Acquired resistance to tyrosine kinase inhibitors targeting <em>EGFR</em> mutations in patients with non–small cell lung cancer are leading to next-generation therapies equipped to circumvent the mutations that arise from initial treatment. A review of these mechanisms, and the latest agents being developed to address them, shows that the pipeline holds promise for the future.
John V. Heymach, MD, PhD
Acquired resistance to tyrosine kinase inhibitors (TKIs) targetingEGFRmutations in patients with nonsmall cell lung cancer (NSCLC) are leading to next-generation therapies equipped to circumvent the mutations that arise from initial treatment.
Among targeted agents, EGFR TKIs have maintained the spotlight since first-generation agents gefitinib (Iressa), erlotinib (Tarceva), and afatinib (Gilotrif) showed a leap forward in efficacy. For patients with advanced disease, median life expectancy with EGFR TKI therapy ranges from 20 to 30 months, compared with less than 1 year prior to their introduction, when platinum-based chemotherapy was the best available option.1,2
“The era in which chemotherapy was the only option for [patients with NSCLC is] drawing to a close,” John V. Heymach, MD, PhD, chair of thoracic head and neck medical oncology at The University of Texas MD Anderson Cancer Center, said at the 2018 American Society of Clinical Oncology Annual Meeting.3“Virtually all [patients with NSCLC] can receive a nonchemotherapy regimen with immunotherapy or, if they have a driver mutation, with an appropriate targeted agent.”
Benefit from EGFR-targeted TKIs requiresEGFRmutation positivity, and these benefits may be halted by the development of resistance mechanisms. For virtually all patients, resistance is inevitable, and disease progression often occurs within 1 to 2 years of starting a TKI.4
Efforts to overcome resistance have defined the recent landscape of TKI research, from second-generation to now third-generation agents and combination regimens. Compared with those first-generation agents, recent results show improved response rates and extended median overall survival (OS), with potential to overcome previously untreatable resistance mechanisms. A review of these mechanisms, and the latest agents being developed to address them, shows that the pipeline holds promise for the future.
Generally, there are 2 types of resistance to TKIs: intrinsic, present prior to targeted therapy; and acquired, developing during treatment and typically appearing after a period of initial clinical benefit. Intrinsic, or acquired, resistance is usually conferred by a mutation, although alternative pathway activation, phenotypic changes, or histologic changes can also play a role.5 Some of these phenomena are poorly understood, while other resistance mechanisms have relevant agents readily available for treatment or in development.
The most common mechanism of acquired resistance is the T790M substitution mutation at exon 20 (FIGURE), occurring in approximately half of patients treated with TKIs.6The high prevalence of acquired resistance from T790M mutations prompted the development of third-generation TKIs, which are active against exon 19, exon 21, and T790M mutations.7First- and second-generation agents are ineffective against the T790M mutation for 2 reasons: first, the larger size of methionine, compared with threonine, interferes with target binding; and second, increased affinity of EGFR for adenosine triphosphate (ATP) increases likelihood that EGFR binds to ATP instead of the intended TKI.8,9The second most common acquired resistance mechanism encountered with first- and second-generation TKIs isMETgene amplification, which occurs in about 1 in 5 patients.10
As usage of first- and second-generation TKIs led to awareness of the T790M mutation andMETamplification resistance mechanisms, the FDA approved the third-generation TKI osimertinib (Tagrisso) in 2015 for patients whose tumors are positive for theEGFRT790M mutation.11Unfortunately, this and other third-generation agents are associated with another common acquired resistance mechanism, C797S. The C797S point mutation also occurs at exon 20, but in contrast with T790M, resistance to third-generation TKIs is due to blocked covalent binding at the cysteine residue. Prevalence remains unclear, although published data suggest that approximately one-third of patients treated with osimertinib are affected.12Unfortunately, research also suggests that the C797S mutation confers resistance to all other third-generation TKIs.13
Discussion of the T790M and C797S mutations that are encountered when treating patients with existing EGFR-targeted TKIs brings to mind the evolving paradigm in the age of targeted cancer therapies: If each generation leads to more resistance mechanisms, then how can resistance be overcome? Second- and third-generation TKIs remain in the spotlight; however, efforts to develop the next generation of TKIs are clearly seen in recent trials exploring combination approaches that may mitigate or overcome resistance. Although existing methods may not be able to beat all resistance mechanisms, they can provide a clearer picture of the complex molecular landscape inherent to NSCLC, which may someday give rise to new methods of overcoming resistance entirely.
About 1 in 10 patients withEGFR-mutant NSCLC have an in-frame insertion at exon 20 ofEGFR.14This mutation confers intrinsic resistance to TKIs via steric hindrance at the kinase binding pocket, resulting in very low treatment response rates that have historically been approximately between 3% and 8%. Owing to its greater flexibility and smaller size than other TKIs, the investigational agent poziotiniban inhibitor of both EGFR and HER2 kinases—escapes steric hindrance and may be an effective treatment option for this patient population.
A recent phase II trial with poziotinib, involving 50 patients withEGFR-mutant NSCLC, demonstrated a best response rate of 55% and a confirmed objective response rate of 43%, greatly exceeding previous responses.15Many had been heavily pretreated with either chemotherapy or PD-1/PD-L1 inhibitor therapy (86% and 54%, respectively). Heymach, principal investigator on the study, presented findings at the International Association for the Study of Lung Cancer 19th World Conference on Lung Cancer (WCLC) in September 2018 in Toronto, Canada.
“Durable responses have been observed, with 6 treated patients already treated for a year thus far,” Heymach said. “This compares favorably with historical response rates of less than 8% for approved TKIs and less than 19% for standard-of-care, second-line agents.”15
“This encouraging activity has prompted a confirmatory, international, multicenter study inEGFRandHER2exon 20 mutant patients, which is currently enrolling, and now includes a first-line cohort, as well as development of a separate pan-tumor basket study,” Heymach added.
Adverse events (AEs) were manageable and similar to those of other TKIs, including paronychia, diarrhea, and rash.
In September 2018, the FDA approved dacomitinib (Vizimpro) as a first-line agent for advanced NSCLC withEGFRexon 19 deletion or exon 21 L858R substitution mutations as detected by an FDA-approved test. Approval was based on results from the phase III ARCHER 1050 trial; dacomitinib was the first second-generation TKI to demonstrate superiority over first-generation agents, when their efficacy was compared head-to-head.16
The trial involved 452 patients with unresectable disease who had no prior therapy or had been without systemic therapy for at least 12 months. Patients exhibitedEGFRexon 19 deletion or exon 21 L858R substitution mutations. Following randomization, patients were randomized 1:1 to receive either dacomitinib or standard treatment with gefitinib. Progression-free survival improved from 9.2 months for patients treated with gefitinib to 14.7 months for patients treated with dacomitinib (P<.0001).17The median OS in patients treated with dacomitinib was 34.1 months, compared with 26.8 months for gefitinib (HR, 0.76; 95% CI, 0.582-0.993; 2-sidedP= .044).18
On the negative side, the increased kinase inhibition provided by dacomitinib comes with lower tolerability. In the ARCHER 1050 trial, 66.1% of patients receiving dacomitinib required a dose reduction, compared with just 8.0% of the cohort receiving gefitinib.16
Heymach reflected on this trade-off: “It’s been nearly 15 years since EGFR-targeted therapies were introduced, helping extend survival for thousands of patients in the time since,” he said. “The second generation of these therapies is more effective, but can also cause greater [adverse] effects, so patients and their doctors will need to weigh the risks and benefits.”
Early results from a phase I/II trial show promise for CK-101 (RX518), an emerging third-generation TKI. Developers at Checkpoint Therapeutics suggest that CK-101 may provide greater tolerability than osimertinib with similar efficacy.19 CK-101 is designed to be a first-line agent for classic exon 19 and exon 21 mutations, and a second-line agent against the T790M mutation.
Investigators presented phase I/II results at the 2018 WCLC meeting.19During the dose-escalation and expansion portions of the trial, 37 patients withEGFR-mutant advanced NSCLC received CK-101.20 Patients were eligible if they had no prior experience with an EGFR-targeted TKI or exhibited a T790M mutation and experienced disease progression with an EGFR TKI. Results suggested high tolerability; the most common AEs were nausea (16%), diarrhea (14%), increased lacrimation (14%), and vomiting (11%). Most of these were grade 1/2, with the exception of 1 patient who experienced grade 3 diarrhea. No grade 4 AEs were reported.
During the dose-expansion phase, 19 patients who received 400 mg of CK-101 twice daily and were evaluable for response achieved a partial response rate of 42% (n = 8). Response rates were higher in the group of 8 treatment-naïve patients, among whom 75% achieved a partial response. Compared with osimertinib, these early results suggest that CK-101 may have similar response rates and fewer AEs, but accurate comparison requires results from a phase III trial, which is currently planned.
Similar to osimertinib and CK-101, nazartinib (EGF816) is a third-generation TKI selective for classic activating mutations and T790M mutations. Preliminary phase II trial results showed the benefits of nazartinib in 40 patients, who received the agent for a median duration of exposure of 16.1 weeks.21The most frequent AEs were maculopapular rash (30%), diarrhea (28%), and stomatitis (23%). A quarter of the patients (25%) experienced grade 3 AEs. No grade 4 AEs were reported. Responses to nazartinib were evaluated in 24 patients with an unconfirmed overall response rate (ORR) of 75%, and confirmed ORR of 67% as assessed by blinded independent central review. Both assessments were made up of 1 complete response to treatment and 15 partial responses, with 2 additional partial responses awaiting confirmation. The investigators noted that response rates were high and the agent was well tolerated, despite the fact that 40% of patients had brain metastases. As with CK-101, more reliable comparisons with osimertinib depend on phase III trials.
Some emerging agents againstEGFR-mutant NSCLC are being developed for use in combination with existing EGFR-targeted therapies to overcome or slow TKI resistance. Among the leaders in this arena is DS-1205, being developed by Daiichi Sankyo. DS-1205 is a highly selective inhibitor of AXL, a cell surface receptor in the TAM family of kinases, that is being evaluated in phase I trials as combination therapy with gefitinib and osimertinib, seperately, for treatment of metastatic NSCLC.22,23This novel strategy is based on research suggesting AXL upregulation may be an unaddressed mechanism of TKI resistance.
In 2017, investigators using an HCC827 model showed that adding DS-1205 to osimertinib or erlotinib delayed resistance to either agent in a dose-dependent manner. In an erlotinib-resistant model, adding DS-1205 resensitized the tumor to erlotinib. These promising results further support the role of AXL in TKI resistance, and in October 2018, Daiichi Sankyo announced that the first phase I trial has begun in humans.23DS-1205 will be studied in combination with gefitinib in patients who have experienced disease progression during treatment with erlotinib, gefitinib, or afatinib or developed disease while on osimertinib and do not have a T790M mutation, a patient population with an unmet need.
BLU-667 targets the oncogenic driver RET, a specific mechanism of resistance activated in multiple cancers, including NSCLC.24In NSCLC,RETfusions appear relatively uncommonly, occurring in approximately 1% to 2% of patients.25A proof-of-concept study showed that BLU-667 overcame resistance to osimertinib in 2 patients with advanced NSCLC who had acquiredCCDC6-RETfusions.26 Following disease progression, 8 weeks of treatment with a combination of osimertinib and BLU-667 resulted in tumor reductions of 78% in both patients. The combination was well tolerated, without either patient experiencing grade 3 or higher AEs. Treatment is ongoing.
Although targeted therapies are now standard of care forEGFR-mutated NSCLC, success is brief and the agent pipeline has shifted toward overcoming resistance mechanisms. Third-generation TKIs, such as the FDA-approved osimertinib, improve tolerability, but even these meet resistance via C797S and other point mutations.
Fourth-generation agents to address C797S mutations are on the horizon, and investigators are focused on investigating novel combination therapies to join DS-1205 and BLU-667. These next-generation combinations target separate oncogenic drivers entirely, thereby overcoming resistance in a different way.
Second-generation TKIs awaiting approval, such as the novel agent poziotinib that tackles exon 20 in-frame insertions, may be able to help treat a patient population that currently has poor response rates.
Research in the field ofEGFR-mutated NSCLC is fast-paced and promising. As response rates improve, so does understanding of the complex molecular network that drives NSCLC.