Improving diagnostic strategies for ovarian cancer in Filipino women using ultrasound imaging and a multivariate index assay

1Introduction

Ovarian cancer is the 7th most common cancer and the 8th leading cause of cancer mortality among women worldwide [1] . In the Philippines, it ranks 5th among female cancers, with 5395 new cases documented in 2020. It accounted for 3379 cancer deaths in the past year, making it the 7th leading cause of all cancer mortality in the country [2] . Currently, there is no standard strategy for screening and early detection of ovarian cancer that reduces ovarian cancer mortality. In our setting, combined clinical assessment and ultrasound remain the gold standard for preoperative diagnosis of ovarian cancer and eventual referral to a gynecologic oncologist [3] . Based on the 2016–2020 annual statistics of our study center, the Philippine General Hospital, 15.4% of all gynecologic patients had an adnexal mass on ultrasound. Most of these ovarian masses were benign (74.1%), while 25.9% were malignant.

Several ultrasound-based prediction models and morphologic scoring systems developed to guide sonographers in the preoperative diagnosis of ovarian cancer are in use in the Philippines today. These include the Sassone and Lerner scoring systems, the Risk of Malignancy Index (RMI), and the various models introduced by the International Ovarian Tumour Analysis (IOTA) group, such as Logistic Regression 1 (LR1), Logistic Regression 2 (LR2), and Assessment of Different NEoplasia in the AdneXa (ADNEX) 4567 . Proper training in applying these criteria is deemed highly critical as there are a limited number of gynecologic oncologists in the country. One of the challenges in using these models is that their diagnostic performances depend on the expertise of the ultrasonographer. The diagnostic performance of these tests is higher when performed by a Level III sonographer, defined by the International Society for Ultrasound in Obstetrics and Gynecology as a practitioner who is likely to spend most of their time undertaking gynecological ultrasound and/or teaching, research, and development in the field, compared to a Level I sonographer who has performed a minimum of 300 examinations under the supervision of a Level-II practitioner or an experienced Level-I practitioner with at least 2 years’ regular practical experience [89] . Hence, the applicability of these tests in areas without trained ultrasonographers might be low.

Most recently, the local use of multivariate index assays (MIA), which includes MIA or Ova1®, and a second-generation multivariate index assay (MIA2G, Overa®) have been made available. These assays were shown to be effective for the evaluation of ovarian tumors and demonstrated higher sensitivity for early-stage ovarian malignancy compared to clinical assessment alone, CA-125, and modified ACOG guidelines with consistent performance across menopausal status 1011121314 . These MIAs are increasing in utility worldwide as adjuncts to traditional screening methods but have yet to be maximized on a national level due to economic constraints and the lack of local experience [10111315] .

Furthermore, various international guidelines have already added screening criteria involving biochemical markers. However, these criteria have not been adopted as a local standard practice primarily due to their unavailability in most areas in the Philippines. Previous studies showed that adding biochemical markers, such as CA-125, MIA, and MIA2G, to ultrasound tests can influence the diagnostic performance for ovarian cancer [1516] . Moreover, the diagnostic performance of these tests is affected by how these tests are combined. When the ultrasound test and the MIA or MIA2G are combined in parallel, the sensitivity and NPV improved. On the other hand, sequential or serial tests with ultrasound imaging followed by either MIA or MIA2G improved the specificity and PPV of the test for ovarian cancer [1516] .

In this research, we endeavored to evaluate ultrasound imaging and biomarker assays’ clinical performance in preoperative assessment, and in distinguishing between benign and malignant ovarian masses in Filipino patients. This project also assessed the overall utility of these assays in our local setting when used alone or in combination using parallel or serial testing.