Hemorrhage-Induced Bias of [18F]FDG PET Signal Interpretation in Pancreatic Ductal Adenocarcinoma

Pancreatic ductal adenocarcinoma is a particularly aggressive form of cancer, known for its ability to spread quickly and its resistance to treatment. Because the disease often advances silently, doctors rely heavily on imaging to understand what is happening inside a patient's body. One of the most common tools for this is a scan called a PET scan, which uses a special radioactive sugar to light up areas where cells are consuming energy at a high rate. Since cancer cells typically eat sugar much faster than healthy cells, these scans usually show bright spots where tumors are growing. However, this method is not perfect. The sugar can also light up areas where the body is fighting inflammation or where tissue has died, creating a confusing picture that makes it difficult to tell exactly how much active cancer remains.

A team of researchers at the Technical University of Munich and the University Medical Center Hamburg-Eppendorf has uncovered a specific reason why these scans can sometimes be misleading. They discovered that when a tumor bleeds internally, the blood itself can trick the scan into seeing a bright, active tumor even when the tissue is actually dead. In their study, the scientists looked at pancreatic tumors in mice and a small group of human patients to see how bleeding inside the tumor affected the imaging results. They found that while dead tissue usually appears dark on these scans because it is not using energy, dead tissue that is also bleeding appears bright. This happens because the bleeding triggers an immune response, sending in white blood cells that are hungry for sugar, which then light up the scan just like a living tumor would.

The researchers began by growing two different types of pancreatic tumors in mice, representing the main variations of the disease found in humans. They then performed the sugar-based scans on these animals and compared the images to detailed maps of the tumors' internal structure. In tumors that had died but did not bleed, the scans correctly showed dark, empty spaces where the cells had stopped working. But in tumors that had died and also contained pockets of bleeding, the scans showed bright, glowing areas. The researchers measured the brightness of these spots and found that the bleeding tumors showed significantly higher levels of sugar consumption than those without bleeding. This difference was clear in both the mouse models and the small group of human patients they examined, where tumors with bleeding showed much higher scan values than those without.

To understand why this was happening, the team looked closely at the tissue under a microscope. They discovered that the bleeding areas were filled with immune cells, specifically a type of white blood cell called a macrophage, which acts as a cleanup crew for the body. These cells rush to the site of the bleed to clear away the damaged blood and tissue. Because these immune cells are very active, they consume large amounts of sugar, causing the scan to light up. The researchers noted that this effect was so strong that it made the dead, bleeding tissue look just as active as a living, growing tumor. This means that a doctor looking at a scan might mistakenly believe a tumor is still growing and active when, in reality, it has died and is simply being cleaned up by the immune system.

This finding challenges the standard way doctors interpret these scans. Usually, when a scan shows a dark, cold spot, it is assumed to be dead tissue and is ignored. When a spot is bright, it is assumed to be active cancer. This study suggests that a bright spot could actually be a dead tumor that is bleeding and inflamed. The researchers tested whether adjusting the numbers to account for the rest of the body's sugar levels would fix the problem, but the difference remained. The bleeding tumors still looked much brighter than the non-bleeding ones, regardless of how the data was calculated. This indicates that the presence of blood is a major factor that can skew the results, leading to an overestimation of how much cancer is actually present.

The study also looked at whether different types of pancreatic tumors reacted differently to this phenomenon. They found that the effect happened in both major subtypes of the disease, meaning it is a general issue rather than one limited to a specific kind of tumor. In the human patients included in the test group, those with bleeding tumors had scan values that were more than double those of patients without bleeding. While the group of human patients was small, the results mirrored what was seen in the mice, suggesting that this is a real biological phenomenon that affects human diagnosis as well. The researchers emphasized that this does not mean the scans are useless, but rather that they need to be read with more care.

To solve this problem, the authors suggest that doctors should not rely on the sugar scan alone. They recommend combining it with other types of imaging, such as MRI, which can clearly show the physical structure of the tumor and detect the presence of blood. By looking at both the metabolic activity from the sugar scan and the structural details from the MRI, doctors can distinguish between a living tumor and a dead, bleeding one. This approach would help prevent the misclassification of a patient's condition, ensuring that treatment decisions are based on the true state of the disease rather than an artifact of internal bleeding.

The researchers acknowledge that their study has limits, particularly the small number of human patients they were able to examine. They also note that the mouse models, while useful, do not perfectly replicate the complex environment of a human tumor. However, the consistency of the findings across different models and the clear biological mechanism they identified provide a strong basis for further investigation. They propose that future studies should look at larger groups of patients to confirm how often this happens and to develop better ways to spot it. Until then, the key takeaway is that bleeding inside a tumor can create a false signal of activity, and recognizing this is crucial for accurately understanding the progress of pancreatic cancer.