What’s the Fuss About?
In May of 2023, the FDA approved a new treatment for osteoarthritis (OA) pain in dogs. This was Librela (bedinvetmab). A monoclonal antibody that targets nerve growth factor (NGF), Librela was a completely new category of treatment for OA pain in dogs. This was greeted with significant excitement because, although there are many treatments used for this condition (e.g. 1, 2), all of them have their pros and cons, and it is not always possible to adequately maintain comfort and function in every dog with existing tools.
That excitement has subsequently been diminished by emerging concerns about the safety of Librela. Reports of adverse effects to the FDA by owners and veterinarians, negative media coverage, formation of anti-Librela groupsonline, and even a lawsuit against the manufacturer have all left dog owners confused and frightened about the effects of this drug. So what is actually happening, and what should vets and dog owners do?
To begin with, we should consider the context. Osteoarthritis is a common and serious condition that causes significant pain and disability in aging dogs and can be a reason for euthanasia due to diminished quality of life. As mentioned, there are many therapies available for OA. Some are well-understood and there is strong scientific evidence to support their use, such as non-steroidal anti-inflammatory drugs (NSAIDs. See 3, 4, 5). The value of others is less clear (e.g. glucosamine, gabapentin, acupuncture). Other factors, such as obesity and activity level, also play a role in the management of OA.
No one treatment alone is going to be a perfect solution for all patients, and the best approach will vary with circumstances., The most effective and best studied pain relievers are NSAIDs. While these drugs do have potential risks, they are well understood and the vast majority of dogs with OA can benefit from them safely. However, there seems to be great anxiety among pet owners and even vets about NSAIDS, and an unfortunate reluctance to use them appropriately. This can lead to inadequate treatment of pain in individual dogs, or to the use of dubious or ineffective treatments. This reluctance is also part of why Librela was greeted with such enthusiasm when it was introduced. There was a hope, encouraged by the manufacturer, that this would be a safer alternative first-line treatment than NSAIDs.
Where We Started
So was such hope justified by the scientific evidence when Librela was first introduced? Well, sort of.
The initial data used to gain FDA approval did show evidence of benefit. About 45% to 48% of treated dogs were judged as responding well, which was more than the improvement seen in the placebo groups (the placebo effect, contrary to what a lot of people believe, is common in veterinary patients and clinical research studies).
In a study done in Europe, only 17% of the placebo group showed an adequate response; considerably less than the 45% response in the Librela group. In a US study, however, the placebo response was about 36%, which is not nearly as different from the 48% response seen with Librela. Overall, the drug appeared to be effective based on the mechanism of action (biologic plausibility), the preclinical research in dogs, the response to similar drugs in humans, and the clinical studies done for Librela approval in several countries (Librela has been approved and used in many European countries since 2021)
The initial approval also included evaluations of safety. Studie in laboratory Beagles lasting form 2 weeks to 6 months were done at doses ranging from the usual clinical dose up to 10 times this amount, and no serious adverse effects were reported. Clinical dogs also reported few serious adverse effects compared with the placebo groups.
This is not, of course, a perfect or complete assessment of the possible risks of Librela. However, there is an inevitable compromise in the development of new medicines between the strength of the information produced to show safety and efficacy and the time, cost, and risk to study subjects needed to obtain that information.
A “perfect” study for determining safety would be to use the drug in thousands of dogs with arthritis, with all the complex mix of ages, breeds, medical conditions, and other treatments seen in the real world, for many years. However, this is impossible to achieve in most cases, due to logistics and cost, and requiring this would result in virtually no medicines ever being approved for use in veterinary patients.
The majority of drugs we have available in veterinary medicine are used those approved for humans and used “off-label” in our patients (see this post for a more detailed discussion of the drug approval process in veterinary medicine). We typically have a lot less information about the safety and effectiveness of these drugs than we originally had about Librela. So while the initial data were imperfect, it would be wrong to say that the approval was unusual or inappropriate.
Where We Are Now?
So what has changed? Since the US approval, there have been a lot of anecdotal reports to the FDA about bad things happening to dogs getting Librela. These have ranged from minor to severe, and they have included a tremendous variety of events. Some of the most concerning are neurologic changes and severe physical disability leading to euthanasia. This has raised the concern that some of these events may have been caused by the Librela.
Deciding whether this is actually the case or not, however, is quite a challenge. The majority of these adverse events have happened in older dogs, many of whom have medical conditions other than OA and are getting drugs or other medical treatments besides Librela. This is a population that is at high risk for all kinds of medical problems, including those we are hearing about. Determining what, if any, role Librela has in these issues isn’t as simple as just assuming any problem a dog getting Librela has is caused by the drug.
One thing the FDA has done to try and sort this out is something called disproportionality analysis. Essentially, they do statistical analysis to see if the reported problems are more common dogs getting Librela than in dogs getting other drugs, and they attempt to compare similar dogs based on age and other factors.
This is not a clear indicator that Librela is the cause, but it can raise the level of suspicion if there are a lot more instances of a particular problem than are seen in similar dogs getting other drugs. The FDA did find reason for concern in this analysis for several of the types of problems reported, inclduing some of the most serious.
The agency also looked at whether the reported problems were more common in dogs also using some other medical treatment along with Librela. Sometimes, this was the case, and they found that about 25% of the cases evaluated were also taking the medicine gabapentin. Again, this doesn’t show that there is truly a risk to combining these drugs, but it suggests this is a question that needs to be looked into.
Not surprisingly, the manufacturer, Zoetis, has tried to reassure vets and dog owners about the safety of Librela. In an email to vets, the company pointed out that despite the apparently large number of reports of adverse events, they are actually quite uncommon relative to the number of doses of the drug they have sold (the company cannot, of course, know how many doses have actually been given to how many dogs, but the sales figures are a rough estimate of this since vets don’t tend to keep a lot of unused drugs in stock). The estimated adverse event rate is less than 10 per 10,000 dogs treated by this estimate, though there is significant uncertainty behind this number.
It is also important to remember that people who have negative experiences with any drug are a lot more likely to tell others about it than those who have positive experiences, which is one of the reasons anecdotes are not a very reliable guide to the true safety of any treatment. In online discussion groups of veterinarians, there are often many anecdotes indicating vets have used the drug frequently without seeing any problems, and there are stories of dogs saved from euthanasia when they responded to Librela after failing other OA treatments. Dueling anecdotes often leaves us with more questions than answers.
We can try to make some sense of the limited data we have by looking at the way Librela works and experiences in other species. The drug reduces the amount of nerve growth factor (NGF). This is an important chemical with many roles in the body. Most critically, it is involved in development of the nervous system, which is why drugs like this should not be used in growing or pregnant animals. However, NGF is also part of the pathway that creates pain in OA, which is why reducing it can help control pain. It may also be involved in the maintenance and repair of nerve tissue and the bone and cartilage of the joints, so adverse effects regarding nerve function or bone and cartilage health might plausibly be related to the drug. However, there is not yet sufficient evidence for us to know if and when this is the case and which patients might be at greatest risk.
There has also been a lot of concern focused on the fact that similar drugs for reducing NGF have failed to be approved or have been restricted by the FDA due to concerns about adverse effects (e.g. 6, 7). The most serious problem was the development of Rapidly Progressive Osteoarthritis (RPOA), which can involve the dramatic and devastating collapse of joints. This has mostly been seen in humans taking both NSAIDs and an anti-NGF drug at the same time. It is not clear if RPOA happens in dogs or not.
While RPOA was not reported in the initial clinical studies, these involved fairly small numbers of dogs and the subjects were not allowed to take NSAIDs during the trial. Some vets have reported joint damage consistent with RPOA in dogs taking Librela, but again these haphazard case reports are not yet sufficient to know when, if, and in which dogs this problem might occur.
As always, we have to bear in mind that humans and dogs share a lot of aspects of our biology, but there are also important differences that affect how we handle drugs. Humans can chew sugarless gum with xylitol safety while this chemical easily kills dogs. Dogs can take the antibiotic enrofloxacin routinely while humans get unpleasant hallucinations from it. The evidence in humans is worth paying attention to, and it can be a clue to what might be happening in dogs, but it is not sufficient proof in itself to make a confident judgement.
What Do I Do?
The frustrating reality is that right now we don’t have a firm, final answer to how risky Librela is for each patient. The best we can do is integrate what we know with the level of uncertainty and the needs of each individual patient and make a choice now, knowing that our decision might be different in the future when there is more, better information.
Though I am involved in canine aging research, and I have even worked a bit on osteoarthritis, in my clinical practice I mostly see emergency and surgical patients, so I haven’t used much Librela myself. Many of my colleagues in the practice have used it, and anecdotally they have seen mostly positive results and few of the concerning adverse effects discussed here.
My own approach, then, is based not so much on anecdote or personal experience but on my analysis of the evidence and uncertainty at this moment. My judgement will certainly change as new information comes in, which is, after all, the whole point of evidence based medicine. For now, here’s what I tell my clients and colleagues-
- Like all drugs, Librela has both risks and benefits
There is reasonable but limited evidence for beneficial effects on OA pain, though the response rates in the published studies are less than 50%, and this may include 15%-30% placebo effects. I think it is appropriate to use for OA pain with the caveats below.
- While the risks appear quite low based on studies for approval, there are a lot of reports of more serious issues now that it is on the market. We don’t know for sure if these are actual problems with the drug, but they might be.
Anyone who says Librela is absolutely killing and disabling dogs in droves is going well beyond the data. Anyone who says Librela is a clearly safe, first-line OA drug with minimal side-effects is also going beyond the data. The uncertainty about the real risks and the factors that determine which dogs can safely use it and which can’t is enormous at this time. Because of this,
- Librela should not be the first thing we try for OA treatment.
We know a lot more about the safety and benefits of some other treatments. NSAIDs are the best first-line drug treatment, and the fears about them are excessive, irrational, and quite likely part of the reason we may have jumped too early and enthusiastically to using Librela. Some dogs truly cannot use NSAIDs due to other conditions or medications, but I routinely see dogs who could benefit from these drugs not using them because of unreasonable anxiety on the part of vets and dog owners.
There also other therapies for OA with decent supporting evidence, including weight loss and a few supplements (e.g. fish oils, green-lipped muscle). Rehabilitation (aka physical therapy) is a plausible and promising option, though we have little data as yet about what works for what cases.
There are also lots of implausible, unproven, or useless therapies out there, from CBD to glucosamine, acupuncture to cold laser, tramadol to gabapentin. While most of these are likely to have minimal risks, they don’t have strong evidence that they work either. If we use the current concern over Librela to justify greater use of such treatments, we will very likely be leaving a lot of dogs with inadequate pain control. It is also worth remembering the basic priciple of biology that there are no benefits without risks. Or, as I like to call it, McKenzie’s Law:
- In dogs that can’t use NSAIDs or other treatments with good supporting evidence, or dogs who aren’t sufficiently comfortable on these, Librela can be effective and even life-saving. I think it is worth trying in these cases, with appropriate discussion of the potential risks and the uncertainty around these, and with proper monitoring.
- Librela should probably not be given along with NSAIDs.
We don’t yet know if dogs will have the same issue of rapid cartilage destruction seen in humans. Some cases that look like this condition have been reported, but it’s too soon to know if/when/how often/to whom this happens. Most of the dogs who could safely use NSAIDs probably won’t need Librela anyway, but this limitation seems prudent right now.
Some specialists recommend x-rays of joints before and regularly during use to watch for signs of cartilage destruction. This might increase safety, but it can be expensive and may need repeated sedation, so not always practical.
- Librela probably shouldn’t be used in dogs with neurological symptoms, such as weakness or incoordination (aka ataxia). Partly this is because it won’t help if they don’t have pain or OA anyway, and a lot of the time hind limb weakness in of large-breed older dogs is due to neuromuscular disease rather than OA. However, this might also be a group at higher risk of adverse neurological effects based on both the reported adverse effects and the mechanism of action of the drug.
- I recommend regular recheck visits to look for any problems in dogs taking Librela.
I would typically recommend and exam 1 month after first dose and then ever three months for at least a few visits, and x-rays of any joints that seem to be getting worse. This is a pretty arbitrary schedule, and other approaches are also reasonable, but this level of monitoring also isn’t a bad idea for dogs with chronic musculoskeletal issues anyway, especially if we are trialing multiple treatments.
I also recommend the use of objective tools for measuring pain and disability, such as a validated pain scale or mobility tool. Placebo effects influence owners and vets alike, and these can make ineffective treatments look like they work.
Bottom Line
- Librela is likely mildly to moderately effective for OA pain, though we do not yet know which dogs will respond best and for how long.
- Librela has risks as well as benefits, but we don’t really know how common these are. We also cannot predict very accurately which dogs are most likely to have adverse effects.
- Librela should not be used in dogs with neurologic disease or taking NSAIDs
- Librela should not be the first choice for most dogs with OA (weight management and NSAIDS are the best first-line for many dogs, and there are other options with some supporting evidence if these are not possible or sufficient).
- Dogs on Librela should be closely monitored on a regular basis
Things will change. As more information is produced, we will have a better sense of how well Librela works, what the risks are, and which dogs are more or less likely to experience both benefits and risks. My hope is that we will be cautious but not hysterical in our use of this drug while more research is done. We don’t want to cause harm, but we also don’t want to abandon a potentially useful treatment without sufficient cause.