Malignant bone tumors in childhood are uncommon, with approximately 300 diagnosed in children under the age of 15 each year in the United States.1 Despite their rarity, children and adolescents with these tumors have a significant impact on the health professionals involved in their care. Prior to the past decade, the diagnosis of a malignant bone tumor usually meant a rapid demise, and in an attempt to salvage even a small proportion of these patients, amputations were usually performed.
Fortunately, significant advances have been made since the early 1970s in treating these tumors. Fifty percent or more of all children and adolscents with newly diagnosed osteosarcoma or Ewing's sarcoma will be long-term survivors with the potential to be cured of the malignancy. This improved survival is primarily due to more effective chemotherapy programs. Major advances in surgical techniques have allowed for "limb-salvage" procedures to be performed in selected cases- with good functional results.
The vast majority of primary malignant bone tumors of childhood are either osteosarcoma or Ewing's sarcoma. Other malignant bone tumors, such as fibrosarcoma, chondrosarcoma, and reticulum cell sarcoma are very rare in childhood and will not be discussed. Several other pediatric malignancies, especially neuroblastoma, have a predilection for spread to bone and are considered elsewhere in this issue of Pediatric Annals.
Osteosarcoma is the most common primary malignant bone tumor of the pediatric and adolescent age groups.1 It is a malignancy predominantly of adolescents and young adults with approximately 50% of cases occurring in the second decade.2'3 Approximately three per one million children under the age of 15 years in the US are newly diagnosed each year.1 The incidence is highest in the adolescent age group, with approximately seven new cases perone million adolescents per year.4 Osteosarcoma is slightly more common in males than in females, with a male to female ratio of 1.5:1.
Primary osteosarcoma may arise within any bone of the skeleton; but 90% arise within the long, tubular bones, with the extremities being most commonly involved. The primary tumor usually arises in the metaphyseal ends of the long bones. Nearly 50% of all tumors involve the region around the knee, that is, the distal femur and proximal tibia, with more than one-third of cases involving the distal femur. Other common sites include the proximal humerus, ilium, and proximal femur. Primary tumors may also arise in the facial bones, skull, spine, and ribs, and very rarely at other sites.2'3
Osteosarcoma is seen most frequently during the adolescent growth spurt and is found in the most rapidly growing end of the most rapidly growing bone, namely, the distal femur.6 Prior to the age of 13 years, the incidence of osteosarcoma is similar for boys and girls. At puberty, however, as boys start growing more rapidly as well as taller than girls, the incidence of osteosarcoma is higher in boys than girls.1
Osteosarcoma may be induced by ionizing radiation, as occurs in radium-dial painters,9 and it has also been induced within the field of radiation exposure following radiotherapy for other childhood malignancies.10
Children with the familial form of retinoblastoma are unusually susceptible to the development of osteosarcoma. These second malignancies may be seen within the field of radiation exposure, at a distant site, and even in patients who have received no prior radiotherapy.11,12
Several pre-existent skeletal defects such as bone cysts, osteocartilaginous exostoses, osteochondromata, fibrous dysplasia, and giant cell tumors may infrequently predispose to the development of osteosarcoma in children. Adults with Paget's disease (osteitis deformans) also have an increased risk.
The usual presenting complaint is pain, which frequently follows slight trauma to the area. Trauma, however, has not been implicated as causative of osteosarcoma; but the resulting pain calls attention to the abnormal bone and the pain is generally persistent and progressive in severity. The presence of a mass is commonly a later finding. A radiographic examination performed at the onset of symptoms may be negative; whereas a bone scan is likely to be positive. As radiographic changes may not be seen for several months, very close monitoring is warranted in any child or adolescent with suspicious symptomatology. At initial presentation, most patients do not have constitutional symptoms and likewise, the physical examination is unremarkable except for the findings at the primary site.
A definitive diagnosis of osteosarcoma must be made by biopsy, although the clinical presentation and radiographic findings often are strongly suggestive of the diagnosis. The serum alkaline phosphatase is frequently elevated and may be useful in monitoring the response to preoperative chemotherapy.13 Following definitive surgery, a rising alkaline phosphatase is strongly indicative of the imminent development of metastatic disease and tissue levels of alkaline phosphatase obtained from the primary tumor biopsy may correlate with prognosis.14 Other routine laboratory studies, including the complete blood count, erythrocyte sedimentation rate, and serum biochemical determinations are generally within normal limits.
In addition to the routine anteroposterior and lateral radiographs of a suspected lesion, conventional tomograms and magnification radiographs may be of value in assessing the primary lesion. Osteosarcoma is classically a mixed lytic and sclerotic lesion located in the metaphysis of a long bone with associated cortical bone destruction. Pathologic fractures may be seen and an associated soft tissue mass is common. The extracortical soft tissue extension of a highly ossified osteosarcoma may display transverse or radiating striations emanating from the involved cortex in a characteristic "sunburst" perpendicular configuration. Such striations or spicules represent osteoid produced by extracortical tumor tissue. A lamellated "onionskin" periosteal reaction, or reactive bone, may also be seen (Figure I).15
Figure 1. Typical radiographic appearance of an osteosarcoma of the right distal femur in an 11-year-old boy. There is elevation of the periosteum, the typical "sunburst" soft tissue calcification, and an associated soft tissue mass (Courtesy of John H. Miller, M.D.).
Computed tomography (CT scan) and angiographic evaluation of the primary tumor may be of value in assessing the extent of tumor involvement both within the medullary canal of the bone and of any associated soft tissue mass (Figure 2). These studies will also aid in demonstrating involvement of the neurovascular bundle. Likewise, the CT scan can be useful in determining the presence of "skip" lesions, that is, metastases within the medullary canal of the involved bone. This is critical information for the surgeon if a radical resection is contemplated, since clear surgical margins and a functional extremity are mandatory if such a procedure is to be undertaken.
Figure 2. A computed tomographic scan of the distal femurs from the same patient described in Figure 1. The right distal femur demonstrates sclerotic bone, soft tissue calcification, and a large soft tissue mass. This study was obtained following an open biopsy (Courtesy of John H. Miller, M. D.).
A radionuclide bone scan, using a technetium-99m phosphonate compound, will show a marked increase in osteoblastic activity at the site of the primary tumor (Figure 3). This method will also demonstrate macroscopic bony metastases within the primary involved bone, or in other bones, and may do so prior to the appearance of any abnormality on a radiographic skeletal survey (Figure 4). The bone scan should be performed prior to the diagnostic biopsy, since a surgical incision will cause changes at the biopsy site. A baseline bone scan may be useful in assessing response to therapy if chemotherapy is given prior to a definitive surgical procedure.
A posteroanterior and lateral radiograph of the chest should be performed to detect the presence of metastatic disease. High resolution rapid CT scanning of the chest at l.O cm intervals may be superior to a regular radiograph or conventional tomographic evaluation of the chest in detecting pulmonary metastases (Figure 5). Thus, a CT scan of the chest should be performed in all patients with newly diagnosed osteosarcoma.16
The diagnostic biopsy is generally obtained through a surgical incision, although a needle biopsy or aspiration biopsy is being used successfully in several centers. Surgical biopsy permits exposure of the tumor with a better chance of obtaining adequate tissue for definitive diagnosis. Since radiotherapy oran en bloc resection may be part of the treatment regimen, depending upon the histologic diagnosis, the biopsy site should take these possible subsequent therapeutic modalities into account. Preferably, the biopsy will be obtained by the same surgeon who will participate in the definitive management of the patient following diagnosis and after consultation with the pediatric oncologist and radiotherapist.
The location of the biopsy should be such that the entire site can be included within the surgical specimen if an en bloc resection is to be considered later. The most important axiom in making the incision is that it should be longitudinal rather than transverse, in case a subsequent limb salvage procedure is performed. The biopsy site should also be located so that it does not compromise the possible later administration of radiotherapy.
Figure 3. A technetium-99m phosphonate bone scan from the same patient described in Figure 1 demonstrating markedly increased uptake in the right distal femur (Courtesy of John H. Miller. M. D.).
The surgical incision can generally be less than 5 cm to 8 cm in length, but should be sufficient so that an adequate sample is obtained. The pathologist can assess the need for further tissue on frozen section examination. If the diagnosis can be made on a biopsy of a soft tissue mass, that is preferable to going through an already compromised bone where the risk of a pathologic fracture will be significantly increased. Specimens should be obtained for routine histological examination, appropriate cultures, and for electron microscopy should this be necessary for diagnosis later.
The gross appearance of the primary tumor generally corresponds to the roentgenographic appearance. Usually, osteosarcoma is a bulky tumor extending beyond the bone to produce a soft tissue mass that may encircle the bone. The intramedullary portion is usually more sclerotic than the soft tissue component and the epiphyseal plate often limits the growth toward the joint. Some tumors spread along the marrow cavity, but usually such spread is contiguous and is apparent grossly (Figure 6). "Skip" areas of medullary involvement are an uncommon finding.5 The histologic appearance of the conventional osteosarcoma is highly variable. Most of these lesions have a predominantly spindle cell stroma and the cells, in their histology, appear to be highly malignant. Mitotic figures, including abnormal ones, are easily found, the nuclei are large and irregular, and malignant giant cells may be present. The normal trabeculae of the involved bone are usually destroyed by the tumor and tumor cells may permeate among and envelop the residual bony trabeculae, especially at the margins.5
Conventional osteosarcomas may be subdivided into three main histologic categories, depending on the predominant differentiation of the tumor cells. Approximately 50% of osteosarcomas produce osteoid in sufficiently large amounts to be called "osteoblastic osteosarcomas." The osteoid is a pink hyaline material that may or may not be calcified. Usually, it is very fine, being present between individual tumor cells to produce a "lace-like" pattern. Rarely, however, osteoid may take the form of larger trabeculae. In about one-fourth of osteosarcomas, the predominant differentiation is toward cartilage. Nevertheless, a tumor that is predominantly cartilaginous but shows foci of anaplastic cells that produce osteoid should be considered an osteosarcoma.
The remaining one-fourth of conventional osteosarcomas show a spindle cell stroma, with a herringbone pattern similar to that seen in fibrosarcoma and osteoid is present in only minimal amounts as fine pink material between tumor cells. Such tumors are classified as fibroblastic osteosarcoma. It may be difficult to differentiate some of these osteosarcomas from "pure" fibrosarcomas.5 Other types of osteosarcoma are uncommon in childhood.5
Osteosarcoma is highly malignant and rapidly disseminates hematogenously, with pulmonary metastases generally being the first site of recurrent disease. Prior to recent advances in therapy, metastatic disease occurred within six months to nine months from diagnosis with death ensuing six months later.15 Following the development of pulmonary metastases, dissemination to other bones and visceral organs frequently occurs. With more effective chemotherapy, the incidence of metastases has decreased and those patients who still do develop pulmonary metastases tend to have fewer with a later onset.
Surgery: Primary Tumor Surgical ablation of the primary tumor has been a prerequisite for potential cure in osteosarcoma and until recently, the procedure of choice in patients with extremity lesions was amputation. Enneking and Kagen observed secondary smaller foci of osteosarcoma anatomically separate from the primary lesion, but in the same extremity, in 10 out of 40 patients, and it was therefore concluded that local recurrences would occur unless disarticulations were performed.17 Subsequently, however, the expected high rate of local recurrence has not been observed by others, even when cross-bone amputations have been performed.18'19
During the past decade, several teams of investigators have attempted to avoid amputation in a select group of patients with osteosarcoma of the extremity. Instead, en bloc resection of the involved bone and soft tissue with a prosthetic replacement and functional restoration has been utilized. The principles of good cancer surgery must still be maintained, ie, surgical ablation of the primary tumor must still be complete, yielding gross and microscopic margins completely free of tumor.13'20-23 Chemotherapy cannot be relied upon to eradicate disease in the region of the primary tumor, although it may make an inoperable lesion operable."'20
Limb-salvage procedures require highly specialized surgical techniques. When properly performed in a select group of patients, it may be of significant benefit in the patient's long-term rehabilitation with respect to function of the involved extremity and the patient's psychologic well-being. These procedures have been most widely utilized for primary osteosarcoma of the femur. Indications for such a procedure at this site include: 1) complete or nearly complete linear growth; 2) a small tumor with minimal involvement of soft tissue; 3) lack of involvement of the neurovascular bundle on angiography; 4) a properly placed initial biopsy incision; and 5) absence of demonstrable metastases. 3'"' Although the results are preliminary using this surgical approach, they do appear to be very encouraging. The functional capacity of the extremity following the procedure is generally good and complications have been minimized with increasing surgical experience and the use of better prosthetic devices.
Figure 4. A) Radiograph of the pelvis and the left proximal femur demonstrating normal bony architecture. The right leg is surgically absent, having been removed for an osteosarcoma of the distal femur. B) Technetium-99m phosphonate bone scan from the same patient. There is markedly increased uptake in the neck of the femur, whereas the radiograph is unremarkable. This was later confirmed to represent metastatic osteosarcoma (Courtesy of John H. Miller. M. D.).
Figure 5. A CT scan of the chest in a patient with osteosarcoma demonstrating a solitary pulmonary metastasis of the left lung (arrow). Fibrotic changes from a prior surgical procedure are seen in the right hemothorax. A radiograph of the chest obtained at the same time did not reveal the pulmonary metastasis (Courtesy of John H. Miller, M. D.).
For lesions of the upper extremity, resections are clearly advantageous over amputations, since no prosthetic device can duplicate the function of the human hand. For other sites, the value of the procedure is less certain and may result in an extremity that is less functional than might be the case following an amputation. An example of this may be a tumor of the proximal tibia, where an above-the-knee amputation and a good prosthesis results in a very functional lower extremity. In certain expendable bones, such as the fibula, ulna or clavicle, en bloc resection alone may suffice.
Figure 6. Amputation specimen of an osteosarcoma of the distal femur demonstrating intramedullary and soft tissue involvement (Courtesy of Hart Isaacs, M.D.).
Figure 7. Typical osteoblastic osteosarcoma, demonstrating tumor cells surrounded by abundant osteoid (Hematoxylin and eosin stain, magnification ? 450) (Courtesy of Hart Isaacs, M.D.).
Surgery: Thoracotomy Although metastases are infrequent at initial presentation, the vast majority of patients who develop metastatic disease do so in the lungs.3 Many of those patients who previously would have died are now being salvaged and may be long-term survivors as a result of an aggressive surgical approach. CT scans are performed at regular intervals during the first two years following diagnosis in order to detect small lesions (Figure 5). When pulmonary metastases are found, aggressive use of multiple or bilateral thoracotomies have resulted in prolonged survival and potential cure of 25% to 50% of those patients who have operable pulmonary metastases.24"27 The role of chemotherapy is less certain, since many of the long-term survivors had no chemotherapy following the development of pulmonary metastases or no chemotherapy since their last thoracotomy. In those in whom chemotherapy was utilized, the agents and regimens used varied widely.
Chemotherapy Prior to the use of chemotherapy, several large reviews reported a cure rate of approximately 15% to 20% in osteosarcoma.2'3'28'29 In 1971, Marcove and associates reported on 145 consecutive cases of operable osteosarcoma of the extremities in patients under the age of 21 years who did not have metastases at the time of diagnosis.29 The five-year disease-free survival in these patients was 17.4%, with most recurrences within two years of diagnosis. AU of these reviews were retrospective and were subject to variability in the initial evaluation performed, the interpretation of the pathological material, and in the treatment administered. In addition, these reviews encompassed a time period prior to the use of radionuclide bone scanning and CTscanning of the chest. Thus, metastatic disease present at diagnosis may have been missed in many patients.
Prior to the early 1970s, numerous single agents and combination chemotherapeutic regimens were used to treat patients with metastatic osteosarcoma and all of these chemotherapy programs were found to be ineffective. Since that time, regression of metastatic osteosarcoma has been demonstrated following the use of single agents, such as high-dose methotrexate with citrovorum factor rescue,30 adriamycin,31 and cisplatinum,32 as well as following the use of the combination of bleomycin, cyclophosphamide, and dactinomycin (BCD).33
In the past, more than 80% of patients who underwent "curative" amputation subsequently developed pulmonary metastases. Thus, it was postulated that micrometastases were present at diagnosis in the majority of patients, and that adjuvant chemotherapy might be effective in preventing the development of overt metastatic disease. Based upon this background, adjuvant chemotherapy in the treatment of patients with surgically ablated osteosarcoma was begun in the early 1970s.
Studies using various agents, either singly or in combination as adjuvant therapy for patients immediately following their primary surgical treatment, were instituted. Since the results with surgical ablation alone were very disappointing, none of these studies used a concurrent control arm. The agents most commonly used included adriamycin, varying doses of high-dose methotrexate with citrovorum factor rescue, and vincristine. Other agents, such as cis-platinum and cyclophosphamide, have been incorporated into several of the chemotherapy regimens. For those studies wit h sufficient followup at this time, the two-year disease-free survival is approximately 40% to 50%, regardless of the adjuvant chemotherapy program utilized.34"38
As these studies were ongoing, a retrospective review of their experience and a prospective controlled study were being conducted at the Mayo Clinic. In their review, Taylor and co-workers found that the prognosis of patients less than 21 years of age at diagnosis, who were treated surgically for osteosarcomas arising in the extremities, improved steadily between 1963 and 1974.39 The two-year disease-free survival of about 20% in the early part of their study period was typical ofthat of other reports. However, between 1969 and mid-1974, the threeyear disease-free survival had increased to approximately 35%. This improvement was not attributable to adjuvant therapy nor any basic change in the type of patient treated.
A prospective randomized study was performed in which postoperative observation alone was compared to an adjuvant chemotherapeutic regimen consisting of vincristine and high-dose methotrexate with citrovorum factor rescue. In this study, Edmonson and colleagues observed no difference in survival between the two treatment groups. In a preliminary report, approximately 52% of each group were expected to be continuously free of disease at two years.40
Thus, the improved results seen with adjuvant chemotherapy, as summarized above, may not have been due to the therapy itself, but to the selection of a group of patients with a better prognosis. During this period of apparent improved rate of relapse-free survival, more sophisticated diagnostic techniques, such as bone scanning and CT scanning, may have selected out a group of patients with metastatic disease who had previously been missed. In addition, it is statistically improbable for a wide spectrum of adjuvant chemotherapeutic regimens to give similar results despite the use of varying agents, dosages, and administration schedules.
Therefore, since the only controlled study was that performed by the Mayo Clinic,40 it is difficult to assess the role of adjuvant chemotherapy in the treatment of osteosarcoma. The only possible exception to this statement is the ongoing study at the Memorial SloanKettering Cancer Center. In their study, Rosen and coworkers demonstrated the importance of treatment with preoperative high-dose methotrexate with citrovorum factor rescue.41 They also showed the prognostic importance of the histologic response of the tumor during such treatment to ultimate outcome.
According to Rosen's protocol, prior to performing definitive surgery, chemotherapy is administered for a period of four weeks to 16 weeks. The definitive surgical procedure is then performed and multiple histologic sections are examined from the surgical specimen. The effect of preoperative chemotherapy on the primary tumor is noted and a value is assigned according to the extent of tumor destruction attributable to preoperative chemotherapy. The histologic grading of the effect of preoperative chemotherapy on primary osteosarcoma ranges from a grade I response, with little or no effect of chemotherapy noted, to a grade IV response with no viable-appearing tumor cells noted in any of the histologic sections. The histologic responders who were continued on a regimen of high-dose methotrexate with citrovorum factor rescue, adriamycin and BCD (bleomycin, cyclophosphamide and dactinomycin) postoperatively have done well and 98% of the patients remain continuously free of disease with a median time on study of 31 months.41
Patients with a poor histologic response who were continued on the same postoperative regimen did poorly, with 67% relapsing at a median time on study of 13 months. Finally, patients with a poor histologic response treated with cis-platinum, adriamycin and BCD did well with a 93% rate of disease-free survival at a median time on study of 18 months.41
At the present time, two national cooperative group studies have been proposed in order to answer several of the important questions in the management of patients with osteosarcoma. In one study, children and adolescents will be treated in a fashion similar to that being done at the Memorial Sloan-Kettering Cancer Center, in an attempt to duplicate the results of a single institution study in a cooperative group. This non-randomized study will utilize preoperative chemotherapy in an effort to determine the optimal postoperative treatment for the individual patient by observing the histologic response of the primary tumor.
In the other study, children will be randomized to receive either no adjuvant chemotherapy postoperatively or a chemotherapy program based upon the Memorial Sloan-Kettering Cancer Center regimen. The chemotherapy regimen will include high-dose methotrexate with citrovorum factor rescue, bleomycin, cyclophosphamide, dactinomycin, adriamycin and cis-platinum. One of the objectives of this study is to determine if intensive multiagent chemotherapy, given in an adjuvant fashion following surgical ablation of the primary tumor, will significantly improve the disease-free survival for patients with non-metastatic osteosarcoma of the extremity when compared to a concurrent control group.
1. Young JLJr. Heise HW.Silverberg E, et al: Cancer incidence. survival and mortality for children under 15 years of age. New York, American Cancer Society. Professional Education Publication No. 3022-PE, 1978.
2. Dahlin DC, Coventry MB: Osteogenic sarcoma; a study of 600 cases. J Bone Joint Surg 1967; 49A:I0l-l 10.
3. McKenna RJ, Schwinn CP, Soong KY, et al: Sarcomata of the osteogenic series (osteosarcoma, fibrosarcoma, chondrosarcoma, parosteal osteogenic sarcoma and sarcomata arising in abnormal bone). J Bone Joint Surg 1966; 48A:l-26.
4. Glass AG, Fraumeni JF Jr: Epidemiology of bone cancer in children. Journal of the National Cancer Institute 1970; 44:187-199.
5. Dahlin DC, Unni KK; Osteosarcoma of bone and its important recognizable varieties. Am J Surg Pathol 1977; 1:61-72.
6. Rosen G: Management of malignant bone tumors in children and adolescents. Pediatr Clin North Am 1976; 23:183-213.
7. Fraumeni JF Jr: Stature and malignant tumors of bone in childhood and adolescence. Cancer 1967; 20:967-973.
8. Tjalma RA: Canine bone sarcoma: estimation of relative risk as a function of body size. Journal of the National Cancer Institute 1966; 36: 1 1 37-1 150.
9. Aub JC, Evans RD, Hempelmann LH. et al: The late effects of internallydeposited radioactive materials in man. Medicine 1952; 31:221-329.
10. Li FP, Cassady J R. Jaffe N : Risk of second tumors in survivors of childhood cancer. Cancer 1975; 35:1230-1235.
11. Jensen RD, Miller RW: Retinoblastoma; epidemiologic characteristics. N Engl J Med 197 1 ; 285:307-3 1 1 .
12. Schimke RN, Lowman JT, Cowan GAB: Retinoblastoma and osteogenic sarcoma in siblings. Cancer 1974; 34:2077-2079.
13. Rosen G, Marcove RC, Caparros B, et al: Primary osteogenic sarcoma; the rationale for preoperative chemotherapy and delayed surgery. Cancer 1979; 43:2163-2177.
14. Levine AM, Rosenberg SA: Alkaline phosphatase levels in osteosarcoma tissue are related to prognosis. Cancer 1979; 44:2291-2293.
15. H uvos AG: Bone Tumors: Diagnosis. Treatment and Prognosis. Philadelphia. WB Saunders Co. 1979.
16. Muhm JR. Brown LR, Crowe JK, et al: Comparison of whole lung tomography and computed tomography for detecting pulmonary nodules. AJR 1978; 131:981-984.
17. Enncking WF, Kagan A: "Skip" metastases in osteosarcoma. Cancer 1975; 36:2192-2205.
18. Douglass HO Jr. Wang J. Takita H. et al: Improvement in the results of treatment of osteogenic sarcoma. Surg Gynecol Obstet 1975; 140:693-700.
19. Pratt C, Shanks E, Hustu O, et al: Adjuvant multiple drug chemotherapy for osteosarcoma of the extremity. Cancer 1977; 39:51-57.
20. Marcove RC. Rosen G: En bloc resections for osteogenic sarcoma. Cancer 1980; 45:3040-3044.
21. Jaffe N, Watts H, Fellows KE, et al: Local en bloc resection for limb preservation. Cancer Treat Rep 1978; 62:217-223.
22. Sim FH, Ivins JC, Pritchard DJ: Surgical treatment of osteogenic sarcoma at the Mayo Clinic. Cancer Treat Rep 1978; 62:205-211.
23. Weisenburger TH, Eilber FR, Grant TT. et al: Multidisciplinary "limb salvage" treatment of soft tissue and skeletal sarcomas. IntJ Radial Oncol Biol Phys 1981; 7:1495-1499.
24. Rosen G, H uvos AG, MosendeC.et al: Chemotherapy and thoracotomy for metastatic osteogenic sarcoma; a model for adjuvant chemotherapy and the rationale for the timing of thoracic surgery. Cancer 1978; 41:841-849.
25. Han M. Telander RL. Pairolero PC, et al: Aggressive thoracotomy for pulmonary metastatic osteogenic sarcoma in children and young adolescents. J Pediatr Surg 1981; 16:928-933.
26. Rosenberg SA, Flye MW, Conkle D, et al: Treatment of osteogenic sarcoma. IL Aggressive resection of pulmonary metastases. Cancer Treat Rep 1979; 63:753-756.
27. Burgers JMV, Breur K, van Dobbenburgh OA. et al: Role of meta statecto my without chemotherapy in the management of osteosarcoma in children. Cancer 1980; 45:1664-1668.
28. Friedman MA. Carter SK: The therapy of osteogenic sarcoma; current status and thoughts for the future. J Surg Oncol 1972; 4:482-510.
29. Marcove RC. Mike V. Hajek JV, et al: Osteogenic sarcoma in childhood. New York State J Med 1971; 71:855-859.
30. Jaffe N, Traggis D, Geiser C, et al: Favorable response of metastatic osteogenic sarcoma to pulse high-dose methotrexate with citrovorum rescue and radiation therapy. Cancer 1973; 31:1367-1373.
31. Cortes EP, Holland JF. Wang JJ, et al: Adriamycin (NSC- 1 23 127) in 87 patients with osteosarcoma. Cancer Chemotherapy Report 1975:6:305-313.
32. Ochs JJ, Freeman AI, Douglass HO Jr, et al: Cis-dichlorodiammineplatinum (II) in advanced osteogenic sarcoma. Cancer Treat Rep 1978; 62:239-245.
33. Mosende C, Gutierrez M, Caparros B, et al: Combination chemotherapy with bleomycin, cyclophosphamide and dactinomycin for the treatment of osteogenic sarcoma. Cancer 1977; 40:2779-2786.
34. Rosenberg SA, Chabner BA, Young RC. et al: Treatment of osteogenic sarcoma. 1. Effect of adjuvant high-dose methotrexate after amputation. Cancer Treat Rep 1979; 63:739-751.
35. Jaffe W, Frei E HL Watts H, et al: High-dose methotrexate in osteogenic sarcoma; a 5-year experience. Cancer Treat Rep 1978; 62:259-264.
36. Ettinger LJ, Douglass HOJr, Higby DJ.etal: Adjuvant adriamycin and cisdiamminedichloroplatinum (cis-platinum) in primary osteosarcoma. Cancer 1981;47:248-254.
37. Cortes EP, Holland JF. Glidewell O: Amputation and adriamycin in primary osteosarcoma; a 5-year report. Cancer Treat Rep 1978;62:271-277.
38. Sutow WW, Gehan EA. Dyment PG. et al: Multidrug adjuvant chemotherapy for osteosarcoma: Interim report of the Southwest Oncology Group Studies. Cancer Treat Rep 1978; 62:265-269.
39. Taylor WF. Ivins JC. Dahlin DC. et al; Trends and variability in survival from osteosarcoma. Mayo Clin Proc 1978; 53:695-700.
40. Edmonson JH, Green SJ, Ivins JC, et al: Methotrexate as adjuvant treatment for primary osteosarcoma, letter to the editor. ?' Engl J Med 1 980; 303:642-643.
41. Rosen G, Caparros B. Huvos AG. 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; 49:1221-1230.