Does big business mean bad science?: A2 milk

Ten years ago, in a meeting room at Auckland University medical school, representatives from the Dairy Board sat down with executives from an American food giant and agreed a deal to supply hundreds of thousands of tonnes of milk powder.

The American giant was Abbott Laboratories, the world's third biggest producer of infant formula, and the milk powder was A2 a type of milk the Dairy Board believed had none of the health risks associated with its standard product, A1. Those health risks were thought to include diabetes, heart disease and mental illnesses such as autism and schizophrenia. The agreement never led to a full supply contract why, is not known. Indeed, details of the meeting and the Dairy Board's relationship with Abbott are now disputed by those involved.

But that should now be no surprise in the story of A1 and A2 milk all facts are disputed, science is inconclusive and the stakes are higher than in the decades-long battle over tobacco.

That's why Lincoln University professor Keith Woodford sat down three years ago and began writing Devil in the Milk.

The book, published this week, draws together the strands of scattered scientific studies on whether A1 milk could be a health risk, and, if so, how. It also records the about-face of the Dairy Board, since merged into Fonterra, as it shifted from driving A2 research to denial and denigration of A2's health claims.

Woodford is clearly in the pro-A2 camp, presenting evidence that science has been led astray by commercial interests. "The way science is set up in New Zealand is that when you have a controversial issue like this, if there's one group that want to close it down they can do so extremely effectively," he told the Sunday Star-Times.

Once the dairy industry began to see A2 as a risk, rather than an opportunity, the incentive to fund research into A1's potential health risks dwindled away, Woodford said.

The industry's view was summed up by former Dairy Board chairman Warren Larsen, who told Australian TV in 2003: "You never do anything that destroys the (market) category. Nothing. And in this case that's precisely what the A2 Corporation, in my view, has done."

Professor Bob Elliott, a key figure in diabetes research who did much of the initial work into A1's health effects, supports Woodford's view.

"The association between A1 consumption and diabetes is very strong, and there is a plausible mechanism for it," he said last week. "The epidemiology can't be laughed off. (The authorities) are trying to stop smoking on evidence that is probably less solid than that, and the evidence that dietary fat is bad for you is much less solidly based.

"Having said all of that, the case is not closed by any means. But there are powerful forces out there which would want you to ignore it."

Elliott is at least as close to the science as Dr Jeremy Hill, now group director of technology at Fonterra, who was a dairy chemist in the 1990s for the NZ Dairy Research Institute.

In 1993 Hill alerted Elliott to the potential for a link between A1 milk and diabetes, and worked closely with him for several years on related research.

Today, however, Hill sees things differently. "My view, and it's not just my view, is that the balance of scientific evidence to date means that it's extremely unlikely that there's a health effect of A1 and A2, or a disease causing effect of diabetes or heart disease or autism or schizophrenia," he told the Star-Times.

"That could always change with new evidence, but quite frankly that's not transpired. And I think, if anything, it's going the other way."

These competing views of the science have left the public at a loss, and hamstrung the only company to market an A2 product New Zealand's A2 Corporation. Fair trading law means A2 can make no claims about risks associated with A1 milk because the science is disputed.

Woodford, however, says there is disturbing evidence that science has been misused in a PR campaign that obscures the real issues.

As an example, he cites an international study of mice and rats, known as the Food and Diabetes trial, that found no link between A1 milk and diabetes. Its results, published in 2003, were widely quoted as ending the debate over A2 milk. One distinguished academic wrote "any reader of the literature must surely take the findings of experienced researchers in Ottawa, London and Auckland as the latest (perhaps the final) word on this subject."

What was not widely publicised, however, was that the study and the interpretation of its results was seriously flawed.

The rodents used in the trial were supposed to be fed diets supplied by NZDRI that clearly distinguished between A1 and A2 milk types supplemented by two kinds of base diet infant formulas Prosobee and Pregestimil. A control group was fed a purely cereal-based diet. After the study was done, but before the results were published, it emerged that Pregestimil had been contaminated with beta-caso-morphin-7 a substance found in A1 milk but not in A2, and the very thing scientists thought may have been a trigger for diabetes.

This substance, known as BCM7, is what Woodford calls "the milk devil".

The contamination was not mentioned in the published results, nor was it revealed to the whole industry at the time.

For Elliott, one of the scientists involved in the trial, this "invalidates the study and its conclusions must be withdrawn".

Hill, also a co-author of the study, disagrees. "The results are pretty consistent," he told the Star-Times. "It was the non-milk-containing diet that was far more diabetogenic in both animal models."

Except, Woodford said, the animals used in the trial were specially bred to be prone to diabetes when fed a cereal-based diet, so that result was only to be expected. It was the A1/A2 comparison that mattered.

Hill's comment also seems at odds with his memo to Larsen in October 2000, in which he wrote: "(An) important result from the trial was that a hypoallergenic infant formula (Pregestimil) also produced high levels of diabetes. NZDRI has since shown that Pregestimil contains a high amount of BCM7. This result is not known outside the NZ dairy industry."

These arguments over the validity of the FAD study cover such intricate details that only a book could discuss them fully, but the evidence in Woodford's book blasts a big hole in any scientific confidence that A1 milk carries no health risk.

While the case is far from proven the A1 theory, for example, can't explain why the incidence of diabetes has doubled in the last 20 years what concerns Elliott is that further studies are required. But "who's going to do it? Some milk company? Fonterra's certainly not going to do it they've washed their hands of it."

The A2 story was just a regrettable dead end, Hill says. "As the originator of the idea with Bob, had it held up obviously I would have been delighted to have seen such an effect and been able to make a contribution in that way... Had there been an effect, New Zealand would have been in a position to have owned the intellectual property we could have potentially been in a very favourable long-term position. But the thing that would have to be clear is knee-jerk reactions around the potential health consequences could have a very damaging effect and still could. But that would only be the case if it was valid and we weren't able to react to the situation.

"Until the scientific evidence says otherwise, then really there isn't a cause for concern. Our position has to be that we are an ethical marketer of dairy products. So if we are promoting any particular benefit then it has to be based on valid science."

Meanwhile Elliott's diabetes work got a fillip last week the company he co-founded 20 years ago, Living Cell Technologies, won ethical approval to begin a New Zealand clinical trial of its pig cell transplant technology. No stranger to the long struggles involved in scientific progress, he remains convinced that A1's health risks are a live issue.

"The numbers are there, it can't be ignored. You ignore it at your peril."

- © Fairfax NZ News