(L-476) Quantification of Key Biomarkers for Point-of-Care Cervical Cancer Screening

Introduction:: Cervical cancer is the fourth most common cancer in women and the fourth leading cause of cancer-related deaths among them globally. Low- and middle-income countries account for roughly 84% and 88% of cervical cancer cases and deaths, respectively. Current cervical cancer screening techniques, including the Papanicolaou smear test and human papillomavirus (HPV) DNA testing, are expensive and time-consuming, possess low sensitivity or specificity, require trained personnel, and depend on sophisticated instrumentation, limiting their implementation in these low-resource settings. Therefore, there is an urgent need to develop a low-cost, rapid point-of-care (POC) assay capable of diagnosing early-stage cervical cancer with high sensitivity and specificity. Cervical intraepithelial neoplasia lesions and invasive cervical cancer have a highly heterogenous tumor microenvironment and are not all HPV-dependent, limiting the utility of a single biomarker. Multiplex detection of cervical precancer- and cancer-associated biomarkers would enable more reliable, sensitive, and specific identification of these lesions. This study quantifies four known key cervical cancer protein biomarkers in four immortalized cervical cancer cell lines and one normal primary cervical epithelial cell strain. This quantification is the critical first stage to integrate these proteins into a POC test.


Materials and Methods:: Commercial enzyme-linked immunosorbent assays (ELISAs) (from Biomatik and CUSABIO) were used to quantify four known critical cervical cancer proteins labeled as A, B, C, and D in four distinct cancer cell lines and one normal primary cervical epithelial cell strain. Radioimmunoprecipitation assay lysates of HPV-positive HeLa, Ca Ski, and HT-3 cervical cancer cell lines, HPV-negative C-33 A cervical cancer cell line, and normal primary cervical epithelial cells (PCS-480-011™) were evaluated. Bicinchoninic acid assays were utilized to normalize the data obtained through the ELISAs to gain a better understanding of the protein expression trends. Dunnett’s multiple comparisons tests were carried out to analyze statistical significance against the primary cervical epithelial cell lysates.


Results, Conclusions, and Discussions:: The biomarker-to-total protein ratios are shown in Fig. 1. The ratios for proteins A and C were lower in all the HPV-positive cell line lysates than those of the normal ones. This was mainly due to the normal lysates’ roughly six-fold lower total protein concentration and relatively similar biomarker level compared to the HPV-positive cancer lysates. Of note, overall protein C expression at sub-ng/mL levels will require signal enhancement for POC detection. On the other hand, for proteins B and D, the ratio was higher in C-33 A and Ca Ski lysates, respectively. The smaller normal lysate ratios observed here may be attributed to the comparatively low biomarker levels measured for these lysates. Of note, protein B expression was considerably higher in the HPV-negative C-33 A lysates, pointing to the possible influence of HPV infection on protein expression.

To sum up, the quantification of these proteins underscores the importance of targeting a combination of signature biomarkers. Also, these measurements provide insights into POC protein detection feasibility, highlighting the significance of integrating enzymatic signal enhancement into the diagnostic platform to enable biomarker identification at clinically relevant concentrations. Given the high protein expression variability across cell lines, supervised machine learning will be explored to obtain a binary decision threshold, preventing false-positive and false-negative results. Future work will validate these data with clinical cervical swab specimens and finalize the limit of detection requirements for the POC cervical cancer detection device.


Acknowledgements (Optional): : This project is funded by NIH R01CA246315.


References (Optional): :