Symptom triggered screening for ovarian cancer: a pilot study of feasibility and acceptability

Abstract

Purpose: Our goal was to determine if symptom-based ovarian cancer screening was feasible in a primary care clinic and acceptable to women and practitioners. In addition, we wanted to describe the outcomes for a pilot group of screened women.

Methods: A prospective study of 2262 women over age 40 with at least one ovary participated in symptom-based screening using a symptom index (SI). The first 1001 were in a non-intervention study arm and 1261 were screened for symptoms and referred on to testing with CA125 and transvaginal ultrasound (TVS) if the SI result was positive. Patients and practitioners were surveyed about acceptability of study procedures. All patients were linked to the Western Washington SEER Cancer Registry to determine if ovarian cancer was diagnosed in any women.

Results: Of the eligible women visiting the clinic, 72.5% were interested in participating, and the participation rate was 62.1%. Of the 1261 who participated in the screening arm 51 (4%) were SI positive and 47 participated in CA125 (45/47 normal) and TVS (32/47 normal). Two endometrial biopsies and one hysteroscopy D&C were performed secondary to study enrollment (pathology negative). No laparotomies or laparoscopies were performed secondary to study involvement. A survey of patient acceptability, on a scale of 1-5, revealed a mean score of 4.8 for the acceptability of SI screening and 4.7 for TVS and CA125 testing among SI positive women. Providers also rated the SI procedures highly acceptable with a mean score of 4.8. Two participating patients were diagnosed with ovarian cancer; one had a true positive SI in the non-intervention arm and one had a false negative SI in the screened arm.

Conclusion: While our pilot study is not large enough to assess sensitivity or specificity of a symptom-based screening approach, we did find that this type of screening was feasible and acceptable at the time of a primary care visit and referred approximately 4% of women for additional diagnostic testing. Symptom-based screening also resulted in minimal additional procedures.