Breast cancer has been playing chess while we've been stuck playing checkers. For decades, this cellular villain has watched us throw the exact same screening playbook at every single woman who walks through the door - annual mammograms, same age cutoffs, rinse and repeat - like a heist movie where the security guards never change their patrol route. A new commentary in JAMA by Pesapane and Cassano argues it's time we finally got smart about our strategy.
The One-Size-Fits-All Problem (a.k.a. The Netflix Algorithm We Never Built)
Think about how Netflix knows you're a "dark Scandinavian thriller" person while your roommate gets recommended nothing but baking shows. That level of personalization? We've had it for entertainment for over a decade. But breast cancer screening? Still operating like a Blockbuster Video circa 2003.
The current approach is brutally simple: hit a certain age, get a mammogram at a set interval, repeat until someone tells you to stop. The US Preventive Services Task Force updated its recommendations in 2024 to say all women should start biennial screening at 40 - a step forward, sure, but still fundamentally treating every woman's breast cancer risk as identical. That's like giving everyone the same prescription glasses because, hey, you all have eyes.
Pesapane and Cassano, writing from the European Institute of Oncology in Milan, call this out directly: our screening paradigm was "designed during an era with limited risk stratification, few prevention pathways, and limited ability to tailor imaging intensity" (DOI: 10.1001/jama.2026.1263). Translation: we built this system when we didn't know any better, and now we do.
Enter the WISDOM Trial (The Avengers Assemble Moment)
If this story has a superhero origin arc, it's the WISDOM randomized clinical trial, whose results dropped in December 2025 like a Marvel post-credits scene everyone had been waiting for. The trial enrolled over 28,000 women and compared personalized, risk-based screening against the standard annual mammogram approach.
Here's the plot twist worthy of a Succession finale: risk-based screening didn't just match annual screening - it actually found fewer late-stage cancers. The risk-based group had 30 per 100,000 person-years of stage IIB+ cancers compared to 48 per 100,000 in the annual screening group. One-third fewer advanced cancers. That's not a rounding error. That's the screening equivalent of upgrading from a flip phone to a smartphone.
The WISDOM approach used genetic sequencing of nine susceptibility genes, a polygenic risk score, and clinical risk models to sort women into four buckets: low risk (26% - could safely wait until 50 to start screening), average risk (62% - every two years), elevated risk (8% - annual), and high risk (2% - twice per year with alternating mammography and MRI). Personalized. Precise. Actually logical.
The AI Plot Thickens
Meanwhile, artificial intelligence is crashing this party like an uninvited character who turns out to be essential to the plot. The FDA authorized Clairity Breast in 2025 - the first AI tool that can analyze a routine mammogram and predict a woman's five-year breast cancer risk. Trained on over 421,000 mammograms, this tool catches patterns in breast tissue that human eyes and traditional risk calculators miss entirely. It's now included in NCCN guidelines, which is basically the Rotten Tomatoes "Certified Fresh" of cancer screening.
This matters especially for women with dense breast tissue, where standard mammography performs about as well as trying to find a polar bear in a snowstorm. AI-based assessment has shown superior accuracy in exactly these cases.
Why "More Screening" Isn't Always "Better Screening"
Here's the uncomfortable Black Mirror episode nobody wants to watch: for every 1,000 women screened biennially from age 40 to 74, about eight avoid dying from breast cancer. That's genuinely important. But those same 1,000 women also rack up around 1,540 false positives, 210 unnecessary biopsies, and 12 cases of overdiagnosis - cancers detected that would never have caused harm. That's a lot of anxiety, needles, and treatment side effects for tumors that were basically just sitting in the corner minding their own business.
Risk-based screening flips this equation. Low-risk women get fewer unnecessary scares. High-risk women get the intensive surveillance they actually need. Resources go where they matter most. It's the difference between a targeted Severance-style operation and just... carpet-bombing everything.
The Bottom Line
Pesapane and Cassano are making a case that shouldn't be controversial but somehow still is: stop screening everyone the same way. The tools exist. The WISDOM trial proved the concept works. AI is making risk prediction cheaper and more accessible. The only thing standing between us and smarter screening is institutional inertia - the healthcare system's version of "but we've always done it this way," which, if you've watched any season of any hospital drama ever, you know is the villain's favorite line.
The era of personalized breast cancer screening isn't coming. It's here. We just need to stop treating every woman like she's the same character in the same story with the same ending.
References:
-
Pesapane F, Cassano E. Screening for Breast Cancer. JAMA. Published online April 6, 2026. DOI: 10.1001/jama.2026.1263. PMID: 41941217.
-
Esserman LJ, Eklund M, Veer LJ, et al. Risk-Based vs Annual Breast Cancer Screening: The WISDOM Randomized Clinical Trial. JAMA. 2025. DOI: 10.1001/jama.2025.23039. PMID: 41385349.
-
US Preventive Services Task Force. Screening for Breast Cancer: US Preventive Services Task Force Recommendation Statement. JAMA. 2024;331(22):1918-1930. DOI: 10.1001/jama.2024.5534. PMID: 38687503.
-
Pesapane F, Cassano E, et al. Artificial Intelligence-Driven Personalization in Breast Cancer Screening: From Population Models to Individualized Protocols. Cancers. 2025;17(17):2901. PMID: 40940998.
-
ACR Appropriateness Criteria - Female Breast Cancer Screening: 2025 Update. Journal of the American College of Radiology. 2025. Available at: JACR.
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.