Yesterday we circulated comments from Dr Arpad Pusztai on a statement by the UK's Advisory Committee on Novel Foods and Processes (ACNFP) on a study carried out by Dr Ermakova from the Russian Academy of Sciences.
Dr Ermakova found that where female rats were fed on genetically modified soya their offspring were five times more likely to die within three weeks of birth than those of mothers fed on normal soya.
Dr Pusztai took issue with the way the ACNFP in their statement drew a critical comparison between Dr Ermakova's study and one by Brake and Evenson. ACNFP stated:
"”¦Dr Ermakova's findings are not consistent with those described in a peer-reviewed paper published in 2004.1 In a well controlled study no adverse effects were found”¦" http://food.gov.uk/multimedia/pdfs/acnfpgmsoya.pdf
Dr Pusztai argued that this critical comparison was invalid, as was the implication that Brake and Evenson's research was of a superior quality. Dr Pusztai pointed out that Brake and Everson had a completely different - and highly specific - focus in their GM soya study (testicular development in young male rats) and that aspects of their study were very poorly described.
More than one of our subscribers, however, felt unclear as to why the ACNFP's comparison was invalid. One wrote to us:
"I probably have absolutely no clue what I'm talking about, but doesn't the Brake/Evenson paper's abstract state that, as in the Ermakova study, pregnant "test subjects" were fed on GM and various controls, and then their offspring "tested"? It would seem that analyzing the male offspring is not terribly different, because there were both male and female offspring (presumably) in the Ermakova study, and, in both cases, the mothers were fed GM and the offspring analyzed."
At our request Dr Pusztai commented further on the question of the validity of using the Brake and Evenson study as a means of questioning the Ermakova study. His further comments are given below.
For Dr Pusztai's previous comments:
http://www.lobbywatch.org/archive2.asp?arcid=6149
For ACNFP's statement:
http://food.gov.uk/multimedia/pdfs/acnfpgmsoya.pdf
---
Dear Jonathan,
The only passage in the Brake and Evenson paper relevant to the Ermakova study is when they say that during breeding they observed no mortality in the offspring. Now, obviously, we have to accept their statement as I have no reason to doubt its validity. However, and equally, we also have to accept Ermakova's statement unless and until it can be proven what she said was untrue. It would be quite a bit patronizing to say that Ermakova's work is inferior to (perhaps because she is Russian) or that she is less trustworthy than Brake and Evenson (B and E).
The two studies cannot be compared because there were major differences in the two studies. Clearly, Ermakova used rats while Brake and Evenson used mice. Moreover, the two groups used RR soybeans from two totally different sources. Ermakova gave its source in contrast to B and E who did not (apart from saying that it was by a seed dealer and taken from the middle of fields in South Dakota). Ermakova used whole (undefatted) soybeans as a paste and gave it to the rats in addition to the normal rat chow while B and E used extruded and dried meals and formulated diets with these, incorporating about 21% of the transgenic soya into the diet but without specifying how much conventional soya was put into the control diet. Neither of the two groups specify the amount of diet consumed by their respective animals. Basically, the description of both studies lack essential nutritional details, particularly the B and E study.
My previous comments on the B and E paper as you quote me below were given in a different context and referred to the validity of their conclusions on testicular development, which was the objective of their study. Perhaps, I should have not come back to these in the context of the Ermakova study or the ACNFP's comments to it. What I did say was that because of the poor nutritional design of the feeding, in fact to be precise no design at all, the B and E work's conclusions probably cannot be justified. We have no idea what happened to these male mice once they were born, no feeding protocol, no data, no weights, no feed intake, no growth pattern related to feed intake, absolutely nothing. If this is what ACNFP thinks the gold standard in a nutritional experiment then God save us from the consequences of their judgment when they pronounce on the safety of new products, proteins, including GM foods.
When will people ever understand that in a comparison of biological properties of animals any differences in feed intake can have immeasurably greater effect than whether the animals are fed GM- or non-GM feed. Without strict pair-feeding of a group of animals of very similar starting weights one has not a cat-in-hell's chance to find significant differences unless the differences are colossal? So when B and E say that the testicular development of their male mice was not influenced by feeding them on GM soybean or non-GM soybean they have wasted a lot of money, work and effort to come to a conclusion that would not stand up to proper scientific scrutiny. The tragedy is that most people do not see this or do not want to see it and as such these papers are a waste of our time. I think that a little bit of humility would not come amiss for members of ACNFP.
So, do I know whether Ermakova is right or wrong? No, I do not! However, if I still had a lab and animal house, I can assure you, I would find it out as quickly as possible rather than patronize a la ACNFP well-intentioned scientists who at least try to do something useful for us all.
Best wishes
Arpad