PET-CT CAN CHARACTERIZE ACCURATELY ADRENAL MASSES IN CANCER PATIENTS

A recent report examined the prevalence and nature of adrenal masses that were identified incidentally on computed tomography (CT) in patients with no known malignancy.^[1] A computer search of the abdominal CT reports in 65,231 consecutive patients over a 3-year period at a single institution resulted in identification of an adrenal mass in 3307 (5%) of patients; 973 of whom (1.5% of all patients and 29% of patients with adrenal mass) had 1049 adrenal masses had no known malignancy or suspicion for a hyperfunctioning adrenal mass. With the use of histopathology, further imaging follow-up, and characterization and clinical follow-up serving as validation criteria, the following types of adrenal lesions were identified: 75% adenomas, 6% myelolipomas, 4% hamartomas, 1% cysts, 0.3% pheochromocytomas, and 0.1% cortisol-producing adenoma, and no malignancy. The focus of the following complementary study was to assess the ability of positron-emission tomographic CT (PET-CT) in characterizing adrenal masses in patients with known cancer.

PET/CT for the Characterization of Adrenal Masses in Patients With Cancer: Quantitative Versus Qualitative Accuracy in 150 Consecutive Patients

Boland GWL, Blake MA, Holalkere NS, et al
AJR. 2009;192:956-962

Summary

The study cohort included 150 consecutive patients with a wide variety of known underlying malignancies, with lung cancer, lymphoma, and colorectal cancer as the most common etiologies -- comprising about two thirds of all cancers in the group. Patients were referred to 2-[fluorine 18]fluoro-2-deoxy-D-glucose (FDG) PET-CT for evaluation of extent of disease. Validation for the detected adrenal lesions was by histology and all relevant prior and follow-up CT scans (unenhanced, contrast-enhanced, and delayed contrast-enhanced washout studies). Qualitative analysis was performed by comparing the level of FDG uptake in the adrenal gland with that of the liver using the following scoring scheme:

• 0 -- adrenal activity less than liver activity;
  
  
• 1 -- equal to liver;
  
  
• 2 -- moderately higher than liver; and
  
  
• 3 -- markedly higher than liver.

Semiquantitative analysis was performed by using the typical mean standardized uptake value formulation with the region of interest drawn over two thirds of the adrenal lesion on fused PET-CT image.

A total of 165 adrenal lesions were found, with 139 lesions (84%) ultimately determined as benign and 26 lesions (16%) as malignant. The analysis of the unenhanced CT demonstrated a mean diameter of 2.6 cm (range, 1.2-6.9 cm) for malignant adrenal lesions and 1.8 cm (range, 1-4.3 cm) for the benign lesions. The mean CT density was 36 Hounsfield units (HU) (range, 22-67 HU) for the malignant lesions and 1.1 HU (range, -66 to 35 HU) for the benign lesions. All malignant lesions showed CT density greater than 10 HU. Using 10 HU as the unenhanced CT density threshold, the sensitivity and specificity for detection of adrenal malignancy were 100% and 66%, respectively. The analysis of FDG PET showed a standardized uptake value range of 2.3 to 26.1 for malignant lesions and 0.5 to 3.3 for benign lesions. Using the lowest malignant standardized uptake value value of 2.3 as threshold, the sensitivity and specificity for detection of adrenal malignancy were 100% and 94%, respectively. For the qualitative PET analysis, using a score of 2 as the threshold yielded a sensitivity of 100% and a specificity of 97% for detection of adrenal malignancy. The authors concluded that PET-CT is highly accurate in differentiating benign from malignant adrenal masses in patients with known cancer.

Viewpoint

Based on their study, the authors offer a recommendation for a stepwise diagnostic procedure for the characterization of adrenal masses in patients with known cancer. In the first step, if there is no evidence for benign disease with unenhanced CT (<>