Treating sensitizing EGFR-mutant non-small cell lung cancer (NSCLC) with tyrosine kinase inhibitors (TKIs) leads to better outcomes compared with standard chemotherapy.
Hongbin Chen, MD, PhD
Assistant Professor of Oncology
Department of Medicine
Roswell Park Cancer Institute
Buffalo, New York 14263
Treating sensitizingEGFR-mutant non-small cell lung cancer (NSCLC) with tyrosine kinase inhibitors (TKIs) leads to better outcomes compared with standard chemotherapy. However, acquired resistance to first-generation EGFR TKIs, primarily mediated by the secondary gatekeeperT790Mmutation, results in disease progression. Previous attempts with subsequent-generation irreversible nonselective EGFR TKIs have not been successful in overcoming this resistance. Efforts to overcome resistance using more selective and potent TKIs are under way. Preliminary data from early-phase clinical trials are encouraging of mutant-specific third-generation EGFR TKIs that targetT790Mand otherEGFRmutations in NSCLC while sparing wild-type EGFR.
As a receptor tyrosine kinase, EGFR belongs to the ErbB family, which includes 4 members: ErbB1/ EGFR, ErbB2/HER2, ErbB3, and ErbB4. Ligand binding of EGFR activates intracellular signaling cascades through homodimerization or heterodimerization.1It has been a decade since 3 different research groups first reported somatic mutations in the tyrosine kinase domain of theEGFRgene in non-small cell lung cancer (NSCLC).2-4Most of theseEGFRmutations occur between exon 18 and exon 21, which encode the adenosine triphosphate (ATP)- binding pocket of the tyrosine kinase domain. The 2 most commonEGFRmutations found in patients with NSCLC are deletions of exon 19, with elimination of amino acids leucine-arginine-glutamic acidalanine at positions 747-750 (also known as LREA), which accounts for about 45% of allEGFRmutations, and a substitution mutation in exon 21, with substitution of arginine for leucine at position 858 (also known as L858R), which accounts for about 40% of allEGFRmutations. These are considered actionable mutations that are sensitive to targeted therapy using EGFR tyrosine kinase inhibitors (TKIs).5Using multiplexed assays to genotype oncogenic drivers in 1007 patients, the Lung Cancer Mutation Consortium (LCMC) identified sensitizingEGFRmutations in 17% of tumors, consisting of 10% exon 19 deletions, 6% L858R, 0.5% G719X, and 0.5% L861Q.6Median survival was significantly longer among patients who had an oncogenic driver mutation and received genotype-directed therapy compared with those patients who had an oncogenic driver mutation but did not receive genotype-directed therapy.
Despite the initial efficacy in treatingEGFR-mutant NSCLC with first-generation EGFR TKIs, such as erlotinib (Tarceva) or gefitinib (Iressa), disease progression normally occurs after approximately 1 year, indicating the development of resistance to TKIs.7The most common acquired resistance mutation is theT790M“gatekeeper” mutation in exon 20, leading to substitution of methionine for threonine at position 790 in the kinase domain, which accounts for 50% to 60% of patients with disease progression while taking a first-generation EGFR TKI.8Efforts to overcome resistance using more selective and potent TKIs are under way. Data from early-phase clinical trials are emerging in mutant-specific third-generation EGFR inhibitors.EGFR TKIs were originally tested in the second-line and third-line treatment settings among unselected patients with advanced/metastatic NSCLC. In the BR.21 trial, erlotinib significantly prolonged overall survival (OS) and progression-free survival (PFS) with a better response rate (RR) than best supportive care in patients with stage 3B or 4 NSCLC.9EGFRmutation status was not checked at the time of this trial.10Only a small subset of patients had tumors tested later forEGFRmutation, which seemed to be predictive of response to erlotinib.11
The SATURN study demonstrated improved PFS with erlotinib as maintenance therapy in patients with nonprogressive disease following first-line platinum-doublet chemotherapy.12Although patients with activatingEGFRmutations derived the greatest PFS benefit from erlotinib maintenance therapy, significant PFS benefits were also observed with erlotinib in the wild-type (WT) EGFR subgroup.13The TAILOR study demonstrated the superiority of chemotherapy to EGFR TKI for second-line treatment in patients with WT EGFR tumors.14
During the past 5 years, several large phase III trials have demonstrated impressive efficacy of first-generation EGFR TKIs among patients with sensitiveEGFRmutations in advanced/metastatic NSCLC. Most of the trials, such as IPASS,15WJTOG3405,16NEJ002,17and OPTIMAL,18were conducted in an Asian population, while the EURTAC trial enrolled European patients.19It should be pointed out that these improvements were consistent across all of these trials regardless of the various chemotherapy regimens used as control groups in different studies or the ethnicity (Asian or European) and geographic locations. However, OS was not significantly different between the EGFR TKI and the chemotherapy arms, likely due to the confounding effects of high crossover to the EGFR TKI arm from the chemotherapy arm.20-23Based on the positive findings of the pivotal EURTAC trial, erlotinib has been approved for the first-line treatment of patients with metastatic NSCLC whose tumors harbor EGFR exon 19 deletions or exon 21 (L858R) substitution mutations; a companion diagnostic test for patient selection, cobas EGFR Mutation Test, has also been approved.24Afatinib (GILOTRIF) is an oral, irreversible, selective ErbB family EGFR TKI that targets ErbB1/EGFR, ErbB2/HER2, and ErbB4/HER4. It is an aniline-quinazoline derivative that covalently binds directly to the ATP-binding site in the kinase domains of EGFR, HER2, and ErbB4.25The efficacy of afatinib in NSCLC is being formally evaluated in the LUX-Lung clinical trial program.26LUX-Lung 3 and LUX-Lung 6, 2 multicenter, international, open-label, randomized phase III trials, have demonstrated improvement in PFS among patients with metastatic NSCLC whose tumors tested positive forEGFRmutations.27,28Health-related quality-of-life measures were also significantly better with afatinib.29,30In a pooled analysis of these 2 trials, afatinib significantly prolonged OS (27.3 vs 24.3 months) in the overallEGFR-mutant population.31However, the improved OS was observed only inEGFR-mutant cancer with exon 19 deletions (31.7 months with afatinib vs 20.7 months with chemotherapy), but not in cancer withL858Rmutation (22.1 months vs 26.9 months). This is the first analysis to demonstrate that genotype-directed therapy for patients withEGFRmutations can improve survival. In the absence of direct head-to-head trial data comparing efficacy of the first-generation with second-generation EGFR TKIs, afatinib is a viable treatment alternative to erlotinib in the first-line treatment ofEGFR-mutant NSCLC, for which the drugs currently have identical indications.32
In the second-line or third-line treatment setting after failure of erlotinib or gefitinib and chemotherapy, afatinib has not been effective at overcoming acquired resistance as a single agent.33,34In addition, the first-generation and second-generation EGFR TKIs target WT EGFR as well as mutant EGFR, resulting in frequent “on-target” toxicities, such as skin rash and diarrhea. This in turn prevents these drugs from reaching adequate blood concentrations to sufficiently inhibitEGFR T790M.
A phase Ib study of intermittent administration of high doses of afatinib as a means of achieving plasma levels active againstT790Mmutations is ongoing (NCT01647711). An in vitro model suggested that inhibition of EGFR by currently available second-generation EGFR TKIs may not be sufficient to physiologically prevent the emergence of cells that are still dependent on EGFR signaling.35Dual inhibition of EGFR with afatinib and cetuximab, an EGFR monoclonal antibody, successfully overcame T790M-mediated resistance in preclinical models.36
In a phase Ib study combining afatinib and cetuximab among 126 heavily pretreated patients with advancedEGFR-mutant lung cancer and acquired resistance to erlotinib/gefitinib, overall RR was 29% and was comparable inT790M-positive tumors (32%) andT790M-negative tumors (25%).37Median PFS was 4.7 months, and the median duration of the responses were 5.6 months for patients withEGFR T790M-positive tumors and 9.5 months for those withT790M-negative tumors. Grades 3 and 4 treatment-related adverse events (AEs) were noted in 44% and 2% of patients, respectively. The most common grade 3 AEs were rash (20%) and diarrhea (6%). This supports the preclinical hypothesis that a significant proportion of tumors in patients with acquired resistance to erlotinib/gefitinib remain dependent on EGFR signaling for survival, butT790M-independent mechanisms of resistance likely also exist.
Dacomitinib (PF-00299804) is another secondgeneration, irreversible pan-HER TKI. Although it demonstrated significantly improved PFS compared with erlotinib in a randomized phase II trial,38the phase III ARCHER 1009 trial recently reported that dacomitinib was not superior to erlotinib in secondline or third-line therapy of advanced NSCLC.39The randomized phase III BR.26 trial also failed to show OS benefit with dacomitinib compared with placebo in heavily pretreated patients.40Additionally, in another phase II study, dacomitinib revealed no activity againstT790M-mutant tumors.41TheT790Mmutation increases the ATP affinity of the oncogenicL858Rmutant and alters the drug binding within the ATP-binding pocket of EGFR, thus reducing the potency of any ATP-competitive TKI as the primary mechanism by which theT790Mmutation confers drug resistance.42The presence ofT790Mdefines a clinical subset of resistance to TKIs, but also bears a relatively favorable prognosis and more indolent progression in comparison with the absence ofT790M.43Treatment options for patients who developT790M-mutant NSCLC are limited, usually including cytotoxic chemotherapy or clinical trials. Recent development of more potent and highlyselective third-generation pyrimidine EGFR TKIs yielded exciting data in targetingT790M-mutant NSCLC as well asEGFR-sensitizing mutations, while sparing classical toxicities from inhibition of WT EGFR in the skin and gastrointestinal tract. Several such inhibitors, including rociletinib (CO-1686), AZD9291, and HM61713, have entered early-phase clinical study and will be discussed below.
Rociletinib (CO-1686) is a novel, irreversible, and orally delivered TKI that specifically targets the mutant forms of EGFR, includingT790M, while exhibiting minimal activity toward the WT receptor.44As a 2,4-disubstituted pyrimidine molecule, rociletinib covalently modifies the conserved cysteine-797 residue in the ATP-binding pocket of the EGFR kinase domain and forms high affinity forEGFR T790Mmutant.45It induces tumor regression in EGFR-mutant NSCLC tumor xenograft and transgenic models with minimal activity against the WT EGFR.
In a first-in-human dose-finding phase I/II study, rociletinib was administered twice daily in 72 patients withEGFR-mutated advanced NSCLC who were previously treated with an EGFR TKI, and tumor biopsies were screened for central EGFR genotyping. 46 The median age of patients was 59 years; 75% of patients were women and 14% were Asian. The median number of previous lines of therapy was 3, and the median number of previous EGFR TKIs was 2, with immediate prior TKI in 75% of patients. Rociletinib demonstrated dose-linear pharmacokinetics consistent across efficacious dose range with a half-life suitable for twice-a-day dosing. Treatment-related AEs occurring in >10% of patients were nausea, hyperglycemia/impaired glucose tolerance (IGT), diarrhea, vomiting, decreased appetite, myalgia, and QTc prolongation. Grade 3 hyperglycemia/IGT occurred in 22% of patients and was well managed with oral hypoglycemic agents. No grade 3/4 diarrhea was observed, and 3 patients (4%) experienced skin rash (all grade 1). Objective RR within therapeutic dose range was achieved in 58% of 40 patients with centrally confirmedT790Mmutation-positive tumors. Median PFS has not yet been reached, but the current estimate exceeds 12 months. Rociletinib has demonstrated promising efficacy against EGFR-activating andT790M-mutant NSCLC while sparing WT EGFR signaling across all dose levels in this phase I/ phase II early expansion cohort.
Rociletinib received the US Food and Drug Administration (FDA)’s breakthrough therapy designation in May 2014 for mutant NSCLC in patients with theT790Mmutation after progression on EGFR-directed therapy. The rociletinib development program, labeled TIGER (Third-generation Inhibitor of mutant EGFR in lung cancER), is investigating the drug as a second-line treatment in patients with EGFRT790Mmutations and as a first-line therapy in comparison with erlotinib (Table 1).47,48A companion diagnostic tool to identify patients whose tumors express the activatingEGFRmutation and theT790Mresistance mutation is also in development.
Click to view
TIGER-X, part of the TIGER program, is enrolling patients in this phase II expansion cohort study to evaluate rociletinib in 2 groups of patients. The first group enrolls patients directly after progression on their first and only EGFR-directed TKI therapy, such as erlotinib or afatinib, who have developed theT790Mmutation. The second group includes patients withT790M-positive disease after progression on their second- or later-line TKI therapy or subsequent chemotherapy.
Because of improved tolerance due to mutation selectivity over WT EGFR, coupled with activity againstT790Mand other mutations, rociletinib use in first-line therapy is being explored in TIGER1, a randomized phase II/III registration study versus erlotinib in newly diagnosed patients. The phase II portion of this study is expected to begin in late 2014.
TIGER2 is a single-arm, global registration phase II study under way in patients withT790M-positive disease directly after progression on their first and only TKI therapy. TIGER3 is planned as a randomized, comparative phase III study of rociletinib versus chemotherapy in patients with acquired TKI resistance in the second-line treatment setting ofEGFR T790M-mutant NSCLC.
AZD9291 is a novel oral and potent selective third-generation irreversible inhibitor of both EGFR mutation sensitizing andT790M-resistance mutants that spares WT EGFR.49This mono-anilino-pyrimidine compound is structurally distinct from other third-generation EGFR TKIs and offers a pharmacologically differentiated profile from earlier generation EGFR TKIs. In preclinical testing, the drug potently inhibits signaling pathways and cellular growth in bothEGFR-mutant andEGFR/T790M-mutant cell lines in vitro, with lower activity against WT EGFR cell lines, translating into profound and sustained tumor regression in EGFR-mutant tumor xenograft and transgenic models.
A phase I dose-escalation study (NCT01802632) of AZD9291 (mesylate salt form) in patients withEGFR-mutation-positive NSCLC and acquired resistance to EGFR TKIs reported results after enrollment of 31 and 201 patients into the escalation and expansion cohorts, respectively (median age 61/60 years; 65%/62% women; 71%/63% Asian).50At doses of 20 mg to 240 mg daily, the overall RR (confirmed and unconfirmed) was 53% (95% CI: 46%-60%) among all evaluable patients, 64% (95% CI: 55%-73%) in patients withEGFR T790-mutation-positive disease (N = 89), and 22% (95% CI:12%-36%) in patients withEGFR T790mutation-negative disease. The overall disease control rate in patients withEGFR T790mutation-positive disease was 94%, with no dose-limiting toxicities observed. The most common AEs were lowgrade diarrhea (30%), rash (24%), and nausea (17%), with grade 3/4 AEs occurring in 16% of patients and 6 patients requiring a dose reduction. There were 5 reports of interstitial lung disease, most of which (n = 4) occurred with a 160-mg dose of AZD9291 and were resolved without fatalities. AZD9291 appears to have robust efficacy and is well tolerated in patients withEGFRmutation-positive NSCLC and acquired resistance to EGFR TKIs. The response evaluation among patients withT790M-mutant tumors is especially impressive at this early stage.
Afatinib + cetuximab37
T790M+: 18.9 weeks
T790M-: 10 weeks
EGFR indicates epidermal growth factor receptor; NR, not reported/reached; NSCLC, non-small cell lung cancer; PFS, progression-free survival; TKI, tyrosine kinase inhibitor.
HM61713 is anEGFR-mutant selective inhibitor that showed strong anticancer activity in several EGFRmutant lung cancer cell lines, includingT790Mmutation- harboring cell line, with low activity against WT EGFR.51A phase I dose-escalation and expansion study of HM61713 (NCT01588145) assigned patients withEGFRmutation-positive advanced NSCLC to treatment arms according to time since EGFR TKI (Arm A: <4 weeks; Arm B: ≥4 weeks). Among 83 patients (all Asian) enrolled to date, median age was 59 years, with 62.7% female, 91.6% ECOG 0-1, and a median of 2 (1-7) prior lines of chemotherapy.T790Mmutation was positive in 48 patients (57.8%). Pharmacokinetics was dose linear over the dose range (75-500 mg/day). Treatment-related AEs were mostly grade 1/2; those occurring in ≥10% were nausea, skin exfoliation, headache, rash, decreased appetite, diarrhea, pruritus, constipation, dry skin, vomiting, productive cough, upper abdominal pain, cough, dyspepsia, and dyspnea. Overall RR was 21.7% and disease control rate was 67.5% (61.9% and 73.2% in Arm A and B, respectively). Response rate was 29.2% inT790mutation-positive patients and 11.8% inT790mutation-negative patients, with median PFS estimated to be 18.9 and 10 weeks, respectively. Therefore HM61713 showed good safety profile and promising antitumor activity in patients withEGFR-mutated NSCLC who failed to respond to EGFR TKIs, especially in patients withT790Mmutation. An additional expansion study will be opened at higher dose level in the third quarter of 2014.WZ4002 is a third-generation mutant-specific EGFR TKI with an anilinopyrimidine core.52In vitro studies showed that it was more potent againstEGFR T790M, and up to 100-fold less potent against WT EGFR, than quinazoline-based EGFR inhibitors.53In vivo studies indicated it was also effective in murine models of lung cancer driven byEGFR T790M. Combining WZ4002 with crizotinib, a MET inhibitor that has been approved for first-line therapy in anaplastic lymphoma kinase (ALK) rearrangementpositive NSCLC,54was effective in overcoming multiple resistance mechanisms to reversible EGFR TKI in vivo, including T790M.55However, no clinical trial of WZ4002 has been initiated so far.
Further development of AZD9291 directed atT790M-mutant NSCLC as a single agent or in combination with other agents is ongoing. A phase II, open label, single-arm study (AURA2, NCT02094261) of AZD9291 (80 mg orally, once daily) and a phase III, open label, randomized study (NCT02151981) of AZD9291 (80 mg orally, once daily) versus platinumbased doublet chemotherapy are currently accruing patients with locally advanced/metastatic NSCLC whose tumors have progressed after EGFR TKI and harbor aT790mutation.AP26113 is a second-generation ALK TKI that has been shown to be highly effective against crizotinibresistant tumors harboring secondary gatekeeper mutations.56AP26113 is also a dual ALK/EGFR inhibitor that has activity against EGFR T790M in cell and mouse models of NSCLC.57An ongoing phase I/ II study (NCT01449461) is enrolling patients with EGFR-resistant NSCLC as well as ALK-positive NSCLC.58Among 51 evaluable patients with ALK-positive NSCLC after prior crizotinib treatment, 35 (69%) responded. The most common treatmentemergent AEs (≥20%) were generally grade 1/2 in severity.
Immunotherapy using an antibody against programmed cell death 1 (PD-1) has shown activity in NSCLC.59Preclinical data support EGFR pathway activation of PD-1 ligand (PD-L1) expression and immune escape inEGFR-driven lung tumors.60Treatment ofEGFR-mutant NSCLC cell lines with an EGFR inhibitor down-regulates PD-L1 expression. A phase I study (NCT01454102) combined nivolumab, an anti-PD-1 antibody, and erlotinib in 21 patients with EGFR-mutant adenocarcinoma (33%T790M-mutant) and prior erlotinib treatment.61The objective RR was reported to be 19% (4/21) and median PFS was 29.4 weeks, with grade 3 treatment-related AEs occurring in 24% of patients (no grade 4 reported). The estimated duration of response was 60 weeks, and 15% of patients had stable disease. Nivolumab in combination with erlotinib may provide durable clinical benefit inEGFRmutation-positive advanced NSCLC, with evidence of activity at TKI resistance.
Recently, a high-throughput cancer cell line screening platform to profile the sensitivity of validated and investigational anticancer small compounds identified indolocarbazole compounds, including a clinically well-tolerated FLT3 (fms-like tyrosine kinase 3) inhibitor, midostaurin or PKC412 (protein kinase C 412), as potent and noncovalent reversible inhibitors ofEGFR T790Mthat spare WT EGFR.62Midostaurin is currently undergoing phase III clinical testing in patients with acute myelogenous leukemia with activatingFLT3mutations.
Other small molecule compounds designed to targetT790M-mutant tumor include ASP8273 (NCT02192697) and EGF816 (NCT02108964), both undergoing investigation in phase I clinical trials.47Treating sensitizingEGFR-mutant NSCLC with TKIs leads to better outcomes compared with standard chemotherapy. Inevitably, disease progression occurs, mostly as a result of acquired secondary mutation ofT790Mthat causes the resistance to first-generation EGFR TKIs. Even irreversible, second-generation EGFR TKIs have not been successful in overcoming this resistance, although recent analysis suggested survival benefits of afatinib in comparison with chemotherapy as front-line therapy. However, treatment-related AEs as class effects may limit the ability to maximize doses in order to overcome such resistance, even in the use of an afatinib and cetuximab combination. The development of third-generation EGFR TKIs that targetT790M mutant-specific NSCLC while sparing WT EGFR offers optimistic data to salvage treatment failure in genotype-selected therapy. The preliminary findings from early-phase clinical trials indicate that it should be possible to developEGFRT790Mmutant-specific inhibitors for which dosing is not limited by on-target toxicities, which may therefore be advantageous relative to currently available irreversible EGFR inhibitors (Table 2). These agents are generally well tolerated and common AEs, such as skin rash and diarrhea, are infrequent, but novel AEs, such as hyperglycemia, have been noted in clinical trials. These AEs are manageable, although the exact mechanism is not clear.
Both rociletinib (CO-1686) and AZD9291 have been granted breakthrough therapy designation by the FDA. The effectiveness of these agents likely provides an incentive to re-biopsy patients with acquired resistance to frontline EGFR inhibitors, as this may well be an eligibility requirement for participation in clinical trials examining these agents in theT790M mutant-enriched patient population or may be a specific indication upon eventual approval. Other noninvasive techniques utilizing cell-free plasma DNA are being explored as a possible approach to detectT790Mmutation.63Future clinical development of these agents with mature survival outcomes is crucial to confirm these early, yet consistently dramatic, efficacy signals and to select the leader in this class of TKIs based upon safety, tolerance, and efficacy. The differences in dosing schedules (twice daily in rociletinib vs once daily in AZD9291) and somewhat unique toxicity profiles (hyperglycemia and QTc prolongation in rociletinib vs interstitial lung disease in AZD9291) currently do not seem to have a major impact on further development. Head-to-head comparisons with earlier generations of EGFR TKIs, such as the TIGER 1 trial, will eventually help to determine the range of indications of these novel agents as well as the superiority in the first-line setting, as they have the advantage of targeting double mutants in both EGFR-sensitizing mutant andT790M-mutant NSCLC. It is thus equally important to address the issue of how to use these targeted agents in an appropriate sequence to optimize efficacy and deter resistance. The race to establish the role of these mutant-specific EGFR TKIs will be rewarding and beneficial to our patients.