Erlotinib and Trametinib in Patients With EGFR-Mutant Lung Adenocarcinoma and Acquired Resistance to a Prior Tyrosine Kinase Inhibitor

PURPOSE

Inhibition of the MEK/ERK pathway is critical for Bcl-2-like protein 11 (BIM)-mediated epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor (TKI)-induced apoptosis, and dysregulation of this pathway may be a mechanism of acquired resistance. Therefore, MEK inhibition with trametinib and an EGFR TKI may resensitize tumors with acquired resistance. Limited targeted therapies are available after progression on EGFR TKIs, and it is in this setting that we completed a phase I/II study of erlotinib and trametinib.

METHODS

Patients with metastatic EGFR-mutant lung adenocarcinoma and acquired resistance to an EGFR TKI received combination erlotinib 75 mg and trametinib 1.5 mg daily until progression or unacceptable side effects. The primary objective was objective response rate determined using RECIST version 1.1.

RESULTS

Twenty-three patients were accrued; patients had received a median of two lines of prior TKI therapy (61% prior osimertinib), and 48% had acquired EGFR T790M. We confirmed one partial response (1/23, 4%, 95% CI, 0 to 22). The median progression-free survival was 1.8 months, and the median overall survival was 21 months. Diarrhea (87%), acneiform rash (87%), and fatigue (52%) were the most common treatment-related adverse events. Two patients who had tumor shrinkage both harbored a BRAF fusion.

CONCLUSION

Addition of trametinib to erlotinib in the acquired resistance setting in an unselected population is not efficacious. Future studies should focus on targeted therapies in molecularly selected populations. Acquired BRAF fusions in patients with EGFR-sensitizing mutations may be a molecular subset where EGFR and MEK combination therapy could be studied further.

INTRODUCTION

Treatment of the 20% of lung cancers with epidermal growth factor receptor (EGFR)-activating mutations has been transformed by the approval of multiple generations of EGFR tyrosine kinase inhibitors (TKIs). Patients with these oncogene-addicted lung cancers have a 70% objective response rate (ORR), a 10-month median progression-free survival (PFS) with erlotinib¹ or afatinib,² and a 19-month PFS with osimertinib as first-line treatment.³ However, all patients develop resistance to EGFR TKIs and subsequent clinical progression. Although acquisition of an on-target EGFR T790M mutation is seen frequently with earlier-generation EGFR TKIs and can be treated with second-line osimertinib,⁴ there are no approved targeted therapies for EGFR T790M-independent resistance or after progression on osimertinib.

CONTEXT

• Key Objective

• In patients with metastatic epidermal growth factor receptor (EGFR)-mutant lung cancers and acquired resistance to an EGFR tyrosine kinase inhibitor (TKI), does combination erlotinib and trametinib result in tumor shrinkage and clinical benefit?

• Knowledge Generated

• In a heavily pretreated population, there was one partial response of 23 patients. Two patients who had tumor shrinkage both harbored a BRAF fusion.

• Relevance

• Addition of trametinib to erlotinib in the acquired resistance setting in an unselected population is not efficacious. Dose reductions and interruptions may have resulted in subtherapeutic levels of MEK inhibition. The use of newer-generation EGFR TKIs such as osimertinib in combination regimens will likely lead to improved tolerability and possibly increased efficacy. Acquired BRAF fusions in patients with EGFR-mutant lung cancers may be a molecular subset where EGFR and MEK combination therapy should be studied further.

One common pathway of resistance is dysregulation of the downstream mitogen-activated protein kinase (MAPK) pathway. Preclinical data have established the necessary^5,6 and sufficient^7,8 role of the pro-apoptosis regulator Bcl-2-like protein 11 (BIM) to induce cell death triggered by EGFR TKIs. Inhibition of the upstream protein MAPK/extracellular signal-regulated kinase (ERK) (MEK) induces BIM expression and may resensitize tumors to EGFR TKIs.^9-11 Inappropriate MAPK pathway activation also induces epithelial-to-mesenchymal transition (EMT), and inhibition of MEK reversed EMT and led to resensitization to EGFR therapy in preclinical models.¹² Also, low levels of the upstream tumor suppressor neurofibromin 1 (NF1) are associated with EGFR TKI resistance and result in incomplete inhibition of RAS/ERK signaling in the presence of EGFR TKI.¹³ Treatment of NF1-deficient lung cancers in a mouse model of lung adenocarcinoma with trametinib and erlotinib restored sensitivity to erlotinib. These converging data suggest the utility of erlotinib and trametinib treatment in EGFR-mutant lung cancers with acquired resistance to EGFR TKIs.

MATERIALS AND METHODS

Patients

Patients with advanced EGFR-mutant lung cancers who developed resistance to an EGFR TKI and had metastatic lung adenocarcinoma, an EGFR-sensitizing mutation, adequate organ function, and RECIST version 1.1 measurable disease were eligible for participation. Please refer to the protocol (Data Supplement) for comprehensive inclusion and exclusion criteria. All patients provided written informed consent to an institutional review board-approved protocol at Memorial Sloan Kettering Cancer Center (New York, NY).

Study Design

This was a single-institution, open-label, phase I/II clinical trial (ClinicalTrials.gov identifier: NCT03076164). The phase I portion was a safety run-in to determine the recommended phase II doses of erlotinib and trametinib. The starting dose of trametinib 1.5 mg and erlotinib 75 mg daily was based on a phase I clinical trial (ClinicalTrials.gov identifier: NCT01192165) of trametinib 1 mg and erlotinib 100 mg daily in solid tumors in which there were no responses in a molecularly unselected study population.¹⁴ Therefore, we hypothesized that a higher dose of MEK inhibition was needed, and a lower dose of erlotinib may still be efficacious. Six patients were enrolled in a safety lead-in to assess for dose-limiting toxicities (DLTs). After the phase II dose was identified, all patients were treated at that dose level. Computed tomography imaging was obtained at baseline and every 8 weeks for RECIST version 1.1 response assessment.

Molecular Profiling

Molecular profiling of tumors was done by multiple assays, including a targeted next-generation sequencing (NGS) custom hybridization capture-based assay (Integrated Mutation Profiling of Actionable Cancer Targets, MSK-IMPACT), a mass spectrometry genotyping assay examining specific hotspot mutations (Sequenom MassArray spectrometer), and polymerase chain reaction (PCR) using fluorescent probes to detect EGFR exon 19, 20, and T790M mutation status. PIK3CA mutation, EGFR/MET/HER2 amplification, and v-raf murine sarcoma viral oncogene homolog B1 (BRAF) fusion detection were assessed via MSK-IMPACT and anchored multiplex PCR (Archer FusionPlex Custom Solid Panel).

Statistics

The primary end point of the phase II portion of this study was ORR. A Simon optimal two-stage design was used to test a null response rate of 5% versus 25%, using 10% type I and II error rates. The first stage consisted of nine patients; at least one response was needed to proceed with the second stage. If three or more patients achieved objective responses, the combination would be worthy of further investigation. Secondary objectives included PFS, overall survival (OS), treatment-related adverse events per the National Cancer Institute Common Terminology Criteria for Adverse Events version 4.0, and association of response with mutation status. PFS and OS were calculated from the time of treatment initiation. Survival analyses for PFS and OS were performed using Kaplan-Meier time-to-event estimates. Patients who did not experience the event of interest at the database lock (May 1, 2020) were censored at the time of the last follow-up date.

RESULTS

Erlotinib and Trametinib Activity and Safety

From March 2017 to October 2019, 23 patients were enrolled on this study. The first six enrolled in the phase I portion (Appendix Fig A1). There were no DLTs identified with daily doses of 75 mg erlotinib and 1.5 mg trametinib, and an additional 17 patients were treated at these doses in the phase II study portion. Patients were heavily pretreated (median prior lines of treatment 2, range, 1-5; Table 1, Appendix Fig A2). Prior EGFR TKIs included erlotinib (74%, 17/23), afatinib with or without cetuximab (35%, 8/23), and osimertinib (61%, 14/23). About half of patients (48%, 11/23) had an acquired T790M mutation, and ten patients (43%, 10/23) had baseline brain metastasis.

FIG 1.

Duration of erlotinib and trametinib therapy. The duration of erlotinib and trametinib combination therapy in 23 patients with epidermal growth factor receptor (EGFR)-mutant lung adenocarcinoma. Boxes to the left represent baseline (a) type of EGFR-sensitizing mutation; (b) presence of a co-occurring TP53 mutation; (c) presence of a co-occurring RB1 mutation; (d) presence of an acquired EGFR T790M mutation; and (e) presence of a co-occurring PIK3CA mutation, EGFR/MET/HER2 amplification, or BRAF fusion. Blank spaces represent testing for co-occurring mutations was not performed. BOR was unknown for two patients who experienced clinical progression and one patient who withdrew consent prior to a follow-up scan. Patients 3, 8, 9, 11, 12, and 20 had EGFR, MET, EGFR, EGFR, EGFR, and HER2 amplifications, respectively. BOR, best overall response; PD, progressive disease; PR, partial response; SD, stable disease.

TABLE 1.

Baseline Patient Characteristics

FIG 2.

(A) Best response of target lesions from baseline. Best overall response to erlotinib and trametinib in 20 response-evaluable patients with epidermal growth factor receptor (EGFR)-mutant lung cancers. Patients not evaluable for response were not included in this plot. Bars represent the best percentage change in the sum of target lesions per RECIST version 1.1. Dashed lines represent cutpoints for RECIST-defined SD. Boxes below each bar represent baseline (a) type of EGFR-sensitizing mutation; (b) presence of a co-occurring TP53 mutation; (c) presence of a co-occurring RB1 mutation; (d) presence of an acquired EGFR T790M mutation; and (e) presence of a co-occurring PIK3CA mutation, EGFR/MET/HER2 amplification, or BRAF fusion. Blank spaces represent testing for co-occurring mutations was not performed. (B) Patient 20 achieved a RECIST best overall response of PR (48% reduction). This patient did not have prestudy mutation testing but was found to have a BRAF fusion identified on their immediate post-treatment biopsy. (C) Patient 13 harbored a BRAF fusion and achieved a mixed response, having developed tumor shrinkage in the chest (40% reduction) but a new lesion in the liver. Patients 2, 4, 5, 8, 11, and 14 had EGFR, HER2, MET, EGFR, EGFR, and EGFR amplifications, respectively. PD, progressive disease; PR, partial response; SD, stable disease.

Of the 23 patients, 20 had RECIST-evaluable disease as two patients discontinued therapy prior to on-treatment imaging because of clinical progression and one withdrew because of toxicity. The median time on treatment was 2 months (Fig 1). One confirmed partial response (PR) was observed (Fig 2). Twelve patients achieved stable disease (SD), and seven patients had progressive disease (PD) as their best overall response (BOR). The ORR was 4% (1/23), and median PFS was 1.8 months (Appendix Fig A3). A PFS > 6 months was seen in one patient (9.4 months, BOR SD) (Fig 2A).

All 23 patients were evaluable for toxicity. The most common treatment-related adverse events (grade 1-3) were rash (87%), diarrhea (87%), fatigue (52%), nausea (30%), and vomiting (30%) (Table 2). Most treatment-related adverse events were grade 1 and 2, with grade 3 rash and diarrhea seen in 3 (13%) patients each. Of 23 patients, 43% (10 patients) required dose reductions and 39% (9 patients) paused treatment for a week or longer. Four patients (17%) permanently discontinued therapy because of toxicity (for rash, diarrhea, left ventricular ejection fraction decrease, and fatigue).

TABLE 2.

Treatment-Related Adverse Events

Characteristics of Patients with Response

The type of EGFR-sensitizing mutation and most co-occurring mutations were not associated with objective response or duration of treatment.

Two patients exhibited tumor shrinkage: one achieved a PR (Fig 2B) and one had a mixed response¹⁵ (Fig 2C). One was a 49-year-old man with EGFR-mutant lung adenocarcinoma with bone metastasis diagnosed in 2015 who received erlotinib (7 months), with development of T790M, followed by osimertinib (13 months) before disease progression. He was on study for 5.5 months and had a BOR PR (48% reduction). Targeted NGS of a growing lung lesion using orthogonal DNA- and RNA-based tests at progression showed a class II hyperactivating ZC3HAV1-BRAF fusion (along with TP53 and PIK3CA mutations) that was likely a pre-existing resistance mechanism to prior therapy.^16-19 The other patient was a 60-year-old woman who was initially diagnosed with EGFR-mutant lung adenocarcinoma with brain metastasis in 2014. She received four cycles of cisplatin and pemetrexed followed by definitive therapy for oligoresidual disease. Upon recurrence (September 2016), she was treated with carboplatin, pemetrexed, and bevacizumab. After identifying an EGFR-sensitizing mutation, she switched to afatinib for 35 months prior to disease progression. Targeted NGS of the primary tumor site (lung) prior to study enrollment also identified a class II hyperactivating ERC1-BRAF fusion, TP53 mutation, and PIK3CA mutation. The BRAF fusion in this case was also confirmed by Archer RNA testing to confirm the fusion transcript. She was on trial therapy with erlotinib and trametinib for 1.7 months. She had RECIST PD with new liver metastasis but a mixed response with 40% shrinkage at the primary site.

DISCUSSION

In our study of patients with EGFR-mutant lung cancers and acquired resistance to EGFR TKI, most patients did not respond to treatment with trametinib and erlotinib. Interestingly, both patients who experienced tumor shrinkage had BRAF fusions. BRAF fusions are MAPK pathway-activating mutations and are found as acquired resistance mutations in approximately 2% of EGFR-mutant lung adenocarcinomas.¹⁹ Preclinical studies and case reports of BRAF fusion-positive cancers suggest sensitivity to direct downstream MEK inhibition.^20-22 With an increased ability to detect gene fusions (NGS and RNA-based assays), future studies of combination EGFR and MEK inhibition should enrich for acquired BRAF fusions.

Treatment-related adverse events were significant, although were not usually treatment-limiting. Dose reductions and interruptions may have resulted in subtherapeutic levels of MEK inhibition. The use of newer-generation EGFR TKIs such as osimertinib in combination therapies will likely lead to improved tolerability and efficacy. The TATTON study included an osimertinib and selumetinib arm that was well-tolerated and demonstrated preliminary efficacy.²³ These more tolerable combination therapies may be investigated as first-line treatment options to delay resistance to EGFR TKI therapy (ClinicalTrials.gov identifier: NCT03392246).¹¹

In conclusion, most patients with EGFR-mutant lung cancers treated with trametinib and erlotinib did not respond to this combination. The combination should only be evaluated further in molecular subsets such as acquired BRAF fusions where combination therapy has a rationale.

APPENDIX

FIG A1.

Consort diagram.

FIG A2.

Duration of prior lines of therapy. BOR, best overall response; PD, progressive disease; PR, partial response; SD, stable disease.

FIG A3.

Progression-free survival and overall survival.

AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/po/author-center.

Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).

Maria E. Arcila

Honoraria: Invivoscribe, Biocartis

Consulting or Advisory Role: AstraZeneca

Travel, Accommodations, Expenses: AstraZeneca, Invivoscribe, Raindance Technologies

Gregory J. Riely

Research Funding: Novartis, Roche/Genentech, GlaxoSmithKline, Pfizer, Infinity Pharmaceuticals, Mirati Therapeutics, Merck, Takeda

Patents, Royalties, Other Intellectual Property: Patent application submitted covering pulsatile use of erlotinib to treat or prevent brain metastases

Travel, Accommodations, Expenses: Merck Sharp & Dohme

Other Relationship: Pfizer, Roche/Genentech, Takeda

Mark G. Kris

Consulting or Advisory Role: AstraZeneca, Pfizer, Daiichi Sankyo

Travel, Accommodations, Expenses: AstraZeneca, Pfizer, Genentech, Daiichi Sankyo

Other Relationship: Genentech/Roche

Helena A. Yu

Consulting or Advisory Role: AstraZeneca, Lilly, Daiichi Sankyo

Research Funding: AstraZeneca, Astellas Pharma, Lilly, Novartis, Pfizer, Daiichi Sankyo

Travel, Accommodations, Expenses: Lilly

Other Relationship: Astellas Pharma

No other potential conflicts of interest were reported.

AUTHOR CONTRIBUTIONS

Conception and design: Sara A. Hayes, Maria E. Arcila, Gregory J. Riely, Mark G. Kris, Helena A. Yu

Financial support: Mark G. Kris, Helena A. Yu

Administrative support: Helena A. Yu

Collection and assembly of data: Jia Luo, Yosef Tobi, Linda Ahn, Sara A. Hayes, Afsheen Iqbal, Kenneth Ng, Maria E. Arcila, Gregory J. Riely, Mark G. Kris, Helena A. Yu

Data analysis and interpretation: Jia Luo, Alex Makhnin, Maria E. Arcila, Mark G. Kris, Helena A. Yu

Manuscript writing: All authors

Final approval of manuscript: All authors

Accountable for all aspects of the work: All authors