TY - JOUR T1 - Outcomes and Cost Analysis of Surveillance Strategies After Initial Treatment for Women With Recurrent Ovarian Cancer JF - International Journal of Gynecologic Cancer JO - Int J Gynecol Cancer SP - 1333 LP - 1342 DO - 10.1097/IGC.0000000000001040 VL - 27 IS - 7 AU - Vinita M. Alexander AU - Alan N. Gordon AU - David H. Howard AU - Namita Khanna Y1 - 2017/09/01 UR - http://ijgc.bmj.com/content/27/7/1333.abstract N2 - Objective The aim of this study was to determine whether there is a survival or cost benefit dependent on detection strategy of recurrent ovarian cancer (ie, imaging, physical examination findings, report of symptoms, or rising cancer antigen 125 [CA-125] levels).Methods/Materials A retrospective chart review of 112 ovarian cancer patients was conducted, and method of detection of recurrent disease was determined from medical records. The following primary outcomes were determined using Cox proportional hazards regression model: progression-free survival (PFS) after diagnosis of recurrence and time to death after diagnosis of recurrence (overall survival [OS]). Several approaches to disease surveillance were proposed, and a cost model was applied.Results Median time to recurrence was 13.5 months. Overall, 6.3% presented with only physical examination findings; 24.1%, with elevating CA-125 levels; 34.8%, with imaging; and 32.1%, with symptoms. Most patients presenting with recurrent disease diagnosed by rising CA-125 were white (62.9%); those with imaging and symptomatic recurrences were blacks (56.4% and 57.1%, respectively). There was a small but not statistically significant OS benefit for recurrence detected via CA-125 (P = 0.85; OS per detection method: PE, 20.7 months; CA-125, 26.8 months; imaging, 17.8 months; and symptoms, 6.6 months). We modeled costs of surveillance in our patient cohort; up to 40.8% of cases of ovarian cancer recurrences would have been missed if no imaging were obtained during surveillance.Conclusions Results indicate minimal differences in PFS and statistically insignificant differences in OS, depending on detection method. Notably, black patients with Medicaid presented most often with symptomatic recurrences, which surprisingly did not affect patient OS and PFS; and interestingly, pr\ivate or self-pay insurance was associated with decreased OS among black patients. From our cost analysis, we estimate that the most cost-effective surveillance strategy for the first year costs $9.2 million annually and includes office visit biannually, biannual CA-125 levels, and annual asymptomatic imaging. ER -