Role of Salivary MicroRNA as a Marker of Progesterone Resistance in Endometriosis: Preliminary Results from a Single-Institution Experience

Abstract

This feasibility study explores the potential of salivary microRNAs (miRNAs) as non-invasive biomarkers for diagnosing endometriosis and assessing treatment response. Almost one-third of patients with endometriosis do not respond to the standard progestin treatment due to various mechanisms of progesterone resistance. MiRNAs, recognized for their stability in body fluids and role in gene regulation, may offer new diagnostic and prognostic opportunities as they are involved in the pathogenic pathways of endometriosis and progesterone resistance. We sequenced salivary miRNAs in three cohorts of patients: control women without endometriosis and patients with endometriosis who responded and did not respond to standard progestin treatment. This aims to identify the differential miRNA expression profiles associated with therapeutic response to dienogest. The preliminary results demonstrate the feasibility of miRNA sequencing from saliva and reveal distinct miRNA profiles between responders, non-responders, and controls. Key miRNAs, including mir-3168, the mir-200a family, and mir-93-5p, emerged as potential biomarkers, showing significant differential expression linked to both endometriosis presence and treatment response. Further validation of these findings in larger cohorts could pave the way for miRNA-based diagnostic and prognostic tools, potentially reducing diagnostic delays and personalizing treatment approaches for endometriosis patients, also with new target therapies.

Keywords: biomarkers; endometriosis; miRNA; progesterone resistance; prognosis.

Figure 1

Patient’s Global Impression of Change (PGIC) for responders vs. non-responders.

Figure 2

Principal Component Analysis (PCA) of miRNAs differentially expressed in the three groups (green = non-responders, blue = controls, and red = responders). PCA was performed using as predictors all the miRNAs identified as differentially expressed in at least one patient group in the comparison of each patient group to the control state.

Figure 3

PCA plot of the controls (blue) vs. all patients with endometriosis (responders and non-responders) (green).

Figure 4

Volcano plot showing the relationship between fold change and statistical significance (FDR-adjusted p-value) of miRNAs differentially expressed in non-responder patients compared to responders. The blue points in the plot represent the significantly (FDR ≤ 0.1) downregulated miRNAs (see Table 3).

Figure 5

Heatmap showing the unsupervised clustering of samples based on the expression levels of the three differentially expressed miRNAs (let7c-5p, mir200a-3p, and mir3168) in responder and non-responder patients. The color scale represents the level of upregulation (red) and downregulation (blue) of miRNA. The clustering based on the three miRNAs effectively separates responders (purple bar) from non-responder patients (green bar).

Figure 6

Volcano plot showing the relationship between fold change and statistical significance (p-value) of miRNAs differentially expressed in non-responder patients compared to responders. The blue and red points in the plot represent the significantly (p-value ≤ 0.05) downregulated and upregulated miRNAs, respectively (see Table 4).

Figure 7

Heatmap showing the unsupervised clustering of samples based on the expression levels of the eight differentially expressed miRNAs (let7c-5p, mir200a-3p, mir3168, mir23b-3p, mir27b-3p, mir141-3p, mir155-5p, and mir143-3p) in responder and non-responder patients. The clustering based on the eight miRNAs was effective in discriminating responders (purple bar) from non-responders (green bar) in all cases except one non-responder.

Figure 8

Volcano plot showing the relationship between fold change and statistical significance (FDR-adjusted p-value) of miRNAs differentially expressed in patients compared to healthy controls. The blue and red points in the plot represent the significantly (FDR ≤ 1) downregulated and upregulated miRNAs, respectively (see Table 5).

Figure 9

Heatmap showing the unsupervised clustering of samples based on the expression levels of differentially expressed miRNAs (mir93-5p and mir3168) in patients (green bar) compared to controls (purple bar). A group of patients clustered together with healthy subjects.

Figure 10

Volcano plot of miRNAs differentially expressed in patients compared to controls with a p-value ≤ 0.05.

Figure 11

Heatmap showing unsupervised clustering of samples based on the expression levels of differentially expressed miRNAs in patients (green) compared to controls (purple).

Figure 12

Volcano plot of miRNAs differentially expressed in responders compared to controls.

Figure 13

Heatmap showing unsupervised clustering of samples based on the expression levels of differentially expressed miRNAs (mir93-5p, mir205-5p, and mir143-3p) in responders (green) compared to controls (purple).