Ovarian Cancer Detection Rate Doubles With New Algorithm

Liam Davenport

May 14, 2015

Detection of ovarian cancer was significantly improved in a study that used a risk algorithm based on changes in serum cancer antigen 125 (CA125) levels over time rather than relying on traditional fixed CA125 cutoff values.

As part of the ongoing United Kingdom ovarian cancer trial involving more than 200,000 women, researchers found that the risk algorithm could detect 87% of ovarian cancers, compared with fewer than 50% of cancers identified using cutoff values.

The findings could substantially improve the early detection of ovarian cancer, which is associated with 6-month survival rates of less than 5% if it is diagnosed at an advanced stage, say the researchers.

The study was published online May 11 in the Journal of Clinical Oncology.

"There is currently no national screening programme for ovarian cancer, as research to date has been unable to provide enough evidence that any one method would improve early detection of tumours," lead author Usha Menon, MD, Women's Cancer, Institute for Women’s Health, University College London, in the United Kingdom, commented in a release.

"These results are therefore very encouraging. They show that use of an early detection strategy based on an individual's CA125 profile significantly improved cancer detection compared to what we've seen in previous screening trials."

Dr James Brenton, an ovarian cancer expert at the charity Cancer Research UK, which supported the study, nevertheless cautioned: "A blood test to find women at risk of ovarian cancer is an exciting prospect, but this work still needs to be tested in women to see if it can save lives."

Looking at Changes Over Time

For the multimodal screening arm of the United Kingdom Collaborative Trial of Ovarian Cancer Screening (UKCTOCS), 46,237 women aged 50 years or older underwent incidence screening, in which use of annual serum CA125 levels was compared with use of an algorithm for assessing risk for ovarian cancer.

The algorithm assigned women to one of three categories: normal risk, which involved the patient's returning for annual screening; intermediate risk, which involved repeat CA125 measurement; and elevated risk, which involved repeat CA125 measurement plus assessment with transvaginal ultrasound. Women with persistently elevated risk were referred for clinical assessment.

During a total of 296,911 women-years of annual incidence screening, 2822 women underwent clinical assessment, and 640 underwent surgery, which was laparoscopic in 415 cases.

Of the women who underwent surgery, 133 had primary invasive epithelial ovarian or tubal cancers (iEOCs), 17 had borderline iEOCs, and four had nonepithelial ovarian cancers. Twenty-two cases were detected within 1 year of screening.

The algorithm detected 87.1% of the iEOCs, compared with 41.3%, 48.4%, and 66.5% of cases that would have been detected if a fixed CA125 cutoff of >35 U/ml, >30 U/ml, and >22 U/ml, respectively, had been used.

The area under the curve for the algorithm was significantly higher than for the single-threshold rule, at 0.915 vs 0.869 (P = .0027).

The sensitivity and specificity of the algorithm for detecting iEOCs was 85.8% and 99.8%, respectively, with 4.8 surgeries required per each iEOC identified.

Explaining why changes in CA125 levels were more effective in detecting ovarian cancer than a single measurement assessed against a threshold, Dr Menon told Medscape Medical News that, early in the course of disease, levels may rise but remain below the threshold.

With a cutoff of 30 U/ml, which is currently used for postmenopausal women in the United Kingdom, Dr Menon explained that a woman's CA125 level may rise from 8 U/ml to 15 U/ml.

She said: "So if I just looked at today's values, 15 U/ml is below 30 U/ml, so it's normal and you can come back in a year, [whereas] in our algorithm, it would trigger repeat testing."

She added: "What we have understood is that, for an individual, the value is very stable with time, especially in the older woman."

"If I'm usually running 8 U/ml, I will run 7 U/ml or 8 U/ml year on year. So for me, 15 is kind of high, which is why looking at the change over time gave us the ability to pick up the disease when it was below the normal cutoff."

Dr Menon clarified: "It's not that the people who were picked up below 30 U/ml would never been picked up, but they would have been picked a year later, or when they presented as an incident cancer."

Will Improved Detection Results in Survival Benefit?

The next stage is to see whether the improved detection of ovarian cancer translates into fewer women dying of the disease.

The full UKCTOCS trial is due to report by the end of the year. The report will include a comparison of the number of deaths from ovarian cancer in the screening arm vs the control arm, in which women had no screening.

"We have been working very hard to get to this point, but now we are at a point where, depending on what the answer is, it would be possible for the [UK National Health Service] to make a decision with regard to ovarian cancer screening," Dr Menon said.

She added: "If, after picking up these women earlier, there is no difference in the number of women who die from ovarian cancer in the screening arm vs the number of women in the control arm, then it would not justify an ovarian cancer screening program because, if we pick them up, we could not make a difference; we did not pick them up early enough."

One recurring issue associated with screening programs is with regard to treatment. Is Dr Menon concerned that, as a result of a potential nationwide ovarian screening program, a large number of women could end up having unnecessary procedures or treatment?

She replied: "In ovarian cancer, we don't at this point know what overdiagnosis and overtreatment might mean."

"In prostate, as well as in breast and lung cancer, there is this concept of cancers which might never have presented during the lifetime of the person, so we overdiagnose and overtreat in screening."

She added: "Because we are really at the beginning of ovarian cancer screening, that is not very clear at this point, because we do not have a precursor lesion that we pick up in ovarian cancer, like we pick up ductal carcinoma in situ in breast cancer."

In conclusion, Dr Menon said that the study is "great, encouraging news."

However, she cautioned that although the screening strategy appears to be more effective than previous approaches, "we do need this final answer, which will be available later this year, as to whether there is a difference in deaths between the two arms."

The study was supported by the Medical Research Council, Cancer Research United Kingdom, and the Department of Health, with additional support from the Eve Appeal and by researchers at the National Institute for Health Research University College London Hospital (UCLH) Biomedical Research Centre. The United Kingdom Collaborative Trial of Ovarian Cancer Screening Biobank was funded in part by the Special Trustees of Bart's and the London and Special Trustees of UCLH. All relevant financial relationships of the authors are listed in the original article.

J Clin Oncol. Published online May 11, 2015. Abstract

Comments

3090D553-9492-4563-8681-AD288FA52ACE
Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.

processing....