September 29, 2023

The need-to-know this morning

Structure Therapeutics said its oral GLP-1 medicine led to roughly 5% weight loss after four weeks in a Phase 1 trial.
Madrigal Pharmaceuticals, a company awaiting FDA approval of its treatment for NASH, raised $500 million in a stock offering.

podcast

When is negative data positive enough?

Who gets credit for scientific breakthroughs? And wasn't biotech supposed to bounce back? We cover all that and more this week on "The Readout LOUD," STAT's biotech podcast.

Our colleague Elaine Chen joins us to explain the story of Svetlana Mojsov, a chemist who played a vital role in the discovery of GLP-1 who has spent decades fighting for proper recognition. We also discuss the latest news in the life sciences, including the FDA hearing on a BrainStorm Cell Therapeutics' polarizing ALS treatment, the end of the road for a once-vaunted drug developer, and the sorry state of biotech stocks.

Listen here.

gene therapy

What to do with Taysha's languishing gene therapies?

The tide's unequivocally turned for Taysha Gene Therapies, a once well-funded biotech that is now running out of options. Back in 2020, the company raised more than $300 million, with the intention of licensing and developing gene therapies for at least 18 different rare neurological diseases. But the company has since shelved nearly every one of these programs, and last week, another one was abruptly stopped.

Now, Taysha is developing a single gene therapy, which targets a far more prevalent (but still rare) disease called Rett syndrome. In today's biotech climate, "I think the world of bringing 16 to 18 programs simultaneously is just unrealistic," Taysha CEO Sean Nolan told STAT.

Now, the company is trying to figure out how to offload the remaining drugs, many of which are still in the preclinical stage. Some of the families who have supported Taysha's efforts suggest that the company give the programs to the family foundations and let them determine the best path forward.

artificial intelligence

Will AI-driven data analysis give us what we need?

Code Interpreter, the artificially intelligent ChatGPT plug-in tool that writes and runs its own Python code to churn and process data, could be a bit problematic. It's not so much that AI-driven statistical analyses done on a massive scale could be prone to error — but that's part of the problem. The issue, opines metascience expert Jordan Dworkin of the Federation of American Scientists, lies more in whether or not queries completed by an ever-obsequious AI bot will actually help generate real knowledge.

"Without broader shifts in norms and incentives, the speed offered by AI tools could threaten the potential value of some of biomedical science's most promising new offerings, like preprints and large public datasets," he writes. The concern, then, is that knowledge might end up coming from meta analyses that are also AI-generated, and end up missing true expert judgment in terms of choosing what topics are worth studying in the first place.

medicare

Physician payments from drug industry cost Medicare millions

Ophthalmologists who got money from drug companies were less likely to prescribe cheaper medicines for age-related macular degeneration. This caused Medicare to pay an extra $643 million over a six-year period, a new study shows.

Specifically, the doctors who were courted by pharma would prescribe Avastin, an effective but older cancer medicine used off-label to treat macular degeneration, about 28% of the time. They'd choose costlier drugs approved specifically to treat the eye disease 72% of the time.

In contrast, physicians who didn't receive any payments would prescribe Avastin 46% of the time — nearly twice as often as those who had received payments from the drug industry. On average, Avastin cost $70.86 per Medicare claim for macular degeneration. In contrast, Lucentis cost $333.55 and Eylea cost $921.65.