The Protein With a Secret Second Job: How Medicine Is Finding Hope by Unlearning

Ninety-three percent. That's the number that made Bradley Davis, a urologist at the San Francisco VA, pause. In 2024, 93% of veterans with low-risk prostate cancer chose active surveillance over immediate treatment — up from just 27% in 2005. Three decades after UCSF first pioneered "watch and wait" for these patients, the approach has quietly become the national standard of care.

That number is more than a statistic. It's proof that medicine can change direction — and the same revolution is happening in labs across the world, where researchers are rethinking what we thought we knew about some of our most stubborn diseases.

The protein with a secret second job

At Peter Mac in Australia, a team led by associate professor Shom Goel has been wrestling with a frustrating mystery: why does breast cancer treatment stop working for some patients?

The answer, published in Nature, upends a decades-old assumption. The retinoblastoma (Rb) protein — long celebrated as one of the body's natural brakes on runaway cell growth — has a second, hidden role. It doesn't just switch genes off. It can switch some on, unexpectedly activating a group of estrogen-responsive genes in hormone receptor-positive breast cancer, the most common subtype.

"Rb is not simply a brake," the researchers effectively discovered. CDK4/6 inhibitors work by switching Rb back on to slow tumors — but this newfound activation of estrogen genes may explain why most cancers eventually resist. It's a paradox that could point toward better combination treatments.

Maps and targets from the cellular frontier

The same spirit of reexamination is reshaping the search for new cancer therapies. At Harvard Medical School, Steven Gygi and his team built an AI tool called KinoPlex that maps the three-dimensional structures of 1.8 million protein sites where kinases — cell-communication molecules — might act. Humanity had fully characterized fewer than 1% of them before this. Published in Nature Biotechnology, KinoPlex could help doctors match a patient's specific cancer with the right one of around 100 existing kinase-inhibiting drugs.

From Harvard to Karolinska Institutet in Sweden, researchers reported in Nature Communications that macrophage-targeted immunotherapy only works when tumor immune cells have a functional response system — a finding that could predict which patients will respond and open new combination therapies.

And at UC Irvine, a framework published in Frontiers in Pharmacology proposes repurposing existing, safety-tested drugs to prevent the dangerous brain inflammation that can complicate CAR T-cell therapy — the immunotherapy that's transformed hope for people with aggressive blood cancers.

The gut's quiet influence

The rethinking extends beyond cancer. At the University of Wisconsin–Madison, professors Barbara Bendlin and Federico Rey spent nearly a decade chasing a clue: the gut bacteria of Alzheimer's patients differ from those of healthy people. Now, in Nature Communications, they show how a compound called imidazole propionate (ImP), produced by some gut bacteria, plays a role in the brain changes underlying Alzheimer's disease. Targeting that compound could reduce risk — a path to prevention that starts in the digestive tract.

For the ones who need help most

And some of the most hopeful news comes from a Southern California classroom. A UCLA study in the Journal of Research on Adolescence found that among teenage girls in a program combining structured breathing with social-emotional learning, those who had experienced the most childhood trauma showed the largest gains in social connectedness.

School-based programs usually help the least-needy most. Here, the reverse happened. The study's message — that the students who need help most can benefit most — joins a wider theme running through all these findings.

From a protein with a secret double life to a gut compound nobody expected, from 93% of veterans choosing watchful waiting to teenage girls breathing their way toward connection — the story is the same. Medicine is no longer assuming it knows how the body works. It's asking again, looking closer, and finding that the most stubborn problems often yield to a willingness to be surprised.

The tools to act on these discoveries — repurposed drugs, AI maps, deeper biological understanding — are already here. The next breakthroughs won't come from more of the same. They'll come from looking at what we thought we knew and asking one gentle question: what if we're wrong?