Research ArticleExperimental Studies

Peripheral Nerve Sheath Tumors in Patients With Neurofibromatosis Type 1: Morphological and Immunohistochemical Study

REINHARD E. FRIEDRICH, LOUISA K. N. NÖRNBERG and CHRISTIAN HAGEL
Anticancer Research March 2022, 42 (3) 1247-1261; DOI: https://doi.org/10.21873/anticanres.15592

Abstract

Background/Aim: The aim of the present investigation was to characterize the growth pattern and antigen profile of peripheral nerve sheath tumors (PNST) in a large series of tumors obtained from patients with Neurofibromatosis type 1 (NF1) focusing on morphological characteristics of diffuse plexiform neurofibroma (DPNF). Materials and Methods: Tissue micro-array (TMA) analysis was applied to study 520 formalin-fixed, paraffin-embedded human PNST of 385 patients with confirmed NF1 diagnosis. PNST originated from all areas of the body and were classified as cutaneous neurofibroma (CNF, n=114), diffuse neurofibroma (DNF, n=109), DPNF (n=108), plexiform neurofibroma (PNF, n=110), and malignant peripheral nerve sheath tumor (MPNST, n=22). Histomorphology and antigen expression patterns of the tumors were determined [S100, epithelial membrane antigen (EMA), CD90, mast cell tryptase, and neurofilament]. Results: Benign PNST showed significantly more S100-positive tumor cells than MPNST (p<0.001). EMA expression was most pronounced in perineurium of DPNF. The number of mast cells in CNF, DNF and DPNF was significantly higher compared to PNF and MPNST (p<0.001 for both comparisons, Mann-Whitney U-test). Conclusion: DPNF show some distinct cellular characteristics. A high number of EMA positive cells possibly indicates the dissemination of perineural cells to the surrounding tissue. Concerning mast cell density, DPNF resemble DNF and CNS rather than PNF. Close contact of tumor cells in DPNF, DNF and CNF with the immune system is a prerequisite for permanent immunological reactions in contrast to PNF in which tumor cells are partitioned from the immune system by the perineurium and blood-nerve barrier of blood vessels. It is assumed that these morphological distinctions may reflect in part the biological differences between the entities. While PNF is a known precancerous stage in NF1 patients, DPNF are not rated as such. Furthermore, the morphologic differences between benign nerve sheath tumors may be important for the efficacy of drugs to access tumor cells.

Key Words:

Neurofibromatosis type 1 (NF1) is an autosomal dominant hereditary neurocutaneous disease. NF1 is likely the most common autosomal dominant inheritable syndrome in humans associated with a predisposition to develop cancer (1). Calculated incidence is 1:2,500 to 1:3,000 individuals living at birth (2, 3). About half of the affected individuals have new mutations of the NF1 gene, while in the other half, heredity can be demonstrated (4). Neurofibroma, a benign peripheral nerve sheath tumor (PNST), is the hallmark of NF1 (5). Neurofibromas often arise in the skin and in large numbers, are highly visible and often impose a heavy psychological and social burden on patients (1). However, the variability and unpredictability of symptoms and findings make NF1 a disease difficult to diagnose and to treat in individual cases (6, 7).

In our histological studies on PNST in NF1, it has repeatedly been noted that PNF may show focal loss of perineural integrity allowing tumor growth in continuity from the intraneural to extraneural space. This morphological characteristic contradicts the morphological definition of PNF and has been discussed as a special feature within this tumor group (8, 9). The frequent occurrence of this finding in NF1-associated PNST has prompted us to preliminary define these tumors as a separate subtype [diffuse plexiform neurofibroma (DPNF)] and to study them with morphological techniques in comparison with other neurofibroma of NF1 patients (Figure 1). Institutional morphological classifications so far have not included the DPNF as an entity in the canon of PNST. This morphological differentiation pattern in neurofibroma has already been described (10) and is synonymously applied to clinical descriptions of large benign PNSTs (11). The purpose of this study is to test the hypothesis that DPNF has extensive morphological similarity to PNF but can be distinguished from PNF by its protein expression pattern. Identifying some morphological characteristics of DPNF may be relevant for the assessment of therapeutic measures in NF1, e.g., chemotherapy to reduce the size of PNST.

Figure 1.
Figure 1.

Diffuse-plexiform neurofibroma (DPNF). (A) Plexiform neurofibroma partly enclosed in perineurium (black arrows). To the right side the continuity of the perineurium is disrupted and the tumor infiltrates the adjacent soft tissue (red arrowheads). (B) Immunohistochemistry of DPNF depicting an endoneurial neurofibroma surrounded by a diffuse neurofibroma, the perineurium (black arrows) is labelled by antibodies against epithelial membrane antigen (EMA), a discontinuity of the perineurium can be noticed (red arrowhead). A, H&E stain; B, EMA immunohistochemistry (chromogen DAB, counter stain hemalum), scale bars 20 μm.

Classification of neurofibromas. In this study, neurofibromas are classified according to following descriptions of the lesions:

Cutaneous (dermal) neurofibroma (CNF). CNF is the most common tumor in NF1. However, CNF is also well-known as sporadic neoplasm, thus not sufficient as a single finding for diagnosis of the syndrome (6, 12-17). The tumors develop in the dermis or subcutis and may arise in almost all body regions. The terms ‘cutaneous‘ and ‘dermal‘ are generally used interchangeably to denote this entity. CNF are tumors confined to the upper layers of the skin and do not develop invasive growth. In this presentation, the term ‘cutaneous neurofibroma’ is used to describe tumors of the skin that are likely to emanate from the terminal axons in the skin and do not develop beyond the subcutaneous fat in deeper body layers. Current efforts to classify these tumors into further subgroups are not pursued here (18).

Plexiform neurofibroma. Plexiform neurofibroma (PNF) occurs in approximately 25-30% of NF1 patients (1, 16). PNF is very characteristic of NF1 and a major clinical determinant of the disease (13, 14, 19-21). PNF often present congenitally or get noticed in early childhood. Growth characteristics and timing of PNF onset substantiate current estimates that PNF arises during embryogenesis (1). PNFs are intra-neurally localized neoplasms, arising from peripheral nerves (12, 13) and grow in nerve fascicles under the skin or deep in the body. In addition, nerves inside organs may also be the origin of PNF. Predilection sites of PNF formation are major nerves of the trunk, extremities, and head (1, 16). Because of the frequently irregular tumor mass, large differences in diameter within the affected tumorous nerve, and the high number of affected nerves, radical surgical resection can be difficult or impossible and repetitive measures be necessary (12, 13, 16, 18-22). Histologically, PNF is recognized by its prominent intra-fascicular growth. The perineurium is still intact and surrounds the bloated nerve fascicles. The tumor shows a low cellularity, although cell density may vary. The tumor tissue comprises a myxoid matrix with wavy spindle-shaped nuclei and a variable proportion of collagen bundles (14, 15, 19).

PNF can degenerate and develop into malignant peripheral nerve sheath tumor (MPNST). The risk of developing MPNST is approximately 8-12% in NF1 patients (4). It is estimated that approximately 30% to half of all patients with MPNST have NF1, so the genetic background of the patient should always be checked when diagnosing an MPNST. Accordingly, persistent pain or suddenly increasing tumor growth may be important signs of malignant degeneration of PNF in NF1 patients (4).

Diffuse neurofibroma. Diffuse neurofibroma (DNF) forms plaque-like thickenings in the dermis and subcutis, especially on the head and neck (14, 15). There may be remnants of perineurium in the tumor. This subtype remains benign, despite its invasive growth.

Histologically, extensive plaque-like tumor cell infiltrates are found in dermis and subcutis (12). Tumor growth does not destroy the structures of the skin but spares the adnexa and spreads flatly along the subcutaneous structures to the fatty tissue and the skeletal muscle (12-15). As a special feature of this tumor subtype, pseudo-Meissner bodies have been described. These skin appendages are spherical, balloon-like structures expressing the S100 protein. The capsule-like margin of the lesions is stained by epithelial membrane antigen (EMA). Pseudo-Meissner bodies are functionless, but their morphology resembles the phenotype of skin mechanoreceptors (13).

Diffuse plexiform neurofibroma (DPNF). DPNF is characterized by remnants of peripheral nerve fascicles and tumor cells in the surrounding tissue. Although the perineurium is visible, it is no longer intact in at least one site (Figure 1). The cellular content of the fascicles is higher than in PNF, but still below the surrounding tissues and in DNF. The morphological pattern of this subtype is a ruptured nerve fascicle, from which spindle-shaped cells grow into the surrounding tissue. This subtype is of biological interest in that the outbreak of tumor cells from an organ capsule usually is a criterion of malignancy. However, this criterion does not apply in the case of DPNF.

Recent surgical reports on problems in developing treatment concepts and discussions on individual approaches to therapeutic solutions to NF1 associated PNST also refer to DPNF (23-25). However, it is not certain in all reports whether the tumor description “diffuse” combined with “plexiform” is a micromorphological characteristic of the tumor, i.e., a histological diagnosis (a type of PNF or a differentiation pattern within a PNF), or the clinical characteristic of tumor extension (invasive, ill-defined soft tissue mass) (26). The different use of the terms “plexiform” and “diffuse” in medical specialties contributes to the fact that the exchange of information on this lesion is difficult to achieve (20). This study aims to contribute to what are some characteristics of morphologically defined DPNF.

Malignant peripheral nerve sheath tumor (MPNST). MPNST arise from peripheral nerves (27, 28). MPNST accounts for approximately 5-10% of human soft-tissue sarcomas (29). In up to 50% of patients with MPNST association with NF1 can be proven (11, 14, 30). Sporadic and NF1-associated MPNST are similar in tumor biology. However, some clinical differences have been identified distinguishing sporadic and syndromic MPNST (30-32). Different PNS cells may be involved in MPNST development (Schwann cells, perineural cells, fibroblasts) (29). The most common progenitor lesions of NF1-associated MPNST are PNF (3, 17), in contrast to sporadic MPNST, which arise de novo in nerves. The preferred localization of both MPNST types are major nerve branches of extremities and trunk (13-15, 19, 30). However, NF1-associated MPNST appear to be diagnosed somewhat more frequently in the extremities (30). MPNST typically arise in adulthood. Patients with sporadic MPNST are on average older than NF1 patients with this diagnosis (30). MPNST are highly malignant tumors that metastasize very early, especially in the lungs, bones, and brain (15). At present, radiation therapy and chemotherapy have little or no impact on patients’ long-term survival (1). Surgical resection of the tumor with wide margins to healthy tissue is the treatment of choice (32), but the prognosis is generally unfavorable because the likelihood of tumor recurrence is 40-68% (13-15). The 5-year survival rate of MPNST patients was reported to be 19-28% (33).

Because MPNST are often highly de-differentiated, the identification of the tumor entity may be difficult. In this study, only tumors from NF1-patients were examined. Due to the high mortality of NF1 patients affected by MPNST, the study of tumorigenesis and the malignant transformation of benign tumors into an MPNST is of great importance. Reliable and predictive biomarkers are needed.

Materials and Methods

Patient cohort and samples. The human tissue samples were processed and diagnosed in the Institute of Neuropathology, UKE. The tissue samples originated from PNST surgery of NF1 patients performed in the Department of Oral and Craniomaxillofacial Surgery, UKE, between 1981 and 2012. The samples were routinely formalin-fixed and paraffin-embedded. All patients met the current diagnostic criteria for diagnosing NF1 as proposed by the US NIH (1, 4). The present cohort comprised 520 human peripheral nerve sheath tumors from 385 patients (women: 54.4%, men: 43.8%). There were 228 samples from male patients and 283 from female patients. Data on gender were missing in 9 cases.

The samples comprised 136 CNF, 123 DNF, 113 DPNF, 126 PNF and 22 MPNST. Atypical neurofibroma and other rare PNST entities were excluded from evaluation. Classification of tumors was performed by an experienced neuropathologists (CH) on 8 standard staining sections of each tumor sample: Hematoxylin-eosin, Periodic Acid Schiff reaction (PAS), Elastica-van Gieson (EvG), Turnbull blue, S100 protein, EMA, neurofilament, and Ki-67 protein. For the present study, a section was taken from each paraffin block and stained according to H&E. The tumors were re-classified according to the neurofibroma subtype and typical areas were marked on the slides. The marked regions were then punched out from the block for construction of a tissue micro array (TMA). From each tumor 2 punches were taken. The TMA consisted of up to 148 tissue punches. TMA sections were processed in the staining automat Ventana Benchmark XT (Roche Deutschland, Grenzach-Wyhlen, Germany) for immunohistochemical staining. Only samples that could be assessed without doubt were included in the evaluation.

Table I shows the technical characteristics of the antibodies and the evaluated topography of the antigens in the tissue. Procedural details are described elsewhere in detail (34-37).

View this table:
Table I.

Antibodies: Clones, dilutions, providers, and order numbers.

Ethics. This study was accepted by the local authority of Eppendorf University Hospital (appointed by the University of Hamburg) as a prerequisite for the preparation of a medical dissertation (LKNN). All patients gave informed consent regarding the scientific evaluation of tumor data prior to their treatment in the hospital. All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Declaration of Helsinki and its later amendments or comparable ethics standards. Data were anonymized prior to analysis, and the investigators studying the tissue samples were blinded for diagnosis, identity of individuals, and assignment of the single case to a diagnostic group. These investigations were carried out in accordance with the Hamburg Health Service Act (Hamburgisches Gesundheitsdienstgesetz). This type of study does not require an ethics vote.

Statistics. Nominal and ordinal scaled variables were checked using the Chi² test. For evaluation of metric data, the Mann-Whitney U-test was used, as the assumption of a normal distribution was not sufficiently fulfilled. The Welch test was used instead of the t-test for comparison of age because a normal distribution could be assumed, but not equality of variances. The level of significance was set at 5% for all analyses. All statistical computations were performed as exploratory analyses. Statistics were calculated using SPSS for Windows ver. 22.

Results

Tumor type and patient age at sampling. The mean age of all patients (n=513) was 30.9 years (missing data in 7 patients). Concerning the type of tumor, the average age for CNF (n=136) was 39.4 years, for DNF (n=123) 32 years, for DPNF (n=113) 25.6 years, for PNF (n=125) 24.6 years, and for MPNST (n=16) 38 years. Figure 2 shows the mean age the patients with consideration of the tumor subtype. Taken together, patients with CNF, DNF and DPNF showed a significantly higher age at operation compared to patients with PNF (Figure 2).

Figure 2.
Figure 2.

Mean age at operation of NF1 associated peripheral nerve sheath tumors. The group of patients with tumors that do not show a tendency for malignant transformation (CNF, DNF and DPNF, boxed) were significantly older at operation than patients suffering from PNF which is regarded a precancerous lesion (***p<0.001, Welch test, n=513, SE: standard error of the mean).

Tumor subtypes and localization. Tumor localization was distinguished (head/neck, trunk, and extremities). No information on the localization was available in 66 cases (12.7%). Based on the total group of 454 tumors with information on topography, 33% (150/454) were localized in the head and neck region, 40.5% (184/454) in the trunk and 26.4% (120/454) in the extremities. The findings are summarized in Table II. When comparing tumor subtypes in relation to the localization, substantial differences were noted in that MPNST were never observed in the head and neck region (p<0.001, Chi2 test, Figure 3).

View this table:
Table II.

Distribution of the tumor groups depending on the region of the body.

Figure 3.
Figure 3.

Tumor subtype vs. localization. In contrast to benign neurofibroma subtypes (all CNF, DNF, DPNF and PNF), MPNST were not observed in the head and neck region (light blue columns, ***p<0.001, Chi-square, n=454).

Immunohistochemistry. S100 protein. S-100 immunoreactivity was assessed as present (positive) or missing (negative). Neurofibroma and MPNST showed nuclear and cytoplasmatic S100 protein expression in the tumor cells. For technical reasons, no immunoreactivity was observed in 30 (6.4%) of the examined neurofibromas. Therefore, the evaluation of the immune reaction refers to n=468 tumors. Benign tumors were S100-positive in 99.8% (467/468) and negative in 0.2% (1 DNF). MPNST were immunoreactive to S100 antigen in 81.8% (18/22) and negative in 18.2% (4/22). The median of S100 positive cells in benign and malignant tumors (n=490) was 73.8% (65.9% and 79.7%, respectively), based on total number of cells per area evaluated (0.12 mm²). The results are summarized in Figure 4 (boxplots) and illustrated in Figure 5.

Figure 4.
Figure 4.

Quantification of S100 expressing cells in NF1 associated peripheral nerve sheath tumors. MPNST demonstrated a significantly lower percentage of S100 positive cells than benign neurofibroma subtypes (all CNF, DNF, DPNF, PNF) (***p<0.001, Mann-Whitney U-test, n=468).

Figure 5.
Figure 5.

S100 expression in benign neurofibroma vs. MPNST. (A) Strong labelling of tumor cells with antibodies against S100 protein in a DNF (patient no. 157). (B) No detection of S100 protein in MPNST (patient no. 1016). S100 immunohistochemistry (chromogen DAB, counter stain hemalum), scale bars 20 μm.

EMA. The immunoreaction of the anti-EMA antibody identified perineural cells, as expected. In addition, EMA-positive immunoreactions were sporadically detected in solid, fragmentary structures that appeared to wrap around or “envelope” an adjacent tissue particle (“wrapped/enveloped structures”) which presumably could be allocated to small nerve fascicles.

A total of 41 tumors were excluded from the evaluation because of technical deficiencies (7.8%). Most of the tumor population (n=479) showed no EMA expression (47.4%, 227/479). EMA expression was present in 29.9% (143/479) of the tumors in perineuria or other surrounding structures. Evaluation of EMA immunoreaction in PNST is summarized in Table III and illustrated in Figure 6 and Figure 7.

View this table:
Table III.

Immunoreactivity of anti-epithelial membrane antigen (EMA) antibody in defined compartments of peripheral nerve sheath tumors of patients with neurofibromatosis type 1 (NF1).

Figure 6.
Figure 6.

EMA expression in NF1 associated peripheral nerve sheath tumors. (A) No demonstration of EMA in tumor cells of a DNF (note the labelling of epithelial cells in sweat glands, patient no. 314). (B) Diffuse PNF with EMA-positive perineurium (patient no. 370). (C) DNF with minute nerve fascicles surrounded by EMA-positive perineurium (patient no. 32). (D) PNF with multifocal membranous staining of cells in tumor tissue (patient no. 463). EMA immunohistochemistry (chromogen DAB, counter stain hemalum), scale bars 20 μm.

Figure 7.
Figure 7.

Detection of EMA in NF1 associated peripheral nerve sheath tumors (n=479).

CD90. CD90 immunohistochemistry was performed to determine the fibroblast content of the nerve sheath tumors. The original total number of tumors in this study was reduced by 8.7% (n=45) to 475 tumors due to technical reasons. CD90 was positive in 68.8% (327/475) and negative in 31.2% (148/475) of the tumors. No subtyping of the tumors was performed in this evaluation. As expected, CD90 immunoreaction was primarily found in the compartments rich in connective tissue like perivascular area of the tumors and in perineural structures (Figure 8 and Figure 9), but also – although significantly less frequent – in a patchy distribution within the tumor (Figure 9C). Figure 8 shows the distribution of tumor subtypes relative to different CD90 staining and Figure 9 illustrates the findings.

Figure 8.
Figure 8.

CD90 expression in NF1 associated peripheral nerve sheath tumors. CD90 expression in scattered cells within the tumor tissue was less frequently observed than in the perineurium or perivascular cells (***p<0.001, chi-square, n=475).

Figure 9.
Figure 9.

Detection of CD90 in NF1 associated peripheral nerve sheath tumors. (A) No expression of CD90 in a CNF (patient no. 505). (B) Perivascular demonstration of CD90 in a DNF (patient no. 435). (C) DNF showing labelling of perivascular structures and intra-tumoral spindle shaped cells (patient no 504). (D) PNF with perineural CD90 labelling (patient no. 381). CD90 immunohistochemistry (chromogen DAB, counter stain hemalum), scale bars 20 μm.

Mast cell tryptase. Mast cells were identified by anti-mast cell tryptase antibody. All antibody-labeled cells were evaluated in an area of 0.79 mm². The median mast cell count was 45 (13.8/66.8) in all tumors studied (n=485). For technical reasons, 35 cases were excluded from the evaluation. The median mast cell count was 57 (41.5/73) for CNF (n=131), 56.5 (44.75/78.235) for DNF (n=117), 58.5 (33.1/76.6) for DPNF (n=98), 8 (4/12) for PNF (n=118), and 9 (0.5/28) for MPNST (n=21). Mann-Whitney U-test showed highly significant differences between the group CNF/DNF/DPNF and PNF (p<0.001), as well as between the group CNF/DNF/DPNF and MPNST regarding the median number of mast cells in the tumor. Thus, PNF and MPNST had significantly fewer mast cells than the group of other (benign) neurofibromas. Figure 10 shows mast cell tryptase staining and tumor subtypes. Figure 11 illustrates the findings.

Figure 10.
Figure 10.

Quantification of mast cells in NF1 associated peripheral nerve sheath tumors. Both, PNF and MPNST showed significant lower mast cell densities than CNF/DNF/DPNF (boxed) (***p<0.001 for both comparisons, Mann-Whitney U-test, n=485).

Figure 11.
Figure 11.

Detection of mast cells in NF1 associated neurofibroma. (A) Mast cells in a CNF (arrows, patient No. 47). (B) Mast cells in a PNF (arrows, patient no. 486). Mast cell tryptase immunohistochemistry (chromogen DAB, no counter stain), scale bars 20 μm.

Neurofilament. Neurofilament expression of the nerve sheath tumor was always detectable in axons of intact nerve fascicles (Figure 12A). Furthermore, bundles of residual axons were observed within PNF (Figure 12B). A total of 502 tumors were evaluated (96.6%); 45.6% (229/502) of the tumors were neurofilament-negative, 13.7% (69/502) had intact nerves in the tumorous tissues that were neurofilament antibody-labeled, and 40.6% (204/502) of the specimens showed residual axons in the tumor tissue.

Figure 12.
Figure 12.

Detection of axons in NF1 associated peripheral nerve sheath tumors. (A) Neurofilament immunohistochemistry depicting axons in a small intact nerve fascicle in a DNF (patient no. 32). (B) PNF with loosely scattered axons in the tumor tissue (patient no. 274). Neurofilament immunohistochemistry (chromogen DAB, counter stain hemalum), scale bars 20 μm.

MPNST were predominantly negative for neurofilament staining (81.8%, 18/22). Table IV and Figure 12 summarize and illustrate the results. The statistical tests showed no significant differences in neurofilament expression patterns between the tumor subtypes.

View this table:
Table IV.

Immunoreactivity of anti-neurofilament antibody in neural structures of peripheral nerve sheath tumors in patients with neurofibromatosis type 1 (NF1).

Discussion

This study shows the distribution pattern of several immunohistochemically identified antigens in cytoplasmic and membranous compartments of various PNST of patients with NF1. In this context, emphasis is placed on the differentiation of a tumor type termed DPNF that has received poor attention so far. It is shown that DPNF can reproducibly be classified by its morphological characteristics within the group of benign PNST. Some spectra of the antibodies’ reactions in the entities indicate that DPNF differs from PNF in terms of protein expression. However, the major difference of this tumor is extra-capsular tumor growth without any signs of malignant, invading and destroying characteristics. We consider the differentiation of PNSTs in NF1 to be noteworthy, because, for example, pharmacological concepts of reducing tumor growth must consider how and in what amount active ingredient act in an either compartmentalizing, sealing tissue environment (e.g., nodular PNF) or diffusely spreading, usually slowly growing neoplasm with poorly defined borders (38). Surgical measures for reduction or resection of PNST of the NF1 patient significantly alter the tumor milieu (34). A PNF without leakage of vessels and the perineurium may allow only restricted circulation (of agents) across capsule-like borders. This indication is even more important because diffuse neurofibroma differs only slightly in ultrastructural features from normal nerve (39, 40).

Neurofibromas are composite tumors (14, 39, 40). Neurofibromas consist of Schwann cells, fibroblasts, perineural cells and some mast cells. The proportion of each cell class was determined and showed clear differences (30, 34). In the present study, the distributional differences of cells were examined in a large number of NF1-affected individuals. Cell entity-specific antibody reaction patterns were analyzed in PNST. Furthermore, the different cell types of the tumors in relation to the diagnostic groups were also determined. The selection of the sample judged to be representative of the tumor subtype represents an unavoidable bias of the comparative morphological study. The large number of tissue samples collected for the TMA should reduce this bias. On the other hand, the technique allowed the standardized immunohistochemical analysis of the samples.

S100 protein expression. Schwann cells express S100 protein (39, 40) and are the tumor cells of neurofibroma (41, 42). The proportion of Schwann cells in a single neurofibroma is about 60%-85% (43, 44). In this study, the percentage of S100 expression in Schwann cells of all neurofibromas was 73.8% and thus within the known range.

Another criterion is the qualitative assessment of S100 protein expression, that is, how many tumors in a study express the protein. There are clear differences in S100 positivity if tumor biology is considered. In benign PNSTs, S100 positivity in this study is 99.8% and thus within the range of published data. In contrast, S100 in MPNST is less reliably expressed: the ratio of S100 positive tumors is about 50% to 70% (45). In this study, the proportion of S100 positive MPNST was slightly higher, at 81.8%. One cause of diminished S100 expression may be the de-differentiation of the MPNST. Anaplastic nerve sheath cells can show expression pattern lacking cytoplasmic proteins characteristic for Schwann cells. On the other hand, it is also possible that human MPNSTs are not derived from Schwann cells, the expression of the S100 protein is to be regarded as an immunological phenocopy of the Schwann cell, or the variable S100 positivity of MPNST is an indication that this tumor arises from Schwann cell progenitors that only facultative express S100 (46, 47).

Another aspect is the reaction spectrum of the antibody used. The polyclonal anti-S100 antibody used in this study identifies S100A. The S100 protein family is very large and contains several subtypes for the A-type alone. Antibodies to these subtypes have different affinities for individual PNSTs (48). Although a polyclonal antibody very likely has greater potential to label somewhat differentiating epitopes, cell entity-specific binding differences of the antibody cannot be ruled out depending on the distribution pattern of the respective antigens (49). Hence, quantification of the immune reaction may not be useful for the S100.

EMA expression. Antibody against the EMA was assessed for various reasons. On the one hand, the expression pattern should exclude EMA-positive perineurioma, which is a significant differential diagnosis of neurofibroma (13-15). On the other hand, remains of the perineurium should be reliably identified and the tumor subtypes clearly defined in neurofibromas (50, 51). The immunohistology staining patterns present in this study allowed the safe exclusion of perineuriomas because the tumor cells did not express EMA, neither diffuse nor intense, the latter patterns are a hallmark of perineurioma (52). In addition, none of the tumors studied had the growth pattern of a perineurioma. However, EMA immunohistochemistry can lead to diagnosis of plexiform-diffuse neurofibroma in some cases, as these were better identified by the EMA-positive perineural residues than in conventional histological staining.

A multifocal EMA positivity was observed in 22.8% of the neurofibromas. Noteworthy 29.9% of diffuse plexiform neurofibromas demonstrated EMA-positive structures which presumably were small nerve fascicles.

CD90 expression. CD90 is a surface antigen expressed by various cell populations (fibroblasts, neurons, blood stem cells, endothelia). In the present study, CD90 was used as a fibroblast marker. However, so far there is no antibody that marks exclusive and reproducible all fibroblasts. Therefore, it cannot be ensured that all fibroblasts have been identified by the antibody or other cells were CD90 positive, without necessarily belonging to this cell class. An alternative means of detection would have been the determination of the fibroblast surface protein (34).

Due to their morphology and localization, the majority of CD90 positive cells were identified as pericytes and perineural cells, which are known to be associated with fibroblasts. Since CD90 positive cells were identified only sporadically in the tumorous tissues, the proportion of ‘free’ fibroblasts can be assumed to be small (<20%) in nerve sheath tumors in NF1.

According to the literature, the proportion of fibroblasts in human neurofibromas is about 10% to 20% (44). In this study, it was not the percentage of total fibroblasts that was determined but the proportion of CD90 positive neurofibromas (68.8%). More than half of the CD90-positive cellular staining is due to the labeling of pericytes and perineural cells. Only a relatively small proportion of neurofibromas had free fibroblasts in the tumor.

Mast cell density. According to the literature, mast cells in neurofibroma are discussed as promoters of tumorigenesis and tumor growth (54-56). The results of this study argue against such a central function of mast cells in tumor development and spread in PNST of patients with NF1. The median values of mast cells were highest in CNF [median (M): 57] compared to other neurofibroma subtypes. However, this entity remains benign and does grow for years in many cases (6). These clinical parameters correlate with a benign histopathological phenotype with low proliferation rate. We have already communicated this result in a previous study on proliferation indices in CNF (53), confirming earlier studies on low proliferation activity of neurofibroma in NF1 (39, 40). In addition, the number of mast cells was significantly lower in both PNF (M: 8) and MPNST (M: 9) than in the benign variants of PNST (CNF, DNF, DPNF). If mast cells support the tumor progression from the PNF to the MPNST, then it would be plausible to assume this cell type is increased in number in PNF and MPNST. However, this was not the case. Earlier reports show the reduction of nerve sheath tumors using mast cell inhibitors (54-56). An explanation of this effect would be that inflammatory processes mediated by mast cells always lead to edema and thus to a tumor volume increase (34). This increase would be attributable to the inflammatory process and not to actual (cellular) tumor growth, e.g., in neurofibroma. Accordingly, the effect of blocking mast cell effects in tumors would be a reduction of intra-tumoral fluid. These effects may be therapeutically important (34), especially in acute situations, without affecting the long-term mechanisms controlling tumor proliferation. On the other hand, (nodular) PNF, described as ‘solid’ tumors (14), morphologically are closed compartments, and are supposed to have less exchange of fluid. It can be assumed that proteins secreted extracellularly in an encapsulated tumor have a longer residence time on site than in tumors with undefined margins. Therefore, for example, factors secreted by mast cell possibly are retained longer in place and thus can act over a longer period in intraneural neurofibroma compared to ‘diffuse’ lesions. This could increase mast cell effects on tumor cells and thus make a substantial contribution to tumor progression in slowly proliferating tumors such as PNF, even if mast cells are not increased in number. However, mast cells are substantially lower in number in PNF than in non-proliferating CNF. It is possible that differential effects of mast cells in NF1-associated PNST can be more accurately identified by typing these cells (57).

Neurofilament detection. Evidence of neurofilament reveals persistence for residual intact innervation in tumor tissue. Neurofilament-positive axons can be detected in PNSTs, namely in the plexiform subtype (45). The application of the anti-neurofilament antibody should clarify if and if so, how many axons are present in the tumor tissue. In all investigated nerve sheath tumors, axons were detected in the tumor tissue, which supports current estimates about the suspected origin cells of the tumors. As expected, most PNF (66.4%) contained axons in tumor tissue. PNF also had the highest value in terms of the highest number of axons in the tumor (M: 35). The evaluation of neurofilament distribution in PNST reveals the lack of association between the tumor type and axon degeneration in benign PNSTs. This assessment corresponds to the clinical experience, e.g., even very large PNF, for example in formerly called ‘elephantiasis neuromatosa’, can contain functional nerves, and organs embedded in a tumor mass, e.g., muscles, may be functional, too.

DPNF. The DPNF in this study is a morphologically definable entity. The morphological term is not new (8). Previous investigators have suggested that local defects of the perineurium in the PNF allow the intraneural tumor to grow into the perineural area. However, it has not yet been clarified whether this is a separate entity or a focally conspicuous differentiating feature of the PNF. In this study, we tested whether DPNF can be distinguished from PNF by the expression pattern of characteristic proteins of PNST cells. The expression pattern of the proteins in DPNF is closer to CNF and DNF than PNF. This distinction is novel. The protein differentiation of DPNF from PNF could correlate with the different biological properties of the tumors. While PNF is precancerous and malignant degeneration of PNST is a major factor of reduced life expectancy in NF1 patients compared to the normal population, neurofibromas with the eponymous morphological features of ‘diffuse’ invasive growth are not considered to be precancerous. On the other hand, NF1-associated nodular PNF have been well described arising in adulthood within congenitally manifest massive soft tissue PNST (58). Indeed, there are authors who have defined DPNF as congenital tumors (50, 51). The present study suggests that there are age differences in the manifestation of the differentiated PNST. However, the mean age of patients at the time of surgical treatment reflects only differences in temporal treatment need and not the chronology of the underlying tumor biology.

Furthermore, the histologic basis of these tumors often cannot be fully determined from a single tissue sample because tumor regions with strictly intraneural PNF and those with perineural defects may not be homogeneously distributed and thus escape diagnosis in a representative tissue sample (20). Further investigation is needed to determine whether DPNF forms a separate entity of nerve sheath tumors or is a differentiation within PNF.

It is plausible to assume that topography, i.e., the whole of preexisting peri-tumorous structures, exerts a decisive influence on tissue differentiation already at the origin of nerve sheath tumors. For example, cutaneous neurofibromas arise in early contact with a very collagen-rich environment, plexiform neurofibromas develop in intraneural compartments, and diffuse neurofibromas grow in contact with various soft tissues. Another factor that is difficult to decipher is the timing of tumor development and the analysis of the respective parallel differentiation processes in tumorous and non-tumorous cells (58).

Conclusion

NF1-associated PNST are heterogeneous entities that contain very different cells. The tumor tissue shows distinct differences in structure and differences in tumor biology are associated with morphologically characterized entities. However, in all PNST, the proportion of Schwann cells or tumor cells derived from Schwann cells (73.8%) is much higher than other cellular compartments. Single axons were still detectable in all benign tumor subtypes. The number of mast cells does not correlate with estimates of tumor biology in benign PNSTs. Presumed effects of mast cells on potential de-differentiation of PNST cannot be explained by an increasing number of mast cells. The DPNF is a morphologic variant of NF1-associated PNST that should be further characterized with respect to the clinical significance of the defining morphological sign, i.e., a defect in the perineural sheath in a PNF.

Acknowledgements

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Footnotes

  • ↵* These Authors contributed equally to this study.

  • Authors’ Contributions

    Histological and immunohistological evaluation (LKNN, CH), clinical diagnosis and treatment of patients (REF, CH), drafting the manuscript (REF, CH), final approval of the manuscript (all Authors).

  • Conflicts of Interest

    The Authors have no conflicts of interest regarding the work presented.

  • Received November 2, 2021.
  • Revision received December 17, 2021.
  • Accepted January 17, 2022.

References

  1. ↵
    1. Ferner RE and
    2. Gutmann DH
    : Neurofibromatosis type 1 (NF1): diagnosis and management. Handb Clin Neurol 115: 939-955, 2013. PMID: 23931823. DOI: 10.1016/B978-0-444-52902-2.00053-9
    OpenUrlCrossRefPubMed
  2. ↵
    1. Burger PC and
    2. Scheithauer BW
    1. Burger PC and
    2. Scheithauer BW
    : Neoplastic: Familial tumor syndromes. In Burger PC and Scheithauer BW (eds.): Diagnostic pathology. Neuropathology. 1st edition, Amirsys, Salt Lake City, pp. 1-7, 2012.
  3. ↵
    1. Rodriguez FJ,
    2. Stratakis CA and
    3. Evans DG
    : Genetic predisposition to peripheral nerve neoplasia: diagnostic criteria and pathogenesis of neurofibromatoses, Carney complex, and related syndromes. Acta Neuropathol 123(3): 349-367, 2012. PMID: 22210082. DOI: 10.1007/s00401-011-0935-7
    OpenUrlCrossRefPubMed
  4. ↵
    1. Boyd KP,
    2. Korf BR and
    3. Theos A
    : Neurofibromatosis type 1. J Am Acad Dermatol 61(1): 1-14; quiz 15-6, 2009. PMID: 19539839. DOI: 10.1016/j.jaad.2008.12.051
    OpenUrlCrossRefPubMed
  5. ↵
    1. von Recklinghausen FD
    : Über die multiplen Fibrome der Haut und ihre Beziehung zu den multiplen Neuromen. Verlag August Hirschwald, Berlin, 1882.
  6. ↵
    1. Ruggieri M
    : The different forms of neurofibromatosis. Childs Nerv Syst 15(6-7): 295-308, 1999. PMID: 10461778. DOI: 10.1007/s003810050398
    OpenUrlCrossRefPubMed
  7. ↵
    1. Friedrich RE and
    2. Nuding MA
    : Optic pathway glioma and cerebral focal abnormal signal intensity in patients with neurofibromatosis type 1: Characteristics, treatment choices and follow-up in 134 affected individuals and a brief review of the literature. Anticancer Res 36(8): 4095-4121, 2016. PMID: 27466519.
    OpenUrlAbstract/FREE Full Text
  8. ↵
    1. Tonsgard JH
    : Clinical manifestations and management of neurofibromatosis type 1. Semin Pediatr Neurol 13(1): 2-7, 2006. PMID: 16818170. DOI: 10.1016/j.spen.2006.01.005
    OpenUrlCrossRefPubMed
  9. ↵
    1. Upadhyaya M,
    2. Cooper DN
    1. Stemmer-Rachamimov A and
    2. Nielsen GP
    : Pathologic and molecular diagnostic features of peripheral nerve sheath tumors in NF1. In: Neurofibromatosis Type 1. Upadhyaya M, Cooper DN (eds). Springer Verlag, Berlin - Heidelberg - New York, pp. 429-443, 2012.
  10. ↵
    1. Fetsch JF,
    2. Michal M and
    3. Miettinen M
    : Pigmented (melanotic) neurofibroma: a clinicopathologic and immunohistochemical analysis of 19 lesions from 17 patients. Am J Surg Pathol 24(3): 331-343, 2000. PMID: 10716146. DOI: 10.1097/00000478-200003000-00001
    OpenUrlCrossRefPubMed
  11. ↵
    1. Riccardi VM
    : Neurofibromatosis: phenotype, natural history, and pathogenesis. The Johns Hopkins University Press, Baltimore, pp. 28 (elephantiasis and DPNF), pp. 31 (DPNF are congenital in origin), pp. 214 et seqq. (neurofibrosarcoma), 1992.
  12. ↵
    1. Kumar V,
    2. Abbas AK,
    3. Aster JC
    1. Kumar V,
    2. Abbas AK,
    3. Aster JC,
    4. Cotran RS and
    5. Robins SL
    : Neoplasia. In: Robbins & Cotran pathologic basis of disease (Robbins Pathology), 9th edition. Kumar V, Abbas AK, Aster JC (eds.). Philadelphia (PA), Elsevier/Saunders, pp. 282-340, 2014.
  13. ↵
    1. Kumar V,
    2. Abbas AK,
    3. Aster JC
    1. Kumar V,
    2. Abbas AK,
    3. Aster JC,
    4. Cotran RS and
    5. Robins SL
    : Peripheral nerves and skeletal muscle. In: Robins & Cotran pathologic basis of disease (Robbins Pathology), 9th edition. Kumar V, Abbas AK, Aster JC (eds.). Philadelphia (PA), Elsevier/Saunders, pp. 1227-1250, 2014.
  14. ↵
    1. Scheithauer BW,
    2. Woodruff JM and
    3. Erlandson RA
    1. Scheithauer BW,
    2. Woodruff JM and
    3. Erlandson RA
    : Neurofibroma. In: Tumors of the peripheral nervous system (AFIP Atlas of Tumor Pathology). Scheithauer BW, Woodruff JM and Erlandson RA (eds.). Washington, DC, Armed Forces Institute of Pathology (Fasc. 24), pp. 177-218, 1999.
  15. ↵
    1. Louis DN,
    2. Ohgaki H,
    3. Wiestler OD and
    4. Cavenee WK
    1. Scheithauer BW,
    2. Louis DN,
    3. Hunter S,
    4. Woodruff JM and
    5. Antonescu CR
    : Tumours of the Cranial and Paraspinal Nerves. In: WHO classification of tumours of the central nervous system, 4th edition. Louis DN, Ohgaki H, Wiestler OD and Cavenee WK (eds.). Lyon, distributed by WHO Press, World Health Organization, IARC Press, pp. 152-162, 2007.
  16. ↵
    1. Ganten D and
    2. Back W
    1. Krone W and
    2. Kehrer-Sawatzki H
    : Neurofibromatosen. In: Molekularmedizinische Grundlagen von hereditären Tumorerkrankungen. Ganten D and Back W (eds). Berlin, Springer, pp. 88-233, 2001.
  17. ↵
    1. Rodriguez FJ,
    2. Folpe AL,
    3. Giannini C and
    4. Perry A
    : Pathology of peripheral nerve sheath tumors: diagnostic overview and update on selected diagnostic problems. Acta Neuropathol 123(3): 295-319, 2012. PMID: 22327363. DOI: 10.1007/s00401-012-0954-z
    OpenUrlCrossRefPubMed
  18. ↵
    1. Ortonne N,
    2. Wolkenstein P,
    3. Blakeley JO,
    4. Korf B,
    5. Plotkin SR,
    6. Riccardi VM,
    7. Miller DC,
    8. Huson S,
    9. Peltonen J,
    10. Rosenberg A,
    11. Carroll SL,
    12. Verma SK,
    13. Mautner V,
    14. Upadhyaya M and
    15. Stemmer-Rachamimov A
    : Cutaneous neurofibromas: Current clinical and pathologic issues. Neurology 91(2 Suppl 1): S5-S13, 2018. PMID: 29987130. DOI: 10.1212/WNL.0000000000005792
    OpenUrlAbstract/FREE Full Text
  19. ↵
    1. Love S,
    2. Louis DN and
    3. Ellison DW
    : Tumors of the peripheral nerves. In: Greenfield’s Neuropathology, 8th edition. Hodder Arnold Education Publishers, pp. 2053-2260, 2008.
  20. ↵
    1. Friedrich RE and
    2. Diekmeier C
    : Peripheral nerve sheath tumors of the upper extremity and hand in patients with neurofibromatosis type 1: topography of tumors and evaluation of surgical treatment in 62 patients. GMS Interdiscip Plast Reconstr Surg DGPW 6: Doc15, 2017. PMID: 29214122. DOI: 10.3205/iprs000117
    OpenUrlCrossRefPubMed
  21. ↵
    1. Friedrich RE and
    2. Tuzcu CT
    : Surgery for peripheral nerve sheath tumours of the buttocks, legs and feet in 90 patients with neurofibromatosis type 1. In Vivo 35(2): 889-905, 2021. PMID: 33622881. DOI: 10.21873/invivo.12329
    OpenUrlAbstract/FREE Full Text
  22. ↵
    1. Friedrich RE,
    2. Hagel C,
    3. Brehme Z,
    4. Kluwe L and
    5. Mautner VF
    : Ki-67 proliferation-index (MIB-1) of neurofibromas in neurofibromatosis type 1 patients. Anticancer Res 23(2A): 953-955, 2003. PMID: 12820329.
    OpenUrlPubMed
  23. ↵
    1. Yoshinaga A,
    2. Tsuge I,
    3. Yoshinaga D,
    4. Ogino S,
    5. Sakamoto M,
    6. Saito S and
    7. Morimoto N
    : Reduction of intraoperative bleeding in diffuse plexiform neurofibroma resection using the LigaSure vessel sealing system. Plast Reconstr Surg 148(2): 344e-346e, 2021. PMID: 34254965. DOI: 10.1097/PRS.0000000000008163
    OpenUrlCrossRefPubMed
    1. Sekiyama H,
    2. Nobeyama Y and
    3. Nakagawa H
    : Successful treatment by negative-pressure wound therapy for ulcer located on diffuse plexiform neurofibroma. J Dermatol 44(12): e313-e314, 2017. PMID: 28734012. DOI: 10.1111/1346-8138.13983
    OpenUrlCrossRefPubMed
  25. ↵
    1. Kitano D,
    2. Osaki T,
    3. Nakasone M,
    4. Nomura T,
    5. Hashikawa K and
    6. Terashi H
    : Two cases of debulking surgery for lower limb diffuse plexiform neurofibroma with transcatheter arterial embolisation. Int J Surg Case Rep 55: 132-135, 2019. PMID: 30738368. DOI: 10.1016/j.ijscr.2019.01.018
    OpenUrlCrossRefPubMed
  26. ↵
    1. Ehara Y,
    2. Koga M,
    3. Imafuku S,
    4. Yamamoto O and
    5. Yoshida Y
    : Distribution of diffuse plexiform neurofibroma on the body surface in patients with neurofibromatosis 1. J Dermatol 47(2): 190-192, 2020. PMID: 31849109. DOI: 10.1111/1346-8138.15194
    OpenUrlCrossRefPubMed
  27. ↵
    1. Reilly KM,
    2. Kim A,
    3. Blakely J,
    4. Ferner RE,
    5. Gutmann DH,
    6. Legius E,
    7. Miettinen MM,
    8. Randall RL,
    9. Ratner N,
    10. Jumbé NL,
    11. Bakker A,
    12. Viskochil D,
    13. Widemann BC and
    14. Stewart DR
    : Neurofibromatosis type 1-associated MPNST state of the science: Outlining a research agenda for the future. J Natl Cancer Inst 109(8): djx124, 2017. PMID: 29117388. DOI: 10.1093/jnci/djx124
    OpenUrlCrossRefPubMed
  28. ↵
    1. Weiss SW,
    2. Goldblum JR and
    3. Enzinger FM
    : Enzinger and Weiss’s soft tissue tumors. 4th ed., St. Louis, Mosby, 2001.
  29. ↵
    1. Zou C,
    2. Smith KD,
    3. Liu J,
    4. Lahat G,
    5. Myers S,
    6. Wang WL,
    7. Zhang W,
    8. McCutcheon IE,
    9. Slopis JM,
    10. Lazar AJ,
    11. Pollock RE and
    12. Lev D
    : Clinical, pathological, and molecular variables predictive of malignant peripheral nerve sheath tumor outcome. Ann Surg 249(6): 1014-1022, 2009. PMID: 19474676. DOI: 10.1097/SLA.0b013e3181a77e9a
    OpenUrlCrossRefPubMed
  30. ↵
    1. Hagel C,
    2. Zils U,
    3. Peiper M,
    4. Kluwe L,
    5. Gotthard S,
    6. Friedrich RE,
    7. Zurakowski D,
    8. von Deimling A and
    9. Mautner VF
    : Histopathology and clinical outcome of NF1-associated vs. sporadic malignant peripheral nerve sheath tumors. J Neurooncol 82(2): 187-192, 2007. PMID: 17111191. DOI: 10.1007/s11060-006-9266-2
    OpenUrlCrossRefPubMed
    1. Barnard ZR,
    2. Agarwalla PK,
    3. Jeyaretna DS,
    4. Farrell CJ,
    5. Gerstner ER,
    6. Tian D and
    7. Curry WT Jr.
    : Sporadic primary malignant intracerebral nerve sheath tumors: case report and literature review. J Neurooncol 104(2): 605-610, 2011. PMID: 21327709. DOI: 10.1007/s11060-011-0531-7
    OpenUrlCrossRefPubMed
  32. ↵
    1. Eckert JM,
    2. Byer SJ,
    3. Clodfelder-Miller BJ and
    4. Carroll SL
    : Neuregulin-1 beta and neuregulin-1 alpha differentially affect the migration and invasion of malignant peripheral nerve sheath tumor cells. Glia 57(14): 1501-1520, 2009. PMID: 19306381. DOI: 10.1002/glia.20866
    OpenUrlCrossRefPubMed
  33. ↵
    1. Ferrari A,
    2. Bisogno G,
    3. Macaluso A,
    4. Casanova M,
    5. D’Angelo P,
    6. Pierani P,
    7. Zanetti I,
    8. Alaggio R,
    9. Cecchetto G and
    10. Carli M
    : Soft-tissue sarcomas in children and adolescents with neurofibromatosis type 1. Cancer 109(7): 1406-1412, 2007. PMID: 17330850. DOI: 10.1002/cncr.22533
    OpenUrlCrossRefPubMed
  34. ↵
    1. Hagel C,
    2. Behrens T,
    3. Prehm P,
    4. Schnabel C,
    5. Glatzel M and
    6. Friedrich RE
    : Hyaluronan in intra-operative edema of NF1-associated neurofibromas. Neuropathology 32(4): 406-414, 2012. PMID: 22129111. DOI: 10.1111/j.1440-1789.2011.01276.x
    OpenUrlCrossRefPubMed
    1. Friedrich RE,
    2. Beer C,
    3. Glatzel M and
    4. Hagel C
    : Vascular endothelial growth factor, basic fibroblast growth factor and epithelial growth factor receptor in peripheral nerve sheath tumors of neurofibromatosis type 1. Anticancer Res 35(1): 137-144, 2015. PMID: 25550544.
    OpenUrlAbstract/FREE Full Text
    1. Friedrich RE,
    2. Behrendt CA,
    3. Glatzel M and
    4. Hagel C
    : Vascular innervation in benign neurofibromas of patients with neurofibromatosis type 1. Anticancer Res 35(12): 6509-6516, 2015. PMID: 26637864.
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Friedrich RE,
    2. Naber U,
    3. Glatzel M and
    4. Hagel C
    : Vessel and mast cell densities in sporadic and syndrome-associated peripheral nerve sheath tumors. Anticancer Res 35(9): 4713-4722, 2015. PMID: 26254361.
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Böckelmann LC and
    2. Schumacher U
    : Targeting tumor interstitial fluid pressure: will it yield novel successful therapies for solid tumors? Expert Opin Ther Targets 23(12): 1005-1014, 2019. PMID: 31825703. DOI: 10.1080/14728222.2019.1702974
    OpenUrlCrossRefPubMed
  39. ↵
    1. Lassmann H,
    2. Jurecka W,
    3. Lassmann G,
    4. Gebhart W,
    5. Matras H and
    6. Watzek G
    : Different types of benign nerve sheath tumors. Light microscopy, electron microscopy and autoradiography. Virchows Arch A Pathol Anat Histol 375(3): 197-210, 1977. PMID: 143775. DOI: 10.1007/BF01102988
    OpenUrlCrossRefPubMed
  40. ↵
    1. Lassmann H,
    2. Jurecka W and
    3. Gebhart W
    : Some electron microscopic and autoradiographic results concerning cutaneous neurofibromas in von Recklinghausen’s disease. Arch Dermatol Res 255(1): 69-81, 1976. PMID: 816259. DOI: 10.1007/BF00581680
    OpenUrlCrossRefPubMed
  41. ↵
    1. Kluwe L,
    2. Friedrich R and
    3. Mautner VF
    : Loss of NF1 allele in Schwann cells but not in fibroblasts derived from an NF1-associated neurofibroma. Genes Chromosomes Cancer 24(3): 283-285, 1999. PMID: 10451710. DOI: 10.1002/(sici)1098-2264(199903)24:3<283::aid-gcc15>3.0.co;2-k
    OpenUrlCrossRefPubMed
  42. ↵
    1. Kluwe L,
    2. Friedrich RE and
    3. Mautner VF
    : Allelic loss of the NF1 gene in NF1-associated plexiform neurofibromas. Cancer Genet Cytogenet 113(1): 65-69, 1999. PMID: 10459349. DOI: 10.1016/s0165-4608(99)00006-0
    OpenUrlCrossRefPubMed
  43. ↵
    1. Peltonen J,
    2. Jaakkola S,
    3. Lebwohl M,
    4. Renvall S,
    5. Risteli L,
    6. Virtanen I and
    7. Uitto J
    : Cellular differentiation and expression of matrix genes in type 1 neurofibromatosis. Lab Invest 59(6): 760-771, 1988. PMID: 2462129.
    OpenUrlPubMed
  44. ↵
    1. Sheela S,
    2. Riccardi VM and
    3. Ratner N
    : Angiogenic and invasive properties of neurofibroma Schwann cells. J Cell Biol 111(2): 645-653, 1990. PMID: 1696266. DOI: 10.1083/jcb.111.2.645
    OpenUrlAbstract/FREE Full Text
  45. ↵
    1. Louis DN,
    2. Ohgaki H,
    3. Wiestler OD and
    4. Cavenee WK
    (eds): WHO Classification of tumours of the central nervous system, 4th edition. Lyon, distributed by WHO Press, World Health Organization, IARC Press, 2007.
  46. ↵
    1. Gupta G and
    2. Maniker A
    : Malignant peripheral nerve sheath tumors. Neurosurg Focus 22(6): E12, 2007. PMID: 17613203. DOI: 10.3171/foc.2007.22.6.13
    OpenUrlCrossRefPubMed
  47. ↵
    1. Carroll SL and
    2. Ratner N
    : How does the Schwann cell lineage form tumors in NF1? Glia 56(14): 1590-1605, 2008. PMID: 18803326. DOI: 10.1002/glia.20776
    OpenUrlCrossRefPubMed
  48. ↵
    1. Nonaka D,
    2. Chiriboga L and
    3. Rubin BP
    : Differential expression of S100 protein subtypes in malignant melanoma, and benign and malignant peripheral nerve sheath tumors. J Cutan Pathol 35(11): 1014-1019, 2008. PMID: 18547346. DOI: 10.1111/j.1600-0560.2007.00953.x
    OpenUrlCrossRefPubMed
  49. ↵
    1. Hirose T,
    2. Sano T and
    3. Hizawa K
    : Ultrastructural localization of S-100 protein in neurofibroma. Acta Neuropathol 69(1-2): 103-110, 1986. PMID: 3083636. DOI: 10.1007/BF00687045
    OpenUrlCrossRefPubMed
  50. ↵
    1. Riccardi VM
    : Neurofibromatosis type 1 is a disorder of dysplasia: the importance of distinguishing features, consequences, and complications. Birth Defects Res A Clin Mol Teratol 88(1): 9-14, 2010. PMID: 19691086. DOI: 10.1002/bdra.20616
    OpenUrlCrossRefPubMed
  51. ↵
    1. Riccardi VM
    : The genetic predisposition to and histogenesis of neurofibromas and neurofibrosarcoma in neurofibromatosis type 1. Neurosurg Focus 22(6): E3, 2007. PMID: 17613220. DOI: 10.3171/foc.2007.22.6.4
    OpenUrlCrossRefPubMed
  52. ↵
    1. Macarenco AC and
    2. Macarenco RS
    : Cutaneous lipomatous sclerosing perineurioma. Am J Dermatopathol 30(3): 291-294, 2008. PMID: 18496437. DOI: 10.1097/DAD.0b013e31816f150a
    OpenUrlCrossRefPubMed
  53. ↵
    1. Friedrich RE,
    2. Hagel C,
    3. Kohlrusch FK,
    4. Schanze I,
    5. Wieland I and
    6. Zenker M
    : Mosaic neurofibromatosis type 1 with multiple cutaneous diffuse and plexiform neurofibromas of the lower leg. Anticancer Res 40(6): 3423-3427, 2020. PMID: 32487640. DOI: 10.21873/anticanres.14327
    OpenUrlAbstract/FREE Full Text
  54. ↵
    1. Yang FC,
    2. Ingram DA,
    3. Chen S,
    4. Zhu Y,
    5. Yuan J,
    6. Li X,
    7. Yang X,
    8. Knowles S,
    9. Horn W,
    10. Li Y,
    11. Zhang S,
    12. Yang Y,
    13. Vakili ST,
    14. Yu M,
    15. Burns D,
    16. Robertson K,
    17. Hutchins G,
    18. Parada LF and
    19. Clapp DW
    : Nf1-dependent tumors require a microenvironment containing Nf1+/— and c-kit-dependent bone marrow. Cell 135(3): 437-448, 2008. PMID: 18984156. DOI: 10.1016/j.cell.2008.08.041
    OpenUrlCrossRefPubMed
    1. Staser K,
    2. Yang FC and
    3. Clapp DW
    : Pathogenesis of plexiform neurofibroma: tumor-stromal/hematopoietic interactions in tumor progression. Annu Rev Pathol 7: 469-495, 2012. PMID: 22077553. DOI: 10.1146/annurev-pathol-011811-132441
    OpenUrlCrossRefPubMed
  56. ↵
    1. Zhu Y,
    2. Ghosh P,
    3. Charnay P,
    4. Burns DK and
    5. Parada LF
    : Neurofibromas in NF1: Schwann cell origin and role of tumor environment. Science 296(5569): 920-922, 2002. PMID: 11988578. DOI: 10.1126/science.1068452
    OpenUrlAbstract/FREE Full Text
  57. ↵
    1. Tucker T,
    2. Riccardi VM,
    3. Sutcliffe M,
    4. Vielkind J,
    5. Wechsler J,
    6. Wolkenstein P and
    7. Friedman JM
    : Different patterns of mast cells distinguish diffuse from encapsulated neurofibromas in patients with neurofibromatosis 1. J Histochem Cytochem 59(6): 584-590, 2011. PMID: 21525187. DOI: 10.1369/0022155411407340
    OpenUrlCrossRefPubMed
  58. ↵
    1. Li S,
    2. Chen Z and
    3. Le LQ
    : New insights into the neurofibroma tumor cells of origin. Neurooncol Adv 2(Suppl 1): i13-i22, 2019. PMID: 32642729. DOI: 10.1093/noajnl/vdz044
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Peripheral Nerve Sheath Tumors in Patients With Neurofibromatosis Type 1: Morphological and Immunohistochemical Study
REINHARD E. FRIEDRICH, LOUISA K. N. NÖRNBERG, CHRISTIAN HAGEL
Anticancer Research Mar 2022, 42 (3) 1247-1261; DOI: 10.21873/anticanres.15592
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords