Molecular Epidemiology of EGFR and KRAS Mutations in Lung Adenocarcinomas Based on Clinical Testing of 2787 Consecutive Cases
S Dogan, E Brzostowski, MF Zakowski, M Ladanyi. Memorial Sloan-Kettering Cancer Center, New York, NY
Background: Almost 50% of lung adenocarcinomas harbor EGFR or KRAS mutations but the etiology of most of these mutations is unclear, except for the subset of KRAS mutations known to be smoking-associated (G12C, G12V). Associations between types of somatic mutations and demographic and epidemiologic data can provide clues to distinct etiology or biology. Here, we took advantage of our clinical testing data on over 2700 lung adenocarcinomas studied since 2004 to examine correlations between types of EGFR or KRAS mutations and demographic data in unprecedented detail.
Design: Of 2787 cases tested clinically for the two major EGFR mutations (exon 19 deletions and the L858R mutation in exon 21) using sensitive mutation-specific PCR assays, 522 (19%) were mutant, including 320 exon 19 deletions (61%) and 202 L858R mutations (39%). Routine testing for KRAS mutations (codons 12 and 13) by PCR-sequencing (after microdissection if needed) started in 2006 and in 2200 cases received since, we detected 553 (25%) mutations, including G12C (37%), G12V (22%), G12D (18%), G12A (10%), and other G12 and G13 mutations (13%). We examined correlations with demographic data using the Fisher exact test or unpaired t-test.
Results: Comparison of EGFR and KRAS mutant cases showed differences in sex ratio (p=0.001; favoring males for latter) but not age. Among EGFR mutant cases, men showed a higher ratio of exon 19 to L858R than women (2.4 vs 1.4; p=0.007). Patients with EGFR L858R tended to be older than exon 19 mutants (average age 68 vs 64; p=0.0001), reflected by an exon 19 to L858R ratio of 3.5 under age 50 (p=0.008) and of 1.1 in patients 70 and over (p=0.002). Among KRAS mutant cases, compared to other KRAS mutations types, the sex ratio for G12C was skewed more towards women (p=0.006) and G12D, the KRAS mutation most often seen in never smokers, was relatively more common under age 40 (p=0.02). Notably, of 15 patients under age 40 with mutations, 7 had EGFR exon 19 deletion, 4 KRAS G12D, 2 KRAS G12V, 1 EGFR L858R, and 1 KRAS G12S.
Conclusions: The distinctive age distribution and sex ratio associations of certain EGFR and KRAS mutations may signal differences in etiology and/or biologic potential. For instance, the lower age of patients with EGFR exon 19 deletion could reflect slightly faster progression when untreated compared to EGFR L858R, as suggested by some studies. Annotation of this unique dataset for stage, smoking history, and ethnic origin is ongoing and should provide further insights.
Tuesday, March 23, 2010 1:00 PM
Platform Session: Section F, Tuesday Afternoon