For women with a high inherited risk of ovarian cancer, risk-reducing surgery remains the only proven preventive option, albeit at a significant personal cost. Professor Adam Rosenthal explores the rationale, evidence and future potential of Risk of Ovarian Cancer Algorithm-based surveillance as a means of enabling earlier diagnosis, greater choice and more personalised care for patients not yet ready to undergo surgery.

Women at high inherited risk of ovarian cancer – such as those with germline BRCA1 and BRCA2 pathogenic variants – are advised to undergo risk-reducing salpingectomy and oophorectomy before natural menopause, as this is the only way to reliably prevent this poor-prognosis cancer.

However, surgery carries a cost and entails loss of fertility, premature menopause and the need for hormone replacement therapy (HRT) to mitigate these effects, with no guarantee that patients will feel the same way after surgery, even if they take HRT.

Women in this situation face a stark choice: surgery and its side effects, or ‘do nothing’ and hope they are not diagnosed with advanced disease. As with many cancers, earlier diagnosis improves outcomes, creating a need for a surveillance option that can detect ovarian cancer earlier in those not yet ready to undergo risk-reducing surgery.

ROCA testing for ovarian cancer

The Risk of Ovarian Cancer Algorithm (ROCA) test involves collecting a blood sample every four months to measure serum cancer antigen 125 (CA125) using a specialised assay and a high-quality analyser to minimise intra-assay variation. The algorithm takes into consideration the woman’s age, high-risk status and menopausal status.

Once more than two longitudinal samples have been taken, the algorithm can assess the natural variation around a woman’s individual CA125 baseline and determine whether the next sample’s result falls within that range. It then calculates the risk of the women currently having ovarian cancer, based on all the above factors.

The risk score is expressed as a ratio. Ratios greater than 1:1000 are considered normal and any result worse than that triggers intervention. This means a repeat ROCA test at six weeks (risk 1:500–1:1000), a repeat test at six weeks with a transvaginal ultrasound scan (1:499–1:34) or direct referral to a gynaecology rapid access clinic with a transvaginal ultrasound scan (<1:34).

Trial results in the UK1,2 and US3 using the ROCA test in high-risk populations showed remarkably consistent results. Incident cancers in BRCA-carriers were downstaged, with around half diagnosed at stage 1 or 2, rather than nearly all at stage 3.

This earlier diagnosis resulted in clinical benefits, including less extensive surgery, which reduces morbidity and length of inpatient stay, reduced need for neo-adjuvant chemotherapy to shrink otherwise inoperable tumours, and a very high rate (>90%) of complete cytoreduction at primary surgery – that is, no visible tumour remaining at the end of the operation.

Although the data were non-randomised, so a survival advantage could not be claimed, the above outcomes are strong prognostic indicators. It is therefore reasonable to expect better outcomes when cancers are detected earlier through surveillance.

Ovarian cancer surveillance can offer reassurance

The choice to undergo surveillance is highly personal. Understandably, many women prioritise reducing their cancer risk over preserving fertility or avoiding early menopause, opting for risk-reducing surgery.

Others prefer to delay surgery to have children or to maintain their natural hormonal function. For these women, surveillance offers reassurance that if ovarian cancer is diagnosed before they have risk-reducing surgery, it can be detected earlier, potentially leading to better outcomes.

However, it is essential that surveillance is not promoted as an alternative to risk-reducing surgery. Only surgery can prevent ovarian cancer. Surveillance is nevertheless clearly better than doing nothing and should therefore be an option for those women who wish to have children or avoid surgical menopause. This view was endorsed by recent NICE guidance.

It is crucial that services offering surveillance are adequately resourced to enable an annual review of all patients who have not yet undergone surgery. This allows them to re-evaluate their decision regularly and consider whether they are at a point in their lives where they are willing to proceed with surgery, which is the safest option.

From a health services perspective, earlier diagnosis of ovarian cancer should reduce surgical morbidity, potentially improve survival and reduce costs, as advanced ovarian cancer is expensive to treat. In fact, a cost-effectiveness analysis conducted by a team from the London School of Economics found that surveillance will likely be cost-saving in the NHS setting.

The role of multidisciplinary working

Multidisciplinary collaboration is essential. Clinical geneticists play a key role in counselling and confirming a woman’s risk status, which includes them assessing the pathogenicity of her genetic variant or verifying that a genetic test performed outside of an NHS laboratory is reliable.

Gynaecologists with relevant expertise are experienced in counselling women about their options, the effects of risk-reducing surgery – particularly if undertaken prior to menopause – and the limitations of surveillance. Menopause experts may be needed to optimise post-operative HRT regimens.

Clinical psychologists can help high-risk women navigate emotionally difficult choices. For women with a prior breast cancer, breast oncologists are needed to advise on the appropriateness of risk-reducing surgery and, in certain circumstances, whether HRT may be an option.

Finally, expert gynaecological pathologists meticulously examine the removed fallopian tubes and ovaries to exclude the presence of microscopic cancers. Failure to detect these can seriously compromise outcomes in the minority of women found to have one.

How will emerging research help further pave the way for ovarian cancer progress?

Trials worldwide, including in Europe,4–8 are exploring risk-reducing salpingectomy with delayed oophorectomy. As ovarian cancers in BRCA carriers overwhelmingly originate in the fallopian tubes, it seems logical to assume that this two-stage surgical approach might reduce ovarian cancer risk while preserving hormonal function until natural menopause.

The degree of protection it provides has yet to be determined, and the current view is that completion oophorectomy closer to the age of natural menopause is essential.

It will be at least five years before definitive safety data are available. Until then, this approach should be considered experimental and not offered outside of a clinical trial.

However, if the data confirm the anticipated safety levels, it could be a game-changing option for women who currently delay risk-reducing bilateral salpingo-oophorectomy because of understandable concerns about premature menopause.

This approach would not be suitable for women delaying the standard surgery in order to conceive spontaneously, although IVF remains possible after salpingectomy.

For the time being, it makes sense to offer surveillance to women who have had bilateral salpingectomy but have not yet completed oophorectomy. High-grade serous cancer developing on the ovary would still be expected to trigger an elevated ROCA score even if the fallopian tubes had been removed. Once oophorectomy is complete, surveillance can cease.

Where next for risk prediction, screening and surveillance?

The introduction of the ROCA test in the NHS is, to my knowledge, the first instance of cancer surveillance being offered to a high-risk population outside established national screening programmes.

Existing breast and bowel screening programmes already include high-risk groups, such as BRCA carriers and patients with Lynch syndrome, respectively, who require more frequent testing from a younger age.

As evidence accumulates that ovarian cancer surveillance leads to earlier diagnoses and improved outcomes, I am hopeful that surveillance programmes will be introduced for many high-risk populations.

Increasing numbers of people are now being identified as having an inherited high risk of cancer. Several factors contribute to this trend: traditional family-history-based testing, which can miss about half of those at high risk; the mainstreaming of germline testing among newly diagnosed cancer patients; cascade testing of relatives of carriers; and decreasing costs of genetic testing.

Lower prices make broader population screening more cost-effective and enable access to more affordable private testing.

Risk-prediction models are also improving to the point where they are beginning to enter routine use in the clinic. These provide more reliable risk estimates to aid decision-making, for example, by showing individuals what their risk is over the next one, five or 10 years. This is crucial information for deciding when to undergo risk-reducing surgery.

A number of promising technologies are beginning to enter clinical trials, including urine markers for diagnosing renal tract cancers in patients with Lynch syndrome, DNA-methylation-based tests on vaginal samples for detecting breast and gynaecological cancers, and MRI screening for people at high risk of prostate cancer.

I feel strongly that funders need to invest more in this type of research. High-risk populations are ideal cohorts for evaluating whether a novel early-detection technology can improve cancer outcomes.

Trials in these populations are also logistically easier to undertake and less costly than trials in the general population, as higher cancer incidence means that fewer participants and shorter follow-up periods are required to demonstrate clinical benefits.

Before introducing a new national general-population screening programme, randomised controlled trial evidence demonstrating a mortality reduction is generally needed to justify the programme’s very high cost and to offset the harms screening inevitably causes to some healthy individuals.

In contrast, in many high-risk populations, the high incidence of poor-prognosis cancers means that evidence of earlier diagnosis and consequent clinical benefits may be sufficient to justify surveillance.

I believe that although mortality reduction is crucial for general population screening programmes, high-risk groups may be willing to accept smaller clinical benefits, such as decreased morbidity or extended survival, as these are clearly better than doing nothing and waiting for symptoms. By then, many cancers are more advanced, harder to treat and linked to higher morbidity.

Finally, successful early detection in high-risk populations raises the possibility that a novel screening technology may also work in the general population and therefore has the potential to help everyone.

Author

Adam Rosenthal FRCOG PhD
Consultant gynaecologist, University College London Hospitals NHS Foundation Trust, and professor of gynaecological cancer prevention at University College London, UK

References

  1. Rosenthal A et al. Evidence of Stage Shift in Women Diagnosed With Ovarian Cancer Screening During Phase II of the United Kingdom Familial Ovarian Cancer Screening Study. J Clin Oncol 2017;35(13):1411–20.
  2. Philpott S et al. The avoiding late diagnosis of ovarian cancer (ALDO) project; a pilot national surveillance programme for women with pathogenic germline variants in BRCA1 and BRCA2. J Med Genet 2023;60(5):440–9.
  3. Skates S et al. Early Detection of Ovarian Cancer using the Risk of Ovarian Cancer Algorithm with Frequent CA125 Testing in Women at Increased Familial Risk – Combined Results from Two Screening Trials. Clin Cancer Res 2017;23(14):3628–37.
  4. Steenbeek M et al. Association of Salpingectomy With Delayed Oophorectomy Versus Salpingo-oophorectomy With Quality of Life in BRCA1/2 Pathogenic Variant Carriers. A Nonrandomised Controlled Trial. JAMA Oncol 2021;7(8):1203–12.
  5. Van Bommel M et al. Salpingectomy With Delayed Oophorectomy Versus Salpingo-Oophorectomy in BRCA1/2 Carriers: Three-Year Outcomes of a Prospective Trial. BJOG 2025;132(6):782–94.
  6. Steenbeek M et al. TUBectomy with delayed oophorectomy as an alternative to risk-reducing salpingo-oophorectomy in high-risk women to assess the safety of prevention: the TUBA-WISP II study protocol. Int J Gynecol Cancer 2023;33(6):982–7.
  7. Gaba F et al. Preventing Ovarian Cancer through early Excision of Tubes and late Ovarian Removal (PROTECTOR): protocol for a prospective non-randomised multi-center trial. Int J Gynecol Cancer 2021;31(2):286–91.
  8. Sia J et al. Estimands for Clinical Effectiveness of Risk-Reducing Early Salpingectomy in Women With High Risk of Ovarian Cancer. JAMA Netw Open 2025;8(9):e2532195.