International Journal of Molecular Medical Science, 2024, Vol.14, No.5, 264-273 http://medscipublisher.com/index.php/ijmms 5 3.3 Biomarkers beyond HPV While HPV-related genetic and epigenetic alterations are central to cervical carcinogenesis, other molecular markers beyond HPV have also been identified as potential contributors to disease progression. Recent research has focused on microRNAs (miRNAs) and long non-coding RNAs (lncRNAs), which play crucial roles in regulating gene expression and are often dysregulated in cancer. Specific miRNAs, such as miR-21, miR-34a, and miR-143-3p, have shown altered expression in cervical cancer and precancerous lesions, correlating with disease severity (Wittenborn et al., 2020). These miRNAs can serve as both diagnostic and prognostic markers, adding valuable information beyond traditional HPV testing. Additionally, protein biomarkers like p16INK4a, Ki-67, and cyclin E—associated with cell proliferation and cycle regulation—have been proposed as surrogate markers for HPV-related oncogenesis and are being explored for their potential in cervical cancer screening and diagnosis. These biomarkers could enhance the accuracy of screening by helping to distinguish between benign HPV infections and lesions at higher risk of progressing to invasive cancer, ultimately improving early detection and patient management. 4 Advances in Genetic Marker-based Screening 4.1 Next-generation sequencing (NGS) Next-Generation Sequencing (NGS) has transformed the field of genetic screening by providing high-throughput, comprehensive analysis of genetic alterations linked to cervical cancer (Zhong, 2024). NGS enables the simultaneous sequencing of multiple genes, including those associated with HPV integration and host cell mutations, offering detailed insights into the genetic landscape of cervical lesions (Lee et al., 2020). By detecting somatic mutations, single nucleotide variants (SNVs), and copy number variations (CNVs), NGS can uncover specific genetic signatures related to the progression of precancerous lesions to invasive cervical cancer. For example, mutations in genes like PIK3CA, TP53, and PTEN have been identified in cervical cancer patients through NGS, providing potential markers for early detection and personalized risk assessment (Espinosa et al., 2013). Moreover, NGS-based liquid biopsy approaches, which analyze circulating tumor DNA (ctDNA) in the blood, are being investigated for their potential as a non-invasive screening method that can monitor genetic changes in real time, supporting early detection and treatment monitoring. 4.2 DNA Methylation panels DNA methylation, an epigenetic modification that regulates gene expression, has emerged as a promising marker for the early detection of cervical cancer. The methylation of specific gene promoters, both in the host and the HPV genome, has been linked to the silencing of tumor suppressor genes and the progression of cervical intraepithelial neoplasia (CIN) to invasive cancer. Methylation panels, which assess the methylation status of multiple genes, have been developed as triage tools to identify high-grade lesions among women with HPV-positive results. For example, a panel that includes methylation markers such as ANKRD18CP, C13ORF18, and EPB41L3 has shown high sensitivity and specificity in distinguishing CIN2+ lesions from lower-grade abnormalities (Li et al., 2021). Using DNA methylation panels alongside HPV testing has the potential to enhance the accuracy of cervical cancer screening, thereby reducing unnecessary colposcopies and biopsies by providing a more precise risk assessment. Additionally, these panels can be applied to self-collected samples, facilitating broader population-based screening, particularly in low-resource settings where access to healthcare facilities is limited. 4.3 Machine learning in screening The integration of machine learning and artificial intelligence (AI) into cervical cancer screening has the potential to significantly enhance the interpretation of genetic and epigenetic data, leading to more accurate and efficient screening strategies (Hou et al., 2022). Machine learning algorithms can analyze complex datasets, including genomic, methylation, and imaging data, to identify patterns linked to cervical cancer risk. For instance, ensemble learning techniques have been used to predict cervical cancer risk by combining genetic markers with clinical features, thereby improving the robustness and accuracy of risk stratification models. By incorporating gene-assistance modules and data correction mechanisms, these machine learning models can tackle the challenges of variability and ambiguity in cervical cancer screening, providing personalized risk assessments and
RkJQdWJsaXNoZXIy MjQ4ODYzNQ==