Integrating Artificial Intelligence for Advancing Multiple-Cancer Early Detection via Serum Biomarkers: A Narrative Review.

The concept and policies of multicancer early detection (MCED) have gained significant attention from governments worldwide in recent years. In the era of burgeoning artificial intelligence (AI) technology, the integration of MCED with AI has become a prevailing trend, giving rise to a plethora of MCED AI products. However, due to the heterogeneity of both the detection targets and the AI technologies, the overall diversity of MCED AI products remains considerable. The types of detection targets encompass protein biomarkers, cell-free DNA, or combinations of these biomarkers. In the development of AI models, different model training approaches are employed, including datasets of case-control studies or real-world cancer screening datasets. Various validation techniques, such as cross-validation, location-wise validation, and time-wise validation, are used. All of the factors show significant impacts on the predictive efficacy of MCED AIs. After the completion of AI model development, deploying the MCED AIs in clinical practice presents numerous challenges, including presenting the predictive reports, identifying the potential locations and types of tumors, and addressing cancer-related information, such as clinical follow-up and treatment. This study reviews several mature MCED AI products currently available in the market, detecting their composing factors from serum biomarker detection, MCED AI training/validation, and the clinical application. This review illuminates the challenges encountered by existing MCED AI products across these stages, offering insights into the continued development and obstacles within the field of MCED AI.

Long short-term memory model - A deep learning approach for medical data with irregularity in cancer predication with tumor markers.

BACKGROUND: Machine learning (ML) has emerged as a superior method for the analysis of large datasets. Application of ML is often hindered by incompleteness of the data which is particularly evident when approaching disease screening data due to varied testing regimens across medical institutions. Here we explored the utility of multiple ML algorithms to predict cancer risk when trained using a large but incomplete real-world dataset of tumor marker (TM) values. METHODS: TM screening data were collected from a large asymptomatic cohort (n = 163,174) at two independent medical centers. The cohort included 785 individuals who were subsequently diagnosed with cancer. Data included levels of up to eight TMs, but for most subjects, only a subset of the biomarkers were tested. In some instances, TM values were available at multiple time points, but intervals between tests varied widely. The data were used to train and test various machine learning models to evaluate their robustness for predicting cancer risk. Multiple methods for data imputation were explored and models were developed for both single time-point as well as time-series data. RESULTS: The ML algorithm, long short-term memory (LSTM), demonstrated superiority over other models for dealing with irregular medical data. A cancer risk prediction tool was trained and validated for a single time-point test of a TM panel including up to four biomarkers (AUROC = 0.831, 95% CI: 0.827-0.835) which outperformed a single threshold method using the same biomarkers. A second model relying on time series data of up to four time-points for 5 TMs had an AUROC of 0.931. CONCLUSIONS: A cancer risk prediction tool was developed by training a LSTM model using a large but incomplete real-world dataset of TM values. The LSTM model was best able to handle irregular data compared to other ML models. The use of time-series TM data can further improve the predictive performance of LSTM models even when the intervals between tests vary widely. These risk prediction tools are useful to direct subjects to further screening sooner, resulting in earlier detection of occult tumors.

Aspartate ß-Hydroxylase (ASPH) Expression in Acute Myeloid Leukemia: A Potential Novel Therapeutic Target.

BACKGROUND: Aspartate ß-hydroxylase (ASPH) is an embryonic transmembrane protein aberrantly upregulated in cancer cells, associated with malignant transformation and, in some reports, with poor clinical prognosis. OBJECTIVE: To report the expression patterns of ASPH in acute myeloid leukemia (AML). METHODS: Cell surface expression of ASPH was measured via 8-color multiparameter flow cytometry in 41 AML patient samples (31 bone marrow, 10 blood) using fluorescein isothiocyanate (FITC)-conjugated anti-ASPH antibody, SNS-622. A mean fluorescent intensity (MFI) of 10 was used as a cutoff for ASPH surface expression positivity. Data regarding patient and disease characteristics were collected. RESULTS: ASPH surface expression was found on AML blasts in 16 samples (39%). Higher ASPH expression was seen in myeloblasts of African American patients (p=0.02), but no correlation was found between ASPH expression and other patient or disease characteristics. No association was found between ASPH status and CR rate (p=0.53), EFS (p=0.87), or OS (p=0.17). CONCLUSIONS: ASPH is expressed on blasts in approximately 40% of AML cases, and may serve as a new therapeutically targetable leukemia-associated antigen.

Improving Multi-Tumor Biomarker Health Check-up Tests with Machine Learning Algorithms.

BACKGROUND: Tumor markers are used to screen tens of millions of individuals worldwide at annual health check-ups, especially in East Asia. Machine learning (ML)-based algorithms that improve the diagnostic accuracy and clinical utility of these tests can have substantial impact leading to the early diagnosis of cancer. METHODS: ML-based algorithms, including a cancer screening algorithm and a secondary organ of origin algorithm, were developed and validated using a large real world dataset (RWD) from asymptomatic individuals undergoing routine cancer screening at a Taiwanese medical center between May 2001 and April 2015. External validation was performed using data from the same period from a separate medical center. The data set included tumor marker values, age, and gender from 27,938 individuals, including 342 subsequently confirmed cancer cases. RESULTS: Separate gender-specific cancer screening algorithms were developed. For men, a logistic regression-based algorithm outperformed single-marker and other ML-based algorithms, with a mean area under the receiver operating characteristic curve (AUROC) of 0.7654 in internal and 0.8736 in external cross validation. For women, a random forest-based algorithm attained a mean AUROC of 0.6665 in internal and 0.6938 in external cross validation. The median time to cancer diagnosis (TTD) in men was 451.5, 204.5, and 28 days for the mild, moderate, and high-risk groups, respectively; for women, the median TTD was 229, 132, and 125 days for the mild, moderate, and high-risk groups. A second algorithm was developed to predict the most likely affected organ systems for at-risk individuals. The algorithm yielded 0.8120 sensitivity and 0.6490 specificity for men, and 0.8170 sensitivity and 0.6750 specificity for women. CONCLUSIONS: ML-derived algorithms, trained and validated by using a RWD, can significantly improve tumor marker-based screening for multiple types of early stage cancers, suggest the tissue of origin, and provide guidance for patient follow-up.

Anti-tumor activity of antibody drug conjugate targeting aspartate-ß-hydroxylase in pancreatic ductal adenocarcinoma.

Pancreatic ductal adenocarcinoma (PDAC) is an extremely aggressive malignancy with very limited treatment options. Antibody drug conjugates (ADCs) are promising cytotoxic agents capable of highly selective delivery. Aspartate-ß-hydroxylase (ASPH) is a type II transmembrane protein highly expressed in PDACs (97.1%) but not normal pancreas. We investigated anti-tumor effects of an ADC guided by a human monoclonal antibody (SNS-622) against ASPH in human PDAC cell lines and derived subcutaneous (s.c.) xenograft as well as a patient-derived xenograft (PDX) murine model with spontaneous pulmonary metastasis. The cytotoxic effects exhibited by several candidate payloads linked to SNS-622 antibody targeting ASPH+ PDACs were analyzed. After i.v. administration of SNS-622-emtansine (DM1) ADC, the primary PDAC tumor growth and progression (number and size of pulmonary metastases) were determined. The PDAC cell lines, s.c. and PDX tumors treated with ADC were tested for cell proliferation, cytotoxicity and apoptosis by MTS and immunohistochemistry (IHC) assays. SNS-622-DM1 construct has demonstrated optimal anti-tumor effects in vitro. In the PDX model of human PDAC, SNS-622-DM1 ADC exerted substantially inhibitory effects on tumor growth and pulmonary metastasis through attenuating proliferation and promoting apoptosis.

Lambda phage-based vaccine induces antitumor immunity in hepatocellular carcinoma.

BACKGROUND AND AIMS: Hepatocellular carcinoma (HCC) is a difficult to treat tumor with a poor prognosis. Aspartate ß-hydroxylase (ASPH) is a highly conserved enzyme overexpressed on the cell surface of both murine and human HCC cells. METHODS: We evaluated therapeutic effects of nanoparticle lambda (λ) phage vaccine constructs against ASPH expressing murine liver tumors. Mice were immunized before and after subcutaneous implantation of a syngeneic BNL HCC cell line. Antitumor actively was assessed by generation of antigen specific cellular immune responses and the identification of tumor infiltrating lymphocytes. RESULTS: Prophylactic and therapeutic immunization significantly delayed HCC growth and progression. ASPH-antigen specific CD4+ and CD8+ lymphocytes were identified in the spleen of tumor bearing mice and cytotoxicity was directed against ASPH expressing BNL HCC cells. Furthermore, vaccination generated antigen specific Th1 and Th2 cytokine secretion by immune cells. There was widespread necrosis with infiltration of CD3+ and CD8+ T cells in HCC tumors of λ phage vaccinated mice compared to controls. Moreover, further confirmation of anti-tumor effects on ASPH expressing tumor cell growth were obtained in another murine syngeneic vaccine model with pulmonary metastases. CONCLUSIONS: These observations suggest that ASPH may serve as a highly antigenic target for immunotherapy.

A Radiolabeled Fully Human Antibody to Human Aspartyl (Asparaginyl) ß-Hydroxylase Is a Promising Agent for Imaging and Therapy of Metastatic Breast Cancer.

There is a need for novel effective and safe therapies for metastatic breast cancer based on targeting tumor-specific molecular markers of cancer. Human aspartyl (asparaginyl) ß-hydroxylase (HAAH) is a highly conserved enzyme that hydroxylates epidermal growth factor-like domains in transformation-associated proteins and is overexpressed in a variety of cancers, including breast cancer. A fully human monoclonal antibody (mAb) PAN-622 has been developed to HAAH. In this study, they describe the development of PAN-622 mAb as an agent for imaging and radioimmunotherapy of metastatic breast cancer. PAN-622 was conjugated to several ligands such as DOTA, CHXA″, and DTPA to enable subsequent radiolabeling and its immunoreactivity was evaluated by an HAAH-specific enzyme-linked immunosorbent assay and binding to the HAAH-positive cells. As a result, DTPA-PAN-622 was chosen to investigate biodistribution in healthy CD-1 female mice and 4T1 mammary tumor-bearing BALB/c mice. The 111In-DTPA-pan622 mAb concentrated in the primary tumors and to some degree in lung metastases as shown by SPECT/CT and Cherenkov imaging. A pilot therapy study with 213Bi-DTPA-PAN-622 demonstrated a significant effect on the primary tumor. The authors concluded that human mAb PAN-622 to HAAH is a promising reagent for development of imaging and possible therapeutic agents for the treatment of metastatic breast cancer.

PAN-811 inhibits oxidative stress-induced cell death of human Alzheimer's disease-derived and age-matched olfactory neuroepithelial cells via suppression of intracellular reactive oxygen species.

Oxidative stress plays a significant role in neurotoxicity associated with a variety of neurodegenerative diseases including Alzheimer's disease (AD). Increased oxidative stress has been shown to be a prominent and early feature of vulnerable neurons in AD. Olfactory neuroepithelial cells are affected at an early stage. Exposure to oxidative stress induces the accumulation of intracellular reactive oxygen species (ROS), which in turn causes cell damage in the form of protein, lipid, and DNA oxidations. Elevated ROS levels are also associated with increased deposition of amyloid-beta and formation of senile plaques, a hallmark of the AD brain. If enhanced ROS exceeds the basal level of cellular protective mechanisms, oxidative damage and cell death will result. Therefore, substances that can reduce oxidative stress are sought as potential drug candidates for treatment or preventative therapy of neurodegenerative diseases such as AD. PAN-811, also known as 3-aminopyridine-2-carboxaldehyde thiosemicarbazone or Triapine, is a small lipophilic compound that is currently being investigated in several Phase II clinical trials for cancer therapy due to its inhibition of ribonucleotide reductase activity. Here we show PAN-811 to be effective in preventing or reducing ROS accumulation and the resulting oxidative damages in both AD-derived and age-matched olfactory neuroepithelial cells.

Neuroprotective activity of 3-aminopyridine-2-carboxaldehyde thiosemicarbazone (PAN-811), a cancer therapeutic agent.

3-aminopyridine-2-carboxaldehyde thiosemicarbazone (3-AP) is a highly-hydrophobic small molecule that was originally developed for cancer therapy (Triapine, Vion Pharmaceuticals) due to its ability to inhibit ribonucleotide reductase, a key enzyme required for DNA synthesis. 3-AP has a high affinity for divalent cations, chelating the Fe(2+) at the R2 subunit of the enzyme and inhibiting formation of a tyrosyl radical essential for ribonucleotide reduction. We have demonstrated that 3-AP is also a potent neuroprotectant (as such, it is referred to as "PAN-811"). In vitro it completely blocks ischemic neurotoxicity at a concentration of 0.5 microM (EC(50) approximate, equals 0.35 microM) and hypoxic toxicity at 1.2 microM (EC(50) approximate, equals 0.75 microM). Full protection of primary cortical and striatal neurons can be achieved with 3-AP when it is added to the medium at up to six hours after an ischemic insult. 3-AP also suppresses cell death induced by neurotoxic agents, including staurosporine, veratridine and glutamate, indicating activity against a central target(s) in the neurodegenerative process. 3-AP acts via neutralization of two important intracellular effectors of excitatory neurotoxicity; calcium and free radicals. Its reported ability to elevate anti-apoptotic proteins is likely to be a consequence of the suppression of excessive intracellular free calcium. In a rat model of transient ischemia, a single bolus delivery of 3-AP 1 h after the initiation of ischemic attack reduced infarct volume by 59% when administered i.c.v. (50 mug per rat) and by 35% when administered i.v. (1 mg/kg). In Phase I clinical trials in cancer therapy 3-AP had no cardiovascular, CNS or other major adverse effects. Thus, 3-AP has a high potential for development as a novel, potent neuroprotectant for the treatment of neurodegenerative diseases.

A multifunctional cytoprotective agent that reduces neurodegeneration after ischemia.

Cellular and molecular pathways underlying ischemic neurotoxicity are multifaceted and complex. Although many potentially neuroprotective agents have been investigated, the simplicity of their protective mechanisms has often resulted in insufficient clinical utility. We describe a previously uncharacterized class of potent neuroprotective compounds, represented by PAN-811, that effectively block both ischemic and hypoxic neurotoxicity. PAN-811 disrupts neurotoxic pathways by at least two modes of action. It causes a reduction of intracellular-free calcium as well as free radical scavenging resulting in a significant decrease in necrotic and apoptotic cell death. In a rat model of ischemic stroke, administration of PAN-811 i.c.v. 1 h after middle cerebral artery occlusion resulted in a 59% reduction in the volume of infarction. Human trials of PAN-811 for an unrelated indication have established a favorable safety and pharmacodynamic profile within the dose range required for neuroprotection warranting its clinical trial as a neuroprotective drug.