Sponsored by

Looking for unbiased, fact-based news? Join 1440 today.

Join over 4 million Americans who start their day with 1440 – your daily digest for unbiased, fact-centric news. From politics to sports, we cover it all by analyzing over 100 sources. Our concise, 5-minute read lands in your inbox each morning at no cost. Experience news without the noise; let 1440 help you make up your own mind. Sign up now and invite your friends and family to be part of the informed.

Subscribe to 1440 today.

In this week’s newsletter

✅ Cellscapes and synthoids. ✅ A great result - for mice. ✅ New milestone for Insilico’s AI drug ✅ Right to try gets practical. ✅ Giving AI the data it needs. ✅ How axolotls regenerate limbs.  ✅ One step closer to “living reviews”. ✅ Biology’s got reinforcement.

🤙🏼 Want to connect? Add me on LinkedIn. 🙏🏼 Not subscribed to the LEVITY podcast on Youtube yet? Do it here. 🎧 More of a listener? The LEVITY podcast is also available on Spotify, Apple Podcasts and other places.

🏡 Some housekeeping before we get started:

• We’ve got a very special episode of the LEVITY podcast coming up in just a few days (June 17, 6pm CET) featuring Daniel Ives. This is likely Daniel’s first major interview since the publication of a paper where his company, Shift Bioscience, is detailing the discovery of a new and very exciting rejuvenation gene. Make sure you’re subscribed to the LEVITY podcast - you won’t want to miss this one! Also check out my interview with Daniel from last year:

• LEVITY is about to launch its very first sweepstakes contest - and we’ve partnered with a brand whose products I’m genuinely a big fan of. The prizes? They’re going to be really good.

  Keep an eye out over the next month or so - all the details on how to enter (and boost your chances of winning) will be shared right here in the newsletter and on the podcast.

• It’s been three years since my last book came out, and lately, the itch to write another one has grown stronger. I already have a title in mind and, even better, there’s some early interest from the publishing world. Will it be related to longevity? Absolutely. And this time, it will be my first book written in English.

• I had a longer piece ready as the main feature for this newsletter, but I’ve decided to hold onto it for now - with vacation coming up in late July, I’m trying to build up a bit of material in advance so I can keep the newsletters coming regularly while I’m away. In the meantime, as you’ll see below, there’s still plenty to dig into.

News from around the longevity and health space.

Cellscapes and synthoids

Consider still images, motion pictures and video games. The first is a snapshot of an event, the second a story about an event and the third is interacting with an event. For a long time cell biology has mostly been stuck in the still image phase - capturing static moments of cellular life rather than understanding the dynamic stories unfolding within our bodies. But what if we could reach the video game level, where we don't just observe but actually design and control the action?

That's the vision behind Cellscapes, a 10-year initiative launched by Seattle's Allen Institute that aims to create a new mathematical language for describing how human cells move, communicate and organize themselves into tissues and organs.

The team also plans to build “synthoids” - custom-designed communities of cells whose behaviors can be manipulated and programmed to test how cells make decisions and organize into tissues.

A great result - for mice

Anytime a mouse study shows lifespan extension that surpasses rapamycin, it’s worth stopping to take notice. In this case, a combination of rapamycin and the anti-cancer drug trametinib extended mouse lifespan by about 30% - outperforming rapamycin alone.

That said, I’m a bit surprised by how much attention this result has received. There are plenty of caveats: the control mice in the study were short-lived to begin with; most lab mice die of cancer anyway, so it’s not exactly shocking that a cancer drug helped them live longer; and perhaps most importantly, the dose of trametinib used was way beyond what humans would typically get - raising some fairly obvious questions about side effects and real-world relevance.

Andrew Steele has covered the result in more detail, well worth a read.

New milestone for Insilico’s AI drug

Long-time readers of this newsletter are familiar with Insilico Medicine by now. You might also recall that Insilico Medicine’s Phase 2a clinical trial of Rentosertib, an AI-discovered and -designed drug for idiopathic pulmonary fibrosis (IPF), demonstrated improved lung function.

A recent peer-reviewed publication of the Phase 2a data in Nature Medicine marks the first time these findings have been formally validated in a scientific journal.

Right to try gets practical

What if there was a longevity drug you could access without having to wait years for FDA approval? That idea is one of the driving forces behind so-called longevity cities like Infinita and Viva.* The goal is to create regulatory sandboxes where individuals can make informed choices about experimental treatments, fully aware of the risks involved.

Another advantage: these environments could attract the kind of longevity moonshots that often struggle to get off the ground in more heavily regulated systems weighed down by red tape.

* For more on this, check out our LEVITY episodes with Niklas Anzinger and Laurence Ion.

Until recently, this was impossible to do on U.S. soil. But that’s now changing: Montana has passed SB 535 into law, allowing clinics to offer experimental medical treatments that haven’t yet received FDA approval - as long as they’ve passed basic safety testing (phase I trials).

But wait - didn’t Montana already have a “right to try” law? Yes, that was SB 422, which removed the requirement that only terminally ill patients could access experimental treatments. The new law goes a step further by creating a licensing system that allows clinics to legally operate and provide these treatments more broadly.

Giving AI the data it needs

The UK government has announced the creation of the OpenBind consortium. And why do we care about that? Well, OpenBind will generate the largest-ever dataset on drug–protein interactions, twenty times larger than anything previously collected, enabling the development of AI models to identify new drugs, slashing drug development costs (by reducing the need for trial-and-error) by up to £100 billion, and accelerate treatments for previously untreatable diseases.

Among the participants in the project is Isomorphic Labs, a Google DeepMind spin-off.

How axolotls regenerate limbs

Axolotls famously shrug off trauma: lop off a limb and, within weeks, a perfect replacement sprouts. A new Nature Communications study nails down part of the recipe. An axolotl enzyme trims the amount of retinoic acid - a molecule humans also make and smear on their skin - so just the right pattern re-emerges, while the gene Shox sets the limb’s ultimate size.

What makes the discovery even more interesting is the animal’s broader biology. As you may recall from a Steve Horvath study last year: an axolotl’s aging clock races only during its first four years, then almost flat-lines for the rest of its decade-plus life. Across that long plateau the methylome stays eerily calm, with barely any of the age-related gains that mark mammalian aging.

Together, the two studies sketch a dual blueprint for human therapy. Fine-tuning retinoid levels could guide scarred tissues toward proper healing, while manipulating epigenetic circuits might keep those tissues biologically young.

One step closer to “living reviews”

Systematic reviews are the foundation of evidence-based medicine. They guide clinical practice, drug approvals, and billions in healthcare spending. But they're painfully slow - often taking years to complete, and quickly going out of date as new studies emerge.

Now that might be about to change. In a new paper, researchers (including George Church) show that their AI system otto-SR can replicate an entire Cochrane Review issue in just 2 days. What normally requires the equivalent of 12 years of expert labor was handled by LLMs (large language models) parsing PDFs, extracting data, assessing bias, and running meta-analyses - in two days. Yes, twelve years of work was done in two days.

We might be headed for a reality with fully automated “living reviews” that update continuously as new trials get published. Regulators, doctors, and payers would no longer wait years for evidence updates - they'd get real-time dashboards that adjust as the data evolves.

Of course, we’re not there yet. AI remains brittle and prone to hallucinations. And fully automated reviews risk becoming black boxes that nobody can audit. These hurdles don’t feel unsurmountable, however. And when they’re cleared we might get evidence synthesis running at superhuman speed.

Worth your time.

Understanding gene editing

Here’s a lovely visual guide to different gene editing methods.

Biology’s got reinforcement

A great thread from the brilliant Bo Wang on how reinforcement learning (a popular type of machine learning) not only helps scientists model biology but also “intervene in it, design it, reason about it”.

LEVITY turns one!

The LEVITY podcast has just celebrated its first birthday 🎂🥳🎁 and we’re closing in on 1 000 subscribers on Youtube. One of our latest episodes, with Valter Longo, has racked up over 25 000 views. These are small but encouraging milestones - and we’re just getting started!

Pushing the frontier

Only when we have 94-year-olds who look and - more importantly - have the health of 25-year-olds can we say the mission is accomplished. But for now, one must concede that it’s remarkable how much William Shatner still looks like he could be in his 60s.

Xeno life

Remember the replacement paper I discussed a few weeks ago? It often pointed to the rapid advances in xenotransplantation. Tim Andrews, now living with a pig kidney, is a great example of how far we've come.

Hey, you’ve made it all the way here! Thank you so much for reading! 🫶🏼