Tracting the intractable

There is a tendency to view drug design as ‘rational’ and the market as typically ‘irrational’…?

Clearly we run into?definitional challenges?as soon as we examine the words too hard, but there is opportunity in bridging?rational?drug design with?irrational?choosers (the ‘false Development world vs real Commercial world’ challenge?I wrote about here). The opportunity remains to define unmet need not solely by biology or chemistry but by listening harder to people who prescribe, buy and use our drugs. Perhaps, therefore, this is not an intractable problem.

The word ‘intractable’, however, is fascinating.?

In the ‘real world’, intractable means ‘not easily governed, managed, or directed’. In medicine, intractable means ‘unstoppable’. For example, intractable?diarrhea?is diarrhea that can't be stopped, even with medication, and intractable pain is pain that can't be stopped, even with medication.

In the world of?computer science,?intractable problems?are problems for which there exist no?efficient?algorithms to solve them. It is a kind of problem that takes the time it takes to solve, going through solutions one by one - the ‘brute force’ solution, such as finding prime numbers. (It doesn’t mean ‘undecidable’, as covered neatly?here.)

The computer science definition is perhaps the more relevant usage for asymmetric learning. If the problem ‘would molecule?x?work in disease?y?’ could only be understood by putting molecule?x?into disease?y?and watching, and waiting, then that would be an intractable problem. The race would then simply be for those who go first. But, we don’t believe that. We do believe that we can get a clue: smaller studies, translational models, predictive models and more, are all to gain those clues.?

So, the question is: can we make the process of gaining clues more efficient, more tractable? Our task is simple: we need to find out where a drug will work, and who would use it. That is, we have to come from both ends in - like revealing the sculpture in the stone.

“The sculpture is already complete within the marble block, before I start my work. It is already there, I just have to chisel away the superfluous material.” Michelangelo Buonarroti

The drug will do what the drug will do. Our competitive advantage comes from finding out where it will work faster than the others do. Of course, the thing the drug?does?is likely to be of interest in a whole range of diseases. Among that range may well sit the opportunity or opportunities for a commercial proposition. But, we need to recognise that we’re coming at it from two different directions - it is not just an ‘if’ problem, but a ‘then’ problem - the evaluation of unmet need in a whole range of diseases is more complicated than assuming a single destination, but it does increase the probability of finding a suitable destination. We’ve introduced complexity, but also opportunity for a faster learning process - we are no longer relying only on the rational science of receptor binding and in vitro ‘omics’, but on the ‘irrational’ behaviours of customers.

Feeding unmet need into the process, rather than assuming it is there if we ‘follow the science’ gives us a more iterative?opportunity-seeking?culture. It comes closer to a game of skill (like poker) rather than a game of luck or brute force - clues we have now can be judged against a range of possible outcomes, rather than one fixed destination.