By Robert L. Coleman, MD
Professor, University of Texas; M.D. Anderson Cancer, Center, Houston
Dr. Coleman reports no financial relationships relevant to this field of study.
Synopsis: Borderline ovarian tumors present in early stage, infrequently relapse, are generally cured with surgical resection, and may be conservatively treated in women desiring continued fertility. However, nearly one-third of recurrences are malignant and pursue an aggressive course like their invasive primary counterparts.
Source: du Bois A, et al. Borderline tumours of the ovary: A cohort study of the Arbeitsgmeinschaft Gynakologische Onkologie (AGO) Study Group. Eur J Cancer 2013;49:1905-1914.
Borderline ovarian tumors are unique neoplasms that demonstrate proliferation without infiltrative destructive growth or stromal invasion. Clinically, they carry a much more favorable prognosis compared to invasive epithelial ovarian cancer. However, their rarity has limited clear information regarding prognostic factors and no prospective studies exist that evaluate therapeutic strategies. To provide clarity into these important clinical parameters, the AGO study group carried out a retrospective cohort study, collecting data from consecutive patients with ovarian borderline tumors treated between 1998 and 2008 in 24 German centers. All tumors underwent central pathological review, and case patients were prospectively followed for outcomes of recurrence, therapy, and death. Of 1042 patients undergoing independent pathological review, borderline ovarian tumor was confirmed in 950 patients; of the 92 patients excluded, 43% had primary invasive disease and the remaining benign disease. The histology of the borderline tumors was serous (67%) and mucinous (31%). Most were diagnosed in stage I (82%); 8% and 10% were stage II and III, respectively. Overall, 74 patients (8%) experienced relapse and 43 (5%) died within the observation period. Multivariate analysis revealed higher stage, incomplete staging, tumor residuals, and organ preservation as independent prognostic factors for disease recurrence. While the presence of implants was prognostic, neither microinvasion nor micropapillary growth pattern impacted progression-free survival (PFS). Of 74 relapsed patients, 30% had malignant transformation to invasive ovarian cancer with 5-year PFS and overall survival of 12% and 50%, respectively. Prognosis of ovarian borderline tumors correlates with tumor-related as well as surgery-related factors. The balance between recurrence risk and organ preservation and fertility-sparing surgery is an important issue deserving further research.
It has been well described that, in general, ovarian borderline tumors are uncommon and of good prognosis.1 However, the natural history of the disease is not well understood, particularly in regard to recurrence, due to the rarity of the condition and of these events. This paper represents the largest collection of cases with central path review and provides better clarity to prognostic factors previously intimated from retrospective work. The most surprising outcomes in this review were the lower than expected recurrence risk (8% vs historically 11-15%) and invasive recurrence risk (2% vs historically 4%).2 Also of note were the recurrence risks in patients undergoing conservative surgery (organ preserved) and incomplete staging. Fortunately, neither of these factors increasing the risk for recurrence (by up to 5-fold) led to poorer survival. This was largely due to "in organ" recurrence for those undergoing fertility-sparing surgery and the frequency of non-invasive recurrent disease (70% of recurrences). However, somewhat surprising was their identification of high-grade invasive disease (34%) in the 22 patients who recurred with invasive disease.3 The outcomes of these patients was no different than expected from invasive ovarian cancer in general. Finally, a good take-home message from the study was the utility of expert pathology review. In this trial, all cases were independently re-reviewed. In nearly one in eight cases, a different opinion was reached with a 60:40 split on benign and malignant diagnoses. Since this was a retrospective re-review of the original histology, intervention was carried out locally based on the original histology (read locally as borderline ovarian cancer). However, when the outcomes of these discrepant groups were analyzed, the outcomes were vastly different than expected; median PFS in the borderline re-read as invasive was 6 years, and median PFS in the group re-read as benign was not reached (greater than 12.5 years). Overall for the confirmed borderline tumors, the median PFS was 12.5 years. This differential strengthens the validity of the revised histological assessment.