Thursday, January 16, 2014

If humans were rats...

Monsanto safety test assures half would be dead
the rest would suffer from  kidney and liver disease in under 90 days...


Dear Food and Chemical Toxicology: shouldn't you retract Hammond's study?

     

“Results of a 13 week safety assurance study with rats fed grain from glyphosate tolerant (Round-Up-Ready) corn." 

Hammond B, Dudek R, Lemen J, Nemeth M.
Food Chem Toxicol. 2004 Jun;42(6):1003-14


I am Not a Sprague Dawley rat, but my cousins said: Hammond isn't conclusive



Veterinary medical perspectives* 

  

Food and Chemical Toxicology, the journal which recently retracted the  infamous  Seralini study reporting  kidney & liver disease, as well as unexpected breast canceralso published four safety assurance studies by Hammond et al.  

The reason for the retraction was simply bizarre. It was retracted because the results were found  to be inconclusive : "Ultimately, the results presented (while not incorrect) are inconclusive, and therefore do not reach the threshold of publication for Food and Chemical Toxicology"

The feeding trial I will discuss on Sprague-Dawley rats (same rat strain as Seralini's) by Hammond et. al. was published in 2004 and cited by Seralini, but curiously has not drawn any attention.  It is worthwhile to read and understand the study  that triggered Seralini's investigation-- full text of the study is linked in the title. It represents a standard safety test on the most prevalent transgenic crop in the food supply. Please take this opportunity to read the study linked in the title--few GMO studies are OPEN ACCESS. In fact the majority are neither peer reviewed nor published. 

 The study is a safety assessment to ensure that the RoundUp Ready corn (NK 603) is as safe as conventional corn involving two components. Firstly, it is an evaluation of safety of the newly introduced trait, Round Up Resistance conferred by the Round Up Ready transgene (CP4 EPSPS). Secondly, it is an assessment for possible toxicity due to unintended pleiotropic effects resulting from the insertion of the trait (transgene). [*] 

Problem is, the study does not provide robust evidence to support either claim, is just as  inconclusive as Seralini- and should likewise be retracted. The following dissection should clarify the reasons why  it fails to rule out allergenicity,  kidney and liver toxicity in rats- never mind humans-in spite of being titled a Safety Assurance Study

Safety of introduced CP4 EPSPS trait

ALLERGENICITY  OF Round Up Ready (CP4 EPSPS)  transgene protein has not been ruled out



Committees of global experts have created decision trees largely based on assessment of IgE-mediated food allergenicity for risk assessment.[8] The WHO guidelines state that a transgene with greater than 6 (continuous) amino acid homology to a known allergenic epitope or > 35% of (discontinuous) sequence , should be further screened on 25 individual serum samples with high IgE titers to that air borne or food borne allergen.[2] CP4 EPSPS protein has homology of seven amino acids  to a known and prevalent allergen- dust mite allergen (der p 7).[1] 
Hammond cites Harrison et al, a study on digestibility of CP4 EPSPS protein as evidence for lack of  allergenicity of the introduced transgene. While digestibility is a preliminary step in screening novel proteins, it is not a substitute for immunologic studies. Hammond also cites studies on Round Up Ready soybeans as evidence of lack of allergenicity. However, studies on  Round Up Ready soybeans are not a substitute for studies on Round Up Ready corn.  Moreover, a targeted analysis on serum from patients with dust mites allergies was not performed  on 25 serum samples in the studies.  The largest inhibition ELISA on CP4 EPSPS was reported for Round Up Ready soybeans on a trivial sample of 4 [9]  and thus the studies would not uncover cross reactivity  between Cp4 EPSPS and der p 7  in soybeans.  

Research to verify lack of cross-reactivity is not cited by Hammond et al, and a literature search failed to uncover such studies on 25 serum samples with high levels of IgE to dust mites.[9]   Inhibition ELISA studies can and should  be conducted to assess for the presence of allergenic cross-reactivity between the EPSPS CP4 protein in corn and der p 7 to definitively rule out allergenicity [10] This is a minimal standard of risk assessment for known allergens, while everyone accepts that even this standard leaves large gaps since not all allergies are IgE mediated.  



Toxicity due to pleiotropic effects resulting from insertion into the corn genome [*]


Random integration of a transgene may cause gene disruptions leading to sequence changes, production of new proteins or formation of new metabolites or altered levels of existing metabolites that could compromise safety. This includes the potential production of new allergens or toxins.[5]

A proteomics study on maize demonstrated that 43 proteins were up- or down-regulated in transgenic seeds with respect to their controls specifically related to the insertion of a single gene into a maize genome by particle bombardment.[4]  
Non-targeted “omics” profiling studies might be able to  detect unintended and unexpected changes  that targeted compositional analysis underpinning the current risk assessment by Hammond et al would not, making this safety assurance study more conclusive. [11]




Experimental design flaws


Invalid reference groups.

The experiment was conducted on 400 rats fed two doses of corn (11% & 33%). A balanced experimental design using 400 rats would consist of 50 rats/sex/dose on transgenic corn compared to 50 rats/sex/dose of its isogenic parent line grown side by side. The reason for this is that agronomic factors (soil, fertilizers), and environmental influences (location, weather, stress) are factors that should be considered during “GE versus non-GE” evaluations. [3]  The environment has been shown to play an important effect in the protein, gene expression and metabolite levels of maize samples tested, where 5 proteins, 65 genes and 15 metabolites were found to be differentially expressed.[5]  Approximately 100 total proteins were differentially expressed as a consequence of  environmental influence in a proteomics study. [4]


See Table 1. Hammond's reference groups A-F are grown in random geographic locales such as Indiana, Iowa and Colorado, while the genetically engineered corn was grown in Ohio!

Thus reference control  A-F in Table 1  not grown in Ohio--are inappropriate reference groups used to establish invalid reference ranges. Shrinking the experimental group to 80 (20/sex/dose)  while increasing the control group to 320 animals serves to increase type II error (false negatives), reducing sensitivity of detecting adverse health effects.  


I could stop right here. If the reference ranges for non-GE corn are invalid, it isn't possible to compare GE and non-GE for statistical differences within 2 standard deviations as the study does. 
But I shall go on.


DATA is reported for HALF the animals: Selection Bias


Elementary school arithmetic for non-scientists.  Number of experimental animals (N) in Tables should be 20 because there were 20 rats/sex/group.  Don't feel too bad if you aren't finding it easy, because the "esteemed"  scientists and editors at  Food and Chemical Toxicology  evidently flunked elementary school arithmetic as well.

There is a very good reason a system of blinding has been used for hundreds of years in research to prevent bias. Scientists often want a particular outcome from an experiment, and it is  possible to choose the animals to test in such a way as to achieve desired results.  In a non-blinded experiment, where the researchers know which animals are exposed to which food, and test results are published for half the animals ( N 4-10, instead of 20 in Tables 2,3,4), the researchers are free to cherry pick the rats based on overt clinical signs.  If the transgenic corn was engineered with a metabolite toxic to the kidneys, the most  affected rats would be identifiable, as they would be drinking excessively and urinating excessively. Rats with liver disease could likewise be avoided for sampling  because their white skin would glow yellow  due to jaundice.  Bias does not have to be due to deliberate deception- human nature makes scientists vulnerable to confirmation bias. 

Data and conclusions of a non-blinded experiment reporting data for half the animals should be declared invalid by design.

 I could stop right here, but I shall go on. 





http://www.largeverdicts.com/illustration/kidney-nephron.jpg



KIDNEY TOXICITY

Sprague Dawley rats suffer from spontaneous kidney disease, chronic progressive nephropathy (CPN) which confounds toxicology studies.  CPN primarily affects males and is felt to be exacerbated by ad libidum feeding and high protein diets. 

Blood tests associated with kidney function (BUN, creatinine)  are not significantly elevated until advanced stages of disease, which makes the “normal”  tests reported in Tables 4 & 5 meaningless. 
A urinalysis does indicate progression and loss of renal function by declining urine specific gravity, elevated levels of urine protein/ albumin, and casts- prior to elevation of blood tests.
 These inexpensive and crucial functional tests are not published in this study nor any Hammond et al feeding trial.
I could stop right here, but I shall go on.

The  findings reported in Table 7 are vague and not descriptive of the specific histopathological changes.  The earliest lesions of CPN in young rats are convolutions of a single proximal tubule showing basophilia and crowded nuclei, representing simple tubule hyperplasia, with thickened basement membrane (especially of  the Bowman's capsule),  which is not reported in this study. The study instead reports mononuclear cell infiltrate, which is  generally a feature of inflammatory/ immune mediated processes. However, CPN is not an inflammatory or vascular disease, and it has no immunological or autoimmune basis. [6]
 Regeneration is reported in 17/20 rats, which is disproportionately high to the number of animals in whom degeneration is reported, while these processes occur simultaneously. Thus without a  review of histopathology slides by independent pathologists assurances of absence of reno-toxicity can not be made.  

Furthermore, pathologists recommend histopathological grading on a scale of 0-4 (kidneys graded by the study pathologist as minimal = 1 (less than 25% of renal tubules affected); mild = 2 (T 25–50%); moderate =3 (T 50–75%); or marked = 4 (T 75%); [7] [6] which likewise was not done in this study, making a valid comparison of severity between the groups impossible.

In any case, sample sizes of 9-10 for whom results are published are too low to parse toxic  effects  fron confounding  spontaneous kidney disease - unless the potential  toxin has a very strong effect  in a very short period of time. 

So, if you recall the main, and I believe,  valid  criticism of Seralini as having sample sizes too small  to draw conclusions, the same is the case with this study. The sample size is too small to rule out kidney toxicity, separate and apart from the  other flaws I listed above.  GMO toxicity would be masked by spontaneous kidney disease which doesn't  manifest until more than 75% of renal function has been destroyed.  

The difference is that Hammond et al  reaches the  questionable conclusion that this  GMO corn does not cause harm, even as half the laboratory animals disappeared, while Seralini accounts for all of his rats but concludes that the corn is harmful. 

I could stop right here, but I shall go on.


Liver toxicity 

Tables 2,3,4  report normal liver enzymes which would suggest that the corn is not hepatotoxic. Unfortunately they do not. 

Liver enzyme tests are not liver function tests-they are tests of inflammation.  In fact, liver enzymes can appear relatively normal in the face of liver failure.  Bilirubin is a marker of jaundice most often associated with hepato-biliary disease.   In cats the most common form of liver disease is known as hepatic lipidosis or  fatty liver disease. The same disease is recognized  in rats and in  people, and it is called  non-alcoholic fatty liver disease (NAFLD).
 Its prevalence rate  is rising- especially among kids.

 I present an illustrative  case report   with  blood work and a urinalysis in a cat with imminent liver failure, in whom liver enzymes were nearly normal.  The test most crucial for diagnosis was serum and urine bilirubin. 

Total bilirubin (t.bili) in Table 4 is reported for 4-6 out of 20 male rats and 7-8 out of 20 female rats. Bilirubin values in the urine are not published for the rats at all. 


The study reports mulifocal chronic inflammation in most of the rats in Table 7 which is histopathological indicator of  chronic inflammatory hepatitis. Histopathological inflammation is usually associated with elevation of liver enzymes, interestingly reported to be within normal ranges, which doesn't make sense.  The clinical test veterinarians usually perform to assess animal liver function ( in contrast to inflammation) is the bile acid test  - available for rodents, but wasn't performed.

The liver represents a suitable model for monitoring  effects of a diet, due to its key role in controlling the whole metabolism.
 A study examining effects of Round Up Ready soy diet were studied on liver of  female mice in order to elucidate possible interference with ageing.  Several proteins belonging to hepatocyte metabolism, stress response, calcium signalling and mitochondria were differentially expressed in GM-fed mice, indicating a more marked expression of senescence markers in comparison to controls. Moreover, hepatocytes of GM-fed mice showed mitochondrial and nuclear modifications indicative of reduced metabolic rate. The study demonstrated that GM soybean intake can influence some liver features during ageing while liver enzymes (the tests reported by Hammond et al) were not afffected at all.  The mechanisms remain unknown, underlining the importance of  investigating  long-term consequences of GM-diets and the potential synergistic effects with ageing, xenobiotics and/or stress conditions. [12] [13]



Statistics are invalid


Tables 2-4 do not report precise number of rats. Means and standard deviations cannot be calculated for ranges of rats.

A  study with four times as many control/reference animals (N=320) as experimental (N=80) is intuitively grossly imbalanced  which raises questions about the accuracy of the statistics cited. 

------------------------------------------------------------------------------------------------------------


I hope I've clarified  why this study by Hammond et al does not meet its stated goal of reassuring me of safety of this crop to rats. The other three studies linked above relied on  the same flawed experimental design. By definition, no study on rats can meaningfully show safety to a human, since many adverse effects of food additives and drugs are often discovered only when large randomized epidemiological studies are performed on humans. 


 In spite of  lack of  " conclusiveness"  factor  I outlined above, this study has not been retracted,  which suggests that the agricultural biotechnology science establishment is promoting scientific double standards. This is scandalous for many distinct reasons, not the least of which is that science is uncovering multiple mechanisms  which could have caused the rats in Serallini's study to develop breast cancer. 

I could go on and say more ....but  I won't.   Hopefully, I've given  you enough food for thought.  





References

1.BMC Structural Biology Screening of transgenic proteins expressed in transgenic food crops for the presence of short amino acid sequences identical to potential, IgE – binding linear epitopes of allergens Gijs A Kleter  and Ad ACM Peijnenburg

   2.Evaluation of  Allergenicity of Genetically Modified Foods Report of a Joint FAO/WHO Expert Consultation on Allergenicity of Foods Derived from Biotechnology  January 2001 Food and Agriculture Organization of the United Nations (FAO) Rome, Italy  http://www.who.int/foodsafety/publications/biotech/en/ec_jan2001.pdf

8/29/2014 Edit: Snap! The WHO  link went dead. Try  : http://www.fao.org/docrep/007/y0820e/y0820e00.HTM


       3. ISB NEWS REPORT • MAY 2010 Molecular Profiling Techniques as Tools to Detect Potential Unintended Effects in Genetically Engineered Maize. Eugenia Barros. http://researchspace.csir.co.za/dspace/bitstream/10204/4465/1/Barros1_2010.pdf




   4.  J Proteome Res. 2008 May;7(5):1850-61.doi:10.1021/pr0705082. Epub 2008 Apr  Proteomics as a complementary tool for identifying unintended side effects occurring in transgenic maize seeds as a result of genetic modifications. Zolla L, Rinalducci S, Antonioli P, Righetti PG.

5. Comparison of two GM maize varieties with a near isogenic non-GM variety using transcriptomics, proteomics and metabolomics.  Eugenia Barros,Sabine Lezar, Mikko J. Anttonen,, Jeroen P. van Dijk, Richard M. Röhlig, Esther J. Kok3, and Karl-Heinz Engel

6.Toxicologic Pathology, 32:171–180, 2004A Contemporary Overview of Chronic Progressive Nephropathy in the Laboratory Rat, and Its Significance for Human Risk Assessment

GORDON C. HARD1 AND KANWAR NASIR KHAN 

7. Toxicol. Sci. (2012) 128 (2): 346-356. Chemically Exacerbated Chronic Progressive Nephropathy Not Associated with Renal Tubular Tumor Induction in Rats: An Evaluation Based on 60 Carcinogenicity Studies by the National Toxicology Pgram Ronald L. MelnickKathleen M BurnsJerrold M. Ward and James Huff  


8. Environ Health Perspect. 2003 Jun;111(8):1114-21.
Clinical and laboratory investigation of allergy to genetically modified foods.
http://www.ncbi.nlm.nih.gov/pubmed/12826483


9.http://www.allergome.org/script/dettaglio.php?id_molecule=2952&year=1

Yonsei Med J. Aug 31, 2006; 47(4): 505–512. doi:  10.3349/ymj.2006.47.4.505 
Evaluating the Allergic Risk of Genetically Modified Soybean

10. Clinical and Molecular Allergy 2007; 5:2 Assessment of allergen cross-reactivity Rob C Aalberse
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC1797810/


11.Molecular profiling — a tool for addressing emerging gaps in the comparative risk assessment of GMOs Jack A. HeinemannBrigitta Kurenbach,, David Quist  
http://www.sciencedirect.com/science/article/pii/S0160412011001322


12.   2008 Nov;130(5):967-77. Epub 2008 Jul 22.
A long-term study on female mice fed on a genetically modified soybean: effects on liver ageing.

13.  2002 Aug;27(4):173-80.Ultrastructural morphometrical and immunocytochemical analyses of hepatocyte nuclei from mice fed on genetically modified soybean.

Sourcehttp://www.ncbi.nlm.nih.gov/pubmed/12441651

Footnotes:
[*] The study as designed by Hammond et al  does not permit evaluation for  pleotropic effects since there is a confounding factor that has not been segregated out. The corn was treated with Round Up- a mixture of glyphosate, AMPA and proprietary adjuvants. The authors  state that glyphosate wasn't detected at 250 ppb, however, the methodology isn't listed. Typically residues are measured by High Pressure Liquid Chromatography in tandem with Mass Spectrometry or ELISA. Since methods weren't published reported glyphosate residues are questionable. If indeed the levels of glyphosate were <250ppb they might not represent field conditions, and if greater-these additional chemicals make it  impossible to tease away any potential  metabolic toxicity caused by transgene insertion on the genome  from chemical toxicity of Round Up and its adjuvants. It is also the reason that Seralini in his retracted paper came up with the  "complicated design"  separating rats into groups exposed to transgenic corn alone ( to evaluate effects of transgene insertion on the genome in isolation of Round Up)  and those exposed to Round Up in the drinking water. 

*Scandalous! GMO safety studies are not designed, interpreted or reviewed by doctors in veterinary medicine-the only animal health professional with expertise  to interpret veterinary tests- even while the majority of GMO safety studies are performed on animals using  clinical veterinary tests.