Finds higher thyroid malignancy in Hispanics

Thyroid nodule care runs on risk categories. A doctor looks at an ultrasound, maybe does a needle biopsy, sometimes adds molecular testing, and then decides whether a nodule looks safe to watch or risky enough to remove. The problem is that those tools were built to be broadly useful, not necessarily equally calibrated for every population. A new UC San Diego analysis suggests that, for Hispanic patients, the same category label may hide a meaningfully higher cancer risk. (pmc.ncbi.nlm.nih.gov) ### What was the new finding? The study looked at 308 adults and 421 thyroid nodules evaluated at a tertiary academic center from January 2023 through June 2024. About 29.2% of the patients were Hispanic. Researchers compared how well three common tools lined up with the final answer from surgery — ultrasound TI-RADS scores, Bethesda biopsy categories, and molecular testing. Hispanic patients ended up with higher overall odds of malign(pmc.ncbi.nlm.nih.gov)d Bethesda categories was similar between groups. (pmc.ncbi.nlm.nih.gov) ### Why is that surprising? Because the whole point of these systems is to turn messy biology into a usable risk bucket. TI-RADS sorts nodules by ultrasound features. Bethesda sorts needle-biopsy samples by how suspicious the cells look. If two groups land in similar buckets, you would expect roughly similar cancer rates within those buckets. But that did not fully happen here. Basically, the label looked the same while the underlying risk did not. (pmc.ncbi.nlm.nih.gov) ### Which category showed the clearest gap? TI-RADS 4 stood out. Among nodules in that ultrasound category that later went to surgery, 32.7% were malignant in Hispanic patients versus 12.0% in non-Hispanic patients. That came out to an odds ratio of 3.57. This is the kind of result that gets attention because TI-RADS 4 is not the most extreme category — it is exactly the gray zone where management decisions can get tricky. (pmc.ncbi.n([pmc.ncbi.nlm.nih.gov)bout biopsy and molecular testing? The sharpest difference showed up in indeterminate biopsy results — Bethesda III or IV — when molecular testing came back suspicious. In that subgroup, malignancy was 78.6% in Hispanic patients versus 34.8% in non-Hispanic patients, with an adjusted odds ratio of 8.72. The catch is that this was a narrower subgroup, so the confidence interval was wide. Still, the direction is hard to ignore: (pmc.ncbi.nlm.nih.gov)ients than clinicians assume. (pmc.ncbi.nlm.nih.gov) ### Does this mean the tools are broken? Not exactly. It means the tools may be less portable than people hope. Risk systems are usually validated at the population level, but bedside decisions happen in specific clinics, in specific communities, with specific referral patterns. A category can still be useful while being imperfectly calibrated for one group. Think of it like a weather forecast that gets the season right but misses the local microclimate. (pmc.ncbi.nlm.nih.gov) ### Is there bigger context here? Yes — and it points in the same direction. A 2024 California registry study found that Hispanic thyroid cancer patients had more advanced disease features at diagnosis, including larger tumors and more nodal involvement, and Hispanic ethnicity remained independently linked to higher mortality risk after adjustment. So this new nodule study did not appear out of nowhere. It fits a broader pattern that H(pmc.ncbi.nlm.nih.gov), or care pathways. (pmc.ncbi.nlm.nih.gov) ### What should doctors and patients take from this? Not “operate on everybody.” More like: be careful with false reassurance. If a Hispanic patient has a nodule that looks intermediate-risk on ultrasound or sits in an indeterminate biopsy category, the conversation around follow-up, repeat sampling, molecular testing, or surgery may need a little more caution than the category name alone suggests. (pmc.ncbi.nlm.nih.gov)ot that thyroid risk tools stopped working. It is that they may not mean exactly the same thing for everyone — and Hispanic patients may be one group where that difference matters. (pmc.ncbi.nlm.nih.gov)