Revolutionizing Endometriosis Diagnosis: Validation of a Saliva-Based microRNA Signature

Highlight

• A saliva-based microRNA (miRNA) signature demonstrates high diagnostic accuracy for endometriosis.
• The prospective, multicenter validation study shows 96.6% overall accuracy with excellent sensitivity and specificity.
• The biomarker outperforms conventional imaging modalities in diagnostic precision.
• This noninvasive diagnostic tool holds promise for earlier, less invasive endometriosis detection and management.

Study Background

Endometriosis is a chronic gynecological condition characterized by the presence of endometrial-like tissue outside the uterus, causing chronic pelvic pain, dysmenorrhea, and infertility. Diagnosis remains challenging, largely relying on invasive laparoscopic surgery with histological confirmation or imaging techniques that have variable sensitivity and specificity depending on lesion location and disease severity. This diagnostic delay contributes to prolonged patient morbidity and increased healthcare costs.

Emerging molecular biomarkers including circulating microRNAs (miRNAs) have shown promising potential for noninvasive disease detection. miRNAs are small, non-coding RNAs involved in post-transcriptional regulation of gene expression and have been implicated in the pathophysiology of endometriosis. Saliva as a diagnostic medium offers a convenient, noninvasive alternative to blood sampling, with the potential for wider accessibility and patient acceptability.

Study Design

The ENDOmiRNA study was a prospective, multicenter external validation trial designed to evaluate the diagnostic accuracy, biological reproducibility, and clinical utility of a saliva miRNA signature for endometriosis diagnosis. The study recruited 971 female patients aged 18 to 43 years presenting with signs and symptoms suggestive of endometriosis from diverse medical centers.

Diagnostic confirmation was established via imaging modalities (transvaginal ultrasound, magnetic resonance imaging), laparoscopic surgery, or both. Importantly, all patients classified as controls underwent laparoscopy to confidently exclude endometriosis. The saliva miRNA signature assessment was performed blinded to clinical diagnoses to minimize bias.

Primary endpoints included the diagnostic accuracy, sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and likelihood ratios of the saliva miRNA test compared to the reference standard of imaging and/or laparoscopy with histology.

Key Findings

The study population demonstrated a high prevalence of endometriosis (77%). The saliva miRNA signature yielded an outstanding overall diagnostic accuracy of 96.6% (95% CI, 95.2 to 97.6%), with a sensitivity of 97.3% (95% CI, 96.4 to 98.0%) and specificity of 94.1% (95% CI, 91.0 to 96.4%).

Positive predictive value was 98.2% (95% CI, 97.3 to 98.9%), and negative predictive value was 91.3% (95% CI, 88.3 to 93.4%). The positive likelihood ratio was robust at 16.6, indicating a strong ability to rule in disease, whereas the negative likelihood ratio was 0.03, confirming excellent capability to rule out endometriosis.

Among patients with surgical confirmation (laparoscopy and histology), the saliva miRNA signature showed low rates of misclassification (4.6%), underestimation (2.4%), and overestimation (2.2%). Comparatively, imaging modalities had substantially higher misclassification (27.2%), underestimation (15.1%), and overestimation (12.2%) rates. This suggests that the saliva miRNA test surpasses the diagnostic accuracy of current imaging techniques.

Implications of Performance Metrics

The high sensitivity implies the saliva miRNA signature reliably identifies patients with endometriosis, minimizing false negatives, which is critical to avoid missed diagnoses and treatment delays. Similarly, high specificity reduces false positives, sparing patients unnecessary invasive procedures or treatments. The strong likelihood ratios further support the test’s robust clinical diagnostic value.

Expert Commentary

This large-scale, well-designed multicenter validation of a saliva-based miRNA signature is a landmark in the quest for noninvasive and accurate endometriosis diagnostics. It addresses previous limitations of small sample sizes and single-center cohorts. The blinded assessment methodology strengthens the study’s internal validity.

The biological plausibility is supported by miRNAs’ established role in modulating gene expression relevant to endometriosis pathogenesis, including inflammation, angiogenesis, and cellular adhesion. Salivary detection simplifies sample acquisition and may facilitate wider screening or early diagnosis in primary care and resource-limited settings.

Nevertheless, replication in ethnically diverse populations and assessment in community settings will be important to assess generalizability. Moreover, cost-effectiveness analyses and integration into clinical workflows require further exploration.

Limitations include the high disease prevalence in the cohort, which might overestimate PPV and affect generalizability. The study did not report on the diagnostic performance according to endometriosis subtypes or stages, which could inform clinical applicability further.

Conclusions

The ENDOmiRNA multicenter validation study robustly confirms that a saliva microRNA signature is a highly accurate, noninvasive diagnostic tool for endometriosis. This advancement has the potential to reduce diagnostic delay, lower reliance on invasive procedures, and improve patient-centered care.

Future research should focus on implementation pathways, cost-benefit analyses, and evaluating the signature’s performance across disease severities and diverse populations to firmly establish its role in routine clinical practice.

Funding and Trial Registration

This study was funded by Ziwig and is registered at ClinicalTrials.gov under number NCT05244668.

References

Bendifallah S, Roman H, Suisse S, et al. Validation of a Saliva Micro-RNA Signature for Endometriosis. NEJM Evid. 2025 Nov;4(11):EVIDoa2400195. doi:10.1056/EVIDoa2400195. Epub 2025 Oct 28. PMID: 41147827.