Adnexal Mass Triage: Solving the Clinical Dilemma

The Clinical Dilemma: To Operate or To Observe?

The Problem:

You have a patient with a localised adnexal mass. Some will be malignant, but others will be benign. Distinguishing them is the central challenge of gynaecologic oncology.

The Constraint:

• No Biopsy Allowed: Unlike other cancers, pre-operative biopsy of a mass limited to the ovary is generally contraindicated.
• Why? Puncturing the capsule can spill tumour cells into the peritoneal cavity, upstaging a localised Stage IA cancer to Stage IC. This worsens prognosis and forces the need for chemotherapy.

The Stakes:

• Malignant: Requires immediate, extensive staging surgery (laparotomy/hysterectomy/omentectomy). Delay leads to stage progression and worse survival.
• Benign: Can often be observed or treated with simple cystectomy. Mistakenly performing radical surgery on a benign mass causes unnecessary morbidity (surgical menopause, infertility, surgical risks).

The Gap:

Conventional tools—CA-125 and standard Ultrasound—have significant limitations. CA-125 has poor specificity (false positives in endometriosis/fibroids), and ultrasound interpretation is subjective.

This guide details the investigations and algorithms developed specifically to enhance the value of CA-125 and solve this dilemma.

Quick Summary of Algorithms

Algorithm	What it Uses (Inputs)	What it Does (Function)
RMI	CA-125 + Ultrasound Score + Menopausal Status	Identifies High Risk: Uses clinical factors to filter out benign causes of elevated CA-125.
IOTA Simple Rules	5 Benign + 5 Malignant Ultrasound Features	Standardizes Diagnosis: Reduces subjectivity by using strict visual criteria.
IOTA ADNEX	3 Clinical Variables (Age, CA-125, Center type) + 6 Ultrasound Variables	Stratifies Risk: Calculates precise % probability for benign vs. 4 types of malignancy.
ROMA / HE4	Serum HE4 (+ CA-125 + Menopause for ROMA)	Improves Specificity: Distinguishes cancer from endometriosis (which raises CA-125 but not HE4).
MIA (OVA1)	5 Serum Proteins (including CA-125 & inflammatory markers)	Safety Net: Maximizes sensitivity to ensure no cancer is missed by generalists.

The First Enhancement of CA-125: Integrating Clinical Factors

Risk of Malignancy Index (RMI)

• How it enhances CA-125: It combines the raw CA-125 value with Ultrasound Morphology and Menopausal Status. This filters out many “false positive” CA-125 elevations caused by benign premenopausal conditions.
• Formula: RMI = U X M X CA-125
  • U (Ultrasound): Score 0, 1, or 3 based on features (solid parts, ascites, etc.).
  • M (Menopause): Score 1 (Pre) or 3 (Post).
• Diagnostic Performance:
  • Threshold: RMI >200 → suggestive of malignancy.
  • Specificity: Typically 92–97%, excellent for ruling in malignancy.
  • Sensitivity: 61–87%, variable by study and population.
• Limitation: It still relies heavily on CA-125, so it frequently misses early-stage cancers where marker levels are normal (Low Sensitivity).

Standardising Imaging: The IOTA Evolution

The International Ovarian Tumuor Analysis (IOTA) group revolutionised ultrasound by moving away from subjective “gut feelings” to evidence-based standardisation.

The Foundation: IOTA Simple Rules (Published 2008)

• Description: A set of 5 Benign Features (e.g., unilocular, acoustic shadows) vs. 5 Malignant Features (e.g., ascites, vascularity).
• Role: A Bedside Triage Tool. It allows one to differentiate between benign and malignant. If only benign features are present, the mass is classified as benign; if only malignant features are present, it is classified as malignant.
• Limitation: It is a binary tool. If a mass has conflicting features (both Benign and Malignant traits), the result is “Inconclusive” (~20% of cases).

The Evolution: IOTA ADNEX Model (Published 2014)

• Description: The “Next Generation” tool. It takes the standardized inputs from the Simple Rules era but feeds them into a sophisticated mathematical algorithm (logistic regression).
• Role: A Precision Calculator. Unlike Simple Rules, it does not have an “Inconclusive” category.
• The Upgrade: Instead of a simple “Yes/No,” it calculates the precise percentage risk for 5 specific outcomes: Benign, Borderline Tumor, Stage I Cancer, Stage II-IV Cancer, or Metastasis.

Enhancing Biomarkers: Fixing Specificity & Sensitivity

Human Epididymis Protein 4 (HE4)

• How it enhances CA-125: CA-125 is an “acute phase reactant” raised in inflammation (endometriosis). HE4 is a protease inhibitor raised in cancer but NOT in inflammation.
• Diagnostic Gain:High Specificity. It solves the “false positive” problem in premenopausal women.
  • Rule of Thumb: High CA-125 + Normal HE4 = Likely Benign (Endometrioma).
• Limitation: HE4 is less sensitive than CA-125 for mucinous ovarian tumors and can be falsely elevated in renal failure or with age. Thus, it complements rather than replaces CA-125.

Risk of Ovarian Malignancy Algorithm (ROMA)

• How it enhances CA-125: It mathematically combines CA-125 + HE4 + Menopausal Status to generate a predictive probability.
• Diagnostic Gain: Superior to CA-125 alone for stratifying pelvic masses into Low vs. High Risk, particularly in the challenging premenopausal group.

Multivariate Index Assays (MIA): OVA1 & Overa

• How it enhances CA-125: CA-125 alone misses ~50% of Stage I cancers. MIA analyzes 5 proteins (including inflammatory markers and CA-125) to capture these “marker-negative” cancers.
• Diagnostic Gain: High Sensitivity (92–97%). It acts as a “Safety Net.” If the test is elevated, referral is mandatory even if the mass looks benign.
• Limitation: High false-positive rate (Low Specificity).

Assessment of Masses in High-Risk Individuals

Standard algorithms often fail in specific high-risk populations due to altered baseline risks or confounding biomarkers. Tailored strategies are required.

Patients with Endometriosis

• The Challenge: Endometriosis causes chronic inflammation, leading to chronically elevated CA-125 and “complex” cystic masses (endometriomas) on ultrasound. This creates a high rate of false positives (“Red Herrings”).
• The Solution: HE4 Integration.
• Rationale: Unlike CA-125, HE4 is typically NOT elevated in endometriosis.
• Diagnostic Strategy:
  • High CA-125 + Normal HE4: Strongly suggests benign endometrioma.
  • High CA-125 + High HE4: Raises suspicion for malignancy (specifically Endometrioid or Clear Cell carcinomas, which can arise from endometriosis).

BRCA1/2 Mutation Carriers

• The Challenge: These patients are prone to High-Grade Serous Carcinoma, which often originates in the fallopian tube and grows rapidly. Single-point screening (one ultrasound or one CA-125) often misses the window of curability.
• The Solution: Longitudinal Tracking (ROCA Test).
  • Rationale: The Risk of Ovarian Cancer Algorithm (ROCA) tracks the rate of change of CA-125 over time rather than a fixed cutoff. A rapid rise (e.g., 10 to 20 U/mL) triggers an alarm even if the value is “normal” (<35 U/mL).
• Evidence: The UKFOCSS Phase 2 Trial showed that ROCA screening in BRCA carriers significantly increased the detection of early-stage (Stage I/II) cancer compared to annual surveillance (down-staging), allowing for less extensive surgery.
• Guideline Status:
  • NICE (UK): Explicitly recommends ROCA for high-risk women deferring preventative surgery.
  • NCCN (USA): Recommends Risk-Reducing Surgery (RRSO) as the only proven life-saving intervention. Surveillance (TVS + CA-125) is an option for those deferring surgery, but guidelines note it has not been proven to reduce mortality.
• Future Direction:Circulating Tumor DNA (ctDNA).
  • Since most BRCA-associated cancers carry TP53 mutations, “Liquid Biopsies” detecting tumor DNA in blood are being developed to identify malignancy before it is visible on imaging.

The Future Enhancement: Artificial Intelligence

AI & Radiomics

• How it enhances diagnosis: Humans have limited visual perception. AI models use Deep Learning (specifically Convolutional Neural Networks) to analyze ultrasound images pixel-by-pixel, identifying “texture” features invisible to the eye.
• Current Models: This has moved beyond concept to validation. Several AI models have been trained on large datasets (e.g., IOTA) to perform automatic segmentation and malignancy prediction.
• Diagnostic Gain: Studies show these models can match or outperform expert sonographers in differentiating benign vs. malignant masses, reducing inter-observer variability.

Guideline Consensus: Who Recommends What?

Broad-based clinical guidelines differ by region but generally agree on the role of these tools for triage (deciding who performs the surgery).

• RCOG (UK/Commonwealth):
  • Recommended Tool: RMI I.
  • Role: Primary Triage. It is the “Standard of Care.” A score >200 mandates referral to a cancer center. It is favored for its simplicity and high specificity.
• ESGO / IOTA (Europe):
  • Recommended Tool: IOTA Simple Rules or ADNEX Model.
  • Role: Preferred Diagnostic. They explicitly state these models are superior to RMI (better sensitivity). They recommend using them to discriminate between benign, borderline, and invasive tumors.
• ACOG (USA):
  • Recommended Tool: OVA1 and ROMA (Multivariate Assays).
  • Role: Referral Decision Support. ACOG states these FDA-cleared tests are useful for generalists to determine if a patient requires referral to a gynecologic oncologist, noting they have higher sensitivity than CA-125 alone. However, they emphasize that expert ultrasound (if available) is often the most effective modality.

Summary Comparison Table

Tool	Clinical Role	Strength	Weakness
RMI	The “Rule In” Tool	High Specificity. If >200, it is almost certainly cancer.	Misses early cancer (Low Sensitivity).
ADNEX	The Modern Standard	Best overall accuracy. Gives stage-specific risk.	Requires app/calculator.
HE4/ROMA	The Specificity Fix	Excellent for ruling out endometriosis false-positives.	Less sensitive for some mucinous tumors.
OVA1	The Safety Net	Near-perfect Sensitivity. Ensures no cancer is missed.	High false-positive rate (Over-triage).

Key References

1. ACOG Practice Bulletin No. 174: “Evaluation and Management of Adnexal Masses.” American College of Obstetricians and Gynecologists, 2016. (The US clinical standard).
2. RCOG Green-top Guideline No. 62: “Management of Suspected Ovarian Masses in Premenopausal Women.” RCOG, 2011. (Establishes RMI as the triage standard in UK/Commonwealth).
3. ESGO/ISUOG/IOTA/ESGE Consensus Statement: “Pre-operative diagnosis of ovarian tumors.” Ultrasound in Obstetrics & Gynecology, 2021. (The European/Modern Standard).
4. Kaijser J, et al. “Presurgical diagnosis of adnexal tumours… a systematic review and meta-analysis.” Hum Reprod Update, 2014. (The definitive data comparing models).
5. Van Calster B, et al. “Evaluating the risk of ovarian cancer before surgery using the ADNEX model…: a multicentre external validation study.” Lancet Oncol, 2014. (Validation of the modern gold standard).
6. Ueland FR, et al. “Effectiveness of a multivariate index assay in the preoperative assessment of ovarian tumors.” Obstet Gynecol, 2011. (Pivotal trial for OVA1/MIA).