Prognostic value of lemur tyrosine kinase-3 (LMTK3) polymorphism in Japanese (J) patients (PTS) with localized gastric adenocarcinoma (GAC)

Background LMTK3 is an estrogen receptor a (ERa) regulator. Recent studies show that [rs808419(r8) and rs9989661(r9)] and LMTK3 expression are prognostic in breast and colon cancers. Our group demonstrated that r9AA is associated with shorter time to recurrence in Caucasian(C) and Hispanic(H) females(F) with GAC. We investigated the significance of LMTK3 polymorphism in J PTS with GAC. Methods Blood or tissue samples of 169 J PTS who had surgery with/without adjuvant chemotherapy (ACT) were analyzed. Genomic DNA was extracted using the QIAmp kit; all samples were analyzed using PCR-based direct DNA-sequencing. The endpoints of the study were disease-free survival (DFS) and overall survival (OS). Kaplan-Meier curves and log-rank test were used for univariate analysis. Multivariate analysis was performed to test the interaction between polymorphism and gender adjusting for other variables. Results 60 F and 109 males were enrolled in this study, 17% stage(s) IB, 31% s II, 36% s III, 17% s IV (AJCC-6). The median age was 67(31-88). 65% of PTS received S-1 based ACT. Median follow-up was 4 years(ys). Prognosis was worse in men with r9 AA than AG/GG, at 1 year 67% (95% CI 40-83%) with AA vs 99% (95% CI 91-99%) of AG/GG were alive (p= 0.039). Median survival was not reached in the AG/GG group; in the AA group median DFS and OS was 1yr (p= 0.03) and 2ys (p= 0.039) respectively. In the multivariate analysis adjusting for s, age, and ACT, males carrying AA had increased risk of disease recurrence (HR 3.84 95%CI 1.86-7.92, p< 0.001) and dying (HR 3.47 95%CI 1.58-7.62 p=0.002) compared to those with AG/GG (HR=1, reference). Conclusions r9 AA was associated with significantly worse DFS and OS in J male with GAC. These results confirm our previous findings that LMTK3 is an independent prognostic factor for localized GAC; interestingly the relationship between gender and prognostic significance is the opposite in J vs. C/H. The gender disparity can be due to the differences in the etiology (histological subtypes), management strategies, allele frequency, and degree of estrogen exposure in the two populations. Additional studies are warranted to identify the underlying biological mechanism.