2 Clarke Drive
Suite 100
Cranbury, NJ 08512
© 2024 MJH Life Sciences™ and OncLive - Clinical Oncology News, Cancer Expert Insights. All rights reserved.
ROS1 and TRK are two emerging targets that have significant therapeutic promise for patients with non–small cell lung cancer, although they are not commonly considered while doing mutation testing.
Jonathan Riess, MD
ROS1 and TRK are two emerging targets that have significant therapeutic promise for patients with non—small cell lung cancer (NSCLC), although they are not commonly considered while doing mutation testing, according to a talk by Jonathan Riess, MD, at the 2016 International Lung Cancer Congress.
“I think the word needs to get out in the community about these targets,” said Riess, assistant professor of medicine, cancer, hematology oncology, UC Davis Comprehensive Cancer Center. “In the NCCN guidelines there is a clear recommendation for compressive genetic profiling in order to look for other mutations with newer targeted therapy drugs that are available, at least in clinical trials, that have early evidence of efficacy. I think the key point is to test for ROS1 and TRK, because you need to know that these are present in the patient’s tumor in order to guide them to the appropriate therapy.”
ROS1 alterations are present in about 1% to 2% of NSCLC, mainly among those with adenocarcinoma histology. This corresponds to about 2500 patients annually, said Riess. ROS1 fusions (Chromosome 6) share sequence homology with ALK. In most cases, FISH and IHC are used to detect these fusions, along with DNA sequencing methods, he said.
Crizotinib (Xalkori) is approved for locally advanced or metastatic ROS1-positive NSCLC and there are several promising agents being investigated that targeted TRK. Crizotinib was approved for patients with ROS1-positive metastatic NSCLC in March 2016, following an initial FDA approval for ALK-positive NSCLC patients in 2011.
The ROS1 approval for crizotinib was based on a multicenter, single-arm phase I trial known as Study 1001, which included 50 patients with ROS1-positive metastatic NSCLC treated with 250 mg of crizotinib orally twice daily. The primary outcome measures were objective response rate (ORR) and duration of response.1
The study demonstrated an ORR of 66% by an independent radiology review (95% CI, 51-79). There was 1 complete response and 32 partial responses. The median duration of response was 18.3 months (95% CI, 12.7-not reached). The safety profile of crizotinib was generally consistent with that observed in patients with ALK-positive NSCLC.
“Crizotinib has had a tremendous impact,” said Riess. “It was a very effective therapy with a median PFS of 18.3 months. For our current treatments for lung cancer it doesn’t get much better than that.”
While the availability of this therapy has made a significant difference, it does not work for all patients. There are several mechanisms of acquired resistance to crizotinib in ROS1-rearranged NSCLC, explained Riess.
The first resistance mechanism to be described was the G2032R mutation, which was identified in 2013. This mutation prevents crizotinib from binding to ROS1 due to steric hindrance, which allows only one ATP to bind. L1951R and L2026M are also common “gatekeeper mutations” that interfere with crizotinib binding to ROS1.
There are several novel therapies in development to overcome these mechanisms of resistance, he added. These agents include lorlatinib, ceritinib (Zykadia), cabozantinib (Cabometyx), foretinib, DS6061b, and entrectinib (RXDX-101), said Riess.
As an example of the activity seen with these agents, Riess described lorlatinib (PF-06463922), a next-generation ALK/ROS1 TKI. Phase I/II data for lorlatinib were presented at the 2016 ASCO Annual Meeting,2 in which the agent demonstrated durable clinical responses, including intracranial responses, in patients with ALK and ROS1-positive NSCLC.
Patients included in the ongoing phase I/II dose-escalation study had ALK-positive or ROS1-positive NSCLC with or without brain metastases and were treatment naïve or had disease progression after one or more TKIs. The primary objective was to identify the maximum tolerated dose and a recommended phase II dose. Other objectives were safety and efficacy by RECIST v1.1, including intracranial activity.
Of the 54 patients treated as of November 30, 2015, 41 had ALK-positive and 12 had ROS1-positive tumors, and 1 tumor had a mutation status not recorded at the cutoff date. There was a marked reduction in tumor target lesion size for a majority of patients with ROS1-mutant NSCLC.
Of the 11 evaluable patients, tumor size decreased by >30% for 55% of those treated with various doses of lorlatinib. The ORR was 39%. For those with previously treated ALK-mutant tumors, the ORR was 46%. Responses to lorlatinib were seen in a patient with ALK G1202R resistance mutations. Prior to experiencing a response, this patient had received prior crizotinib, ceritinib, and alectinib.
The most common treatment-related adverse events (AEs) were hypercholesterolemia (54%) and peripheral edema (37%). Hypercholesterolemia was the most common (9%) grade (G) ≥3 treatment-related AE and most frequent reason for dose delay/reduction.
The data on lorlatinib is promising, said Riess. “This had activity in both ROS1-positive TKI-naïve patients and also in patients who have had previous exposure to crizotinib and then progressed,” he said. “It is theoretically able to overcome resistance mutations.”
Potential for TRK
TRK is also being underutilized in NSCLC, said Riess, even though an FDA-approved TRK-targeted agent is not yet available. “NTRK1, NTRK2, and NTRK3 are expressed predominantly in the nervous system in normal tissue, not in cancer, and are specifically activated by neurotrophins,” he said. “It was discovered that these TRK fusions can act as an oncogene in 1982, but there haven’t been any great therapies that targeted them up until recently.”
TRK fusions have been described in a multitude of cancers, but in NSCLC they occur in 0.3% to 3% of patients. The two therapies that are furthest along in development for these oncogenes are entrectinib and LOXO-101.
The oral, selective TRK inhibitor LOXO-101 was recently investigated in a phase I ongoing dose escalation study of advanced or metastatic solid tumors with NTRK gene fusions, said Riess. Out of the 6 patients included in the study, 5 had responses, with an ORR of 83%. There is a phase II basket trial exploring LOXO-101 that is currently enrolling for patients with TRK gene fusions (NCT02576431).3
Other TRK inhibitors in development include DCC-2701, MGCD516, PLX7486, TSR-011, and XL-184. All of these are currently being investigated in ongoing clinical trials, Riess said.
Several methods are available for biomarker testing for TRK, including next-generation sequencing (NGS). The alteration has come to the forefront because of the increasing use of NGS, said Riess. “Oncologists need to do a comprehensive NGS panel and if it’s positive for TRK refer these patients for clinical trials,” said Riess. “These are potentially actionable”
References:
<<<
View more from the 2016 International Lung Cancer Conference