Canine parvovirus infection - role for interferons?

We are all familiar with the potential for canine parvovirus (CPV) to cause severe morbidity and mortality. Treatment is symptomatic and supportive and antibiotics are used in many/all cases (there is more to say about this but that is for another blog some other time!). The ability to provide good intensive care can significantly impact on outcome but nevertheless some dogs will die. None of the medical therapies currently used are expected to have any effect on modulating or dampening down the actual pathology of the disease (although early enteral nutrition likely helps to ameliorate intestinal changes to some degree). A treatment that could reduce disease severity and even lower mortality is therefore something that is undoubtedly very desirable but does such an option exist?

Canine interferons are not to my knowledge currently available for therapeutic purposes but recombinant feline interferon- ω (rFeIFN-ω) has been used in the management of cats with various viral disorders and has also been explored in dogs. Interferons may have multiple mechanisms of action in viral diseases for example modulating cellular and immune functions and impairing virus replication. Recombinant feline interferon- ω is produced in silkworm larvae using a baculovirus vector.

I have been aware for probably at least ten years or so of the potential use of rFeIFN-ω in the treatment of CPV infection in dogs. However it is not something I have ever done. I remember this subject being discussed during one of our journal clubs when I was an ECC resident (so between 2006-2009) and I remember the conclusion of that discussion being that it was probably better not to spend the money on this treatment but to use it instead to provide more intensive and/or more prolonged hospitalisation. And that as they say was that. I have never revisited the debate. This week I read something in which the use of rFeIFN-ω in CPV infection was mentioned and the way it was presented – albeit it only in a couple of short sentences – would lead one to ask the question why aren’t we/more people using it? So I went to see what I could find in the literature…as far as I can tell there are 4 papers that get referenced on this subject and in particular the one by de Mari et al (2003). As always, I have just selected some bits to mention:

Ishiwata K, Minagawa K, Kajimoto T. Clinical effects of the recombinant feline interferon-omega on experimental parvovirus infection in beagle dogs. J Vet Medical Sci 1998. 60:911-917.

3-4 month old Specific pathogen free experimental Beagles

Oral inoculation with CPV-2a (isolated from a Japanese dog in 1991)

Diagnosis of clinical infection: appropriate clinical signs + seroconversion + viral DNA in faeces by PCR (a similar approach to confirming diagnosis was adopted in all papers included in this blog post)

3 different experiments but data was pooled and upshot was 12 dogs treated with IFN and 17 dogs with placebo (saline)

Used a scoring system to score clinical signs with respect to pyrexia, vomiting, appetite, faecal condition

Their main conclusion was that treatment with IFN reduced the severity of clinical signs (no effect though on faecal shedding of viral DNA or on seroconversion). One dog died in the placebo group, none in the IFN group, so any effect on mortality clearly could not be evaluated.

Prospective study which is good but some of the obvious shortcomings of this study from a clinical evidence-based medicine point-of-view include:

  • Very small sample size; single breed; single age group
  • Experimental infection of SPF dogs rather than natural field infection
  • No mention of randomisation or blinding
  • No mention of any medical therapy being provided to the dogs

The fact that only one dog in the placebo group died seems very low and may suggest low virulence of the inoculated strain. I suppose one could argue that if this study is interpreted as showing that the use of IFN did indeed reduce the severity of clinical signs, its beneficial effect may be even greater in infections with more virulent strains, more severe signs and greater mortality.

Minagawa T, Ishiwata K, Kajimoto T. Feline interferon-omega treatment of canine parvovirus infection. Vet Microbiol 1999. 69:51-53.

4 month old experimental Beagle dogs

36 dogs in total inoculated with a CPV strain which had been confirmed as pathogenic in a separate preliminary study

19 dogs received IFN (this was made up of two separate groups receiving a lower versus higher IFN dose)

17 control dogs received saline placebo

Very little detail provided but they concluded that IFN treatment resulted in less severe signs, improvement in existing signs, and earlier recovery

In this paper the authors also report the results of ‘field’ trials conducted in veterinary practices in Japan:

93 dogs in total

72 dogs treated with IFN: mortality 19.4%

21 dogs treated with saline placebo: mortality 61.9%

Conventional therapy including fluid therapy was also provided at clinician discretion but no further details are given in the paper

The authors conclude that in this field trial using IFN led to significantly reduced mortality, and that it also led to less severe clinical signs and improvement in existing signs

Some of the comments in the first paper with respect to the quality of evidence also apply to these studies in this paper. On the face of it the difference in mortality rates in the field trial is striking however the number of dogs in the control group was considerably smaller, there is no information provided about the medical therapy offered to any of the dogs, no mention of what sample size would be needed to demonstrate a survival benefit (no power calculation)….any beneficial effect on mortality if present will very likely be much smaller than if we took these data as presented at face value.

Martin V, Najbar W, Gueguen S, et al. Treatment of canine parvoviral enteritis with interferon-omega in a placebo-controlled challenge trial. Vet Microbiol 2002. 89:115-127.

Ten experimental Beagle dogs 8- 9 weeks old

Randomly assigned, double-blinded

5 dogs treated with IFN: 1 died

5 dogs placebo-controlled: all 5 died

Subcutaneous fluids were provided as needed

The authors report that using IFN led to a statistically significant positive difference with respect to clinical signs, deterioration, loss of body weight, temperature

De Mari K, Maynard L, Eun HM, LeBreu B. Treatment of canine parvoviral enteritis with interferon-omega in a placebo-controlled field trial. Vet Rec 2003. 152:105-108.

Field study with natural infection

92 dogs in total; 32 breeds; 1-28 months old; 1-30 kg

19 French practices

Randomised, double-blinded, placebo-controlled

43 dogs treated with IFN: 3 died (7%)

49 dogs treated with placebo: 14 died (28.6%)

Medical treatments at clinician discretion* but steroids could not be given and the dogs could not be vaccinated during the study period

Used a scoring system to categorise clinical signs

There were no statistically significant differences between the two groups with respect to a variety of parameters at the time the treatment protocol began suggesting randomisation had been successful.

Their main conclusion was that the use of IFN led to a statistically significant reduction in mortality; it also led to a more rapid improvement in clinical signs (although there were no statistically significant differences in the treatment provided to survivors in either group).

(* Although on the one hand this almost certainly led to some lack of consistency and standardisation in terms of accounting for other factors that may have influenced progression/outcome, it is realistic to real life practice. ‘Dirty’ but realistic data?)

My thoughts:

SO where does all this leave us? Well, my novice reading of the evidence as it exists thus far is that it may well be that there is a clinical benefit to using rFeIFN-ω in dogs clinical for CPV infection. However we need more and better evidence. The de Mari et al study was a prospective double-blinded randomised controlled trial in clinical patients with field infections which is at least an attempt at better methodology than many veterinary studies. Obviously the 19 different practices, each with very small numbers presumably, may confound findings. More similar studies, perhaps with better standardisation of other medical therapies being provided to the patients, would be good. With sample sizes that are large enough to show an effect. Using a primary end-point of survival and secondary end-points for example of severity of clinical signs, presence/severity of neutropenia etc.; perhaps a standardised scoring system for clinical signs that could be used by different investigators would be useful when it comes to performing meta-analyses. Some other things that we might need to clarify are what dosing protocol to use and whether the same effects are seen with CPV-1 vs. CPV-2a vs. CPV-2b…and whether these effects are consistent in different parts of the world if there are geographical differences in strains etc.

One of the points made in all the papers mentioned above was that as far as the investigators could tell there were no significant adverse effects associated with rFeIFN-ω use in any of the treated dogs. As such if we consider the risk/harm-benefit assessment for using this treatment in dogs with parvovirus, it is unclear but there could be some benefit, and it is unclear but it is very likely that there is no significant risk/harm. When considered in those terms, why wouldn’t we use this treatment?

Which brings me onto my last point, the cost. Unfortunately I do not know the cost of rFeIFN-ω and moreover how much it would end up costing to treat an individual clinical patient using whatever dosing protocol is agreed. However I am going to hazard a guess and suggest that it is an expensive treatment for use in cats and clearly this expense will increase, potentially quite substantially, as we move to bigger and bigger dogs. If we are talking for example about many hundreds of pounds per dog, are we convinced enough of a beneficial effect to be comfortable spending the carer’s money on this rather than several more days of good intensive supportive care? On the other hand if using IFN leads to less severe signs and/or quicker improvement in signs, would at least some of the expense be returned in savings from shorter and less intensive hospitalisation? What if expense is not an issue, we should give it a go right? Or no?

I am not aware of how available rFeIFN-ω is and whether increasing use for CPV infection in dogs would pose any challenges in terms of stocks but this may be something else we need to keep in mind.

The de Mari et al paper was published in 2003 and unless I have missed one or more, none have been published since then. Why is this? Is it a financial thing preventing people from investigating this? A lack of product availability? Or what?

As always, love to hear any thoughts/comments here, on Facebook, or wherever. If you routinely use rFeIFN-ω in dogs with CPV infection, I would love to hear from you. Likewise if anyone has any details on the cost of rFeIFN-ω therapy, that would be great.