Sep 24, 2026 · 46 min · 11 segments
Synopsis: The drug discovery industry faces a persistent challenge: 9 out of 10 drugs entering the clinic fail, underscoring the need for a better path from biological insight to transformative…
Robert PlengeGuest
Rahul ChaturvediHost
Now I'd like to switch gears a bit and get your macro perspective on a handful of things.

So just to start off, let's, you know, for the average listener right now who doesn't live in this space at the same depth that you do, something we've talked about previously is that nine out of 10 drugs that enter the clinic end up failing, and of the ones that actually make it, only one in three is truly novel.

Talk to us about why you think that is and what actually goes wrong along the way.

I think there's now a lot of very good evidence to support why things fail, and largely it comes back to this concept, and we can talk more about causal human biology.

The drug works exactly as it was predicted in preclinical models from a pharmacology perspective, but it doesn't do what we thought that it would do in terms of human physiology.

It doesn't have the effect on a disease state or a physiologic state, so it binds, it inhibits, but it doesn't have the physiologic effect.

And when you look at why drugs fail, most things fail because of they just don't work.

And if you ask why that is, lots of different reasons, but I think it does sort of boil down to this concept of we don't fully understand human biology, and specifically causal human biology.

So not correlative human biology, not causal mouse biology or animal model biology, but causal human biology.

What had me transition from academics to industry in two thousand and twelve, two thousand thirteen was understanding human genetics and how it could actually inform on causal human biology.

This was now sort of 10 years or so into the GWAS and complex trait genetics revolution, and began to understand genetics can point you to targets and molecular mechanisms and patient populations and inform on this causal relationship.

And then there began to be, I think, data on things that I had published when I was in academics, but also others have published since that time, that if you had that human genetic support, things are two to threefold more likely to be successful medicine.

So that success rate isn't one in 10, maybe it's two in 10 or maybe it's three in 10, and I think that story is sort of proven out.

Genetics is one area, in cancer it's somatic cell mutations, longitudinal profiling, so you can follow patients and people over time and you can see something emerges, a consequence of that trait then has an influence on downstream events.

So that, I think, is another example, longitudinal profiling of causal human biology.

So the principles that I thought a lot about in terms of human genetics, those can be applied in a number of other settings as well.

But, you know, that's the primary reason why it's one out of 10 and not three, four, five out of 10, and the more that we can actually understand causal human biology and then to make a medicine match to that molecular mechanism of action, I think we'll begin to chip away at this failure rate problem.

problem.And now roughly a quarter of the entire global pipeline is aimed at 38 targets, and we've talked a little bit about novel targets, novel target pairing, and novel targets entering the pipeline fell from about 100 a year to now roughly 30 or so.

Now I'd like to switch gears a bit and get your macro perspective on a handful of things.

So just to start off, let's, you know, for the average listener right now who doesn't live in this space at the same depth that you do, something we've talked about previously is that nine out of 10 drugs that enter the clinic end up failing, and of the ones that actually make it, only one in three is truly novel.

Talk to us about why you think that is and what actually goes wrong along the way.

I think there's now a lot of very good evidence to support why things fail, and largely it comes back to this concept, and we can talk more about causal human biology.

The drug works exactly as it was predicted in preclinical models from a pharmacology perspective, but it doesn't do what we thought that it would do in terms of human physiology.

It doesn't have the effect on a disease state or a physiologic state, so it binds, it inhibits, but it doesn't have the physiologic effect.

And when you look at why drugs fail, most things fail because of they just don't work.

And if you ask why that is, lots of different reasons, but I think it does sort of boil down to this concept of we don't fully understand human biology, and specifically causal human biology.

So not correlative human biology, not causal mouse biology or animal model biology, but causal human biology.

What had me transition from academics to industry in two thousand and twelve, two thousand thirteen was understanding human genetics and how it could actually inform on causal human biology.

This was now sort of 10 years or so into the GWAS and complex trait genetics revolution, and began to understand genetics can point you to targets and molecular mechanisms and patient populations and inform on this causal relationship.

And then there began to be, I think, data on things that I had published when I was in academics, but also others have published since that time, that if you had that human genetic support, things are two to threefold more likely to be successful medicine.

So that success rate isn't one in 10, maybe it's two in 10 or maybe it's three in 10, and I think that story is sort of proven out.

Genetics is one area, in cancer it's somatic cell mutations, longitudinal profiling, so you can follow patients and people over time and you can see something emerges, a consequence of that trait then has an influence on downstream events.

So that, I think, is another example, longitudinal profiling of causal human biology.

So the principles that I thought a lot about in terms of human genetics, those can be applied in a number of other settings as well.

But, you know, that's the primary reason why it's one out of 10 and not three, four, five out of 10, and the more that we can actually understand causal human biology and then to make a medicine match to that molecular mechanism of action, I think we'll begin to chip away at this failure rate problem.

problem.And now roughly a quarter of the entire global pipeline is aimed at 38 targets, and we've talked a little bit about novel targets, novel target pairing, and novel targets entering the pipeline fell from about 100 a year to now roughly 30 or so.
The rest of this transcript — segmented and speaker-labeled, so you land on the exact moment something was said
Search every transcript — by keyword, by phrase, or by meaning, across every show Radar indexes
Trends — what is surging across podcasts, measured against its own baseline
Alerts — when a name you follow appears in a newly indexed episode
No account is needed to search Radar.