F-18 Fluoro-2-Deoxyglucose Positron Emission Tomography (PET)/Computed Tomography (CT) Imaging in Melanoma: Normal Variants, Pitfalls, and Artifacts

Abstract

Multimodality imaging has revolutionized diagnostic imaging for several oncologic pathologies including melanoma. Although F-18 fluoro-2-deoxyglucose positron emission tomography/ computed tomography [18F]FDG PET/CT has a high sensitivity in stage III and IV melanoma, several normal variants, and imaging pitfalls may result in falsely increased or reduced tracer uptake that may negatively impact diagnostic accuracy. In addition to normal physiologic tracer uptake, differences in the biological and molecular characteristics of different types of melanoma are also responsible for pitfalls. For instance, [18F]FDG PET/CT has a low sensitivity for detecting brain metastases due to normal physiologic [18F]FDG uptake in brain tissue while hepatic metastases from cutaneous melanoma are more [18F]FDG-avid than hepatic metastases from uveal melanoma. With the introduction of immunotherapies for melanoma, treatment response assessment using [18F]FDG PET/CT has a reduced specificity. This is due to hypermetabolic immune-related adverse effects such as hepatitis, dermatitis, and colitis resulting in false-positive uptake. In addition, immune therapy-induced initial increase in tumor uptake followed by disease response (pseudo-progression) is a cause of false-positive scan interpretation. Specific technical artifacts impact disease detection in [18F]FDG PET/CT melanoma imaging. The identification of small metastatic lymph nodes and lung nodules may be limited by the resolution of the PET/CT camera (partial volume effect). Computed tomography (CT) attenuation correction results in less apparent skin and subcutaneous lesions. Pictorial illustrations will be central to this paper for the description of these normal variants, imaging artifacts, and pitfalls. It is critical for the imaging specialist to have a clear understanding of these potential limitations of ¹⁸F-FDG PET/CT imaging in individuals who are referred with melanoma.

Keywords: FDG; PET; PET/CT; [18F]FDG; artifacts; melanoma; pitfalls; variants.

Figure 1

[18F]FDG PET/CT for staging malignant melanoma of the upper lip (A) and back (B–D). Maximum intensity projection (MIP) shows symmetrical brown fat uptake in the cervical, supraclavicular, mediastinal, and paraspinal regions (A). Fused sagittal (B) and trans-axial (C) images show increased tracer uptake in the 6th, 7th, and 8th anterior-lateral consecutive ribs with callus formation noted on CT. These findings represent fractures from previous trauma and not metastatic disease. Generalized muscle uptake (D) was noted in a patient who inadvertently injected his insulin on the morning of the scan.

Figure 2

[18F]FDG PET/CT for suspected recurrent cutaneous melanoma. Fused trans-axial image shows a focus of uptake in the left gluteal region (arrow) mimicking subcutaneous involvement. There was no associated CT change. This focus corresponds to recent intramuscular injection site as confirmed from the clinical history.

Figure 3

Cutaneous melanoma of the right thigh with known inguinal nodal metastases. [18F]FDG PET/CT shows rim enhancing parietal (A) and occipital (E) brain lesions with no discernable metabolic activity (B,C,F,G). Brain metastases was confirmed on MRI (D,H).

Figure 4

Recurrent melanoma with a 9 mm left lung nodule (arrows). There is no metabolism on the [18F]FDG PET/CT trans-axial fused (A) and PET only (B) images. CT features are suspicious for malignancy (C).

Figure 5

Recurrent cutaneous melanoma with widespread metastases. [18F]FDG PET/CT coronal image shows intense avidity of liver lesions (arrow).

Figure 6

Malignant melanoma of the left heel in a known retroviral disease patient. Fused [18F]FDG PET/CT images show cervical (A), submandibular (B), and abdominopelvic lymphadenopathy (C,D). Based on their proximity to the primary site and intense hypermetabolism, the abdominopelvic lymph nodes represent metastatic sites. Conversely, the low grade cervical and submandibular lymph nodes most likely represent retroviral disease.

Figure 7

Metastatic malignant melanoma diagnosed from biopsy of inguinal lymph nodes. [18F]FDG PET/CT was done to identify the primary site of disease. The MIP image (A) shows bilateral inguinal lymphadenopathy and no other sites of metabolically active disease. However, close inspection of the non-attenuation corrected images (B) reveals a small subcutaneous focus in the lateral aspect of the right upper thigh, the likely site of primary disease. This focus is poorly visualized on the corresponding corrected images (C).

Figure 8

Patient with recurrent metastatic cutaneous melanoma had an [18F]FDG PET/CT for re-staging. Fused (A) and PET-only (B) trans-axial images show a bulky nodular thyroid gland with multiple calcifications. The largest nodule is in the left lobe (arrows) with intense hyper-metabolism (SUV_max 20.9) and associated calcifications. Fine needle aspiration cytology of the thyroid nodules was suspicious for papillary thyroid carcinoma (Bethesda Category V). Histology from total thyroidectomy specimen confirmed the diagnosis of a secondary malignancy and patient subsequently received radioactive iodine therapy.

Figure 9

[18F]FDG PET-only (A) and fused PET/CT trans-axial (B) images show increased diffuse subcutaneous uptake (SUV_max 6.03) in a case of invasive malignant melanoma of the left foot. Patient recently had a wider excision after initial excision showed disease extension to the resection margins. Maximum intensity projection (C) of a left forearm melanoma post-excision 2 months prior to PET scan shows linear subcutaneous uptake in the excision site (SUV_max 3.8). Focal uptake in the right cubital fossa corresponds to the site of radiotracer injection.

Figure 10

[18F]FDG PET-only sagittal image shows diffusely increased bone marrow uptake (arrows). Patient is a stage 4 melanoma 6 weeks post chemotherapy.

Figure 11

Hyper-progression: Malignant melanoma with failed response to chemotherapy. [18F]FDG PET/CT prior to initiation of ipilimumab shows abdominopelvic lymphadenopathy (A). Repeat PET/CT (B) after two cycles of ipilimumab shows disease progression which is persistent after subsequent treatment cycles (C,D). Immunotherapy was ultimately terminated and other treatment options explored.

Figure 12

MIP image showing tracer extravasation following multiple injection attempts in the right forearm. This intense areas of uptake can mask pathologic cutaneous, subcutaneous and nodal foci, reducing scan sensitivity.

Figure 13

Metallic prosthetic right shoulder implant with overcorrection artifact resulting in mildly increased uptake (red arrows) on the fused (A,B) and attenuation corrected (C) images. There is no demonstrable peri-prosthetic uptake on the non-attenuation corrected image (D). The implant produces streak artifacts (yellow arrows) best appreciated on the CT images (E,F).

Figure 14

Right axillary lymph node misregistration artifact due to limb motion (A–C). Intense uptake in the right axilla does not conform to a specific anatomical structure (white arrows) while the axillary lymph node demonstrates low grade uptake (yellow arrow). Misregistration between PET and CT images (D–F) due to voluntary head motion.

Figure 15

Attenuation correction artifact of a chemotherapy port (A,B) results in cutaneous uptake on the corrected image (C) which can be mistaken for cutaneous or subcutaneous metastatic disease. There is no uptake on the corresponding non-attenuation corrected image (D).