Abstract Background: In EGF30008 the addition of lapatinib (La) to first-line treatment with letrozole (Le) significantly improved progression-free survival (PFS) in patients (pts) with hormone receptor positive (HR+), HER2 positive (HER2+) metastatic breast cancer (hazard ratio (HR) = 0.71; p=0.019), but not in HR+, HER2 negative (HER2−) pts. This study assessed the ability of VeriStrat (VS), a mass-spectrometry blood-based pretreatment assay correlating with clinical outcome from EGFR-TKI therapy, to stratify pt outcome in EGF30008. The VS assay assigns VS Good (Good), VS Poor (Poor), or Indeterminate labels to a serum sample based on a specific 8-peak signature in the mass-spectra (Taguchi F et al., JNCI 2007). Methods: Blinded to clinical data, pretreatment serum was analyzed using standard VS procedurei. Statistical analyses were performed using Mantel-Haenszel and Cox proportional hazards methods; correlations were evaluated using Fisher's exact and χ2 tests. Results: Of the 1286 pts randomized (HER2+=219; HER2−=952; HER2 unknown=115), 1163 pts had serum available; a VS label was assigned to 1046 pts (961 Good; 80 Poor; 5 Indeterminate); 117 were not evaluable (hemolyzed). In the overall population there was no significant difference in PFS between Good and Poor groups within the Le+La arm (p=0.53). In contrast, PFS of the Good group was longer than that of the Poor group within the Le only arm (HR=0.36, p<0.001) and comparable with that of Good pts within the Le+La arm (median PFS 10.8 mo vs 11.0 mo). An interaction test showed significantly different HRs for PFS in Good vs Poor between treatment arms (p=0.002). In the HER2+ population both Good (n= 169) and Poor (n= 13) groups received significant PFS benefit from addition of La to Le: in Good HR=0.71, p=0.046, medians 3.0 mo (Le) vs 8.0 mo (Le+La); in Poor HR=0.17, p=0.02, medians 2.3 mo (Le) vs 8.6 mo (Le+La). In the HER2− population median PFS for the Poor group (n=58) increased from 3.1 to 11.0 mo (HR=0.57, p=0.068) with addition of La to Le, whereas the Good group (n=702) received no significant benefit, median 13.6 mo on Le and 13.8 mo on Le+La (PFS HR= 0.85, p=0.09). In multivariate analysis VS remained independently significant and predictive of differential treatment effect in the overall population. No significant correlation of VS classification with HER2 status (p=0.35) or prior adjuvant hormonal therapy (p=0.33) was found. VeriStrat was also not significantly correlated with baseline HER2 ECD levels (low <15 ng/ml vs. high ≥15 ng/ml), p = 0.23, or ER expression (H-score:lower quartile:<160 vs. rest), p=0.18. Conclusions: In the overall population VeriStrat predicted PFS in Le alone but not in the Le+La combination, with a significant differential treatment effect. VeriStrat identified a subset of pts with inferior PFS on Le therapy. Adding La to Le improved outcome in HER2+ pts, and interestingly suggested an improvement in the VS Poor subset of HER2− pts. HER2− VS Good pts do not gain benefit with the addition of La. No correlation between VS and investigated predictors of benefit from the addition of La to Le in HER2− pts was observed. If these results can be validated in a prospective study, the addition of La to Le may be a potential treatment option for HER2− HR+ VS poor pts. Citation Information: Cancer Res 2011;71(24 Suppl):Abstract nr S1-4.