Skip to Content
chevron-left chevron-right chevron-up chevron-right chevron-left arrow-back star phone quote checkbox-checked search wrench info shield play connection mobile coin-dollar spoon-knife ticket pushpin location gift fire feed bubbles home heart calendar price-tag credit-card clock envelop facebook instagram twitter youtube pinterest yelp google reddit linkedin envelope bbb pinterest homeadvisor angies

Doctor checking patient

Jeffrey Skolnick, Georgia Institute of Technology CSO, Ovarian Cancer Institute
& Benedict Benigno CEO, Ovarian Cancer Institute

The devastating effects that ovarian cancer bestows on a woman’s life are now well known; However, an occasional recapitulation can allow us to focus more keenly on the mission of the Ovarian Cancer Institute. The following will provide you with a brief insight:

  1. There are no early warning signs
  2. Symptoms are vague and frequently indicative of a problem elsewhere
  3. Pelvic pain and bleeding are rare initial symptoms
  4. The pap smear is utterly useless
  5. The CA 125 test, highly touted as a screening test is worse than useless
  6. Almost all ovarian cancers are diagnosed in an advanced stage
  7. The cure rate in advanced stages is less than 20%
  8. The cure rate in stage I (the cancer is limited to the ovaries) is 92%!
  9. Until relatively recently, treatment has remained the same for over 40 years
  10. A highly-accurate diagnostic test for early-stage OC is oncology’s most holy grail!

For over 20 years, the researchers at the Ovarian Cancer Institute have worked relentlessly toward the discovery of a highly accurate test for early-stage disease, and we have achieved results that are astonishing, results that have transcended all expectations. 100% specificity and 98% sensitivity ain’t shabby! A paper is being written for publication in a peer-reviewed journal as I write this, and I intend to present it at the next meeting of the Society of Gynecologic Oncology. The annual meeting of this society is attended by virtually all GYN-ONCOLOGISTS in the world. There is a press box where journalists take notes and publish their views on the papers presented, which will give our test instant worldwide coverage without spending a dime on PR!

Some board members have asked us why we have not published anything, since our results are vastly superior to anything that we know of. Most publications on this subject are not very scientific and come to conclusions that, by and large, are not justified by what is written in the text. The following will provide you with an insight into these issues.

Overview:

One of the critical needs for the effective treatment of most cancers is the ability to diagnose the cancer at a very early stage as this dramatically increases survival rates. For example, the five-year survival rate of stage 1 Ovarian Cancer (OC) is over 90%; yet, this drops to less than 20% for stage 4 OC 1. What is particularly insidious about OC is that stage 1 OC is often asymptomatic 2 and being relatively rare 3, diagnostic tests with a low false positive rate are required. Thus, it represents a particularly hard application of any diagnostic methodology. For this reason, although its incidence is low, we believe that if we could develop a valid diagnostic test for OC, the same conceptual framework should work for most, if not all, cancers. Thus, we first tested our AI algorithms on OC.

OC-DX Development and Validation of a Blood Test to Diagnose Ovarian Cancer

There are an estimated 236,511 women living with Ovarian Cancer (OC) in the United States 4. In 2023, there were 19,710 new cases and 13,270 deaths from OC making it the fifth-highest cause of cancer deaths in women 3. This high mortality rate reflects the fact that >70% of women are diagnosed with late-state OC 5. Given that 1/78 women, on average, will develop OC, a diagnostic tool with a low false positive and negative rate is sorely needed 3. In the past year, we developed a metabolic fingerprint diagnostic tool, OC-Dx, that likely achieves this goal. Unlike alternative approaches that focus on the changes in a limited number of metabolites or human proteins, this test works by learning the differences in the metabolic profiles found in the blood sera of normal women and women at various stages of the most common form of OC, serous epithelial OC. This information is inputted into a machine learning model, OC-Dx, trained on blood samples for 183 normal women, 91 women with stage 1 serous epithelial OC, and 188 women with stages 2-4 serous epithelial OC. In testing on these samples with appropriate jackknifing, 100% of normal women are classified with 100% probability as being normal, 93.4% of stage 1 women are classified with 100% probability as having OC with the remaining women diagnosed with 97.4% probability as having OC. The sensitivity of stage 1 diagnosis is .99.8%. For the stage 2-4 OC cohort, 88.8%% of stage 2-4 women are classified with a 100% (95%) probability of having OC, with the remaining women diagnosed with a 98.6% probability of having OC. This gives a sensitivity of 99.84%  Thus, the test approaches the long-sought 100% sensitivity and 100% specificity.

FDA-Approved Competition:

In the following, we present and summarize the status of the 3 currently FDA-approved tests related to Ovarian Cancer serum testing:        

Ova 1 is a protein-based assay that uses proprietary OvaCalc Software to incorporate the values for five analytes from separately run immunoassays into a single numerical score between 0.0 and 10.0 6The five analytes are Cancer Antigen 125 (CA 125), Transferrin (TRF), Apolipoprotein A-1 (APO A-1), Beta-2 Microglobulin (B2M), and Prealbumin (TT). This score is then interpreted in the context of menopausal or premenopausal status. The test is not intended to be used as a screen or standalone diagnostic assay and is intended for use in conjunction with imaging studies and clinical assessments in women for whom surgical intervention is planned.

Overa is an assay that uses 5 protein biomarkers: Cancer Antigen 125 (CA 125), Transferrin (TRF), Apolipoprotein A-1 (APO A-1), Follicle-Stimulating Hormone (FSH), and Human Epididymis Protein 4 (HE4) the results of which are combined to calculate a single cancer risk score using proprietary software – OvaCALC. This score is then interpreted using menopausal or premenopausal status 6 The assay is intended for use as a prognostic indicator of a woman’s likelihood that malignancy is present when the physician’s independent clinical and radiological evaluation does not indicate malignancy.

ROMA 7 is an assay that combines HE4, CA 125, and menopausal status into a numerical score. According to the initial FDA 510K clearance received, ROMATM is intended to aid in assessing whether a premenopausal or post-menopausal woman who presents with an ovarian adnexal mass is at high or low likelihood of finding malignancy on surgery. It is indicated for women who meet the following criteria: over age 18; ovarian adnexal mass present for which surgery is planned, and not yet referred to an oncologist. ROMATM must be interpreted in conjunction with an independent clinical and radiological assessment

Table 1. Comparison of OC-DX With the Competitiona

Test Sensitivity (%) Specificity
OC-DX (OUR TEST) 99.8 100
Ova1 91.6 42.1
Overa 83.5 64.8
Roma 87.3 85.5

aThe best results of the test are reported.

Based on Table 1, OC-DX is clearly superior with a 100% specificity which is essential in the case of a disease like Ovarian Cancer where only a tiny minority of individuals are likely to have the condition. In addition to these tests, there are a plethora of other non-FDA-approved tests on the market. These are tools that are used to help a physician reach a clinical decision. We note that at the moment, OC-DX is one of these clinical decision-support tools. Thus far, the competition as in the case of the FDA has far too low specificity and often is not validated by peer-reviewed publications.

Next Steps

A possible disadvantage of the work done to date is that we knew which women had OC and which did not. While we were very careful to ensure that this information was not used in testing OC-Dx, to truly validate the model we need to demonstrate its performance in a blind test. In our proposed Blinded Expansion Cohort Study designed to blindly validate the performance of OC-Dx, over an approximately one-year period, we propose to identify at least 100 unique women for blood serum collection. These women are about to undergo surgery for an ovarian mass, but whether it is cancer or not is unknown to both the clinician and us. In addition, the precise identity of the women will only be known to the clinician. We will then apply OC-Dx to predict the cancer status of the particular patient. It is only at this point that the pathology report will made available to us. Thus, this study will provide an objective, unbiased evaluation of the performance of OC-Dx. If OC-Dx performs as expected, we will then establish a company to make OC-Dx available to gynecologists as a diagnostic tool.

As you can see, we are cliff-hanging (CEOs are allowed to occasionally coin a new adverb) close to achieving this goal, a goal that would have been impossible to imagine without the hard work and constant encouragement of you, our board of directors. We will end these remarks with a tantalizing step into the future. Once we have brought our test for early-stage ovarian cancer to the highest level of accuracy possible, it will be easy to apply the same methodology to other cancers, especially prostate cancer, where the PSA test is so pathetic.

It will be poetic justice to ruminate on the fact that the prostate was saved because the ovaries got there first!

References

  1. Stages of Ovarian Cancer. (2023). OC-DX: Development and validation of a blood test to diagnose Ovarian Cancer
  2. Patient education: Ovarian cancer diagnosis and staging (Beyond the Basics). (2023). https://www.uptodate.com/contents/ovarian-cancer-diagnosis-and-staging-beyond-the-basics/print#:~:text=During%20the%20early%20stages%20of,symptoms%20(urgency%20and%20frequency).
  3. Key Statistics for Ovarian Cancer. (2023). https://www.cancer.org/cancer/types/ovarian-cancer/about/key-statistics.html.
  4. Cancer Stat Facts: Ovarian Cancer. (2023). https://seer.cancer.gov/statfacts/html/ovary.html.
  1. Torre, L.A., Trabert, B., DeSantis, C.E., Miller, K.D., Samimi, G., Runowicz, C.D., Gaudet, M.M., Jemal, A., and Siegel, R.L. (2018). Ovarian cancer statistics, 2018. CA Cancer J Clin 68, 284-296. 10.3322/caac.21456.
  2. Bristow, R.E., Smith, A., Zhang, Z., Chan, D.W., Crutcher, G., Fung, E.T., and Munroe, D.G. (2013). Ovarian malignancy risk stratification of the adnexal mass using a multivariate index assay. Gynecol Oncol 128, 252-259. 10.1016/j.ygyno.2012.11.022.
  3. Oncologists. SOG. Multiplex Serum testing. (2013). https://www.sgo.org/resources/multiplex-serum-testing-for-women-with-pelvic-mass.

Join OCI in the Fight Against Ovarian Cancer Today.