Background: The use of trastuzumab has been established as the standard first-line treatment of HER2 positive (+) AGC. However, the role of anti-HER2 agents in the second-line treatment of HER2+ AGC has not been clearly established yet. TyTAN is the first randomized study to compare the efficacy and safety of adding lapatinib (L) to paclitaxel (P) vs P alone in the second-line treatment of HER2+ AGC. Methods: Eligibility required patients (pts) with AGC, amplification of HER2 by fluorescence in situ hybridization (FISH), and one prior regimen containing fluoropyrimidines and/or cisplatin. Pts were randomized 1:1 to L (1500mg QD) and P (80mg/m², Day 1, 8, 15 q4w) or P alone. The treatments were given until disease progression or unacceptable toxicity. Stratification variables were prior trastuzumab treatment and gastrectomy status. Primary endpoint was overall survival (OS). Results: From March 2008 to June 2011, 1923 pts were screened and 430 pts were HER2+ AGC. 261 out of 430 pts were enrolled. All pts were from Asian countries: Japan (100), China (95), Korea (46), and Taiwan (20). Median OS was 11.0 months for L+P and 8.9 months for P alone in the intent-to-treat (ITT) population (HR 0.84; p=0.2088). In a pre-planned subgroup analysis, median OS in HER2 immunohistochemistry (IHC) 3+ subgroup was 14.0 months for L+P and 7.6 months for P alone (HR 0.59; p=0.0176). The endpoints in efficacy and AEs of special interest for L+P are summarized below (Table). Conclusions: Although OS was prolonged in L+P arm by 2 months, the difference was not statistically significant. HER2 IHC 3+ subgroup demonstrated statistically significant prolongation of OS by adding L. Clinical trial information: NCT00486954.