Vibrational spectroscopy for detecting endometrial cancer in blood samples: A systematic review and meta-analysis of preclinical studies

Blood-based vibrational spectroscopy shows potential for endometrial cancer detection, but clinical validation is needed before routine implementation.

Abstract

Endometrial cancer (EC) is increasingly prevalent worldwide, highlighting the need for non-invasive blood-based diagnostic triage tools. ATR-FTIR spectroscopy enables rapid, label-free biochemical profiling of plasma or serum for experimental cancer detection. To date, no systematic review or meta-analysis has evaluated the experimental performance of infrared spectroscopy for discriminating EC from non-cancer in blood-based samples. This study synthesizes available evidence to characterize the strength, consistency, and heterogeneity of the underlying spectroscopic signal across preclinical and proof-of-concept studies. MEDLINE, Web of Science, EMBASE, Scopus, Google Scholar, and CENTRAL were searched without language restrictions. Eligible studies evaluated ATR-FTIR spectroscopy of plasma or serum using histopathology as the reference standard. Pooled sensitivity, specificity, likelihood ratios, and diagnostic odds ratios were estimated using a bivariate random-effects model, with assessment of heterogeneity, threshold effects, and publication bias. Five case–control studies comprising 1376 participants were included. For plasma-based analyses, pooled sensitivity was 0.61 (95% CI: 0.59–0.68) and specificity was 0.73 (95% CI: 0.69–0.76), with a diagnostic odds ratio of 4.23 (95% CI: 3.33–5.37). For serum-based analyses, pooled sensitivity and specificity were both 0.62 (95% CI: 0.59–0.65), with a diagnostic odds ratio of 2.65 (95% CI: 2.16–3.25). Substantial heterogeneity and significant threshold effects were observed. Current evidence supports reproducible spectroscopic differences between EC and non-cancer blood samples under experimental conditions. However, methodological heterogeneity and retrospective case–control study designs limit clinical interpretability. These findings provide a benchmark for future prospective validation rather than immediate clinical application.

Introduction

Endometrial cancer is one of the most common gynecologic malignancies worldwide, with more than 400,000 new cases diagnosed each year. In the United States alone, approximately 66,200 new cases of uterine corpus cancer are expected in 2023. The disease predominantly affects postmenopausal women, with the highest incidence occurring after age 60; however, incidence is also rising among younger women. Over the past two decades, uterine corpus cancer incidence has increased by approximately 1% per year among women aged 50 years and older and by nearly 2% per year among women younger than 50 years, reflecting a growing burden in both older and reproductive-age populations [1], [2], [3]. In contrast to the number of new cases, mortality has decreased in countries with high human development index (HDI) scores [4], [5].

Unlike ovarian cancer, which often presents at an advanced stage, most endometrial cancers present with abnormal uterine bleeding and are potentially curable if detected early. However, the diagnostic pathway is invasive, resource-intensive, and frequently applied to large numbers of women who ultimately have benign disease, creating an important unmet clinical need for more efficient triage tools. In this context, blood-based vibrational spectroscopy could serve as a non-invasive front-line test to help prioritize women who require urgent invasive investigation [1].

The evaluation of endometrial thickening by transvaginal ultrasound (TVS) is often used as the first evaluation for patients with uterine bleeding, despite the low intra and interobserver agreement [6], [7]. Endometrial biopsy is recommended for symptomatic patients with endometrial thickening on TVS or as an initial step for those with premenopausal or postmenopausal bleeding [3], [8]. The histological findings are the most important prognostic factor in EC and central features in the latest revision of the International Gynecological and Obstetrics Federation (FIGO) 2023 [9], [10], [11].

Despite extensive investigation of circulating biomarkers such as HE4 and CA-125, no blood-based biomarkers are currently in routine clinical use with sufficient sensitivity and specificity to reliably distinguish endometrial cancer from benign endometrial diseases [12], [13], [14], [15]. During the initial stages of tumors or in the process of deterioration, tissues, blood, and other fluids change the composition, and the metabolism of critical biomolecules [16]. Research is underway to identify early cancer detection markers, including circulating tumor DNA (ctDNA) and circulating microRNAs [17], [18], [19]. However, their high cost and technical difficulties still need to be addressed in clinical practice [19].

Vibrational spectroscopy (VS) is a promising tool for classifying pathological systems from standard systems and identifying structural variations in biomolecules based on their chemical bonds [20], [21], [22]. The mid-infrared fingerprint region (approximately 1800–900 cm⁻¹) contains many of the most information-rich vibrational bands of proteins, lipids, nucleic acids, and carbohydrates, although diagnostically relevant absorption features also occur in the high-wavenumber region [21], [22], [23]. Biofluids, such as plasma and blood serum, can facilitate examination of endometrial tissues without cutting them open [24]. They also contain information on important biomarkers, such as proteins, lipids, carbohydrates, and nucleic acids, which give rise to spectroscopic patterns in healthy and cancerous samples [25]. Although IR can detect subtle changes in the chemistry of biosamples, accurate classification of these data is highly dependent on the development and application of data processing tools and classifiers. There are three main steps involved in processing such data: (1) pre-processing, (2) feature extraction between samples, and (3) classification [26]. Combining spectral data with chemometrics, which applies mathematical and statistical models, enables the extraction of chemical information from spectra and provides a more reliable method for discriminating healthy biofluids from biofluids obtained from patients with cancer [27], [28], [29]. When analyzing EC, VS in blood, plasma, and serum studies is limited. Comparative studies using different data processing tools and classifiers can introduce bias when evaluating diagnostic accuracy tests [28], [30]. To date, no systematic review or meta-analysis has systematically evaluated the experimental performance and methodological readiness of infrared spectroscopy for discriminating endometrial cancer from non-cancer in blood-based samples. Rather than estimating clinical diagnostic accuracy, the present study synthesizes the available evidence to characterize the strength, consistency, and sources of heterogeneity of the underlying spectroscopic signal across preclinical and proof-of-concept studies. By integrating results across diverse analytical pipelines and experimental configurations, this review provides a benchmark for the current state of the field and a framework for the design of future prospective clinical validation studies.

Escrito por: Dra. Ana Rita Panazzolo
Especialista em Ginecologia e Obstetrícia
CRM-SC 6848 | RQE 1714