Model-Based Prebiopsy Risk Estimation and Systematic Core Omission During Transrectal Rigid Magnetic Resonance Imaging–Ultrasound Fusion Biopsy
In a cohort of men undergoing transrectal rigid MRI-ultrasound fusion biopsy, the study demonstrates that systematic cores should be retained because omitting them results in a clinically significant miss rate of nearly 9%, with imaging categories alone failing to reliably identify high-grade disease.
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
Every year, millions of men face a difficult decision: whether to undergo a needle biopsy to check for prostate cancer. The process usually begins with a powerful magnetic resonance imaging scan, which acts like a high-resolution map of the prostate gland. If this scan reveals a suspicious area, doctors traditionally take two types of samples. First, they aim directly at the suspicious spot, guided by the map. Second, they take a series of samples from the rest of the gland in a standard pattern, just to be safe. For years, a central question has lingered in urology: if the map shows a clear target, is it still necessary to take those extra, random samples from the rest of the gland? Some hoped that aiming only at the target would be enough, sparing patients from unnecessary needles and reducing the risk of finding harmless, slow-growing cells that would never cause harm. However, the answer depends heavily on how the doctor combines the map with the ultrasound images during the procedure.
A new study involving nearly 900 men sought to answer this question under very specific conditions. The researchers focused on men who had a positive scan and underwent a biopsy using a particular technique called rigid fusion. In this method, the computer software aligns the pre-scan map with the live ultrasound image by moving and rotating the map until it fits, without stretching or warping the image to match the shape of the prostate during the exam. This is a common approach in many hospitals, distinct from other methods that use flexible, elastic matching or a different needle path. The team wanted to know two things: what factors before the biopsy could best predict the presence of dangerous cancer, and exactly how many dangerous cancers would be missed if doctors stopped taking the standard, random samples and only targeted the suspicious spots.
The study found that predicting cancer risk is more complex than looking at the scan alone. While the severity of the suspicious area on the scan is important, it is not the only factor. The researchers discovered that a man's age, the density of a protein called PSA in his blood relative to the size of his prostate, and the results of a physical exam where a doctor feels the prostate all work together to determine the likelihood of finding clinically significant cancer. When these factors are combined, they provide a much clearer picture of risk than any single piece of information could offer. However, even with this detailed risk calculation, the study delivered a firm conclusion regarding the biopsy strategy: under the rigid fusion technique used in this group, skipping the standard random samples is not safe.
When the researchers compared the results of the targeted samples alone against the results of the full set of samples, they found that the targeted approach missed a significant number of dangerous cancers. Specifically, if the doctors had only taken the samples aimed at the suspicious spots, they would have failed to detect nearly one-quarter of the clinically significant cancers that were actually present. This means that for every 100 men who had dangerous cancer, the targeted-only method would have missed about 24 of them. The risk of missing cancer was not evenly distributed; it was highest in men whose scans showed the least certain suspicious areas and in those with lower levels of the blood protein relative to their prostate size. In these groups, the targeted approach missed more than 40 percent of the dangerous cancers.
The study also looked at the most aggressive forms of cancer, which are the ones that require immediate treatment. Here, the pattern was slightly different. The level of suspicion on the scan was less important for predicting these high-grade tumors than the physical exam results and the blood protein density. This suggests that a scan showing a highly suspicious area does not automatically guarantee the presence of the most dangerous cancer, nor does a less suspicious area rule it out. The most critical finding remains the cost of omission. The researchers calculated that by skipping the standard random samples, doctors would miss dangerous cancers in nearly 9 out of every 100 men who undergo the procedure. This is a substantial number, representing a risk that patients and doctors must accept if they choose to stop the standard sampling.
The authors emphasize that these findings apply specifically to the rigid fusion technique used in their study. Other methods, such as those using elastic software matching or a different needle path, might yield different results. However, for the vast majority of procedures currently performed using rigid software alignment, the evidence is clear. The standard practice of taking both targeted and systematic samples should be maintained. Relying solely on the targeted approach, even with a sophisticated risk calculation, would leave a quarter of dangerous cancers undetected. The study concludes that while doctors can use a man's age, blood tests, and physical exam to better understand his individual risk, they should not use that information to justify removing the safety net of the standard random samples. The goal is to ensure that no man leaves the clinic with a dangerous cancer that could have been found with a few extra, routine needles.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.