Feb 15, 2011 (CIDRAP News) After a review of scientific methods that the FBI used in probing the 2001 anthrax mailings, a committee of the National Research Council (NRC) announced today that the available scientific evidence by itself is not adequate to reach a definitive conclusion about the source of the anthrax spores used in the attacks.
The anthrax mailings to three media offices and two US senators in the fall of 2001 led to 22 anthrax cases, including 5 deaths. After 7 years of investigation, the FBI concluded in 2008 that government microbiologist Bruce Ivins had perpetrated the attacks. Ivins committed suicide in July 2008 as the FBI was preparing to file charges.
After the FBI announced its conclusions in August 2008, a number of experts expressed doubt about the findings and called for an independent review. In September 2008 FBI Director Robert Mueller announced that the bureau would ask the National Academy of Sciences (NAS) to conduct such a review.
The NRC, part of the NAS, released the 190-page report on its findings at a news conference today. The NRC committee focused on the biological, physical, and chemical methods, especially the "microbial forensics," used in studying and identifying the anthrax in the letters; it did not cover more traditional forensic areas such as analyses of hair, fibers, and fingerprints.
The FBI hung its case primarily on genetic evidence that it said linked the mailed anthrax to anthrax that was in Ivins' custody at the US Army Medical Research Institute of Infectious Diseases (USAMRIID) in Frederick, Md. The anthrax spores were stored in a flask called RMR-1029.
The NRC committee's key finding was that the link between the attack anthrax and the anthrax in Ivins' custody was not definitive.
"Spores in the mailed letters and in RMR-1029 . . . share a number of genetic similarities consistent with the FBI finding that the spores in the letters were derived from RMR-1029," the NAS said in a press release. "However, the committee found that other possible explanations for the similaritiessuch as independent, parallel evolutionwere not definitively explored during the investigation."
Further, the panel determined that the FBI's data provided "leads as to the origin of the anthrax spores in the letters," but the data did not rule out other possible sources, the NAS said. Committee members said today they could not quantify the probability that the letter samples actually trace back to the USAMRIID flask.
The NRC also emphasized the FBI's own finding that the anthrax in RMR-1029 was not the immediate source of the mailed anthrax, because one or more "intermediate growth steps would have been required to produce the anthrax in the attack letters." The FBI described the RMR-1029 anthrax as the "parent material" to what was mailed.
The panel also determined that silicon was present in the mailed anthrax, but it agreed with the FBI's conclusion that there was no evidence that the substance had been added to "weaponize" the anthrax by aiding its airborne dispersal.
In releasing their report today, committee members said completion of the work was delayed when in November 2010 the FBI notified the panel about additional relevant materials, even though the panel had repeatedly asked for all materials. "The additional material provided new insights and new information about overseas samples and resulted in a new section of the report," said Alice P. Gast, chair of the panel and president of Lehigh University in Pennsylvania.
The committee also learned last fall about the existence of classified information about the investigation. But because of the lateness of the revelation, the need for timely release of the report, and an understanding with the FBI that all materials the NRC reviewed would be made public, the committee decided not to review the classified material, the report says.
One of the committee's two recommendations is that the classified materials should be reviewed. A second recommendation is that investigative bodies should take great pains to educate the public and policymakers about the goals of forensic science and its limitations when used to investigate a biological attack.
Beyond its overall finding that the scientific evidence permits no definitive conclusion about the origins of the mailed anthrax, the committee makes nine main points in its report. The first is that the FBI correctly identified the letter anthrax as the Ames strain of Bacillus anthracis and also had good reason to conclude it had not been genetically engineered.
A second point is that the letter anthrax, when cultured, produced several different types of colonies, or morphotypes, and specific genetic sequences associated with these types provided a way to assess relationships among different anthrax samples studied. However, the development and testing of the assays for these different strains took a long time and slowed the investigation.
The FBI created a repository of Ames strain anthrax samples gathered from labs around the world and then compared them with the attack anthrax in an attempt to find the source, the report explains. The genetic analyses of the repository samples were consistent with the finding that the letter spores were derived from RMR-1029, but, as noted above, did not prove such a relationship, the committee found.
When pressed by a reporter to describe in simple terms the extent to which the panel agreed or disagreed with the FBI conclusion in the case, Gast said, "We do say the results are consistent with a link between the letter samples and RMR-1029, but they're not definitive because there are other possible explanations. We can't quantify that for you, unfortunately."
Gast said further, "What we're saying in a practical sense is that you can't rely solely on the science, and any statements that rely on the science as a foundation for a definitive conclusion can't be made because there are uncertainties, particularly in this field of microbial forensics."
David A. Relman, vice chair of the NRC panel, said the FBI found four specific mutations in the mailed anthrax and designated them as the genetic signature of the material. Among 947 repository samples that were tested, 8 were positive for all four mutations, he said. Seven of these were said to have come from RMR-1029, and the other one came from another lab but was said to have originated from RMR-1029, he explained. Relman is a professor at Stanford University School of Medicine.
In a statement today, the FBI said that the NRC committee "concluded that it is not possible to reach a definitive conclusion about the origins of the B. anthracis in the mailings based on the available scientific evidence alone. The FBI has long maintained that while science played a significant role, it was the totality of the investigative process that determined the outcome of the anthrax case."
The agency added, "The FBI believes that today's report will increase the public's understanding of the exhaustive effort that resolved one of the most extensive investigations in the history of the FBI."
In other findings, the NRC committee said:
- It is difficult to draw conclusions about the amount of time needed to prepare the anthrax used in the attacks or the "skill set" the perpetrator required.
- "There was inconsistent evidence of B. anthracis Ames DNA in environmental samples that were collected from an overseas site." The data on these samples deserve a more thorough scientific review.
- There are other tools, methods, and approaches available today for a scientific investigation like this one, including the use of high-throughput DNA sequencing. The use of these tools might clarify the inferred association between the RMR-1029 and the attack anthrax.
- The FBI built an appropriate organization structure to accommodate the complexity of the anthrax investigation and received the advice of prominent experts.
Feb 15 NAS news release
NAS press conference Webcast link
Full NAS report
Feb 15 FBI statement
Sep 16, 2008, CIDRAP News story on FBI's request for the IOM review