Skip to main content
Premium Trial:

Request an Annual Quote

Perren Cobb on a Nationwide Effort to Bring Microarrays to the ICU

Premium
J. Perren Cobb
Associate Professor; Burn, Trauma, Surgical Critical
Care Section, Division of General Surgery, Department of Surgery
Washington University School
of Medicine

At A Glance:

Name: J. Perren Cobb

Background: 2000 — present: Associate Professor; Burn, Trauma, Surgical Critical Care Section, Division of General Surgery, Department of Surgery, Washington University School of Medicine.

1996 — present: Director, Cellular Injury and Adaptation Laboratory, Washington University School of Medicine.

1995 — 1999: Assistant Professor; Burn, Trauma, Surgical Critical Care Section, Division of General Surgery, Department of Surgery, Washington University School of Medicine.

Education and Training: 1986 — 1994: Resident, Division of General Surgery, Department of Surgery, University of California-San Francisco

1986: MD, University of Louisville School of Medicine

1982: BA, Vanderbilt University


A study concerning the development of protocols for collecting, sharing, and analyzing genomic data from trauma patients was published in the March 29 issue of the Proceedings of the National Academy of Sciences by the Inflammation and Host Response to Injury Large Scale Collaboration Program, a national collaboration of over 70 physicians and scientists.

Funded through a $37 million grant from the National Institutes of Health, the study is the result of more than five years of work from the group, which uses genomic analysis to develop new therapies for critically injured patients. It may also lead the way for establishing biomarkers that could lead to in vitro diagnostics for the 5 million Americans that are admitted to intensive care units annually. Thirdly, the publication marked the emergence of national research network for developing therapies for trauma patients — which other countries like Canada and Germany have, but the US still lacks. To learn more about this study and its potential, BioArray News spoke last week with J. Perren Cobb, the lead author of the article published in PNAS, and an associate professor of surgery and of genetics at Washington University, where part of the study was conducted.

What prompted you to begin this project?

The background is that there are 5 million people admitted to intensive care units every year in the US. People are getting sick, whether it's cancer, or AIDS, or infection, or trauma, trauma being the focus of the Inflammation and Host Response to Injury Program, people end up in the unit — and despite the incredibly sophisticated technology, we are actually pretty poor at guessing who is going to get well and who is not going to get well or who is going to respond to a given course of therapy and who's not.

What we'd like to be able to do is to tell [patients' loved ones] exactly what's going on, We have monitored their progress with these very sensitive biomarkers and we know that they're on the right track or they're not on the right track and this is the specific drug that we're giving them that affects this specific gene to help them get better. So we can't do that right now. The hope is in studying the systemic response to inflammation. This inflammation causes a lot of the body's organs to fail, and if you get enough organs sick enough the person dies.

So how did that collaboration start?

There is no institute in the United States [devoted to this]. If you think about it, if you ask the question, 'Which federal agency is responsible for driving the research imperatives for the critically ill or injured?' There are no federal agencies, despite the fact that there's a huge number of people who are dying of this problem. The number one cause of death in Americans between the ages of 2 and 40 is trauma. It's the number 5 cause of death overall. So there are a lot of people dying of this and investigators have been collaborating for a long time trying to get a big chunk of money from the NIH to address these issues.

About five years ago the NIH doubled its budget for five years, and when money was more flush (this is prior to 9/11) the National Institutes of General Medical Sciences got together and said, 'There's this new revolutionary technology called genomics and there's a number of big problems that need to be tackled. And it's clear that these big problems for any lab or university. It's clear that we are going to have a big grant, a large-scale collaborative research program that we are going to fund at a level of $37 million over five years.'

There were a number of applications submitted and we submitted an application to cover patients, in particular, trauma patients. There were ultimately five of these glue grants funded $125 million worth of grants. And it so happens that we are 70 private universities across 20 or so medical centers, and we are focused on understanding the human response to systemic inflammation with injury as the inciting event.

What kinds of tools are currently state-of-the-art in assessing the health of critically-injured patients?

The standard tools are monitoring devices — the kind of devices that monitor heart rate and blood pressure and collecting blood samples, and monitor things like sodium and potassium, looking at typical proteins in the blood and measuring the concentration of white blood cells. There are tools we can use to monitor — for example, heart function, function of the kidneys. So there are tools that we've been using for decades to look at function and what these genomic tools allow us to do is for the first time ever, look genome-wide at differences in RNA abundance using gene chips.

Do you think your collaboration would spin out into a network similar to the ones they have in Germany or Canada?

Yeah, that's our hope. As I indicated, there's no federal agency that oversees or drives research imperative for that. We have been interested in helping to catalyze an intramural group at the NIH and there has been a meeting of a number of different institutes working in this direction. That's from an NIH, intramural standpoint. From an extramural standpoint, I'm working very hard with a number of different individuals at the NIH to create a venue where people who are interested in this can go at the same place and the same time to talk about things with regards to genomics and the large number of patients that are affected by this each year. That symposium is called Functional Genomics in Critical Injury, and we're having our third meeting in a little over a month on April 21-22 at the NIH. So we're convening this meeting to get everyone on the same page to figure out how to apply these technologies and how best to use them because they are so expensive.

How did you come to use microarrays in your study and what came out of that?

The hypothesis that we're testing is that we can collect samples longitudinally from patients that are really sick and that the genome-wide expression profiles of circulating white blood cells will be informational with regard to outcome. We are hoping that [by comparing the samples of] the patients who live versus the patients who die, we'll be able to generate microarray expression profiles and compare the relative differences in RNA abundance for those who live versus those who die. We'll be able to, number one, understand the host response to injury; and, number 2, hopefully identify some potential regulatory nodes in those interaction networks, and potential therapeutic targets that will allow us to develop novel therapies.

So what was the motivation for creating the protocol?

Nobody's ever tried to do this before where you longitudinally follow patients over several days, collect samples, and try to make sense out of the information. We wanted to make sure that if we got a sample from say University of Washington and a sample from Washington University, or wherever, you'd want to make sure that there wasn't so much noise in the system.

Since nobody's ever done this before we had to pick a platform that was robust and reproducible. We also had to develop protocols for isolating white blood cells and generating high-quality RNA from those samples. And this had to be true regardless of where the sample was collected — they were all sent to a single, central repository and then doled out for the appropriate sampling studies, so number one, we had to create the network; number two, we had to generate the patient samples; and, number three, we had to write the protocols and validate them to make sure the amount of variance from site to site was minimized, etc.

So we had this reproducible methodology system-wide that we could apply to technology and get some useful information out of it. This first paper in PNAS is the description and the discovery for all of these protocols and these methods -all of which will be available shortly through our website.

We want to share this with people so others that want to look at different conditions won't have to reinvent the wheel. This acts as a springboard. Now that we've validated our methods, we can with great confidence go into the clinic, into the ICU, and start collecting samples, and collect those data and analyze, and so far we've collected data from 100 patients from centers around the country.

How large a sample group will you need to feel you've completed this study?

What we'd like to be able to do is a power analysis before we start a study, which gives us an idea exactly how many subjects we'll need to study. But nobody's ever done anything like this before and it's hard for us to know. We are planning on accruing in the first five years of this glue grant samples from 250 patients. And we will easily meet that goal by the end of our program, which is a year from September.

And what platform are you working on?

Affymetrix.

Have they helped in the study, or with developing the protocols?

We have done this independently. Affymetrix are strong partners in the sense that they're obviously interested in what we're doing and the results that we have — but these protocols have been developed independent of the company.

Are you aware of any other companies that are working on an IVD specifically for trauma patients?

No, no, I am not. And we are very interested in identifying and working with our colleagues in industry.

I read that in addition to developing the protocol, you also did a genome-wide monitoring of neutrophil. How did that fit into this study?

Well this is one of those things where you need to crawl before you walk, before you run. The first thing we did is look at a sampling of a mixed cell types. There are several different types of white blood cells that are circulating in your blood. There are neutrophils, monocytes, et cetera. What we've done for this first study is look at all of them together. The protocol we use isolates the white blood cells from the red blood cells. It isolated the plasma. So what we've done is isolated the circulating white blood cells. We also believe that we can isolate particular subsets of white blood cells and it's a more difficult protocol. We developed a protocol for isolating all white blood cells and we've developed expression profiles for the mixture. We are interested in this point in looking at isolated white blood cells — looking at just neutrophils or just lymphocytes to learn a little bit more about the biology of the underlying response.

What kind of breakthrough would that be for people in your field and how would patients benefit in the long run?

There would be two. The principal motivation for us is to identify the biomarkers that will help clinicians identify who is responding and who's not responding to a given therapy. Another big piece of it is that these analytical tools we use will give us a list of genes that will help us identify one group from another but it doesn't tell us anything about what those individual genes do. So that's where you need the isolated cell studies so we can look at a particular cell type and say that we know in a lymphocyte that this particular gene goes up, this other gene goes down. And we can build the network on how those two may interact and do specific studies. The early studies for the first five years are focused on identifying the markers. The follow-up studies focus on the biology of why those markers are informational.

So when can we expect a result from this?

Since the human genome was mapped about five years ago people have been coming up and asking, 'Well, how come all diseases aren't cured?' And it's not so simple. Having the sequence doesn't tell you what the genes are or how they function. My guess, looking at my foggy crystal ball, is that this work will take decades, and I know that certainly those that are interested in intellectual property and drug companies and their investors don't want to hear that things are going to take decades — they want results in a number of months, rather than a matter of years. My hope is that we can get the biomarkers in the near future. Understanding the biology and tweaking individual drugs is going to be a lot more difficult. It's a lot easier said than done, but I think we're well on the way to realizing the potential in the biomarker field, and I have no doubt that the biological follow up will be to come. As far as when that will be? That's anybody's guess.

But I think there's a very high likelihood that we'll develop a diagnostic.

There's no doubt that this technology is going to pay off in the next 5-10 years. It's just a question of in what domain and how long it's going to take.

File Attachments