We use both our own and third-party cookies for statistical purposes and to improve our services. If you continue to browse, we consider that you accept the use of these.

Multidimensional Biomarkers & Machine Learning Based Approaches for Precision Medicine

 In Advanced Applications, Cofactor Genomics, Molecular Diagnostics, Q&A

Multidimensional Biomarkers & Machine Learning Based Approaches for Precision Medicine

By Dave Messina

As part of the Clinical Biomarkers & World CDx Conference held in Boston last month, I had the honor of moderating a panel on a fascinating and timely topic,  “Multidimensional Biomarkers & Machine Learning Based Approaches for Precision Medicine”.

And what a panel — attendees were treated to a remarkable group of leaders in our field:

  • Gabriel Bien-Willner, Chief Medical Officer, Palmetto GBA
  • Bonnie Anderson, Chairman & CEO, Veracyte
  • Wendell Jones, Principal Bioinformaticist & Scientific Advisor, Q2 Solutions
  • Yuri Fesko, Chief Clinical Officer, Oncology Strategic Collaborations & Medical Director, Oncology, Quest Diagnostics
  • Douglas Adkins, Professor, Department of Medicine Oncology Division, Medical Oncology, Washington University School of Medicine in St. Louis

The conversation was lively and insightful, as the panelists discussed the utility of multidimensional biomarkers in the translational and clinical space, detailing the enhanced role of multiplexing in early biomarker selection and the critical need to employ the power of multianalyte diagnostics in a clinical setting. 

“Clinicians are hungry to have predictive biomarkers for their treatments. And when they’re available, we utilize them greatly and depend upon them significantly,” explained Dr. Douglas Adkins. “But we realize that with any one biomarker there’s a lot of error bar in its predictive abilities. So the need to integrate multiple biomarkers makes good sense, to help make a better world for a clinician to select a therapy and give patients confidence that what we’re going to give them actually has a good chance of working.”

Collectively, this group represented a range of stakeholders, from that of  diagnostic developer to payer to global CROs serving pharma to practicing oncologist. Despite these seemingly different viewpoints, there was strong consensus that multidimensional biomarker approaches, including those using machine learning, are essential to distill the complex biological signals of disease into an actionable diagnostic result.

“I would love to see the diagnostic field advance to where we take advantage of the comprehensiveness of the data and be able to inform on things that are important for treatment decisions at the same time as the diagnosis,” said Dr. Gabriel Bien-Willner.

In the wide-ranging session, panel members shared early successes, a commitment to well-validated studies, and a recognition of the impact multidimensional biomarkers are already having on patient care. Watch the entire panel discussion below.

 

Recommended Posts

Start typing and press Enter to search

Steve Daniel