…and how can we lower that number?

How much R&D budget is required to finance a single approved drug?

Well… ask ten different people and you’ll get ten different answers.

Answers will vary but all ten numbers will have one thing in common…they will all be enormous.

Many analysts have published their cost models to arrive at a best estimate that encompasses all R&D costs to produce a single FDA approved medicine.

In terms of accounting detail, I like this 2010 analysis published by Paul Stevens.

Paul, Steven M., et al. “How to improve R&D productivity: the pharmaceutical industry’s grand challenge.” Nature reviews Drug discovery 9.3 (2010): 203-214.

I re-arranged Paul’s key figure to create the image below.

The author has assigned an out of pocket cost to each step of the R&D process, then he multiplied the number by an 11% cost of capital. The final tally is 1.7B in costs to get one drug launched into commercial markets.

Now, these numbers are from a faraway land called 2010. Prices have gone up since then, and furthermore every cost analysis model is biased by the parameters of the model.

Another (more simple) approach is to estimate the total current R&D spend in the U.S. and then divide this cost by the average number of drugs approved annually.

This is a more course methodology, but it quickly gets you to an approximation.

When you tally up total R&D budgets from pharma/biotech based in the U.S., you end up with roughly 130B. If you average the annual approved drugs from the past 10 years, we see about 45 new medicines launched per year. Divide that up… and we arrive at 2.9B per new medicine.

Okay, so this is more than I have in my checking account.

Where do we go from here?

Since the 1950s, we have seen a persistent trend of fewer approved medicines per 1B dollars in spending. Jack Scannell coined the phrase, ‘Eroom’s Law’ for this phenomenon, since it moves in the opposite direction of Moore’s Law. This is one of the fundamental differences between Biotech and Tech.

After chatting with Jack and reading his papers on Eroom’s Law, I’ve paraphrased the following reasons to explain why we keep slipping down Eroom’s slide.

I’ve worked on the lab bench side of early drug discovery for 20 years and worked on the industry/biotech side for 10 years… and these all resonate with me.

On an optimistic note, if we slid down this slide, then surely we can grab a hold of the rails and army crawl our way back up. Right?

Yes, of course.

But how?

Well, I’m not sure, but I have some ideas. And you will be pleased to note that only one of my ideas is ‘AI will fix it’.

Here are four areas to focus on to improve R&D efficiency…

As you can see, my main theme with these ideas is to avoid the expensive, late-stage clinical costs.

My goal for the year is to make headway on the first topic. I’m using an AI platform to arrive at new drug targets for unmet medical needs. The premise here being that higher quality drug targets lead to candidate drugs with a higher chance of R&D success.

I’m using an AI platform called PandaOmics, which is designed by a company named In Silico. If you have an interest in exploring PandaOmics, you can use this link to set up a chat with their people. I’m two months into using their software for drug discovery and am deeply impressed. They generate drug target candidates by integrating multi-omic biological data (genomic, RNA expression, proteomic) and combine with LLM text mining from published literature and clinical trial data.

It’s a brave new world… and I am trying to stay on top of it.

Talk soon,

Kevin

I teach industry seminars for the biotech industry. If you need an educational component to your organization, then please reach out. Contact info. below.

Kevin Curran PhD

Rising Tide Biology

www.risingtidebio.com

kevin@risingtidebio.com

If you find this newsletter interesting…then please use link below to share with colleagues.