Imaging-based biological age estimation predicts competing mortality risk in lung cancer screening
This study demonstrates that an imaging-based biological age estimation derived from low-dose CT scans significantly improves the prediction of non-lung cancer mortality risk in lung cancer screening participants, offering a more accurate assessment of competing risks than chronological age alone.
Original paper licensed under CC BY 4.0 (https://creativecommons.org/licenses/by/4.0/). This is an AI-generated explanation of the paper below. It is not written or endorsed by the authors. For technical accuracy, refer to the original paper. Read full disclaimer
Imagine you are standing in a vast library, but instead of books, the shelves are filled with people's bodies. For decades, doctors have used a simple rule to guess how long someone might live: look at the date on their birth certificate. This is their "chronological age." If you were born in 1960, you are 64 years old. But here is the twist: two people born in the same year can feel and function very differently. One might be running marathons, while the other is struggling to climb a flight of stairs. This difference is called "biological age." It's like a car's odometer versus its engine health; two cars might have driven the same number of miles, but one has a rusted engine and the other is tuned to perfection.
In the world of lung cancer screening, doctors use special low-dose CT scans to look for tiny lumps in the lungs that could turn into cancer. The goal is to catch these lumps early, before they become dangerous. However, there is a tricky problem. The people who get screened are often older smokers. While they are at risk for lung cancer, they are also at risk for many other things that can end their lives, like heart disease or breathing problems. Sometimes, a person might be so sick from these other causes that finding a lung cancer lump early won't actually help them live longer. Doctors need a way to tell the difference between someone who is "old" because of their birth year and someone who is "old" because their body is worn out. They need a way to measure the wear and tear on the engine, not just the miles on the odometer.
This is exactly what a team of researchers from University College London and other institutions set out to do. They asked a simple but powerful question: Can we use the pictures from a lung scan to build a "biological age" score that predicts who is most likely to die from causes other than lung cancer?
The researchers looked at a massive group of over 12,000 people who had participated in a lung cancer screening trial in London. They didn't just look at the lungs; they used a clever computer program (a type of artificial intelligence) to measure four specific things in the body that change as we get older: how much fat is hiding inside our muscles, how dense our bones are, how much calcium is building up in our arteries, and how much fat is stored around our heart. Think of these as the "rust," "cracks," "gunk," and "sludge" that accumulate in a machine over time.
By combining these four measurements, the team created a single number called the "age gap." This number tells you how much older or younger your body feels compared to your actual birth year. If your age gap is positive (like +5), your body looks 5 years older than you are. If it's negative (like -5), your body looks 5 years younger.
The results were fascinating. The team found that this "age gap" was a crystal ball for predicting who might die from non-lung cancer causes. For every 10 years your biological age is older than your real age, your risk of dying from something like heart disease or COPD doubled. In fact, this biological age score was even better at predicting non-lung cancer deaths than just looking at your birth year alone. However, when it came to predicting who would actually get lung cancer, the age gap wasn't as helpful. It was a great predictor for the "other" risks, but not a magic wand for the lung cancer risk itself.
The study suggests that this method could help doctors have better conversations with patients. Imagine a 55-year-old man who looks biologically like he's 65 because of his scan results. Even though he is young by the calendar, his body is showing signs of heavy wear and tear. The doctor might say, "Your lungs look okay for now, but your body is showing signs of being much older. The risk of you getting sick from something else is high, so maybe we should focus on fixing your heart and lungs first, rather than just watching for cancer." Conversely, a 70-year-old woman whose scan shows she is biologically 60 might be a great candidate to keep getting screened, because her engine is still running strong.
The researchers are careful to say that this isn't a final solution yet. They tested their idea within one specific group of people in London, and they need to see if it works for everyone else in the world. They also note that while their computer model is smart, it's not a replacement for a doctor's judgment. But, they have shown that a simple scan can do more than just find lumps; it can tell a story about how fast a person's body is aging. By using this "biological age" score, doctors might be able to tailor screening plans so that they help the people who need them most, without wasting time on those whose bodies are already fighting too many other battles. It's a step toward making screening not just about finding cancer, but about understanding the whole person.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.