In my last biotech related post, I introduced you all to Karuna Therapeutics, a company that turned a $100k investment into $14 billion with a little bit of gumption, over a decade of clinical trials, and a couple hundred million dollars of investment1. I received a very strong response from that post, which was great, including from a couple new friends in the repurposing world. Fun times all around!
While talking to some of the people who responded to my post, I realized that I may have left some of you with the misleading impression that Karuna is an exceptional or interesting company because they made money off someone else’s molecules. That is not the case. Karuna is an exceptional company because they made a ton of money off of just two molecules which are both, at this point, generic and which Karuna simply combined. That sort of repurposing story is rare in biotech, although getting more common, including with my own company2.
But, making money off of other people’s molecules in general isn’t uncommon at all. In fact, I’d say it’s the rule rather than the exception. Nowadays, biotech is awash in small molecules, biologics, gene therapies, and anything else you can think of putting into a mouse’s body. This is why I cringe when I see yet another startup with yet another plan to apply AI or what-have-you to the discovery phase, claiming that their algorithms will allow them to discover 1000 drug candidates before their competitors find one. Even if they’re right, who cares? They can take their 1000 drug candidates and add them to the pile3.
Instead, the real way to be successful in biotech these days is to be able to pick the right candidates, and carefully shepherd them through the requisite trials with the money you have. This is a delicate balancing act, involving structuring your company around development milestones. Each milestone you achieve should make investors more convinced that your company’s drug will make billions of dollars for you or whoever you sell the drug to.
That brings me to the topic of today’s post. Viking Therapeutics is a publicly-traded company which, as of today (February 27, 2024) is worth $8 billion with fewer than 30 employees. So, you know, they’re worth pretty close to $300 million per employee. This is more than twice as much as they were worth yesterday, before they announced that their obesity therapeutic, a combination GLP-1/GIP agonist, worked better than Eli Lilly’s flagship GLP-1/GIP agonist Zepbound (a.k.a. Mounjaro) in a phase 2 trial4.
The reason why Viking Therapeutics was able to achieve this valuation with just 27 employees is because they didn’t develop any of their therapeutics themselves. They licensed all their drugs from Ligand Therapeutics, who, in turn, acquired their drugs from…somewhere. It’s hard to say exactly where, because Ligand’s entire business model is based around finding unused drugs in the patent portfolios of universities and biotech companies (especially big pharma), buying them for a small amount, and then licensing them for more. Ligand is about as tightlipped with their sourcing as Costco is, for obvious reasons.
It’s also difficult to know exactly how much Viking paid for their GLP-1/GIP agonist. Viking’s a bit shy about that. What we do know is that, for their original compounds licensed from Ligand, Viking actually didn’t pay anything up front. Instead, they entered into a complicated financing agreement with Ligand, where they gave Ligand stock, a board seat, and agreed to pay Ligand incentive payments if the molecules reached certain scientific and sales milestones (around $50-$150 million total depending on the molecule). They also agreed to pay single digit royalties if the product made it to market. So we can assume they probably had roughly the same deal for their GLP-1/GIP dual agonist, although probably with greater incentive payments, given the size of the market.
Given Viking’s mix of programs which have been variously continued and discontinued, it’s also hard to piece together exactly how much they spent on their clinical trials. Each year they’ve reported about $30-40 million in research and development costs, which, for them, are driven mostly by trials. That number has increased to about $60 million for their most recent annual report. So, it’s probably something $20-30 million for the phase 2 obesity trial that added the extra $4 billion to their stock price.
None of this is too different from Karuna’s repurposing journey. What is different is that Viking, unlike Karuna, had much more flexibility with their development journey because they basically chose their molecule out of a catalog. This gave Viking a lot more options than Karuna, but also a lot more chances to mess up.
The most important choices Viking had to make were their first choices: which indications and which molecules. This is tougher than it sounds. The biggest indications are, generally, the most competitive, especially if there’s a clear path to market (i.e. the milestones are well laid out and widely accepted). Smaller, less lucrative indications are less competitive, but, not only will you be competing for a smaller prize, you’ll be forging your own path, with the risk that potential partners will not accept your milestones.
So, Viking, back in 2014, had to choose which molecules they wanted to license from Ligand, and hope that, in the 10-ish years it’d take to get any of those molecules approved as a drug, those molecules would be recognizable as best in class for their indication or subindication5, that indication or subindication would be valuable, and, of course, those molecules would not have encountered any of the normal problems along the way, like toxicity or manufacturing issues. Viking chose 3 compounds that were ready for phase 2 trials and three that were preclinical, in a mix of major and minor indications.
Of their original 6 compounds, 10 years later, literally none of them have really panned out. One of the phase 2 compounds in a major indication (NASH) is still in phase 2 trials, but is likely ironically going to be superseded by the GLP-1 agonist revolution; and one of the preclinicals has made it to phase 1 in a rare indication, but it’s a small prize to win anyways. The rest have disappeared from Viking’s website. This is not to knock Viking’s work or thought process, but just to point out that predicting the future is really hard.
What has panned out for Viking, on the other hand, is successfully catching the GLP-1 agonist/GIP agonist waves. Biotech, like techtech, is prone to waves of sentiment, and the insane success of Ozempic and its ilk have been one of the biggest waves in a long time. Not only have they made obesity an actually treatable medical condition6, but they’ve had a ton of success treating a lot of conditions associated with obesity, like fatty liver (NASH) and chronic kidney disease.
The funny thing is that Viking caught that wave relatively late, which put them behind the 8-ball. But that brings me to the next important part of Viking’s success: execution. They first announced preclinical data for their GLP-1/GIP dual agonist in November, 2021, which was a full 5 months after Wegovy, a GLP-1 agonist, had already been approved for weight loss, and after tirzepatide (Mounjaro), Eli LIlly’s flagship GLP-1/GIP dual agonist, had already released human results. They needed to move fast.
So they did. They started with comparing preclinical weight loss results in obese mice treated with their drug vs. tirzepatide and vs. semaglutide (Wegovy), and continuing with preliminary human results by March 2023 and then confirmatory human results vs. tirzepatide by February 2024. With each step, they tried their best to show that their drug would be better than tirzepatide, including starting to develop an oral version of their drug (which, for the record, is similar to what my company is doing with our combination, but with a different best-in-class molecule. Shoot me an email if you want to hear more). This, of course, wouldn’t have been possible if they didn’t choose the right molecule, but it was important to utilize the right molecule to the greatest extent possible.
This tight timeline doesn’t leave them much time for toxicology and manufacturing, so I imagine they had some results already from Ligand, although it’s hard to say without being able to see their FDA filings. It’s also possible they just crammed those development steps in there and got lucky with how smoothly they went.
Regardless, the combination of choosing a molecule that could be best in class, an indication that was both lucrative and wasn’t totally solved by the time Viking got good human results, and executing a development program well meant that Viking was able to achieve their impressive valuation, and, probably, get acquired relatively soon. Good job on becoming billionaires, guys!
And in my last, last post, I introduced you to my vice of nostalgia, if you weren’t already familiar with it. I apologize for the inconsistency of the topics of this newsletter, and I invite anyone who feels weirded out by it to apply for a refund.
Another great example of making money off of combinations generic molecules is Axsome Therapeutics’ dextromethorphan/bupropion for major depressive disorder. Dextromethorphan you might know as the ingredient in cough syrup that teenagers use to get high. Bupropion you probably know better as Wellbutrin. Combine them together, like Axsome did, and go through a regulatory shitshow, like Axsome did, and you get a $4 billion company.
To be fair, if a startup pitches that they can discover drug candidates for “undruggable” targets, or receptors that are widely known to be biologically relevant but are difficult to develop drugs for, that is actually a legitimate pitch and something to build a company off of. But that’s not what most of these AI-for-biotech companies are pitching.
For more on the science of this, stay tuned for my upcoming article with Asimov Press
Subindication meaning that it’s a patient subset. Dividing patients into subsets can be super common in some indications, like cancer, where you’ll often see new cancer drugs marketed as the best option for patients who have prostate cancer, have been castrated, and have failed 4 treatments already. I have mixed feelings about this because I think it actually leaves room for scientific misconduct. That’s a story for another time.
Yes, I am throwing all weight loss surgeries under the bus here, as well as every weight loss therapeutic that’s involved people pooping a lot or people being physically unable to eat. The sheer number of patients who have hopped on the GLP-1 agonist train already dwarfs all of those combined by a long shot.