Two simple reasons why experiments on lab mouse do not adequately recreate physiologically relevant human disease models
These days we frequently come across of discussions about relevance or non-relevance of laboratory mouse models to study human diseases. Mouse models have their supporters and opponents.
Usually, those in favor say that it’s the cheapest model at hand. However, this is not exactly true. Genetically modified mice that actually represent the main attractiveness of using mouse models can easily cost $200 per one male and female pair. And we are talking about the simplest genetic modifications. More fancy mutants can cost $500 or above apiece.
In the US, the FDA’s position is important to consider. FDA’s Investigational New Drug (IND) application requires incorporation of pre-clinical studies (related to proof of concept and toxicology). However, except mice and rats (and some birds), all other mammals are USDA covered species and hence fall under more complex guidelines. For example, USDA covered species include non-human primates, dogs, cats, guinea pigs, hamsters, rabbits, and any other warm-blooded animal. So, basically USDA and FDA regulations naturally make mouse the main experimental target animal for drug testing or academic laboratory manipulation.
Now, those who oppose mouse experimentation say that mouse models are waste and inhumane since they do not adequately represent human disease models. There are sufficient data to make such conclusion. Without going into detailed analysis, I want to suggest two simple reasons that many people haven’t even heard of.
1. Laboratory mice in any university or research organizations are kept in plastic cages at room temperature, around 22-25℃ (72-77℉). However, this is sub-optimal temperature for mouse whose natural affinity is for higher temperature (30℃). Basically lab mice are constantly feeling a little “cold”. And we know enough that this can affect their physiology, for example development of insulin sensitivity, obesity, type 2 immunity, overall inflammatory response.
2. Laboratory mice are fed only specially formulated dry pellets and water. Nothing else. However, in natural environment, mice eat almost everything including vegetables and fruits. Now, such non-natural food composition for lab mice affects their GI tract physiology and of course their microbiota. Research articles are full of protocols where mice are treated with antibiotic and combination of different antibiotics. It appears that this type of antibiotic treatment does not affect lab mice behavior such as their appetite and does not induce loss of weight or other GI tract disturbance that one would expect to happen from antibiotic treatment in ordinary mammals, including humans. Now, would you still think that testing new antibiotics safety profile in mouse could predict any side effects in humans?
The real issue is not whether mouse model is “representative” or not, but rather the question is, if not mouse, what else should we use for pre-clinical studies?
posted by David Usharauli
These days we frequently hear that precision medicine will change how patients are treated. It is true that precision medicine, alternatively called personalized medicine, will change how diseases are diagnosed and treatments prescribed. However very rarely such reports mention that this near future medicine would be so prohibitively expensive that only limited number of individuals would have sufficient financial resources to benefit from it.
Let me explain my point. In the 20th century, the pharmacological drugs mainly consisted of chemically synthesized small molecules. These drugs would typically target small, conservative regions of the enzymes or receptors to mediate their pharmacological effects (aspirin, β2-blockers, etc). Luckily, big proportion of human population was responsive to such small drugs, though existence of protein variants (alleles) among ethnic groups or sometimes just differences between male and female metabolism affected their relative effectiveness.
One area of medicine that really suffered from generic approach to drug activity was cancer therapy. Since tumors and healthy cells frequently express similar molecules targeted by small chemical drugs, cancer therapy was associated with high level of drug side effects.
One way to overcome drug side effects is to make them more selective, more precise. Now, small molecule chemistry would not have much usefulness here since smaller the region the drug targets, the more non-selective its effect. However, selectivity could be achieved by increasing the size of region targeting by drugs. For example, monoclonal antibodies could interact with the larger area of the target protein and their use could allow more precise differentiation between tumor and healthy cells expressing the same target protein differing in just one (or more) amino acid.
Since every human potentially expresses a unique combination of alleles, application of precision medicine would require availability of pharmacogenomics data for each patients. No matter how easy the host genotyping could become in the future, it still will be more expensive that current generic approach. In addition, genomic data would reveal allele variants selectively expressed by certain ethnic groups. If these ethnic groups would not have political or social-economical resources for whatever reason, big pharma would not spend money to develop allele-selective medicine for these groups, and even if Government would provide financial incentives along the line of orphan drug legislation, it would be still very expensive (as orphan drugs are these days, costing on average between $10,000 to $350,000 per year).
Precision medicine will bring better disease management but current healthcare system would not be able to afford it. However, I think if following conditions are met, precision medicine could work as intended, at least in the US.
1. Introduction of upper limit on cost for each class of drug (based on overall performance and developmental cost).
2. Introduction of upper limit on profits for each class of drug (that would take effect after developmental cost is fully recovered. For example, 100% profit law would mean that if drug development cost was 1 Billion, then the company would be allowed to make 2 Billion in free market and afterwards drug cost would decrease step-wise (10%, each subsequent year), until it reaches manufacturing cost.
3. Increase in Medicare/Medicaid contribution and coverage.
posted by David Usharauli