Combination of real-time PET and CT is a highly sensitive tool for identifying non-small cell lung cancer

The combination of real-time PET and CT is a highly sensitive tool for identifying non-small cell lung cancer (NSCLC) and can assist in identifying patients whose cancer has not yet spread to lymph nodes, according to results of a study by researchers at Johns Hopkins.

The findings are important because they suggest that combination PET/CT should be included during the early part of diagnosis, when physicians are trying to determine the stage of the cancer. The seriousness of NSCLC is rated from N0 to N3, with N0 being cancer that has not spread from the original tumor; and N1 to N3 referring to increasingly serious stages that have spread to lymph nodes.

However, the study also found that the combination of PET and CT is less sensitive in identifying benign nodules, that is, non-cancerous lumps of tissue. Moreover, the overall process of determining the stage of NSCLC remains difficult, especially in differentiating between N1 and N2 stages of cancer, the Hopkins researchers note.

"PET is already widely used to evaluate cancer nodules and determine their stage," says Stephen C. Yang, M.D., associate professor of surgery and oncology and chief of the Division of Thoracic Surgery at Johns Hopkins. "Our study will help show whether a combination of PET and CT can accurately determine the cancer stage in patients who might be effectively treated surgically rather than require chemotherapy, especially in the N0 stage." Yang presented these results this week at the 84th annual meeting of the American Association of Thoracic Surgeons in Toronto.

In the Hopkins study, PET/CT correctly identified 94 of 100 (94 percent) malignant lesions (tumors) in patients, but only eight of 22 (36 percent) of lumps that were not cancerous. The technique correctly identified the stage of cancer in 61 of 100 patients (i.e., 56 with stage N0, two with N1 and three with N3).

However, combination PET/CT overstaged the tumor (identified cancer as being more severe) in 20 patients, while understaging (identifying as less severe that it really is) in 19.

The researchers concluded that while PET/CT is highly sensitive for identifying malignant lesions, it is less accurate for benign nodules. Moreover, while it is good at predicting N0 disease, it is not efficient in distinguishing between N1 and N2 disease.

The study used real-time PET/CT, which provides images immediately rather than requiring time to develop pictures from data gathered by the individual imaging machines.