Abstract
Aim
The sinonasal tract hosts numerous types of undifferentiated neoplasms, having small round cell morphology. The aim of this study was to determine whether sinonasal small round blue cell tumors (SRBCT) have distinct imaging features on computed tomography (CT), magnetic resonance imaging (MRI), and 18-fluorodeoxyglucose positron emission tomography (18F-FDG PET)/CT.
Methods
Seventy-three patients (43 male; Mage = 61.2 years) with histopathologically proven sinonasal SRBCT were retrospectively reviewed. Imaging features of SRBCTs including location, maximum dimension, margin characteristics, presence of calcification, sclerotic bone changes, intratumoral necrosis, tumor extension, bone destruction, bone remodeling, perineural spread, T1- and T2-weighted MRI signal intensity, qualitative features on diffusion-weighted imaging and 18F-FDG PET/CT, and pattern of contrast enhancement were analyzed using Fisher’s exact test or the chi-square test. The maximum standardized uptake values (SUVmax) and apparent diffusion coefficient (ADCmean) values of SRBCT were compared by utilizing the Kruskal–Wallis test.
Results
There was a significant difference between SRBCT type regarding the tumor location (p = 0.006), 18F-FDG uptake pattern (p = 0.006), involvement of the orbit (p = 0.016) and pterygopalatine fossa (p = 0.043), the presence of perineural spread (p < 0.001), bone destruction (p = 0.034), and intratumoral necrosis (p = 0.022). Bone destruction and necrosis were more common in rhabdomyosarcoma. Perineural spread was common in sinonasal adenoid cystic carcinoma (ACC). Qualitative 18F-FDG uptake features as well as tumor location were significantly different between sinonasal ACC and sinonasal undifferentiated carcinoma. The ADCmean and SUVmax values were not statistically different between SRBCT types.
Conclusions
Sinonasal SRBCTs have numerous distinct imaging features on CT, MRI, and 18F-FDG PET/CT that could be useful in the differentiation between lesions when the histopathologic diagnosis is inconclusive.
Keywords: Sinonasal malignancy, small round blue cell tumor (SRBCT), 18-fluorodeoxyglucose positron emission tomography scan (18F-FDG PET/CT), diffusion-weighted imaging (DWI), magnetic resonance imaging (MRI)
Introduction
Malignancies of the sinonasal cavity are associated with substantially greater heterogeneity compared to malignancies of the upper aerodigestive tract, where squamous-cell carcinoma (SCC) predominates.1 Even though SCC is the most frequent sinonasal neoplasm,2 there exists a growing variety of some other histologies, such as tumors of epithelial, neuroectodermal, mesenchymal, and lymphoproliferative lineage. Probably the most challenging diagnostic category of sinonasal malignancies is the small round blue cell tumor (SRBCT), which constitutes a heterogeneous group of malignant neoplasms characterized by undifferentiated tumor cells with small-sized nuclei and scant cytoplasm.3 These types of tumors share numerous overlapping histopathological features, as well as displaying substantial variation between cases.4 Differentiation of tumor subtypes is crucial, since some of them are managed by conservative medical treatment, some by local surgery, and others by chemoradiotherapy.5 Hence, it is essential to have a well-developed differential diagnosis for the sinonasal SRBCT category.
The diagnostic classification of SRBCTs of the sinonasal region is challenging for the pathologist, considering that the histopathological and immunophenotypic features could overlap.6 The diagnostic problem may be more pronounced if a small biopsy specimen is given to the pathologist.7,8 Other factors such as tumor site of origin, imaging, and clinical findings need to be combined with the histopathological analysis to achieve the correct diagnosis. It is essential to differentiate sinonasal SRBCTs because of their distinct treatment strategies.9 Therefore, noninvasive imaging techniques, such as computed tomography (CT), magnetic resonance imaging (MRI), and 18-fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT), may possibly play an important role in obtaining information about the biological properties of these tumors.10
To our knowledge, a comprehensive description of the imaging appearance of sinonasal SRBCTs using various imaging methods including diffusion-weighted imaging (DWI) and 18F-FDG PET/CT has not been reported in the literature. The aim of this study was to determine whether sinonasal SRBCTs have distinct imaging features on CT, 18F-FDG PET/CT, and MRI.
Methods
Patient population
Our Institutional Review Board waived informed consent and approved the design of this retrospective study. After performing a search of the electronic database and the pathology registry of our institution, we identified 82 patients with surgically confirmed SRBCTs of the sinonasal region who underwent CT, MRI, and 18F-FDG PET/CT from January 2009 to January 2018. All sinonasal tumors treated in our institution were discussed in the head and neck tumor boards. Six cases was excluded because of the visible artifacts from dental work. Additionally, three patients who met the inclusion criteria were excluded because their images revealed motion artifacts. From the 73 patients having MRI, apparent diffusion coefficient (ADC) calculation was not feasible in five patients due to the presence of visible artifacts from the bone–air interface. Otherwise, conventional T1- or T2-weighted images were not affected significantly for the qualitative and quantitative assessment. In a further eight patients, the tumor was not visible on the high b-value images, despite being visible on conventional T1- or T2-weighted images, thus precluding the drawing of regions of interest (ROIs). Patients were staged based on the American Joint Committee on Cancer criteria (AJCC),11 and the University of California Los Angeles staging system 12 was used for olfactory neuroblastomas (ONBs). The primary tumor anatomic subsite and the extent of the invasion were mostly determined by evaluating pretreatment 18F-FDG PET/CT imaging. CT and MRI scans, physical examination findings, operative reports, and surgical pathology reports were reviewed in cases where the tumor anatomic subsite and extent of invasion could not be determined accurately.
Technique
18F-FDG PET/CT imaging
18F-FDG PET/CT scans conducted at our institution were performed following routine preparation, involving fasting for six hours. Patients were required to have blood glucose levels of <180 mg/dL before the scan. A Siemens Biograph 16 PET/CT scanner (Siemens Medical Systems, Hoffmann Estates, IL) was performed with a continuous spiral technique using a 16-slice helical CT after intravenous injection of 3.7–5.5 MBq/kg of 18F-FDG one hour prior to PET acquisition. Patients were imaged from the vertex to the feet with the arms raised. Additionally, dedicated head and neck 18F-FDG PET/CT images were gathered with the arms down to reduce attenuation artifacts for head and neck tumors. Automated co-registration of the CT and PET scans data was conducted with commercially available software (Syngo.Via® version VA 30; Siemens Healthcare, Forchheim, Germany).
The CT data were obtained utilizing the following parameters: 50–120 mAs, 130 keV, 3 mm section width, and 5 mm table feed per rotation, and were acquired following injection of 2 mL/kg of non-ionic contrast agent (Omnipaque 300; GE Healthcare, Princeton, NJ).
MR imaging
All MR examinations were conducted on a 3-Tesla MRI scanner (Siemens MAGNETOM Trio or Siemens MAGNETOM Skyra; Siemens, Erlangen, Germany) with a 16-channel head coil. Conventional MR images and DWI were obtained in the same procedure. In every patient, T2-weighted fast spin-echo images (TR/TE/NEX, 3000–5000 ms/90–105 ms/2) and precontrast T1-weighted spin-echo images (TR/TE/NEX, 450–600 ms/10–14 ms/1) with or without fat-saturation were acquired, accompanied by contrast-enhanced T1-weighted spin-echo images after the intravenous injection of 0.1 mmol/kg of gadobutrol (Gadovist; Bayer Schering Pharma, Berlin, Germany). Images were acquired in a minimum of two planes with 3–4 mm section thickness, 0–0.4 mm intersection gap, 288 × 224 matrix, and 24 cm field of view (FOV). DWI was performed by using multislice, spin-echo, single-shot echoplanar imaging. Imaging parameters were a TR/TE of 5500/92 ms, FOV of 20 × 24 cm, 256 × 128 matrix, and section thickness of 4 mm, with an interslice gap of 1 mm. Images were obtained with a diffusion-weighted b factor of 0, and 1000 s/mm2 to obtain precise ADC maps.
Image analysis
Experienced head and neck radiologists with dedicated nuclear medicine training who were blinded to tumor histology interpreted all the images with emphasis on location, maximum dimension (<3 cm or ≥3 cm), margin (well-defined or ill-defined), tumor extension, presence of calcification, bone destruction, bone remodeling, sclerotic bone changes, perineural spread, intratumoral necrosis, signal intensity (SI) on T1- and T2-weighted MRI, qualitative imaging features on DWI and 18F-FDG PET/CT, and contrast enhancement pattern on MRI. A well-defined margin was defined as smooth or lobulated margin without spiculation in the tumor border. The presence of calcification was determined on precontrast CT scans. The SI of the lesion on T1- and T2-weighted MR images was compared to the brain stem. Enhancement pattern at the solid portions of the lesion was classified as homogeneous or heterogeneous (Figures 1 and 2). The homogeneous enhancement was defined as an even SI in >90% in the tumor. Non-enhancing areas with hypoattenuation on CT scans or hypointense T1-weighted image and hyperintense T2-weighted image areas were considered as intratumoral necrosis. DWI at b-values of 0 and 1000 s/mm2, as well as ADC maps, were interpreted qualitatively and quantitatively to establish whether the lesion displayed facilitated or restricted diffusion (Figures 1 and 2). ADCmean values were measured on ADC maps by placing ROIs over the tumors.
Figure 1.
Right nasal cavity diffuse B-cell lymphoma extending into right frontal sinus in a 59-year-old man. (a) Axial contrast-enhanced fat-saturated T1-weighted image shows homogeneous contrast enhancement (arrow); (b) axial b = 1000 s/mm2 diffusion-weighted imaging (DWI) and(c) apparent diffusion coefficient (ADC) map reveal restricted diffusion (arrows) in the mass.
Figure 2.
Right maxillary sinus malignant melanoma in a 71-year-old man. (a) Axial contrast-enhanced fat-saturated T1-weighted image shows heterogeneous contrast enhancement (arrow); (b) axial b = 1000 s/mm2 DWI and (c) ADC map reveal facilitated diffusion (arrows) in the mass.
Images were specifically assessed for findings of perineural tumor spread (PNTS) on the basis of Ginsberg’s criteria,13 including bone erosion, sclerotic margins, and widening of the normal diameter of the foramina, fissures, or canals where nerves normally traverse the skull base, and replacement of normal perineural fat with tumor, enhancement of the nerves, and increased size of the nerve in question.14
The degree of 18F-FDG uptake was categorized visually as avid or mild (Figures 3 and 4). The 18F-FDG uptake pattern was also grouped as homogeneous or heterogeneous. For the semi-quantitative analysis of 18F-FDG uptake, the maximum standardized uptake value (SUVmax) was calculated by generating a ROI analysis over the most 18F-FDG-avid portion of the lesion.
Figure 3.
Right nasal cavity adenoid cystic carcinoma in a 75-year-old man. (a) CT image shows soft-tissue mass (arrow). (b) 18-fluorodeoxyglucose positron emission tomography 18F-FDG PET/computed tomography (CT) image reveals mild FDG uptake (maximum standardized uptake value (SUVmax)=3.36; arrow).
Figure 4.
Right nasal cavity malignant melanoma in a 71-year-old man. (a) Axial T2-weighted magnetic resonance image shows soft-tissue mass extending into the ethmoid sinuses (arrow). (b) 18F-FDG PET/CT image reveals intense FDG avidity (SUVmax=18.0; arrow).
Statistical analysis
The Kolmogorov–Smirnov (K-S) test was performed to evaluate conformity to a normal distribution. All data were revealed to be parametric with a normal distribution. Insufficient previous data were available to guide a sample size calculation for this study. Fisher’s exact test or the chi-square test was conducted to compare the frequencies of imaging findings. Corrections for multiple comparisons were performed using the modified Benjamini–Hochberg correction method with a false discovery rate of 5%. The Kruskal–Wallis test with Dunn’s post hoc comparisons was used to detect any statistically significant difference between ADCmean and SUVmax values in tumor types. All statistical analyses were performed with commercially available software (IBM SPSS Statistics for Windows v23.0; IBM Corp., Armonk, NY). The null hypotheses of no difference were rejected if p-values were <0.05.
Results
A total of 73 patients formed the basis of the current study (43 male; Mage = 61.2 years; range 20–93 years). The mean maximum transverse diameter (±SD) of the tumors was 4.6 ± 1.7 cm. Tumor types and locations are described in Table 1. CT, 18F-FDG PET/CT, and MRI features of SRBCTs are described in Tables 2 and 3.
Table 1.
Patient characteristics.
| Tumor type (n) | Characteristic (number of patients or
mean) |
|||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Sex |
Age | Site of primary tumor |
T stage |
|||||||||
| Male | Female | Nasal Cavity | Maxillary S. | Sphenoid S. | Ethmoid S. | Frontal S. | T1 | T2 | T3 | T4 | ||
| ACC (n = 12) | 10 | 2 | 56.3 | 1 | 8 | 1 | 1 | 1 | 1 | 5 | 3 | 3 |
| Lymphoma (n = 10) | 4 | 6 | 64.4 | 8 | 2 | 0 | 0 | 0 | 1 | 3 | 3 | 3 |
| M. Melanoma (n = 8) | 3 | 5 | 73.2 | 7 | 1 | 0 | 0 | 0 | 1 | 3 | 3 | 1 |
| Nasopharyngeal Ca. (n = 1) | 0 | 1 | 66 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 | 1 |
| NUT midline (n = 1) | 1 | 0 | 20 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| ONB (n = 9) | 5 | 4 | 53 | 9 | 0 | 0 | 0 | 0 | 4 | 2 | 0 | 3 |
| Rhabdomyosarcoma (n = 10) | 7 | 3 | 69.3 | 7 | 0 | 1 | 1 | 1 | 1 | 3 | 3 | 3 |
| SCC (n = 15) | 9 | 6 | 63.8 | 6 | 7 | 2 | 0 | 0 | 0 | 2 | 7 | 6 |
| Small-cell carcinoma (n = 1) | 1 | 0 | 67 | 1 | 0 | 0 | 0 | 0 | 0 | 0 | 0 | 1 |
| SNUC (n = 6) | 4 | 2 | 47.6 | 3 | 0 | 1 | 2 | 0 | 0 | 0 | 0 | 6 |
ACC: adenoid cystic carcinoma; M., malignant; Ca.: carcinoma; NUT: nuclear carcinoma of the testis; ONB: olfactory neuroblastoma; SCC: squamous-cell carcinoma; SNUC: sinonasal undifferentiated carcinoma; S.: sinus.
Table 2.
Imaging features of tumors, including the invasion of the adjacent tissue, presence of calcification, bone destruction, bone remodeling, and sclerotic bone changes.
| Tumor type | ACF | Orbita | Ptery. | Cavernous | Expansion | Destruction | Sclerosis | Calcification |
|---|---|---|---|---|---|---|---|---|
| ACC | 2/12 (17%) | 4/12 (33%) | 7/12 (58%) | 1/12 (8%) | 10/12 (83%) | 8/12 (67%) | 3/12 (25%) | 1/12 (8%) |
| Lymphoma | 2/10 (20%) | 5/10 (50%) | 0/10 (0%) | 0/10 (0%) | 6/10 (60%) | 6/10 (60%) | 0/10 (0%) | 0/10 (0%) |
| M. melanoma | 1/8 (12%) | 3/8 (37%) | 1/8 (12%) | 0/8 (0%) | 6/8 (75%) | 6/8 (75%) | 0/8 (0%) | 1/8 (12%) |
| Nasopharyngeal Ca. | 1/1 (100%) | 0/1 (0%) | 1/1 (100%) | 0/1 (0%) | 1/1 (100%) | 1/1 (100%) | 0/1 (0%) | 0/1 (0%) |
| NUT midline | 1/1 (100%) | 1/1 (100%) | 1/1 (100%) | 0/1 (0%) | 1/1 (100%) | 1/1 (100%) | 0/1 (0%) | 0/1 (0%) |
| ONB | 4/9 (44%) | 0/9 (0%) | 1/9 (11%) | 1/9 (11%) | 6/9 (67%) | 3/9 (33%) | 2/9 (22%) | 0/9 (0%) |
| Rhabdomyosarcoma | 3/10 (30%) | 1/10 (10%) | 4/10 (40%) | 1/10 (10%) | 10/10 (100%) | 10/10 (100%) | 0/10 (0%) | 2/10 (20%) |
| SCC | 5/15 (33%) | 10/15 (67%) | 7/15 (47%) | 3/15 (20%) | 11/15 (73%) | 12/15 (80%) | 1/15 (7%) | 0/15 (0%) |
| Small-cell carcinoma | 0/1 (0%) | 1/1 (100%) | 1/1 (100%) | 0/1 (0%) | 1/1 (100%) | 1/1 (100%) | 0/1 (0%) | 0/1 (0%) |
| SNUC | 5/6 (83%) | 4/6 (67%) | 3/6 (50%) | 1/6 (17%) | 5/6 (83%) | 5/6 (83%) | 0/6 (0%) | 0/6 (0%) |
| p-Value | 0.072 | 0.016 | 0.043 | 0.876 | 0.660 | 0.034 | 0.336 | 0.584 |
ACF: anterior cranial fossa; Ptery., pterygopalatin fossa.
Table 3.
CT, MRI, and 18F-FDG PET/CT features of tumors.
| Tumor type | Margin | Enhancement P. | Necrosis | PNTS | T1-weighted image | T2-weighted image | DWI | FDG U. | FDG pattern |
|---|---|---|---|---|---|---|---|---|---|
| ACC | Well-defined: 8 | Homo: 4 | 7/12 | 10/12 | Hypo: 8 | Hypo: 3 | Non-restricted: 5 | Mild: 0 | Homo: 3 |
| Ill-defined: 4 | Hetero: 8 | Iso: 3 | Iso: 3 | Restricted: 5 | Avid: 7 | Hetero: 4 | |||
| Hyper: 1 | Hyper: 6 | ||||||||
| Lymphoma | Well-defined: 6 | Homo: 5 | 2/10 | 0/10 | Hypo: 3 | Hypo: 0 | Non-restricted: 1 | Mild: 0 | Homo: 7 |
| Ill-defined: 4 | Hetero: 5 | Iso: 5 | Iso: 7 | Restricted: 8 | Avid: 9 | Hetero: 2 | |||
| Hyper: 2 | Hyper: 3 | ||||||||
| M. Melanoma | Well-defined: 6 | Homo: 0 | 6/8 | 1/8 | Hypo: 4 | Hypo: 1 | Non-restricted: 1 | Mild: 0 | Homo: 3 |
| Ill-defined: 2 | Hetero: 8 | Iso: 1 | Iso: 4 | Restricted: 4 | Avid: 7 | Hetero: 4 | |||
| Hyper: 3 | Hyper: 3 | ||||||||
| Nasopharyngeal Ca. | Well-defined: 0 | Homo: 0 | 0/1 | 1/1 | Hypo: 0 | Hypo: 1 | Non-restricted: 1 | Mild: 0 | Homo: 1 |
| Ill-defined: 1 | Hetero: 1 | Iso: 1 | Iso: 0 | Restricted: 0 | Avid: 1 | Hetero: 0 | |||
| Hyper: 0 | Hyper: 0 | ||||||||
| NUT midline | Well-defined: 1 | Homo: 0 | 1/1 | 1/1 | Hypo: 0 | Hypo: 1 | NA | Mild: 0 | Homo: 0 |
| Ill-defined: 0 | Hetero: 1 | Iso: 1 | Iso: 0 | Avid: 1 | Hetero: 1 | ||||
| Hyper: 0 | Hyper: 0 | ||||||||
| ONB | Well-defined: 8 | Homo: 2 | 3/8 | 2/9 | Hypo: 6 | Hypo: 1 | Non-restricted: 2 | Mild: 1 | Homo: 5 |
| Ill-defined: 1 | Hetero: 7 | Iso: 3 | Iso: 3 | Restricted: 4 | Avid: 7 | Hetero: 3 | |||
| Hyper: 0 | Hyper: 5 | ||||||||
| Rhabdomyosarcoma | Well-defined: 4 | Homo: 1 | 9/10 | 4/10 | Hypo: 6 | Hypo: 2 | Non-restricted: 2 | Mild: 0 | Homo: 3 |
| Ill-defined: 6 | Hetero: 9 | Iso: 4 | Iso: 2 | Restricted: 6 | Avid: 5 | Hetero: 2 | |||
| Hyper: 0 | Hyper: 5 | ||||||||
| SCC | Well-defined: 12 | Homo: 3 | 10/15 | 8/15 | Hypo: 7 | Hypo: 1 | Non-restricted: 2 | Mild: 0 | Homo: 7 |
| Ill-defined: 3 | Hetero: 12 | Iso: 7 | Iso: 11 | Restricted: 12 | Avid: 11 | Hetero: 4 | |||
| Hyper: 1 | Hyper: 3 | ||||||||
| Small-cell carcinoma | Well-defined: 1 | Homo: 0 | 0/1 | 1/1 | Hypo: 1 | Hypo: 0 | Non-restricted: 0 | Mild: 0 | Homo: 1 |
| Ill-defined: 0 | Hetero: 1 | Iso: 0 | Iso: 0 | Restricted: 1 | Avid: 1 | Hetero: 0 | |||
| Hyper: 0 | Hyper: 1 | ||||||||
| SNUC | Well-defined: 5 | Homo: 1 | 3/6 | 3/6 | Hypo: 3 | Hypo: 1 | Non-restricted: 0 | Mild: 0 | Homo: 6 |
| Ill-defined: 1 | Hetero: 5 | Iso: 3 | Iso: 3 | Restricted: 6 | Avid: 6 | Hetero: 0 | |||
| Hyper: 0 | Hyper: 2 | ||||||||
| p-Value | 0.287 | 0.387 | 0.022 | <0.001 | 0.381 | 0.125 | 0.193 | 0.841 | 0.006 |
CT: computed tomography; MRI: magnetic resonance imaging; 18F-FDG PET: 18-fluorodeoxyglucose positron emission tomography; P.: pattern; PNTS: perineural tumor spread; U.: uptake; Homo: homogeneous; Hetero: heterogeneous.
No significant differences regarding the mean age or sex were determined based on tumor types (p > 0.05). Most tumors presented at an advanced stage; 64% (47/73) were staged T3 or T4.
There was a significant difference between tumor types regarding the location (p = 0.006), PNTS (p < 0.001), bone destruction (p = 0.034), intratumoral necrosis (p = 0.022), 18F-FDG uptake pattern (p = 0.006), orbita invasion (p = 0.016), and pterygopalatine fossa invasion (p = 0.043). No significant differences were noted between tumor types regarding qualitative MR signal intensities on T1-weighted imaging (p = 0.381) and T2-weighted imaging (p = 0.125), qualitative diffusivity characteristics on ADC map (p = 0.193), margin characteristics (p = 0.287), invasion of the anterior cranial fossa (p = 0.072) and cavernous sinus (p = 0.876), accompanying calcification (p = 0.584), sclerosis (p = 0.336), bone remodeling (p = 0.660), contrast enhancement pattern (p = 0.387), and 18F-FDG avidity (p = 0.841).
The ADCmean and SUVmax values were not different between tumor types according to the Kruskal–Wallis test (p = 0.055 and p = 0.443, respectively).
According to the tumor location, maxillary sinuses were the most common sites for adenoid cystic carcinoma (ACC; 8/12; 66%; p < 0.001) and SCC (7/15; 46%; p = 0.002), whereas the nasal cavity was the most common site for ONB (9/9; 100%; p < 0.001). Sinonasal undifferentiated carcinomas (SNUCs) originated mostly from ethmoid sinuses (2/6; 33%; p < 0.001). Intratumoral necrosis was more commonly observed in rhabdomyosarcoma than in lymphoma (90% vs. 20%; p < 0.001). PNTS was more common in ACC than in lymphoma (83% vs. 0%; p < 0.001). ACC showed markedly heterogeneous 18F-FDG uptake in 6/7 (86%) cases, whereas none of the SNUCs (0/10) had heterogeneous 18F-FDG uptake (p = 0.003). The frequency of bone destruction was significantly higher in rhabdomyosarcoma (10/10; 100%) than ONB (6/9; 66%; p < 0.001). The frequency of orbital invasion was significantly higher in SCC (10/15; 66%) than ONB (0/9; 0%) and rhabdomyosarcoma (1/10; 10%; p < 0.001). The frequency of pterygopalatine fossa invasion was significantly higher in ACC (7/12; 58%) than lymphoma (0/10; 0%), malignant melanoma (1/8; 12%), and ONB (1/9; 11%; p = 0.015).
Discussion
The current study suggests that an imaging approach may offer complementary diagnostic yield in sinonasal SRBCT. The tumors in our series tended to be large, expansile, and highly aggressive. Without distinguishing the different histology, our results showed a wide variation of SUVmax (range 2.1–54.3) and ADCmean (range 0.043–2.24 × 10−3 mm2/s). Rhabdomyosarcoma in our series demonstrated a highly aggressive appearance such as adjacent tissue invasion and bone destruction. The frequency of bone destruction and intratumoral necrosis were significantly higher in rhabdomyosarcoma, which is in accordance with previous reports.15,16 Kato et al.17 evaluated CT and MRI features of maxillary sinus ACCs. In their cohort of maxillary sinus ACCs, PNTS was an uncommon imaging finding. However, pterygopalatine fossa invasion and PNTS were commonly seen in our series of sinonasal ACCs.
A substantial difference in SUV parameters between sinonasal tumor types was revealed by Ozturk et al.,18 with SNUC and SCC showing very high SUV values. In another study by Gencturk et al.,19 potential contributions of quantitative DWI parameters were evaluated in sinonasal neoplasms. ADCmean was found to be significantly higher in ACC than SCC, lymphoma, SNUC, and neuroendocrine carcinoma. However, the current study revealed that ADCmean and SUVmax values could be overlapped between various SRBCT types. Grouping all SRBCT types could have made the projection of results difficult due to the smaller number of patients when assessing the role of quantitative 18F-FDG PET/CT and DWI features.
In our cohort, none of the patients with ONB had orbital invasion, which is relatively low compared to reported rates of orbital invasion. In a previous study by Lund et al.,20 frank orbital involvement was found in four (10%) patients, and invasion of orbital periosteum was found in eight (19%) patients. In another study by Rimmer et al.,21 9/95 patients with ONB were found to have orbital periosteum invasion, three patients had an invasion of the orbital contents, and four patients had an invasion of the globe itself. Compared to the above references, the frequency of orbital invasion in our cohort with ONB was relatively low. There are several possible explanations for this seemingly conflicting result. In our study, orbital periosteum invasion without radiological evidence of orbital involvement was not regarded as an orbital invasion. Also, 6/9 ONB cases in our series were at the early stages (T1–T2).
Tumors of the skull base and sinonasal region pose a difficult diagnostic challenge, and differentiation of malignant tumors from benign lesions are the cornerstones of appropriate management. Abdel Razek et al.22,23 analyzed the DWI in assessing malignant and benign skull base and sinonasal lesions. They concluded that DWI could be a promising, noninvasive approach to differentiate malignant tumors from benign skull-base and sinonasal lesions and to determine the pathological grade of malignant tumors. Advanced imaging modalities, including dynamic susceptibility contrast perfusion-weighted MRI (DSC-pMRI) and diffusion tensor imaging (DTI), may also contribute to the differentiation of malignant from benign salivary gland tumors.24 In a study by Yu et al.,25 the fractional anisotropy (FA) values of malignant salivary gland tumors were found to be significantly higher than those of benign tumors (0.26 ± 0.06 vs. 0.17 ± 0.05, respectively). In another study by Abdel Razek et al.,26 the mean FA and mean diffusivity (MD) of malignant salivary gland tumors (0.41 ± 0.07 and 0.89 ± 0.15 × 10−3 mm2/s) were significantly higher than those of benign tumors (0.19 ± 0.07 and 1.28 ± 0.42 × 10−3 mm2/s, respectively). The DSC-pMRI was utilized to determine the major salivary gland tumors as well.27 The dynamic susceptibility contrast percentage threshold value of 26.5% was found to be useful in differentiating malignant from benign major salivary gland tumors, with an area under the curve of 0.96. In our study, a combination of 18F-FDG PET/CT and DWI increased the accuracy of sinonasal SRBCT characterization. Further multicenter studies with a large number of patients using advanced imaging modalities such as DTI and DSC-pMRI may improve characterization of the sinonasal SRBCT.
The strengths of this retrospective study involve strict inclusion criteria and a detailed analysis of pathologic specimens (including immunohistochemistry for each case) by the experienced head and neck pathologist.28,29 Immunohistochemically, cytokeratin, CK7, CK8, and EMA immunoreactivity are useful for distinguishing nonkeratinizing SCC from other SRBCTs. In SNUCs, the tumor cells are immunoreactive for cytokeratins, with no amplification of Epstein–Barr virus RNA. CD56 staining is an important immunohistochemical finding in neuroendocrine tumors. The presence of fibrillary cell processes, Homer–Wright rosettes, and S100-positive cells are histological findings in favor of ONB. Diffuse staining for S-100 and HMB45 could be used to discriminate malignant melanoma. Rhabdomyosarcoma usually expresses myogenic markers such as MyoD1 and desmin. Immunophenotypically, diffuse B-cell lymphoma could be differentiated from other SRBCTs using B-lymphocyte markers of CD20+, CD79+, CD3–, and CD56.3,7
Despite these interesting preliminary results, several limitations of our study must be taken into account. First, a statistical analysis of different rhabdomyosarcoma, lymphoma, and ACC subtypes was not performed due to the small sample size of each tumor type. Second, there was an imbalance between the number of available tumor types, including a low number of small-cell carcinoma (n = 1), nasopharyngeal carcinoma (n = 1), and nuclear carcinomna of the testis (n = 1). Third, each patient was recruited from a head and neck cancer center, and thus most of the SRBCT in our series demonstrated advanced local stage at the time of diagnosis, which might partly explain the high rate of PNTS.30,31
In conclusion, sinonasal SRBCTs have numerous distinct imaging features on CT, MRI, and 18F-FDG PET/CT, which could be used to suggest more likely diagnostic consideration. Even though confirmation of these tumors still depends on tissue sampling, the immunohistochemical analysis may not provide definitive information due to overlapping features and should therefore be used in panels with a list of differential diagnoses guided by the radiological findings.
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Conflict of interest
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
References
- 1.Ozturk K, Gawande R, Gencturk M, et al. Imaging features of sinonasal tumors on positron emission tomography and magnetic resonance imaging including diffusion weighted imaging: a pictorial review. Clin Imaging 2018; 51: 217–228. [DOI] [PubMed] [Google Scholar]
- 2.Kim SH, Mun SJ, Kim HJ, et al. Differential diagnosis of sinonasal lymphoma and squamous cell carcinoma on CT, MRI, and PET/CT. Otolaryngol Head Neck Surg 2018; 159: 494–500. [DOI] [PubMed] [Google Scholar]
- 3.Bridge JA, Bowen JM, Smith RB. The small round blue cell tumors of the sinonasal area. Head Neck Pathol 2010; 4: 84–93. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Franchi A. An update on sinonasal round cell undifferentiated tumors. Head Neck Pathol 2016; 10: 75–84. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Robin TP, Jones BL, Gordon OM, et al. A comprehensive comparative analysis of treatment modalities for sinonasal malignancies. Cancer 2017; 123: 3040–3049. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tilson MP, Bishop JA. Utility of p40 in the differential diagnosis of small round blue cell tumors of the sinonasal tract. Head Neck Pathol 2014; 8: 141–145. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Thompson LD. Small round blue cell tumors of the sinonasal tract: a differential diagnosis approach. Mod Pathol 2017; 30: S1–S26. [DOI] [PubMed] [Google Scholar]
- 8.Chang S, Hirschowitz S, Lu DY, et al. Fine-needle aspiration of bilateral small round blue cell tumors in neck in a 65-year-old man: significance of a wider differential. Cytojournal 2018; 15: 7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Robbins KT, Ferlito A, Silver CE, et al. Contemporary management of sinonasal cancer. Head Neck 2011; 33: 1352–1365. [DOI] [PubMed] [Google Scholar]
- 10.Sen S, Chandra A, Mukhopadhyay S, et al. Imaging approach to sinonasal neoplasms. Neuroimaging Clin N Am 2015; 25: 577–593. [DOI] [PubMed] [Google Scholar]
- 11.Lydiatt WM, Patel SG, O’Sullivan B, et al. Head and Neck cancers—major changes in the American Joint Committee on Cancer eighth edition cancer staging manual. CA Cancer J Clin 2017; 67: 122–137. [DOI] [PubMed] [Google Scholar]
- 12.Tajudeen BA, Arshi A, Suh JD, et al. Esthesioneuroblastoma: an update on the UCLA experience, 2002–2013. J Neurol Surg B Skull Base 2015; 76: 43–49. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Ginsberg LE, DeMonte F. Imaging of perineural tumor spread from palatal carcinoma. AJNR Am J Neuroradiol 1998; 19: 1417–1422. [PMC free article] [PubMed] [Google Scholar]
- 14.Ong CK, Chong VF. Imaging of perineural spread in head and neck tumours. Cancer Imaging 2010; 10 Spec no A: S 92–98. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Wang X, Song L, Chong V, et al. Multiparametric MRI findings of sinonasal rhabdomyosarcoma in adults with comparison to carcinoma. J Magn Reson Imaging 2017; 45: 998–1004. [DOI] [PubMed] [Google Scholar]
- 16.Kim JH, Yoon DY, Baek S, et al. CT, MRI, and FDG PET/CT findings of sinonasal sarcoma: differentiation from squamous cell carcinoma. J Korean Soc Radiol 2016; 75: 17–25. [Google Scholar]
- 17.Kato H, Kanematsu M, Sakurai K, et al. Adenoid cystic carcinoma of the maxillary sinus: CT and MR imaging findings. Jpn J Radiol 2013; 31: 744–749. [DOI] [PubMed] [Google Scholar]
- 18.Ozturk K, Gencturk M, Caicedo-Granados E, et al. Utility of FDG PET/CT in the characterization of sinonasal neoplasms: analysis of standardized uptake value parameters. AJR Am J Roentgenol 2018; 211: 1354–1360. [DOI] [PubMed] [Google Scholar]
- 19.Gencturk M, Ozturk K, Caicedo-Granados E, et al. Application of diffusion-weighted MR imaging with ADC measurement for distinguishing between the histopathological types of sinonasal neoplasms. Clin Imaging 2019; 55: 76–82. [DOI] [PubMed] [Google Scholar]
- 20.Lund VJ, Howard D, Wei W, et al. Olfactory neuroblastoma: past, present, and future? Laryngoscope 2003; 113: 502–507. [DOI] [PubMed] [Google Scholar]
- 21.Rimmer J, Lund VJ, Beale T, et al. Olfactory neuroblastoma: a 35-year experience and suggested follow-up protocol. Laryngoscope 2014; 124: 1542–1549. [DOI] [PubMed] [Google Scholar]
- 22.Razek AA, Sieza S, Maha B. Assessment of nasal and paranasal sinus masses by diffusion-weighted MR imaging. J Neuroradiol 2009; 36: 206–211. [DOI] [PubMed] [Google Scholar]
- 23.Abdel Razek A, Mossad A, Ghonim M. Role of diffusion-weighted MR imaging in assessing malignant versus benign skull-base lesions. Radiol Med 2011; 116: 125–132. [DOI] [PubMed] [Google Scholar]
- 24.Abdel Razek AAK. Routine and advanced diffusion imaging modules of the salivary glands. Neuroimaging Clin N Am 2018; 28: 245–254. [DOI] [PubMed] [Google Scholar]
- 25.Yu J, Du Y, Lu Y, et al. Application of DTI and ARFI imaging in differential diagnosis of parotid tumours. Dentomaxillofac Radiol 2016; 45: 20160100. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Khalek Abdel Razek AA. Characterization of salivary gland tumours with diffusion tensor imaging. Dentomaxillofac Radiol 2018; 47: 20170343. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Abdel Razek AA, Samir S, Ashmalla GA. Characterization of parotid tumors with dynamic susceptibility contrast perfusion-weighted magnetic resonance imaging and diffusion-weighted MR imaging. J Comput Assist Tomogr 2017; 41: 131–136. [DOI] [PubMed] [Google Scholar]
- 28.Li Q, Cui W, Abulajiang G, et al. Application of immunohistochemistry in the diagnosis of small round blue-cell tumors of soft tissue. Clin Lab 2014; 60: 1383–1392. [DOI] [PubMed] [Google Scholar]
- 29.Simons SA, Bridge JA, Leon ME. Sinonasal small round blue cell tumors: an approach to diagnosis. Semin Diagn Pathol 2016; 33: 91–103. [DOI] [PubMed] [Google Scholar]
- 30.Dercle L, Hartl D, Rozenblum-Beddok L, et al. Diagnostic and prognostic value of 18F-FDG PET, CT, and MRI in perineural spread of head and neck malignancies. Eur Radiol 2018; 28: 1761–1770. [DOI] [PubMed] [Google Scholar]
- 31.Gil Z, Carlson DL, Gupta A, et al. Patterns and incidence of neural invasion in patients with cancers of the paranasal sinuses. Arch Otolaryngol Head Neck Surg 2009; 135: 173–179. [DOI] [PubMed] [Google Scholar]