23andMe faces lawmaker grilling over DNA privacy

23andMe executives this week fielded questions from lawmakers concerned that millions of people's genetic data will be sold when the company's assets are acquired out of bankruptcy. On Monday, 27 states and the District of Columbia filed a lawsuit to prevent the sale of 23andMe's genetic data without explicit user consent, and some lawmakers wondered why that shouldn't be the policy.

As STAT's Katie Palmer reports, cofounder Anne Wojcicki and interim CEO Joe Selsavage appeared before the House Oversight Committee on Tuesday, which examined the privacy and national security risks posed by the sale. A hearing of the Senate Judiciary Committee followed on Wednesday, where Wojcicki was not present. One interesting revelation: Since 23andMe filed for bankruptcy, 1.9 million of its 15 million DNA testing customers have asked the company to delete their data.

In May, Regeneron Pharmaceuticals entered into an agreement to buy 23andMe for $256 million, but the court reopened the auction to final bids last week.

Artificial intelligence

New partnership on health care AI standards

An advocacy group for health care artificial intelligence is teaming up with the leading health industry accreditation organization in a move that may raise some eyebrows among people skeptical of industry's influence on standards.

STAT's Brittany Trang this week reported that the Coalition for Health AI (CHAI) has teamed up with the Joint Commission on the development and adoption of AI best practices and an AI certification program for health care organizations. CHAI is a 3,000-member group that includes tech titans like Microsoft and Amazon, health care giants like CVS Health and May0 Clinic, in addition to a bevy of hopeful startups. The Joint Commission accredits and certifies more than 23,000 groups including hospitals, nursing homes, labs, and more.

The deal catapults CHAI back into a position power when it comes to health AI standards. Under the Biden administration it had recruited health department regulators into its ranks and pushed the idea that CHAI-anointed "assurance labs" that certified tech could be part of the answer to the tricky task of regulating how AI is used in health care. The plan lost steam when concerns about industry capture of regulation forced the officials to step down.

pharma

Waves of layoffs at AI biotechs

From Brittany Trang: Layoffs have hit three of the most established AI biotech companies in the last several weeks.

On Tuesday, AI drug discovery company Recursion announced layoffs affecting approximately 20% of its workforce. On LinkedIn, CEO Chris Gibson attributed this to both a realignment of the company's priorities and an "increased usage of AI tools, including those built in-house," saying that the company is shifting from gathering biological data to analyzing it with AI. Notably, the company cut nearly half its pipeline last month, partially in response to poor Phase 2 clinical trial results.

Last month, Daphne Koller's company Insitro announced it was cutting 22% of staff, or around 60 people. The company recently terminated its oncology efforts and doesn't have any clinical-stage candidates yet. It hopes some will be ready for the clinic by 2026.

And in late May, Schrödinger announced that it was laying off 60 people, or about 7% of its employees. The company did not give a reason. CEO Ramy Farid told me last year that he's worried about his physics and machine learning company getting roped in with the AI hype cycle. Today Schrödinger announced the first clinical data from its in-house pipeline, which was a positive readout from its Phase 1 dose-escalation study of a treatment for people with B-cell cancers who had failed previous treatments.

Fundraises, news from Amazon and Cigna, and more

Autonomize AI raised a $28 million Series A round led by Valtruis, The Cigna Group Ventures, and Tau Ventures. Autonomize develops AI "copilots" for administrative tasks like prior authorization, medical record review, clinical research, and quality metrics reporting.
Speaking of Cigna, the insurance giant this morning announced a number of new AI tools around customer experience, including an AI assistant.
Eli Health raised a $12 million Series A round led by BDC Capital's Thrive Venture Fund to launch its at-home hormone testing system, beginning with a test for cortisol.
Fertility benefits company Progyny is the latest offering to join Amazon's portal that helps people connect with health programs covered by their insurance.
Tech focused health benefits company Capital Rx acquired health navigation company Amino Health, which will become the company's new navigation solution Judi Care. Capital Rx and Amino Health have raised $362 million and $160 million in funding respectively, according to Pitchbook. This is the latest merger in the benefits navigation space and follows the high-profile merger of Transcarent and Accolade earlier this year.

policy

The FDA on its hopes for AI

On Wednesday, FDA commissioner Marty Makary and CBER director Vinay Prasad outlined "Priorities for a New FDA" in JAMA. In the "Unleashing AI" section, they said the agency believes "AI can make a first-pass review of documentation received by the FDA as part of an application that often exceeds 500,000 pages and aid in generating standardized tables." The FDA last week rolled out a new AI tool named Elsa.

Medicine

AMA releases positions on physician tracking, AI

The American Medical Association made a pair of notable tech proclamations at its annual meeting this week.

First, the organization advocated for guardrails around the collection of "personal and biological data" to ascertain clinician well being and levels of burnout. Essentially, it calls for consent, minimizing data collection to what is relevant and necessary, and that the data be used only for the intended purpose. Clinician burnout is a hot topic in health care, and this document suggests wide-scale data collection to address the problem. I can understand that doctors might be worried the data might be used for other purposes, like — and I'm riffing here — assessing performance or fitness to work.
Second, AMA adopted a new policy calling for transparency around AI tools used in clinical care. In particular, the position says tools should be "explainable" meaning that they should "provide explanations behind their outputs that physicians, and other qualified humans, can access to interpret and act on when deciding on the best possible care for their patients." This is not a new idea, but it's interesting to see Big Doctor pushing for it.

More around STAT

Check out more exclusive coverage with a STAT+ subscription

Read premium in-depth biotech, pharma, policy, and life science coverage and analysis with all of our STAT+ articles.

What we're reading

'Uber for getting off antidepressants' launches in the US, Wired
Why autism researchers aren't convinced NIH can keep science and politics separate, STAT