Ultrasound to Differentiate Benign From Malignant Ovarian Tumors—Are We There Yet?

Adnexal (ovarian) tumors present a complex problem. Ovarian cancer (Ovca) is the second most common gynecologic cancer in the United States with the highest mortality rate of all gynecologic cancer, 7th among all cancers, and with a general survival rate of 50%.1 Thus, missing Ovca when performing any kind of test (false negative) will have grave consequences but suspecting it when not present (false positive) can have almost as critical results with morbidity and mortality secondary to (unnecessary) intervention.

The purpose of this post is not to review the differential diagnosis of ovarian tumors nor to discuss chemical markers such as CA125 or cancer-specific signal found on cell-free DNA (cfDNA) but to concentrate on ultrasound. Some tumors are relatively easy to recognize because of defined ultrasound characteristics: corpus luteum with the classic “ring of fire” or endometrioma with the ground-glass appearance content, for instance (image 1a and b). Conversely, a large, multilocular lesion with solid components and profuse internal Doppler blood flow leaves little doubt about its malignant nature (image 2).

Image 2: A large, multilocular lesion with solid components.

What are the ultrasound characteristics we look at?

  1. Size: Unilocular cystic ovarian tumor < 10 cm in diameter or simple septated cystic ovarian tumor < 10 cm in diameter rarely, if ever, are neoplastic.2
  2. Volume: Normal volume for premenopausal and postmenopausal ovaries are < 20 cm3 and 10 cm3, respectively.
  3. Appearance: Risk of malignancy in simple, unilocular anechoic cyst, less than 5 cm is < 1% in premenopause and about 2.8% in postmenopause.3
  4. Blood flow criteria: The rationale is that arteries formed by neovascularization in malignant tumors lack tunica media, resulting in lowered impedance (= less resistance to blood flow). Thus, resistance indices will be lower in cancer than in benign tumors. Malignancy was suspected with Doppler indices: pulsatility index (PI)<1 and/or resistive index (RI)<0.4.4 However, too much overlap makes reliance on only Doppler unjustified.

A very important point is that the expert performs very well when analyzing the ultrasound images of an ovarian mass, with a sensitivity of 92–98% and a specificity of 89%. The issue is how to help the non-expert decide whether he/she can continue the care of the patient or needs to refer her to a specialist. Based on several ultrasound criteria, scoring systems were implemented. The first one, in 1990, included appearance (unilocular, unilocular solid, multilocular, multilocular solid, or solid cyst) and presence of papillae (graded according to their number: 0 [none], 1 [one to five], or 2 [more than five]). This method had a sensitivity (true positive rate, or chance that person testing positive actually has Ovca) for malignancy of 82% with a specificity (true negative rate or chance that person with a negative test does not have Ovca) of 92%.5 Two important additional scoring systems were described later: the Morphology Index (MI) combining tumor volume, wall structure, and septal structure and the Risk of Malignancy Index (RMI), the product of ultrasound morphology score, CA 125 level, and menopausal status.6 Additional systems included the Logistic Regression 1 (LR1) and 2 (LR2). None of the published scoring systems were superior to image assessment by an expert, including in a meta-analysis of 47 articles, including over 19000 adnexal masses7 and, in reality, were not used widely in clinical practice.

The International Ovarian Tumor Analysis (IOTA) models

In 2000, a large group of European experts (gynecologists, radiologists, statisticians, biology, and computer experts) published a standardized terminology for the characterization of adnexal masses.8

The two important systems are the Simple Rules (SR) and the Assessment of Different NEoplasias in the adneXa (ADNEX) model. These were externally validated in numerous centers across the world but not in the USA.9 Recently, however, validation on the largest hitherto US population was published.10 This study showed for the first time that the models were effective in this population, regardless of menopausal status or race. These models are easy to learn and are geared towards non-experts.11 It is important to note that the IOTA group was one of the first to incorporate acoustic shadow as a key feature, and the acoustic shadow has been shown to be an important sonographic feature to consider.12

  1. Simple Rules: The IOTA Simple-Rules consist of 2 sets of 5 elements each: benign and malignant.13 Three simple rules are applied: if only benign characteristics are present, the mass is classified as benign. If only malignant features are present, the mass is considered malignant. If no features or both are, the findings are inconclusive. This model works well in about 80% of cases. The other 20% should be referred to an expert.
  2. ADNEX model14: This is a multiclass prediction model to differentiate between benign and malignant tumors and allows automatic calculation of sub-classification of malignant tumors into borderline tumors, Stage I, and Stage II–IV primary cancers, and secondary metastatic tumors. “The advantage of this model is that it gives a personalized risk score for each patient, based on age, whether the patient is seen at an oncology center or not, maximal diameters of the lesion and the solid parts, number of cysts and papillary projections, whether acoustic shadows are present, whether ascites is present and CA125 value (if available, not mandatory for calculation). With a cut-off value for malignancy risk set at 10%, the ADNEX model (with CA125) had a sensitivity of 94.3%, with a specificity of 74%, positive predictive value of 76%, and negative predictive value of 93.6%.”14

The O-RADS model

In 2020, the American College of Radiology convened an international multidisciplinary committee that developed an ultrasound model based on an MRI model used in mammography (the BI-RADS atlas), the O-RADS model (the Ovarian-Adnexal Reporting and Data System) to facilitate differentiation between benign and malignant ovarian tumors.15 It relies on the sonographic nomenclature developed by the IOTA group, but it classifies tumors into 1 of 6 categories (O-RADS 0–5), from normal to high risk of malignancy. O-RADS also includes guidelines for the management of the findings. It should be noted that the O-RADS first model did not take into account the presence or absence of an acoustic shadow, although this has now been amended.

A description of the most recent common ultrasound scoring systems (SR, ADNEX, and O-RADS) is available in the Journal of Ultrasound in Medicine (JUM): Yoeli-Bik R, Lengyel E, Mills KA, Abramowicz JS. Ovarian masses: The value of acoustic shadowing on ultrasound examination. J Ultrasound Med 2023; 42:935–945.    

References

  1. https://www.cancer.org/cancer/types/ovarian-cancer/about/key-statistics.html
  2. Saunders et al. Risk of malignancy in sonographically confirmed septated cystic ovarian tumors. Gynecol Oncol 2010; 118:278–282.
  3. Valentin et al. Risk of malignancy in unilocular cysts: a study of 1148 adnexal masses classified as unilocular cysts at transvaginal ultrasound and review of the literature. Ultrasound Obstet Gynecol 2013; 41:80–89.
  4. Bourne et al. Transvaginal colour flow imaging: a possible new screening technique for ovarian cancer. BMJ 1989; 299:1367–370.
  5. Granberg S et al. Tumors in the lower pelvis as imaged by vaginal sonography. Gynecol Oncol 1990; 37: 224–229.
  6. Yamamoto Y, Yamada R, Oguri H, Maeda N, Fukaya T. Comparison of four malignancy risk indices in the preoperative evaluation of patients with pelvic masses. Eur J Obstet Gynecol Reprod Biol 2009; 144:163–167.
  7. Meys EM et al. Subjective assessment versus ultrasound models to diagnose ovarian cancer: A systematic review and meta-analysis. Eur J Cancer 2016; 58:17–29.
  8. Timmerman D, Van Calster B, Testa A, et al. Predicting the risk of malignancy in adnexal masses based on the simple rules from the international ovarian tumor analysis group. Am J Obstet Gynecol 2016; 214:424–437.
  9. Abramowicz JS, Timmerman D. Ovarian mass-differentiating benign from malignant: the value of the International Ovarian Tumor Analysis ultrasound rules. Am J Obstet Gynecol 2017; 217:652–660.
  10. Yoeli-Bik R, Longman RE, Wroblewski K, Weigert M, Abramowicz JS, Lengyel E. Diagnostic performance of ultrasonography-based risk models in differentiating between benign and malignant ovarian tumors in a US cohort. JAMA Netw Open 2023; 6:e2323289.
  11. Valentin L, Ameye L, Jurkovic D, et al. Which extrauterine pelvic masses are difficult to correctly classify as benign or malignant on the basis of ultrasound findings and is there a way of making a correct diagnosis? Ultrasound Obstet Gynecol 2006; 27:438–444.
  12. Yoeli-Bik R, Lengyel E, Mills KA, Abramowicz JS. Ovarian masses: The value of acoustic shadowing on ultrasound examination. J Ultrasound Med 2023; 42:935–945.
  13. Timmerman D, Testa AC, Bourne T, et al. Simple ultrasound-based rules for the diagnosis of ovarian cancer. Ultrasound Obstet Gynecol 2008; 31:681–90.
  14. Van Calster B, et al. Evaluating the risk of ovarian cancer before surgery using the ADNEX model to differentiate between benign, borderline, early and advanced stage invasive, and secondary metastatic tumours: prospective multicentre diagnostic study. BMJ 2014; 349:g5920.
  15. Andreotti RF, Timmerman D, Strachowski LM, et al. O-RADS US risk stratification and management system: a consensus guide-line from the ACR ovarian-adnexal reporting and data system committee. Radiology 2020; 294:168–185.

Appendix

Classification of primary ovarian tumors

  1. Ovulatory: functional or corpus luteum cyst; theca lutein cyst; polycystic ovary
  2. Infectious or inflammatory: tubo-ovarian abscess; hydrosalpinx
  3. Benign: serous or mucinous cystadenoma; endometrioma; mature cystic teratoma (most common primary benign tumor of the ovary); paraovarian/paratubal cysts
  4. Borderline: serous, mucinous
  5. Malignant
  6. Epithelial: high-grade serous carcinoma (HGSC; 70 to 80%); endometrioid carcinoma (10%); clear cell carcinomas (10%); mucinous carcinoma (3%); Low-grade serous carcinoma (LGSC; <5%); Brenner tumor; carcinosarcoma or malignant mixed müllerian tumor (MMMT); undifferentiated,
  7. Germ cell (20%): teratoma: immature, specialized teratomas of the ovary (struma ovarii, carcinoid tumor); dysgerminoma; yolk sac tumor: endodermal sinus tumor; embryonal carcinoma; choriocarcinoma: <1% of ovarian tumors; malignant mixed germ cell tumor
  8. Sex cord / stromal ovarian tumors (8–10%): fibrothecoma (fibroma, thecoma); Sertoli-Leydig cell tumor; granulosa cell tumor (juvenile or adult); small cell carcinoma

Jacques S. Abramowicz, MD, is a professor in the Department of Obstetrics and Gynecology at the University of Chicago.

Interested in learning more about gynecologic ultrasound? Check out the following posts from the Scan:

Mastering Ovarian Tumor Analysis: Join the AIUM-IOTA Partnership Course for Advanced Gynecologic Ultrasound

Ovarian lesions are a common finding among women, with etiologies ranging from ovarian changes related to normal hormonal function to aggressive malignancies. Therefore, the proper diagnosis and management of ovarian lesions are critical to women’s health. Here, I’ll give a brief description of ovarian tumor analysis, including descriptors, pattern recognition, and the application of the International Ovarian Tumor Analysis (IOTA) group’s Simple Rules, the IOTA ADNEX model, and O-RADS ultrasound characterization.

An ultrasound image of ovarian lesions.

Descriptive Analysis of Ovarian Tumors

The first step in the diagnosis of ovarian tumors is descriptive analysis. This step involves a detailed examination of the tumor’s characteristics, including its size, shape, texture, and location. This information is obtained through various imaging techniques, such as ultrasound, MRI, and CT scans. The following descriptors are used in descriptive analysis:

  • Size: The size of the tumor is measured in centimeters and is one of the critical factors in determining the type of tumor.
  • Shape: The shape of the tumor is described as either round or irregular. An irregular shape is often associated with malignant tumors.
  • Texture: The texture of the tumor is classified as either solid, cystic, or mixed.
  • Location: The location of the tumor is described as either unilateral or bilateral. Unilateral tumors are located on one ovary, while bilateral tumors are located on both ovaries.

Pattern Recognition of Ovarian Tumors

An essential aspect of ovarian tumor analysis is pattern recognition. It involves identifying specific patterns associated with malignant and benign tumors. The following patterns are commonly observed in ovarian tumors:

  • Solid: Solid tumors are characterized by the absence of cystic components and are often associated with malignancy.
  • Cystic: Cystic tumors are characterized by the presence of fluid-filled spaces and are typically benign.
  • Mixed: Mixed tumors have both solid and cystic components and can be either benign or malignant.

Application of the Simple Rules, the IOTA ADNEX Model, and O-RADS Ultrasound Characterization

The Simple Rules, the IOTA ADNEX Model, and O-RADS ultrasound characterization are 3 widely used methods for differentiating ovarian tumors.

  • The Simple Rules: The Simple Rules are a set of guidelines that assist in the diagnosis of ovarian tumors. The rules are based on the tumor’s size, shape, texture, and location. According to the Simple Rules, a tumor is considered benign if it meets all 3 of the following criteria: 1) it is purely cystic, 2) it is less than 10 cm in size, and 3) it has a thin, smooth wall.
  • IOTA ADNEX Model: The IOTA ADNEX Model is a predictive model that uses a combination of clinical and ultrasound findings to diagnose ovarian tumors. The model considers the tumor’s size, shape, texture, location, and other factors, such as the patient’s age and menopausal status. Then, the model provides a probability score for each tumor, indicating the likelihood of malignancy.
  • O-RADS Ultrasound Characterization: O-RADS is a standardized ultrasound reporting system that categorizes ovarian tumors based on their likelihood of malignancy. The system uses a 5-point scale, ranging from 1 (very low risk) to 5 (very high risk). The O-RADS system considers the tumor’s size, shape, texture, location, and vascularity.

The proper diagnosis and management of ovarian lesions are critical to women’s health. Descriptive analysis, pattern recognition, and the application of the Simple Rules, the IOTA ADNEX Model, and O-RADS ultrasound characterization are essential aspects of ovarian tumor analysis. These methods aid in accurately diagnosing and differentiating ovarian tumors and can guide appropriate treatment decisions.

Are you a healthcare professional looking to enhance your skills in gynecologic ultrasound and ovarian tumor analysis? Look no further than the Advanced Gynecologic Ultrasound course offered by the American Institute of Ultrasound in Medicine (AIUM) in partnership with the International Ovarian Tumor Analysis (IOTA) group.

This course offers a unique and valuable opportunity for healthcare professionals looking to enhance their skills in gynecologic ultrasound and ovarian tumor analysis. The comprehensive curriculum, hands-on training, and networking opportunities make it a worthwhile investment for healthcare professionals looking to improve patient outcomes and advance their careers. Register now for the course, taking place this June, at the AIUM Headquarters in Laurel, Maryland.

Sources
https://www.cancer.org/cancer/types/ovarian-cancer/about/what-is-ovarian-cancer.html
https://acsjournals.onlinelibrary.wiley.com/doi/pdf/10.1002/cncr.11339
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC5620878/
https://pubmed.ncbi.nlm.nih.gov/18504770/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC4402441/
https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9728190/
https://www.mdpi.com/2075-4418/13/5/885

Arian Tyler, BS, is the Digital Media and Communications Coordinator for the American Institute of Ultrasound in Medicine (AIUM).