Abstract
Background: Biphenotypic sinonasal sarcoma (BSNS) is a rare spindle cell sarcoma distinctly arising in the sinonasal area, with dual myogenic and neural differentiation, and characterised by the presence of PAX3 gene fusion, typically with MAML3. Although the majority may be indolent, up to 25% of cases reported in the literature are locally aggressive, with invasion of adjacent critical structures in the head and neck region. Case Report: We report 3 cases of BSNS reviewed at our institution between 2016-2020 in addition to the current literature. Patient 1 underwent surgery followed by adjuvant radiotherapy but relapsed 24 months later and was not fit for systemic anticancer therapy and managed with palliative care. Due to comorbidities, patient 2 was recommended for active surveillance, with a view to intervening with radiotherapy should there be evidence of clinical progression. At 60 months, the nasal cavity mass remained stable on serial imaging. Patient 3 underwent primary surgical R0 resection and was offered adjuvant post operative radiotherapy 60 Gy/30 fractions/6 weeks but opted for active surveillance and has no clinical or radiological evidence of recurrence 22 months after surgery. Conclusion: The primary management for BSNS is surgical resection. We recommend discussing the role of postoperative adjuvant radiotherapy 60 Gy/30 fractions/6 weeks in patients who are fit for treatment. In clinical practice, dose levels will be constrained by surrounding normal tissues. At present, the role of systemic anticancer therapy is undefined. A prospective registry of ultra-rare cases may provide an evidence base with which to select optimal treatment strategies for BSNS in the future.
Biphenotypic sinonasal sarcoma (BSNS) is a rare, low-grade spindle cell sarcoma with dual myogenic and neural differentiation, and fascicular/herringbone architecture (1, 2). The biphenotypic nature may arise due to characteristic PAX3 gene fusions, which are involved in both skeletal muscle and neural crest development. BSNS was first described less than a decade ago and histological diagnosis remains a challenge, especially in institutions without access to ancillary molecular diagnostics facilities. Due to the morphologic and immunophenotypic diversity of BSNS, tumours were likely previously mistaken for various other, both benign and malignant, mesenchymal neoplasms. The histologic differential diagnosis of BSNS is broad and includes low grade malignant peripheral nerve sheath tumour (LG-MPNST), solitary fibrous tumour (SFT), hemangiopericytoma, mesenchymal neoplasms with myoid or rhabdomyoblastic differentiation, and other fusion-associated sarcomas (1-4).
Around 100 cases have been described to date. BSNS is more frequently described in females (3:1) and the mean age of onset is in the 5th decade of life (1, 3, 4). There is a varied natural history and although the majority may be indolent, up to 25% of cases are locally aggressive, with extra-sinonasal extension into adjacent structures such as the orbit (25%) and cribriform plate (10%).
Clinical presentation. BSNS typically presents with symptoms of nasal obstruction, sinusitis, facial pressure, epistaxis or purulent rhinorrhoea. The most common sites of disease are the nasal cavity and ethmoid sinus, but BSNS can also arise from the sphenoid, maxillary, and frontal sinuses (1-4). Tumours can extend intracranially and involve the cribriform plate and orbit. Radiological diagnosis is made using magnetic resonance imaging (MRI) and computed tomography (CT), with T2 MRI typically showing a hypointense to isointense soft tissue mass infiltrating the bone whilst CT often displays a hyperostotic bone reaction (1-4).
Histopathological features and immunophenotyping. The characteristic histopathologic features of BSNS are relatively cellular neoplasms composed of spindle cells in intersecting fascicles or herringbone patterns, with an infiltrative tumour edge, and myogenic and neural differentiation with immunohistochemistry (1, 2, 4, 5). Hypercellularity and hemangiopericytoma-like staghorn vessels have also been described. Immunohistochemistry for S100 protein and smooth muscle actin (SMA) is consistently positive, but SOX10 is generally negative. Around 60% of tumours harbour PAX3-MAML3 fusions (t[2; 4][q35;q31.1]) (6-8) although alternate PAX3 fusion partners include FOXO1, WWTR1, NCOA1 and NCOA2. PAX3 is a transcription factor that plays a critical role in both skeletal muscle and neural crest development and blocks terminal differentiation, which likely accounts for the dual phenotype of BSNS. Tumours display a low to intermediate grade histology with a low mitotic rate and a lack of tumour necrosis.
Clinical management. Current published clinical experience on the diagnosis, management, and clinical outcomes of BSNS relies on case reports and small case series (1, 2, 4-7, 9-17).
Chitguppi et al. published a literature review that included published case reports and series of patients diagnosed with BSNS between 2012-2019 (1). A total of 7 published series with 95 patients and one new case report were included. Patient demographics, presenting symptoms, anatomical subsite, immunotyping, staging investigations, treatment details, and survival outcomes were reviewed. Primary surgical resection with or without postoperative radiotherapy were the most frequent treatment modalities (1). No patients underwent adjuvant chemotherapy. However, long-term follow up outcome data were only available for 36% of patients (34/95 patients) (1). In patients with follow up data, no cases of distant metastatic relapse were reported but local recurrence was described in around a third of patients (11/34 patients), with a mean age of 49 years (range=24–69 years). In the cohort of patients who relapsed, 4 patients underwent salvage surgery, with 1 patient also receiving post operative radiotherapy. No details of management of local recurrence were reported in 7 cases.
The efficacy of adjuvant chemotherapy for BSNS is unknown as there is currently no published data on systemic anticancer therapy specific to this sarcoma subtype.
We report 3 cases of BSNS reviewed at our institution between 2016-2020.
Case Report
Case 1. A 71-year-old woman presented with a 3-month history of nasal obstruction and facial pressure. She had a past medical history of an ethmoid schwannoma resected 7 years earlier and a history of ovarian cancer treated with primary surgery and adjuvant chemotherapy 16 years prior. Other comorbidities included osteoporosis, asthma, and hypertension. Her mobility was severely limited due to osteoporosis requiring a wheelchair with a WHO performance status of 2.
MRI sinuses showed a 20×30 mm mass arising in the left posterior nasal cavity protruding posteriorly into the nasopharynx lying along the medial margin of the left medial pterygoid. The mass involved the nasal septum and crossed the midline (Figure 1A and B).
Magnetic resonance imaging (MRI) scans of sinuses with intravenous contrast. Well defined, non-invasive 2×3 cm soft tissue mass arising in the left nasal cavity, which extends as far as the pterygoid plates and posteriorly across the midline. Isointense on T1 and T2 MRI sequences. A) AXIAL MRI T1 fast spin echo (FSE) fat suppression (F/S) + Gadolinium (GAD); B) Coronal MRI Short T-1 inversion recovery (STIR) sequence.
She underwent primary surgical resection with an involved resection margin (R1). Final histopathology showed left and right septal tumour fragments, altogether measuring 40 mm. Sections showed respiratory-lined mucosa overlying a cellular tumour composed of short fascicles and whorls of ovoid and uniform spindle cells with hyperchromatic nuclei, scant neoplasm, and indistinct cell borders. Numerous intervening dilated and branching blood vessels were noted. The mitotic index was 7/10 high power fields (HPF) with no necrosis demonstrated. The tumour showed strong and diffuse positivity for TLE1, CD56, BCL-2 and beta-catenin, with weak focal staining for AE1/3 and desmin. There was only patchy S100 protein, and PAX5 was weakly positive. SMA and CD34 were negative.
Postoperative MRI and staging CT of the thorax/abdomen/pelvis did not show any residual or metastatic disease. The patient underwent adjuvant radiotherapy to the ethmoid sinus and nasal cavity, 60 Gy/30 fractions/6 weeks with intensity modulated radiotherapy (IMRT). She tolerated treatment well apart from Grade 1 fatigue.
Unfortunately, 24 months following adjuvant radiotherapy she experienced local in-field relapse. Restaging MRI showed a 3.9×2.0 cm soft tissue mass filling the nasopharynx, which displayed enhancement and restriction on diffusion weighted MRI, with the anterior margin abutting the posterior wall of the right maxillary antrum. Restaging CT of the thorax/abdomen/pelvis did not show evidence of distant metastases. She underwent further palliative debulking surgery with good symptomatic benefit (improvement in symptoms of nasal blockage and voice quality). Due to a reduced WHO performance status of 3, palliative chemotherapy was not thought to be suitable, and she therefore continued with supportive care and died 8 months later.
Case 2. An 84-year-old woman was referred following an incidental finding of a left paranasal sinus mass diagnosed at the time of medical investigations following a transient ischaemic attack (TIA). She was asymptomatic from the left paranasal mass. Past medical history included nasal surgery for a polyp 40 years ago, hypertension, gout, and cataracts. WHO Performance status was 1. On clinical examination, there was a mild left-sided proptosis but with normal cranial nerve function and visual fields.
MRI and CT sinuses showed a diffusely enhancing, solid polypoid mass arising in the left nasal air space invading the left maxillary antrum, measuring 4.3×2.5×4.2 cm. CT scan showed curvilinear areas of calcification. The mass displaced the bony nasal septum to the right, involving the medial wall of the orbit with displacement of the left optic nerve causing the observed proptosis. Superiorly, the mass abutted the lamina papyracea with suggestion of disruption but no intracerebral invasion. Staging CT did not show any distant metastases.
Biopsy showed a cellular spindle cell tumour infiltrating bone, expressing SMA and S100 protein, in the absence of cytokeratins and melanocytic markers. SOX10 immunostaining was negative, with focal nuclear positivity for beta catenin.
Due to comorbidities, the multidisciplinary team (MDT) recommended active surveillance, initially with 3 monthly MRI scans of the head and neck and a chest x-ray, with a view to intervening with radiotherapy should there be evidence of clinical progression. Over 60 months, the nasal cavity mass remained stable on serial imaging with no appreciable changes. The patient remains clinically stable and continues with surveillance with 6 monthly MRI of the head and neck and chest x-ray.
Case 3. A 53-year-old female presented with symptoms of nasal discomfort. She was fit and well with no significant co-morbidities and a WHO performance status of 0. She subsequently underwent MRI of the head and neck that showed a large right sided sinonasal mass.
Biopsies showed a spindle cell neoplasm with dual neural and focal myoid immunophenotype, with features favouring biphenotypic sinonasal sarcoma. A focal cluster showed cells with a more ‘rhabdomyoblast-like’ phenotype, with abundant eosinophilic cytoplasm. Immunohistochemistry was multifocally strongly positive for SMA, and for S100 protein (including tumour nuclei). There was also focal strong bcl-2 and cytoplasmic beta-catenin, more limited focal CD56, and weak focal CD99 immunostaining. The tumour was negative for SOX10, MelanA, h-caldesmon, MyoD1, AE1/AE3, CAM5.2, CK7, BerEP4, EMA, STAT6, and WT1. The MIB proliferation fraction was low. There was no evidence of a translocation involving the FOXO1 gene at 13q14 or the SS18 gene at 18q11.2 with fluorescent in-situ hybridization (FISH).
She underwent primary surgical resection via lateral rhinotomy and craniofacial approach with resection of the lamina papyracea, periorbita, cribiform plate, and dura. Frozen sections at the time of surgery were negative and the resection was classified R0 (negative resection margin). The MDT recommendation was to offer adjuvant post operative radiotherapy of 60 Gy/30 fractions/6 weeks. However, the patient declined and opted for active surveillance. There is no clinical or radiological evidence of recurrence 22 months after surgery and active surveillance continues with MRI of the face and neck and CT of the chest/abdomen/pelvis.
Discussion
Biphenotypic sarcoma is a very rare and distinct sinonasal malignancy with unique clinicopathological features when compared to other head and neck sarcomas (Figure 2). It is now possible to distinguish BSNS as a separate pathological entity with advanced molecular techniques. Table I summarises the clinicopathological features in our case series.
Histological Features of biphenotypic sinonasal sarcoma. A) The tumour is composed of cellular fascicles of uniform spindle cells arranged in a herring-bone pattern. B) The cells have ovoid vesicular nuclei, indistinct nucleoli, and scanty cytoplasm. C) Immunohistochemistry for S100 protein is positive in many nuclei of lesional cells. D) Immunohistochemistry for MyoD1 is focally positive in nuclei.
Clinicopathological features.
Distinctive pathological features of BSNS include a herringbone pattern, hemangiopericytoma-like staghorn vessels, and “pseudo-gland formation” due to the entrapment of hyperplastic respiratory epithelium. However, none of these features are exclusive to BSNS and a range of sinonasal malignancies including low-grade malignant peripheral nerve sheath tumour (LG-MPNST), glomangiopericytoma, cellular schwannoma, solitary fibrous tumour, and malignant melanoma can mimic BSNS. Immunophenotyping is a pre-requisite for diagnosis, and it is important to be aware of a range of diagnostic myogenic and neural immunomarkers to guide accurate diagnosis and management. Of note, alveolar rhabdomyosarcoma shares a similar driving genetic alteration with BSNS, involving the PAX3 gene, and can occur in sinonasal sites (6-8). However, whilst alveolar rhabdomyo-sarcoma is often characterised by a high risk of distant metastases and poor prognosis, BSNS appears to have a much more indolent clinical course. Recurrences tend to occur within 5 years of diagnosis with locally aggressive disease.
Table II provides a summary of published literature reviews and case series for BSNS (1, 2, 4-7, 9, 11, 13-17). In a combined literature review of 8 publications including case series/case studies of patients with BSNS, 36% (34/95) of patients were followed up for a median duration of 4.61 years (range=3 months-28 years) (1). Of these, 32% (11/34) of patients experienced a local recurrence with a median age of 49 years (range=24-69 years). The median time to recurrence was 2.4 years (1). Of these, treatment details were only available for four patients who underwent salvage surgery, one of whom also received adjuvant post-operative radiotherapy. Overall, specific treatment details were poorly described in the literature. Radiotherapy dose was only described in two cases. No details were reported of specific systemic anticancer therapy regimes either as adjuvant or palliative treatment.
Summary of literature reviews and case series for biphenotypic sinonasal sarcoma.
Primary treatment is definitive surgery aiming for an en bloc R0 surgical resection where technically feasible and patient of adequate performance status. Although the benefit of adjuvant radiotherapy in this setting remains uncertain, a discussion is recommended based on the patient’s age, performance status, proximity to critical structures, and potential challenges posed by salvage surgery in the event of recurrent disease. Therefore, we recommend discussing the role of adjuvant radiotherapy of 60 Gy/30 fractions/6 weeks in patients with WHO Performance Status 0-2 who are fit for treatment. Published doses of radiotherapy have been at the 60 Gy in 30-fraction level. There are not enough published data on radiotherapy and specifically radiotherapy dosage in this disease to make inference on an ideal dose/fractionation. In clinical practice, dose levels will be constrained by surrounding normal tissues. The uncertain benefits of adjuvant radiotherapy must be balanced against the established risks of radiation when the tumour volume lies in close proximity to critical normal tissue structures, e.g., optic apparatus and brain stem.
The optimal clinical target volume (CTV) remains undefined but is based on the adaptation of principles of both head and neck cancer and sarcoma target volume delineation. This includes the reconstructed gross tumour volume (GTV) with a 1.5 cm margin in all planes and adapted according to fascial planes and involved neighbouring soft tissue. The CTV to PTV margin in usually 0.3-0.5 cm based on local set up variability. Co-registration of staging MRI and planning CT can enhance target volume definition.
In patients with inoperable disease, it is reasonable to consider short-term surveillance in the first instance to monitor tumour behaviour, as in asymptomatic patients with very indolent disease, delaying treatment with primary radiotherapy may avoid unnecessary treatment-related toxicity.
Following treatment, we recommend active surveillance with clinical review and MRI of the face and neck and a chest x-ray every 3 months in years 1-2, every 6 months in years 3-5, and annually in years 6-10. As the propensity for distant metastases is rare, we think it is reasonable to reserve re-staging with CT of the chest/abdomen/pelvis for patients reporting symptoms suspicious for metastases or if abnormalities are detected on chest radiography.
Our small case series shows varied clinical outcomes with different approaches to individual patient management, from active surveillance to combined modality treatment. Radiotherapy was considered in all three cases although only delivered in one, the only case where recurrence was seen to date.
Further research is needed to improve our understanding of the pathological and molecular basis for tumour development and progression in BSNS, with the aim of risk stratification to identify patients that may benefit from adjuvant radiotherapy. A clearer understanding of tumour biology and molecular drivers would also provide insight into potential druggable targets and the development of future therapeutic treatments.
Conclusion
At present, the optimal treatment modality for BSNS remains inconclusive due to the lack of published detailed treatment information, long-term clinical outcome, and survival data. Surgery remains the main treatment modality where technically feasible and patient of adequate performance status. The benefit of adjuvant radiotherapy remains uncertain but should be discussed taking into consideration individual patient factors such as age, performance status, proximity to critical structures, and feasibility of salvage surgery in the event of disease recurrence. The role of systemic anticancer therapy is undefined. As clinical trials in rare sarcoma types are challenging to deliver, a prospective registry of ultra-rare cases may help improve our understanding of tumour biology and patterns of clinical behaviour in order to develop an evidence base with which to select optimal treatment strategies in rare sarcoma subtypes.
Footnotes
Authors’ Contributions
ABM and YA conceived and designed the analysis; YA collected the data; SZ, RLJ, CB, RS, CF, KT, MH contributed data and the writing of the paper. All Authors reviewed and approved the final manuscript.
Conflicts of Interest
The Authors have no conflicts of interest to declare in relation to this study.
- Received May 3, 2022.
- Revision received October 28, 2022.
- Accepted October 31, 2022.
- Copyright © 2023 International Institute of Anticancer Research (Dr. George J. Delinasios), All rights reserved.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.