Steven J. Skates, PhD, discusses a study on early detection of ovarian cancer using the Risk of Ovarian Cancer Algorithm (ROCA), in which CA125 was tested in women with high risk who chose to postpone surgery.
In ovarian cancer, the standard of care for women who are high-risk remains risk-reducing salpingo-oophorectomy (RRSO). There are women who want to postpone this surgery, however, and experts are searching for the optimal way to monitor these patients.
In a study on early detection of ovarian cancer using the Risk of Ovarian Cancer Algorithm (ROCA), CA125 was frequently tested in women with high risk who chose to postpone surgery. The study included 3692 women with either a strong familial history of breast and/or ovarian cancer, orBRCA1/2mutations. Patients were screened every 3 months for significant increases in their CA125 baseline.
Co-lead author Steven J. Skates, PhD, co-developer of the ROCA, says that this approach improves on current screening techniques; however, he adds that it is important to remember the success of preventative surgery.
“Surgery essentially reduces the risk down to normal levels, and screening will not make it that far. But, for those high-risk women who want to delay surgery despite standard medical advice to the contrary and want to have surveillance in the meantime, I think we've made a step toward providing a more effective way to do that and there is a clear path ahead of us to improve even more.”
In an interview withTargeted Oncology, Skates, of the Massachusetts General Hospital Cancer Center and the Biostatistics Unit, says that the best option for women who choose to postpone surgery is frequent CA125 testing with the ROCA.
TARGETED ONCOLOGY:Could you provide some background information on this study?
We started this back in 2000, and at that time, we thought that early detection of ovarian cancer by repeating screening might be a better way to find the cancer in early stages than what was being done at the time. We assembled 2 cohorts, 1 was through the National Cancer Institute’s (NCI) Cancer Genetics Network, and that was the main component of that cohortalthough there were additional NCI groups that contributed to that—but in total, about 2500 women. Around the same time, the Gynecologic Oncology Group (GOG) was planning a study in women who were choosing between prophylactic oophorectomy and surveillance. The surveillance arm used this longitudinal CA125 approach that I developed in the late 1990s. GOG began this study in 2003 and enrolled about 1500 women—so in total we had about 4000 women.
We were using this longitudinal CA125 ROCA to determine whether a significant rise in CA125 above each woman's personal baseline may indicate they have ovarian cancer. ROCA was based on prior screening studies where the CA125 blood samples were taken on a regular basis and interpreted with a cut point. In retrospective analysis of those studies, it was pretty clear that those that had ovarian cancer had a rising CA125 before diagnosis. From this retrospective analysis we developed the longitudinal algorithm and implemented it in prospective screening studies. In the prospective studies, when the algorithm identified a rise resulting in an intermediate risk, we flagged them for referral to ultrasound. If it was a sufficiently large rise to result in an elevated risk, we referred them to ultrasound and a review by a gynecologic oncologist.
Then, following completion of two retrospective studies on women who had undergone RRSO which found that it reduced the risk of ovarian cancer by 95%, RRSO became the standard of care for women at high risk. We continued to follow and screen the women enrolled in the 2 cohorts who did not want to undergo immediate surgery. In reporting results for women who did not want to undergo immediate surgery, despite the fact that RRSO is the standard medical advice, instead of looking at CA125 at 1 point in time and evaluating that, we established a baseline for each woman, and a variation about that baseline. From there we looked for significant rises above that baseline when compared to the variation. Once the 2 cohort studies were completed, we reported the results in Clinical Cancer Research.
TARGETED ONCOLOGY:What do you want community oncologists to take away from this report?
The first thing I want to emphasize, is that this isn't, in any way, shape, or form, a replacement for RRSO. That is still the standard of care and still gives the patient far better protection. However, if, for whatever reason, the patient decides not to undergo immediate preventative surgery, then this is a better option for finding ovarian cancer early, or, reducing the risk of diagnosis for advanced ovarian cancer.
In the high-risk population of women who carry theBRCA1gene, in studies that were done in the late 1990s, early 2000s, about 10% of those women with invasive ovarian cancer were detected in early-stage. We've increased that to about 50%, so this baseline approach is better than just using a single cut point. It doesn't detect every patient in early stage, but it identifies more than what other screening studies obtained using a single CA125 cut point.
Our protocol did ask the women to come in every 3 months for a blood test, so 1 contribution of these studies may be the importance of frequency of screening. So, frequent screening plus the algorithm is a better approach for women who don't want immediate surgery. We do not want to influence their decision, but if they do make the decision of surveillance, then there is a better option than the single cut point interpretation of CA125.
But, I do not think any of these early detection approaches will reach the effectiveness of surgery.
TARGETED ONCOLOGY:Are there any next steps regarding this population of patients?
There are 2 populations of interestthere are the women who are high-risk, and there are the women who are at general risk. For the latter, I am a part of a study being done in the UK in postmenopausal normal risk women. But, the focus of the 2 studies in question was on high-risk women. We believe there could be additional biomarkers in the blood that might increase sensitivity for detection in early stage. We are currently undertaking an analysis of biospecimens to try to identify candidate proteins that might be serum biomarkers that could add to the value of longitudinal CA125. That certainly is an opportunity to increase sensitivity for early-stage disease.
Another possibility is, what is now commonly known as liquid biopsies, which is seeking to detect tumor DNA in the blood. This might be another source of signal that might complement the protein markers that we have been investigating. There is an NCI Early Detection Research Network of researchers that are interested in early detection for many different cancers, and we are one of the ovarian cancer biomarker development labs. We are looking both in the liquid biopsy spacecirculating tumor DNA—and the protein biomarker space, to increase sensitivity for early-stage ovarian cancer.
Skates SJ, Greene MH, Buys SS, 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 [published online January 31 2017].Clin Cancer Res.doi:10.1158/1078-0432.CCR-15-2750.