Memorandum submitted by the Shellfish
Association of Great Britain (M5)
SUMMARY
Transfer of DSP testing to CEFAS, Weymouth,
in June 2001 coincided with false, "atypical" positive
response to mouse bioassay. The mice died within minutes as a
result of a convulsive CNS reaction symptomatic of solvent carryover.
The response could not be duplicated on the same material at other
EU laboratories. Similar procedural or artefactual responses have
been recorded in many countries but eliminated or replaced by
chemical testing. An empirical approach to quickly determine the
implication to human health was ignored by the FSA which pursued
a supposed novel toxin. The intent of the EU Directive is to monitor
for specific named toxins associated with specific planktonic
algae. CEFAS showed lack of due diligence in modifying the established
bioassay and there was no basis for the FSA presumption of a new
toxin and the closure of cockle fisheries. The economic impact
on the cockle industry was huge. After two years, the FSA submitted
to independent audit and solvent studies. These identified widespread
procedural malpractice and the carryover in the bioassay of ether
and acetone solvents, and "dirty water" from the cockle
matrix, all of which the test is designed to remove if conducted
properly. The presence of solvents is consistent with the anaesthesia
and rapid death in mice. Independent peer review confirmed the
serious non-conformance, outside the terms of the licence to conduct
the test. The Home Office demanded the immediate elimination of
solvent carryover. A comprehensive technical analysis by FRAME
confirmed the gross misuse of animals and ill-judged application
of science; it recommended actions to ensure the abuse could never
happen again. Industry has proposed a bio-security programme (CIVIT)
which uses only best practice non-animal methodologies and gives
better consumer safety than the mouse bioassay. The FSA misapplication
of the precautionary approach, poor science and lack of transparency
has serious implication for all FSA/CEFAS food safety work, and
the lack of control on the use of animals to wider testing programmes.
1. Background
Diarrhetic Shellfish Poisoning (DSP), together
with Paralytic Shellfish Poisoning (PSP) and Amnesic Shellfish
Poisoning (ASP), testing on cockles and other bivalve molluscs
in England, Wales and Scotland was conducted by Fisheries Research
Services (FRS, Aberdeen) operating under the National Reference
Laboratory (NRL) for toxins from 1996 to June 2001. Following
tender, the contract for England and Wales was awarded to the
Centre for the Environment, Fisheries and Aquatic Sciences, Weymouth
(CEFAS) in June 2001. CEFAS unilaterally modified the testing
methodology, failing to include safeguards adopted to minimise
false positives, and from the beginning produced what they described
as "positive" results. Industry registered concern at
the integrity of the results. For one year, the Food Standards
Agency (FSA) did not declare that the responses were atypical
and the results were recorded as positive DSP. In June 2002, the
FSA advised that the mice had been dying as a result of a convulsive
central nervous system response within minutes of injection. The
normal diarrhetic positive response is death between 2 and 24
hours. The mice died long before a DSP test was completed and
thus false positive results could have camouflaged positive results
for real DSP which would have triggered product recall. Cockle
fisheries in England and Wales were closed on FSA advice following
the atypical positive results; the FSA advocated a precautionary
approach to the possible existence of a novel toxin but continued
to describe the result as "atypical positive for DSP".
Industry queried how the transfer of testing to CEFAS coincided
with areas as far apart as South Wales, the Thames and the Wash
becoming simultaneously contaminated.
2. Failure to Duplicate Results
FRS continued to conduct DSP testing on Scottish
cockles after June 2001 using its established method. The atypical
false positives at CEFAS, Weymouth, were not experienced at FRS
Aberdeen and duplicate cockle samples sent to Aberdeen from closed
fisheries in England and Wales could not replicate the atypical
result. Industry also undertook duplicate testing in Holland and
France using split samples, and has never replicated an atypical
response. From the beginning, the NRL requested the FSA to undertake
tests on possible solvent carryover but the FSA failed to do this.
Industry also advised FSA that the death was symptomatic of solvent
carryover response. The "atypical result" has never
been experienced when a proper extraction procedure has been in
place.
3. Non-uniformity in Testing
The FSA accepted that the FRS test was in accordance
with the EU Directive. Product from Scotland, Holland, France
or New Zealand is brought into England and Wales without being
subjected to the CEFAS test. Ninety percent of UK cockles are
exported to EU Countries.
4. International Modification to Testing
Methodology
Similar anomalies using the DSP intra-peritoneal
mouse injection (MBA) have been recorded in most countries, particularly
on cockles. All modified the test to address the random procedural
or artifactual responses, or changed to chemical testing (by liquid
chromatography mass spectrometry (LCMS)), having established the
necessary reference standards. Holland and Germany do not use
the mouse test on animal abuse principle. France, Spain, Portugal,
Ireland, Italy, New Zealand and Canada have modified procedures
to ensure that solvent is not carried over. Ireland, New Zealand,
Canada, and the US have adopted LCMS for the monitoring programme,
retaining a modified mouse test for background screening only.
FRS is equipped to undertake a similar LCMS programme. Yasumoto,
the Japanese chemist responsible for devising the mouse test,
has publicly recorded at international conference that the test
is unreliable, particularly with cockles, and should be replaced
for routine monitoring by LCMS as quickly as possible.
5. Identification of Implication to Human
Health
In January 2003, the Shellfish Association of
Great Britain (SAGB) presented FSA an empirical approach to identify
whether the atypical response was procedural, artifactual, toxic
only to mice, or toxic to humans. The principle that human toxicological
risk of the causative agent of the atypical MBA result can be
reasonably assessed by oral application of test material to rats
was accepted by the FSA. Yasumoto had agreed to conduct the testing
at no cost in October 2002 but the FSA has yet to supply the material.
The SAGB also arranged for the Dutch Authorities to conduct oral
testing to any FSA prescribed operating procedure. The SAGB also
queried sample transport conditions without ice, noting that cockles
were mailed (Post Office) live and dry in polythene bags causing
stress and likely degradation. The Agency undertook to implement
the empirical work at a meeting on 15 January 2003. A reputable
drug company could complete this work in four weeks. The FSA also
agreed to consult closely with industry on proposed experimental
work and modification to standard operating procedures. The FSA
did not do this and instead pursued possible routes towards identification
of a novel toxin with implication to human health and did not
attempt to rule out solvent carryover or an artifactual reaction
in mice to something emerging from the cockle matrix.
6. Intent of EU Directive
The intent of the relevant Directive 91/492
to permit free trade in bivalve molluscs within Europe, without
disparity, whilst protecting public health in a consistent way
throughout Europe. Article 2 defines "Marine Toxins"
as poisonous substances accumulated by bivalve molluscs feeding
on plankton containing toxins, naming the specific toxins. It
requires avoidance of malpractice with regard to origin and destination,
checks on microbiological quality, checks for toxin producing
plankton and biotoxins in live bivalve molluscs, and for chemical
contaminants.
7. Absence of Associated Algae
At no stage during the 30 month period have
the algae associated with DSP (Dinophysis spp or Prorocentrum
lima) been significantly present in any waters testing atypical
positive. In France, testing for the DSP toxin only begins when
a relevant associated algal bloom is present. CEFAS appears not
to have undertaken any additional phytoplankton monitoring in
order to identify any DSP causative organisms.
8. Abuse of Power
The SAGB contends that the FSA has no powers
to enact a precautionary closure if it is not for a named biotoxin
covered by the Directive or a contaminant as separately defined.
The Home Office licence is issued to conduct the specific DSP
mouse bioassay and the killing of mice to seek a possible novel
toxin appears to be an abuse.
9. Lack of Due Diligence
The Shellfish Association of Great Britain (SAGB)
suggests that CEFAS had not shown due diligence in modifying the
mouse bioassay (MBA) from that used under NRL supervision by the
FRS, with regard to:
— Possible solvent carryover resulting
from the removal of the overnight stand and other safeguards;
— Possible particulate matter carry-over
from inadequate filtration;
— Possible increase in aqueous-soluble
fraction resulting from vigorous shaking in a separating funnel;
— Fifty percent increase in ether volume
without proportionate increase of distilled water backwash.
Aqueous fractions are not the subject of the
DSP test and should not be present in the final injected extract.
The final water backwash is specifically to remove water-soluble
fractions. The CEFAS method uses the least water volume of all
member states and appears to be insufficient to remove all aqueous
fractions.
10. Discrepancy in Interpretation of Directive
Commission Decision 2002/225/EC of 15 March
2002 lays down the rules for the chemical detection of named marine
biotoxins including the DSP complex. The latter consists of Okadaic
Acid, Dinophysistoxins, Yessotoxins, Pectenotoxins and Azaspiracids.
Article 5 states that when the results of the analyses performed
demonstrate discrepancies between the different methods,
the mouse bioassay should be considered the reference method.
When chemical analysis has demonstrated the absence of all the
above DSP complex, but mice die rapidly in a way which is totally
inconsistent with DSP poisoning and prior to the end-point of
a DSP test having been reached, there is per se a "discrepancy"
between the methods. In such circumstances, the FSA is incorrect
in insisting that the mouse test has priority as the reference
method.
11. Lack of Basis for the Presumption of
a New Toxin
Prior to the meeting convened by the FSA on
1 October 2003 to present the results of its findings, the SAGB
strongly contended the FSA presumption of a new toxin in the light
of:
— Consequences of the CEFAS modification
to the MBA test;
— A multitude of substances definitely
non-toxic to humans, are known to give positive reactions during
mouse bioassay;
— Other Countries, including other
EU Member States, have not presumed procedural or artifactual
mouse bioassay problems to be due to a new unknown human toxin;
— Failure by the FSA to initiate studies
to either verify the presence of a novel toxin or to investigate
its source organism;
— Dutch cockles, not subject to mouse
bioassay, are not subject to the same investigation on importation;
— No human response has been observed
among those consuming cockles throughout the thirty month period;
— The failure by FSA to verify results
using duplicate testing and controls;
— Parallel industry testing conducted
on CEFAS "atypical positives" produced confirmed negative
results at all other laboratories;
— An FSA modified MBA operating procedure
imposed on FRS in June 2003 produced rapid mouse deaths at the
FRS laboratory;
— Confirmation by CEFAS and FRS technicians
that there was a clear solvent smell when mice died rapidly. This
was also identified by the Rowett Vetinerary staff who concluded
that the mouse deaths are symptomatic of central nervous system
disorder due to solvent presence;
— There is absolutely no evidence that
the unknown agent which is causing the mice to die is toxic to
humans.
12. Implications to Industry
The closures of the main cockle fisheries in
Burry Inlet (South Wales), the Thames (Leigh on Sea) and Wash
has had huge economic impact on fishermen, processors and marketing
companies. It has decimated an industry employing more than 2,000
people and with final value added sales around £20 million
per year. Burry Inlet was totally closed for 14 months, with the
exception of two weeks, before industry initiated zoning was accepted
by the FSA. The repeated closures in the three major fisheries
have resulted in overexploitation of small cockles and failure
to take, and subsequent loss of, large cockles. Stock structures
are no longer in equilibrium with optimum exploitation. The value
of small cockles is less than half that of large cockles. The
discontinuity of regular supply for contracts resulted in loss
of premium prices and the customer in some cases. Business planning
became impossible. The closures have also damaged the reputation
for the integrity of the product in the market.
13. FSA Reports
Independent audit of the three laboratories
(CEFAS, FRS and DARD (Department of Agriculture and Rural Development))
by Professor Makin and studies on solvent carryover by the Macauley
Institute and Central Science Laboratories (CSL) were published
by the FSA on 1 October 2003.
14. The Makin Report
This identified many areas of quality assurance
concern and lack of quality controls. Makin identified significant
variations in procedures between and within the three laboratories,
many operator deviations from the standard operating procedures
(SOPs) and different MBA end points at each site. He expressed
specific concern relating to the DSP analyses at CEFAS and surprise
that the deviations from the SOP had not been addressed by CEFAS
quality management or the United Kingdom Accreditation Authorities
(UKAS). Makin criticised the failure to use control samples or
any attempt to demonstrate reproducibility. Critically, he noted
the formation of emulsions and "dirty water" carryover
at CEFAS. This would inevitably result in the transfer of compounds
from the cockle matrix into the mouse injection extract. He observed
the clinical symptoms in mice which were wholly consistent with
an anaesthetic effect.
The FSA has yet to explain how a confirmed DSP
negative audit mussel homogenate produced positive symptoms or
death at all three laboratories, as observed by Makin.
15. Misleading Conclusion
Makin's overall conclusion that "no evidence
emerged from the audit that the atypical response is due to the
presence of ether" does not correspond with the detail of
his observations and is misleading. On 13 November 2003, Makin
verbally agreed that he could equally have concluded "that
no evidence has emerged that a novel toxin is responsible for
the atypical response". He denied FSA influence in reaching
his earlier conclusion. The FSA retains total editorial control
and ownership of contracted research apparently without provision
for peer review.
16. Solvent Carryover Reports
The Macauley and CSL reports both showed conclusively
that there was substantial ether and acetone carryover occurring
in the DSP MBA, particularly at CEFAS. Although variable, the
solvents were often at high levels. On 16 October 2003, the Home
Office confirmed its concern at the high level of solvent (ether
and acetone) found in the material injected into mice. It stated
that a high level of ether, sufficient to cause the clinical signs,
was found in many samples, while levels above the LD50 were found
in some samples. The Home Office stipulated that it is essential
that immediate steps be taken to eliminate the problem of solvent
carryover and requesting what measures are being taken to ensure
that ether and acetone are not present in the injectate used in
the mouse bioassay for DSP.
17. Independent Peer Review
Industry sought independent "high level"
peer review and analyses of the FSA reports which were undertaken
by Mackenzie, Cockrill, Parsons and Askew, Appendices III to VI
[Not printed]. All were presented at a meeting with Local Health
Authorities on 15 October 2003; the FSA was invited but declined
to attend. The papers confirmed a liturgy of malpractice, multiple
solvent and cockle juice carryover and resultant anaesthesia,
yet rapid recovery of animals that did not die. The synergistic
effect of acetone and ether combined, transfer of cockle water
and a possible cockle factor were the very likely cause of the
atypical response. It constituted serious non-conformance with
the MBA protocols and the atypical positives should have been
totally disregarded as in other Countries. The DSP MBA was specifically
designed to exclude solvent, water and cockle juice and tests
should have been immediately suspended until nil solvent carryover
was demonstrated; to continue without this, the assay is useless.
The New Zealand Authorities advised the FSA on 13 June 2003 that
vacuum drying of extracts to constant weight was required to remove
the risk of solvent residues affecting results.
The SAGB contends that the standards expected
of Government Laboratories falls far short of those expected from
private Laboratories.
18. Procedural Mechanism for the Malfunction
of the Bioassay
The procedural mechanism by which water transferred
to the injectate has been described separately by Askew. The cockle
factor may be associated with the stress caused in transporting
the samples which are mailed (GPO), live and dry in polythene
bags from the fishery to the laboratory. Cockles tend to stress
more than other molluscs because of their small size and low tolerance
to dessication. Such stress can cause the production of biogenic
amines (eg histamines or other neuroactive compounds extracted
by the acetone (Mackenzie)). In other countries, cockles are delivered
by the quickest route in ice with temperature recorders;
samples are rejected if the cockle meat ever exceeds 10ºC
(Smales, pers com).
19. Non-conformance
Mackenzie has presented a further paper "What
is different about cockles? Methodological hypotheses for the
atypical response" (Appendix VII)[Not printed]. He concluded
that failure to ignore seriously non-conforming data was a significant
departure from good analytical practice and from UKAS accreditation.
The paper analyses the competing hypotheses and overwhelmingly
concludes that an artefact is responsible for the atypical response.
The FSA is actively "spinning" against this despite
weight of evidence for it. There is absolutely no evidence for
the presence of a cockle toxin, the mouse response corresponds
closely to human solvent abuse deaths, and Occam's razor points
to artefact as the cause.
20. Misuse of Laboratory Animals
Industry has repeatedly advised the FSA that
the lack of reproducability, absence of controls and non-conforming
data constituted a failed result outside the terms of the licence
to conduct the test, and hence a misuse of laboratory animals.
This advice has been treated with contempt. A comprehensive peer
reviewed technical paper describing the gross misuse of animals
and of scientific methodology has now been independently produced
by the Fund for the Replacement of Animals in Medical Experiments
(FRAME) (Combes, ATLA, 31, 1-16, 2003 at Appendix VIII.)[Not printed].
The paper critically assesses the development, regulatory use
and methodological deficiencies of the MBA. It discusses how testing
for DSP toxins could and should have been improved and made more
humane by applying the Three Rs concept of Reduction, Refinement
and Replacement, and by the proper validation of the test method
used. It concludes that the MBA should not have been developed
for the routine screening of shellfish samples, as it has a substantially
severe endpoint (death) and is not used as part of a tiered-testing
strategy with non-animal methods. It states that during the UK
monitoring programme for DSP toxins, the assay has been used without
an optimised and universal protocol, and apparently without due
regard to the principles of basic scientific methodology. In view
of this, the atypical results obtained for cockle samples cannot
be relied on as evidence of a human health hazard. It recommends
how to replace this gross misuse of animals and ill-judged application
of science to ensure that it is never allowed to occur again.
Subsequent to this, the SAGB notified the Parliamentary Bioethics
Committee of its concerns over the FSA response to its alerts
on animal ethics aspects (Appendix IX)[Not printed]
21. Industry Initiative
Throughout the 30 month period of closures,
industry has actively sought to be positive and proactive towards
a resolution despite the negative and single minded FSA self-justification
of its own mal-administration. Industry seeks only to ensure the
safety of its cockle products and the growing problems of the
Government monitoring has required the development of its own
bio-security programme, CIVIT, Cockle Industry Voluntary Initiative
on Toxin Testing (Appendix X)[Not printed]. CIVIT will provide
a state of the art consumer protection, comfortably exceeding
that provided by Government monitoring. It will put the UK cockle
industry in the lead with regard to consumer safety within Europe,
based on the most advanced non-animal methodologies in line with
best international practice. The FSA has yet to comment on the
CIVIT proposal. The FSA is resisting acceptance of chemical testing
despite FRS being equipped to undertake the work.
22. Misapplication of the Precautionary Principle
to a Flawed Test
The precautionary principle has been misapplied
by the FSA, used as a scapegoat for its inability to address the
procedural/artefactual causes of the MBA response. Industry believes
that the Agency now believes its own spin, seeking control to
camouflage extremely poor science. It is paranoid about the release
of information. For the past 10 years, Industry has worked closely
with FRS, the NRL, and with Environmental Health Officers, but
the FSA has banned communication, going so far as to threaten
Officers under the Official Secrets Act. The FSA has introduced
new SOPs (June and November) supposedly to prevent solvent carryover
but "unofficial" communication with the laboratories
indicates that this is not being achieved. The FSA fails to accept
that the basic methodology is flawed and has deviated so far from
the prescribed method as to fail to meet the Directive. The developed
World is now using non-animal chemical methods; New Zealand Government
recently introduced its fully validated chemical testing regime,
stating that the use of mice is fraught with scientific validity
and animal welfare issues. It stated that the mouse bioassay also
produces false negatives which is an even greater risk to food
safety.
Is it reasonable to apply a precautionary principle
after 30 months without any substantiated evidence?
23. Implications to Other Areas of Work
There is concern that the poor scientific, transparency
and consultative standards demonstrated on the DSP MBA issue applies
to other areas of the FSA/CEFAS food safety work. Although the
need for FSA independence is appreciated, it is imperative that
central FSA control on funding does not jeopardise the integrity
of scientific research. DEFRA must also be independent to represent
industry. The lack of control on the use of animals also has major
implication to the integrity of the diverse animal testing programmes
conducted in the United Kingdom. Most foodstuffs subjected to
30 months of intensive analysis for a novel toxin would yield
a result, but cockles have not.
6 January 2004
| 
