Pulse afatinib for ERBB2 exon 20 insertion mutated lung adenocarcinomas

Abstract

Introduction

Genomic aberrations involving erb-b2 receptor tyrosine kinase 2 (ERBB2) are driver oncogenes in ∼2% of lung adenocarcinomas. However, the use of daily dosing of ERBB2 tyrosine kinase inhibitors (TKIs) - including afatinib - has been fraught by plasma concentrations that barely achieve preclinical model inhibition, significant patient-reported toxicities and limited clinical activity. We hypothesized that alternative dosing strategies could improve tolerability and efficacy.

Methods

We profiled lung cancer cell lines against TKIs and retrospectively evaluated the toxicity/response to pulse afatinib (280mg once weekly) in lung cancers with ERBB2 mutations.

Results

An ERBB2 exon 20 insertion mutated lung cancer cell line had 50% inhibitory concentration to afatinib higher than the reported plasma concentration of afatinib 40mg daily. Three patients with advanced ERBB2 mutated lung adenocarcinomas were treated with off-label pulse afatinib. The 280mg weekly dose was well tolerated with no reported rash and minimal diarrhea. One TKI-naïve patient achieved a partial response for 5 months and another stable disease for 11 months.

Conclusions

Pulse afatinib at a weekly dosing scheme induced anti-tumor activity in ERBB2 exon 20 insertion mutated lung adenocarcinomas. Future clinical trials of alternative dosing schemes of ERBB TKIs as monotherapy or in combination with other therapies are warranted for ERBB2 mutated tumors.

Introduction

Somatic mutations of the erb-b2 receptor tyrosine kinase 2 (ERBB2) gene - alternatively known as HER2 - were identified in 1.7% of all lung adenocarcinomas [1] analyzed by The Cancer Genome Atlas (TCGA). Most ERBB2 mutations congregate as inframe insertions within exon 20 of this ErbB family member [2], akin to epidermal growth factor receptor (EGFR) exon 20 insertion mutations [3]. Preclinical models have consistently demonstrated that ERBB2 exon 20 insertion mutants are transforming in lung models and define oncogene addiction [2]. Dual EGFR/ERBB2 TKIs have been analyzed in these same preclinical models. This line of research disclosed that ERBB2 mutated lung adenocarcinomas can be inhibited by irreversible second generation ERBB TKIs, such as afatinib, dacomitinib and neratinb; however, at concentrations that were close to 100-fold higher than those necessary to inhibit EGFR-exon 19 deletion or L858R mutated models [4-6].

The dose-limiting toxicities of afatinib, dacomitinib and neratinib are related to inhibition of wild-type EGFR in the skin and gastro-intestinal tract; therefore limiting achievable plasma concentrations in patients to nanomolar concentrations that may not fully inhibit ERBB2 exon 20 mutant proteins [7, 8]. In as much, clinical trials and case series of afatinib, dacomitinib and neratinib monotherapy used at their daily dosing schemes have been disappointing to date. The overall response rates (ORRs) to daily dacomitinib, neratinib and afatinib have been reported at below 15-30%, with short periods of disease control and frequent rash/diarrhea requiring dose reductions or discontinuation [9, 10].

We hypothesized that an alternative treatment strategy for irreversible ERRB TKIs, such as intermittent pulsatile doses, could improve tolerability (by decreasing skin toxicities) and efficacy (by achieving intermittent plasma concentrations that would exceed the threshold for anti-proliferative inhibition of ERBB2 exon 20 insertion mutated lung adenocarcinomas). Here, we report preclinical models that support our hypothesis and retrospectively compile the response/toxicity of three patients with advanced lung adenocarcinoma treated with pulse afatinib at a dose of 280mg once weekly.

Methods

Cell culture, cell proliferation assays and reagents

NCI-H1781 (H1781) cells, harboring ERBB2-G776>VC, were purchased from ATCC (Manassas, VA). Cells were maintained in RPMI 1640 medium (Mediatech, Manassas, VA) supplemented with 10% fetal bovine serum and grown at 37°C in a humidified atmosphere with 5% CO₂. These cells were overlaid in 96-well plates, allowed to attach overnight and then treated with or without two TKIs (erlotinib and afatinib) for 72 hours. Cell viability was determined by CellTiter 96 Aqueous One solution proliferation kit (Promega, Madison, WI) according to the manufacture's protocol. Experiments were performed in triplicate. Inhibitory proliferation curves and the 50% inhibitory concentration (IC₅₀) were generated using the GraphPad Prism 6 software (GraphPad Software, La Jolla, CA). We used IC₅₀ values obtained in identical conditions and previously published [11] for NCI-H3255 (H3255, EGFR-L858R) and NCI-H1975 (H1975, EGFR-L858R+T790M) to contrast H1781 to an EGFR/ERBB2 TKI-sensitive and TKI-resistant cell line, respectively. Erlotinib and afatinib were purchased from LC Laboratories (Woburn, MA). All reagents were dissolved in dimethyl sulfoxide and stored at −80°C.

Tumor and data collection

Patient-tumor pairs followed at Beth Israel Deaconess Medical Center (BIDMC) with a diagnosis of lung adenocarcinoma were identified through an ongoing Institutional Review Board-approved study [12]. Pathologic data, tumor genotype, toxicity to afatinib and radiographic parameters were gathered from retrospective chart extraction. Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 was used to fit target and non-target lesions. Data was collected and managed using REDCap electronic data capture hosted at BIDMC.

Afatinib dosing choice and treatment

Afatinib - an irreversible EGFR/ERBB2 TKI - has a label dose of 40 mg daily for advanced EGFR mutated NSCLCs (equal to 280 mg dose in one week). For the pulse dose of afatinib in ERBB2 mutated NSCLCs, the clinicians/patients at our center used an off-label dosing scheme of the full weekly dose once weekly (e.g., 280 mg once weekly). The rationale was based on prior experience with pulsatile dosing schemes of the EGFR TKI erlotinib, for which doses exceeding the weekly total dose can be given once weekly.

Results

Preclinical models

We selected three representative lung adenocarcinoma cell lines to contrast the anti-proliferative effects of the reversible EGFR TKI erlotinib and the irreversible EGFR/ERBB2 TKI afatinib. H3255 (EGFR-L858R) represents an erlotinib/afatinib hypersensitive cell, H1975 (EGFR-L858R+T790M) characterizes an erlotinib/afatinib resistant cell, and H1781 (ERBB2-G776>VC) embodies an ERBB2 exon 20 mutated lung cell line.

Figure 1 depicts IC₅₀ values of dose-dependent proliferation experiments for the aforementioned TKIs. We also plotted, based on patient-level pharmacokinetic data [7, 8], the reported median trough plasma concentration of erlotinib 150 mg/day (∼2000nM, Figure 1A) or afatinib 40mg/day (∼60nM, Figure 1B). H3255 showed an IC₅₀ value that is many folds lower than the plasma concentrations of erlotinib or afatinib. H1975 showed an IC₅₀ value that exceeds the plasma concentrations of erlotinib or afatinib (Figure 1A and B, respectively).

Figure 1.

Response to ERBB TKIs in EGFR and ERBB2 mutated cell lines. A. 50% inhibitory concentration (IC₅₀) of erlotinib using nanomolar (nM) concentrations. B. IC₅₀ of afatinib using nM concentrations. Cell lines used and genotype are indicated in the figure. The expected median trough plasma concentration of erlotinib 150mg/day (∼2000nM) and afatinib 40mg/day (∼60nM) are plotted (see main text for references).

For the ERBB2 mutated cell line H1781, the IC₅₀ value for erlotinib at 6377nM is consistent with lack of significant anti-ERBB2 activity of this reversible EGFR TKI (Figure 1A). The IC₅₀ value for afatinib at 91nM was lower than that for H1975 but still higher than the plasma concentration of afatinib 40mg daily (Figure 1B). The lower IC₅₀ values of all cell lines to afatinib compared to erlotinib reflect the potency of this irreversible EGFR/ERBB2 TKI against mutant and wild-type proteins.

These preclinical models supported our hypothesis that the most promising inhibitory concentrations of afatinib for ERBB2 mutated lung cancers exceed the achievable clinical concentrations of this TKI. We sought to therefore identify patients that were treated at our multidisciplinary thoracic oncology clinic with alternative doses of afatinib.

Patients with ERBB2 mutated lung adenocarcinomas treated with pulse afatinib

We identified, by retrospective review, 8 patients with ERBB2 mutated lung adenocarcinomas (6 had advanced/recurrent disease). Five received at least one ERBB2 TKI, with 2 receiving dacomitinib as part of a phase II trial (NCT00818441, both without a response). Three patients with advanced lung adenocarcinoma and ERBB2 exon 20 mutated lung adenocarcinomas were treated with pulse afatinib (Table 1). An off-label dose of 280 mg once weekly was chosen by providers plus patients to potentially minimize skin tolerability and maximize efficacy against ERBB2 mutants. These patients were never or light former smokers and most had received prior anti-cancer therapy for brain metastases or systemic tumor burden (Table 1). Only one patient had received a prior ERBB2 TKI: neratinib plus temsirolimus as part of a clinical trial (NCT01827267). The other two cases were TKI-naïve. The ERRB2 genotypes and other genomic events identified using comprehensive genomic profiling with targeted next generation sequencing [13] can be found on Table 1.

Table 1.

Tumor genotype, clinical characteristics, prior therapies and toxicity to pulse afatinib.

Lung adenocarcinoma tumor genotype	Clinical characteristics	Brain metastases/ radiotherapy	Prior cytotoxic chemotherapy	Prior ERBB2 TKI	Toxicity pulse afatinib (CTCAEv4)
ERBB2-A775_G776insYVMA	65 year-old/White/Male/Never smoker/ECOG PS1	Yes/3000cGy whole brain	Yes/carboplatin+pe metrexed (SD×5months)	Yes/neratinib+te msirolimus (SD×3months*)	Grade 0 rash Grade 0 diarrhea No SAE
ERBB2-V747_G748insGSP TP53-Y220C	64 year-old/White/Female/5-pack year smoking/ECOG PS1	Yes/2000cGy whole brain to cervical spine	Yes/carboplatin+pe metrexed (PR×3months)	No	Grade 0 rash Grade 1 diarrhea No SAE
ERBB2-E740_A741insAYVM TP53-R342*	71 year-old/Black/Female/Nev er smoker/ECOG PS1	Yes/Surgery plus 2000cGy whole brain	No	No	Grade 0 rash Grade 1 diarrhea No SAE

ECOG, Eastern Cooperative Oncology Group; TKI, tyrosine kinase inhibitor; PS, performance status; SD, stable disease; PR, partial response; CTCAE, Common Terminology Criteria for Adverse Events; pulse afatinib (280mg once weekly); SAE, serious adverse event;

^*best response with increase in target lesions using RECIST version 1.1.

Grade 1 diarrhea (increase of <4 stools per day over baseline) using common terminology criteria for adverse events (CTCAE, version 4) occurred in two cases, mostly within the first 24-48 hours after the pulse afatinib dose. Diarrhea responded to loperamide for symptomatic management. No other CTCAE significant events were identified as linked to afatinib by reviewing clinical notes and laboratory studies performed during routine clinical care (Table 1).

Radiographic responses to afatinib in lung adenocarcinomas harboring ERBB2 mutations

The ORR using RECIST v1.1, by retrospective imaging review by the investigators, for the three cases was 33.3% (1/3) with some degree of decrease in target lesions in all cases (Figure 2A). The patient with overt progressive disease within one month of pulse afatinib therapy harbored an ERBB2-A775_YG776insYVMA lung adenocarcinoma, had received prior neratinib and progressed in the central nervous system (Table 1, Figure 2A and 2B) with death occurring less than a month after discontinuation of afatinib.

Figure 2.

Response to pulse afatinib. A. Waterfall plot of the maximum response to afatinib 280mg once weekly. Progressive disease (PD), stable disease (SD) and partial response (PR) are indicated in different colors. B. Swimmer plot of the duration of response, duration of use of afatinib and sites of PD in each patient. Post-PD afatinib is shown, when applicable. C. Computed tomography (CT) example of prolonged period of disease control in the patient whose lung adenocarcinoma harbored an ERBB2-V747_G748insGSP mutation. Shown are pre-afatinib and month 8 of pulse afatinib representative CT scan cuts. The red arrows highlight a pleural-based lesion.

The patient with an ERBB2-E740_A741insAYVM lung adenocarcinoma displayed 31% regression in thoracic target lesions and stability of bone plus brain metastases, compatible with partial response by RECIST (Figure 2A). The response lasted 5 months prior to brain metastases progression (Figure 2B); when the patient elected to transfer her type of care to best supportive measures only.

The patient with an ERBB2-V747_G748insGSP lung adenocarcinoma displayed 13% regression in thoracic target lesions (Figure 2A and 2C), lack of new systemic lesions and control of brain metastases for 11 months prior to development of new liver metastases (Figure 2B). The patient's clinical condition had significantly improved while on afatinib without major toxicities (Table 1) and she elected to continue on post-progression pulse afatinib 280mg weekly that is ongoing at time of this report without further symptomatic progression (Figure 2B).

Discussion

The use of TKI monotherapy with the EGFR/ERRB2 TKIs afatinib, dacomitinib and neratinib for ERBB2 exon 20 insertion mutated lung adenocarcinomas has led to subpar results in the clinic. As an example, a phase II trial of dacomitinib at its maximum tolerated doses (MTDs) of 30-45mg daily only resulted in an ORR of 12% (3/26) for patients with advanced ERBB2 exon 20 insertion mutated lung tumors [9]. Most patients displayed skin/gastrointestinal toxicities and the median overall survival was below 9 months from start of dacomitinib [9]. In one of the largest cohorts ever published of ERBB2 mutated lung adenocarcinomas, the ORR of 29 patients that were treated with daily MTDs of neratinib or lapatinib or afatinib was only 7.4%, with a median progression-free survival of 3.4 months and median survival of 6.5 months [10]. The ORRs to standard cytotoxic chemotherapies and or combinations of chemotherapy with anti-ERBB2 antibodies (such as trastuzumab) seem to exceed the reported outcomes with ERBB2 TKIs alone [10]. Grounded on preclinical studies of combinations of ERBB TKIs and mTOR inhibitors [6], a completed clinical trial (NCT01827267) will report on the efficacy of neratinib plus temsirolimus but the preliminary activity seems to show ORR of only 21% (3/14), median PFS of 4 months and diarrhea in all patients with ERBB2 mutated lung adenocarcinoma [14]. Due to the lack of approved precision therapies for this important cohort of lung cancers, the National Comprehensive Cancer Network guidelines suggest that daily therapy with off-label afatinib is a possible therapeutic strategy for chemotherapy-refractory ERRB2 mutated lung adenocarcinomas [15]. However, it is clear that alternatives to ERRB TKI daily monotherapy are urgently needed.

In this report, we explored the preclinical rationale for the need of alternatives to daily irreversible ERBB2 TKI (such as afatinib) monotherapy in ERBB2 mutated lung cancers and report – to the best of our knowledge – the first three cases treated with pulse afatinib (280mg once weekly) in the advanced disease setting. We confirmed that an ERBB2 exon 20 insertion mutated cell line is only maximally inhibited by afatinib concentrations that are beyond the reported median plasma concentrations for the most usual dose of afatinib (40mg/day) in the clinic. This observation sparked a hypothesis that an intermittent pulsatile dose of afatinib could potentially achieve doses that exceed necessary inhibitory concentrations in the clinic and – at the same time – avoid skin toxicities due to lack of daily cutaneous EGFR inhibition. Although our case series is limited by the non-prospective enrollment of patients plus lack of detailed pharmacokinetic measures, the retrospective medical record review presented here pleasantly confirmed that afatinib 280mg once weekly was not associated with rash; in clear contrast to near universal rash/mucositis reported with afatinib 40mg daily [3]. The lack of significant cutaneous toxicities has been reported also with pulse erlotinib at doses up to 1500mg once weekly in EGFR mutated lung adenocarcinomas [16]. The ORR of 33.3% and the prolonged disease control (11 months) of one of the patients to pulse afatinib compare favorably with the aforementioned lackluster activity of ERBB2 TKI monotherapy [9, 10].

Our small case series should be thought as a springboard for future clinical development of pulsatile dosing schemes of ERBB TKIs. Indeed, a clinical trial to determine the MTD of intermittent high-dose afatinib (3 days of pulse doses every 14 days of a 28-day cycle) is near completion (NCT01647711). Additional prospective studies evaluating the systemic and intracranial pharmacokinetics of pulsatile doses of EGFR/ERBB2 TKIs are warranted to understand the peak and trough concentrations achievable in patients with these schemes. Future precision oncology developments of alternative dosing schemes of ERBB TKIs as monotherapy or in combination with other therapies are warranted for ERBB2 mutated lung adenocarcinomas and other malignancies.