Novel BMPR1B::AFF2 in a Sinonasal Region Non-Keratinizing Squamous Cell Carcinoma

Abstract

Introduction

Molecular diagnostic tools may help resolve difficult differential diagnoses, and in some cases, they uncover entirely novel findings.

Methods

We performed immunohistochemistry and molecular analysis on a cytologically bland squamous tumor of the sinonasal region.

Results

We found a previously unreported BMPR1B::AFF2 fusion in a difficult-to-characterize nasal and nasopharyngeal tumor exhibiting aggressive local behavior and recurrence.

Conclusion

While the DEK::AFF2 fusion is well-characterized in sinonasal region tumors, alternate fusion partners are becoming increasingly recognized.

Introduction

The widespread use of molecular diagnostic tools has revolutionized head and neck tumor classification, especially in salivary glands and the sinonasal tract. More precise, molecular-based subclassification may improve prognoses and access to targeted therapies [1]. Herein, we report a novel AFF2 fusion of a recurrent papillary squamous cell carcinoma of the nasal cavity, nasopharynx, and middle ear.

Case and Images

A 50-year-old woman with a reported history of unreviewed “inverted papilloma” presented with progressive otologic and sinonasal symptoms. CT revealed a small papule just anterior to the right eustachian tube orifice. The biopsy revealed a proliferative squamous lesion with a complex exophytic and endophytic papillary architecture. The squamous epithelium lining the papillary fragments was thickened but dyscohesive, with parabasal hyperplasia, immature swirling, occasional mitoses in the basal and parabasal layers, and an overall lack of maturation towards the surfaces (Fig. 1A–D). Rare mucocytes were seen as well. Atypical mitoses and necrosis were absent.

Fig. 1.

A The epithelial tumor is exophytic and papillary. B The fibrovascular cores are lined by variably dyscohesive squamous epithelium. C The interconnecting ribbons of squamous epithelium have basal/parabasal hyperplasia resulting in an alternating blue and pink appearance. D The tumor cell nuclei are small, round, and uniform in size and shape. The mitotic rate was low. A single mucin droplet is seen (lower middle)

By immunohistochemistry, the tumor was diffusely positive for p63 and CK5/6 and negative for p16, BRAF V600E, nuclear beta-catenin and p53 (wild-type). High-risk HPV RNA in situ hybridization was also negative, as was polymerase chain reaction for EGFR hotspot mutations. DEK::AFF2 associated squamous cell carcinoma was considered a top diagnostic consideration, so fluorescence in situ hybridization (FISH) was performed using a two-color probe set designed to span the DEK (6p22.3) and the AFF2 (Xq28) gene loci (fusion probe design; Agilent Technologies, Santa Clara, CA) using clinically validated methods. The fusion FISH assay was reported as negative (Fig. 2). As a result, a diagnosis “atypical papillary squamous epithelial neoplasm” was issued.

Fig. 2.

Interphase FISH analysis with orange-labeled DEK locus-spanning and green-labeled AFF2 locus-spanning probes reveals three AFF2 (green) signals in a subpopulation of cells (arrows), suggestive of a rearrangement of this locus

Over the following year, the patient developed worsening right-sided otorrhea, granulation tissue, recurrent ear bleeding, progressive hearing loss, and nasal obstruction. Repeat CT imaging of the sinuses and temporal bones showed complete opacification of the right external auditory canal and tympanic cavity. Trans canal exploration of the right middle ear revealed a friable, papillomatous mass completed filling the middle ear and external auditory canal, with erosion of the tympanic membrane and ossicular chain. The lesion extended into the round window niche. Surgical debulking was performed, however complete excision was not achieved due to the extent and friability of the mass.

Histological examination of the recurrent tumor revealed essentially identical findings as the biopsy (Fig. 3A). At this time, however, AFF2 protein immunohistochemistry was available, and found to be positive (Fig. 3B). Accordingly, RNA-sequencing was performed, which was initially negative for AFF2 fusions. Upon further communication with the molecular pathologist, however, it was discovered that AFF2 was not included in the extensive, but still targeted, gene panel. Upon adding AFF2, a fusion was identified between exon 1–2 of BMPR1B and exons 10–21 of AFF2. The final diagnosis was recurrent non-keratinizing squamous cell carcinoma harboring BMPR1B::AFF2. The prior FISH assay was reexamined and while indeed technically negative for fusion, was found to have three signal copies of the probe designed to span the AFF2 locus in 78 (39%) of the 200 interphase cells analyzed, suggestive of a potential rearrangement of this gene. Specifically, although this abnormal FISH pattern could simply represent a gain (ie. three copies of the AFF2 gene locus or the X chromosome), it more likely represented a split of one of the AFF2 spanning probe signals into two probe signals due to an underlying structural rearrangement of the AFF2 locus (for an overall total of three probe signals per cell when including the non-rearranged AFF2 homologue locus signal). When viewing this FISH pattern in concert with the clinicohistopathologic findings favoring a diagnosis of DEK::AFF2 associated squamous cell carcinoma, an interpretation suggesting an AFF2 rearrangement with a variant fusion partner was markedly heightened.

Fig. 3.

A The recurrent tumor had a nearly identical appearance histologically. B AFF2 immunohistochemistry was diffusely positive

Following this diagnosis, a CT scan of the neck was performed which did not reveal any evidence of metastatic disease. The patient remains under active surveillance. While radiotherapy was considered as a therapeutic option, no additional therapy has been offered to the patient.

Discussion

Since their initial descriptions, sinonasal region squamous cell carcinomas harboring DEK::AFF2 have become well-recognized [2, 3]. These carcinomas are notorious for being underdiagnosed as squamous papillomas because they are often deceptively bland at the cellular level. Although DEK::AFF2 is the most common fusion partnership, it is also becoming increasingly evident that alternate AFF2 fusion partners may be molecular drivers for head and neck cancers [4]. Breimer, et al. recently described a series of cases harboring variant AFF2 fusions, including EWSR1::AFF2, H3-3A::AFF2, NUCKS1::AFF2, and CHD4::AFF2, suggesting that AFF2 is the crucial gene for driving this family of tumors, similar to NUTM1 for NUT carcinoma [4]. Our current case adds BMPR1B to the list of alternate gene partners. In the Breimer, et al. series, some of the alternate fusion cases had morphological features identical to DEK::AFF2 cases, while others had atypical findings in the form of neuroendocrine or extensive glandular differentiation. In the current case, the findings were those of a classic DEK::AFF2 case: papillary architecture with bland cytomorphology, basal hyperplasia with swirling and dyscohesion. These features not only suggest the possibility of squamous papilloma, but also of an ameloblastic neoplasm either arising from maxillary bone or peripherally in the sinonasal tract.

Our case also highlights challenges relating to molecular confirmation of this emerging tumor. FISH was negative, because the probe was designed to identify DEK::AFF2, the only fusion partnership known when the assay was designed. A second review by an expert in cytogenetics did reveal three copies of the probe signal spanning the AFF2 gene locus suggestive of rearrangement but with a fusion partner other than DEK (variant fusion). The discovery of alternate AFF2 fusions suggests that a potential superior FISH testing strategy may be utilization of an AFF2 break apart design up front or as a reflex assay for DEK::AFF2 fusion FISH negative but clinicohistopathologically suspicious tumors to capture more cases. In our case, even RNA-sequencing was initially falsely negative, underscoring the importance of communication with the molecular laboratory to ascertain that the gene(s) of interest are indeed able to be evaluated. Ultimately, the only testing modality that was clearly positive initially was AFF2 immunohistochemistry. Although this antibody’s specificity is still under evaluation, it appears to be a highly sensitive marker for carcinomas with AFF2 fusions, regardless of the partner [4, 5]. This again may be analogous to NUT carcinoma, where NUT immunostaining has proven to be the most sensitive testing technique available, superior to FISH and sequencing.

In summary we identified a non-keratinizing squamous cell carcinoma of the sinonasal region which harbored a novel BMPR1B::AFF2 fusion. The case highlighted the strengths and weaknesses of the prevailing AFF2 fusion testing strategies, in particular, underscoring the sensitivity of AFF2 immunostaining.

Author Contributions

TRS and JA Bishop wrote the manuscript text. JA Bridge provided figures. All authors reviewed and edited the manuscript.

Funding

This study was funded by the Jane B. and Edwin P. Jenevein M.D Endowment for Pathology at UT Southwestern Medical Center. No external funding was obtained for this study.

Data Availability

No datasets were generated or analysed during the current study.

Declarations

Conflict of interest

The authors declare no competing interests.

Ethical Approval

All procedures performed in this retrospective data analysis involving human participants were in accordance with the ethical standards of the institutional review board (UT Southwestern IRB 112017-073).

Informed Consent

The IRB-approved study did not require informed consent.