Abstract
Background. The purpose of this work was to study clinical and histopathological tumor characteristics of patients treated for synchronous or metachronous skeletal osteosarcoma at The Norwegian Radium Hospital from January 1, 1980 to January 1, 2008. Patients and methods. The hospital sarcoma database and patient records were reviewed to identify cases with synchronous or metachronous skeletal osteosarcoma. Patients with more than one skeletal lesion in the absence of pulmonary or other soft tissue tumor manifestations were included in the study, and histopathological slides from these tumors were reviewed. Results. Among a total of 297 registered osteosarcoma patients, six with synchronous (2.0%) and 10 with metachronous (3.4%) skeletal osteosarcomas were identified. All tumors were of high-grade malignancy. Treatment at the time of the first osteosarcoma diagnosis was in most cases wide resections and multi-agent chemotherapy according to international protocols, whereas the treatment for metachronous tumors was individualized and in general much less intensive. One patient was diagnosed with Li-Fraumeni syndrome, two other individuals may be suspected to have the same syndrome, and yet another patient had previously been treated for a bilateral retinoblastoma. Thirteen patients are dead, 11 from metastatic osteosarcoma, one from myelodysplastic syndrome, and one from wound infection and methotrexate-related nephrotoxicity; whereas three patients are still alive with no evidence of osteosarcoma. Conclusions. The prognosis for patients with synchronous and metachronous skeletal osteosarcoma is poor. However, because long-term survival is seen, aggressive treatment to selected cases, e.g., patients with an osteosarcoma predisposing syndrome and/or late occurring metachronous tumours, is justified. Revealing a possible clonal relationship between these tumors, e.g., by karyotyping, may be of interest for estimating prognosis and guide therapy intensiveness.
Osteosarcomas are malignant primary skeletal tumors most often occurring in adolescents and such neoplasms are slightly more common in males than in females Citation[1–3]. Metastatic spread is predominantly haematogenous, and the lungs are by far the most common site of metastatic disease Citation[3]. After the introduction of chemotherapy for osteosarcoma (OS), extrapulmonary metastases are seen more often Citation[4], and in earlier autopsy studies the frequency of such extrapulmonary OS manifestations was higher than observed clinically Citation[5]. Although imaging studies are far better today, it is still clear that we are not able to detect all micrometastatic disease foci Citation[6]. The frequency of multicentric bone involvement at some point of the OS disease is 30–45% in autopsy studies and 10–30% in clinical studies Citation[4], Citation[5], Citation[7], Citation[8].
In some cases two or more skeletal OS lesions are seen in the same patient, at the same time (synchronous, SOS) or separated in time (metachronous, MOS), without detectable lung or other non-skeletal disease foci Citation[9]. MOS are further divided into early (metachronous tumor developing less than 24 months from first OS diagnosis) and late (metachronous tumor developing more than 24 months from first OS diagnosis) groups Citation[7]. Such SOS and MOS are considered rare and are found at a frequency of about 1–3% of osteosarcoma cases Citation[10], Citation[11]. In general, patients with SOS or MOS face a less favorable prognosis than those with only one skeletal osteosarcoma lesion without visceral metastases, but they are not incurable Citation[7], Citation[12], Citation[13]. It is debated whether such multiple skeletal OS are one primary tumor with metastatic lesion(s), or if they represent separate and clonally unrelated malignancies Citation[9], Citation[12], Citation[14]. Theoretically, MOS can arise as a consequence of resistant or dormant, distant metastatic tumor cells that are not killed by the chemotherapy administered when the primary tumor is treated. Such tumor cells may for some reason re-enter the cell cycle and, at a later stage, become clinically manifest Citation[10], Citation[15]. The proposed estimate that as many as 80% of OS patients have clinically undetectable micrometastases in lungs or other organs at the time of the primary diagnosis Citation[12], Citation[15], Citation[16] lends some support to such a metastatic theory for metachronous tumors. Likewise, this estimate can be taken in favor of a view stating that the occurrence of SOS is likely to be a sign of metastatic disease. Other arguments and theories favoring a metastatic pathogenesis are summarized by Currall and Dixon Citation[9] and include one dominant lesion, bone-to-bone metastases via venous plexuses or intraosseous embolisation, lymphatic spread, clinical and/or radiological underestimation of lung metastases, and a similar chemotherapy response in the primary tumor and the synchronous/metachronous lesion. If this is the case, one would expect that most patients with such multiple skeletal lesions later will develop further overt metastatic disease. Conversely, different histological sub-typing of tumor pairs, different response to chemotherapy between synchronous/metachronous lesion and primary tumor, no detectable pulmonary metastases, and the presence of a syndrome predisposing to OS can be viewed as arguments supporting that the tumors are clonally independent Citation[12]. A third possibility is that some, but not all, SOS and MOS are clonally related. Knowledge on these matters may be important for prognostication and choice of therapy.
The sarcoma database of The Norwegian Radium Hospital, including about 70% of sarcoma cases in Norway Citation[17], was reviewed to identify patients with SOS and MOS. Clinical data and tumor histopathological characteristics for these patients, including survival, treatment intensity for metachronous tumors, and relevant co-morbidity, are reported and discussed. Furthermore, the number of OS patients with two synchronous skeletal OS foci or at least one metachronous skeletal OS lesion found concurrently with the first non-skeletal metastatic OS focus is also reported.
Patients and methods
All patients diagnosed with OS at our institution after January 1, 1980 are registered in the hospital sarcoma database Citation[17]. This database was used to identify patients with SOS or MOS. Relevant patient records for all cases identified were reviewed. Cases diagnosed between January 1, 1980 and January 1, 2008 were included in the study, as were patients with both single and multiple metachronous lesions. OS patients with skeletal lesions and pulmonary metastases diagnosed concurrently were excluded from further analysis in the present study. However, the number of such patients is also reported. Because the hospital sarcoma database does not include information on the clinical course of all patients following diagnosis of the first metastatic lesion, it was not possible to determine the total frequency of OS patients having skeletal metastases at some (later) point of their disease. The presence or absence of lung metastases was determined using CT scans for all patients, except patient 2 who underwent only a chest x-ray at the time of the diagnosis of the metachronous OS.
The term index OS is used for the first OS lesion diagnosed. Classification of tumors as either synchronous or metachronous based on the time interval between their detection is not entirely consistent in previous reports Citation[7], Citation[12]. Herein, tumors were classified as synchronous if both/all tumors were apparent on the initial bone scan of each patient. In those cases where OS lesions developed in areas with scintigraphic findings considered unspecific or clinically irrelevant at the time of the primary diagnosis, initial bone scans were re-evaluated by two experienced nuclear medicine specialists (T.V.B (author) and O.M.A (Acknowledgment)). They were asked to re-evaluate specific scintigraphic lesions questioned at the time of the primary diagnosis. The lesions under scrutiny were graded from 1 to 4 (1 – non-malignant/irrelevant; 2 – probably non-malignant/irrelevant; 3 – probably malignant; 4 – clearly malignant). Furthermore, histopathological characteristics of all tumors included in the study were re-examined and tumor diagnosis, subtype, and grade were re-evaluated by an experienced sarcoma pathologist (author B.B.) according to the WHO classification Citation[18]. Histological response to chemotherapy was evaluated in the surgical specimens according to the criteria used in the on-going EURAMOS1 protocol and classified as good (<10% viable tumor cells), poor (≥10% viable tumor cells), or not evaluable (see www.euramos.org). It seems that bone marrow micrometastatic disease has prognostic impact in OS patients Citation[15]. Hence, we report micrometastatic disease status for patients where such data were available. The study was approved by the regional ethical committee, the institutional study board, and informed consent was obtained from patients still alive.
Results
Patient and tumor characteristics for SOS and MOS
A total of 747 patients with a malignant bone tumor diagnosis were identified in the hospital sarcoma database and 297 of these (40%) were diagnosed with OS. Of the 297 OS cases, six had SOS (2.0%), whereas MOS developed in 10 patients (3.4%). Clinical and histopathological data for these 16 patients are summarized in and . Ten patients were male and six were female. Femur was the most frequent index tumor localization (four cases). Two patients had multiple synchronous tumors (>two, included index tumor) and two patients had multiple metachronous lesions (>one). The interval between the index and the metachronous OS varied from two to 59 months; seven of 10 metachronous tumors developed less than 24 months after the initial OS diagnosis and were therefore classified as early MOS. Five metachronous tumors were detected following radiological examinations initiated because patients experienced localized pain, two were found during treatment for the index tumor, and three were identified in routine imaging at follow-up. Patient 8 had radiotherapy for a uterine cervical cancer 14 years prior to the diagnosis of sacral OS, and this OS may well have been radiotherapy-induced Citation[19].
Table I. Clinical data and histopathological characteristics of tumors in patients with synchronous and metachronous osteosarcoma
One of the patients (11) was diagnosed with Li-Fraumeni syndrome (LFS) and had a germline TP53 mutation. Patient 3 was diagnosed with an anaplastic astrocytoma between the index and metachronous osteosarcoma, and this may raise the suspicion of LFS, although the family cancer history is unknown. Another patient (6) was surgically treated for a well-differentiated liposarcoma before the diagnosis of the metachronous OS. This patient ultimately succumbed to a myelodysplastic syndrome without signs of OS. Although we do not have knowledge on the family history of this patient, LFS may be suspected. Patient 5 was treated for bilateral retinoblastoma 13 years prior to the OS diagnosis and therefore probably had an RB1 mutation known to predispose to OS Citation[3].
OS patients with multiple skeletal OS lesions and visceral metastases
In the same time period as the above-mentioned cases were diagnosed at our hospital, the number of OS patients with more than one skeletal OS lesion and simultaneous non-skeletal metastases at primary diagnosis was five (1.7%). Furthermore, 10 patients (3.4%) were diagnosed with one or several OS lesions simultaneously with non-skeletal metastases at some point after completion of treatment for the primary OS.
Treatment
All patients except four received multi-agent chemotherapy for their index OS (). Treatment was administered according to the following protocols: SSGII/T-10 Citation[20], Citation[21], SSGVIII Citation[22], ISG/SSGI Citation[23], ISG/SSGII Citation[24], EURAMOS1 (on-going study of resectable OS in patients < 40 years) Citation[25], and EURO-B.O.S.S. (on-going study of bone sarcomas in patients aged 41–65) Citation[25]. All protocols include(d) high-dose methotrexate (HD-MTX), doxorubicin, and cisplatin, whereas ifosfamide was used in all protocols except SSGII/T-10. Patient 1 was scheduled for such multi-agent chemotherapy, but succumbed to a combination of methotrexate-related renal failure and postoperative wound infection following the first cycle of HD-MTX. Three patients did not receive chemotherapy at all: patients 5 and 8 because they were judged to have a poor prognosis due to multiple synchronous tumors, and patient 7 because of significant cardiovascular morbidity related to a congenital, inoperable ventricular septum defect. Most patients received multi-agent chemotherapy both in a neoadjuvant and in an adjuvant setting. Two patients (1 and 13), however, had primary surgery and therefore only adjuvant chemotherapy. Except patients 5 and 8 who had more than 10 skeletal lesions at primary diagnosis, all patients had surgery of their index OS. Patients 5 and 6 had palliative and adjuvant, respectively, radiotherapy for their index tumors.
Table II. Treatment data and survival for patients with synchronous and metachronous osteosarcoma
For the second synchronous tumor, the treatment was chemotherapy and surgery for three patients and no OS-specific therapy in three patients. The treatment for the metachronous tumors was individualized and in most patients much less intense compared to therapy given for the index OS. Patients 1 and 12 had metachronous tumors occurring so early that they were affected by chemotherapy administered for the index tumor. Only three patients received chemotherapy dedicated for their metachronous tumors; patients 4 and 11 are still alive whereas patient 10 died from lung metastases nine months after diagnosis of the metachronous tumor. In total for patients with metachronous tumors: one had surgery only, four had surgery combined with other treatment modalities, two had no surgery but other forms of therapy, and three (included 1 and 12) had no active osteosarcoma treatment at all.
Histopathology, bone scan re-evaluation, and bone marrow status at index OS diagnosis
A histopathological diagnosis was obtained and re-evaluated in all index tumors and in 13 of 16 metachronous/second synchronous tumors (). All tumors were classified as OS of high-grade malignancy. The histopathological sub-typing of the tumors was as follows: seven index tumors were classified as osteoblastic, two as chondroblastic, one as fibroblastic, one as telangiectatic, and five tumors as mixed variants. Patients 7, 8, and 12 did not have histopathological verification of their metachronous/second synchronous lesions, whereas for patients 1 and 6 the metachronous lesions were not possible to sub-type. In the other 11 metachronous/synchronous OS, the histological subtype was concordant with the index tumor in six cases and discordant in five cases.
Re-evaluation of bone scans was done in nine patients (). Two patients (6 and 14) were re-classified from metachronous to synchronous, whereas three patients (10, 12, and 15) were considered not to have initial bone scintigraphic lesions indicative of synchronous tumors. Furthermore, patients 5, 7, and 8 had initial bone scans suggestive of synchronous tumors that were never verified histologically. The re-evaluation of the scintigrams from these three patients confirmed that lesions highly suggestive of synchronous OS were present. Also, in patient 3, the re-evaluation of the scan taken 59 months from the primary diagnosis confirmed findings highly suggestive of OS.
Table III. Results of bone scan re-evaluation in nine patients with synchronous and metachronous osteosarcoma
Seven patients had bone marrow specimens sampled at the time of the index OS diagnosis (). Five of these (3, 7, 13, 14, and 16) had micrometastatic bone marrow disease, whereas patients 6 and 10 had bone marrow samples scored as negative. Four of the five patients with primary micrometastatic bone marrow disease are dead from OS, whereas one (13) is still alive with no evidence of OS. The two patients with no micrometastatic bone marrow disease are both dead, one of them (6) without evidence of OS.
Table IV. Micrometastatic bone marrow status at the time of the index osteosarcoma diagnosis
Patient outcome
At the time of writing, three patients are alive, all without signs of OS (4, 11, and 13). For most of the other 13 patients, the presumed cause of death was based on investigations performed when they were still alive. Metastatic disease including pulmonary metastases from OS was the presumed cause of death in 11 of these patients, myelodysplastic syndrome in one patient (6), and a combination of methotrexate-related nephrotoxicity and wound infection in one patient (1).
Discussion
Synchronous and metachronous skeletal osteosarcomas are rare entities which carry a poor prognosis, but in recent reports it has been emphasized that long-term survival is possible if the patient receives adequate treatment Citation[7], Citation[12]. The relative incidence of SOS and MOS of 2.0% and 3.4%, respectively, found in our material, is in the expected range based on earlier reports Citation[10], Citation[12]. Our results also show that the simultaneous detection of pulmonary and skeletal metastases at primary diagnosis or first metastatic diagnosis in patients with OS is relatively rare (5.1%). Nonetheless, not taking micrometastatic disease into consideration, the skeleton, second only to the lungs, seems to be the most frequent organ involved in metastatic disease in OS Citation[5].
In two of the patients with MOS, the metachronous tumor was found two months after the diagnosis of the index tumor, and the classification of these as MOS is not obvious. In fact, these two tumors would have been assigned to different groups based on different criteria in earlier publications Citation[7], Citation[12], Citation[26]. Furthermore, although the initial clinical evaluation of some uncertain scintigraphic findings concluded that they were not clinically relevant, the disease course nonetheless showed that some of these minor scintigraphic abnormalities turned out to be skeletal OS lesions. Also, in five of six patients with SOS, one of the OS lesions was larger and more illustrious than the other lesions. A possible interpretation of these findings is that the number of patients with metastatic disease at the time of the primary OS diagnosis was underestimated. If diagnosed today, MRI would have been performed in most of these cases to further explore such uncertain/unspecific scintigraphic findings, thereby reducing the uncertainty. There is no method to determine the frequency of possible underestimation of pulmonary metastases in the patients included in this study. Except at the time of diagnosis of the metachronous tumor of patient 2, all patients went through a chest CT looking for pulmonary metastases when any OS lesion was found. However, because many of the patients were diagnosed and treated several years ago, the sensitivity of their CT scans was not as high as it would have been today.
Histological sub-typing was concordant in six and discordant in five of the 11 evaluable tumor pairs. Although it is tempting to claim that this finding corresponds to a connection and no connection, respectively, it is no more than a weak indication. Metastases may look different than the primary tumor and two unrelated tumors may be histologically of the same sub-type. In the five patients where response to chemotherapy was evaluated for tumor pairs, four had a concordant response. The interpretation of these findings when it comes to a possible relationship between the tumor pairs is difficult; a different response to chemotherapy between two tumors does not prove that they are unrelated, because a metastasis may be genetically dissimilar to its primary tumor and therefore respond in a different manner.
The time interval between metachronous skeletal tumors is worth noticing. It has been reported that patients with early metachronous tumors (<24 months interval) have a poorer prognosis than those with late metachronous tumors (>24 months interval) Citation[7]. Two of the three patients with late metachronous tumors in our series are still alive four and a half (11) and nearly 25 (4) years after the diagnosis of their MOS, whereas none of the patients with early metachronous tumors are alive. A possible interpretation is that most of the early tumors are metastatic, whereas among the late tumors the proportion of clonally independent tumors is higher.
Micrometastatic bone marrow disease at the time of the primary osteosarcoma diagnosis has been shown to have a certain prognostic value Citation[15]. The fact that patient 13 had positive micrometastatic bone marrow status at the time of the primary diagnosis, lends some support to a theory suggesting that the SOS in this patient were clonally dependent. However, because sub-clinical metastatic disease is thought to be common in OS patients at the time of the primary diagnosis Citation[12], the mentioned theory remains unresolved. Furthermore, because this patient is still alive 30 months after the diagnosis of SOS, one could alternatively speculate that the two OS lesions were in fact clonally independent. Patient 6 had negative micrometastatic bone marrow status at the time of the primary OS diagnosis and succumbed 48 months later – without any evidence of OS. It is likely that the patient, given metastatic disease at primary diagnosis, would have had evidence of OS disease at the time of death. When held together, these findings may therefore indicate a clonal independence between the two OS lesions.
Early occurring metachronous tumors, rapid disease progression ultimately involving also lungs, multifocal (>one) metachronous skeletal lesions, unspecific primary bone scan findings later manifesting as OS lesions, and possibly also similar histological sub-typing may indicate metastatic OS. Most of our patients seem to fall into this category, and although their chest CT scans were normal, this by no means excludes the possibility of pulmonary metastases Citation[6]. The presence of an OS predisposing syndrome, absence of bone marrow micrometastases, and/or late occurring solitary metachronous tumors, such as in patients 3, 4, 6, and 11, may indicate unrelated primary tumors. Some synchronous tumors may also be unrelated, as might be the case in our patient 13. Nonetheless, the above-mentioned arguments provide, at best, only circumstantial evidence for the presence or not of a clonal relationship between index OS and synchronous/metachronous lesions. Cell culture and karyotypic analysis could, if performed and successful, have provided more information on possible clonal associations between index OS and synchronous/metachronous lesions. A problem is the karyotypic complexity often seen in these neoplasms Citation[27], but the finding of the same marker chromosome in two skeletal OS would strongly suggest their clonal relationship.
Two of the three patients that received chemotherapy for their metachronous tumor are alive several years from diagnosis of their metachronous OS, as is one of the patients with synchronous tumors. Although the herein reported material is small, this fact emphasizes, in accordance with earlier reports Citation[7], Citation[12], that some patients with MOS and SOS are curable, and that selected cases should be treated intensively. Our results may be interpreted in two different ways: the selection of patients for chemotherapy in the setting of a metachronous tumor was good, or, too few patients received chemotherapy for their metachronous tumor. Hence, even a few more patients could have become long-term survivors with more aggressive treatment.
Acknowledgements
We thank MD PhD Olav Magne Aas for his help with re-evaluation of bone scintigrams. Declaration of interest: The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.
Related Research Data
References
- Dorfman HD, Czerniak B. Bone cancers. Cancer 1995; 75: 203–10
- Kansara M, Thomas DM. Molecular pathogenesis of osteosarcoma. DNA Cell Biol 2007; 26: 1–18
- Malawer, MM, Helman, LJ, O'Sullivan, B. Sarcomas of bone. VT Devita, Hellman, S, Rosenberg, SA. Cancer: Principles and practice of oncology, 7th ed. Philadelphia: Lippincott-Raven publishers; 2005; 1638–86.
- Giuliano AE, Feig S, Eilber FR. Changing metastatic patterns of osteosarcoma. Cancer 1984; 54: 2160–4
- Jeffree GM, Price CH, Sissons HA. The metastatic patterns of osteosarcoma. Br J Cancer 1975; 32: 87–107
- Kayton ML, Huvos AG, Casher J, Abramson SJ, Rosen NS, Wexler LH, et al. Computed tomographic scan of the chest underestimates the number of metastatic lesions in osteosarcoma. J Pediatr Surg 2006; 41: 200–6
- Aung L, Gorlick R, Healey JH, Shi W, Thaler HT, Shorter NA, et al. Metachronous skeletal osteosarcoma in patients treated with adjuvant and neoadjuvant chemotherapy for nonmetastatic osteosarcoma. J Clin Oncol 2003; 21: 342–8
- Lockshin MD, Higgins IT. Bone metastasis in osteogenic sarcoma. Arch Intern Med 1966; 118: 203–4
- Currall VA, Dixon JH. Synchronous multifocal osteosarcoma: Case report and literature review. Sarcoma 2006; 3 Pages, Article ID 53901
- Rodriguez EK, Hornicek FJ, Gebhardt MC, Mankin HJ. Metachronous osteosarcoma: A report of five cases. Clin Orthop Relat Res 2003; 411: 227–35
- Unni, KK. Dahlin′s bone tumors. General aspects and data on 11.807 cases, 5th ed. PhiladelphiaPennsylvania: Lippincott Williams & Wilkins; 1996.
- Jaffe N, Pearson P, Yasko AW, Lin P, Herzog C, Raymond K. Single and multiple metachronous osteosarcoma tumors after therapy. Cancer 2003; 98: 2457–66
- Bacci G, Fabbri N, Balladelli A, Forni C, Palmerini E, Picci P. Treatment and prognosis for synchronous multifocal osteosarcoma in 42 patients. J Bone Joint Surg Br 2006; 88: 1071–5
- Fitzgerald RH, Jr, Dahlin DC, Sim FH. Multiple metachronous osteogenic sarcoma. Report of twelve cases with two long-term survivors. J Bone Joint Surg Am 1973; 55: 595–605
- Bruland OS, Hoifodt H, Saeter G, Smeland S, Fodstad O. Hematogenous micrometastases in osteosarcoma patients. Clin Cancer Res 2005; 11: 4666–73
- Bruland OS, Pihl A. On the current management of osteosarcoma. A critical evaluation and a proposal for a modified treatment strategy. Eur J Cancer 1997; 33: 1725–31
- Aksnes LH, Hall KS, Folleraas G, Stenwig AE, Bjerkehagen B, Taksdal I, et al. Management of high-grade bone sarcomas over two decades: The Norwegian Radium Hospital experience. Acta Oncol 2006; 45: 38–46
- Fletcher, CDM, Unni, KK, Mertens, F. World Health Organization classification of tumours. Pathology and genetics of tumours of soft tissue and bone. Lyon: IARC Press; 2002.
- Patel SR. Radiation-induced sarcoma. Curr Treat Options Oncol 2000; 1: 258–61
- Rosen G, Caparros B, Huvos AG, Kosloff C, Nirenberg A, Cacavio A, et al. Preoperative chemotherapy for osteogenic sarcoma: Selection of postoperative adjuvant chemotherapy based on the response of the primary tumor to preoperative chemotherapy. Cancer 1982; 49: 1221–30
- Saeter G, Alvegard TA, Elomaa I, Stenwig AE, Holmstrom T, Solheim OP. Treatment of osteosarcoma of the extremities with the T-10 protocol, with emphasis on the effects of preoperative chemotherapy with single-agent high-dose methotrexate: A Scandinavian Sarcoma Group study. J Clin Oncol 1991; 9: 1766–75
- Smeland S, Muller C, Alvegaard TA, Wiklund T, Wiebe T, Bjork O, et al. Scandinavian Sarcoma Group Osteosarcoma Study SSG VIII: Prognostic factors for outcome and the role of replacement salvage chemotherapy for poor histological responders. Eur J Cancer 2003; 39: 488–94
- Ferrari S, Smeland S, Mercuri M, Bertoni F, Longhi A, Ruggieri P, et al. Neoadjuvant chemotherapy with high-dose Ifosfamide, high-dose methotrexate, cisplatin, and doxorubicin for patients with localized osteosarcoma of the extremity: a joint study by the Italian and Scandinavian Sarcoma Groups. J Clin Oncol 2005; 23: 8845–52
- Brach del Prever A, Smeland S, Tienghi A, Hall KS, Aglietta M, Bernini G, et al. High-risk osteosarcoma (OS): Preliminary results of the ISG-SSG II protocol. J Clin Oncol 2005 ASCO annual meeting proceedings 2005; 23(16S)9002
- Carrle D, Bielack SS. Current strategies of chemotherapy in osteosarcoma. Int Orthop 2006; 30: 445–51
- Amstutz HC. Multiple osteogenic sarcomata–metastatic or multicentric? Report of two cases and review of literature. Cancer 1969; 24: 923–31
- Sandberg AA, Bridge JA. Updates on the cytogenetics and molecular genetics of bone and soft tissue tumors: Osteosarcoma and related tumors. Cancer Genet Cytogenet 2003; 145: 1–30