Egfr Intron 1 Ca Dinucleotide Repeat As Predictor Of Toxicity And Efficacy In Lung Cancer Patients Treated With Erlotinib

14566 Background: The length of the CA dinucleotide repeat in epidermal growth factor receptor (EGFR) intron 1 has been proposed to play an important role in the biological effects of tyrosin kinase inhibitors. Thus, we compared this polymorphism with the grade of skin toxicity in non-small cell lung cancer (NSCLC) patients treated with erlotinib. An exploratory analysis of clinical benefit rate and progression-free survival was already planned. Methods: Metastatic or locally advanced unresectable NSCLC patients treated orally with 150 mg/d of erlotinib in our institution were eligible. Rash was assessed according to a modified scale based on the Common Toxicity Criteria v.3.0; clinical benefit rate was defined as the sum of complete response, partial response or stable disease; progression-free survival was the time measured from the treatment start to progression. Only cases without EGFR pathogenic mutations were included in the efficacy analysis. Genomic DNA was isolated from peripheral blood; the length of the EGFR intron 1 CA dinucleotide repeat was evaluated by PCR and analyzed on a capillary sequencer. EGFR mutations were determined in neoplastic tissue. Intron 1 CA alleles were considered “short” if the sum of CA repeats of both alleles was ≤33 (group A), and “long” otherwise (Group B). Results: 23 patients were evaluated (7 retrospectively and 16 prospectively). The most common EGFR genotypes were: 16 and 20 CA repeats (with 46 and 28% frequency, respectively). Rash severity was higher in group A (p=ll seven cases with 0 grade skin toxicity belonged to Group B. 2 cases presented EGFR mutations; thus, 21 were finally included in the efficacy analysis. Clinical benefit rate was 62% (CI 95% 24% to91%) in group A and 33% (CI 95% 7%-70%) in group B. Median progression-free survival was 3.5 months (CI 95% 0–7.2) and 1.3 months (CI 95% 0–2.9) in groups A and B respectively. Conclusions: the length of the CA repeat in EGFR intron 1 accurately predicts skin toxicity of erlotinib in NSCLC. In absence of pathogenic mutations, this polymorphism seems to identify a subgroup of patients that could obtain benefit of this treatment. Larger studies are needed to address this hypothesis. No significant financial relationships to disclose.