NEW YORK – New research indicates that self-reported endometriosis is associated with a significantly increased risk of clear-cell and endometrioid invasive ovarian cancers, said Stacey A. Missmer, Sc.D.
In addition, for the first time, low-grade serous invasive ovarian cancer also was linked to endometriosis. No association was found for mucinous or high-grade serous invasive ovarian cancer or borderline tumors of either subtype in this analysis of pooled data from 13 ovarian cancer case-control studies, according to Dr. Missmer, who described the findings at the annual congress of the Endometriosis Foundation of America, but was not involved in the study.
The findings, published in Lancet Oncology (2012;13:385-94), show that "clinicians need to be aware of the increased risk of specific ovarian cancer subtypes in women with endometriosis," wrote lead author Celeste Leigh Pearce, Ph.D., and her colleagues.
"The hope is that we will develop a risk stratification model that combines genetic and epidemiological risk to better stratify women into high-risk, intermediate-risk, and low-risk categories, allowing better individualization of prevention and early detection approaches such as risk-reduction surgery and screening," wrote Dr. Pearce of the gynecology department at the University of Southern California, Los Angeles, and her coauthors.
The 13 case-control studies, which were part of the Ovarian Cancer Association Consortium, included 13,226 controls and 7,911 women with invasive ovarian cancer. Of those, 818 controls (6.2%) and 738 women with ovarian cancer (9.3%) reported a history of endometriosis. There also were 1,907 women with borderline ovarian cancer, and 168 of these women (8.8%) reported a history of endometriosis.
"Clinicians need to be aware of the increased risk of specific ovarian cancer subtypes in women with endometriosis."
A threefold increased risk of clear-cell invasive ovarian cancer in patients with self-reported endometriosis (P less than .0001) was found. A twofold increased risk was noted for low-grade serous (P less than .0001) and endometrioid invasive ovarian cancers (P less than .0001). "On the basis of evidence, including the results of molecular studies, endometriosis should be thought of as a precursor lesion for clear-cell and endometrioid ovarian cancers, whereas the type of association with low-grade serous ovarian cancers requires further follow-up," wrote the authors.
No association was found between endometriosis and the risk of invasive mucinous or high-grade serous ovarian cancer or borderline tumors of either subtype.
For serous ovarian cancer, the association with endometriosis depends on the grade. The risk for the invasive low-grade serous ovarian cancer was significantly stronger than that for its high-grade counterpart (odds ratio, 1.94; 95% confidence interval, 1.21-3.11; P = .01). The authors suggested that this is an additional piece of evidence suggesting that the pathogenesis of low-grade and high-grade serous ovarian cancers might differ.
Commenting on the study, Dr. Missmer said that the absolute risk of ovarian cancer in women with endometriosis was small. For most patients in the studies, the endometriosis was self-reported and not pathologically confirmed, noted Dr. Missmer, who is the director of epidemiologic research in reproductive medicine at Brigham and Women’s Hospital, Boston. She said that the findings indicated three possible causal associations: endometriosis causes cancer, long-term treatment of endometriosis influences cancer risk, or both are independent pathologies with common risk factors.
In an accompanying comment published in Lancet Oncology (2012;13:326-8), Charlie Gourley, Ph.D., of the University of Edinburgh Cancer Research UK Centre, said that the main strengths of the study are its statistical power and robust methods, which give it the power to allow definitive ovarian cancer subgroup analysis. He remarked that the main truly novel finding was the association between the history of endometriosis and low-grade serous ovarian cancer, but he was somewhat surprised by the lack of association between serous borderline tumors and endometriosis.
"By identification of an association between particular histological subtypes and not to others, weight is added to the increasing evidence that these histological subtypes are distinct disease entities," wrote Dr. Gourley. He also suggested that while the results do not mandate targeted ovarian cancer screening of patients with endometriosis, such screening should be considered.
Dr. Missmer said she had no relevant financial disclosures. Dr. Pearce and Dr. Gourley likewise reported no conflicts of interest.