THREE British researchers claim this week that their attempts to publish
important findings about the safety of the synthetic growth hormone BST were
blocked for three years by Monsanto, the biotechnology company that
manufactures the hormone. Their research suggested that cows given the drug
are more likely to develop mastitis, a painful inflammation of the udder. BST
can increase milk yield by up to 15 per cent.
The authors believe that if their findings had been published earlier, they
might have influenced the US Food and Drug Administration (FDA), which gave
Monsanto a licence to sell BST last November. In the European Union,
agriculture ministers agreed in December 1993 to a one-year moratorium on the
use of the drug. This decision had nothing to do with the question mark over
the welfare of dairy cows, but was intended to prevent Europe鈥檚 milk lake from
growing any larger. Agriculture ministers meet again later this year to decide
whether to lift the moratorium.
In this week鈥檚 Nature, Erik Millstone of the Science Policy Research Unit
at the University of Sussex, Eric Brunner of University College London and Ian
White, a statistician at the London School of Hygiene and Tropical Medicine,
present their analysis of eight trials of BST carried out by Monsanto. The
company maintains that it was well within its rights to block publication of
these results until its own researchers and their academic collaborators had
published their analysis of the data.
Advertisement
鈥淢illstone was given the data for his own use,鈥� says Doug Hard of Monsanto.
鈥淗e wanted to publish before the others, and in the scientific world, that鈥檚
not acceptable.鈥� Monsanto finally published its data in August, clearing the
way for the British researchers to publish.
In Monsanto鈥檚 paper in the Journal of Dairy Science, the company鈥檚
researchers conclude that the incidence and severity of mastitis is linked
mainly to milk yield: cows which produce more milk are more likely to develop
the condition. They base their findings on the results of 15 trials, including
the eight analysed by Millstone and colleagues. They say a cow with a low milk
yield given BST has the same risk of mastitis as an untreated cow that
produces a high yield naturally.
In its assessment of BST, the FDA concluded that 鈥渢he risk of clinical
mastitis was slightly higher in dairy cows treated with [BST]鈥�. But it ruled
that the increase was of no concern.
Millstone and his colleagues disagree. Their work is based on 鈥渟omatic cell
counts鈥�, measures of the numbers of white blood cells in a cow鈥檚 milk. These
cells, called leucocytes, help to combat infections and higher counts are
linked with worse infections. 鈥淥n average, BST treatment produced a 19 per
cent increase in somatic cell counts relative to controls,鈥� they say.
Millstone and his two colleagues say it was impossible for them to tell
whether the inflammation was caused directly by BST, or by the increased milk
output. But he stresses that whether the mechanism is direct or indirect is
irrelevant. 鈥淐ows鈥� health may deteriorate for other reasons, but that鈥檚 no
reason to say that one more insult [from BST] doesn鈥檛 matter,鈥� he says.
Millstone and his colleagues sent their analysis to the FDA, but are
鈥渃oncerned鈥� that their findings were kept out of the public domain at a time
when animal welfare issues surrounding the use of BST were unresolved.
鈥淣aturally, we accept that Monsanto and their principal investigators are
entitled to decide when and how their raw data are published,鈥� Millstone
writes in Nature. 鈥淏ut we do not agree that they have any right to prevent or
delay the publication of our work analysing their data.鈥�
Millstone first asked Monsanto for data relating to trials of BST in 1989.
On 18 October that year, the manager of Monsanto鈥檚 Animal Sciences Division at
the time, Neil Craven, wrote back saying: 鈥淎s you know, we request that the
raw data be kept confidential 鈥� We hope that you will discuss any
interpretation of the data with us before disclosing it to third parties.鈥�
The British researchers have tried to publish their results twice before,
in the Veterinary Record and the British Food Journal. On both occasions, the
paper was withdrawn at the last moment. Millstone says he and his colleagues
kept their promise to discuss with Monsanto any papers intended for
publication. But after seeing their submission to the Veterinary Record,
Monsanto wrote to Millstone saying: 鈥淲e 鈥� must refuse permission for you to
publish these results.鈥� The second attempt at publication fell through when
the British researchers refused to indemnify the British Food Journal鈥檚
publisher against breach of copyright.
On 5 February 1992, Hard wrote to Millstone saying that: 鈥淎s the raw data
are confidential, all subsequent analyses are as well.鈥� He added that Monsanto
had submitted a paper containing the data to the Journal of Dairy Science. In
a letter to Brunner on 5 March 1993, the journal denied having received a
paper from Hard 鈥渙n the topic of BST and somatic cell counts鈥�.
Brunner queries the time it took Monsanto to publish its own analysis of
the data. 鈥淭hey said we should show them the paper before we show it to third
parties,鈥� he says. 鈥淲e did that, and never at any stage have they made any
criticism of our work 鈥� and never did any of the referees who looked at our
work on behalf of the journals.鈥�
Hard says the peer review process takes time, especially when 20 groups of
authors are involved. 鈥淭hat can take a couple of years,鈥� he says. 鈥淭hen
getting all the authors to agree takes an additional amount of time.鈥�