8045 Background: The prevalence of the tumorigenic KIF5B:RET fusion gene in NSCLC tumors has been estimated at 0.2–6% (Jiu et al 2012; Lipson et al 2012). We retrospectively analyzed tumor samples from 4 Phase III NSCLC trials of vandetanib, a TKI that selectively targets RET, VEGFR and EGFR signaling, to determine the prevalence of RET fusions and other RET biomarkers, and any potential association with outcome to vandetanib (V). Methods: The studies evaluated were ZODIAC (NCT00312377; docetaxel ± V 100mg), ZEAL (NCT00418886; pemetrexed ± V 100mg), ZEPHYR (NCT00404924; V 300mg vs placebo) and ZEST (NCT00364351; V 300mg vs erlotinib). RET biomarkers evaluated included RET fusions (including KIF5B:RET) and RET gene copy number (assessed by a 4-probe FISH assay), as well as RET protein expression (by IHC). Results: Of 4089 patients randomized across the 4 studies, 1291 and 1234 had tumor samples available for FISH and IHC analysis, respectively, with evaluable data obtained for 944 and 1102. RET fusions (in >10% of tumor cells) were detected in 7 of 944 samples (vandetanib, n=3; comparator, n=4), at a prevalence of 0.7% (95% CI, 0.3–1.5%). None of the 3 vandetanib-treated RET fusion-positive patients had an objective RECIST response, although there was radiologic evidence of tumor shrinkage in 2. Overall, 2.8% (n=26) of samples had RET amplification (innumerable RET clusters, or ≥7 copies in >10% tumor cells), 8.1% (n=76) had lower RET gene copy number gain (4–6 copies in ≥40% tumor cells) and 8.3% (n=92) were RET expression positive (signal intensity ++ or +++ in >10% of tumor cells). There was no difference in ORR between vandetanib and comparator for the RET amplification-positive subset (both 8.3% [1/12]), the RET copy number gain subset (9.8% [4/41] vs 9.1% [3/33], respectively) or the RET protein expression-positive subset (15.2% [7/46] and 13.6% [6/44], respectively). Conclusions: The prevalence of RET fusions was estimated at 0.7%. There were too few vandetanib-treated patients with RET fusions to make any firm conclusion regarding association with efficacy. Evidence from the other RET biomarkers tested suggested that these do not infer a differential advantage in patients treated with vandetanib. Clinical trial information: NCT00312377; NCT00418886; NCT00404924.