Prostate cancer occupies a strange place in American medicine. It is the most frequently diagnosed cancer in U.S. men, with more than 333,000 new cases a year, and also among the least deadly, with only about 4% of cases ultimately proving lethal and a five-year survival rate near 99%. That combination sounds like a triumph, and in part it is. But it is also, in part, an artifact of how the disease is found, and the growing debate over whether to lean on MRI instead of biopsy is, underneath the technical detail, a debate about that artifact.
A test that works too well
The standard diagnostic path begins with a PSA blood test. If the level of that protein is elevated, the next step is usually a systematic biopsy, in which the urologist samples tissue from across the prostate in a grid-like pattern to hunt for cancer. This process is genuinely effective at detecting cancer, including at very early stages, and for a long time that was considered its great virtue.
The problem is that it may be too effective. Up to 70% of new prostate cancer diagnoses are low-grade tumors that will never need treatment, because the odds of them turning deadly are negligible. Finding them anyway is not a neutral act. A man told he has cancer rarely hears the word "harmless," and many go on to surgery or radiation that carries real risks of incontinence and impotence, or at minimum to years of anxiety and repeat biopsies, all for a tumor that was never going to hurt him. This is overdiagnosis, and it flows directly from a test that is very good at finding cancer and very bad at telling which cancers matter. The systematic biopsy's core strength, that it finds so much, is also its central flaw, because it cannot distinguish the cancer that will kill you from the one that will simply sit there for the rest of your life.
The survival number hides as much as it reveals
The reassuring 99% survival statistic is entangled with this problem in a way that is easy to miss. When a diagnostic system sweeps up enormous numbers of cancers that were never going to be fatal, the survival rate is pushed upward almost automatically, because those patients were always going to survive their disease regardless of what anyone did. Counting them as cancer "survivors" inflates the number without reflecting any lives saved.
So the framing of prostate cancer as the cancer you are overwhelmingly likely to survive is true, but it partly measures the overdiagnosis rather than any special success at defeating a lethal disease. Some of that survival figure is treatment working, and some of it is the statistical shadow of finding cancers that never needed to be found. Keeping those two apart matters, because a great deal of the argument for changing how the disease is diagnosed rests on recognizing that a chunk of what the current system detects is not a win at all.
Why the better test finds less
This is what makes the alternative approach counterintuitive. The proposed shift is to perform an MRI before any biopsy, and then to biopsy only the suspicious lesions the scan flags, a targeted rather than systematic approach. Its advantage is not that it finds more cancer. Its advantage is that it finds less, specifically less of the insignificant kind, while still catching the cancers that are actually dangerous.
That inverts the intuition most people bring to medical testing, where a more sensitive test that catches more is assumed to be better. In a disease defined by overdiagnosis, the more discriminating test is the better one, because the goal is not to find the maximum amount of cancer but to find the right cancer. The evidence for this has become substantial. Randomized trials, including the Swedish STHLM3-MRI and GOTEBORG-2 studies published in the New England Journal of Medicine, found that MRI-directed biopsy detects clinically significant cancer about as well as systematic biopsy while sharply cutting the number of unnecessary biopsies and the diagnosis of indolent tumors. Pre-biopsy MRI is now standard practice in much of the world, recommended by European and major U.S. cancer-network guidelines, yet in the United States such scans were performed in only about a third of prostate cancer cases as of 2022.
The case for caution is real
It would be a mistake to treat that lag as pure inertia or bad faith, because there are legitimate clinical reasons to be careful, and honesty requires giving them weight. MRI interpretation varies considerably from reader to reader, depends heavily on expertise, and most of the strong trial evidence comes from specialized centers with deep experience in prostate imaging, which raises a real question about whether community practices can reproduce those results. An MRI-first strategy is only as good as the MRI, and a mediocre scan read by an inexperienced radiologist could miss a cancer that a systematic biopsy would have caught.
And systematic biopsy does find more cancer overall, including a small number of significant cancers that MRI-targeting misses. The trials are reassuring on this point, the missed significant cancers tended to be low-volume and intermediate-risk, the kind managed with active surveillance anyway, but "tended to be" is not "always were," and missing an aggressive cancer is a grave harm that weighs heavily against the harms of overdiagnosis. So there is a genuine tradeoff rather than a clear winner: MRI-first reduces the real damage of overdiagnosis and overtreatment, but carries its own real risk of missed cancers, contingent on imaging quality that may not travel well from expert centers to everywhere else. Reasonable clinicians can, and do, weigh that tradeoff differently, and the caution is not merely a cover story.
The incentives that quietly favor the status quo
That said, clinical caution is not the only force holding the status quo in place, and the less flattering forces deserve to be named plainly. Prostate biopsies are lucrative for urologists, who can perform them within their own practices, which gives the people ordering the tests a financial reason to keep ordering them. Finding indolent cancers compounds the effect, because those cancers are typically placed on active surveillance, which generates a stream of follow-up appointments and repeat biopsies over the years. A shift toward MRI-first would move revenue away from urology and toward imaging, and would shrink that downstream stream of monitoring biopsies.
This is not an accusation that individual doctors are consciously choosing profit over patients; most are surely doing what they sincerely believe is thorough care. But the structural incentive is real and points in one direction, and it is a textbook principal-agent problem: the physician who decides which diagnostic pathway to use benefits financially from the more-testing pathway, which happens also to be the one that overdiagnoses and overtreats. When the clinically cautious choice and the financially rewarding choice coincide, it becomes very hard to disentangle how much of the resulting practice is caution and how much is incentive, and the coincidence itself is a reason to scrutinize the practice rather than take it at face value.
The patient instinct that misfires
There is a final force pushing in the same direction, and it comes from patients themselves. People facing a possible cancer often want as much information as they can get, and they frequently push for more testing rather than less, on the deeply ingrained belief that catching cancer early and knowing everything about it can only help.
For most cancers, that belief is sound. For the large indolent majority of prostate cancers, it can be exactly wrong. Knowing about a tumor that was never going to harm you does not extend your life; it can shorten your peace of mind and expose you to treatments whose side effects are worse than the disease would ever have been. The "catch it early" mantra, one of the most valuable messages in all of cancer care, misfires precisely here, where early detection of an indolent cancer is not a rescue but a trap. That is a genuinely hard thing to communicate to a frightened patient, and the difficulty of communicating it is part of why the overdiagnosing approach persists, since it is far easier to order another test than to explain why not finding something can be the better outcome.
What the debate is really about
Strip away the acronyms and the procedural detail, and the fight over MRI versus biopsy is a fight over a single confusion at the center of how this disease is managed: the equation of finding more cancer with providing better care. For most of oncology that equation holds. For a cancer as overwhelmingly indolent as this one, it breaks, and clinging to it produces a system that manufactures diagnoses nobody benefits from and some people are harmed by, then points to the resulting survival statistics as proof it is working.
The better test, the one that finds fewer cancers, is unsettling precisely because it asks doctors and patients to want less detection, to accept that a scan showing nothing suspicious is often good news rather than a missed opportunity, and to trust a more discriminating tool over the reassuring thoroughness of sampling everything. The evidence increasingly says that trust is warranted, at least where the imaging is good, and the rest of the wealthy world has largely acted on it. Whether the United States follows will depend not only on the data, which is already fairly clear, but on whether medicine can override the intuition, the incentives, and the patient demand that all, for once, happen to be pulling the wrong way at the same time.
Primary sources
- STAT News (reporting by Annalisa Merelli) for the framing of the biopsy-versus-MRI debate, the figures on prostate cancer incidence (more than 333,000 new U.S. cases annually) and low lethality (about 4% fatal, roughly 99% five-year survival), and the estimate that up to 70% of new diagnoses are low-grade and need no treatment.
- STAT News for the finding that pre-biopsy MRI was used in only about a third of U.S. cases in 2022, the observations about variable MRI interpretation, and the value of systematic biopsy in finding more cancer overall.
- STAT News for the financial incentives for urologists to perform biopsies and monitor indolent cancers, patient demand for more testing, and specialist advocacy for pushing further toward imaging.
- The New England Journal of Medicine for the GOTEBORG-2 population-based noninferiority trial showing MRI-targeted biopsy reduces overdiagnosis of clinically insignificant cancer while maintaining detection of significant cancer.
- JAMA and the STHLM3-MRI randomized trial, via PubMed Central, for evidence that MRI in men with elevated PSA reduces unnecessary biopsies while maintaining detection of clinically significant cancer.
- European Association of Urology, NCCN, and American Urological Association guidance for the differing recommendations on pre-biopsy MRI.