Hybrid Imaging Proves More Sensitive and Specific than Solo Scanners

February 19, 2008

In what is possibly the largest study of its kind, Harvard University researchers have shown that PET/CT can reliably tell whether unsuspected adrenal lesions discovered while managing oncologic patients are benign or malignant.

In what is possibly the largest study of its kind, Harvard University researchers have shown that PET/CT can reliably tell whether unsuspected adrenal lesions discovered while managing oncologic patients are benign or malignant.

Incidental adrenal lesions in patients with cancer require accurate characterization because their status can significantly affect staging, therapy, and prognosis. Several imaging tests are available to offer a helping hand, including contrast/noncontrast CT and MRI. None, however, has proven conclusive.

A 2006 Massachusetts General Hospital study of 38 patients found PET/CT to be sensitive and specific for the characterization of incidental adrenal lesions (Blake MA, Slattery JM, Kalra MK. Adrenal lesions: characterization with fused PET/CT image in patients with proved or suspected malignancy-initial experience. Radiology 2006; 238(3):970-977). The results encouraged an investigative group led by Dr. Giles W. L. Boland, an associate professor of radiology at Mass General, to validate the findings in a larger cohort of patients.

Between October 2004 and March 2007, Boland and colleagues evaluated 150 patients with various malignancies, including lymphoma and lung, head, and neck and colorectal cancer. All patients underwent noncontrast PET/CT and a follow-up CT scan six months later. The researchers confirmed that PET/CT was highly effective for differentiating benign from malignant adrenal disease in this patient population. Based on a qualitative analysis, PET/CT yielded sensitivity, specificity, accuracy, and positive and negative predictive values for characterization of benign adrenal "incidentalomas" of 100%, 97%, 98%, 87%, and 100%, respectively.

Boland cautioned the study had several limitations, including the retrospective nature of the sample. He called for careful consideration of many other criteria, including the Hounsfield unit and standardized uptake value measurements, adrenal wash-outs, and lesion size assessments. But the findings showed a way forward, he said.

"PET/CT is indeed an accurate test," Boland said.