RESUMO
PURPOSE: Ovarian cancer is oftendiagnosed late due to vague symptoms, leading to poor survival rate. Improved screening tests could mitigate this issue. This narrative review examines the potential and challenges of integrating artificial intelligence (A.I.) into ovarian cancer screenings, with a focus on improving early detection, diagnosis, and personalized risk assessment. METHOD: A comprehensive review of existing literature was conducted, analyzing studies and discussions within the scientific community. RESULTS: A.I. shows promise in significantly improving the ovarian cancer screening processes, increasing accuracy, efficiency, and resource allocation. However, data quality and bias issues pose considerable challenges, potentially leading to healthcare disparities. CONCLUSIONS: Integrating A.I. into ovarian cancer screenings offers potential benefits but comes with significant challenges. By promoting diverse data collection, engaging with underrepresented groups, and ensuring ethical data use, A.I. can be harnessed for more accurate and equitable ovarian cancer diagnoses.
RESUMO
One unexploited family of cancer biomarkers comprise glycoproteins, carbohydrates, and glycolipids (the Tumor Glycocode).A class of glycolipid cancer biomarkers, the tumor-marker gangliosides (TMGs) are presented here as potential diagnostics for detecting cancer, especially at early stages, as the biological function of TMGs makes them etiological. We propose that a quantitative matrix of the Cancer Biomarker Glycocode and artificial intelligence-driven algorithms will expand the menu of validated cancer biomarkers as a step to resolve some of the challenges in cancer diagnosis, and yield a combination that can identify a specific cancer, in a tissue-agnostic manner especially at early stages, to enable early intervention. Diagnosis is critical to reducing cancer mortality but many cancers lack efficient and effective diagnostic tests, especially for early stage disease. Ideal diagnostic biomarkers are etiological, samples are preferably obtained via non-invasive methods (e.g. liquid biopsy of blood or urine), and are quantitated using assays that yield high diagnostic sensitivity and specificity for efficient diagnosis, prognosis, or predicting response to therapy. Validated biomarkers with these features are rare. While the advent of proteomics and genomics has led to the identification of a multitude of proteins and nucleic acid sequences as cancer biomarkers, relatively few have been approved for clinical use. The use of multiplex arrays and artificial intelligence-driven algorithms offer the option of combining data of known biomarkers; however, for most, the sensitivity and the specificity are below acceptable criteria, and clinical validation has proven difficult. One strategic solution to this problem is to expand the biomarker families beyond those currently exploited. One unexploited family of cancer biomarkers comprise glycoproteins, carbohydrates, and glycolipids (the Tumor Glycocode). Here, we focus on a family of glycolipid cancer biomarkers, the tumor-marker gangliosides (TMGs). We discuss the diagnostic potential of TMGs for detecting cancer, especially at early stages. We include prior studies from the literature to summarize findings for ganglioside quantification, expression, detection, and biological function and its role in various cancers. We highlight the examples of TMGs exhibiting ideal properties of cancer diagnostic biomarkers, and the application of GD2 and GD3 for diagnosis of early stage cancers with high sensitivity and specificity. We propose that a quantitative matrix of the Cancer Biomarker Glycocode and artificial intelligence-driven algorithms will expand the menu of validated cancer biomarkers as a step to resolve some of the challenges in cancer diagnosis, and yield a combination that can identify a specific cancer, in a tissue-agnostic manner especially at early stages, to enable early intervention.
RESUMO
Background: Ovarian cancer (OC) is the deadliest gynecological cancer, often diagnosed at advanced stages. A fast and accurate diagnostic method for early-stage OC is needed. The tumor marker gangliosides, GD2 and GD3, exhibit properties that make them ideal potential diagnostic biomarkers, but they have never before been quantified in OC. We investigated the diagnostic utility of GD2 and GD3 for diagnosis of all subtypes and stages of OC. Methods: This retrospective study evaluated GD2 and GD3 expression in biobanked tissue and serum samples from patients with invasive epithelial OC, healthy donors, non-malignant gynecological conditions, and other cancers. GD2 and GD3 levels were evaluated in tissue samples by immunohistochemistry (n=299) and in two cohorts of serum samples by quantitative ELISA. A discovery cohort (n=379) showed feasibility of GD2 and GD3 quantitative ELISA for diagnosing OC, and a subsequent model cohort (n=200) was used to train and cross-validate a diagnostic model. Results: GD2 and GD3 were expressed in tissues of all OC subtypes and FIGO stages but not in surrounding healthy tissue or other controls. In serum, GD2 and GD3 were elevated in patients with OC. A diagnostic model that included serum levels of GD2+GD3+age was superior to the standard of care (CA125, p<0.001) in diagnosing OC and early-stage (I/II) OC. Conclusion: GD2 and GD3 expression was associated with high rates of selectivity and specificity for OC. A diagnostic model combining GD2 and GD3 quantification in serum had diagnostic power for all subtypes and all stages of OC, including early stage. Further research exploring the utility of GD2 and GD3 for diagnosis of OC is warranted.
RESUMO
INTRODUCTION: The PrismRA® test identifies rheumatoid arthritis (RA) patients who are unlikely to respond to anti-tumor necrosis factor (anti-TNF) therapies. This study evaluated the clinical and financial outcomes of incorporating PrismRA into routine clinical care of RA patients. METHODS: A decision-analytic model was created to evaluate clinical and economic outcomes in the 12-month period following first biologic treatment. Two treatment strategies were compared: (1) observed clinical decision-making based on a 175-patient cohort receiving an anti-TNF therapy as their first biologic after failure of conventional synthetic disease-modifying antirheumatic drugs (csDMARDs) and (2) modeled clinical decision-making of the same population using PrismRA results to inform first-line biologic treatment choice. Modeled costs include biologic drug pharmacy, non-biologic pharmacy, and total medical costs. The odds of inadequate response to anti-TNF therapies and various components of patient care were calculated based on PrismRA results. RESULTS: Identifying predicted inadequate responders to anti-TNF therapies resulted in a modeled 38% increase in ACR50 response to first-line biologic therapies. The fraction of patients who achieved an ACR50 response to any therapy (TNFi and others) within the 12-month period was 33% higher in the PrismRA-stratified population than in the unstratified population (59 vs. 44%, respectively). When therapy prescriptions were modeled according to PrismRA results, cost savings were modeled for all financial variables: overall costs (4% decreased total, 19% decreased on ineffective treatments), total biologic drug pharmacy (4% total, 23% ineffective), non-biologic pharmacy (2% total, 19% ineffective), and medical costs (6% total, 19% ineffective). Female sex was the clinical metric that showed the greatest association with inadequate response to anti-TNF therapies (odds ratio 2.42, 95% confidence interval 1.20, 4.88). CONCLUSIONS: If PrismRA is implemented into routine clinical care as modeled, predicting which RA patients will have an inadequate response to anti-TNF therapies could save > $7 million in overall ineffective healthcare costs per 1000 patients tested and increase targeted DMARD response rates in RA.
