Cause of SARS disputed

Head of Canadian lab not convinced that coronavirus causes SARS.

Apr 11, 2003
Robert Walgate(walgate@scienceanalysed.com)

The director of one of the World Health Organization's global network of 11 laboratories investigating SARS (Severe Acute Respiratory Syndrome), told The Scientist yesterday (April 10) that the new coronavirus implicated as the cause of the disease is certainly around in the environment but is unlikely to be the causative agent. Frank Plummer is director of Canada's National Microbiology Laboratory in Winnipeg.

Canada is the Western country hardest hit by SARS, which arrived in Toronto before WHO announced its global alert on the disease March 12. It has seen 190 SARS cases, in two waves, and 11 deaths, Donald Low told The Scientist. Low is chief microbiologist at Mt Sinai Hospital, Toronto. He dealt with the first case in Canada and has just emerged from quarantine.

But according to Plummer "The proportion of our samples [from Canadian SARS patients] that show the coronavirus is going down." He said earlier this week that about 60% were showing the virus by PCR test; now it's only 50%.

"Of course, the case definition of SARS is a little loose," said Plummer "but many of the Toronto cases are epidemiologically linked, and we are finding some of the best-characterized cases are negative. So it's puzzling. As is the fact the amounts of virus we are finding, when we find it, are very small – only detectable by very sensitive PCR."

Their samples — from about 50 cases of probable SARS and 30 cases of suspected SARS — are primarily, but not exclusively, nasopharyngeal swabs. "That's what the majority of labs around the world are testing… it's where you find most respiratory viruses… It's strange [that there's so little virus there] because it seems to be transmitted by close contact," said Plummer.

"Coronavirus could be the etiology – but I'm not impressed," he said. On the basis of the Canadian data, "the chances of having SARS if you have this virus are increased by about a factor of two – compared with if you don't."

Plummer had heard the recent results from Hong Kong, published this week in The Lancet, which showed 45 out of 50 patients to have the new coronavirus, "but I don't understand what the difference is. The coronavirus is definitely around in the environment, it's definitely circulating," he said "but the relationship with SARS, based on our data, appears to be fairly weak. It could be that it causes SARS and we are sampling the wrong sites. Or we are getting poor specimens. But we don't believe it's because of our testing. It could be you need more than one thing to get SARS. Or it could be an epidemic of coronavirus superimposed on an epidemic of something else, and we're trying to work through that."

All 11 laboratories in the WHO network have been discussing these results among themselves and, according to Plummer, "the results from different labs around the world — except for what's been published in The Lancet — appear to be not that different from ours."

"I think everyone is under huge pressure to get stuff out, and the journals are bugging people for papers," said Plummer. "And some people are slapping things together. My position is I remain to be convinced."

As for the tests themselves "We've been using primarily PCR but also some cultures." The sequences Plummer's lab is looking for are "the same as everyone else's," provided by the US Centers for Disease Control. The coronavirus is both avian-like and bovine-like, and is thought to represent a new family of coronaviruses. It was found in early SARS samples by means of a microarray developed by Joseph DeRisi at the University of California in San Francisco.

"And we have positive results in people who don't meet the case definition. That's not unusual; Hong Kong, I believe, is finding asymptomatic positives. People in Winnipeg [which is well outside the Canadian SARS foci] test negative, there's no doubt about that. But we have two people from outside of Toronto [the main focus] with no travel history to the Far East and no known contact with SARS patients who test positive; and people who've either traveled to the Far East or who have had contact with SARS patients, but are not themselves probable or suspect cases of SARS, who test positive."

The US Centers for Disease Control had, however, reported that it had found no evidence for positive tests in 400 non-SARS controls. "Yes, but I believe they were antibody tests on banked serum of CDC employees. In Canada, we've done tests on banked pharyngeal swabs from a study we did about a year ago during the respiratory virus season, and we also found zero. So yes, this is definitely a new virus, circulating currently, and we probably haven't seen it before — that's what our data suggest — but is it really the whole or part of the cause of SARS?"

As for other agents that have been implicated — paramyxoviruses like metapneumoviruses, and the recently emerged Chinese suggestion that chlamydia is involved — "I think the chlamydia has pretty much been ruled out by most labs outside China; we find metapneumovirus in a few patients, but there's not really any correlation."

But in contrast to Plummer, Julie Hall of the WHO Global Outbreak Alert and Response Network told The Scientist she was, if anything, more certain than before. "I think we're up to 98% [certainty that coronavirus causes SARS]," and she was unaware (on April 9) of any false positives.

"However, the 2% doubt is there over whether coronavirus can do it by itself." Hall said it was still possible that other viruses are needed to cause the really severe symptoms.

She was glad that all possible avenues are being explored, and pointed to the length of time it took to establish that HIV causes AIDS. She admitted that other evidence, from animal studies and serology, would be needed to confirm the new coronavirus as the cause of SARS, and "we don't quite have that yet."