Amplification of the MYC Gene in Osteosarcoma Secondary to Paget's Disease of Bone

Purpose. In a previous series of 25 human osteosarcoma samples studied for MYC gene amplification, we found amplification in two cases (8%), including one arising in association with Paget's disease (pagetic osteosarcoma). Based on this observation, we further investigated the prevalence of MYC gene amplification in pagetic osteosarcomas. Methods. MYC gene amplification was assessed by Southern blot analysis using frozen tissue samples in five cases of pagetic osteosarcoma and 53 cases of primary (non-pagetic) osteosarcoma. Amplification was considered present if the MYC copy number was six or greater. Results. Three out of five patients (60%) with pagetic osteosarcoma showed MYC gene amplification, whereas it was present in only 5/53 patients (9.4%) with primary osteosarcoma. The incidence of MYC amplification in pagetic osteosarcoma was thus significantly higher than that in primary osteosarcoma (p = 0.016). Discussion. The finding that MYC gene amplification may be more common in pagetic than primary osteosarcoma warrants further study and suggests pathogenetic differences between primary osteosarcomas and those arising in the setting of Paget's disease. Three of the four pagetic osteosarcomas from the present study were previously shown to be immunoreactive for p53, suggesting that p53 mutation may also be a frequent genetic lesion in these tumors.


Introduction
Paget's disease is a benign but precancerous condition affecting the mesenchymal cells of bones. Of the secondary sarcomas arising in Paget's disease of bone, osteosarcomas (pagetic osteosarcomas) are by far the most common and make up a substantial proportion of osteosarcomas of late adulthood. They represent an infrequent but highly lethal complication of Paget's disease of bone. In a recent review of the Memorial Sloan-Kettering Cancer Center (MSKCC) experience with 67 cases of pagetic sarcoma, we found that sarcoma may be the initial manifestation of Paget's disease in up to 50% of patients with this complication. However, remarkably little work has been performed on the molecular genetic basis of sarcomas secondary to Paget's disease.
We have recently examined the status of the MYC gene in a large series of osteosarcomas and found that only 2/25 patients showed MYC gene amplification. 3 Of these two patients, one was a 60-year-old man with Paget's disease and a tibial osteosarcoma. Based on this observation, we have further investigated the incidence of MYC gene amplification in pagetic osteosarcoma and compared it with that in sporadic osteosarcoma.

Patients and methods
Five samples from five patients with pagetic osteosarcoma were identified in our osteosarcoma frozen tissue bank. We also examined 58 frozen tissue samples from 53 patients with primary (nonpagetic) osteosarcoma as a control group. All these cases were confirmed as osteosarcoma by histological review (AGH). The results of 25 patients (24 non-pagetic, one pagetic osteosarcoma) included in the current series have been reported previously. 3 All five samples of pagetic osteosarcoma were obtained from primary tumors in patients with previously diagnosed Paget's disease of bone, while the sporadic (non-pagetic) osteosarcoma samples were obtained from 41 primary and 17 metastatic tumors. The primary sites of the tumors included the tibia (two cases), femur (one), humerus (one) and ilium (one) in the pagetic osteosarcomas, and the femur (18), tibia (13), humerus (four), ilium (four) and other sites (14) in the non-pagetic osteosarcomas. MYC gene amplification was assessed essentially as previously described. 3 In brief, we performed quantitative comparison of the Southern blot hybridization signal of a genomic MYC exon-1 probe (Xho I-Xba I fragment) to that of a probe for a reference gene on chromosome 8, neurofilament light polypeptide (NEFL, band 8p21). Signals were quantified directly on a BioRad Phosphorimager (Hercules, CA, USA). A reference probe from the same chromosome was used to control for the nonspecific effect of polyploidization on the quantitation of gene copy number. 4 HL-60 cell line DNA and human placental tissue DNA were used as positive and negative controls for MYC gene amplification, respectively. The MYC gene is amplified approximately 16-fold in the HL-60 cell line. To score a sample as positive, we used a cut-off point of six copies per cell for the MYC gene amplification normalized to the signal in human placenta. Fractional amplification values were rounded off to the nearest whole number.

Results
MYC gene amplification was detected in 3/5 patients (60%) with pagetic osteosarcoma (Fig. 1). A quantitative analysis of pagetic osteosarcoma cases 1, 2 and 3 showed amplification levels of 8, 11 and 6 copies per cell, respectively. The other two MYC NF pagetic OS = T:a. 2 3 4 5 ::r.. Quantitative comparison of the hybridization signal for MYC and the reference probe NF (neurofilament) showed 11 and 6 copies per cell of the MYC gene in pagetic osteosarcoma cases 2, and 3, respectively. Amplification is visually apparent in case 2, but required quantitative confirmation in case 3. Quantitative analysis in case 5 showed five copies of MYC and it was scored as negative for amplification. Pagetic osteosarcoma cases 4 and 5 showed no amplification. HL-60 and human placenta, respectively, were used as positive and negative controls for amplification. pagetic osteosarcoma cases, 4 and 5, showed no amplification. In contrast, only 5/53 patients (9.4%) with sporadic (non-pagetic) osteosarcoma revealed MYC gene amplification, ranging from 6 to 15 copies (mean 10 copies). The incidence of amplification in pagetic osteosarcoma was significantly higher than that in primary (nonpagetic) osteosarcoma (60% vs 9.4%, p=0.016 odds ratio: 14.4) by two-tailed Fisher's exact test (Table 1).
In previous studies, including some of these cases, we found that 3/4 pagetic osteosarcomas showed p53 over-expression, but none showed amplification of MDM2 (Table 1).6 Clinical features such as sex and primary site were Case 4 also had a history of Paget's disease of the nipple due to extensive underlying intraductal carcinoma, representing a rare coincidence of these two eponymous conditions.

Discussion
The MYC proto-oncogene, which maps to chromosome band 8q24, encodes a transcription factor containing dimerization motifs (helix-loop-helix (HLH) and leucine zipper (LZ)). 7 Normally, MYC dimerizes with another HLH/LZ protein, MAX, and interacts with promoters containing the sequence CACGTG. The spectrum of genes whose transcription is physiologically regulated by MYC is a subject of active study. At the cellular level, .MYC is known to drive cell proliferation in response to extracellular signals and to promote apoptosis in proliferating cells upon withdrawal of the same signals. 7 Recently, the cell cycle phosphatase CDC25 has been identified as a direct target of transcriptional activation by MYC, providing the first clear link between MYC deregulation and cell cycle activation. 8 The present study of frozen tumor samples of pagetic and non-pagetic osteosarcomas suggests that MYC gene amplification is a common molecular genetic alteration in pagetic osteosarcoma compared to its low incidence in sporadic osteosarcoma. To our knowledge, this is the first study of molecular genetic alterations in pagetic osteosarcomas.
That the MYC proto-oncogene (formerly c-myc) plays an important role in osteosarcoma has been noted in various animal models and osteosarcoma 9 cell lines, usually in the form of amplification. 2 In the murine SEWA osteosarcoma cell line, tumorigenicity is proportional to the degree of MYC gene amplification. 13 However, few samples of human osteosarcoma have been studied for MYC gene amplification. Masuda et al. 14 found MYC amplification in 1/3 osteosarcomas, and Ikeda et al. 15 in two of four high-grade pediatric osteosarcomas. In an earlier series, we found MYC gene amplification in only 2/25 osteosarcomas (8%), one of which was a pagetic osteosarcoma. 3 More recently, another group found MYC amplification in 1/8 osteosarcomas (13%) of unspecified type. 16 In this study, we found MYC amplification in 9.4% of sporadic osteosarcomas. These data suggest that the true prevalence of MYC gene amplification in sporadic osteosarcoma is lower than expected from limited earlier studies. This discrepancy may be due to the method of quantification and to the use of surgical specimens, where amplification may be harder to detect because of admixed non-tumor tissue. Nevertheless, in the present series, pagetic osteosarcomas revealed a higher prevalence of MYC gene amplification (60%), suggesting a more important role for MYC gene amplification in pagetic osteosarcomas than in primary (non-pagetic) osteosarcomas.
In addition to the potential biological significance of MYC gene amplification, its presence may also be relevant clinically. In some tumor types, the presence of amplification of specific genes predicts a poorer prognosis. 17'18 Evidence is now accumulating that, likewise, MYC gene amplification is associated with poorer prognosis in musculoskeletal sarcomas including osteosarcomas. 16'19 Correspondingly, in some human osteosarcoma cell lines, MYC gene amplification is associated with higher growth rate. 2'1 Finally, in two cases of pagetic osteosarcoma, we found a coincidence of MYC amplification and p53 over-expression (indicative of p53 point mutation), in contrast to a previous study 22 of these two alterations in sarcomas.