Multiphase dual-energy CT parameters combined with CT features in differentiating three common lung cancers
This study demonstrates that combining triphasic dual-energy CT quantitative parameters with conventional CT features significantly improves the diagnostic accuracy for differentiating among adenocarcinoma, squamous cell carcinoma, and small cell lung cancer compared to using either method alone.
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
Lung cancer is not a single disease but a collection of different conditions that behave in distinct ways. The two most common groups are non-small cell lung cancers, which include adenocarcinoma and squamous cell carcinoma, and small cell lung cancer. While they all appear as abnormal growths in the chest, their biology, how they spread, and the treatments that work best for them are vastly different. Adenocarcinoma often starts in the outer edges of the lung, while squamous cell carcinoma and small cell lung cancer tend to grow near the center, close to the main airways. Because the treatment for one type can be ineffective or even harmful for another, doctors need to know exactly which kind a patient has before they begin therapy. Usually, this requires a biopsy, where a needle takes a small sample of tissue for a pathologist to examine under a microscope. However, getting that sample can be difficult if the tumor is deep inside the chest or if the patient is too frail to undergo the procedure. This creates a need for a way to identify the cancer type using only the images from a scan.
A team of researchers at Gaozhou People's Hospital in China set out to improve how doctors distinguish between these three types of lung cancer using a specialized type of imaging called dual-energy computed tomography. Standard CT scans take pictures of the body using X-rays, but dual-energy CT uses two different energy levels simultaneously. This allows the machine to see not just the shape of a tumor, but also its chemical makeup, specifically how much iodine-based contrast dye the tumor absorbs. The researchers focused on a specific protocol where they scanned patients three times after injecting the dye: once when the dye first arrived in the arteries, once when it filled the veins, and a third time after a short delay. They also looked closely at the physical shape of the tumors, noting features like whether the edges were smooth or spiky, if the tumor pulled on the nearby lung tissue, or if it had fused with nearby lymph nodes. By combining these detailed chemical measurements with the visual shape of the tumors, the team aimed to create a reliable method for telling the cancers apart without needing a biopsy in every case.
The study involved 151 patients who had already been diagnosed with lung cancer through tissue samples. The group included 81 people with adenocarcinoma, 38 with squamous cell carcinoma, and 32 with small cell lung cancer. The researchers analyzed the scans of all these patients, measuring how much iodine the tumors held at different times and noting their physical characteristics. They found clear patterns that separated the three groups. Small cell lung cancer, which is known for growing quickly and often having areas of dead tissue inside it, consistently showed the lowest levels of iodine uptake across all three scan phases. It also had a strong tendency to be located in the center of the chest and to merge with nearby lymph nodes. Adenocarcinoma, on the other hand, was usually found on the outer edges of the lung and often caused the lung surface to pull inward, a sign known as pleural retraction. Squamous cell carcinoma occupied a middle ground but had its own distinct signature, often showing uneven enhancement of the dye and a tendency to be located centrally.
The most significant finding was that looking at the scans from all three time points together provided a much clearer picture than looking at just one phase or just the shape of the tumor. When the researchers combined the chemical data from the delayed scan with the physical features of the tumor, their ability to correctly identify the cancer type improved dramatically. For distinguishing between adenocarcinoma and small cell lung cancer, this combined approach was correct 95.1% of the time. When telling apart squamous cell carcinoma and small cell lung cancer, the accuracy reached 89.5%. Even for the more difficult task of distinguishing between the two types of non-small cell cancer, the combined method was successful in 90.1% of cases. The study showed that the delayed phase of the scan, which captures how the dye lingers in the tissue, was particularly useful for making these distinctions, a detail that previous studies had not fully explored for this specific purpose.
The researchers noted that while their method showed great promise, it was tested on a single group of patients from one hospital, so the results would need to be confirmed in larger, wider studies before becoming a standard practice everywhere. They also acknowledged that the measurements relied on human observers drawing circles around the tumors on the screen, which introduces a small amount of human variation. Despite these limitations, the work demonstrates that a careful look at how a tumor absorbs dye over time, paired with a close inspection of its shape, can provide a powerful, non-invasive way to identify lung cancer types. This approach offers a potential path forward for patients who cannot undergo a biopsy, giving doctors a clearer view of the disease and helping to ensure that the right treatment is chosen from the very beginning.
Drowning in papers in your field?
Get daily digests of the most novel papers matching your research keywords — with technical summaries, in your language.