Tenosynovial giant cell tumors in unusual locations detected by positron emission tomography imaging confused with malignant tumors: report of two cases

A tenosynovial giant cell tumor (T-GCT) is a benign, synovial tumor arising from the synovium, bursae, or tendon sheath. It can be intra- or extra-articular and localized or diffuse. A localized extra-articular T-GCT is synonymous with a giant cell tumor of the tendon sheath; it usually occurs in the hand (other sites include the wrist, ankle/foot, and knee) and is more often seen in women than men [1], with a reported annual incidence of about one in 50,000 [2]. Diffuse T-GCT is considered as a locally aggressive and predominantly involves the knee, followed by the hip, ankle, elbow, and shoulder [3]; it has a reported annual incidence of approximately two cases per million, mostly younger than 40 years of age, with no gender predilection [4]. Unusual sites of T-GCTs include extra-articular neck and hip [5, 6].

Positron emission tomography (PET) with fluorine-18 fluorodeoxyglucose (FDG) with computed tomography (FDG PET/CT) is widely used to differentiate malignant from benign tumors and to detect distant metastasis [7, 8]. However, as on FDG is not cancer-specific, false-positive findings of benign diseases have been reported in active inflammation or infection [9‐11]. Schulte et al. reported a 15 % false-positive rate of malignancy among 200 tumor and tumor-like lesions of the bone [12]. The appearance of T-GCT in FDG PET/CT was not fully reported because this modality is not routinely used. However, there are reports of varied FDG uptake (maximum standardized uptake value [SUVmax] = 3.43–25 g/ml) in T-GCTs, which mimic malignant tumors [13‐16].

In this article, we discuss two cases of T-GCT that developed in unusual locations and were initially thought to be malignant tumors on FDG PET/CT, but were histologically confirmed as T-GCT.

T-GCT is a benign soft tissue tumor that arises from the synovium, bursae, or tendon sheath that typically contains areas of intermediate and/or low signal intensity on T1- and T2WIs. A T-GCT is usually diagnosed by MRI [17]. In our patients, the MRI findings were compatible with T-GCTs, even though the masses were located in unusual sites. Moreover, FDG PET/CT is not frequently employed for T-GCT, as the findings cannot be confirmed. Hamada et al. analyzed FDG PET/CT images of 56 soft tissue tumors (19 malignant and 37 benign tumors) and found statistically significant difference in the SUVmax between malignant and benign lesions in early scans (5.50 ± 5.32 g/ml versus 3.10 ± 2.64 g/ml). Moreover, the intensities of benign soft tissue lesions were <6.0 g/ml [18]. Broski et al. reported a mean SUVmax of 8.7 (range, 4.0–14.5) g/ml for FDG PET/CT imaging of 14 T-GCTs, and concluded that a T-GCT could be intensely hypermetabolic, mimicking musculoskeletal metastasis [15]. Selby et al. reported a pigmented villonodular synovitis of the shoulder mimicking a metastatic melanoma [16]. Lindkog et al. reported false-positive results of PDG PET/CT in five cases of benign soft tissue lesions, including two T-GCT lesions, and mentioned the importance of understanding the limitations of PET, including false-positive findings of benign lesions [11]. Hamada et al. reported a positive correlation of FDG SUVmax with the expression of GLUT-1 and hexokinase II in 49 musculoskeletal tumors [19]. However, the expression patterns of GLUT-1 and hexokinase II in T-GCTs have not been fully evaluated. Therefore, the mechanism of high uptake of FDG PET/CT in T-GCT remains uncertain. Although FDG PET/CT is a useful imaging modality for patients with a history of malignancy and the demand has been increasing, the chance of detecting benign lesions is also increased. When a T-GCT is detected in a typical location and MRI findings are compatible to T-GTC, careful observation without histological confirmation is reasonable. However, if a suspected T-GCT is detected in an unusual location, a biopsy should be performed for differentiation from a malignant tumor. Recently, there have been new developments in the medical treatment of T-GCT using a selective colony-stimulating factor 1 receptor (CSF1R) kinase inhibitor [20]. FDG PET/CT might be a useful imaging modality in monitoring this treatment [21].

In conclusion, we encountered two cases of T-GCTs that developed in unusual locations, which were confused with malignant tumors. The final diagnoses of these cases were histologically confirmed as T-GCTs. It is important to realize that a T-GCT may show high FDG uptake, which can resemble from a malignant tumor. Although MRI findings and the location might help to diagnose a T-GCT, careful assessment is mandatory.

Written informed consent was obtained from Patient #1 and from the parents or legal guardians of Patient #2 for publication of these case reports and any accompanying images. Copies of the written consent forms are available for review from the Section Editor of this journal.