Bayesian approach to interpreting somatic cancer sequencing data: a case in point

Area of Science:

• Oncology
• Bioinformatics
• Genomic Medicine

Background:

• Accurate cancer lineage and tissue-of-origin assignment is critical for effective, disease-specific therapeutic strategies.
• Cancers of unknown primary present a diagnostic challenge, impacting patient treatment and outcomes.
• Contemporary treatments are often tailored to specific cancer types, highlighting the need for precise origin identification.

Purpose of the Study:

• To demonstrate the utility of Bayesian analysis in merging diverse data types for cancer tissue-of-origin assignment.
• To provide a practical example of applying iterative Bayesian analysis to a challenging clinical case.
• To improve diagnostic accuracy for cancers of unknown primary.

Main Methods:

• Iterative Bayesian analysis integrating clinical history, histology, and immunohistochemistry (IHC) data.
• Incorporation of cancer DNA sequencing data into the Bayesian framework.
• Calculation of odds ratios (OR) between differential diagnoses using Bayesian principles.

Main Results:

• Bayesian analysis successfully merged multimodal data (clinical, IHC, sequencing) for tissue-of-origin assessment.
• The iterative Bayesian approach provided meaningful assistance in a clinical case differentiating primary lung cancer from metastatic bladder cancer.
• Demonstrated the feasibility of using Bayesian methods to enhance diagnostic confidence in complex cases.

Conclusions:

• Bayesian analysis offers a robust framework for integrating heterogeneous data in oncology diagnostics.
• This approach can significantly aid in assigning the tissue of origin for cancers of unknown primary.
• The methodology holds promise for improving patient management by enabling more precise, data-driven treatment decisions.