2021 NPB-10

This case was originally published in 2021. The information provided in this case was accurate and correct at the time of initial program release. Any changes in terminology since the time of initial publication may not be reflected in this case.

A 56-year-old man with a history of neurofibromatosis type 1 presented with left lower extremity pain, weakness, and an enlarging thigh mass. MRI of the left upper leg showed an 8.7 x 4.9 x 3.9 cm spindle-shaped, soft tissue mass associated with the sciatic nerve. The mass was hypermetabolic on PET scan with a maximal standardized uptake value of 6.1. Innumerable small cutaneous and subcutaneous nodules were also noted on imaging. Given the hypermetabolic appearance on PET scan and slightly increased growth from imaging one year prior, there was clinical concern for malignancy, and the patient underwent en bloc excision of the mass. In addition to the case stains, H3K27me3 showed retained nuclear immunoreactivity in the tumor cells.

Tissue Site
Left thigh

The whole slide image provided is an H&E-stained section from left thigh mass excision.

Neurofibromatosis type 1 (NF1) is an autosomal dominant tumor predisposition syndrome caused by a germline loss of function mutation in the NF1 gene on chromosome 17. The encoded protein, neurofibromin, is a GTPase-activating protein for RAS oncogenes; its loss of function results in constitutive activation of RAS signaling, which promotes cell survival and proliferation. Somatic loss of the remaining wild-type NF1 allele in Schwann cells results in neurofibroma formation in NF1 patients, who may exhibit anywhere from a few to innumerable neurofibromas. While neurofibromas are benign, a subset will undergo transformation to malignant peripheral nerve sheath tumor (MPNST). Plexiform neurofibromas, which are virtually pathognomonic of NF1, have a greater propensity for malignant transformation than other neurofibromas. An estimated 8% to 16% of NF1 patients will eventually develop MPNST. MPNST has a poor prognosis, with a high incidence of recurrence and metastasis and a five-year survival rate of only 20% to 50%. Complete wide surgical resection with negative margins is the only potential cure for MPNST. Ideally, removal of suspicious tumors before malignant transformation may prevent MPNST from developing. Plexiform neurofibromas that cause pain, demonstrate significant growth during the adult years, and/or are FDG-avid on PET scans are concerning, and surgical resection may be indicated. Terminology and histologic criteria for neurofibromas in the process of transforming have historically been a challenge, with the terms atypical neurofibroma and low-grade MPNST used inconsistently. A 2016 consensus meeting on the pathology of NF1-associated atypical nerve sheath tumors proposed the term atypical neurofibromatous tumor of uncertain biologic potential (ANNUBP) for neurofibromas with some concerning features that do not meet the criteria for MPNST. ANNUBP is defined as a neurofibroma with two or more of the following worrisome features: nuclear atypia, hypercellularity, loss of neurofibroma architecture, mitotic activity >1/50 but <3/10 high-power fields. Loss of neurofibroma architecture is defined as a fascicular growth pattern and/or loss of the CD34-positive fibroblastic network. The term ANNUBP may also be used in the setting of a small biopsy with worrisome features that do not meet criteria for MPNST. Accordingly, the term ANNUBP does not signify a discrete tumor type, but rather a clinical setting in which additional tumor resection or sampling, molecular testing, and/or clinical correlation and follow-up may be necessary.

In the case presented here, ANNUBP is the best diagnosis. H&E-stained sections show a plexiform neurofibromatous tumor (Image A) with variable cellularity (Image B). The majority of the tumor shows classic neurofibroma morphology, with wavy spindled cells situated in a background of “shredded carrots” collagen and myxoid material (Image C). However, in the more central regions of the tumor, areas of hypercellularity (Image D) and nuclear pleomorphism (Image E and Image F) are seen. While neurofibromas typically show significant S100 and SOX-10 immunoreactivity, atypical tumors, and even more so MPNSTs, may show decreased staining. In this case, S100 still shows significant immunoreactivity, albeit perhaps slightly less in the hypercellular regions (Image G). Classic neurofibromas typically show a rich CD34-positive fibroblastic network, often described as a “fingerprint” pattern. ANNUBPs may show loss of normal neurofibroma architecture which can be demonstrated by loss of this CD34 network or tumor cells assuming a fascicular growth pattern. In this case, an intact CD34 network can be seen in the areas of neurofibroma (upper left), while there is loss of CD34 staining in the atypical areas (lower right, Image H). While mitotic figures are difficult to find in this case, with only one possible mitotic figure identified after extensive examination, the Ki67 proliferative index is somewhat elevated, estimated up to ~10% (Image K). In this case, the neurofibroma displays areas of hypercellularity, nuclear atypia, and loss of neurofibroma architecture as evidenced by loss of the CD34 staining network, which is consistent with ANNUBP, but it does not show sufficiently elevated mitotic activity or necrosis to qualify it as an MPNST. It is important to emphasize that neurofibromas with only hypercellularity (cellular neurofibroma) or only nuclear atypia (“ancient neurofibroma”) are still considered benign and do not qualify for ANNUBP designation (Table 1).

The molecular underpinnings of the neurofibroma-to-MPNST transformation have been recently elucidated. As mentioned above, the loss of both NF1 alleles results in neurofibroma formation. Loss of CDKN2A/B, or less often SMARCA2, occurs during the transformation of neurofibroma to ANNUBP. CDKN2A/B loss can be illustrated by loss of p16 expression, as is seen in the hypercellular regions of this tumor (Image I). The addition of TP53, EGFR, and/or polycomb repressive complex 2 (PRC2) subunit mutations are thought to drive MPNST transformation. In this case, a p53 immunostain highlights only a few scattered cells, consistent with a TP53 wild-type staining pattern (Image J). Mutations in PRC2 subunits, including EED and SUZ12, are seen in approximately 70% of MPNSTs. Complete loss of H3K27me3 immunoreactivity in the tumor cells with preserved staining in non-neoplastic cells can be used as a surrogate marker for PRC2 subunit mutations and is characteristic of some MPNSTs but is not seen in neurofibromas or ANNUBPs. In this case, H3K27me3 staining was retained (not shown), consistent with a diagnosis of ANNUBP. It is important to note, however, that a mosaic staining pattern for H3K27me3 is nonspecific.

The differential diagnosis of ANNUBP includes cellular neurofibroma, neurofibroma with atypia (“ancient neurofibroma”), hybrid neurofibroma/schwannoma, and low-grade MPNST; the former three are all benign. Cellular neurofibromas display hypercellularity with no other atypical features. Care should be taken not to designate tumors with a dense inflammatory cell infiltrate as hypercellular. Neurofibromas with atypia display nuclear pleomorphism with no other atypical features. Hybrid neurofibroma/schwannoma is characterized by cellular schwannian nodules within a classic neurofibroma. Low-grade MPNST shares many features with ANNUBP but displays elevated mitotic activity (3-9 mitoses/10 HPFs). While not generally in the differential of ANNUBP, it is important to note that other tumors in the differential of MPNST, including desmoplastic melanoma and synovial sarcoma, may also show complete loss of H3K27me3 staining. In addition, cellular schwannomas may show elevated mitotic activity and extensive p53 immunoreactivity but are still considered benign; these tumors do not show loss of H3K27me3 staining.

Thus far, most studies have shown a low risk of recurrence and no risk of metastasis in ANNUBPs. In this case, the patient has done well, with no evidence of recurrence four years after the resection. His tumor nerve margins showed neurofibroma with no atypical features. The hope is that the term ANNUBP will eventually be replaced by more precise terminology and diagnostic criteria so as to better predict the behavior of these tumors and to promote the resection of potentially progressive tumors before they undergo malignant transformation.

Table 1: Suggested classification and terminology for NF1-associated nerve sheath tumors based on 2016 consensus meeting (*Modified from Miettinen et al. Hum Pathol 2017).

• ANNUBP is a neurofibroma with two or more of the following atypical features: cytologic atypia, hypercellularity, loss of neurofibroma architecture, mitotic index >1/50 HPFs and <3/10 HPFs. ANNUBP encompasses entities previously termed atypical neurofibroma and sometimes low-grade MPNST, as well as the clinical situation in which a small biopsy shows some concerning features but is not diagnostic of MPNST.
• Neurofibromas with either isolated cytologic atypia or hypercellularity are considered benign.
• Loss of CDKN2A/B, or less often SMARCA2, is thought to drive transformation of benign neurofibroma to ANNUBP.
• Mutations in PRC2 subunits, EED and SUZ12, as well as TP53 and EGFR, are associated with transformation of ANNUBP to MPNST.
• Complete loss of H3K27me3 staining in the tumor cells is a surrogate marker for PRC2 subunit mutations and is seen in some MPNSTs but not in neurofibromas or ANNUBPs. A mosaic staining pattern of H3K27me3 is nonspecific.

1. Beert E, Brems H, Daniëls B, et al. Atypical neurofibromas in neurofibromatosis type I are premalignant tumors. Genes, Chromosom Cancer. 2011;50:1021-1032.
2. Brohl AS, Kahen E, Yoder SJ, Teer JK, Reed DR. The genomic landscape of malignant peripheral nerve sheath tumors: diverse drivers of Ras pathway activation. Sci Rep. 2017;7(1):14992.
3. Carrió M, Gel B, Terribas E, et al. Analysis of intratumor heterogeneity in neurofibromatosis type 1 plexiform neurofibromas with atypical features: correlating histological and genomic findings. Hum Mutat. 2018;39:1112-1125.
4. Higham CS, Dombi E, Rogiers A, et al. The characteristics of 76 atypical neurofibromas as precursors to neurofibromatosis 1 associated malignant peripheral nerve sheath tumors. Neuro Oncol. 2018;20(6):818-825.
5. Legius E, Brems H. Genetic basis of neurofibromatosis type 1 and related conditions, including mosaicism. Childs Nerv Syst. 2020. [online ahead of print]
6. Lemberg KM, Wang J, Pratilas CA. From genes to -omics: the evolving molecular landscape of malignant peripheral nerve sheath tumor. Genes. 2020;11:691.
7. Meyer A, Billings SD. What’s new in nerve sheath tumors. Virchows Archiv. 2020;476:65-80.
8. Miettinen MM, Antonescu CR, Fletcher CDM, et al. Histopathologic evaluation of atypical neurofibromatous tumors and their transformation into malignant peripheral nerve sheath tumor in patients with neurofibromatosis 1 – a consensus overview. Hum Pathol. 2017;67:1-10.
9. Pemov A, Hansen NF, Sindiri S, et al. Low mutation burden and frequent loss of CDKN2A/B and SMARCA2, but not PRC2, define premalignant neurofibromatosis type 1-associated atypical neurofibromas. Neuro Oncol. 2019;21(8):981-992.
10. Prudner BC, Ball T, Rathore R, Hirbe AC. Diagnosis and management of malignant peripheral nerve sheath tumors: current practice and future perspectives. Neurooncol Adv. 2020;2(1):i40-i49.