0% found this document useful (0 votes)
4 views4 pages

Bayesian AI Model for HDGC Risk Assessment

Artificial Intelligence: A Bayesian Probability/Calculus-Based Algorithm for Hereditary Diffuse Gastric Cancer and Pharmacological Options Authors: Dale Srinivas*, Arvind Kumar Gupta, Anu Singh, Sarika Gupta Affiliations: Coventry & Warwickshire Partnership NHS Trust, UK; Twin Island University, Guyana Date: 26 September 2025 Abstract Hereditary Diffuse Gastric Cancer (HDGC) is a rare autosomal dominant cancer predisposition syndrome associated primarily with pathogenic variants in the CDH1 gene

Uploaded by

Dale Srinivas
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
4 views4 pages

Bayesian AI Model for HDGC Risk Assessment

Artificial Intelligence: A Bayesian Probability/Calculus-Based Algorithm for Hereditary Diffuse Gastric Cancer and Pharmacological Options Authors: Dale Srinivas*, Arvind Kumar Gupta, Anu Singh, Sarika Gupta Affiliations: Coventry & Warwickshire Partnership NHS Trust, UK; Twin Island University, Guyana Date: 26 September 2025 Abstract Hereditary Diffuse Gastric Cancer (HDGC) is a rare autosomal dominant cancer predisposition syndrome associated primarily with pathogenic variants in the CDH1 gene

Uploaded by

Dale Srinivas
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd

Artificial Intelligence: A Bayesian

Probability/Calculus-Based Algorithm
for Hereditary Diffuse Gastric Cancer
and Pharmacological Options
Authors: Dale Srinivas*, Arvind Kumar Gupta, Anu Singh, Sarika Gupta

Affiliations: Coventry & Warwickshire Partnership NHS Trust, UK; Twin Island University,
Guyana

Date: 26 September 2025

Abstract
Hereditary Diffuse Gastric Cancer (HDGC) is a rare autosomal dominant cancer
predisposition syndrome associated primarily with pathogenic variants in the CDH1 gene,
and less commonly CTNNA1. HDGC is characterised by early-onset diffuse gastric cancer
and increased risk of lobular breast carcinoma. We propose a Bayesian
probability/calculus-based AI model that integrates clinical history, genetic testing,
histopathology, and surveillance data to refine diagnostic probability, predict cancer risk,
and optimise preventive and therapeutic interventions.

The Bayesian inference formula is expressed as:

P(HDGC | Evidence) = [P(Evidence | HDGC) * P(HDGC)] / P(Evidence)

An extended malignancy risk model is expressed as:


P(Cancer | Evidence, HDGC) = Σ P(Cancer | Eᵢ, HDGC) * P(Eᵢ | HDGC)

Background
HDGC is associated with a lifetime risk of diffuse gastric cancer of up to 70% in men and
56% in women, and a risk of lobular breast cancer of up to 42% in women with CDH1
mutations. Diagnosis is based on a combination of family history, clinical suspicion, genetic
testing, and histopathology demonstrating signet ring cell carcinoma. Prophylactic
gastrectomy is recommended in confirmed mutation carriers due to the difficulty of early
detection by endoscopy. A Bayesian model can dynamically integrate these heterogeneous
data sources to refine risk assessment and guide management.
Bayesian AI Framework
Clinical features and family history establish prior probability, genetic testing confirms
mutation status, histopathology provides strong diagnostic evidence, and surveillance
findings refine probability estimates over time. Posterior probabilities inform decision-
making regarding prophylactic gastrectomy, surveillance intervals, and systemic therapies.

Table 1: Evidence Integration in Bayesian HDGC Model


Evidence Type Key Features Contribution to Posterior
Probability

Clinical features Family history, early-onset Establishes prior


gastric cancer, lobular probability
breast cancer

Genetic testing CDH1/CTNNA1 mutations Confirms diagnosis

Histopathology Diffuse-type signet ring cell Strong diagnostic evidence


carcinoma

Surveillance Endoscopy, imaging Updates recurrence/cancer


risk

Pharmacological and Therapeutic Options


- Surgical: Prophylactic total gastrectomy for mutation carriers.
- Pharmacological: Chemotherapy (platinum-based, fluoropyrimidines) for advanced cases.
- Targeted therapy: HER2 inhibitors in selected cases, PARP inhibitors under investigation.
- Supportive: Nutritional support, breast surveillance for lobular breast cancer risk.
- Experimental: Gene editing, immunotherapy trials.

Table 2: Therapeutic Options in HDGC


Category Interventions Comments

Surgical Prophylactic gastrectomy Mainstay, prevents gastric


cancer development

Pharmacological Chemotherapy (platinum, For advanced/metastatic


fluoropyrimidines) disease

Targeted therapy HER2 inhibitors, PARP Molecular targeted


inhibitors approaches
Supportive Nutritional support, breast Addresses systemic risks
surveillance

Experimental Gene editing, Future personalised


immunotherapy medicine strategies

Conclusion
Hereditary Diffuse Gastric Cancer is a highly penetrant cancer predisposition syndrome
with poor prognosis if undetected. A Bayesian AI framework provides a powerful method
for integrating family history, genetic, clinical, and surveillance evidence, refining risk
estimates, and guiding precision preventive and therapeutic interventions. This approach
exemplifies the integration of artificial intelligence with clinical genetics and oncology.

References
Guilford, P., et al. (1998). E-cadherin germline mutations in familial gastric cancer. Nature,
392(6674), 402–405.

Hansford, S., et al. (2015). Hereditary diffuse gastric cancer: Updated clinical practice
guidelines. Lancet Oncol, 16(1), e60–e70.

van der Post, R.S., et al. (2015). Hereditary diffuse gastric cancer: Updated clinical
guidelines. J Med Genet, 52(6), 361–374.

Oliveira, C., et al. (2019). E-cadherin alterations in hereditary diffuse gastric cancer. Hum
Mutat, 40(9), 1421–1432.

Bayes, T. (1763). An Essay towards Solving a Problem in the Doctrine of Chances. Phil Trans
R Soc Lond, 53, 370–418.

Pearl, J. (2009). Causality: Models, Reasoning, and Inference. Cambridge University Press.
Figure

You might also like