In 2009, Moore et al. developed an ovarian malignancy risk algorithm (ROMA) by integrating CA125, serum HE4 values and menopausal status.

In 2011, the FDA recommended the use of the ROMA algorithm in clinical practice to help stratify ovarian cancer risk.

ROMA Value (%) = exp(PI) / [1 + exp(PI)] * 100
In which, PI (Predictive Index) is the predictive index calculated as follows:
- Women before menopause:
PI= -12.0 + 2.38*Ln(HE4) + 0.0626*Ln(CA125)
- Postmenopausal women:
PI= -8.09 + 1.04*Ln(HE4) + 0.732*Ln(CA125)
Where, exp(PI) = e^PI and Ln() is the natural logarithm (or base e logarithm).
ROMA assesses patients into two groups: high-risk and low-risk.

In 2015, Karlsen et al. developed the Copenhagen Index by integrating biomarkers CA 125, serum HE4 and patient age to assess the malignancy risk of ovarian tumors before surgery.

CPH-I = -14.0647+1.0649*log2(HE4) + 0.6050*log2(CA 125) + 0.2672*Age/10

Predicted probability PP (Predicted probability) = e^(CPH–I)/(1+e^(CPH–I))

References:

Moore RG, McMeekin DS, Brown AK et al. Gynecol Oncol. 2009;112(1):40-6

 https://pubmed.ncbi.nlm.nih.gov/18851871/

Karlsen MA, Høgdall EV, Christensen IJ et al. Gynecol Oncol. 2015 ;138(3):640-6. https://pubmed.ncbi.nlm.nih.gov/26086566/

RMI is a tool used for ovarian cancer detection. There are various RMI scoring systems (RMI 1, RMI 2, RMI 3, and RMI 4), with similar accuracies.

Formula for RMI 1, 2 and 3: U x M x CA125 (Cut-off of 200 or 250)

Formula for RMI 4: U x M x S x CA125 (Cut-off of 450)

Img

Ultrasound features: multilocular cysts, solid areas, metastases, ascites, and bilateral lesions.

Historical Validation Data
Less than 25 – < 3% risk of malignancy
25 to 250 – around 20% risk of malignancy
Over 250 – around 75% risk of malignancy

References

Yamamoto Y, Yamada R, Oguri H et al. Eur J Obstet Gynecol Reprod Biol. 2009

https://pubmed.ncbi.nlm.nih.gov/19327881/

The ADNEX risk model can be used by medical doctors to diagnose ovarian cancer in women who have at least one persistent adnexal (ovarian, para-ovarian, and tubal) tumor and are considered to require surgery.1 ADNEX estimates the probability that an adnexal tumor is benign, borderline, stage I cancer, stage II-IV cancer, or secondary metastatic cancer (i.e. metastasis of non-adnexal cancer to the ovary).

The ADNEX model uses nine predictors. There are three clinical variables, age, serum CA-125 level, and type of center (oncology referral center vs other), and six ultrasound variables, maximal diameter of lesion, proportion of solid tissue, more than 10 cyst locules, number of papillary projections, acoustic shadows, and ascites. All patients included required surgery as judged by a local clinician. As with all current diagnostic models for adnexal tumors (e.g. RMI, ROMA) it implies that patients selected for expectant management were excluded when creating the model. As a consequence ADNEX cannot be applied to conservatively treated adnexal tumors.

The manuscript describing the model is published in the BMJ, and we append the abstract of the paper below. The ADNEX model has been externally validated in the original paper, and in six subsequent studies. These studies confirm the discrimination between benign and malignant masses. Further calibration results when presented are good. Discrimination between the five tumor types is more difficult to investigate due to the low prevalence of several categories, and the fact that this is a non-standard analysis that needs to be done with care. As a result, we wrote a letter to the editor about one of the validation studies. Nevertheless, available results are good.

ADNEX cannot replace training and experience in ultrasonography and cannot compensate for poor quality ultrasound equipment.

References

Van Calster B, et al. BMJ 2014;349:g5920. https://www.bmj.com/content/349/bmj.g5920

Epstein E, et al. Ultrasound Obstet Gynecol 2016;47:110-6. https://pubmed.ncbi.nlm.nih.gov/25925783/

Araujo KG, et al.. Ultrasound Obstet Gynecol 2017 https://pubmed.ncbi.nlm.nih.gov/27194129/

Szubert S, et al. Gynecol Oncol 2016;142:490-5. https://pubmed.ncbi.nlm.nih.gov/27374142/

Sayasneh A, et al. Br J Cancer 2016;115:542-8. https://pubmed.ncbi.nlm.nih.gov/27482647/

Meys EM, et al. Ultrasound Obstet Gynecol 2017 https://pubmed.ncbi.nlm.nih.gov/27514486/

Van Calster B, et al. Eur J Epidemiol 2012;27:761-70. https://pubmed.ncbi.nlm.nih.gov/23054032/