Utility of Magnetic Resonance Imaging in Evaluating Focal Hepatic Uptake on 18F-FDG PET/CT Without a Computed Tomography Correlate in Patients with Known Malignancy – A Single-Center Retrospective Study

INTRODUCTION

Radionuclide imaging procedures in nuclear medicine are noninvasive, and except for intravenous injections, are generally painless medical tests.^[1]

These imaging scans use radioactive materials called radiopharmaceuticals or radionuclides which are injected into a vein and accumulate in the target organ. A common procedure is positron emission tomography/computed tomography (PET/CT) which is a type of nuclear medicine imaging modality that uses small amounts of radioactive material, most commonly used ¹⁸F-fluorodeoxyglucose(¹⁸F-FDG), to diagnose various diseases, including many types of cancer. PET and CT images are combined to provide combined physiological information and precise anatomical localization in a single examination.

FDG-PET uses ¹⁸F-fluorodeoxyglucose, a short-lived radioactive compound that localizes in tumor cells. Tumor cells have high metabolic activity and use more glucose; ¹⁸F-fluorodeoxyglucose is a glucose analog that, after entering the cell, has the same metabolic pathway as glucose. When phosphorylated, FDG-6-PO4 does not undergo further metabolism or diffuse out of the cell, so it remains trapped, ideal for imaging.^[2] PET/CT scans could detect metabolically active tissue regardless if the activity is related to malignancy, inflammation, or other causes.^[3] Hence, any incidental abnormalities on 18F-fluorodeoxyglucose (FDG) PET/CT have been reported in 6.71%–12% of scans; furthermore, accurate interpretation of incidental focal uptake and knowing when to suggest further investigations are important.^[1]

Metastasis is the most common malignancy of the liver, occurring 20 times more often than primary liver tumors. Metastases can result from a wide variety of neoplasms; colorectal cancer is the most common one, then breast and lung primary cancers. Several types of liver metastases are known to have avid ¹⁸F-FDG uptake, and thus, ¹⁸F-FDG PET/CT has a good detection rate for liver metastases.^[4,5] There are several studies shown ¹⁸F-FDG PET/CT scan to be the most sensitive imaging modality in the diagnosis of liver metastasis from colorectal, esophageal, and gastric cancers. Furthermore, mimickers can result in a false positive and usually are associated with abscesses, cholangitis, and granulomatous lesions.^[6] In addition, the liver is known to demonstrate heterogeneous physiological uptake on ¹⁸F-FDG PET/CT which may be misinterpreted as a focal metastatic lesion. One of the challenges in the assessment of focal hepatic uptake on PET/CT is the absence of corresponding findings on the plain CT examination. This results in decreased confidence about the findings.^[6]

There are a few studies that looked into the utility of magnetic resonance imaging (MRI) in further evaluating patients with known malignancy who had focal hepatic uptake in PET/CT without a CT correlate, however, those studies had small sample sizes. Therefore, this study is an additional study to the literature with a larger sample size in the evaluation of the utility of contrast-enhanced liver MRI in the evaluation of patients with known malignancy who had focal hepatic uptake on ¹⁸F-FDG PET/CT without a CT correlate. It also further analyzes some of the objective ways in determining the presence or absence of hepatic metastasis on PET/CT.^[7]

MATERIAL AND METHODS

The study was done at the Royal Hospital (RH), as a retrospective cross-sectional study. It was approved by the Center of Studies and Research at the Ministry of Health, Muscat, Sultanate of Oman. ¹⁸F-PET/CT and contrast-enhanced liver MRI database at RH were searched electronically to identify patients who underwent contrast-enhanced liver MRI within 3 months of PET/CT with focal hepatic ¹⁸F-FDG uptake, for histopathology-confirmed cases of malignancy between 2015 and 2021. The patients who had a plain CT correlate for the focal FDG uptake were excluded from the study. Thirty-six patients were recruited between 2015 and 2021, who had focal hepatic uptake on FDG PET/CT without a CT correlate and follow-up contrast-enhanced MRI within a 3-month period. All the data collection was carried out using the workstations at the Radiology Department of the RH.

A board-certified nuclear medicine physician blindly reviewed 18F-FDG PET/CT and a board-certified body imaging radiologist blindly reviewed contrast-enhanced liver MRI examination, separately. Assessment of the presence of a focal uptake on FDG PET/CT and MRI was performed with documentation of the number of lesions, measurement of the lesion size, localization according to the hepatic segments, and interpretation of the findings. The results of the histopathology findings and/or follow-up examinations were then collected to determine the final diagnosis. Ultimately, the results of the PET/CT scan and MRI were counterchecked against the final diagnosis. Statistical analysis and correlation between the intensity of the FDG uptake and the presence of focal hepatic lesions on contrast-enhanced MRI were performed using SPSS (Statistical Package for the Social Sciences) software, version 27 and Microsoft Excel 2019 Version.

RESULTS

Thirty-six patients were identified between 2015 and 2021, who had focal hepatic uptake on FDG PET/CT without a CT correlate and follow-up contrast-enhanced liver MRI within 100 days; 18 were female and 18 were male, with a mean age of 51 years. The sites of the primary malignancies with their frequency are listed in Table 1. The most common primary malignancy was colorectal cancer (39%), followed by breast cancer (20%). In the study group, 57 focal hepatic FDG uptakes were identified on PET/CT without CT correlate. The median maximum standardized uptake value (SUV_max) was 5.0 (range = 2.7–16.24). The ratio of median SUVmax of the hepatic lesion to that of normal parenchyma was 1.8 [Fig 1].

Close

Fig 1:

The ratio of the median maximum standardized uptake value of the hepatic lesion to that of normal parenchyma. SUVmax: Maximum standardized uptake value

Export to PPT

Out of the 57 sites of focal FDG uptake in the liver, 37 had correlating lesions on MRI, and the remaining 20 focal areas of FDG uptake were not corresponding to any findings on MRI, and they did not show any abnormalities on the follow-up examinations indicating that they are likely part of the heterogeneous physiological liver uptake of FDG [Fig 2]. The median size of focal hepatic FDG uptake and the focal lesions on MRI are summarized in Table 2. Eighty-one percent (30 cases out of the 37 cases) of the focal hepatic FDG uptake with MRI correlate were confirmed to be metastatic lesions based on imaging features as well as follow-up imaging where they demonstrated interval increase in size and 5 out of the 30 cases were confirmed histopathologically to be metastatic lesions. The remaining 7 cases out of 37 (19%) of focal areas of FDG uptake were confirmed on MRI as benign lesions, two cases were postsurgical changes, two were simple cysts, one was an abscess, one was a case of hemangioma, and one was a case of posttransarterial chemoembolization in cirrhotic liver.

Close

Fig 2:

Focal FDG positron emission tomography/computed tomography uptake with correlation to contrast-enhanced magnetic resonance imaging liver lesions. MRI: Magnetic resonance imaging, FDG: ¹⁸F-fluorodeoxyglucose

Export to PPT

We found no significant difference in the SUV_max of the hepatic lesions and SUV_max ratio between the groups with and without MRI correlate [Table 3]. However, there was a significant difference in the SUV_max of the hepatic lesions between the group with final diagnosis of benign lesions and metastasis with P = 0.016, but there was no significant difference in SUV_max ratio with P = 0.097 [Table 4]. Furthermore, there was a significant difference between the difference of SUV_max of the lesion and the background liver, between the groups of the final diagnosis of benign and metastatic lesions, and there was no significant difference between the difference of SUV_max between the lesion and mediastinum, between the two groups [Table 5].

DISCUSSION

According to our study, 55% of the sites of focal hepatic ¹⁸F-FDG uptake on PET/CT scan without CT correlate in patients with a known primary malignancy had a corresponding abnormality on MRI liver and 81% of these lesions were diagnosed as metastases based on imaging features and follow-up imaging. There was no significant statistical difference of the SUV_max values of the lesions with or without MRI correlate. These findings suggest that focal ¹⁸F-FDG uptake of the liver without CT correlate may warrant contrast-enhanced MRI evaluation in patients with a known malignancy because of the increased risk of hepatic metastasis.

It can be difficult to identify hepatic lesions on PET scans without a CT correlate because the physiologic uptake of normal hepatic parenchyma can vary and be impacted by cancer treatments such as chemotherapy and radiation as well as other variables such as blood sugar levels and hepatic steatosis.^[6] According to our findings, there was no statistically significant difference between the groups with and without MRI correlate in terms of SUV_max and lesion-to-parenchyma SUV_max ratios, or in terms of the final diagnoses of benign lesions versus metastases. In our study, there is a case of a metastatic liver lesion, with SUV_max of 6.5, in contrast, there is a case with a focal hepatic FDG uptake with SUV_max of 7.97, which did not show any lesion at the liver MRI or in the follow-up imaging [Fig 3]. According to this finding, SUV_max may not be a reliable diagnostic standard for hepatic lesions in cancer patients.

Close

Fig 3:

A 61-year-old male with known SCC of the right forearm. (a) Positron emission tomography/computed tomography scan shows focal hepatic FDG uptake at segment V (arrow), (b) Magnetic resonance imaging axial T2-weighted imaging shows no corresponding lesion. FDG: ¹⁸F-fluorodeoxyglucose

Export to PPT

Few studies have investigated the significance of focal uptake on FDG PET/CT scans in patients with known malignancy. One of the studies reported that 41% of focal FDG-avid lesions of any organs without CT correlates in patients with cancer are malignant, and one study that has been published in 2019 at AJR by T. Araki showed that more than half of the cases of focal hepatic uptake on PET/CT without a CT correlate had an MRI correlate, and more than 75% of these lesions were metastases, regardless of SUV_max.^[4,7] These studies confirm our findings that hepatic FDG uptake with MRI correlate has a high predictive value for the presence of hepatic metastases and may warrant further evaluation with MRI in patients with malignancy.

One of the most sensitive imaging modalities for hepatic metastasis is FDG PET/CT,^[6] but there are known interpretational pitfalls. Although they are uncommon, false-positive results can happen in certain benign lesions such as abscesses, cholangitis, and granulomatous diseases. In our study, there is a case of liver hemangioma, which is one of the most common benign vascular liver lesions. They are frequently diagnosed as an incidental finding on imaging, and most patients are asymptomatic.^[8-11] It showed FDG uptake at PET scan with SUV_max value of 3.9 [Fig 4], and there is a case with a hepatic abscess which showed FDG uptake with an SUV_max value of around 6.88, so these findings could represent a mimicker for liver metastasis [Fig 5].

Close

Fig 4:

A 48-year-old female with known colorectal cancer. (a) Positron emission tomography/computed tomography shows focal hepatic FDG uptake at segment II, (b and c) Magnetic resonance imaging axial postcontrast shows progressive peripheral nodular enhancement in keeping with a hemangioma (arrow), (d) Magnetic resonance imaging T2-weighted imaging axial shows T2-hyperintense lesion in segment 2 (arrow). FDG: ¹⁸F-fluorodeoxyglucose

Export to PPT

Close

Fig 5:

A 66-year-old female with colorectal cancer. She was diagnosed to have a liver abscess. (a) Positron emission tomography/computed tomography shows focal hepatic FDG uptake at the right liver lobe, (b) Magnetic resonance imaging T2-weighted imaging shows central diffusion restriction (arrow), (c and d) diffusion-weighted imaging/apparent diffusion coefficient shows central diffusion restriction (arrow), (e) Postcontrast image shows a peripheral rim of enhancement. FDG: ¹⁸F-fluorodeoxyglucose

Export to PPT

Our study has some limitations partly due to its retrospective nature and due to the relatively small sample size and the fact that it is performed in a single institution. In addition, our study mainly correlated focal FDG uptake with contrast-enhanced MRI findings, so patients with indeterminate FDG uptake who did not undergo MRI because of subtle image findings were not captured in our study. Furthermore, PET/CT and follow MRI were not performed simultaneously or in the same week, so may have had some cases of progression of hepatic metastasis that were better detected by MRI.

CONCLUSION

More than half of the areas of focal liver uptake seen on PET/CT without a CT correlate in patients with known malignancy had an MRI correlate in our study, and 81% of those lesions were diagnosed as metastases based on imaging features and follow-up imaging, with no significant statistical difference of the SUV_max values. Therefore, contrast-enhanced MRI is a useful imaging modality for the characterization of focal hepatic FDG uptake without CT correlate in patients with known malignancy. SUV_max and the difference between SUV_max of focal hepatic uptake and the background liver parenchyma appear to be helpful parameters in the distinction between benign and malignant lesions, but there is an overlap of the values. Further research on those parameters is suggested.