ISA claims rubbished

Led by Dr Fred Kibenge, the paper titled “Discovery of variant infectious salmon anaemia virus (ISAV) of European genotype in British Columbia, Canada”, published January 6^th, makes exorbitant claims about the presence of ISAV in BC. Specifically, the researchers claim have found genetic sequences that belong to the virus in farmed Atlantic and wild Pacific salmon, and furthermore, that the virus they detected is a new variant of a European strain.

The authors analyzed over 1,100 samples from market-bought farmed and wild salmon and used two genetic techniques (qPCR and conventional PCR) to try and detect the presence of ISAV. According to their results, 79 of the 1,106 samples were “non-negative” in at least one of the replicates by at least one of the testing methods.

So what exactly are the findings of this paper?

The authors concluded that: “This is the first published report of the detection of ISAV sequences in fish from British Columbia, Canada. The sequences detected… are of European genotype. These sequences are different from the classical ISAV… and this difference suggests the presence of a new ISAV variant of European genotype in BC. Our results further suggest that [ISAV]… can be present without clinical disease in farmed fish and without being detected by virus isolation using fish cell lines.”

They also say that their own tests could not be duplicated and have not been confirmed by accepted international tests, and no disease was present. Moreover, the authors admit that “All virus isolation attempts on the samples were negative, and thus the samples were considered ‘negative’…”

Accompanying the publication, perhaps as a ploy to distract from the non-results of the paper, the authors also released a well-timed and emotionally-charged press release, “Feared Atlantic Farm Salmon Virus Identified in British Columbia”. In the release they say things like “European ISAV is the most feared salmon virus in the salmon farming industry”, “The risk of an outbreak has the potential for severe consequences”, and “The potential that viruses such as ISAV are contributing to widespread decline in sockeye salmon”.

To the untrained eye of the general public, this publication and press release was the cause for instantaneous panic – a European fish virus is now in BC, it’s from Atlantic salmon farms, and it is killing all the wild fish. And mainstream media actively participated in the chaos. Within hours of publication there were articles circulating the internet, such as “Salmon Virus Found in British Columbia, Could Spread to Washington if Not Contained” and “Salmon virus found in Cultus Lake Trout samples”.

However, not all is what it seems, and unfortunately this game of smoke and mirrors is nothing but a repetitious construction of epic proportions. Another attempt to find something that has been exhaustively shown to not exist. Except this time, they were able to find a journal willing (likely regrettably) to publish it.

As one excellent blog (alaskasalmonranching.com) stated: “This is probably more a story about how easy it is to publish crap science, or maybe how you can turn crap science into an exciting press release that then turns into a scary, eye-candy newspaper headline.”

Fortunately, experts in the field were quick to criticize the findings in the paper. One of these was Dr Gary Marty, a Senior Fish Pathologist with the BC Ministry of Agriculture. In addition to acting as the primary veterinarian for fish pathology in BC, Dr Marty has been one of the main voices of reason with respect to fish diseases in BC, often calling into question the debatable “science” and ethical compass of anti-salmon farming activists, such as one of the paper’s co-authors, Alexandra Morton.

“The primary change that this new paper brings to BC is that we now have unconfirmed PCR test results for ISAV that are published in the peer-reviewed scientific literature.  For the past 4+ years, positive PCR test results for ISAV from BC samples have been reported, primarily by Alexandra Morton and Kristi Miller.  None of these results have been confirmed by CFIA using OIE standards, and none of these results have been confirmed by other standards of diagnostic medicine.  The recent paper (Kibenge et al. 2016) does not change this status”, explained Marty in his analysis of the paper.

He pointed out several inherent weaknesses that severely compromise the results of the study:

1.) No confirmatory tests were conducted

Firstly, using PCR alone is insufficient to verify the presence of virus. Because the chance for false positive PCR test results is high in regions of the world like BC that are free of reportable diseases like ISA, the World Organization for Animal Health (OIE) standards for disease confirmation never rely on PCR results alone. Nucleic acid sequencing, as done by Kibenge et al. (2016) is helpful to rule out nonspecific amplification as a source of false positive tests, but sequencing does not rule out the possibility of sample contamination.

“Kibenge et al. (2016) is based solely on PCR results alone; therefore, their results remain unconfirmed,” said Marty. He described several different methods independent of virus isolation that are commonly used to verify and validate the presence of virus in samples.

• in situ hybridization – highlights viral nucleic acid in tissue;
• immunohistochemistry – highlights viral antigens in tissue;
• IFAT – highlights viral antigens in tissue;
• electron microscopy – shows the virus in sections of the tissue;
• consistent repeatability within the laboratory;
• consistent repeatability of test results by an outside laboratory.

These methods could have been initiated by Dr Kibenge’s laboratory independent of the CFIA; for example, samples could have been split and sent to two laboratories; samples positive in one laboratory could have then been independently tested by the second laboratory.

However, Kibenge et al. (2016) did not report results from any other methods mentioned by the OIE manual for demonstrating the presence of ISAV in tissues.

2.) Questionable laboratory credentials

Sample contamination is particularly a problem for laboratories that do research and diagnostics on tested organisms.  Spillover of nucleic acid from research samples can easily contaminate diagnostic samples, thus using another test to confirm results from the PCR is absolutely essential.

And the lab that conducted the work for this paper has been questioned before on potential contamination issues.

In 2012, the OIE audited Kibenge’s lab, and reported that: “Dr Kibenge has five rooms at his disposal for the activities undertaken as the OIE-RL for ISA. The same rooms are in use for research and other diagnostic activities.” Furthermore, “The panel has serious concerns regarding the current setup in the OIE-RL. The primary concerns were the cramped, untidy conditions of the laboratories, particularly the general laboratory [room 329(S)] where both sample preparation and post-PCR analysis were performed in close proximity to each other. … The panel believes that there is a serious risk that the integrity of the test samples will be compromised.” Finally, “Conclusions of the audit were unfavourable and showed that a series of weaknesses in the system have a direct impact on the quality of diagnoses conducted by the OIE Reference Laboratory at AVC.”

OIE accreditation was subsequently removed from Kibenge’s lab, severely compromising the credibility of the diagnostic results produced there.

3.) Lack of repeatability in the results

When extraordinary results are found, like the one claimed by this study, the typical course of action by most investigators is to ensure they are valid by several levels of confirmation, including contracting an outside laboratory to replicate the results. However, these authors did not seek outside confirmation.

In fact, they couldn’t even consistently repeat their own results.

All of their samples (n = 1,106) were tested by qPCR; all samples with positive results with at least one replicate (n = 20) and a subset with negative qPCR results were also tested by conventional PCR.  Of the 10 samples that were positive by qPCR with both replicates, only one sample was also positive by both conventional assays (= 10% of the 10 qPCR positives, and 1/1,106 samples tested = 0.09%).  Only 9% (6/65) of samples that tested positive by conventional PCR were also positive by qPCR, even though the samples tested by conventional PCR were biased towards fish that were already positive by qPCR.  Only 2% (1/50) of the samples positive by segment 8 conventional PCR were also positive by qPCR.

4.) Incorrect assumptions

One of the biggest claims made by the paper, is that a novel mutation is limiting the ability of labs to detect the virus using current techniques. However, according to Marty, this just doesn’t add up.

“The paper reports 10 positive qPCR tests from 1,106 samples tested using the same PCR primers and probe that we have used since October 2014”, said Marty, “If the virus were truly present in the samples that we were testing, we would be detecting it at least as often as Dr. Kibenge’s laboratory; the fact that we are not detecting ISAV in our samples decreases my confidence in the results reported in Kibenge et al. (2016).”

“As part of the BC Fish Health Audit and Surveillance Program, the BC Animal Health Centre tests moribund and recently dead fish from salmon farms.  Since October 2014, samples from 930 fish were tested, and all results were negative - no virus.  Before October 2014, we used different PCR tests for ISAV that did not rely on detecting the reported mutation in the probe sequence; from 2003 through the third quarter of 2014 (n = 6,980), the test results were the same: all negative - no virus.

“Our veterinary virologist, Dr Tomy Joseph, is confident that a single mutation in the probe sequence would not significantly affect our ability to detect levels of virus that might be causing disease”, said Marty.

5.) Contradiction and blatant misinformation

Throughout the paper, the authors offer contradictory statements to that of their press release. For example, in the paper they say “our results further suggest that ISAV- HPR? strains can be present without clinical disease in farmed fish”, yet in their press release, they say “the risk of an outbreak has the potential for severe consequences in BC, the Northwestern United States and Alaska”, “ISA virus is a serious matter”, “concern that the pathogen release from the open net farming industry is far more serious than anyone knew”, and “tragic consequences”.

Furthermore, whether intentional or not, the authors blatantly lie about the facts. They state in the conclusion of the paper that the CFIA completed two years of ISAv surveillance in BC, but they did not test Atlantic farmed salmon…”

This couldn’t be farther from the truth. The CFIA 2014-2015 Annual Report on Aquatic Animal Health, provides the active surveillance activities that have been conducted during the fiscal year 2014-2015.

“British Columbia farmed Atlantic salmon (Salmo salar) were sampled for Infectious Salmon Anaemia virus (ISAv), to increase confidence in the absence of non-pathogenic ISAV in British Columbia. By the end of March 2015, 3312 samples had been tested and found to be negative for ISAV. One more year of active surveillance of farmed Atlantic salmon is planned to be conducted for this agent.”

Of this astoundingly high number of tested fish, not a single one tested positive for the virus: “In 2014, 3312 Atlantic salmon were tested for ISAV. All samples have been negative, increasing our confidence in disease freedom.”

The report goes on to say: “ISAV has never been confirmed in British Columbia. This data from this survey is used to allow the maintenance of strong import controls, and supports export certification. It will be used to support the declaration of British Columbia as free from this agent.”

And these results from BC confirm the same results from Washington, Alaska, and Oregon.

The authors try to justify their methods of sampling from fish obtained in markets. In a scientific study, they think it is acceptable to use uncontrolled, contaminated fish from unknown origins. They say in their press release that: “The researchers were not allowed access to Atlantic salmon from farms for testing and so all farmed salmon samples came from markets in British Columbia.”  However, according to the people actually growing the Atlantic salmon, they were never contacted or requested for samples.

Perhaps the biggest contradiction in this story is how the authors go as far as to implicate ISAv as a reason for the decline of certain Pacific salmon populations. “Detection of the ISA virus was three-fold greater in farmed than wild salmon, but European ISA virus genetic sequence was detected in 72% of the cutthroat trout that reside in Cultus Lake, home to Canada’s most endangered Fraser River sockeye salmon population…”.

Instead of referring to the mounting body of evidence that continues to point to drastic challenges in local environment as a major source of marine mortality for juvenile salmonids, the authors insist that detection of this virus is likely the cause for severe populations declines for sockeye salmon.

Based on results from extensive monitoring over the past decade for ISA and other diseases in BC farm salmon, Marty emphasized: “I am confident that these unconfirmed test results are not a threat to wild or farmed salmon.”

“The unconfirmed PCR test results reported by Kibenge et al. (2016) will not change our approach to rigorous science-based sampling and testing hundreds of farm salmon every year for diseases, including ISA.”

Jeremy Dunn, the Executive Director for the BCSFA, echoed the concerns by Dr Marty, and spoke for the industry in BC: "For over four years Ms Morton has been reporting positive PCR tests for ISAV from BC samples, none of these results have been confirmed by the CFIA using OIEA standards. We have great concerns about the methodology, and the ethics of the researchers involved given their history of reporting false positives with respect to ISA. None of the results reported in this paper have been confirmed by an outside laboratory.”

“Furthermore”, Dunn explained, “ISA has never been detected in fish on the West Coast of North America. This has been confirmed through thousands of tests by CFIA, as well as thousands of tests by authorities in Washington State, Oregon, and Alaska. Farm-raised salmon in BC are healthy and have never shown signs of sickness from ISA. This report claims to find an ISA sequence, but the researchers admit they were unable to verify it using necessary, globally standard follow-up tests. Their study also confirms they found no evidence of the ISA disease in BC Fish".

“We have tested thousands of samples using methods accredited by independent organizations.  Our results are also consistent with several American organizations, and CFIA.  The results of the Kibenge et al. (2016) paper are contrary to all of these results. These contrary conclusions require very strong evidence to be accepted by the broader scientific community.  The Kibenge et al. (2016) paper does not provide that evidence.”

A request to speak to Dr Kibenge was declined, while all scientific questions were deferred to another one of the co-authors, the professional activist and well-known anti-salmon farming zealot, Alexandra Morton.