e20564 Background: In metastatic EGFR-mutant non-small cell lung cancer (mNSCLC), rapid identification of targetable alterations is essential for treatment. Plasma circulating tumor DNA (ctDNA) testing may complement tissue genotyping, particularly when tissue is limited or results are delayed. However, real-world data on ctDNA testing using non-standardized sampling in EGFR-mutant disease remain limited. Methods: We conducted a prospective observational study including patients with EGFR-mutant mNSCLC who underwent opportunistic plasma ctDNA testing using a validated, hospital-based next-generation sequencing assay. Clinical characteristics, ctDNA timing, detected EGFR alterations, and concordance with available tissue molecular results were analyzed descriptively. Results: Our cohort included 29 EGFR-mutant mNSCLC patients; 59% female, 62% never-smokers; median age 67 (32–97). Most patients (86%) had de novo metastatic disease. Among these patients, 13% had bone-only disease, 87% had visceral involvement and 27% had CNS involvement. Plasma ctDNA was collected at metastatic diagnosis in 26 patients and at disease progression in 3 patients. Among those tested at metastatic diagnosis, 27% had ctDNA plasma collected before treatment initiation and 73% on or after treatment initiation. Most ctDNA samples (52%) were collected on the day of treatment initiation, with median time from treatment initiation to collection of 5 days (range 1–30). Osimertinib monotherapy was used in 96% of patients. Among patients tested for ctDNA at stage IV, the median time from tissue biopsy to plasma ctDNA collection was 28 days (range 11-135). Among patients tested at progression, median time from initial tissue biopsy to plasma ctDNA was 37 months. EGFR mutations identified in tissue included L858R (41%), exon 19 deletions (45%), exon 20 mutations (7%), and L747P-exon 19 (7%). Discordance between tissue and plasma ctDNA was observed in 11 patients (38%), with no variants detected in ctDNA despite the presence of EGFR mutations in tissue. Among discordant cases, three patients had plasma collected at recurrence to stage IV, two at disease progression, and six had de novo metastatic disease, with ctDNA collected before treatment initiation in one patient and within 0–7 days after treatment initiation in five patients. EGFR L858R and exon 19 deletions each accounted for 45.5% of discordant cases; exon 20 mutations accounted for 9%. Among discordant patients, metastatic sites included bone-only (n = 1), isolated brain metastasis (n = 1), and visceral disease (n = 9). Conclusions: Plasma ctDNA testing is useful when rapid results are needed; however, in this small real-world cohort, EGFR mutations were missed in 38% of samples. This highlights the ongoing importance of tissue genotyping and the need to better define the timing and role of plasma ctDNA testing in EGFR-mutant lung cancer.
Lopes et al. (Thu,) studied this question.