Neelesh Sharma, MD, PhD

First-generation, reversible, epidermal growth factor receptor (EGFR) tyrosine kinase inhibitors (TKIs) such as erlotinib and geftinib have become the standard of care and first-line treatment option for patients with non-small cell lung cancer (NSCLC) with activating

mutations. Despite initial clinical benefit, overall efficacy of these agents is limited by emergence of drug-resistance mutations, including the gatekeeper

mutation. Substitution of methionine for threonine at position 790 of

leads to increased affinity for adenosine triphosphate (ATP), causing resistance to competitive inhibition by reversible EGFR-TKIs. Second-generation, irreversible EGFR-TKIs were developed to overcome this resistance and demonstrated significant preclinical activity in mouse models of

-mutant lung cancer. However, these agents had limited success in clinical trials because of concurrent inhibition of wild-type EGFR and associated toxicity that prevented achieving adequate drug levels in plasma for sufficient inhibition of EGFR T790M. CO-1686 and AZD9291 are third-generation, irreversible

mutant-specific EGFR-TKIs with little activity on wildtype EGFR and robust activity in T790M preclinical models. In ongoing phase I clinical trials, both agents have shown promising responses in patients with EGFR-TKI-resistant tumors carrying

Developments in last decade have established a central role for epidermal growth factor receptor (EGFR) pathway in the treatment of non-small cell lung cancer (NSCLC). Initial observation that EGFR protein is highly expressed in various tumors, including lung cancer, led to the development of first-generation small molecule EGFR tyrosine kinase inhibitors (TKIs). Advances in structure-activity relationship (SAR) and molecular modeling made it easier to generate novel molecules that were complementary in shape and electrostatics to the EGFR kinase domain topography.

A major breakthrough in this field was the discovery of 4-anilinoquinazolines (gefitinib, erlotinib, and lapatinib), which are potent and selective but reversible inhibitors of EGFR. The quinazoline ring binds in the adenine pocket with hydrogen bonds and the anilino ring binds in an adjacent, unique hydrophobic pocket, causing competitive inhibition for adenosine triphosphate (ATP) binding in the tyrosine kinase domain.

Phase I and II clinical trials with reversible firstgeneration EGFR-TKI (erlotinib, geftinib) revealed dramatic responses in some patients with lung adenocarcinoma who were never smokers, women, or of Asian origin.

Later, it was recognized that these patients had activating mutations in

gene, which was predictive for higher response rates (RR) and survival with EGFR inhibitors.

mutations depends on the histology and ethnicity; it can be as high as 30% to 40% in an East Asian population with adenocarcinoma, and as low as 15% to 20% in Caucasians.

More than 145 different types of nucleotide changes have been reported within the EGFR kinase domain, but most clinically relevant

mutations occur within the first 4 exons encoding the ATP-binding pocket of the kinase domain (exons 18-21).

Deletions in exon 19 that eliminate a common amino acid motif (LREA) and point mutation in exon 21 that involves the substitution of an arginine for a leucine at codon 858 (also known as L858R), account for approximately 85% of

mutations in lung cancer. These mutants preferentially bind to EGFR-TKI as demonstrated by their higher Km for ATP and a lower Ki for erlotinib and gefitinib.

In the last few years, multiple clinical trials have demonstrated that first-line treatment with EGFR-TKIs in lung adenocarcinoma patients with

mutations is superior to traditional platinum-based chemotherapy in terms of objective response rate (ORR), progression-free survival (PFS), and tolerability.

At present, EGFR-TKI has become the standard of care first-line treatment for this subset of NSCLC.

Despite initial clinical benefit with first-generation, EGFR TKIs, such as erlotinib and geftinib in first-line treatment of NSCLC, overall efficacy of these agents is limited by emergence of drug-resistance mutations, including the gatekeeper

Second-generation, irreversible EGFR-TKIs were developed to overcome this resistance and demonstrated significant preclinical activity in mouse models of

These agents had limited success in clinical trials because of concurrent inhibition of wild-type EGFR and associated toxicity that prevented achieving adequate drug levels in plasma for sufficient inhibition of EGFR T790M.

CO-1686 and AZD9291 are third-generation, irreversible

mutant-specific EGFR-TKIs with little activity on wildtype EGFR and robust activity in T790M preclinical models.

Despite a rapid and profound response, almost all patients receiving reversible first-generation EGFRTKI for the treatment of

-mutant NSCLC eventually progress because of emergence of acquired resistance. Several mechanisms of acquired resistance have been described (

), the most common (~50%) being mutation of the gatekeeper T790 residue (

mutation results in an amino acid substitution at position 790, from a threonine (T) to a methionine (M). Threonine 790 is critical for the binding of anilinoquinazoline inhibitors to the EGFR because of its key location at the entrance to a hydrophobic pocket in the ATP binding cleft. Kobayashi et al introduced T790M substitution into the sequence of the wild-type EGFR,

(a frequently identified deletion mutant) and the

(the most common point mutation). 17 These constructs with or without T790M were transiently transfected in COS-7 and NIH-3T3 cells. All construct pairs demonstrated identical levels of expression of total EGFR and phosphorylated EGFR, suggesting that the presence of the

mutation did not alter the production, degradation, activation, or deactivation of the EGFR protein. In contrast, when these cells were treated with gefitinib, the constructs without T790M were completely inhibited at a concentration of 20 nM, whereas constructs carrying the T790M substitution were resistant to concentrations of gefitinib as high as 2 &mu;M. Interestingly, CL-387,785, a specific and irreversible anilinoquinazoline EGFR inhibitor, strongly inhibited EGFR phosphorylation in constructs with T790M with a half-maximal inhibitory concentration (IC50) of 3.0 nM.

Multiple observations indicate that in addition to causing drug resistance,

mutations may also provide growth advantage. Somatic

mutations have been described at the time of diagnosis of lung cancer in the absence of previous EGFR-TKI treatment.

mutation has been reported in a family with a hereditary predisposition to lung cancer, suggesting that this mutation may be involved in development of lung cancer.

Transgenic mice with induced expression of EGFR T790M in mouse lung epithelia develop lung adenocarcinomas.

TABLE 1. Mechanisms of Acquired Resistance to EGFR-TKI

Acquired secondary mutations in tyrosine kinase domain (~60%, mostly

Activation of other receptor tyrosine kinases (eg,

mutation in ABL, which causes resistance to most of the BCR-ABL1 TKIs used for the treatment of chronic myeloid leukemia (CML). It was initially thought that similar to

, methionine substitution for threonine introduces a bulkier amino acid side chain, resulting in steric hindrance with binding of reversible EGFR-TKIs.

substitution in CML imparts resistance to all the tested TKIs, the corresponding

mutation is inhibited in vitro by multiple irreversible EGFR-TKIs (CL-387,785, EKB-569, and HKI-272) that closely resemble the anilinoquinazoline inhibitors. The fact that irreversible inhibitors that form a covalent bond with Cys- 797 at the edge of the ATP-binding cleft can bind and inhibit T790M does not support the steric hindrance as a mechanism of resistance.

Yun et al have described the structural and enzymologic effects of the

mutation in the context of both the wild-type and the

mutant binds geftinib in low nanomolar affinity, which is only slightly weaker than that for

mutant. This small difference in gefitinib binding because of

mutation is unlikely to explain the significant differences in sensitivity. Crystal structure of

mutant in combination with the inhibitor also does not support steric hindrance to inhibitor binding. Further analysis of ATP affinity of wildtype and mutant

mutant. T790M mutation nearly restores the ATP affinity to wild-type levels in the L858R/T790M double mutant. Reversible first-generation inhibitors such as erlotinib compete with ATP for binding to the kinase active site; the enhanced ATP affinity caused by

mutation reduces potency of these drugs. Irreversible EGFR inhibitors can overcome T790M resistance through covalent binding because they are not in a competitive, reversible equilibrium with ATP.

A number of quinazolinebased, second-generation, irreversible EGFR inhibitors are under evaluation, such as BIBW-2992 (afatinib), HKI-272 (neratinib), PF00299804 (dacomitinib), CL387785, EKB- 569, and CI-1033.

These compounds are also ATP mimetic but covalently bind Cys-797 of EGFR and are thus considered &lsquo;irreversible&rsquo; inhibitors. The irreversible covalent binding leads to longer and more potent suppression of kinase activity than do reversible TKIs since the activity is suppressed until the synthesis of new receptors. Many of these inhibitors have shown promising preclinical activity in

-mutant lung cancer cell lines and mice models. In a cell-free system, afatinib is 100-fold more active against the geftinib-resistant double-mutant EGFR

Engelman et al showed that PF00299804, an irreversible pan- EGFR inhibitor, has significant activity in NSCLC cell lines harboring endogenous

mutations, cell lines engineered to express EGFR T790M in cis to an activating mutation and in their xenograft model.

HKI-272 is very effective in NCI-H1975 cell line, harboring both L858R and T790M, as well as in NSCLC cell lines selected for geftinib resistance.

Preclinical efficacy of second-generation, irreversible EGFR-TKIs has translated into only modest clinical efficacy in patients with NSCLC who are resistant to erlotinib or geftinib (

HKI-272 is an irreversible pan-EGFR receptor TKI, which is the first second-generation EGFR-TKI examined in a large-scale study of patients with NSCLC who had previously benefited from an EGFR-TKI such as gefitinib or erlotinib.

In this phase II trial, 137 patients with NSCLC were treated in 3 arms. Patients in arm A and B had progressed after at least 12 weeks of prior EGFR-TKIs and arm C included patients who were TKI-na&iuml;ve. Patients were placed in arm A if they were

-mutation positive, in arm B if they were wild-type, and in arm C if they were light smokers and had adenocarcinoma. Overall, HKI-272 had low efficacy, with 3.4% RR in arm A, 0% RR in arm B or C, and median PFS of 15.3 weeks (90% confidence interval [CI]: 14.7-15.9 weeks) for all patients. Twelve patients (7%) had

mutation, including 11 from arm A with prior TKI exposure and one patient who was TKI-na&iuml;ve from arm C. None of the patients with

mutation had response. All patients initially received neratinib at 320 mg daily, which caused significant diarrhea (grade 3 in 50% of patients); therefore the dose was subsequently reduced to 240 mg with decrease in grade 3 diarrhea to 25%. Pharmacokinetic data were not available for patients on this study but based on the previous phase I study, lowering the dose for excessive diarrhea might have decreased drug bioavailability below the threshold required to inhibit T790M (based on preclinical models).

Afatinib is an orally bioavailable, pyrimidine-based, irreversible inhibitor of EGFR, HER2, and HER4.

It has gained FDA approval for clinical use in the firstline treatment of patients with NSCLC who have the

mutation. Afatinib was evaluated in LUX-Lung 1, a randomized, placebo-controlled, phase IIb/III trial in patients with NSCLC who had progressed after at least 12 weeks of treatment with erlotinib or gefitinib.

Patients were not required to have been tested for

mutation to enter the study. However, the requirement for at least 12 weeks of treatment with EGFR inhibitors served as an enrichment strategy for patients with

mutations and acquired resistance. There was no difference in the median overall survival (OS) between the two groups. Median OS was 10.8 months in the afatinib group (95% CI: 10.0-12.0) versus 12.0 months (10.2- 14.3) in the placebo group (hazard ratio [HR] = 1.08; 95% CI: 0.86-1.35; P = .74). Median PFS was longer in the afatinib group than in the placebo group (3.3 months vs 1.1 months; HR = 0.38; 95% CI: 0.31- 0.48; P <.0001). The confirmed disease control rates (DCRs; ORR plus stable disease) for &ge;8 weeks for afatinib was 58.2% compared with placebo, DCR of 18.5% (P <.0001). Only 8 patients had known

mutation, and responses were not reported in the study. Diarrhea and rash were the most common severe adverse effects (AEs) observed in patients who received afatinib, requiring dose reduction in 21% and 15 % patients, respectively. Incidence of diarrhea in the afatinib arm was 86.9%, of which 16.9% were of grade 3 intensity, based on common terminology criteria for adverse events (CTCAEs). Any grade rash/acne was observed in 76.7% of patients receiving afatinib and grade 3 rash was reported in 14.4% of patients.

Similar to the LUX-Lung 1 trial, the LUX-Lung 4 trial evaluated efficacy of afatinib in Japanese patients with NSCLC in third-line and fourth-line settings after progression on erlotinib and/or gefitinib.

The study enrolled 61 patients; 72.6% had an

-mutant tumor, and 82% of patients fulfilled Jackman&rsquo;s criteria of acquired resistance to EGFRTKIs. The results were consistent with the LUX-Lung 1 trial. In patients meeting the criteria for acquired resistance, median PFS was 4.4 months with partial response (PR) of 5.9%, and DCR of 68.6%. Two patients had acquired

mutations and achieved stable disease, 1 for 9 months, and the other for 1 month. Diarrhea and skin rash were the most commonly reported AEs occurring in almost all patients. Grade 3 diarrhea occurred in 37.1% of patients and grade 3 rash occurred in 27.4% of patients.

Dacomitinib is an oral, irreversible, small-molecule inhibitor of HER1/EGFR, HER2, and HER4 TKIs with antitumor activity in both gefitinib-sensitive and gefitinib-resistant preclinical NSCLC models. A phase II trial examined efficacy and safety of dacomitinib in patients with KRAS wild-type NSCLC who progressed after 1 or 2 chemotherapy regimens and erlotinib.

Patients with adenocarcinoma as well as squamous histology were included. Among

mutation-positive patients, RR was 8% and PFS was 4.5 months. Median PFS for overall population was 12 weeks. Among 6 patients with known

; 3 had stable disease for more than 9, 12, and 12 weeks, respectively, and 3 had progressive disease. Diarrhea (grade 3, 12%), acneiform rash (grade 3, 6%), and fatigue (grade 3, 3%) were the most common treatment-related AEs. Patients with radiographic response to treatment also had improvement in NSCLC-related symptoms of dyspnea, cough, and pain (chest, arm/shoulder), first observed after 3 weeks on therapy. An ongoing phase II study is evaluating efficacy of dacomitinib in 2 groups of patients with NSCLC, those whose tumor has a documented

mutation and those without this mutation (NCT01858389).

BMS-690514 is a second-generation, reversible, multikinase inhibitor with activity on EGFR, HER2, HER4, and vascular endothelial growth factor receptors (VEGFRs) 1 to 3.

In a phase I-IIa study, BMS- 690514 was evaluated in patients who were erlotinib- na&iuml;ve (cohort A) or erlotinib-resistant (cohort B) with advanced NSCLC.

Maximum-tolerated dose (MTD) was determined to be 200 mg daily. DCR (&ge;4 months) and RR, respectively, were 43.3% and 3.3% (cohort A) and 22.6% and 3.2% (cohort B). Disease control was achieved in 7 of 10 (70%) patients with

) and 6 of 21 (29%) patients with wild-type EGFR. Tumor shrinkage was observed in 1 of 2 patients with

mutation. The most frequent treatment-related AEs were diarrhea and acneiform rash.

XL-647 is a second-generation, reversible inhibitor of multiple receptor TKIs, including EGFR, VEGFR2, HER2, and EphB4.35 In preclinical studies, XL-647 showed encouraging efficacy against tumors harboring T790M.

However, a phase II trial in patients with NSCLC who progressed after &ge;12 weeks treatment with erlotinib or gefitinib and/or those with a documented

, demonstrated a disappointing 3% RR and significantly worse PFS in patients with

Of 33 evaluable patients, 12 (36%) had tumors with

mutation. None of the patients with T790M mutation had objective response and 67% (8/12) had progression of disease as best response compared with 14% (3/21) of those without this mutation.

Afatinib showed significant activity in NSCLC cell lines carrying

mutation, but no meaningful response was seen with afatinib alone in transgenic mouse models of EGFR L858R/T790M and EGFR T790M.

Analysis of mRNA profile from the tumors in these models of lung cancer showed higher expression of EGFR ligands, amphiregulin, and epiregulin, in tumors compared with normal lung. Combined treatment with EGFR antibody, cetuximab and afatinib resulted in near complete response in T790M transgenic mouse model. Tumor lysates from the mice that received combined treatment showed marked depletion of both phospho-EGFR and total EGFR levels, consistent with the mechanism of the individual drugs. Based on these results, Janjigian et al conducted a pilot study of afatinib, 40 mg daily, and escalating-dose cohorts of biweekly cetuximab at 250 mg/m

in patients with EGFRmutant lung cancer who had progressed on prior EGFR- TKIs.

Combination of afatinib 40 mg daily and biweekly cetuximab 500 mg/m

was recommended for further phase II evaluation. No dose-limiting toxicities (DLTs) were observed. Grade 1/2 rash and diarrhea were common AEs. Objective responses were seen in 8 of 22 evaluable patients (36%, 95% CI: 0.17-0.59) and disease control was observed in all patients. Because of these encouraging results, this study was expanded, and updated results were presented at the 2012 meeting of the European Society for Medical Oncology (ESMO).

Among 90 patients evaluable for efficacy, DCR was 94% and PFS at 3, 6, and 9 months were 70%, 42%, and 18%, respectively. The OR rate was 40% (95% CI: 27.6-53.5), similar in both T790M+ (38%) and T790M (47%) tumors. The median PFS was 4.7 months, and the median duration of response was 7.7 months. Most common AEs included rash (grade 1/2, 65%; grade 3, 12%) and diarrhea (grade 1/2, 63%; grade 3, 6%).

Acquired resistance develops in some xenograft models of T790M after chronic exposure to afatinib and cetuximab combination.

mutations were detected in these resistant cells. Analysis of resistant tumors by array comparative genomic hybridization (aCGH) revealed additional amplification at the EGFR locus, and immunoblotting of cell lysates showed increased EGFR protein expression. Interestingly, resistant cell lines displayed in vitro sensitivity to AZD9291, a third-generation, irreversible mutant-specific EGFR- TKI.

Third-Generation Nonquinazoline-Based EGFR-TKIs

TABLE 2. Second- and Third-Generation EGFR-TKI in Patients With NSCLC Who Have Acquired Resistance to Erlotinib or Geftinib

Articles

Despite initial clinical benefit of erlotinib and geftinib, overall efficacy of these agents is limited by emergence of drug-resistance mutations, including the gatekeeper T790M mutation.

Neelesh Sharma, MD, PhD | Authors

Articles

Novel Treatment Options for <em>T790M</em>-Mutant Non-Small Cell Lung Cancer