EP482 Associations between thrombocytosis, predictive biomarkers, and survival outcomes in uterine serous carcinoma

2019 
Introduction/Background Thrombocytosis is known to be associated with poor outcome in endometrial cancer, though data on uterine serous carcinoma (USC) is sparse. The purpose of this study was to evaluate thrombocytosis and its association with patient survival as well as other predictive biomarkers. Methodology We performed a retrospective chart review on 167 patients who presented with USC between 2005 and 2017. Data collected included: age, race, ethnicity, place of birth, overall survival (OS), tumor size, CA125 at diagnosis, CBC count at diagnosis, presence of lymphovascular space invasion (LVSI), and immunohistochemical positivity (>1%) for Estrogen Receptor (ER) and Progesterone Receptor (PR) in the tumor cells. Thrombocytosis was defined as platelets ≥450k. Data analysis was performed using Cox proportional hazard models, Spearman rank correlation, and the Kaplan-Meier method, with significance set at p Results 18% of women had thrombocytosis at disease presentation. Thrombocytosis prevalence did not vary by race, ethnicity, or place of birth. Thrombocytosis was strongly correlated with rising CA125 levels (Spearman9s rho=0.37, p=0.01), tumor size (Spearman9s rho=0.36, p=0.04), and presence of LVSI (Spearman9s rho=0.26, p=0.04). LVSI was present in 71% of patients without thrombocytosis vs. 100% of those with thrombocytosis (p=0.04). In the hazards model, stage, LVSI, thrombocytosis, and tumor size were associated with worse survival (HR=1.90, 2.28, 2.94, and 1.02, respectively; all p Conclusion Thrombocytosis is an independent predictor of poor outcome in USC. It additionally correlates with other known poor prognostic factors such as LVSI and rising CA125 levels. Further investigation of these observations is required to understand the biological underpinnings of platelets in tumor progression. Disclosure Nothing to disclose.
    • Correction
    • Source
    • Cite
    • Save
    • Machine Reading By IdeaReader
    0
    References
    0
    Citations
    NaN
    KQI
    []