EP 54

Companion Diagnostic Biomarkers

This episode explores the world of companion diagnostic biomarkers, which are specialized tests that act as biological bouncers to prevent toxic disasters and match tumors to the right drugs. We delve into the high stakes of pharmacogenetics, where understanding an enzyme like DPYD prevents standard treatments from becoming lethal poisons. We also discuss how multi-gene panels spare patients from unnecessary chemotherapy and how AI is digesting the massive complexities of multi-omics to create hyper-efficient drug pipelines.

companion diagnosticspharmacogeneticsmulti-gene expression panelsmulti-omicsAImachine learning

play_arrowListen to episode

Key Concepts

Companion Diagnostics

Companion diagnostics are specialized tests that provide essential information for the safe and effective use of a specific therapeutic product.
They are defined by the FDA under rigid, uncompromising terms.
A companion diagnostic is hardwired into the drug's warning label and a physician cannot legally or medically prescribe the drug unless the test has been run first.

Pharmacogenetics

Pharmacogenetics is the study of how germline genetic variations influence how the body metabolizes a chemical.
A specific subset of the population carries a genetic variation that turns them into ultra-rapid metabolizers for certain enzymes.
This can lead to toxic disasters if not identified and addressed.

Multi-Gene Expression Panels

Multi-gene expression panels are used to look at a functional signature of a tumor, rather than a single biomarker.
They can spare patients from unnecessary chemotherapy and provide a more accurate prediction of recurrence.
Examples of multi-gene expression panels include the 21-gene panel and the 70-gene panel.

Multi-Omics

The integration of multi-omics data, including genomics, transcriptomics, proteomics, and metabolomics, provides unprecedented insights into tumor biology.
This approach can identify druggable vulnerabilities in tumors that were previously invisible.
Multi-omics data can be used to derive functional signatures and identify actionable diagnostic targets.

AI and Machine Learning

AI and machine learning algorithms are used to derive functional signatures from multi-omics data and identify actionable diagnostic targets.
These algorithms can ingest large amounts of data and identify complex, nonlinear relationships that would be difficult for humans to detect.
The use of AI and machine learning can accelerate the discovery of new companion diagnostics and improve patient outcomes.

Episode Summary

check_circleCompanion diagnostics are specialized tests that provide essential information for the safe and effective use of a specific therapeutic product.
check_circleThe FDA codifies the level of urgency for companion diagnostics through a process called pre-market approval (PMA).
check_circlePharmacogenetics is the study of how germline genetic variations influence how the body metabolizes a chemical.
check_circleA specific subset of the population carries a genetic variation that turns them into ultra-rapid metabolizers for certain enzymes.
check_circleCompanion diagnostics can prevent toxic disasters by identifying patients who are poor metabolizers of certain drugs.
check_circleMulti-gene expression panels are used to look at a functional signature of a tumor, rather than a single biomarker.
check_circleThe integration of multi-omics data, including genomics, transcriptomics, proteomics, and metabolomics, provides unprecedented insights into tumor biology.
check_circleAI and machine learning algorithms are used to derive functional signatures from multi-omics data and identify actionable diagnostic targets.
check_circleThe use of companion diagnostics can lead to a higher trial success rate and justify premium pricing for drugs.
check_circleThe future of medicine may involve the use of liquid biopsies and AI algorithms to provide a real-time, high-definition map of a patient's molecular state.

Full Transcript

Discussion

Join the discussion — sign in to leave a comment.