Abstract
Malignant peripheral nerve sheath tumors (MPNST) are aggressive neoplasms, arising either sporadic, in the setting of neurofibromatosis type I (NF1) or prior radiation. Most MPNST occur in adults and their pathogenesis is driven by loss of function mutations in the PRC2 complex, regardless of their clinical presentation. In contrast, pediatric MPNST are rare and their pathogenesis has not been elucidated. In this study we investigate a large cohort of 64 MPNSTs arising in children and young adults (younger than age of 20 years) to better define their clinicopathologic and molecular features. 16 (25%) cases were investigated by MSK-IMPACT, a targeted NGS panel of 505 cancer genes. Most patients (80%) were aged 11-20 years. A history of NF1 was established in half of the cases. Mean tumor size was 8.5 cm. The most common locations included the extremities (34%) and abdomen/pelvis (27%). Histologically, 89% of high-grade MPNST showed conventional features, while remaining 3 cases showed a predominant epithelioid phenotype. Heterologous differentiation occurred in 25% of high grade cases, with half showing rhabdomyoblastic differentiation. Tumors arose in a background of a plexiform neurofibroma (16%), neurofibroma (13%) and schwannoma in 2 cases (3%). Immunohistochemically, H3K27me3 expression was lost in 82% of conventional high-grade MPNST analyzed, while loss of SMARCB1 expression was seen in one epithelioid MPNST. Genomically, all cases showed more than one genetic abnormality, with 53% showing mutations in EED / SUZ12 genes, and 47% cases each harboring alterations in NF1 and CDKN2A/CDKN2B genes. At last follow-up, 30% patients died of disease, 28% were alive with disease and 42% had no evidence of disease. NF1 status did not correlate with overall survival. In conclusion, half of pediatric and young adult MPNST were NF1-related and showed loss of function alterations in PRC2 complex, NF1, and CDKN2A, similar to the adult counterpart. Thus, H3K27me3 loss of expression may be used in the diagnosis of high grade MPNSTs in children. Moreover, a small subset of pediatric MPNST have an epithelioid morphology with different pathogenesis.
Keywords: MPNST, children, Neurofibromatosis, Neurofibroma, H3K27me3
INTRODUCTION
Malignant peripheral nerve sheath tumor (MPNST) represents an aggressive soft tissue sarcoma often arising from a pre-existing peripheral nerve sheath tumor, or in the setting of Neurofibromatosis type I (NF1).1 Additionally, MPNSTs can also arise sporadically or as a radiation-associated tumor. The molecular pathogenesis of adult MPNSTs has been recently characterized demonstrating recurrent and concurrent inactivating mutations in three pathways involving the NF1, CDKN2A/CDKN2B and the PRC2 core components.2–4 These alterations drive the sarcomagenesis of high grade MPNST irrespective of the underlying clinical setting they occur. However, their incidence varies, with >90% of sporadic and radiation-induced MPNST showing PRC2 complex abnormalities, while only two-thirds of high grade NF1-related MPNST show these alterations.2, 5 MPNST driven by loss of function alterations in the PRC2 core components due to mutations in either EED or SUZ12 genes show complete loss of H3K27me3 protein expression by immunohistochemistry, which has become a reliable ancillary diagnostic marker, being lost in 70-80% of all adult high grade MPNSTs.5, 6 In contrast, the less common epithelioid MPNST lack alterations in the PRC2 complex and show instead SMARCB1 gene inactivation.7 MPNSTs occur with predilection in adult patients between 20-50 years of age, while pediatric cases are uncommon and mostly occurring in the setting of NF1.1 In adults, MPNSTs are aggressive tumors with reported rates of LR and distant metastasis ranging from 40-68%.8, 9 Clinicopathologic features such as NF1-association, truncal location, tumor size greater than 5 cm, and high grade morphology are regarded as adverse prognostic factors.10, 11 In children, however, both the clinicopathologic features and molecular alterations have not been yet characterized. In this study we investigate a large cohort of MPNSTs arising in children and young adults using the latest diagnostic criteria to better define the clinicopathologic and molecular features of these tumors in this age group compared to the adult counterpart.
MATERIALS AND METHODS
Patient Selection
Archival and consultation material from pediatric patients with diagnosis of MPNST was retrieved from the Pathology Department files at Memorial Sloan Kettering Cancer Center (2000-2022). The patient’s clinical records were reviewed to record the clinical and epidemiologic parameters, in particular the patients’ clinical history of NF1, clinical management and follow-up information. The study was approved by the Institutional Review Board at MSKCC (IRB# 02-060).
Pathologic Features.
The pathologic features of these cases were re-reviewed and the diagnosis of MPNST was confirmed. Clinicopathologic features including size, location, association with NF1, presence of cytologic atypia, spindle or epithelioid morphology and background of pre-existing peripheral nerve sheath tumor (neurofibroma, schwannoma, plexiform neurofibroma) were recorded for each case. In sporadic MPNSTs the diagnosis was based on multiple features including presence of nerve sheath tumor association, patchy S100 positivity and absence of other immunohistochemical staining for markers such as Cytokeratin, EMA, TLE1, and CD34 to differentiate from histologic mimics such as synovial sarcoma and fibrosarcoma. Where available, molecular studies (FISH for SS18 gene rearrangements or Archer FusionPlex) were used to confirm diagnoses. A 2-tier grading system was used to categorize the MPNSTs into high and low grade based on the cellularity, mitotic activity, presence of necrosis, and loss of S100 immunopositivity.12 Immunohistochemical stains for S100 (Cell Marque, Mo/4C4.9, 1:600), H3K27me3 (Cell Signaling, C36B11, 1:100), Pan-cytokeratin (DAKO, AE1/AE3, 1:400), EMA (Ventana, E29), SOX10 (Biocare, BC34, 1:200), Laminin (Abnova, LAM-89, 1:100), Collagen type IV (Ventana, CIV22), CD34 (Ventana, QBEnd-10), TLE1 (Santa Cruz, M101, 1:50) and BAF47 (BD Bioscience, 25, 1:200) were analyzed and recorded when available.
MSK-IMPACT Assay
Details of the MSK-IMPACT assay have been previously published.13 Briefly, MSK-IMPACT is a comprehensive molecular profiling assay that involves hybridization capture and deep sequencing of all exons and selected introns of up to 505 oncogenes and tumor-suppressor genes, allowing the detection of point mutations, small and large insertions or deletions, and rearrangements. In addition to capturing all coding regions of the genes, the assay also captures >1000 intergenic and intronic single-nucleotide polymorphisms (tiling probes), interspersed homogenously across the genome, aiding the accurate assessment of genome-wide copy number. The assay uses paired tumor / normal DNA for sequencing and analysis. Only somatic alterations seen in the tumor are reported, while germline alterations are filtered out. In total, the probes target approximately 1.2 megabases of the human genome.
Statistical Analysis
Statistical analysis was conducted on the GraphPad Prism platform (version 9; GraphPad Software, San Diego, CA). Associations between MPNST and clinicopathological factors were assessed by Pearson χ2, Fisher exact or Kruskal-Wallis tests according to whether the variables were categorical or continuous. The overall survival was measured from the date of surgery and examined by Kaplan-Meier analysis and the log-rank test. Two-sided P values < 0.05 were considered significant in all statistical analyses.
RESULTS
Clinicopathologic Features
Sixty-four patients were included in the study, including 33 females and 31 males. The clinicopathologic findings are summarized in Supplementary Table 1. Thirteen (20%) patients were younger than age 10, while remaining 51 patients (80%) were aged 11-20 years. A clinical history of NF1 was identified in 37 patients (58%), 8 (13%) of whom were less than 10 years of age and 29 (45%) between 11-20 years of age. Tumor size was available in 47 cases and ranged from 1.8 to 18.5 cm (mean, 8.5 cm). The most common location was the soft tissue of the extremities (22, 34%). Other sites included abdomen / pelvis (17, 27%), head and neck (9, 14%), thorax / chest wall (8, 13%), brain / spinal cord (6, 9%), and 1 each in the penis and paratesticular region.
Most pediatric MPNST are histologically high grade and show a conventional spindle cell morphology.
All except 7 cases (89%) were classified as high grade MPNST, while remaining 11% were diagnosed as low grade MPNST. Histologically, 60 (94%) tumors showed features of conventional MPNST with fascicles of spindle cells and alternating hypocellular and hypercellular areas. (Figure 1A) The tumor cells showed primitive morphology with hyperchromatic spindled nuclei and scant pale eosinophilic cytoplasm. (Figure 1B) Mitotic figures were conspicuous, and areas of geographic necrosis were noted. (Figure 1C) Three (6%) of the tumors showed pure epithelioid morphology.
Figure 1: Morphologic features of pediatric high-grade malignant peripheral nerve sheath tumors.
A) Spindle cells with elongated nuclei in intersecting fascicles. (H&E, 200x) B) Primitive appearing morphology with fascicles of dense spindle cells with scant pale eosinophilic cytoplasm. (H&E, 200x) C) Areas of geographic necrosis within the tumor. (H&E, 100x) D) Immunohistochemical stain for H3K27me3 showing complete loss of expression in the tumor cells (note the endothelial cell positivity – internal positive control).
Heterologous differentiation was noted in 16 high grade MPNST cases (25%), with 9 of the cases showing rhabdomyoblastic differentiation, 3 showing osteosarcomatous / chondrosarcomatous differentiation, 2 with glandular differentiation and 1 each with primitive neuroectodermal tumor (PNET) and angiosarcoma features. (Figure 2A–2C)
Figure 2: Morphologic variants of pediatric MPNST.
A) High grade MPNST in the pelvis of a 16 year old male with heterologous rhabdomyoblastic differentiation. (H&E, 200x) B) High grade MPNST in the brachial plexus of an 18 year old male, NF1-associated, showing heterologous angiosarcomatous differentiation (H&E, 200x) C) High grade MPNST in the elbow of a 9 year old male, NF1-associated, showing heterologous osteosarcomatous differentiation (H&E, 200x) D) Epithelioid MPNST (20yF, thigh) showing diffuse sheets of epithelioid cells with round nuclei and minimal cytoplasm. (H&E, 200x) Inset showing loss of SMARCB1 (INI1) expression in tumor cells and retained expression in the background endothelial cells. (INI1 stain, 400x)
High grade MPNST tumors arose in a background of plexiform neurofibroma in 10 cases (16%) and intraneural neurofibroma in 8 cases (13%).
Among the NF1-related high grade MPNST, 15 of 37 (41%) arose in a background of neurofibroma, while in the sporadic high grade MPNST group only 2 of 27 (7%) patients arose in the background of neurofibroma. Two (7%) tumors arose in the background of a conventional schwannoma, one of them showing epithelioid morphology (epithelioid MPNST), while the other showed a mixed spindle and epithelioid morphology (conventional MPNST).
Immunohistochemical staining for S100 was available in 45 high grade MPNSTs. Thirty (67%) of high grade tumors showed focal to patchy positivity of S100, while 15 cases (33%) were completely negative. H3K27me3 immunohistochemistry data was available on 22 (34%) high-grade tumors, including 13 (59%) NF1-associated MPNST. Eighteen (82%) cases showed complete loss of nuclear staining (Figure 1D), 1 case showing partial staining and 3 cases retained expression.
Low Grade MPNST rarely occurs in the pediatric age group.
Seven (11%) cases were classified as low-grade MPNST. Three of the seven (43%) low-grade tumors arose in the background of peripheral nerve sheath tumors, one in a plexiform neurofibroma and two in a background of neurofibroma. One of the two cases that arose in a background of neurofibroma had clinically documented NF1.
Immunohistochemical staining data for S100 was available in 4 low-grade MPNST cases, all of which showed diffuse positivity for S100. H3K27me3 immunohistochemistry data was available in one low-grade MPNST which showed retained expression. In three of the cases schwannian differentiation was supported in addition to the diffuse staining for S100, by either immunohistochemical stains for laminin and collagen type IV highlighting the basement membrane material deposition (2 cases) and by electron microscopy showing similar findings (one case).
NF1-associated MPNST are predominantly high grade tumors and show a high rate of H3K27me3 loss of expression.
Thirty-seven (58%) cases in our study cohort were associated with NF1, with the remaining 27 (42%) being sporadic MPNST. Among NF1-MPNST, all except one (97%) were high grade MPNST. In contrast, 21 of the 27 (78%) sporadic MPNST were high grade, while all but one of the 7 low grade MPNST were sporadic MPNST.
The NF1-MPNST group included 20 females and 17 males, while the sporadic MPNST group had 13 females and 14 males. The NF1-MPNST group included 8 patients aged 0-10 years and 29 patients between 11-20 years. In the sporadic group, 5 patients were between 0-10 years and 22 were between 11-20 years of age. The tumors in the NF1-MPNST group were slightly larger (mean size: 8.9 cm) in comparison to those in the sporadic MPNST group (mean size: 7.5 cm).
Immunohistochemical staining for H3K27me3 was performed in 13 NF1-MPNST cases, 12 (92%) of which showed complete loss of expression. In the sporadic group, H3K27me3 expression was lost in 5 of the 8 (62%) high grade tumors analyzed.
Half of high grade conventional MPNST in children show genomic perturbations of NF1, CDKN2A/B and PRC2 core components.
MSK-IMPACT data was available in 16 high-grade MPNST cases, including 15 (94%) conventional MPNST and one (6%) epithelioid MPNST. (Figure 3) Twelve (71%) of the cases were NF1-associated and 5 (29%) were sporadic MPNST. All cases showed multiple genetic abnormalities. The most common recurrent genetic abnormalities noted included mutations / deletions in EED / SUZ12 genes (9 cases), mutations in the NF1 gene (8 cases), and deletions of CDKN2A gene (8 cases). EGFR gains / amplification and TP53 gene mutations were identified in 2 cases each. Although EED, EGFR and TP53 gene alterations were noted only in the NF1-MPNST cases, NF1, SUZ12, and CDKN2A/CDKN2B gene alterations were noted both in NF1- and sporadic MPNST. Tumor mutational burden (TMB) data was determined in 7 of the cases and ranged from 0 to 3.5 (median- 2.5).
Figure 3:
A comprehensive profile of the genomic alterations identified by targeted exome sequencing in 16 pediatric MPNST cases, along with clinico-pathologic features, NF1 status and immunohistochemical H3K27me3 loss of expression. The most common genetic alterations are noted in the PRC2 complex genes (SUZ12 / EED), NF1 and CDKN2A/CDKN2B. The single epithelioid MPNST showed SMARCB1 genetic abnormality. Note the loss of H3K27me3 expression in a majority of the cases.
Targeted RNA sequencing (MSK-ARCHER) performed on 5 cases of high grade MPNST (2 NF1-associated and 3 sporadic) and showed no evidence of gene fusions.
Epithelioid MPNST are rare in pediatric age group and are unrelated to NF1 status.
Three (4%) cases of epithelioid MPNST were identified with large, plump, epithelioid cells with abundant cytoplasm in a lobulated growth pattern. (Figure 2D). No evidence of NF1-association was identified in any of the cases. One of these 3 cases arose in the background of conventional schwannoma. In this case, immunohistochemical stains showed positivity for S100 in the background schwannoma component but loss of S100 staining in the malignant epithelioid component. One of the 3 cases had MSK-IMPACT performed and identified 2 SMARCB1 mutations (missense and splice variant), 1 SMARCA4 mutation (missense) and PIK3CB gene mutation. This tumor also showed strong S100 positivity and loss of INI1 (SMARCB1) expression by immunohistochemistry.
Age, NF1 status, or presence of heterologous elements do not correlate with overall survival.
Follow-up was available in 43 cases (67%), including 41 high grade and 2 low grade tumors, with a duration ranging from 4 to 204 months (median – 33 months). Treatment data was available in 45 cases (70%) and included either single modalities of surgery and chemotherapy or a combination of two or three of these therapies. The most common modality of treatment was a combination of surgery, chemotherapy and radiation therapies (18, 38%), followed by surgery and radiation (10, 21%), surgery and chemotherapy (9, 19%) and only surgery (9, 19%).
Local recurrence was observed in 15 cases (35%) including in one patient with low grade MPNST. Distant recurrence was seen in 12 cases (28%) (all high grade) and both local and distant recurrences were seen in 2 patients (5%). At last follow-up for the high grade MPNST cohort, 13 patients (30%) died of disease, 12 (28%) were alive with disease and 16 (37%) had no evidence of disease. Two patients with low grade MPNST that had follow-up data available had no evidence of disease at 40 and 144 months. Follow-up data was available in one epithelioid MPNST, the patient being alive with disease at 24 months. The 5-year overall survival (OS) of the entire study group was 66.2%, and the 5-year OS for the high grade conventional MPNST was 64.2%. (Supplementary Figure 1A)
Among the 43 patients with follow-up data available, there were 27 (63%) NF1-associated tumors. Statistical analysis showed that there was no significant difference in OS between NF1 vs non-NF1 related MPNST (p=0.65), nor between age groups (younger or older than age 10) (p=0.70), nor the presence or absence of heterologous differentiation (p=0.39). (Supplementary Figure 1B–1D)
DISCUSSION
MPNST is a rare disease that occurs with predilection in adult population. Based on the Surveillance, Epidemiology and End Results (SEER) database, the overall incidence of MPNST among the entire population was 1.46 per million person-years, with a peak incidence in the elderly population aged 75-79 years (3.75 per million person-years), while only 10-20% of the cases (0.56 per million person-years) in the pediatric population (0-19 years).14 There were no statistically significant differences in incidence rates among the different ethnic and racial groups. Within the pediatric age group (0-19 years), the highest incidence of MPNSTs was 15-19 years. In the largest study to date of 159 pediatric MPNSTs (patients less than 21 years) by Meister et al,15 67% of their cases occurred in adolescent children. In keeping to these findings, our study also showed almost 80% of the cases occurred in children in the 11-20 years age group.
MPNSTs have an increased propensity to occur in the setting of NF1 syndrome, however, the reported incidence is highly variable, ranging from 21% to 67%. 16–19 Focusing specifically on the pediatric age group, the reported rates of NF1-related MPNSTs have a wide range from 17% to 72%.15, 16, 20, 21 The study by Bates et al based on the SEER database did not include NF1 status information in the database.14 In our study, 58% of the children with MPNST had clinically diagnosed NF1.
The anatomic location of pediatric MPNSTs is also variable in the literature. In the larger studies by Meister et al.15 and Carli et al.16, the extremities were the most common location, comprising around 40% of the cases. The study by An et al.20 with 11 patients reported that abdominal cavity (45%) is the most common site, followed by head and neck and thorax. In our study, the extremities were the most common site (34%), followed closely by abdomen/pelvis (27%), with head and neck, thorax/chest wall and brain / spinal cord in the range of 12-13% each. The NF1-associated group, in our study, showed a similar anatomic distribution with extremities being the most common site (39%) followed by abdomen / pelvic sites (24%).
The mean reported size of the tumors in our cohort of cases is 8.5 cm. This is identical to those reported in the study by An et al.20 Two other studies on pediatric MPNSTs15, 16 report that tumors greater than 5 or 10 cm, respectively, represent two-thirds of their case cohorts.
Similar to the adult population, a vast majority of the tumors (90%) were classified as high grade. The morphology was mostly that of conventional spindle, predominantly monomorphic, cells with fascicular arrangement. In one-third of our cases a pre-existent benign peripheral nerve sheath tumor was present, including neurofibroma (31%), plexiform neurofibroma (16%) and, in one case, sporadic MPNST with epithelioid morphology arose in a background of conventional schwannoma. Heterologous differentiation was noted in 25% of high grade MPNST study cases, with about half of these cases showing a heterologous rhabdomyoblastic differentiation (malignant Triton tumor). The reported rate of heterologous rhabdomyoblastic differentiation (malignant Triton tumor) in all MPNST is 17.9%.11 The other types of heterologous differentiations noted in our study included osteosarcoma, chondrosarcoma, angiosarcoma, PNET (primitive neuroectodermal tumor) and glandular differentiation.
About 70-80% of all high grade conventional adult MPNSTs show complete loss of expression of H3K27me3 due to alterations in the PRC2 complex (see below) and represents a reliable ancillary marker in the diagnosis of high grade MPNST, in particular in the sporadic and radiation-associated setting.5, 6 The loss of H3K27me3 expression in pediatric MPNST has not been previously established. Our results show complete loss of expression in 82% of the high grade pediatric MPNST cases studied. NF1-MPNST showed higher rate of H3K27me3 loss of expression (92%) in comparison to the sporadic MPNST (62%).
The molecular characterization of pediatric MPNSTs has not been performed to date. Our study investigates one of the largest cohort of molecularly-analyzed pediatric MPNSTs using a targeted exome sequencing platform on 16 high grade lesions. Genetic alterations identified in our cohort of pediatric MPNSTs recapitulated the molecular underpinnings of adult MPNSTs, affecting three main pathways including perturbations in PRC2 core components (EED or SUZ12), NF1 and CDKN2A/CDKN2B.2, 3, 22, 23 These molecular abnormalities were observed in a non-mutually exclusive manner in approximately one-half of the cases. (Figure 3) Other recurrent events identified at a low frequency included gains of the EGFR gene (17%) and mutations in the TP53 gene (5%). Most of the analyzed tumors also showed associated non-recurrent genetic abnormalities. The single epithelioid MPNST included in our analysis (case 64) showed two SMARCB1 (missense and splice site) mutations and an associated SMARCA4 mutation. Indeed, SMARCB1 alterations are the main driver alteration reported in up to 81% in epithelioid MPNST.7 The most common SMARCB1 gene alterations include homozygous deletions, and less common non-sense, frameshift or splice-site gene mutations.7 As a result of the wide application of RNA sequencing in clinical practice, a subset of previously diagnosed non-NF1 related, mainly low grade MPNST occurring in children and young adults have now been reclassified as kinase fusion positive tumors with often S100 and CD34 coexpression.24 However, none of the 5 patients tested, including 4 sporadic high-grade cases, showed the presence of gene fusions by Archer FusionPlex.
Conflicting data regarding the survival impact of NF1 status among high grade MPNST patients still persists in the literature. In the study by Hwang et al.10, encompassing 95 adult MPNST patients, including low and high grade cases, treated at their health system over a 27-year period, 33% of the cases were associated with NF1. Although by univariate analysis the NF1-related MPNST group had a significantly lower 10-year OS rate (45±11%) compared to 60±8% in sporadic MPNST patients with localized disease, multivariate analysis showed no statistically different outcomes between the two groups. In contrast, the study by Le Guellec et al.11 including 106 adult patients with MPNST (68 NF1-MPNST and 38 sporadic MPNST), showed that NF1 status was an independent prognostic factor for survival by multivariate analysis. In their study, NF1 patients had a poor 5-year OS of 34.8%, compared to the sporadic group (68.5%). In the study by Meister et al.15 focusing on children and young adults less than 21 years of age (159 patients), the 5-year and 10-year OS was 54.6% and 47.1%, respectively. In their study, NF1-MPNST occurred in older children (>10 years), had a larger size (> 5 cm) and showed a higher grade compared to patients without NF1 history. NF1-MPNST patients also had a significantly lower 5-year OS (32%) compared with the sporadic MPNST patients (62%) in a univariate analysis. However, there was no statistically significant difference in a multivariate analysis, as well as no difference in response to chemotherapy or radiotherapy between the two groups. Our study showed a 5-year OS of 66.2%, which appears to be better than the survival reported in the other studies of pediatric MPNST patients.15, 20 Our data does not find a statistically significant difference between the OS of NF1-associated MPNST and the sporadic groups in pediatric patients (p value 0.65).
In our study, combinations of surgery, chemotherapy and radiation therapy were used in the management of the majority (73%) of patients, with all 3 modalities combined being used in 38% of the patients. This is in keeping with the approach reported by other investigators,14, 15 where a combinatorial therapeutic approach was the most widely used in the management of these patients.
In conclusion, we report on a large cohort of MPNST in children managed at our tertiary cancer center, which includes the largest number of molecularly analyzed pediatric MPNSTs to date. In the current study most patients were adolescent, of equal gender and in half of the cases a clinical history of NF1 was documented. The tumors most common presentation was in the extremities and abdomen-pelvic locations. In keeping with the adult counterpart, the high grade pediatric MPNST were associated with similar genomic driver events, with genomic perturbations in PRC2 core components (SUZ12 / EED), NF1 mutations and CDKN2A/CDKN2B deletions. Thus, loss of H3K27me3 expression may be used as a powerful immunohistochemical diagnostic marker in both pediatric and adult MPNSTs. Epithelioid MPNST is similarly rare in children and is not associated with NF1 syndrome, but rather harbor SMARCB1 loss of function alterations. Additional investigations to explore different therapeutic approaches are warranted to counter these aggressive tumors with poor prognosis.
Supplementary Material
Supplementary Figure 1: Kaplan-Meyer survival plots. A. Overall survival (OS) of the study group of patients with available follow-up. B. OS of NF1-associated MPNSTs in comparison with non-NF1-associated MPNSTs from the study cohort. C. OS of patients younger than 10 years compared to patients aged 11-20 years. D. OS of high grade MPSNT patients with heterologous differentiation in comparison with high grade MPNST without heterologous differentiation from the study cohort.
ACKNOWLEDGEMENTS
The authors would like to thank Bruce Crilly for preparation of composite figures.
Supported in part by: P50 CA217694 (NA, CRA), P30 CA008748 (NA, CRA), Kristin Ann Carr Foundation (CRA), Cycle for survival (LW, CRA)
REFERENCES
- 1.WHO Classification of Tumours Editorial Board. WHO classification of tumours: soft tissue and bone tumours. 5th ed. Lyon, France: International Agency for Research on Cancer; 2020. [Google Scholar]
- 2.Lee W, Teckie S, Wiesner T, et al. , PRC2 is recurrently inactivated through EED or SUZ12 loss in malignant peripheral nerve sheath tumors. Nat Genet, 2014. 46(11): p. 1227–32. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Zhang M, Wang Y, Jones S, et al. , Somatic mutations of SUZ12 in malignant peripheral nerve sheath tumors. Nat Genet, 2014. 46(11): p. 1170–2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Sohier P, Luscan A, Lloyd A, et al. , Confirmation of mutation landscape of NF1-associated malignant peripheral nerve sheath tumors. Genes Chromosomes Cancer, 2017. 56(5): p. 421–426. [DOI] [PubMed] [Google Scholar]
- 5.Prieto-Granada CN, Wiesner T, Messina JL, et al. , Loss of H3K27me3 Expression Is a Highly Sensitive Marker for Sporadic and Radiation-induced MPNST. Am J Surg Pathol, 2016. 40(4): p. 479–89. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Schaefer IM, Fletcher CD, and Hornick JL, Loss of H3K27 trimethylation distinguishes malignant peripheral nerve sheath tumors from histologic mimics. Mod Pathol, 2016. 29(1): p. 4–13. [DOI] [PubMed] [Google Scholar]
- 7.Schaefer IM, Dong F, Garcia EP, et al. , Recurrent SMARCB1 Inactivation in Epithelioid Malignant Peripheral Nerve Sheath Tumors. Am J Surg Pathol, 2019. 43(6): p. 835–843. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Hruban RH, Shiu MH, Senie RT, et al. , Malignant peripheral nerve sheath tumors of the buttock and lower extremity. A study of 43 cases. Cancer, 1990. 66(6): p. 1253–65. [DOI] [PubMed] [Google Scholar]
- 9.Kourea HP, Bilsky MH, Leung DH, et al. , Subdiaphragmatic and intrathoracic paraspinal malignant peripheral nerve sheath tumors: a clinicopathologic study of 25 patients and 26 tumors. Cancer, 1998. 82(11): p. 2191–203. [DOI] [PubMed] [Google Scholar]
- 10.Hwang IK, Hahn SM, Kim HS, et al. , Outcomes of Treatment for Malignant Peripheral Nerve Sheath Tumors: Different Clinical Features Associated with Neurofibromatosis Type 1. Cancer Res Treat, 2017. 49(3): p. 717–726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Le Guellec S, Decouvelaere AV, Filleron T, et al. , Malignant Peripheral Nerve Sheath Tumor Is a Challenging Diagnosis: A Systematic Pathology Review, Immunohistochemistry, and Molecular Analysis in 160 Patients From the French Sarcoma Group Database. Am J Surg Pathol, 2016. 40(7): p. 896–908. [DOI] [PubMed] [Google Scholar]
- 12.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. Sep;67:1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Cheng DT, Mitchell TN, Zehir A, et al. , Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT): A Hybridization Capture-Based Next-Generation Sequencing Clinical Assay for Solid Tumor Molecular Oncology. J Mol Diagn, 2015. 17(3): p. 251–64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Bates JE, Peterson CR, Dhakal S, et al. , Malignant peripheral nerve sheath tumors (MPNST): a SEER analysis of incidence across the age spectrum and therapeutic interventions in the pediatric population. Pediatr Blood Cancer, 2014. 61(11): p. 1955–60. [DOI] [PubMed] [Google Scholar]
- 15.Meister MT, Scheer M, Hallmen E, et al. , Malignant peripheral nerve sheath tumors in children, adolescents, and young adults: Treatment results of five Cooperative Weichteilsarkom Studiengruppe (CWS) trials and one registry. J Surg Oncol, 2020. 122(7): p. 1337–1347. [DOI] [PubMed] [Google Scholar]
- 16.Carli M, Ferrari A, Mattke A, et al. , Pediatric malignant peripheral nerve sheath tumor: the Italian and German soft tissue sarcoma cooperative group. J Clin Oncol, 2005. 23(33): p. 8422–30. [DOI] [PubMed] [Google Scholar]
- 17.Glover TW, Stein CK, Legius E, et al. , Molecular and cytogenetic analysis of tumors in von Recklinghausen neurofibromatosis. Genes Chromosomes Cancer, 1991. 3(1): p. 62–70. [DOI] [PubMed] [Google Scholar]
- 18.Evans DGR, Salvador H, Chang VY, et al. , Cancer and Central Nervous System Tumor Surveillance in Pediatric Neurofibromatosis 1. Clin Cancer Res, 2017. 23(12): p. e46–e53. [DOI] [PubMed] [Google Scholar]
- 19.Doorn PF, Molenaar WM, Buter J, et al. , Malignant peripheral nerve sheath tumors in patients with and without neurofibromatosis. Eur J Surg Oncol, 1995. 21(1): p. 78–82. [DOI] [PubMed] [Google Scholar]
- 20.An HY, Hong KT, Kang HJ, et al. , Malignant peripheral nerve sheath tumor in children: A single-institute retrospective analysis. Pediatr Hematol Oncol, 2017. 34(8): p. 468–477. [DOI] [PubMed] [Google Scholar]
- 21.Meis JM, Enzinger FM, Martz KL, et al. , Malignant peripheral nerve sheath tumors (malignant schwannomas) in children. Am J Surg Pathol, 1992. 16(7): p. 694–707. [DOI] [PubMed] [Google Scholar]
- 22.De Raedt T, Beert E, Pasmant E, et al. , PRC2 loss amplifies Ras-driven transcription and confers sensitivity to BRD4-based therapies. Nature, 2014. 514(7521): p. 247–51. [DOI] [PubMed] [Google Scholar]
- 23.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): p. 981–992. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Suurmeijer AJH, Dickson BC, Swanson D, et al. , A novel group of spindle cell tumors defined by S100 and CD34 co-expression shows recurrent fusions involving RAF1, BRAF, and NTRK1/2 genes. Genes Chromosomes Cancer, 2018. 57(12): p. 611–621. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Figure 1: Kaplan-Meyer survival plots. A. Overall survival (OS) of the study group of patients with available follow-up. B. OS of NF1-associated MPNSTs in comparison with non-NF1-associated MPNSTs from the study cohort. C. OS of patients younger than 10 years compared to patients aged 11-20 years. D. OS of high grade MPSNT patients with heterologous differentiation in comparison with high grade MPNST without heterologous differentiation from the study cohort.