Rescued from deep inside a thread at Drugmonkey

In DM’s thread on a Nature article concerning the bombing of a UCLA researcher whimple asks to hear from pain researchers:

Good idea, let’s throw out pain research based on animal models altogether.

Yes, this is a possibility that some people are willing to consider, the sarcastic tone of the statement notwithstanding. The best refutation would be evidence that pain research based on animal models has directly resulted in improvements in pain management in animals (including humans as animals).

How about it pain researchers? Can you compile a quickie list of such instances? Even stipulating that the ends can be used to justify the means, if causing pain to animals is “bad”, then this research is going to need to demonstrate that it is “worth it”. There’s a lot of arguing over how bad the badness is, but not a lot arguing over how good the goodness is. In the absence of a list of pain research successes, a cynic might suppose this is because there in fact isn’t any good that has resulted from this research.

To which I responded:

“Yo Whimple, pain researcher here to address your points:
1) Go to any pain research conference and you will quickly learn that most pain researchers don’t work with animals, most work with humans. These include physicians doing clinical work, electrophysiologists doing peripheral nerve recordings, fMRI people doing what fMRIers do, psychologists looking at comorbidity, nurses looking at outcomes, etc., and the list goes on and on.
2) Then there are us basic researchers who work with cells, animal tissues, human tissues and so on. Most of this work does not involve causing any pain at all. Genetic studies are done, patch clamp recordings are made, biochemical experiments on all sorts of channels and kinase are performed all with the aim of understanding the physiology of nociceptors and pain transmission neurons. These studies have led to an exponential increase in our understanding of the channels and kinase (and other proteins) that mediate nociception.
3) There are, obviously, preclinical pain models. Some of the first ones (back in 50s and 60s) were extreme and have been discarded on ethical grounds. Most models used now cause a mild pain stimulus (certainly no worse than the shock applied in fear conditioning). Others involve long-term inflammation and largely the same as preclinical models used to study inflammation or arthritis. In other words, these models are not exclusive to pain research as they are co-opted from other areas of research on important human diseases that also involve pain. Nerve injury models are slightly different and somewhat unique to the pain area but let’s remember that this is the hardest type of human pain to treat and it leads to a terrible quality of life…
4) So what are the treatments that have resulted from this work? A) massive improvement in opiate efficacy and safer formulations based on preclinical work. B) new generation neuropathic pain meds including gabapentin, pregabalin and the clinical rationale for the use of 5HT and NE reuptake blockers and improvements on their clinical use based on real basic science. C) anti TNFalpha, anti IL-6 and anti NGF treatments either now coming out or in late trials. Some of these originated in other areas and some originated in the pain area but the rationale for the clinical use is all based on preclinical basic science. That’s just a start to a long list.
5) If you doubt the importance of basic science pain research take a trip to your local university hospital and go talk to the patients in the pain clinic.”

Whimple then returns with:

Juniorprof, good list. Personally, I think if the studies can be done on humans, they should be done on humans. My concern is that animal research slows down human research because it’s so easy to work with mice (or whatever) so people spend a lot of resources on finding effective treatments for mice and doing “basic research” studies with mice, which don’t necessarily (usually?) translate into humans. My big whatif goes something like this:

Suppose researchers were not allowed to work with animal models, but could do human subjects research with all the current legal and ethical limits on human research currently in place still in effect (cell culture work on all species still allowed). Would that slow down the development of useful human clinical treatments, speed up the development of useful human clinical treatments, or would things progress at about the same pace? If we took all the animal model researchers and said, “If you’re going to use animals, they have to be human,” would that wind up being good or bad for science from the point of view of the taxpayer funding the work and waiting for treatments and cures?

Can pain researchers (for example) say, “without studies in animal X, currently useful product Y would have been impossible to develop.” Where “impossible” really means impossible, not just more difficult or costly or slower to get working?

Is it in the taxpayers’ best interest to make animal work easier to do, or would the taxpayers be better off long-term if animal work was more difficult to do? (leaving aside the above extreme example of animal work not being allowed at all) Usually I get jumped on for asking this kind of question, but I think it’s worth discussing since the answer to this question helps determine the “goodness” side of the “animal research bad” (less than / equal to / more than) “animal research good” equation.

To which I added:
“Whimple,
First, I largely agree, I would prefer to do the work in humans too but in many cases it is simply not possible. On the other hand, we have recently developed a new treatment preclinically that we are now trying to take into a trial in humans. We never would have gotten anyone to believe us without the preclinical data but now that we have it a trial is feasible (its a synergistic drug combo wherein both drugs are approved for human use so we can proceed). From a basic science perspective the trial isn’t very interesting at all. From a treatment perspective its interesting and will certainly be exciting if it translates. From a personal perspective, its about the most exciting thing I can think of… taking treatments form the bench to the clinic is why I wanted to enter this field and I can’t wait to work with some humans!

On your other question, are there examples where the clinical never would have been possible without the preclinical the answer is also yes. My example is not a good one because I likely would have been able to instigate a trial but it would have been much more difficult without the preclinical. On the other hand, the anti-NGF treatments for chronic pain are an excellent example of where it would not have been possible. We have known for a long time that NGF is involved in preclinical pain models and in human pain. We have also known that genetic mutations in humans that block NGF signaling (mostly receptor mutations) cause a terrible disease wherein people with the mutation have profound mental retardation, total lack of pain and inability to sweat. While it was the view of most researchers that this was due to a developmental issue, it was not known if blocking NGF signaling later in life would lead to similar deficits and this made any trial on anti-NGF therapies impossible due to very serious safety issues. After decades of preclinical work it is now known that anti-NGF treatments later in life do not cause similar problems and this animal work has made the safety issue a much smaller issue. So, due to literally thousands of papers on NGF signaling in animal models we have a good idea that these treatments will not cause devastating side effects in humans and these therapies are now late in clinical development. If they gain FDA approval I expect that they will become important treatments for chronic pain disorders. That is but one of several examples wherein animal work was absolutely neccessary to develop new pain treatments.”

Perhaps we can continue the conversation here…

14 responses to “Rescued from deep inside a thread at Drugmonkey

  1. So, due to literally thousands of papers on NGF signaling in animal models we have a good idea that these treatments will not cause devastating side effects in humans and these therapies are now late in clinical development.

    On the other hand, you could have tried this in one terminally ill cancer patient with unmanageable pain and got your answer. As it is, even after thousands of papers in animals, there’s still no answer to what happens in people, until you actually use people.

    Animal research is so seductively easy that researchers get addicted to it. You can’t throw a stick without hitting a lab that has cured cancer in mice, but when are they going to the clinic? Usually never. Translating animal research to humans sucks because if you try it in humans and it doesn’t work you’ve wasted your time with the animals. People are so scared to find out their animal “model” isn’t a model at all that animals is all the wind up ever doing. I think it hurts science.

  2. On the other hand, you could have tried this in one terminally ill cancer patient with unmanageable pain and got your answer
    Hold on there big guy, we have societal and scientific standards that say that this type of practice is not okay. Moreover, what if it did nothing? Would you then conclude that it wasn’t effective for pain even though you likely had a false positive. What if it had no effect on mental function in this patient so you decided it was safe for everyone and then you gave it to a two year old and you forever stunted their intellectual growth (a very real possibility based on the preclinical work).

    You may be correct that animal experiments are easier than human ones but they are only easier when testing certain things. In the pain area human psychophysics gives you a level of information that is unparalleled in animal studies. Many pain researchers do both human and animal work. There is a good reason for that…

  3. Hold on there big guy, we have societal and scientific standards that say that this type of practice is not okay

    No, this is fine. You have informed consent from a study participant, and an experiment with the potential to benefit the participant, with the ultimate goal of benefitting this class of participant. Standard phase I clinical trial stuff really.

    Many pain researchers do both human and animal work.

    Define “many”. What fraction have IRB approval to do anything in humans?

  4. Whimple, but by your rules we would not have any preclinical data and the only available evidence to support such a claim to IRB would be that the treatment would be harmful. How could you justify to IRB that the patient would receive any benefit over the opiates that they were already taking. Then again, the patient would have a very limited option for opiates because few of the newer generation compounds used for cancer pain would have been developed. Finally, in the face of the very limited human data suggesting that anti-NGF is a bad idea who would have developed the anti-NGF treatment in the first place?

    Define “many”. What fraction have IRB approval to do anything in humans?
    I can’t possibly know this. However, of the basic pain scientists I know about 25% have done some project in humans in their career. Usually it happens in collaboration with people who do only human work so their collaborators are likely the ones with IRB approval.

  5. “Would that slow down the development of useful human clinical treatments, speed up the development of useful human clinical treatments, or would things progress at about the same pace?”

    The standard answer to Whimple’s question is that without animal research, we would have had really good iron lungs (efficient, low-powered, maybe in many beautiful colors), but no polio vaccine. It’s a cliche to use that example, but I think it illustrates the point I want to make, which is that limiting research to humans skews the solutions we can consider, because we’re bound with the requirement that we do no harm to the individual human animal we’re using in the research. Having animals, to whom we assign less autonomy and less value (and I do, without a doubt) means we can try things that might harm the individual animal but might benefit others (animals, humans, or perhaps even trees, energy production, or whatever).

    “On the other hand, you could have tried this in one terminally ill cancer patient with unmanageable pain and got your answer.”

    How in the world is that even remotely possible? What kind of study can you validate with an n of one?

  6. The best description of the efficacy of animal rights comes from “Love at Goon Park” by Deborah Blum. (For a good report of the two sides of the animal rights discussion [and good successes from both sides actually talking to each other], I also strongly recommend “The Monkey Wars” by the same author.) “Love at Goon Park” is a biography of Harry Harlow, notorious for his monkey abusive-mother/child-removal experiments, which galvanized the animal rights movement, and which are probably some of the hardest experiments to justify in all of science. But it turns out that directly because of these experiments, medical doctors changed how they handled infants (becoming more concerned with human contact). Survival rates in premature infants, newborns in the NICU, and child survival rates in hospitals in general are directly attributable to the Harlow experiments.

    There is no question that medical science on animals has historically helped human survival and well-being. Anyone who argues differently has simply not seen the data. The questions, however, are (1) are you willing to sacrifice animals for human health? (2) what can we do to reduce the number of animals used? and (3) what can we do to refine our techniques so as to make life better for our animals?

    In answer to question 1, factory farms are much less regulated and much more devastating to animal welfare than scientific labs. Shouldn’t we start there? In answer to questions 2 and 3, in fact, scientists work very hard to reduce the number of animals and to refine the experiments — in large part, because (as I noted over at DrugMonkey), it is good science. Fewer animals is less expensive. Within-animal statistics provides better power and more control of uncontrollable variables. Stressed animals react very differently from unstressed. You need to control for these variables. Scientists know this. Most scientists are actually very conscientious about their animal research. In my experience, the biggest problem with the animal rights fight is that it’s a fight. I have attended many a pro-animal rights rally and had excellent discussions with the people there.

    PS. “On the other hand, you could have tried this in one terminally ill cancer patient with unmanageable pain and got your answer.” This shows a remarkable lack of understanding of cancer. Cancer is a very complex disease with many many subtypes, each of which react differently to different pain medications. We need to stop using “we’re going to cure cancer” as a rallying cry for science. Science doesn’t work that way. It works in surprising ways – like changing the NICU because Harry Harlow discovered that children love their mothers for being mothers more than for their milk.

  7. I am an animal researcher in the addiction neurosciences. I can add my 2 cents:

    Animal models contribute enormously to understanding the pathological changes in the brain that result from exposure to abused drugs, whether it is long term or short exposur, in ways that human laboratory studies and clinical trials cannot. For example, animal models identified dopamine as a critical neurotransmitter involved with drug abuse and now we are developing treatments in humans targeting the dopaminergic system that appear to have some efficacy (e.g. aripiprazole, albeit it is not solely a D1 partial agonist). Animal models are not perfect, no model is perfect. But they are necessary because the process of addiction can be studied from the drug naive state to addiction. It is simply unethical to expose drug-naive humans to drugs of abuse. Of course, enrolling human addicts in a study can be quite informative (and necessary too), but in this case you are only studying the endgame and not the process of addiction.

    Additionally, psychiatric diseases like addiction or other diseases such as cancer, are quite heterogenous. All alcoholics, for example, are not the same alcoholic. This complexity can be addressed, in part, by using animal models where there is precise control over drug dosing, genetics and some control over behavior. Further, specific brain regions and neurotransmitter systems can be directly targeted to ask direct questions about their role in addiction. Thus, animal models are one important “tool in the tool chest” to understand addiction to ___(insert drug of choice here).

    One more thought. I’ve heard the argument before about only using humans with terminal diseases for testing new drugs/therapies. In part, treatments for HIV have progressed this way and perhaps cancer as well, which is good. But there is something unsatisfying about this as an argument against animal research because not every disease or condition is terminal. Of course, I am thinking of psychiatric disorders/diseases. Informed consent issues aside, if you suffered from a mental disorder and had been through every known treatment option to no avail, would you want to be a human test subject on a brand new molecular entity (i.e. new drug) that had never even been tested in another mammalian species, even humans? I would not.

  8. I feel I might introduce my self here. My name is Kate, I’m a newbie here, someone told me that i might find some good information here so… basically that’s why I’m here, and for any good advice i might get also… hope to have good time here

  9. nashville casino depot belterra casino golf new orleans harrah’s casino .point defiance casino le meridien port vila spa and casino las vegas casino host silver slots casino .regency casino mont online casino wiki .ultimate casino ps2 new years casino bonus jackpot junction casino hotel morton mn hotels close to monte casino .may be 7 cedars casino sequim washington .about .Come to .without or red rock casino vegas reviews pachangas casino Before All about rake casino And new years eve horseshoe casino gala maybury casino Such riande granada hotel casino with After follow bingo atlantic city casino tamarindo costa rica casino As so When oldest las vegas casino above joyland online casino brighton indian casino dating casino Other paypal casino us piggs peak flash casino commerce casino stampede casino The best latest no deposit casino bonus i’m with
    shreveport casino packages venetian resort casino macau .So because prime casino no deposit bonus was bella casino Heh, fiesta casino hotel henderson nv .above .After .because Heh, santa ana casino nm riverbend casino as a result of In il casino apartments so golden nuggets casino vodka martini casino royale seneca niagara casino address sega casino ds You could get sands casino bethlehem jobs Usualy i’m with .

  10. tks for the effort you put in here I appreciate it!

  11. Каждому Привет! Заходите и развлекитесь на Сними девочку на ночь в своем городе.

  12. Pingback: A pain drug discovery success story: anti-NGF therapy for osteoarthritis | JUNIORPROF

  13. Man .. Excellent .. Wonderful .. I’ll bookmark your blog and take the feeds additionallyI’m satisfied to find a lot of useful information right here in the post, we want develop more techniques in this regard, thank you for sharing. . . . . .

  14. Радиокомпоненты Кабель! Россия вся электрика Электрика.

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s