Does Taking ‘The Pill’ Reduce Ovarian Cancer Risk?

The use of oral contraceptives and a greater number of live births both reduced the risk of ovarian cancer, according to a new analysis of a population registry.

“Findings from our large-scale, hypothesis-free, machine-learning study suggest that the risk for ovarian cancer is likely to be, at least in part, modifiable and also that it may be possible to develop predictive blood tests that can identify the cancer in its early stages of development,” reported Iqbal Madakkatel, PhD, a research associate at the University of South Australia in Adelaide, Australia, and his colleagues in The International Journal of Gynecological Cancer.

The findings relied on machine learning to analyze 2920 exposures among 221,732 women in the UK Biobank, with 1441 cases of ovarian cancer occurring during a median 12.6 years after baseline. The final model prediction used the 87 predictors that made up the top 3% of contributing factors. These included baseline and personal characteristics, female-specific factors, sociodemographics, lifestyle and environment, cognitive and psychosocial factors, physical measurements, health and medical history, and biomarkers.

Does Taking ‘The Pill’ Prevent or Reduce Ovarian Cancer Risk?

The researchers found that having ever used oral contraception reduced odds of ovarian cancer by 26% (odds ratio [OR], 0.74; 95% CI, 0.66-0.84). Further, those who stopped using oral contraception at ages ≥ 45 years had nearly half the risk for ovarian cancer than those who stopped taking it before age 25 years (OR, 0.57; 95% CI, 0.41-0.80).

These findings match the existing evidence base, Elena Ratner, MD, MBA, a gynecologic oncologist at Yale Cancer Center, New Haven, Connecticut, told Medscape Medical News. “Any woman who has been on a birth control pill for 5 years of cumulative use has a decreased risk for ovarian cancer by about 50%,” Ratner said.

The data supporting a protective effect of oral contraception has been consistent in multiple meta-analyses, agreed Melissa Hodeib, DO, a gynecological oncologist at UCLA Health in Los Angeles. The reason, they both explained, is the hypothesis that more ovulation increases the risk.

“There’s a theory that every time you ovulate, there is an inflammatory reaction or process occurring at the surface of the ovary, and preventing that is what’s really leading to the decrease in ovarian cancer,” Hodeib said. As the follicle ruptures the surface of the ovary each month, the constant trauma to the lining of the ovary may potentially result in carcinogenesis.

That theory may also explain why data on risk associated with IUDs are inconsistent, she added. While the copper IUD does not prevent ovulation, it’s possible that hormonal IUDs involve “partial suppression of ovulation in some women, but it’s not the primary mechanism of action, which I think is why we’ve gotten contradictory data on whether it has the same preventive effect as oral contraceptives.”

What Are the Modifiable Risk Factors for Ovarian Cancer? 

One of the strongest risk factors for ovarian cancer is a family history or presence of any of the several genetic mutations found to be associated with risk, including BRCA1 and BRCA2 mutations, Ratner said. Other nonmodifiable risk factors include early menarche or late menopause — again due to opportunity for ovulation — and endometriosis, Hodeib said.

Even the known modifiable factors, however, primarily have to do with decreasing ovulation, including having more pregnancies and breastfeeding, Ratner said. In this study, participants who had at least two live births had 39% lower odds of ovarian cancer than those with no live births (OR, 0.61, 95% CI, 0.54-0.69).

A second theory of ovarian cancer relates to the cancer beginning in the fallopian tubes, and past studies have found a 50%-60% risk reduction from salpingectomy. But while this paper found a 79% reduction in risk from bilateral oophorectomy, it was surprising, and potentially concerning, that no association with tubal ligation or salpingectomy was found, Rimel said.

A limitation of any machine learning system with large amounts of data is controlling for multiple variables and recognizing that the value of the output is only as good as the input. “Any bias in the system will send you a signal that isn’t necessarily real,” Rimel said. She wasn’t sure what to make of the study’s lack of findings related to fallopian tube removal, but she found its absence was noteworthy.

What Are the Barriers to Screening For and Detecting Ovarian Cancer?

Women diagnosed with stage I ovarian cancer have an excellent chance of survival, but only about 17% of cases are diagnosed this early, according to the Ovarian Cancer Research Alliance. The ROCA test can estimate the likelihood of a woman’s chances of developing ovarian cancer if she carries a confirmed BRCA mutation, Hodeib said, but “we still have not led to an improvement in detection.”

“The biggest issue with ovarian cancer, is that the symptoms themselves are so vague,” Hodeib said. “It doesn’t have a clear-cut trigger to say that something’s wrong.” Symptoms typically include bloating, abdominal discomfort, and a decreased appetite, which leads patients to seek care initially from their PCP. “Their PCP does a GI workup, and then they get a colonoscopy, and months and months are passing as the disease is growing. Then by the time you discover it, it’s stage III or stage IV.”

Another challenge is that ovarian cancer remains so rare, with only about 20,000 new cases a year, Bobbie J. Rimel, MD, a gynecologic oncologist at Cedars-Sinai in Los Angeles, told Medscape Medical News. “Whenever you’re talking about a rare disease, you can’t have a lot of false positive in your screen,” she said.

Finally, another major hurdle is that ovarian cancer does not behave like most other cancers.

“Breast cancer and colon cancer start at a single site and grow from there, and then once they reach a critical size, they spread from there,” Rimel said. “With ovarian cancer, partly because of the location of the ovary, where it’s suspended inside the abdomen, a small tumor can start breaking off like dandelion fluff into the abdomen and remain tiny and nonsymptomatic for a period of time that may be very long, but is not detectable.” When it spreads, it settles on surfaces in the abdomen, making it difficult or impossible to remove only the specific area affected by the tumor.

What New Evidence Does Study Reveal About Risk Factors?

Past studies have found conflicting evidence regarding the influence of height and weight on ovarian cancer risk, Rimel said, but this study found a positive association for both. Greater height increased the risk by 13% per standard deviation above average, and greater weight modestly increased risk (OR, 1.08).

Previous evidence has suggested obesity modestly increases risk, but not as much as it does with uterine cancer, Ratner said. A greater intake of alcohol — more than seven drinks a week — has also been linked to a modest increased risk, as it is with several cancers, Ratner said, but no link to alcohol intake was noted in this study.

Several measures of adiposity in this research, including weight, body fat percentage, and basal metabolic weight, were linked to greater odds of ovarian cancer, ranging from a 14% to 28% increase, depending on the characteristic. Each additional percentage point of body fat above the standard deviation, for example, increased odds of cancer by 28% (OR, 1.28).

The most likely mechanism for the height finding is genetic, Rimel suggested. It’s possible that genetic risk factors for ovarian cancer sit along the same pathways as those related to height.

The novel findings of this study were the identification of various blood biomarkers associated with ovarian cancer risk, experts told Medscape Medical News. The researchers found that higher levels of aspartate aminotransferase levels and alanine aminotransferase each reduced the odds for ovarian cancer by 11%. A higher neutrophil percentage was linked to a modest increased risk (OR, 1.08), and a higher total serum protein was linked to a lower risk (OR, 0.86).

“The finding that 20% of the features suggested by the model as important for ovarian cancer were blood biomarkers is notable, and can potentially inform on mechanisms associated with tumorigenesis, and support the development of predictive blood tests,” the authors wrote.

Could Biomarkers Be Promising Avenues for Future Screening?

A screening test for ovarian cancer has been a bit of a Holy Grail, Ratner suggested, with researchers having sought one for decades with no success. “It’s so complicated because of how this cancer presents,” she said. Even studies using ultrasounds every six months did not show an increase in early detection or survival.

Hodeib similarly noted past failures that tried combining tumor marker tests with ultrasound screening.

“So even though these blood biomarkers they mentioned are exciting because they’re easy tests, you can have other reasons for alterations in these lab values that are not specific to ovarian cancer,” she said. Further, there is the problem of determining a timeline for using these tests. “Are you going to check labs every month, every 3 months, every 6 months?”

Hodeib noted other research looking into glutamine and alanine, both of which were also explored in this study. “I think that speaks to the need to maybe focus a bit more on cancer cells’ metabolism and the tumor microenvironment” to potentially come up with a screening or early detection test, she said. “I hope we find a way to detect this early because it’s just an awful disease to treat.”

The study was funded by Australia’s Medical Research Future Fund, and one author was funded by Australia’s National Health and Medical Research Council. The authors reported no disclosures. Rimel declared serving on the advisory boards of GSK, Merck, and Immunogen. Hodeib and Ratner had no disclosures.

Tara Haelle is a science/health journalist based in Dallas.