1069 Can serum human epididymis protein 4 (HE4) support the decision to refer a patient with a pelvic mass to an oncology center

Introduction/Background* Human epididymis protein (HE)4 has been shown to have a higher specificity than Cancer Antigen (CA) 125 in detecting ovarian cancer (OC). However, almost all studies are performed in oncology centers. As the prevalence of OC influences performance of predictive biomarkers, it is still unclear whether HE4 can be used to support referral decisions in general hospitals. We assessed the accuracy of HE4 with or without Risk of Malignancy Index (RMI) in differentiating benign from malignancy in patients with a pelvic mass from general hospitals. Methodology In this prospective, observational cohort study, we included patients with a pelvic mass between 2017 and 2021 from nine general hospitals. HE4 and CA125 were measured using electrochemiluminescence in preoperative samples. All patients underwent surgery. Accuracies of HE4, RMI, CA125, and combinations hereof, were determined using Area under the Curve (AUC), sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV). Result(s)* We included 311 patients, of whom 82 patients had a malignant pelvic mass. Sixty-six patients had epithelial OC, 6 non-epithelial OC and 10 had ovarian metastases or a pelvic mass of non-ovarian origin. The remaining patients had a benign (n=190) or a borderline pelvic mass (n=39). HE4 had the highest AUC (figure 1). The addition of HE4 in patients with an elevated RMI score at a pre-specified 85% sensitivity had the highest specificity, PPV and NPV (table 1) in differentiating malignant from benign and borderline pelvic masses. Conclusion* HE4 is superior to CA125 and RMI in predicting malignancy in a population with a low prevalence of OC. The addition of HE4 in patients with an elevated RMI score improved the performance of HE4 alone in discriminating malignant from benign including borderline pelvic masses. Although there is still room for improvement, this confirms that HE4 can be used to support referral decisions in a population from general hospitals.