Clinical Value of 18F-FDG PET/CT in Detecting Adrenal Metastasis in Patients with Hepatocellular Carcinoma.

Clinical Value of

F-FDG PET/CT in Detecting

Adrenal Metastasis in Patients with

Hepatocellular Carcinoma

Yin Lin, M.D.

Long-Bin Jeng, M.D.

Hsin-Yi Wang, M.D.

Shih-Chuan Tsai, M.D.

Wan-Yu Lin, M.D.

Chia-Hung Kao, M.D.

*

Hepatocellular carcinoma is a highly prevalent and lethal neoplasm, representing the second most common cancer in Taiwan (1, 2). Liver transplantation, which is considered the ideal treatment for limited HCC, involves removing both tumors and the preneoplastic underlying diseased liver, particularly considering the high prevalence of hepatitis B and C virus carriers in Taiwan (3-5). Because HCC disseminates mainly through the portal vein radicles, the rate of microvascular invasion increases proportionally to the size of the tumor (6). Surveillance programs

aim to reduce disease-specific mortality by detecting HCC at a curable stage. Therefore, detecting extrahepatic metastatic disease is crucial when planning

potential therapy for patients with HCC.

Selecting suitable candidates for LT preoperatively is still a challenge. The lungs, skeletal system, and adrenal glands are the most common sites of nonlymphatic

metastatic HCC (7). F-18 fluorodeoxyglucose (18F-FDG) positron

emission tomography/computed tomography (PET/CT) imaging has been proven to be effective in detecting metastatic HCC (8, 9). Regarding several types of malignancy, 18F-FDG PET/CT images have been proven to

be highly sensitive and effective, specifically in treating adrenal metastasis (10, 11). However, not all of the FDG-avid

lesions in adrenal glands indicate metastatic disease (12).

The purpose of the current study was to evaluate the performance of F-18 FDG EPT/CT by using SUV analysis to distinguish

benign from malignant adrenal masses.

Methods and Materials

Patients

From January 2009 to January 2012, 166 patients diagnosed with HCC at China Medical University Hospital received

18F-FDG PET/CT imaging prior to liver transplantation (certificate number of local institutional review board:

DMR99-IRB-010-2). Of these patients, 5 patients (4 males, 1 female; median age: 51.2 y, range: 33-61 y) were found to have suspected adrenal metastases and were included in this study.

One patient (Case 1) had adrenal lesions on the left, one (Case 3) had adrenal lesions on the right, and 3 (Cases 2, 4, and 5) had bilateral adrenal lesions. Each adrenal lesion was characterized according to its size, its attenuation on CT expressed in Hounsfield units (HUs), and the intensity of 18F-FDG uptake, expressed as a standardized uptake value (SUV). The final diagnoses of all 5 patients were confirmed based on the other imaging findings, clinical follow-up, or histopathologic examination of surgical specimens. Table I

lists the characteristics of all 5 patients and indications of 18F-FDG PET/CT scans. We analyzed the pattern and SUVmax

of the adrenal glands, recorded the clinical features of these patients, and compared the results with other imaging methodologies and histopathologic findings.

18F-FDG PET/CT Imaging

All 5 patients underwent a whole-body 18F-FDG PET/CT

study. The patients fasted for at least 6 hours before undergoing 18F-FDG PET/CT scans. None of them were diabetic. All

of them were asked to rest in a quiet room after intravenous injection of 370 MBq (10 mCi) of 18F-FDG. Sixty minutes after the 18F-FDG injections, a whole-body 18F-FDG PET/CT scan was performed using a Discovery STE scanner (GE, USA). No contrast was used at the time of 18F-FDG PET/CT acquisition.

After the acquisition, data were transferred to a workstation for processing and interpretation. The reconstructed images were displayed in axial, sagittal, and coronal planes. The image

interpretation and analysis were performed qualitatively (visually) and quantitatively. The maximal standardized uptake

value (SUVmax) of the liver and adrenal glands, the retention index (RI) (defined as the ratio of increased SUV activity in the delayed phase to adrenal SUV activity in the early phase), and the standardized uptake ratio (SUR) (defined as the ratio of adrenal SUV activity to liver SUV activity) were computed. Table II presents the quantitative values for the 5 patients. Statistical Analysis

Descriptive analysis was expressed in terms of mean and standard deviation. Linear regression was used to determine an association between CT numbers and SUVmax.

Results

Image Analysis

The average SUVmax of the liver in the 5 patients was 2.85 (ranging from 2.4 to 2.74) in the early images. The average SUVmax of the adrenal glands was 2.31 (ranging from 1.53 to 2.81) in the early images, and 2.36 (ranging from 1.41 to 3.14) in the delayed phase. The average SUR of the 8 adrenal glands was 0.9. The average SUVmax of adrenal metastases was 2.23 (Lesions 4-6, and 8), and the average SUR of

adrenal metastases was 0.83. The average CT number of the adrenal glands was 39 (ranging from 25 to 49). We determined a moderate correlation between SUVmax and the CT

numbers of the adrenal lesions (Figure 1). Compared with

histopathologic reports and established clinical and imaging follow-up results, there was one false-negative metastatic

lesion (Lesion 4, Case 3) and one false-positive adrenal gland lesion (Lesion 7, Case 5).

Initial Staging

The initial staging 18F-FDG PET/CT scan was performed in both Cases 1 and 5, yielding negative findings in bilateral adrenal glands. A 33-year-old female hepatitis B virus (HBV) carrier, identified as Case 1, visited our outpatient department for a regular examination. She was diagnosed as having

HCC with elevated serum alpha-fetoprotein and hepatic masses detected using abdominal sonography. The 18F-FDG PET/CT scan revealed multiple 18F-FDG-avid masses in both lobes of the liver, an 18F-FDG-avid mass in the left adrenal gland, and an 18F-FDG-avid area in the left retroperitoneal area. The 18F-FDG-avid lesions in the left adrenal gland and left retroperitoneal area were recognized as benign processes.

Therefore, the patient received transcatheter arterial chemoembolization (TACE) for local control followed by antiviral

therapy; to date, there has been no evidence of recurrence. Case 5 presented with a massive HCC approximately 12 cm in diameter in the right lobe of the liver, detected using abdominal sonography. The initial staging 18F-FDG PET/CT scan showed no extrahepatic metastases. The patient received living donor liver transplantation (LDLT) after being downstaged with TACE.

Restaging

Cases 2-4 underwent 18F-FDG PET/CT scans for restaging. Case 2 was a 59-year-old male hepatitis C virus (HCV) carrier. HCC was diagnosed in 2002; right partial lobectomy

was performed and 9 courses of TACE were administered between 2002 and 2010. In February 2011, a nodule (3 cm in

diameter) in Segment 8 was noted in the CT images. The pretreatment 18F-FDG PET/CT scan showed mild tracer accumulation

in the bilateral adrenal glands, which were recognized as benign lesions. The patient received LDLT in March 2011; to date, no evidence of HCC recurrence or metastases has been noted. Case 3 was a 50-year-old male HBV carrier. HCC was diagnosed in 2006, after which 6 courses of TACE and radiotherapy were administered for local recurrence. In July 2011, several nodules in the bilateral lobes of the liver were noted. Pretreatment CT images showed a small nodule in the right adrenal gland, which was difficult to differentiate between a benign process and malignancy. The following 18F-FDG PET/CT scan showed no abnormal tracer accumulation in the right adrenal gland (Figure 2). The patient received LDLT in October 2011. Right-side adrenalectomy was also performed, histopathologically revealing metastatic HCC.

Case 4 was a 62-year-old male HBV carrier who received a segmentectomy because of a massive HCC in the right lobe of the liver in January 2011. The preoperative CT image showed a small nodule in the right adrenal gland, which was difficult to distinguish as being a benign process or a malignancy. The postoperative CT images showed progressive enlargement of the bilateral adrenal glands. The restaging 18F-FDG PET/CT scan in October 2011 revealed abnormal 18F-FDG accumulations in the bilateral adrenal glands, which were recognized

as metastatic lesions. The patient received TACE for bilateral adrenal metastases in October 2011. However, the CT image from December 2011 showed progressive enlargement of the bilateral adrenal glands. It was suggested that a second course of TACE be administered, or adrenalectomy be performed, as soon as possible.

Treatment Response Evaluation

Case 5 was a 54-year-old HBV carrier. Seven months after receiving LDLT for HCC, a right adrenal mass approximately 3 cm in diameter was found in the patient. The 18F-FDG PET/ CT scan showed two 18F FDG-avid masses in the bilateral adrenal glands (Figure 3). Subsequent CT images showed an enlarged right adrenal gland; therefore, the patient received a

bilateral adrenalectomy. The histopathologic findings showed metastatic HCC in the right adrenal gland but no malignancy

in the left adrenal gland.

Discussion

Diagnosing malignant adrenal involvement in HCC patients is necessary to determine the appropriate treatment approach and to assess prognosis after LDLT. A high complication rate

and a high recurrence rate are consequences of liver transplantation accompanied by known metastases, which is a

definitive contraindication for liver transplantation. 18F-FDG PET/CT scans are a feasible tool for detecting malignancy and have several advantages.

Several studies have shown that unenhanced CT and F-18 FDG PET can be used to differentiate benign from malignant adrenal masses in cancer patients (13, 14). However,

SUV of 3.1 led to misclassifying most of our patients. Watanabe et al. (15) enrolled 129 patients with extraadrenal primary malignancies, most of whom were lung cancer patients,

and concluded that the adrenal-to-liver SUV ratio was accurate in differentiating benign from metastatic adrenal lesions. In contrast to the findings of previous studies, the average SUVmax of the 8 adrenal lesions was lower than the SUVmax of the liver (mean SUR 5 0.9) and considerably lower in the adrenal metastatic lesions (mean SUR 5 0.83). Only 3 lesions (Lesions 1, 2, and 8) were greater than 1.0, and 2 of them were benign lesions, possibly because of overestimation of the hepatic SUV caused by preclinical presentations, such

as HBV- or HCV-related hepatitis, thus indicating an underestimation of the SUR. In a systematic review and metaanalysis

of the characterization of adrenal masses by using an 18F-FDG PET scan, qualitative (visual) PET analysis demonstrated the most favorable sensitivity and specificity (16). In our cases, 4 lesions (Lesions 5-8) were suspected to be positive in visual

analysis, and 3 of the 4 were proved malignant based on histopathologic and clinical results. We distinguished benign from

malignant lesions visually based on 18F-FDG accumulations in the adrenal glands.

The false-negative adrenal metastatic lesion (Lesion 4, Case 3) showed no greater 18F-FDG accumulations than that in the liver (SUR 5 0.65) in the early phase. This small lesion was less than 1 cm in diameter in the CT images. In the delayed phase, it decreased in capacity of 18F-FDG accumulation at an RI ratio of 7.84%. We concluded that this lesion was a benign process based on the 18F-FDG PET/CT findings, but it proved to be malignant according to histopathologic results.

One possible reason is that the lesion was too small to be visualized or quantitated. The 18F-FDG accumulations of the

small lesions may be underestimated during the reconstruction process, because of so-called partial volume effects. This bias may be minimized by increasing the matrix size (17). Although simple correction methods are routinely available and applicable to reduce biases, the partial volume effect may still be a factor when analyzing PET tumor imaging.

The false-positive lesion (Lesion 7, Case 5) was visualized in the 18F-FDG PET/CT scan, showing an increased SUVmax in the delayed phase (RI ratio 5 4.76%). An enlarged right adrenal mass in this patient was noted in CT images 7 months after LDLT. The 18F-FDG PET/CT images showed 2 18F-FDG-avid

lesions in both adrenal glands with increased 18F-FDG accumulation in the delayed phase. We concluded that the bilateral

adrenal lesions were metastases based on the 18F-FDG PET/CT images. However, the left adrenal gland showed no metastatic tissue after adrenalectomy. False-positive findings have been reported and are encountered in approximately 5% of adrenal lesions identified as positive in 18F-FDG PET/CT images. The possible causes of false-positive results may be adrenal adenomas, adrenal endothelial cysts, and inflammatory and infectious lesions (18).

The diagnostic CT features for distinguishing between metastatic lesion and an adenoma are tumor size and attenuation

value. In our cases, the average number of HU of all adrenal lesions was 39, considerably higher than the thresholds (17). In our case, the false-negative adrenal lesion had similar HU numbers to that of the false-positive adrenal lesion. Despite moderate correlation between the SUVmax (both SUVe and SUVd) and the CT numbers of the adrenal lesions, none of these values is effective in distinguishing a benign adrenal mass from HCC adrenal metastasis.

Conclusion

For differentiating benign from malignant adrenal masses in HCC patients, visual PET analysis appears to be more useful than calculating quantitative standardized uptake values. Both CT images and 18F-FDG PET/CT scans have limitations in distinguishing between malignant and benign lesions in adrenal glands in HCC patients. Larger case numbers are necessary to compare and confirm the results of this study.