Skip to Navigation
University of Pittsburgh
Print This Page Print this pages

September 26, 2002

SCIENCE 2002: Responding to the threat of bioterrorism

Prompt treatment with antibiotics saved the lives of most victims of last year's anthrax attacks. Fortunately, the anthrax involved was not drug-resistant.

"That was the good news. But it's drug-resistant strains of anthrax that we have to worry about," Graham Hatfull said during a Science 2002 session on "Responding to the Threat of Bioterrorism."

Hatfull, who is the Eberly Family Professor in the biological sciences department, pointed out: "The time that it takes to figure out how to deal with the anthrax that appears in bioterror is going to be related to how many people ultimately suffer and die in an attack. We need to know how to treat it early."

To help make that possible, Hatfull and his colleagues are seeking to apply a new methodology involving bacteriophages, viruses that infect bacteria.

The number of bacteriophages is estimated at 1031, he said. "There are more bacteriophages than all other forms of life put together. They've probably been around for 3 billion years, and they've gotten quite good at what they do" — which is, infecting host bacteria in order to replicate themselves.

"And they don't just infect any old bacteria. Their range of hosts is highly specific," Hatfull noted. "Phages" often carry not only the genes they need to reproduce themselves, but additional proteins and toxins as well.

Hatfull and his research team have collected samples of 15 phages known to infect bacillus anthracis or related bacterial strains, and are studying their genetic sequences. The ultimate goal is to alter the phages so that they will carry genetic material inside anthrax and other deadly bacteria.

In the future, scientists might be able to use these genetically altered phages to confirm precisely which bacteria they were dealing with, Hatfull explained.

For example, scientists might create special phages that would carry luciferous genes (which emit luminous photons) into the specific, deadly bacteria that they are known to infect. If scientists subsequently detected luminescence, they would know which bacteria had infected patients.

Hatfull emphasized that such methodologies shouldn't be limited to fighting bioterrorism. They would be equally effective against naturally occurring infectious diseases such as tuberculosis, which kills 3 million people around the world each year, he said.

q Pitt professor Gregory F. Cooper and Carnegie Mellon's Andrew W. Moore described the Pittsburgh-based Real-Time Outbreak Disease and Surveillance (RODS) system, which uses information technology to monitor hospital emergency rooms and other facilities that could identify bioterrorism or naturally occurring disease outbreaks.

The RODS lab is sponsored by Pitt's Center for Biomedical Informatics and Center for Public Health Practice. Pitt recently received a $6 million grant from the Pennsylvania Department of Health for RODS-related research. The system was part of the security program for the 2002 Winter Olympics in Salt Lake City.

Some 44 percent of western Pennsylvania hospitals are participating in RODS and more are expected to join, Moore said. The system uses what he called "a piece of software with a horrible name": WSARE, an acronym for What's strange about recent events?

WSARE alerts biosurveillance researchers to unusual patterns in emergency room admissions, for example. "The question we then have to answer is: Is this sufficiently strange that it's worth us going out and actively looking at some of these cases of individuals who are coming down with a particular disease?" Moore said.

Cooper said RODS researchers are working on a system called PANDA (patient and population-based anomaly detection and assessment). "The goal is to be able to mine data about the population as a whole and compare it with data about individual patients, in order to arrive at an overall assessment of the risks of an unanticipated disease," Cooper said.

The advantage of biosurveillance systems is that they create a higher barrier against bioterrorism attacks, he and Moore said.

The threat is that bioterrorists will begin employing disease agents that don't cause demographic or occupational clusters that current biosurveillance systems can detect, they said.

q A lot of useful biosurveillance data is in "free text" form that can't readily be used by computers, explained Wendy Chapman, a postdoctoral fellow in the medical school's Department of Medicine.

She and other researchers are developing and refining methods for encoding free text information, the simplest method being the manual keyword system that the International Olympics Committee used in Salt Lake City.

The IOC biosurveillance team performed keyword searches on "chief complaint" reports filed by emergency room staff during the Olympics. These brief reports document the major health complaints that ER patients describe upon being admitted.

IOC scientists classified symptoms into eight health syndromes including respiratory and gastrointestinal illnesses. "If a keyword search found that the word 'cough' was frequently turning up in chief complaints, that would alert [scientists] to the possibility of a suspicious outbreak of respiratory disease," Chapman explained.

Researchers also are working on a system that would search free-text reports of chest X-ray results for signs of anthrax infections, she said.

"There's a lot of free-text data out there, and trying to measure the usefulness of it is going to take time," Chapman said.

— Bruce Steele

Filed under: Feature,Volume 35 Issue 3

Leave a Reply