 International Association for the Study of Lung Cancer 2025 World Conference on Lung Cancer; September 6-9, 2025; Barcelona, Spain. Abstract 4540.
7. Nuvalent receives U.S. FDA breakthrough therapy designation for NVL-520. News release. Nuvalent. February 27, 2024. Accessed November 17, 2025. https://investors.nuvalent.com/2024-02-27-Nuvalent-Receives-U-S-FDA-Breakthrough-Therapy-Designation-for-NVL-520