• Login
  • |
  • Register
RSS Feed Print
Genomics in Clinical Trials
Janine Holley
Posted: Thursday, September 29, 2011 10:30 AM
Joined: 11/16/2010
Posts: 7

We had a great panel discussion in Providence earlier this week (if you missed the webcast, the you can listen to the 1-hour recording here). We didn't have time to get to all the audience questions, so I've listed a few of them below.  


  1. How early in a clinical trial (phase 1, 2, 3 etc) should one incorporate genomic markers? 
  2. How would pharma handle DNA variants of unknown significance in clinical trials?   
  3. What sort of database would clinical researchers use to store the genotype and phenotype data of subjects?    


I went back to the panelists to get their thoughts on these questions, and I'd also love to hear what members think. Please add a comment...

Janine Holley
Posted: Friday, September 30, 2011 8:58 AM
Joined: 11/16/2010
Posts: 7

I checked in with Richard Resnick (CEO, GenomeQuest), who was one of our panelists for this session. Below is his response to the question of how pharma would handle DNA variants of unknown significance...


Quote from Richard Resnick:  

  1. Make sure you have a small set of genomic biomarkers early (like, fewer than 20), and use them to power the trial 

  2. Use whole genome or exome sequencing to measure those 20 biomarkers, rather than whatever the traditional tech would be 

  3. You'll find the presence or absence of these genomic biomarkers in each patient sample. Cool. You'll also find about 3m other variants. 

  4. Consider those 3m other variants the so-called "variations of unknown significance (VUS)" from the perspective of the trial, and save them. For later. 

  5. Make sure you get informed consent that allows you to consider these data points in the future. 

  6. Get your drug approved  

  7. Take the whole genome data that you have, and retrospectively combine this with the rich medical make-ups you have taken in on these patients through this trial. 

  8. Discover. Find new biomarkers. Find new mechanisms for disease. Cure cancer. 


The point is, if you are going to be including any biomarkers at all in your trial design, make them genomic and use sequencing - for almost the same price you get everything you wanted around biomarker categorization, and you get a thousand genome project of your own that is rich in genotype/phenotype interactions which will fuel your pipeline for decades. 


Like I said during the panel, it's a great time to start a pharma company.  




Janine Holley
Posted: Friday, September 30, 2011 12:25 PM
Joined: 11/16/2010
Posts: 7

Bradley Smith (Vice President Translational Medicine, Quintiles) was also a panelist. Brad had the following thoughts on Question #1: How early in a clinical trial should one incorporate genomic markers?


Quote from Brad Smith: 

The use of genomic markers in clinical trials must accomplish a number of goals:

1) Clinically validate the predictive value of the genomic marker

2) Establish the laboratory use of the test (technical validation)

3) Discover new markers that may be informative of patient response

Genomic markers must be used as early as possible to accomplish these goals. The opinion of FDA leadership recommends that tests be locked down going into late phase. Therefore, technical and clinical validation must occur in early phase trials. In fact, the latest FDA guidance on co-development of drugs and companion diagnostics recommends that diagnostic plans be included in the product development plan. Therefore, markers and tests must be known and validated as early as possible. 

An adaptive trial design presents an attractive model that takes advantage of the scope of a nex-gen sequencing analysis. In this model, new predictive markers, identified in the course of the analysis, may be enriched for and validated. This approach will require that the trial design address the regulatory questions that may arise. Other considerations that should be addressed include varied demographic and genetic diversity in the trial populations including the diversity in the control groups. Another consideration that will enable the identification and validation of novel genomic markers is the development and use of an integrated database allowing for visibility and early decision making based upon patient molecular profiles and therapeutic efficacy. 

All of these steps will add value to the genomic analysis and permit it to contribute to the success of the drug development program.


Please login to post a new topic or reply to an existing post.
Privacy Policy|Terms of Use