Data Support ctDNA-Based Adaptive Strategies for EGFR+ Advanced NSCLC

“…The APPLE trial supports the longitudinal ctDNA assessment as a relevant tool for making treatment decisions, not only for adapting treatment strategies but also for the assessments of escalating or de-escalating treatment in patients with NSCLC,” the study authors stated.

The serial monitoring of circulating tumor DNA (ctDNA) T790M mutation status demonstrated feasibility when using first-generation EGFR inhibitors to treat patients with advanced EGFR-mutated advanced non–small cell lung cancer (NSCLC), supporting earlier switches to treatment with osimertinib (Tagrisso), according to findings from arms B and C of the phase 2 APPLE trial (NCT02856893).

Among patients initially receiving gefitinib (Iressa) until emergence of ctDNA EGFR T790Mm mutation in arm B or until progressive disease in arm C, respectively, the overall response rate (ORR) was 53.2% (95% CI, 38.1%-67.9%) and 56.8% (95% CI, 41.0%-71.7%). Additionally, the median progression-free survival (PFS) was 10.7 months (95% CI, 8.4-17.5) and 9.8 months (95% CI, 9.1-12.9) in each respective arm.

In arm B, 17% of patients switched to treatment with osimertinib due to a T790M mutation observed during ctDNA analysis, and 51% switched due to disease progression without T790M mutation detected in ctDNA. In arm C, 77% of patients switched to osimertinib. The ORR after switching to osimertinib was 65.6% (95% CI, 46.8%-81.4%) in arm B and 58.8% (95% CI, 40.7%-75.4%) in arm C (P = .5692).

In arm B, the PFS rate with osimertinib at 18 months (PFSR-OSI-18) was 67.2% (84% CI, 56.4%-75.9%), with rates of 87.5% (84% CI, 38.7%-98.1%) for those who switched to osimertinib based on molecular progression and 62.5% (84% CI, 40.5%-78.4%) for those who switched based on progressive disease; the median follow-up was 30 months. The median PFS among those who switched based on molecular progression and those who switched based on progressive disease, respectively, were 22.1 months (95% CI, 21.9-not reached [NR]) and 21.9 months (95% CI, 16.6-NR). Additionally, the PFSR-OSI-18 was 53.5% (84% CI, 42.3%-63.5%) for those in arm C.

In an exploratory comparison between arms B and C, the median PFS was 22.0 months (95% CI, 18.6-NR) vs 20.2 months (95% CI, 14.6-35.0) in each respective arm (Hazard ratio [HR], 0.80; 90% CI, 0.51-1.27; P = .22). A subgroup analysis indicated that there were no significant differences in outcomes between arms B and C, including all EGFR mutation subgroups and those with central nervous system metastases at study entry.

“…The APPLE trial supports the longitudinal ctDNA assessment as a relevant tool for making treatment decisions, not only for adapting treatment strategies but also for the assessments of escalating or de-escalating treatment in patients with NSCLC,” the study authors stated. “In this last scenario, the baseline ctDNA status may help to detect patients with EGFR-mutant NSCLC with worse prognosis.”

In the multi-center, non-comparative phase 2 EORTC-1613-APPLE trial, investigators treated patients with EGFR-mutated advanced NSCLC across 3 arms. Patients received 80 mg of up-front osimertinib daily until disease progression based on RECIST v1.1 criteria in exploratory arm A, 250 mg of gefitinib daily until detection of T790M-positive ctDNA followed by osimertinib in arm B, or gefitinib until disease progression followed by osimertinib in arm C.

The primary end point was PFSR-OSI-18 in arm B. Secondary end points included PFS, ORR, overall survival (OS), and brain PFS (BPFS).

Patients with measurable disease per RECIST v1.1 criteria, an ECOG performance status of 0 or 1, and an archival tissue sample were eligible for enrollment on the trial.

Between November 2017 and February 2020, investigators enrolled 52 patients into arm B and 51 patients into arm C. The median age was 69 years (range, 31-83) in arm B and 61 years (range, 33-89) in arm C. In arms B and C, respectively, most patients were female (75% vs 65%), never smoked (71% vs 59%), and had tumors harboring EGFR exon 19 deletion mutations (64% vs 65%).

In arms B and C, respectively, the median OS was NR and 42.8 months (95% CI, 27.0-NR), and the 18-month OS probability was 87.0% (84% CI, 78.1%-92.5%) and 77.3% (95% CI, 66.9%-84.8%). The median BPFS was 24.4 months (95% CI, 17.9%-28.6%) and 21.4 months (95% CI, 14.5%-42.8%) in each arm (HR, 1.05; 90% CI, 0.65-1.69; P = .57).

Adverse effects (AEs) occurred in 98% of patients in arm B, including treatment-related AEs (TRAEs) in 88%. Grade 3 or higher TRAEs occurred in 19% of patients, with diarrhea being the most common. Additionally, grade 3 or higher AEs and TRAEs during osimertinib treatment occurred in 91% and 9% of patients, respectively.

Overall, 10% of patients in both arm B and C discontinued treatment with gefitinib due to toxicity. No patients discontinued treatment with osimertinib due to AEs. Additionally, 10% of patients in arm B had osimertinib dose reductions compared with none in arm C.

Reference

Remon J, Besse B, Aix SP, et al. Osimertinib treatment based on plasma T790M monitoring in patients with EGFR-mutant non-small-cell lung cancer (NSCLC): EORTC Lung Cancer Group 1613 APPLE phase II randomized clinical trial. Ann Oncol. 2023;34(5):468-476. doi:10.1016/j.annonc.2023.02.012