18F-Fluorodeoxyglucose-positron Emission Tomography/Computed Tomography Imaging of Metastatic Nasopharyngeal Cancer with Emphasis on the Distribution of Bone Metastases.

Distant metastases change the prognosis of patients with nasopharyngeal carcinoma (NPC) which most commonly metastasizes to the bone. 18F-fluorodeoxyglucose (18F-FDG) positron emission tomography/computed tomography (PET/CT) is reported as useful in imaging NPC metastases. Our study assesses the incidence and distribution of bone metastases detected by 18F-FDG PET/CT in NPC. 717 18F-FDG PET/CT scan reports of histologically proven NPC patients imaged in Singapore General Hospital, Singapore, between 2003 and 2009 were reviewed for the total number of metastases (scanned from vertex to mid-thigh) and analyzed for distribution. Of the 709 FDG avid metastases in these reports, 357/709 (50.35%) were locoregional nodal metastasis and 352/709 (49.65%) were distant metastases of which 192/709 (27.08%) of total metastases and 54.54% of distant metastases (192/352) were in the bones. The majority of the bone lesions 125/192 (65.1%) were in the axial skeleton with 109/192 (56.77%) in the lower skeleton (thoracolumbar spine, sacrum, and pelvis). The incidence of bone metastases in our study (27.08%) was higher than that reported in other studies, for example, 15% by Liu et al. and 11% (230 patients) by Caglar et al. Bone metastases have been reported in the femurs and the feet and as such some metastases may have been outside the field of view of the scans. In our study, 27% of FDG avid NPC metastases are in the bones.

Introduction

Nasopharyngeal carcinoma (NPC) is an epithelial neoplasm that shows endemic distribution. The highest prevalence rates are in the South-Eastern Asia including Malaysia, Indonesia, Singapore, and South-Eastern China.[1] It is considered a rare tumor of head and neck occurring in 10–30 persons per 100,000 people per year.[2] The etiology of the disease is linked with the Epstein–Barr virus (EBV).[3]

NPC has the highest rate of lymph node and/or distant metastasis among the head and neck tumors[4] with distant metastasis most commonly affecting the bones (approaching 70–80%) followed by the liver (30%) then the lungs (18%).[356] Bone metastases are reported to be the most frequent site of distant failure causing 30% of deaths in advanced disease with studies suggesting that the sites and number of bone metastasis may be prognostic indicators for survival.[7] Growing evidence shows that long-term survival and complete response can be obtained among a small proportion of patients with bone metastasis, especially those patients with a solitary lesion who received aggressive treatment.[8] Hence, the detection of distant metastases including bone metastases is crucial in the management and prognosis of NPC patients.

NPC is usually investigated by physical examination and direct nasopharyngeal endoscopy with biopsy. Locoregional involvement is evaluated by computed tomography (CT) or magnetic resonance imaging (MRI) scans. Distant metastases are assessed by plain chest radiographs, ultrasonography of the liver and bone scintigraphy.[9] Recently, positron emission tomography (PET)/CT with 18F-fluorodeoxyglucose (18F-FDG) is performed as a routine procedure for diagnosing, staging, recurrence, and follow-up of NPC patients. In addition, it is used to assess local residual disease and response to treatment.

The aim of this study is to evaluate the imaging of the distribution of bone metastases in NPC with 18F-FDG PET/CT.

Materials and Methods

The patients

We reviewed the reports of 18F-FDG PET/CT scans of histologically proven NPC patients performed in the Nuclear Medicine and PET Department, Singapore General Hospital, Singapore, between 2003 and 2009 using the Nuclear Medicine Information System. The study was approved by the SingHealth Centralized Institutional Review Board. The scans were interpreted by nuclear medicine physicians.

We excluded all patients with additional second malignancies from the study. We analyzed the reports of patients with more than one PET/CT scan between 2003 and 2009 individually. We analyzed the data from the reports for the following:

Study indication (staging, response to treatment, recurrence, and follow-up)

Locoregional nodal involvement

Distant metastases, if any and their sites

Sites of any bone lesions were carefully assessed and further analyzed for distribution.

18F-fluorodeoxyglucose positron emission tomography/computed tomography scan

All patients fasted overnight or at least 6 h prior to the scan and received between 333 and 518 MBq (9 and 14 mCi) of 18F-FDG calculated based on body weight. PET/CT system (Biograph, Siemens, LSO crystals) was utilized for image acquisition. A low dose single slice CT with a slice thickness of 5 mm, KVp from 80 to 130, and mAs range 200–400 was done from the vertex of the skull to mid-thigh followed by PET acquisition in the same area at 3 min per bed position. The CT data were utilized for attenuation correction and anatomical correlation. The PET data underwent iterative reconstruction (24 subsets, 3 iterations). Intravenous contrast was given when requested by the referring physician.

Metastases determination

We considered all metastasis to be present if noted on the report. We considered all the equivocal lesions as negative though follow-up reports within the targeted period were reassessed and were used to upgrade the equivocal lesions if later turn to be positive.

Results

Patient demographics

A total of 722 reports of 18F-FDG PET/CT scans performed on NPC patients imaged between 2003 and 2009 were reviewed. Five patients with additional concurrent primary malignancies were excluded, and a total of 717 reports (628 patients) were included in the final analysis. Five hundred and fourteen patients were male and 203 were female [Table 1]. The majority were between 45 and 55 years of age.

Table 1

Age distribution of patients

The ethnic groups of the patients are listed in [Table 2].

Table 2

The patient ethnic groups

The noted indications for the FDG PET/CT studies are listed in [Table 3].

Table 3

The indication for the scan

The total number of FDG avid metastatic lesions in these reports was 709. Of these, 357/709 (50.35%) were locoregional nodal metastases (retropharyngeal 115, cervical 226, and supraclavicular 16) [Table 4].

Table 4

The distribution of locoregional nodes

Distant metastases represented 352 (49.65%) of the 709 lesions. Of these, 192 were bone lesions (27.08%).

The distribution of bone lesions is cervical spine (13/192, 6.77%), clavicles (7/192, 3.65%), ribs (23/192, 11.98%), sternum (8/192, 4.17%), scapulae (13/192, 6.77%), humerus (6/192, 3.13%), thoracic spine (26/192, 13.54%), lumbar spine (27/192, 14.06%), sacrum (28/192, 14.58%), pelvic bones (28/192, 14.58%), and femora (13/192, 6.77%) [Table 5].

Table 5

The distributions of bone metastases

Discussion

NPC is an epidermoid-origin malignant neoplasm classified by the World Health Organization in two histological types of differentiated nonkeratinizing carcinoma (type II) and undifferentiated carcinoma (type III).[6] It is a squamous cell carcinoma with very aggressive behavior, high incidence of locoregional lymph node, and distant metastasis at presentation.

Although that NPC is considered radiosensitive and curable, the 1-year mortality rate can reach 90% when distant metastasis is present at the presentation.[9] In patients with NPC, the presence of distant metastasis influences prognosis and treatment, so accurate evaluation of distant metastases is essential for the management.

Our study showed that about half of the metastatic lesions are at distant sites and that 192 (27%) of the 709 FDG avid metastases were in the bones.

Our incidence of bone metastases is slightly higher than other existing studies, for example, 15% by Liu et al.[6] and 11% (230 patients) by Caglar et al.[10]

The majority of the bone lesions 125/192 (65%) are in the axial skeleton with 109/192 (57%) in the lower skeleton (thoracolumbar spine, sacrum, and pelvis) Figure 1.

Figure 1

18F-fluorodeoxyglucose positron emission tomography/computed tomography scan at the same transaxial level (a-c) show large fluorodeoxyglucose avid sclerotic lesion in the right pelvic bone and another small lesion in the sacrum

Bone metastases have even been reported in the femora and the feet[11] and as such some metastases in our study may have been outside the field of view of the scans.

18F-FDG PET/CT is reported as useful in imaging NPC bone metastases[312] though there is controversy when comparing its usefulness relative to other modalities.

Vellayappan et al. conclude that for newly diagnosed NPC, 18F-FDG PET/CT is accurate in N and M staging but not T staging. They recommend using 18F-FDG PET/CT, together with MRI of the nasopharynx, in routine staging of NPC.[13]

A study comparing 18F-FDG PET/CT with 18F-sodium fluoride (18F-NaF) PET/CT and bone scintigraphy showed that 18F-NaF PET/CT is superior in detecting bone metastases followed by 18F-FDG PET/CT which was superior to technetium-99m based bone scintigraphy.[14]

Yang et al. showed that although 18F-FDG PET/CT is superior at lesion level, there was no statistical difference between 18F-FDG PET/CT and planar bone scintigraphy in detecting bone metastases in NPC.[12]

Recent meta-analysis by Wei et al. comparing 18F-FDG PET/CT, MRI, and single photon emission computed tomography (SPECT) in the diagnosis of local residual/recurrent NPC showed that SPECT and PET/CT are superior to MRI for the detection of local residual/recurrent NPC, and the addition of CT to PET will not significantly improve the diagnostic accuracy. They also concluded that 201TI-SPECT and MIBI-SPECT have the same diagnostic accuracy.[15]

Huang et al. evaluated the differences in prognostic values of static and dynamic PET/CT in NPC. The study concluded that the tumor volume from the static scan is useful in NPC prognosis, but they could not justify the role of dynamic scanning in their small study population.[16]

The PET/CT is also having a role in the management of NPC patients who develop distant metastasis after initial radiation therapy. Chang et al. showed that combining 18F-FDG PET/CT with aggressive treatment approach using locoregional modalities could be of benefit to NPC patients with favorable prognostic factors, even after distant metastasis.[17]

The role of PET in evaluating NPC is growing through the use of new PET radiotracers, for example, 68Ga-DOTA-TOC,[18] 68Ga-DOTA-NOC,[19] 11C-Choline,[20] and 18F-FLT[21] which can evaluate bone metastases from different aspects.

Prognostic biomarkers, for example, osteopontin[22] can have a crucial role in imaging and the management of bone metastases in NPC patients as well.

The study had few limitations as it is a retrospective study. Moreover, we did not account for the number of patients instead of the number of scans, albeit a minority did have multiple scans. Furthermore, the distant metastases were not proven histologically and we did not consider the SUVmax measurements. A recent study done by Xiao et al. showed that SUVmax at the primary site can be helpful biomarker in predicting distant metastasis of NPC patients treated with intensity modulated radiation therapy. Furthermore, combining SUVmax of the primary site with the overall tumor stage could be more precise to predict treatment outcome.[23]

Another limitation for the study is not considering the ethnicity of the patients and other factors in regards to bone metastases distribution.

Future prospective study with a large sample size is recommended to correlate metastases with the EBV titers and the treatment modality which will be of great benefit in the management of patients with NPC.

Conclusion

18F-FDG PET/CT has a role in the management of NPC and in evaluating bone metastases. A prospective study may help further assess the evolving role of PET/CT in NPC bone metastases particularly in light of the development of new PET/CT scanners, advances in scanning/processing technology, and the growing use of new PET tracers and prognostic biomarkers.

Financial support and sponsorship

Nil.

Conflicts of interest

There are no conflicts of interest.