Hongtao Ye

Role of the transcription factor T (brachyury) in the pathogenesis of sporadic chordoma: a genetic and functional-based study

A variety of analyses, including fluorescence in situ hybridization (FISH), quantitative PCR (qPCR) and array CGH (aCGH), have been performed on a series of chordomas from 181 patients. Twelve of 181 (7%) tumours displayed amplification of the T locus and an additional two cases showed focal amplification; 70/181 (39%) tumours were polysomic for chromosome 6, and 8/181 (4.5%) primary tumours showed a minor allelic gain of T as assessed by FISH. No germline alteration of the T locus was identified in non‐neoplastic tissue from 40 patients. Copy number gain of T was seen in a similar percentage of sacrococcygeal, mobile spine and base of skull tumours. Knockdown of T in the cell line, U‐CH1, which showed polysomy of chromosome 6 involving 6q27, resulted in a marked decrease in cell proliferation and morphological features consistent with a senescence‐like phenotype. The U‐CH1 cell line was validated as representing chordoma by the generation of xenografts, which showed typical chordoma morphology and immunohistochemistry in the NOD/SCID/interleukin 2 receptor [IL2r]

The role of epidermal growth factor receptor in chordoma pathogenesis: a potential therapeutic target.

gamma^{rm{null}}

Diagnostic value of H3F3A mutations in giant cell tumour of bone compared to osteoclast-rich mimics.

mouse model. In conclusion, chromosomal aberrations resulting in gain of the T locus are common in sporadic chordomas and expression of this gene is critical for proliferation of chordoma cells in vitro. Copyright

Detection of USP6 gene rearrangement in nodular fasciitis: an important diagnostic tool

Chordoma, the molecular hallmark of which is T (brachyury), is a rare malignant bone tumour with a high risk of local recurrence and a tumour from which metastatic disease is a common late event. Currently, there is no effective drug therapy for treating chordomas, although there is evidence that some patients respond to the empirical use of epidermal growth factor receptor (EGFR) antagonists. The aim of this study was to determine the role of EGFR in the pathogenesis of chordoma. Paraffin‐embedded material from 173 chordomas from 160 patients [sacro‐coccygeal (n = 94), skull‐based (n = 50), and mobile spine (n = 16)] was analysed by immunohistochemistry and revealed total EGFR expression in 69% of cases analysed. Of 147 informative chordomas analysed by FISH, 38% revealed high‐level EGFR polysomy, 4% high‐level polysomy with focal amplification, 18% low‐level polysomy, and 39% disomy. Phospho‐receptor tyrosine kinase array membranes showed EGFR activation in the chordoma cell line U‐CH1 and all of the three chordomas analysed. Direct sequencing of EGFR (exons 18–21), KRAS, NRAS, HRAS (exons 2, 3), and BRAF (exons 11, 15) using DNA from 62 chordomas failed to reveal mutations. PTEN expression was absent by immunohistochemistry in 19 of 147 (13%) analysed chordomas, only one of which revealed high‐level polysomy of EGFR. The EGFR inhibitor tyrphostin (AG 1478) markedly inhibited proliferation of the chordoma cell line U‐CH1 in vitro and diminished EGFR phosphorylation in a dose‐dependant manner, a finding supported by inhibition of phosphorylated Erk1/2. p‐Akt was suppressed to a much lesser degree in these experiments. There was no reduction of T as assessed by western blotting. These data implicate aberrant EGFR signalling in the pathogenesis of chordoma. This study provides a strategy for patient stratification for treatment with EGFR antagonists. Copyright

Frequency of Mouse Double Minute 2 (MDM2) and Mouse Double Minute 4 (MDM4) amplification in parosteal and conventional osteosarcoma subtypes

Driver mutations in the two histone 3.3 (H3.3) genes, H3F3A and H3F3B, were recently identified by whole genome sequencing in 95% of chondroblastoma (CB) and by targeted gene sequencing in 92% of giant cell tumour of bone (GCT). Given the high prevalence of these driver mutations, it may be possible to utilise these alterations as diagnostic adjuncts in clinical practice. Here, we explored the spectrum of H3.3 mutations in a wide range and large number of bone tumours (n = 412) to determine if these alterations could be used to distinguish GCT from other osteoclast‐rich tumours such as aneurysmal bone cyst, nonossifying fibroma, giant cell granuloma, and osteoclast‐rich malignant bone tumours and others. In addition, we explored the driver landscape of GCT through whole genome, exome and targeted sequencing (14 gene panel). We found that H3.3 mutations, namely mutations of glycine 34 in H3F3A, occur in 96% of GCT. We did not find additional driver mutations in GCT, including mutations in IDH1, IDH2, USP6, TP53. The genomes of GCT exhibited few somatic mutations, akin to the picture seen in CB. Overall our observations suggest that the presence of H3F3A p.Gly34 mutations does not entirely exclude malignancy in osteoclast‐rich tumours. However, H3F3A p.Gly34 mutations appear to be an almost essential feature of GCT that will aid pathological evaluation of bone tumours, especially when confronted with small needle core biopsies. In the absence of H3F3A p.Gly34 mutations, a diagnosis of GCT should be made with caution.

H3f3a (histone 3.3) G34w Immunohistochemistry: A Reliable Marker Defining Benign and Malignant Giant Cell Tumor of Bone

Dear Editor, Nodular fasciitis is a lesion that is well known for posing a diagnostic challenge to pathologists. The clinical presentation of sudden onset and fast growing rate of this soft tissue lesion along with histological features, comprising a poorly demarcated cellular myofibroblastic mitotically active lesion, frequently raises concern of malignancy. There are no specific immunohistochemical markers for nodular fasciitis and the diagnosis, until recently, was based on histological features alone. The age of occurrence and anatomical site are unhelpful in making the diagnosis as these lesions can occur at any age and at any anatomic soft tissue site, although the most frequent site is the forearm and the trunk and is more common in young adults [5]. These lesions have traditionally been considered reactive in nature because if unresected, they may involute and are not associated with malignant transformation. If the patient is quizzed, a history of trauma to the site of tumour may be obtained. In 2004, a USP6 gene rearrangement (chromosome 17p13) was identified in aneurysmal bone cysts [2, 3]. This rearrangement has been associated with a variety of fusion partners including CDH11, ZNF9, COL1A1, TRAP150 and OMD [3]. In 2011, it was also reported that nodular fasciitis harboured a USP6 rearrangement; however the most common fusion partner, MYH9 in chromosome 22q12.3, found in 65 % of cases of nodular fasciitis has not been reported in aneurysmal bone cysts [1]. We therefore set out to determine the incidence of USP6 gene rearrangements in a series of cases of nodular fasciitis diagnosed at our service in order to incorporate this test as an ancillary diagnostic tool for this lesion. Thirty-four cases, previously diagnosed as nodular fasciitis, were selected from our files. All cases were analysed by fluorescence in situ hybridisation (FISH) using custom-made break-apart BAC probes as previously described [1]. Material was available from 28 of these cases for analysis by RT-PCR for MYH-USP6 fusion transcript using previously published primers sets [1]. Thirty-one cases including 1 reactive tonsil, 5 B cell lymphomas, 9 PNET/Ewing sarcoma, 5 desmoidtype fibromatoses, 5 myxoid liposarcomas and 6 synovial sarcomas were selected as negative controls. All control samples were diagnosed in conjunction with their characteristic genetic alteration. In addition, 16 cases of primary aneurysmal bone cyst were tested. The presence of break-apart signals was assessed in ‘hot spots’ tumour-rich areas where 50 nonoverlapping nuclei were scored. Cases were classified as positive for USP6 gene rearrangement when 15 %, or above, of cells harboured the break-apart signal. All 34 cases of nodular fasciitis analysed by FISH, and 21 of the 28 cases tested by RT-PCR were informative. Ninety-one percent (31/34) of the cases initially diagnosed as nodular fasciitis showed a clear USP6 gene rearrangement by FISH in more than 15 % of the cells (range, 15–80 %; mean, 35.6 %) (Fig. 1). This result confirms the data of Erickson-Johnson et al. reporting USP6 gene rearrangement in 92 % of the cases [1]. M Fernanda Amary and Hongtao Ye contributed equally to this study.

Recurrent mutation of IGF signalling genes and distinct patterns of genomic rearrangement in osteosarcoma

Sir: Amplification of Mouse Double Minute 2 (MDM2) and Mouse Double Minute 4 (MDM4) exert an oncogenic effect mainly by inhibiting the tumour-suppressive effect of p53. Nutlin-3A, an inhibitor of MDM2, has been shown to induce apoptosis in vitro and in vivo in the p53 wild-type MDM2-amplified osteosarcoma (OS) cell lines OSA and MHM. Parosteal OS are surface neoplasms most commonly – but not exclusively – lowgrade tumours and therefore are not treated with the same multidrug chemotherapy as for conventional, intramedullary, high-grade OS (HG OS). However, these tumours can progress to intermediate and highgrade neoplasms, may recur and may present with late metastatic disease. MDM2 amplification has been reported frequently in parosteal OS, tumours which rarely harbour p53 mutations, and therefore are candidates for treatment with Nutlin-3A, particularly when patients have inoperable metastases. The presence of amplification of MDM4, which can also inhibit p53-mediated apoptosis, has never been investigated in parosteal OS but is important, as its presence could counter the benefit from treatment with Nutlin-3A. The number of patients with HG OS who are likely to benefit from Nutlin-3A is low, because p53 is mutated in a significant proportion of such tumours. Nevertheless, the finding that this drug can induce cell cycle arrest in vitro in the non-MDM2-amplified cell line U2OS, which derives from what was described originally as ‘a moderately differentiated osteogenic sarcoma’, implies that a subset of central OS may benefit from this treatment. However, the number of reports of HG OS that have been screened for the presence of MDM2 amplification is small, with only one study reaching 100 cases, and there are no reports of the status of MDM4 amplification in these tumours. FISH was performed on tissue microarrays containing 24 parosteal OS and 166 HG OS resections with duplicate cores for each tumour. The probes for MDM2 on chromosome 12 and MDM4 on chromosome 1, with their respective chromosome enumeration probes, were provided by Zytovision (Bremerhaven, Germany) [Zytolight SPEC MDM2 ⁄ CEN12 (Z-2013) and MDM4 ⁄ 1p12 (Z-2080) Dual Colour Probes]. At least 50 cells were examined per case. Detection of more than three copies of MDM2 or MDM4 in more than 10% of the cells was defined as copy number gain. Additionally, this gain was identified as gene amplification when more than 10 copies of the gene occurred in more than 30% of cells. Aneuploidy was defined as the presence of copy number gain of MDM2 or MDM4 and an equivalent number of respective enumeration probes. The detection of MDM2 amplification in 20 of 24 (83%) parosteal OS supports previous reports that this is a recurrent genetic event in this subtype (Table 1, Figure 1B). MDM4 amplification was not detected in any of the 24 informative OS analysed, although low copy number gain representing chromosomal aneuploidy was identified in eight cases. MDM2 amplification was a rare event in HG OS, appearing in only six of 155 OS (4%), a frequency similar to that found in a previous report (Table 1, Figure 1C). This finding was not associated with a particular subtype. MDM4 amplification was detected

EGFR inhibitors identified as a potential treatment for chordoma in a focused compound screen

Giant cell tumor of bone (GCTB) is a locally aggressive subarticular tumor. Having recently reported that H3.3 G34W mutations are characteristic of this tumor type, we have now investigated the sensitivity and specificity of the anti-histone H3.3 G34W rabbit monoclonal antibody in a wide variety of tumors including histologic mimics of GCTB to assess its value as a diagnostic marker. We also determined the incidence of H3.3 G34 mutations in primary malignant bone tumors as assessed by genotype and H3.3 G34W immunostaining. A total of 3163 tumors were tested. Totally, 213/235 GCTB (90.6%) showed nuclear H3.3 p.G34W immunoreactivity. This was not the case for the rare variants, p.G34L, M, and V, which occurred most commonly in the small bones of the hands, patella, and the axial skeleton. If these sites were excluded from the analysis, H3.3 G34W expression was found in 97.8% of GCTB. Malignant bone tumors initially classified as osteosarcomas were the only other lesions (n=11) that showed G34W expression. Notably an additional 2 previously reported osteosarcomas with a p.G34R mutation were not immunoreactive for the antibody. A total of 11/13 of these malignant H3.3-mutant tumors exhibited an osteoclast-rich component: when imaging was available all but one presented at a subarticular site. We propose that subarticular primary malignant bone sarcoma with H3.3 mutations represent true malignant GCTB, even in the absence of a benign GCTB component.

Sensitivity of MDM2 amplification and unexpected multiple faint alphoid 12 (alpha 12 satellite sequences) signals in atypical lipomatous tumor

Osteosarcoma is a primary malignancy of bone that affects children and adults. Here, we present the largest sequencing study of osteosarcoma to date, comprising 112 childhood and adult tumours encompassing all major histological subtypes. A key finding of our study is the identification of mutations in insulin-like growth factor (IGF) signalling genes in 8/112 (7%) of cases. We validate this observation using fluorescence in situ hybridization (FISH) in an additional 87 osteosarcomas, with IGF1 receptor (IGF1R) amplification observed in 14% of tumours. These findings may inform patient selection in future trials of IGF1R inhibitors in osteosarcoma. Analysing patterns of mutation, we identify distinct rearrangement profiles including a process characterized by chromothripsis and amplification. This process operates recurrently at discrete genomic regions and generates driver mutations. It may represent an age-independent mutational mechanism that contributes to the development of osteosarcoma in children and adults alike.

Fibroblastic growth factor receptor 1 amplification in osteosarcoma is associated with poor response to neo-adjuvant chemotherapy

Chordoma is a rare malignant bone tumour with a poor prognosis and limited therapeutic options. We undertook a focused compound screen (FCS) against 1097 compounds on three well‐characterized chordoma cell lines; 154 compounds were selected from the single concentration screen (1 µm), based on their growth‐inhibitory effect. Their half‐maximal effective concentration (EC50) values were determined in chordoma cells and normal fibroblasts. Twenty‐seven of these compounds displayed chordoma selective cell kill and 21/27 (78%) were found to be EGFR/ERBB family inhibitors. EGFR inhibitors in clinical development were then studied on an extended cell line panel of seven chordoma cell lines, four of which were sensitive to EGFR inhibition. Sapitinib (AstraZeneca) emerged as the lead compound, followed by gefitinib (AstraZeneca) and erlotinib (Roche/Genentech). The compounds were shown to induce apoptosis in the sensitive cell lines and suppressed phospho‐EGFR and its downstream pathways in a dose‐dependent manner. Analysis of substituent patterns suggested that EGFR‐inhibitors with small aniline substituents in the 4‐position of the quinazoline ring were more effective than inhibitors with large substituents in that position. Sapitinib showed significantly reduced tumour growth in two xenograft mouse models (U‐CH1 xenograft and a patient‐derived xenograft, SF8894). One of the resistant cell lines (U‐CH2) was shown to express high levels of phospho‐MET, a known bypass signalling pathway to EGFR. Neither amplifications (EGFR, ERBB2, MET) nor mutations in EGFR, ERBB2, ERBB4, PIK3CA, BRAF, NRAS, KRAS, PTEN, MET or other cancer gene hotspots were detected in the cell lines. Our findings are consistent with the reported (p‐)EGFR expression in the majority of clinical samples, and provide evidence for exploring the efficacy of EGFR inhibitors in the treatment of patients with chordoma and studying possible resistance mechanisms to these compounds in vitro and in vivo.