Neoadjuvant Osimertinib for the Treatment of Stage I-IIIA Epidermal Growth Factor Receptor-Mutated Non-Small Cell Lung Cancer: A Phase II Multicenter Study

Abstract

Purpose: To assess the safety and efficacy of the third-generation epidermal growth factor receptor (EGFR) tyrosine kinase inhibitor osimertinib as neoadjuvant therapy in patients with surgically resectable stage I-IIIA EGFR-mutated non-small cell lung cancer (NSCLC).

Patients and methods: This was a multi-institutional phase II trial of neoadjuvant osimertinib for patients with surgically resectable stage I-IIIA (American Joint Committee on Cancer [AJCC] V7) EGFR-mutated (L858R or exon 19 deletion) NSCLC (ClinicalTrials.gov identifier: NCT03433469). Patients received osimertinib 80 mg orally once daily for up to two 28-day cycles before surgical resection. The primary end point was major pathological response (MPR) rate. Secondary safety and efficacy end points were also assessed. Exploratory end points included pretreatment and post-treatment tumor mutation profiling.

Results: A total of 27 patients were enrolled and treated with neoadjuvant osimertinib for a median 56 days before surgical resection. Twenty-four (89%) patients underwent subsequent surgery; three (11%) patients were converted to definitive chemoradiotherapy. The MPR rate was 14.8% (95% CI, 4.2 to 33.7). No pathological complete responses were observed. The ORR was 52%, and the median DFS was 40.9 months. One treatment-related serious adverse event (AE) occurred (3.7%). No patients were unable to undergo surgical resection or had surgery delayed because of an AE. The most common co-occurring tumor genomic alterations were in TP53 (42%) and RBM10 (21%).

Conclusion: Treatment with neoadjuvant osimertinib in surgically resectable (stage IA-IIIA, AJCC V7) EGFR-mutated NSCLC did not meet its primary end point for MPR rate. However, neoadjuvant osimertinib did not lead to unanticipated AEs, surgical delays, nor result in a significant unresectability rate.

FIG 1.

CONSORT diagram. DFS, disease-free survival; EFS, event-free survival; ITT, intention-to-treat; MPR, major pathological response; ORR, objective response rate; OS, overall survival.

FIG 2.

Radiographic and pathologic response to treatment in patients who underwent surgical resection. (A) Patient stage (AJCC v7) before osimertinib treatment is indicated, as are the number of cycles of osimertinib received before surgery. Depth of best radiographic response and RECIST response (unconfirmed) are indicated. (B) Waterfall plot showing % pathological regression after osimertinib treatment. Tumors that achieved a MPR or pathological regression > 50% are indicated. (C) Pretreatment EGFR mutation subtypes are indicated by blue and red boxes. Detection and classification of TP53 or RBM10 mutations in either the pretreatment biopsy or resected tumor specimen by next-generation DNA sequencing is indicated. AJCC, American Joint Committee on Cancer; EGFR, epidermal growth factor receptor; INDEL, insertion or deletion; MPR, major pathological response; PR, partial response; SD, stable disease; QNS, quantity not sufficient for next-generation sequencing.

FIG 3.

Postresection (A) DFS, n = 24 and (B) OS, n = 24. Kaplan-Meier curves with 95% CIs of DFS and OS for all patients who underwent surgical resection. (A) Median DFS with 95% CI indicated. (B) Median OS NR. DFS, disease-free survival; NR, not reached; OS, overall survival.

FIG A1.

Radiographic response to treatment in ITT population. (A) Patient stage (AJCC v7) before osimertinib treatment is indicated, as are the number of cycles of osimertinib received before surgery. (B) Waterfall plot of unconfirmed best RECIST 1.1 radiographic response after osimertinib treatment is shown. PR, SD, and PD are indicated. (C) Pretreatment EGFR mutation subtype indicated by blue and red boxes. Detection and classification of TP53 or RBM10 mutations in pretreatment biopsy is indicated. ^aDenotes patients who did not undergo lung cancer surgical resection. AJCC, American Joint Committee on Cancer; ITT, intention-to-treat; PD, progressive disease; PR, partial response; QNS, quantity not sufficient for next-generation sequencing; SD, stable disease.

FIG A2.

Postresection DFS and OS for patients who underwent surgical rection. (A) Kaplan-Meier curves comparing DFS for patients whose tumors showed an MPR (red) compared with those whose tumors did not show an MPR (blue). Median DFS indicated. (B) Kaplan-Meier curves comparing OS for patients whose tumors showed an MPR (red) compared with those whose tumors did not show an MPR (blue). (C) Kaplan-Meier comparison of DFS for patients whose tumors showed a pathological regression (Path R) of >50% (red) compared with those whose tumors did not show pathological regression >50% (blue). Median DFS with 95% CI are indicated. (D) Kaplan-Meier comparison of OS for patients whose tumors showed a pathological regression (Path R) of >50% (red) compared with those whose tumors did not show pathological regression >50% (blue). Median OS with 95% CI are indicated. DFS, disease-free survival; HR, hazard ratio; MPR, major pathological response; NR, not reached; OS, overall survival.

FIG A3.

EFS and OS for ITT population (n = 27). (A and B) Kaplan-Meier curves with 95% CIs of EFS and OS for all ITT patients. (A) Median EFS with 95% CI indicated. (B) Median OS NR. (C) Kaplan-Meier curves comparing EFS for ITT patients whose tumors showed an MPR (red) compared with those whose tumors did not show an MPR (blue). Median EFS indicated. (D) Kaplan-Meier curves comparing OS for ITT patients whose tumors showed an MPR (red) compared with those whose tumors did not show an MPR (blue). (E) Kaplan-Meier comparison of EFS for ITT patients whose tumors showed a pathological regression (Path R) of >50% (red) compared with those whose tumors did not show pathological regression >50% (blue). Median EFS with 95% CI are indicated. (F) Kaplan-Meier comparison of OS for ITT patients whose tumors showed a pathological regression (Path R) of >50% (red) compared with those whose tumors did not show pathological regression >50% (blue). Median OS with 95% CI are indicated. EFS, event-free survival; HR, hazard ratio; ITT, intention-to-treat; MPR, major pathological response; NR, not reached; OS, overall survival.

FIG A4.

Oncoprint showing co-occurring tumor genomic alterations identified by clinical NGS performed on (A) pretreatment biopsies or (B) post-treatment resected tumors. Two pretreatment tumors (arrows) were analyzed by Foundation One CDx, all other samples were analyzed by UCSF500. Alterations identified through these assays and included in the clinical sequencing reports are indicated. (C) UCSF500 analysis of paired pretreatment and post-osimertinib tumors. ^aDenotes alterations that were detectable in pretreatment but not post-treatment samples. ^bDenotes alterations that were only detectable in post-treatment samples. NGS, next-generation DNA sequencing; QNS, quantity not sufficient for next-generation sequencing.