PD-L1 Positivity Rates in Head and Neck Squamous Cell Carcinoma in a UK Reference Centre
A retrospective audit of 98 UK HNSCC cases revealed that PD-L1 CPS ≥20 positivity rates (59.2%) were significantly higher than those reported in the KEYNOTE-048 trial, with oral cavity primaries and resection samples showing the highest expression levels.
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
In the fight against cancer, the immune system is often the body's most powerful ally, yet tumors have learned to wear a disguise that keeps this defense force at bay. One of the most effective disguises involves a protein on the surface of cancer cells called PD-L1. This protein acts like a "do not disturb" sign, signaling to the immune system's soldiers that the cell is healthy and should not be attacked. Doctors can now use a specific medicine, pembrolizumab, to strip away this disguise, allowing the immune system to recognize and destroy the tumor. However, this treatment does not work for everyone, and its success depends heavily on how many of these "do not disturb" signs are present on the cancer cells. To measure this, pathologists use a scoring system that counts the number of cancer cells and immune cells wearing the PD-L1 badge. This number, known as the combined positive score, helps doctors decide if a patient is likely to benefit from the treatment. The challenge lies in ensuring that the score calculated in a local hospital matches the scores seen in the massive clinical trials that originally proved the drug worked, because if the local numbers are different, the rules for who gets the medicine might need to be reconsidered.
Researchers at a major hospital in the United Kingdom decided to check their own records to see how their local results compared to the famous KEYNOTE-048 trial, which established the current rules for using this cancer drug. They looked back at 98 cases of head and neck squamous cell carcinoma, a type of cancer affecting the mouth, throat, and voice box, that were tested between 2021 and 2023. In the original trial that set the standard, about 85 percent of patients had enough of the PD-L1 protein to be considered for treatment, and roughly 43 percent had a very high level of the protein. When the UK team analyzed their own 98 patients, they found that 90.8 percent had a score high enough to qualify for treatment, a number that was slightly higher but statistically similar to the trial. More surprisingly, they found that 59.2 percent of their patients had a very high score, a figure that was significantly higher than the 43 percent seen in the original trial. This difference was not a small fluctuation; it was a clear signal that the group of patients in this local center had a different biological makeup than the group used to design the treatment guidelines.
The researchers dug deeper to understand why their numbers were so high. They noticed that the type of tissue sample mattered greatly. When they looked at patients who had undergone surgery to remove the tumor, known as resections, the rate of high scores was 76.2 percent. In contrast, patients who only had a small needle biopsy showed a lower rate of 54.5 percent. While this difference did not reach the threshold of statistical certainty, it suggested that larger pieces of tissue might capture more of the cancer's true nature than small samples. The location of the tumor also played a major role. Patients with cancers starting in the oral cavity, or mouth, showed the highest rates of the high scores, with nearly three-quarters of them qualifying for the highest treatment category. This group of mouth cancers made up a large portion of the hospital's cases, which likely pulled the overall average up. The study also noted that the doctors who scored the slides were different individuals, and while the testing method itself was standardized, the human element of reading the slides could introduce some variation.
These findings suggest that the local population of patients in this UK center may not look exactly like the patients in the global trials that defined the treatment rules. If a local hospital sees a much higher percentage of patients with high scores, it raises questions about whether the current thresholds for giving the drug are perfectly calibrated for every specific group of people. The researchers emphasized that this was a look back at past data, and they are now planning a new, forward-looking study to confirm if these results hold true when they are collected in real-time without any selection bias. As treatments for cancer move from advanced stages to earlier, curable stages, getting these numbers right becomes even more critical. The goal is to ensure that every patient gets the right treatment based on accurate, reliable data, and that the way doctors count these biological markers remains consistent across different hospitals and regions.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.