Pediatric Annals

Osteosarcoma

Cameron K Tebbi, MD; John Gaeta, MD

Abstract

Primary bone tumors account for 10% of all malignancies of adolescents.1 In the Third National Cancer Survey, the incidence of bone tumors, primarily osteosarcoma and Ewing's sarcoma, for blacks and whites under 20 years old was 4-8 and 5.6 per million, respectively.2 Although bone tumors are seen in all age groups, they are most common in those 15 to 19 years of age. ' The death rate from bone tumors among adolescents is about four times greater than in children, being 11 cases per million population. This peak rate of death from bone tumors parallels the timing of the adolescent growth spurt.

INCIDENCE AND ETIOLOGY

Osteosarcoma is one of the most common malignancies of adolescence,4 exceeded only by leukemia, brain tumors, and lymphomas. By definition, this tumor is a malignant sarcoma of bone that originates from osteon. The tumor is more common in males, with a male to female ratio of 1.75:1. 5'7 Prior to the adolescent years, the frequencies in boys and girls are equal, whereas among teenagers and adults the incidence in males increases. The most common sites of osteosarcoma in order of frequency are femur, tibia, humerus, fibula, scapula, ilium, radius, rib, mandible, and clavicle.816 Extraskeletal osteosarcomas are rarely seen in children.1720

The etiology of osteosarcoma is not clear. Several possible etiologic factors have been suggested, but remain largely unproven. These include accelerated rate of growth,21 presence of congenital malformations and genetic conditions,2237 trauma,'8 and preexisting diseases, 3y as well as environmental factors such as carcinogens,40'42 viral agents,43'46 exposure to radiation and radioactive agents.47'57 In one study, children with cancer who were treated with radiation had a 2.7-fold higher incidence of these tumors than those who had not been irradiated.57 A dose-response correlation was identified with patients who received over 6000 cGy to the bones having a 40-fold higher risk of osteosarcoma.57

The incidence of osteosarcomas peaks during the adolescent growth spurt. Also, these tumors occur in bones undergoing an earlier growth spurt, such as the humerus, at an earlier age than the later-maturing tibia and femur. This time relationship suggests an etiological correlation between rapid growth and tumor formation. The reported greater mean height for patients with osteosarcoma, as compared with the mean height of patients with nonosseous malignancies, is further circumstantial evidence for such a relationship.11,21 However, measurements of levels of growth hormone in osteosarcoma are inconsistent. Growth hormone was found to he elevated hy some,58 but not by others. 59 Abnormal glucose tolerance tests are common. Further evidence for the role of growth rate in etiology of this tumor is the finding of a much higher incidence of bone tumors among larger hreeds of dogs.60'62

Various congenital and genetic conditions are associated with osteosarcoma. Osteosarcomas occur in association with Paget's disease, *9 bilateral retinoblastoma,22·57 polyostotic fibrous dysplasia,63'65 osteogenesis imperfecta,27 Waardenburg-like syndrome,32 Other's syndrome,26 multiple hereditary exostosis,28 solitary enchondroma,29 myositis ossificans,25 and osteopoikilosis.11 Among these conditions, the association of osteosarcoma and retinoblastoma is hest known. The occurrence of osteosarcoma in patients with a history of retinoblastoma is probably not solely due to radiation therapy given to the original tumor. In one report, 40% of patients with retinoblastoma treated with bilateral enucleation subsequently developed osteosarcoma. 59 There is also an increased incidence of osteosarcomas in cases of unilateral retinoblastoma.57 Patients with retinoblastoma treated with radiation therapy are at risk for developing orbital osteosarcoma; this risk is greater in cases of bilateral retinoblastoma.

Children with cancer who are treated with radiation and alkylating agents have an incidence of osteosarcoma 133 times that of the general population.57 The risk of developing osteosarcoma in Paget's disease is over 10%;…

Primary bone tumors account for 10% of all malignancies of adolescents.1 In the Third National Cancer Survey, the incidence of bone tumors, primarily osteosarcoma and Ewing's sarcoma, for blacks and whites under 20 years old was 4-8 and 5.6 per million, respectively.2 Although bone tumors are seen in all age groups, they are most common in those 15 to 19 years of age. ' The death rate from bone tumors among adolescents is about four times greater than in children, being 11 cases per million population. This peak rate of death from bone tumors parallels the timing of the adolescent growth spurt.

INCIDENCE AND ETIOLOGY

Osteosarcoma is one of the most common malignancies of adolescence,4 exceeded only by leukemia, brain tumors, and lymphomas. By definition, this tumor is a malignant sarcoma of bone that originates from osteon. The tumor is more common in males, with a male to female ratio of 1.75:1. 5'7 Prior to the adolescent years, the frequencies in boys and girls are equal, whereas among teenagers and adults the incidence in males increases. The most common sites of osteosarcoma in order of frequency are femur, tibia, humerus, fibula, scapula, ilium, radius, rib, mandible, and clavicle.816 Extraskeletal osteosarcomas are rarely seen in children.1720

The etiology of osteosarcoma is not clear. Several possible etiologic factors have been suggested, but remain largely unproven. These include accelerated rate of growth,21 presence of congenital malformations and genetic conditions,2237 trauma,'8 and preexisting diseases, 3y as well as environmental factors such as carcinogens,40'42 viral agents,43'46 exposure to radiation and radioactive agents.47'57 In one study, children with cancer who were treated with radiation had a 2.7-fold higher incidence of these tumors than those who had not been irradiated.57 A dose-response correlation was identified with patients who received over 6000 cGy to the bones having a 40-fold higher risk of osteosarcoma.57

The incidence of osteosarcomas peaks during the adolescent growth spurt. Also, these tumors occur in bones undergoing an earlier growth spurt, such as the humerus, at an earlier age than the later-maturing tibia and femur. This time relationship suggests an etiological correlation between rapid growth and tumor formation. The reported greater mean height for patients with osteosarcoma, as compared with the mean height of patients with nonosseous malignancies, is further circumstantial evidence for such a relationship.11,21 However, measurements of levels of growth hormone in osteosarcoma are inconsistent. Growth hormone was found to he elevated hy some,58 but not by others. 59 Abnormal glucose tolerance tests are common. Further evidence for the role of growth rate in etiology of this tumor is the finding of a much higher incidence of bone tumors among larger hreeds of dogs.60'62

Various congenital and genetic conditions are associated with osteosarcoma. Osteosarcomas occur in association with Paget's disease, *9 bilateral retinoblastoma,22·57 polyostotic fibrous dysplasia,63'65 osteogenesis imperfecta,27 Waardenburg-like syndrome,32 Other's syndrome,26 multiple hereditary exostosis,28 solitary enchondroma,29 myositis ossificans,25 and osteopoikilosis.11 Among these conditions, the association of osteosarcoma and retinoblastoma is hest known. The occurrence of osteosarcoma in patients with a history of retinoblastoma is probably not solely due to radiation therapy given to the original tumor. In one report, 40% of patients with retinoblastoma treated with bilateral enucleation subsequently developed osteosarcoma. 59 There is also an increased incidence of osteosarcomas in cases of unilateral retinoblastoma.57 Patients with retinoblastoma treated with radiation therapy are at risk for developing orbital osteosarcoma; this risk is greater in cases of bilateral retinoblastoma.

Children with cancer who are treated with radiation and alkylating agents have an incidence of osteosarcoma 133 times that of the general population.57 The risk of developing osteosarcoma in Paget's disease is over 10%; however, this disease is rare before the fourth decade of life.

EXPERIMENTAL ANIMAL MODELS

Osteosarcoma occurs in many animals, including mice,66 rats,67,68 hamsters,68,69 chickens,70,71 dogs,60'62,72 rabbits,40,73 cats, cattle, horses, pigs, and sheep. 74'76 The tumor can occur spontaneously or it may be induced by a variety of techniques, such as exposure to radiation or carcinogens,75 inoculation by viruses,43,67"69 and transplantation of naturally occurring tumors.71 In addition, this tumor can be cultivated in tissue culture media,77 and well defined tissue culture cell lines are available.78

Osteosarcomas can be produced experimentally by injection of radioactive substances that accumulate in bones. Intravenous injection of radium chloride and mesothorium induces osteosarcomas in rabbits.73 Likewise, injection of bone-seeking radionuclides such as radioactive phosphorus, cesium-144, radium-228, plutonium-239, thorium-228, cytrium-91, strontium-90 and others can result in development of osteosarcomas. 79'82 Radionuclides that accumulate on bone surfaces, such as plutonium-239 and thorium-228, tend to give rise to a higher proportion of osteosarcoma in tubular regions of bones than bone volume seekers, such as radium-226 and radium-223.79

Implantation of various radioactive substances such as phosphorus-32, utrium-91, and cesium-144, either directly into bones or subcutaneously, can also result in development of bone tumors, including osteosarcoma.83'85 The rate of induction of bone tumors by radionuclides is roughly proportional to the dose administered, and the latent period increases with decreasing doses.84 Single or multiple bone tumors develop and metastases may occur.85 External irradiation can also induce osteosarcoma.

Several chemical agents have been used in animals to experimentally induce osteosarcoma. In rabbits, injection of zinc-beryllium silicate or beryllium oxide induces osteosarcomas in one third of the animals.40,86-88

Osteosarcoma can be produced by RNA viruses isolated from spontaneously occurring tumors in mice.41·89 The "C" type viral agent, however, is species specific.90 Intravenous injection of SV-40 virus has produced osteosarcoma in weanling hamsters. The same result can be obtained by intraperitoneal injection of Moloney or Harvey murine viruses or sarcoma virus into neonatal rats.91 Subcutaneous inoculation of the BK type human papovavirus into newborn Wistar rats has induced osteosarcoma in less than a quarter of these animals.92 Despite the initial reports, cell-free extracts of human osteosarcomas have not been shown to produce the tumor in experimental animals.93-96

PATHOLOGY

Osteosarcoma is defined as a malignant bone tumor with proliferating cells that have the potential to produce either osteoid or immature bone.97"99 Such a description, although incomplete, is a result of cumulative experience beginning with development of a registry organized by the American College of Surgeons in the 1920s.100,101

Osteosarcoma usually originates in metaphyseal regions of long bones. Central axis lesions are rare, accounting for the primary site in less than 10% of cases in children.8,10 Tumor growth usually extends through the cortex, causing varying degrees of bone destruction and expansion of periosteum (Figure 1). Tumor deposits can occur in the medulla of the bone, separate from the initial tumor site. These are referred to as "skip lesions." They occur in about 20% of cases and are located within a distance of 10 cm from the bulk of primary tumor.

The gross appearance of the tumor is not specific. Histological examination often reveals uniform sheets of osteoblasts with trabecular bone deposition. The appearance varies with the type of the disease. Positive identification requires adequate tissue samples because 15% to 30% of osteosarcomas, proved after surgical resection, may be missed with fine needle aspiration.102,103 Grading of the tumor based on cellular pleomorphism and mitoric acriviry is used to predict the degree of aggressive behavior in the tumor.98 Using this system, more than 80% of children have high grade tumors. The ultrastructure of osteosarcomas varies depending on the cellular elements included in the tumor. I04106 These cells include osteoblasts, chondroblasts, fibroblasts, histiocytes, multinucleated osteoblast-like tumor cells, and atypical primitive mesenchymal cells. Electron microscopy findings are often suggestive but not pathognomonic and may fail to unequivocally distinguish an osteoma or a benign osteoblastoma from this tumor. I04

Figure 1. Cross-section of a resected bone containing an osteosarcoma.

Figure 1. Cross-section of a resected bone containing an osteosarcoma.

While no uniformly accepted classification of osteosarcoma is currently available, it is often divided into five major types. These include the conventional telangiectatic, parosteal, periosteal and small-cell types. 107 Recognition of these subgroups is important because each type has a different prognosis. The conventional or osteosarcoma ordinaire type refers to tumors containing malignant osteoid-producing cells with varying degrees of chondroblastic and osteoblastic differentiation. This entity is the most common type of osteosarcoma. Depending on the predominant tissue content, this type of tumor is subdivided into osteoblastic, fibroblastic, and chondroblastic variants.5,107,108 Telangiectatic osteosarcoma is a rare and highly malignant form of the disease, often involving areas adjacent to the diaphysis of the bone.109,110 Histologically, this tumor is very cellular and contains sparse matrix, areas of osteoid formation, abundant vascular channels, and frequent areas of hemorrhage. Pathological fractures are common.

Parosteal osteosarcoma is most common in older patients and often occurs in the metaphyseal region of the femur.111113 The tumor can reach a significant size prior to diagnosis. It often surrounds the bone shaft without penetration or involvement of the cortical structure. However, at times there is sclerosis of cortex at the site of involvement. Periosteal changes are usually absent in this form of tumor. Histologically, most of these tumors are well differentiated, with presence of irregular bone trabeculae and elongated, fibroblastic-like cellular stroma usually separated by a thick bundle of collagen. The periphery of the tumor shows a significant degree of atypia and anaplasia. The growth of the tumor is relatively slow, but local recurrences and metastasis to the lung, although delayed, do occur.114

Periosteal osteosarcoma is a rare tumor and is most often seen in adolescents. A common site of involvement is the proximal metaphysis of the tibia. The tumor originates from periosteum with only a minimal attachment or penetration of the cortex. The tumor may have a chondrosarcomatous appearance with lobulation and may contain islands of osteosarcoma. Whether this should be classified as parosteal chondrosarcoma or as osteosarcoma is not clear.115

The small-cell osteosarcoma is characterized by presence of malignant small round cells with areas of osteoid formation.116 There may also be areas of differentiation to cartilage and fibrous tissue.

CLINICAL PRESENTATION

Pain is the most common complaint of patients with osteosarcoma. This can be severe and require large doses of analgesic drugs for relief. Frequently, a history of trauma is present. The median duration of symptoms prior to establishment of the diagnosis is three months, but this interval varies from a few weeks to several months."4 Pain often increases with activity, especially weight bearing, and can result in a limp. Depending on the site of the tumor, the mass may or may not be clinically palpable. When therapy is neglected, the tumor can assume a tremendous size (Figure 2). Local edema, tenderness, or decreased range of motion may be present, and occasionally a pulsation or bruit can be found. Pathological fractures are uncommon at presentation, but are seen in up to 30% of cases with the telangiectatic type of osteosarcoma.

Figure 2. Osteosarcoma m a patient whose parents, due to religious beliefs, had not sought treatment until the tumor had achieved a tremendous size and had seeded numerous metastases.

Figure 2. Osteosarcoma m a patient whose parents, due to religious beliefs, had not sought treatment until the tumor had achieved a tremendous size and had seeded numerous metastases.

Osteosarcoma has a propensity to occur in the metaphyses of long bones. Often areas of bone destruction and new bone formation are seen; however, the lesion may be entirely lytic or blastic. In two thirds of cases, a soft tissue mass is present. The initial work-up should include plain roentgenograms of the involved bone, which often are diagnostic. The extension of the tumor through the periosteum may result in a socalled "sun burst" sign (Figure 3), which is demonstrated in 60% of cases."7 Periosteal reaction and the laying down of the newly formed bone can also produce an "onion peel" appearance. Computed tomography (CT) and magnetic resonance imaging (MRI)"8,119 can delineate the extent of the cortical bone involvement and the presence of intramedullary spread, as well as presence of skip lesions and soft tissue extensions (Figure 4). These can assist in planning for the surgical procedure.

The initial evaluation should determine whether there are metastases to chest and skeletal system. At the time of diagnosis, only 10% to 20% of patients have metastases. These figures may be an underestimation; with use of more sensitive techniques, more metastases will be found."5'120 By far, the lung is the most common site of spread of the tumor (Figure 5). Lung involvement may result in pneumothorax. Pneumothorax may occur in the absence of any demonstrable lesion and may be the first sign of metastasis. 121,122 Less common sites of metastases include bones, pleura, lymph nodes, pericardium, kidney, and brain.

Figure 3. Plain lateral roentgenogram of the distal femur showing destruction of the bone. There is also extension of the tumor to a poorly calcified mass, with so-called "sun burst" appearance (arrow).

Figure 3. Plain lateral roentgenogram of the distal femur showing destruction of the bone. There is also extension of the tumor to a poorly calcified mass, with so-called "sun burst" appearance (arrow).

Recognition of pulmonary metastases on plain chest x-rays is not always possible.123 Use of conventional tomography and computed tomography (CT) may disclose the presence of tumor, which may not have been apparent with use of standard techniques. About 10% to 30% of patients with normal standard chest x-rays have metastases identified with linear xray tomograms. Furthermore, as many as 15% of those with negative tomograms may have positive CT scans.123 Even with current technology of CT scanning, tumors less than 3 mm in diameter are not detected. During surgical exploration of the lungs, the number of metastases identified is usually several times larger than the number seen on CT scan. On the other hand, the number of false-positive findings in CT scans of the lung can be substantial, because it is not always possible to differentiate tumors from granulomas, pleural-based lymph nodes, and nodules of scar tissue. Patients who have undergone surgery for removal of metastases may pose additional diagnostic difficulties. The role of MRI in detection of metastases is now under study.

Figure 4. Magnetic resonance image of osteosarcoma in the distal femur, also showing extensive soft tissue involvement.

Figure 4. Magnetic resonance image of osteosarcoma in the distal femur, also showing extensive soft tissue involvement.

Figure 5. Multiple metastases to the lung as shown by CT scan.

Figure 5. Multiple metastases to the lung as shown by CT scan.

Radionuclide scintigraphy with technetium-99 methylene disphosphonate can detect the extent of the primary tumor and presence of metastases.124 Generally, the primary bone tumor shows increased concentration of the radionuclide ("hot spots"); however, "cold spots" at the primary site of osteosarcoma have also been observed. The technique may allow early detection of asymptomatic metastases in bones as well as in soft tissues.125'128 Radionuclide scans also are accurate for detecting intramedullary tumor involvement, but often the scans show a falsely extended uptake pattern in tissues adjacent to the tumor.125 Metastatic lesions can be identified with accuracy by presence of areas of increased radionuclide uptake (Figure 6).

At the time of diagnosis, nearly half of patients have increased levels of serum alkaline phosphatase.129,130 This finding is associated with an increased likelihood of subsequent pulmonary metastasis.131,132 Nevertheless, presence of a normal serum alkaline phosphatase level in patients with detectable metastases is not unusual.133 Nonspecific tests, such as erythrocyte sedimentation rate (ESR) and serum ceruloplasmin level, are of little value in the diagnosis and assessment of patients with osteosarcomas. Despite early enthusiasm, determination of somatomedin level and assay of growth hormone are not useful in assessing these patients.134

Figure 6. Radionuclide scintigraphy with technetium-99 methylene disphosphonate showing lung metastases in a patient who had previously undergone amputation.

Figure 6. Radionuclide scintigraphy with technetium-99 methylene disphosphonate showing lung metastases in a patient who had previously undergone amputation.

Accurate diagnosis of osteosarcoma ultimately depends on histopathological examination of the tumor. Biopsy of the tumor should be obtained prior to definitive surgery. Opinions differ as to whether biopsy should be obtained at the same time as definitive surgery is done.135·136 The difficulty in obtaining immediate and reliable pathological diagnosis at the time of biopsy is a limiting factor. No difference in the rate of relapse-free survival could be shown in a retrospective comparison of patients who have immediate amputations as compared to those who had amputations several days after biopsy.137

THERAPY OF OSTEOSARCOMA

There is no uniformly accepted treatment of osteosarcomas. In general, the current treatment of this tumor includes surgery and chemotherapy. In many centers, surgery is tailored to the extent of the patient's disease. The choice of surgical technique, that is, amputation or limb salvage, depends on several factors. These factors include age, size and degree of skeletal maturity, anatomical location, extent of the tumor, presence of soft tissue and neurovascular involvement, expected postoperative function, and availability of surgical expertise, together with an evaluation of the patient's attitude and desires. In view of the results of recent studies clearly indicating the usefulness of chemotherapy, most centers combine surgery with this modality for treatment of all patients. In those patients who will undergo limb salvage operations, chemotherapy is administered intravenously or intraarterially in advance of the surgical operation.

Amputation

Historically, amputation has been the primary treatment for osteosarcoma. This is often possible because the tumor occurs predominantly in the extremities. A major controversy concerns the level of amputation. Clinically undetected "skip lesions," in the form of intramedullary tumor deposits separated from the major area of tumor, are known to occur. These skip lesions are seen in approximately 20% of cases and are not always detectable with use of x-ray examination, scintigraphic scans, or magnetic resonance imaging. The lesions may be found beyond the epiphyseal plate.138 Extension of tumor into the bony medulla may occur, with tumor extending into a proximal joint.139140 The usual practice is to remove the entire extremity, allowing a safety margin of 7 cm to 10 cm, or to excise the whole bone at or above the contiguous joint. For a proximal lesion of the tibia, an above-theknee amputation or disarticulation of the knee is performed.9 For a proximal femoral tumor, depending on its location, a hip disarticulation or hemipelvectomy is done. 141 A Van Nes rotationplasty, 142 using the inverted heel as a knee joint, is possible in younger children and has produced good functional results.143

In the upper extremities, for tumors of the proximal humerus, shoulder disarticulation or forequarter amputation may be carried out. Many surgeons perform a cross-bone resection for distal femoral lesions, allowing for the safety margin above the limit of medullary extension. Careful suturing techniques are necessary to prevent stump protrusion. The local recurrence rate is less than 5%.144'146 In one study, only two femoral stump recurrences were noted in 100 cases.135 Angiography and prelabeling with tetracycline147 may also contribute to a better delineation of tumor extent. To reduce the spread of micrometastases, many surgeons apply a tourniquet proximal to the tumor.135 The efficacy of this maneuver, however, has not been established in controlled studies.135 Likewise, the preventive effect of anticoagulant therapy in the development of metastases has not been clearly documented.148

Following amputation, rehabilitation should start without delay. Immediate fitting of the patient with a temporary prosthetic limb is of utmost importance. This may have a positive psychological effect by reducing the patient's anxiety and sense of helplessness, as well as providing hope for rehabilitation. Consideration should be given to preparation of a prosthesis prior to the surgery.149 Changes in stump volume due to adjuvant chemotherapy, of course, must be anticipated.

Limb-Sparing Techniques

En bloc resection of tumor as an alternative to amputation is increasingly favored.135,150155 For this technique to be used, the patient's bone growth must be near the stage of epiphyseal closure, and the tumor should preferably be predominantly intraosseous with only a small degree of soft tissue extension.135 Limbsparing procedures are most applicable to patients with a small tumor located above the knee. The soft tissues, as well as vascular and nerve structures, must be free of tumor. A careful preoperative evaluation, including a biomechanical profile, is essential.

Patients are often given preoperative chemotherapy as an initial effort to destroy micrometastases. This therapy may also reduce tumor mass or better delineate the margins of the tumor and can transform a case not suitable for limb-sparing surgery into an operable one. Prophylactic radiation therapy to the lungs in the preoperative or postoperative period has not improved prognosis. 156,157 Intra-arterial preoperative chemotherapy in the area of the lesion can also be used. Allograft replacement, such as a crossleg graft, is possible. Segmental arthrodesis and metallic implants are also used. After bone replacement, most patients are given adjuvant chemotherapy.

Presence of metastases was previously considered a contraindication to limb salvage surgery.154 However, with improved results of adjuvant chemotherapy this is no longer necessarily true.

Thirty-three percent of patients undergoing en bloc resection of distal femoral lesions at a later date may require conventional amputation.155 Complications that may render amputation necessary include infection, tumor recurrence, or prosthesis breakage. When limb-salvaging surgery is done in a lower extremity, a full-leg ischial weight-bearing brace may be needed, because the function of an artificial knee joint is limited.155,158,159 The range of knee motion after segmental knee replacement ranges from 0° to 140° (average 88°). 135

Following en bloc resection of tumors of the shoulder girdle, using a Tikoff-Lindberg procedure, preservation of good hand function is possible.136,155,160 Criteria for performing this procedure include: 1) no involvement of the axillary artery or brachial plexus; 2) no lymph node involvement; and 3) no fixation of tumor to the chest wall.57 The TikofY-Lindberg procedure involves an en bloc upper humeral interscapulothoracic resection. This operation appears to preserve more function than is possible following a forequarter amputation or a shoulder disarticulation. In the Tikoff-Lindberg procedure the distal humerus is fixed to the chest wall with a metallic rod. 1 36 Upper extremity preservation is much better than use of any external prosthetic device attempting to replace forequarter amputation or a shoulder disarticulation.154 Radial nerve palsy may be treated with a cock-up splint or tendon transfer.155 Shoulder padding or plastic fillers can conceal the cosmetic defect. Cryotherapy with liquid nitrogen has been used to control pain.160 There is no evidence that this therapy can affect the growth of the tumor.

Table

TABLE 1Single Agent Useful In Therapy of Osteosarcoma

TABLE 1

Single Agent Useful In Therapy of Osteosarcoma

With the definite improvement in long-term survival with contemporary adjuvant chemotherapy,155,161'166 it is likely that performance of limbsparing procedures will become more common. Improved survival implies that chemotherapy provides better control of micrometastases. So far, available studies have shown no difference in disease-free survival in use of radical surgery as compared to nonradical surgery. 167,168

RADIATION THERAPY

Osteosarcoma is a radioresistant tumor and use of radiation does not now play a major role in the therapy of this disease. After a full course of radiation therapy, viable tumor cells were noted to be still present in as many as two thirds of primary tumors.156 The tumoricidal dose is greater than 6,000 cGy.169,170 Preoperative radiation and surgery has not produced better results than surgery alone. 171,172 Complications frequently occur after radiation. In one study, 23% of patients treated with 8,000 to 9,000 cGy were relapsefree at two years, but 43% of these surviving patients eventually required an amputation because of complications.173

Prior to the use of adjuvant chemotherapy, a radiotherapy regimen developed by Cade was commonly used in Europe.48 The aim of radiation therapy was to spare those 80% of the patients, who would have died, from the added burden of an amputation. Over a period of six to nine weeks, the tumor was irradiated with 6,000 to 8,000 cGy.174 Amputation was performed six to nine months later only in those patients who had no evidence of metastases. The survival rate of patients treated in this way was equal to that of patients whose affected limbs were amputated immediately after diagnosis.

Radiation, even in doses of 4,000 to 6,000 cGy, may not relieve local symptoms. 175 Large doses of radiation ranging from 12,000 to 16,000 cGy after tourniquetinduced hypoxia of local tissue, produced less than 50% local tumor control.176 Likewise, prophylactic whole lung radiation, in general, did not decrease the rate of relapse-free survival of patients with osteosarcoma.177 The use of radiation has only a very small role in current therapy of osteosarcoma.

CHEMOTHERAPY

Until recently, the beneficial effects of chemotherapy in osteosarcoma were controversial. In the Mayo Clinic studies performed in the 1970s, after 18 months of follow-up, the survival rate of patients treated with surgery alone was 52%. These results were no different than the survival rate of patients who received adjuvant chemotherapy with a combination of high dose methotrexate and vincristine. These findings cast doubt on the value of giving chemotherapy in osteosarcoma,178·179 and even resulted in proposal of a theory that the biological nature of this disease may have changed with time. Because of lack of a large randomized study comparing surgery alone with use of surgery and chemotherapy, the notion that adjuvant chemotherapy was useless could not be refuted.

From recent controlled studies, however, it is now clear that chemotherapy does significantly increase the survival of these patients.161·180 Conclusive results were reported from two large multicenter studies. In these studies, the actuarial relapse-free survival rates of patients with osteosarcoma of the extremity treated with adjuvant chemotherapy were compared with survival of control patients not given chemotherapy. 161 The control group was randomly assigned to observation alone following definitive surgery. In one study, at two years, the actuarial relapse-free survival of controls was only 17%, as opposed to 66% for the patients given chemotherapy.161 In a similar randomized study, disease-free survival at two years of 27 patients who were treated with surgery alone was 20% compared to a two-year survival rate of 55% for 32 patients who were given adjuvant chemotherapy. The demographic make up of the patients and the type and extent of the tumors in both groups were similar, thus eliminating any obvious bias to account for the observed differences in survival rates. It should be noted that the survival rate of patients treated with surgery alone in the 1980s145,161 is similar to the survival rates of patients similarly treated in earlier studies, ranging from 17% to 20%. It was concluded from these studies that the natural course of osteosarcoma of the extremity has remained unchanged, and adjuvant chemotherapy increases the rate of prolonged relapse-free survival of patients with this tumor.

Use of Single Agents in Chemotherapy

Used alone, only a few chemotherapeutic agents have shown effectiveness against osteosarcoma. These drugs include methotrexate, melphalan, cisplatin, adriamycin, 5-fluorouracil, cyclophosphamide and mitomycin C (Table). Drugs that when given alone have no demonstrable efficacy include actinomycin diaziquone-D (AZQ), the epiphyllotoxins (etoposide and teniposide), dacarbazine (DTIC), mitoxanthrone HCl (DHAD), acridinyl anisidide (m-AMSA), and vincristine. The presumed lack of efficacy, however, is based on a very small number of trials carried out in patients with advanced disease.

A variety of biological response modifiers, including transfer factor, nonspecific immunotherapy, and interferon, have been tried in the therapy of osteosarcoma.183'186 During the past decade, interferons, because of their antiproliferative activity, have been used in the treatment of osteosarcoma.183,185 Human osteosarcoma xenografts in nude mice were found to respond to human interferon, but not to murine interferon. This finding indicated that the antitumor effects may be related to a direct anticellular action and not due to immune modulation.181 Isolated reports show efficacy of interferon in treatment of a small number of patients with osteosarcoma. In one report, 106 units of interferon given three times a week produced objective reduction in metastatic tumors in two out of three patients.182 A nonrandomized clinical trial done in Sweden, using naturally occurring partially purified interferon, was originally thought to be successful. But after longer follow-up, the results did not indicate enough improvement in survival to warrant wider use of this agent. 183 More recently, the German-Austrian Cooperative Group found that addition of interferon to chemotherapy produced no prolongation of disease-free survival of patients with osteosarcoma.152

In a very limited study, the results of treatment with transfer factor were comparable to combination chemotherapy consisting of methotrexate, doxorubicin, adriamycin, and vincristine.184 Survival at 18 months in patients given transfer factor was 55%. 184 The result of this study has not been confirmed.

Use of Combination Chemotherapy

Chemotherapy regimens containing two or more drugs produce results superior to use of a single agent. 161 Chemotherapy of osteosarcoma has usually consisted of giving these agents prior to surgery or postoperatively.

Postoperative Chemotherapy

Postoperative chemotherapy is administered with the goal of destroying residual tumor cells not removed by surgery. Many drug combinations, and varying doses and schedules, have been tried. The rational basis for multidrug regimens is the finding that chemotherapeutic agents are often more effective when given in combination.

Earlier studies demonstrated that methotrexate in doses ranging from 3 to 7.5 g/m2 followed by citrovorum factor rescue is clearly an effective agent in osteosarcoma. Methotrexate given in this way produced a response rate of approximately 40%. 187,188 Because it was observed that vincristine can facilitate the intracellular influx of methotrexate, this agent was added to the methotrexate-citrovorum regimen. This new study showed a disease-free survival rate of 43% at two years. The addition of adriamycin to this regimen did not improve the results.189 Adjuvant chemotherapy with the classic combination of cyclophosphamide, vincristine, and actinomycin D (VAC) was not effective in patients with osteosarcoma, producing a disease-free survival rate of only 20%. I5

The combination of cyclophosphamide, Oncovin®, phenylalanine mustard and adriamycin (COMPADRI-I) produced significantly better results. In 44 newly diagnosed patients with nonmetastatic osteosarcoma of an extremity, this regimen produced a relapsefree survival rate of 55% after 37 months of followup. 190 This combination of drugs was later modified by addition of high dose methotrexate (COMPADRI-II). Treatment with this combination in 60 évaluable patients yielded a disease-free survival rate of only 37%. In a subsequent study, the same drugs were used; however, doses of methotrexate were given on a weekly basis and the total dose of adriamycin was increased to 360 mg/m2 (COMPADRI-III). The relapse-free survival rate in 44 patients followed for a median of 14. 5 months was 42%. 191 Another study used vincristine, high dose methotrexate with citrovorum factor rescue, and adriamycin in conjunction with radiotherapy and amputation. The rate of prolonged survival was 30%. In one study, patients given L-phenylalamine mustard and cyclophosphamide together with high dose methotrexate showed a prolonged disease-free survival rate of 50%.192 The combination of cyclophosphamide, vincristine, adriamycin, and DTIC produced a disease-free survival rate of 61% with a median follow-up of 60 months.

After adriamycin was shown to be effective in osteosarcoma,193,194 it was added to combination chemotherapy regimens. A combination of intermediatedose methotrexate and adriamycin given at regular intervals for one year postoperatively in 11 patients produce a 64% disease-free survival rate at 42 to 62 months.195 In 22 patients treated with high dose methotrexate and adriamycin for 16 months, the disease-free survival rate after a median of three years follow-up was 54%. u5 Addition of adriamycin to vincristine and high dose methotrexate with citrovorum factor rescue in 16 patients who had undergone amputation resulted in a survival rate of 37% after 24 to 36 months follow-up.196 A combination of adriamycin, vincristine, high dose methotrexate used in 23 patients produced a survival rate of 52%. The use of adriamycin plus vincristine and high dose methotrexate resulted in a disease -free survival rate of 30% in a group of 16 patients who had received radiation therapy and amputation. 192 Administration of six courses of adriamycin followed by high dose methotrexate, cyclophosphamide, and L-phenylalanine mustard in 12 patients produced a disease-free survival rate of 50% after 24 months of follow-up.

The use of adriamycin and vincristine given in combination with high or medium dose methotrexate has not improved the results.197 Only 38% of patients remained disease-free at four years after diagnosis. A regimen consisting of adriamycin, vincristine, mitomycin-C, DTIC, cyclophosphamide, and BCG used in 21 patients with osteosarcoma of the limbs produced a disease-free survival rate of 85%. However, these patients were followed for only six to 29 months.198

Cisplatin has been shown to be a valuable agent for the treatment of osteosarcoma. Given as a single agent, cisplatin produces a partial or complete response in about one third of patients.199'204 This response is observed in newly diagnosed patients as well as in those with recurrent tumor. 205 In one study, this agent was given together with methotrexate. This combination produced complete control of tumor in nine patients for a mean follow-up of 19 months.206 Cisplatin is effective when combined with adriamycin for therapy of osteosarcoma.207,208 This combination produced complete response of tumor in 14 of 22 patients (64%), with a median follow-up of 70 months.

Chemotherapy combinations containing cisplatin, while effective, can cause severe renal toxicity.209 Patients who receive multiple courses of cisplatin show progressive reduction in kidney function as evidenced by decreased creatinine clearance and delayed excretion of serum methotrexate. In one study, the first indication of toxicity was seen after a cumulative dose of 450 mg/m2. Even after termination of therapy, renal impairment persists.209,210 Attempts are being made to reduce the cumulative toxicity of this group of drugs using analogues with less renal and autotoxicity.

To assess the role of chemotherapy in a randomized prospective study, a postoperative chemotherapy regimen consisting of adriamycin, high dose methotrexate, bleomycin, cyclophosphamide, and actinomycin D was given to 32 patients. The control group consisted of 27 patients who received no adjuvant chemotherapy. At a median follow-up of two years, the disease-free survival rate of the group given chemotherapy was 55% compared with 20% for controls. The disease-free survival rate of the controls was comparable to the survival rates reported in earlier studies.145 Likewise, another large multicenter study reported a two-year relapse-free survival rate of 66% in patients with osteosarcoma of the extremity given adjuvant chemotherapy.211 The control group, treated only with surgery, showed a relapse-free survival rate of only 17%. This study confirmed that the natural history of osteosarcoma has not changed and adjuvant chemotherapy increases the rate of prolonged relapsefree survival of patients with high-grade disease.

Preoperative Chemotherapy

The administration of chemotherapy prior to amputation or segmental resection of the involved bone can provide important information regarding the biological responsiveness of the tumor. In effect, giving chemotherapy prior to surgery is an in vivo test for assessment of the tumor's sensitivity to chemotherapeutic agents and helps in the selection of appropriate regimens for postoperative therapy.212 Histological examination of the resected tumor permits immediate assessment of the degree of tumor necrosis and allows for change of therapy or increased doses. Based on these findings, a grading system has been developed.149 Accordingly, four levels of chemotherapy effect, ranging from minor changes (grade I) to presence of areas of a cellular tumor; fibrosis and necrosis admixed with areas of histologically viable tumor (grade II); predominance of tumor destruction with scattered foci of viable tumor (grade III); and complete effacement of viable tumor (grade IV), are identified. Grades I and II response are associated with more frequent recurrence of the disease as compared with those with grade III and IV responses.149,213 One report indicated a positive correlation between a "good" radiologic response and histological grading of the resected bones.214

Other advantages of preoperative chemotherapy include reducing tumor size in order to make possible or to facilitate an en bloc resection and limb preservation, and allowing time for preparation of a custom fit prosthetic bone replacement. A variety of preoperative chemotherapy schedules have been developed. Earliest attempts included therapy of patients in whom it was thought that en bloc resection of their tumor could be achieved with the insertion of a custom-fitted prosthetic bone to replace the resected bone. 147 In this study, newly diagnosed patients with biopsy-proven osteosarcoma received repeated sequential preoperative chemotherapy, including vincristine, high dose methotrexate with citrovorum factor rescue, followed in ten to 14 days by adriamycin for a period of two to three months. After en bloc tumor resection, postoperative chemotherapy included courses of agents used prior to surgery along with high doses of cyclophosphamide two weeks prior to and after each methotrexate course (T5 protocol). The dose of methotrexate was increased when no early response in the primary tumor was noted. In 31 patients treated in this fashion, the disease-free survival was 58%. The overall survival at four years was 77%, with a disease-free survival rate of 50%. 149,215 The survival rate of these patients is clearly superior to the 52% four-year overall survival rate observed in 23 patients who were given similar therapy after amputation. These findings indicate that preoperative chemotherapy produces an improved survival rate. 149

In a subsequent study, a combination of bleomycin, cyclophosphamide, and actinomycin D (BCD) was substituted for cyclophosphamide. Also, more frequent courses of methotrexate with higher doses (12 g2) were given to young children (T7 protocol). This therapy resulted in an overall survival rate of 81%, and a disease-free survival rate of 76% in 37 patients.149,216 Since the degree of necrosis due to chemotherapy showed correlation with the outcome of the disease, the postoperative chemotherapy regimen was modified by substituting cisplatin for methotrexate (TlO protocol) in patients who showed a grade I or II response. In the initial report of this study, out of 57 évaluable patients, 22 had a grade I or II response. A recent update on 87 patients treated with this protocol showed that 77% have survived free of the disease after a median of 58 months follow-up.216 The overall survival was 82%. More recently, in complete responded, preoperative chemotherapy with high dose methotrexate plus bleomycin, cyclophosphamide, and actinomycin D is followed postoperatively by one course of BCD and two courses of high dose methotrexate for a total of 15 weeks (Tl 2 protocol). Of the 51 patients entered in this study, 75% are disease-free and 76% are alive after 25 months of follow-up.216

But not all patients undergoing preoperative chemotherapy are candidates for limb salvage procedures. In two reported studies, only 48% and 58% of patients had limb-sparing operations.217 The degree of necrosis in the resected bone can serve as a prognostic sign and offer guidance in prescribing further chemotherapy. However, even with the finding of over 90% necrosis of tumor, further extensive chemotherapy is indicated. In one study among patients receiving preoperative chemotherapy with methotrexate and cisplatin and having a grade IV necrosis of tumor, the group given 24 weeks of postoperative chemotherapy had a superior rate of prolonged survival as compared with those receiving only four weeks of treatment.218

Intra-Arterial Chemotherapy

Intra-arterial chemotherapy provides localized perfusion and produces maximum concentration of drug in the tumor. This technique exposes the tumor to a local concentration of drug, which cannot be achieved without life-threatening side effects by intravenous chemotherapy. Furthermore, the systemic effects of intra-arterial treatment are similar to those of standard therapy. Intra-arterial therapy is most effective using drugs with a short half-life. The serum concentration of cisplatin achieved by this technique is four times higher in the afferent artery of the tumor as compared with the efferent veins.219 Intra-arterial adriamycin as well as cisplatin and methotrexate are used alone or in combination, with or without intravenous adjuvant chemotherapy, for the treatment of osteosarcoma.220 After intra-arterial infusion, CT scans may show decrease in tumor size.221 The effects of high dose methotrexate with citrovorum factor (MTX-CF) given intravenously were compared with intra-arterial cisplatin. The responses of intra-arterial therapy were found superior to the conventional method of parenteral chemotherapy. Furthermore, two patients who failed on intravenous chemotherapy responded to intra-arterial therapy.222

Continuous infusion of cisplatin and adriamycin, accompanied by obstruction of blood flow after infusion in the hope of increasing perfusion of the tumor, has been tried.223 Severe skin and soft tissue necrosis, however, limits this technique. Intra-arterial infusion of 5-bromodeoxyuridine (BUdR) as a rathosensitizer, combined with hyperfractionated irradiation and adjuvant systemic chemotherapy, has resulted in local control in seven of nine patients in whom primary tumors were not resectable. Significant soft tissue injury occurred in five patients, and distant metastases developed in six. Histopathological study of the tumor after intra-arterial therapy may still demonstrate viable tumor cells in half of the patients, and gross residual tumor may persist even in areas of tumor supplied by an artery.

TREATMENT OF METASTASES

Following treatment of osteosarcoma, local tumor recurrence at the margins of the primary lesion and distant metastasis can occur. The spread of the tumor is most often to the lungs and bones; however, metastases also occur in other organs, including the heart, kidneys, and epidural spaces. 224227 Pulmonary metastases in 20% to 45% of patients are bilateral.228,229 If overt metastases are not treated, the majority of patients die within one year.228,229 Complete evaluation, including radiological examinations and bone scans, is essential in designing a plan for treatment. Spread of tumor ranges from development of only one or two metastatic lesions, which may be resectable without recurrence, to development of multiple and recurring lesions. This behavior may reflect the degree of differentiation of the tumor.212 It is also possible that treatment may alter the pattern of metastases, with nonpulmonary metastases becoming more prevalent. 23° The histological picture of a metastasis may be different from that of the primary tumor. Furthermore, the histological pattern may also vary among metastatic lesions, as shown in lesions excised during consecutive thoracotomies from the same patient.231 Usually, the number of metastases found during surgery is several times higher than the number seen in radiographic studies.

The therapy of metastases includes surgical removal and chemotherapy. Surgical resection helps to control the local growth of the tumor, and prevent pulmonary hemorrhage and airway obstruction. Some patients require repeated resection of metastases. Aggressive surgical treatment is warranted, however, and can prolong survival, including disease-free survival, and may even produce "cures." Based on various studies, nearly half of patients undergoing one or more thoracotomies for resection of metastatic disease, with or without accompanying adjuvant chemotherapy, survive two to three years after the operation. Furthermore, 23% to 56% have a chance of long-term, disease-free survival.232'240 Factors influencing survival in patients with lung metastasis include sex, number of nodules detected radiographically and resected, completeness of resection, tumor doubling time, disease-free interval, and unilateral or bilateral location of lesions. 238'245

The surgical technique used for resection of pulmonary metastasis varies among surgeons. When the location of detectable tumor allows, a midline sternalsplitting incision is advocated.245 This approach allows access to both pleural cavities and the mediastinum; thus, the extent of chest involvement can be fully assessed and complete resection can be done in one operation. Furthermore, this technique is associated with a faster recovery as well as with less postoperative pain and fewer disturbances of pulmonary function.246 Wedge or segmental resections are usually adequate. In one review, only five out of 60 patients needed lobectomies.247 Since lymph nodes, bronchial tree and mediastinum are rarely involved, node dissection is not usually recommended. 160 The surgical mortality rate of this procedure is extremely low,248 and morbidity is not excessive. In addition to pulmonary metastases, resection of other metastases should also be attempted. Involvement of vertebral bodies by metastatic osteosarcoma may require urgent surgical decompression.249 Chemotherapy of metastatic lesions should be tailored to suit the needs of the individual patient, with consideration given to prior treatments.

MANAGING PSYCHOSOCIAL EFFECTS

Cancer diagnosis, plus the medical and surgical treatments, have profound psychological effects. With increased duration of survival of patients, the longterm effects of cancer, including psychosocial disturbances, have become increasingly important. Knowledge of factors that play a role in helping patients cope with their disease and its associated disabilities can serve, to a large extent, to ameliorate the psychosocial difficulties. The peak incidence of osteosarcoma is during the second decade of life, a time when physical attributes and body image are of utmost importance. The psychological trauma caused by cancer itself, as well as by surgery and chemotherapy, during this sensitive stage of life can be devastating. Incomplete development of adaptive mechanisms in adolescents, coupled with the stress of disease, fear of the unknown, alterations in body image, and isolation from peers, as well as interrupted educational, vocational and life plans, can produce a crisis of major proportions.

At the time of diagnosis, an estimate of the level of the patient's emotional maturity and physical development should be made in order to plan a program for rehabilitation and counseling. Prior experience shows that despite all obstacles, when properly managed, these patients can adapt well to the situation. In fact, the majority can live useful and productive independent lives.250'251 In one long-term study of 27 cancer patients who had amputations, mainly for treatment of osteosarcomas, 41% of those older than 18 years of age had attended college and only 14% had not completed high school. In this group, 46% were married and nearly all patients were gainfully employed. Based on educational, vocational, and social achievement of these patients, over 85% were judged to be welladapted. A positive attitude and open discussions with health care providers can help provide a justifiably optimistic view of life, especially during the early and highly distressful times of diagnosis and treatment.

SUMMARY

Osteosarcoma is the most common bone tumor of children and adolescents. The peak incidence of the disease is in the 15 to 19 year age group. The disease is more commonly seen in males than females. While several factors, including exposure to radiation, genetic disorders such as retinoblastoma, and high rate of bone growth, have been associated with osteosarcoma, in most cases no definite etiology can be established. Osteosarcoma usually originates in the metaphyseal region of long bones and extends through the cortex, causing varying degrees of bone destruction and expansion of periosteum. The radiographic appearance caused by this process is often referred to as "sun burst" sign.

Positive diagnosis of osteosarcoma is made by histopathology. The histopathological classification of osteosarcoma can also predict the degree of aggressive behavior of this tumor and thus has prognostic significance. Surgery, including amputation or limb-salvage procedure, is the mainstay of treatment of osteosarcoma. It is now unequivocally established that adjuvant chemotherapy will prolong the survival of patients with this disease. Chemotherapy agents often used include platinum dérivâtes, methotrexate, vincristine, cyclophosphamide, adriamycin, actinomycin D, bleomycin and DTIC. Depending on surgical decision, these agents can be used prior to or after the operation. Immediate fitting with prosthesis and provision of appropriate medical and psychological support in the care of these patients is essential.

REFERENCES

1. Altman AJ, Schwartz AD: Malignant Diseases of Infancy. Childhood and Adolescence. Philadelphia, WB Saunders, 1978. pp 405-426.

2. Third National Cancer Survey: Incidence Data. National Cancer Institute, 1985, Monograph No. 41.

3. Miller RW, Etalager NA: ; PeaW 1974; 85:664.

4- 1985 Cancer Facts & Figures. The American Cancer Society, 1985.

5. Dahlin DC, Coventry MB: ; Bone Joint Surg 1967; 49A:I01-1I0.

6. Geschickter CF. Copeland MM: Tumms of Bone. Philadelphia, JB Lippincott Company, 1949, pp 181-212.

7. Birch JM. Marsden HB, Swindell R: Br) Cancer 1980; 42:215-22).

8. Hayles AB, Dahlin DC, Coventry MB: JAMA 1960; 174:1174-1177.

9. Marcove RC. Mike V. Hajek JV. et al: NY State) Med 1971; 71:855-859.

10. Price CHG, Zhuber K. Saher-Kuntschik M. et al: I Bone Joint Sur« 1975; 57B:34l-345.

11. Scranton PE, De Cicco FA. Totren RS, et al: Cancer 1975; 36:2179-2191.

12. Carón AS, Hajdu SI, Strong EW: AmJ Surg 1971; 122:719-725.

13. Plan C, Shanks E. Hustu O. et al: Cancer 1977; 39:51-57.

14. Huvos AG: Bone Tumors: Diagnosis, Therapy and Prognosis. Philadelphia, WB Saunders, 1979, ? 49.

15. Sutow WW. Sullivan MP. Wilbur JR. et al: J CIm Pharmacol 1975; 15:530-533.

16. Sutow WW. Sullivan MP. Fembach DJ, et al: Cancer 1975; 36:1598-1602.

17. Kauffman SL, Stout AP: Cancer 1963; 16:432-439.

18. Valderrama E, Kahn LB, Wind E: Cancer 1983; 51:1132-1137.

19. Fine G, Stout AP: Cancer 1956; 9:1027-1043.

20. Strauss HK, Jackson M: Surgery 1951; 29:917-928.

21. Fraumeni JF Jr: Cancer 1967; 20:967-973.

22. Kitchin FD: Tumors of the Eye. ed 3. Hagerstown. Harper & Row Publishers Inc. 1976. pp 125-131.

23. Abramson DG. Ronner HJ. Ellsworth RM: AmJ Ophthalmol 1979; 87:624-627.

24. Scimk. RN, Lawmsn JT, Cam, GAb: Lawn 1974; 34~2O77.2O79.

25. Pack Ut Beaun,J RR:)AMA 942; 19:776-779.

26. Beaddock GTE Hadltso VII),) Bone bong Stag `966; 48B:l45'149.

27. Klenennan L. Ockenden BC, Ton,nsend AC:] Bone)aing Sung 1967; 49B:314-323.

28. Glass AG. Feaumeni JFJe:] Noel Cancer Ins; 970; 44:187-199,

29, Rockwell MA, Enneking WE) Bone Ernst Swig 1971: 53A:341.344.

30. Guns Ri, Sedaoo H: Modern Majfrmne 1973; 41:82.

31. Mmdcli ER. Northup CS. Dooglats H&) Bone ,kcu Sing 1978; 60A:405.408.

32, Parry DM, Safyes AW, Molvihill JJ: I Med Gene; 1978; 15:66-69.

53. Fnu,neni )F )r: Bone cancer: Fmnt Radlaa `flier Oncol 1975; 10:17.

34. Li FE Tucker MA, Fraurneni JFJr:] Pediaer 1976: 88:419.423.

35. Miller RW; Recent Reuaks Cancer Ret 1976; 54:50-62.

36. Mulvihill JJ, Gealnick HR. Whang-Beng 1. teal: Cancer 1977; 403115.3122.

37. Bottomley RH, homer AL, Candur PT: Cancer 1971; 28:519-528.

38. Ewing J. Arch Baud 1935: 19:690.728.

39. Price CHO: 3 Bone Joint Sung 1962; 44B:366-761,

40. Teppo 1, Saxen B, Tervo T: Lance; 1977: 1:405.

41. Dutra FR, Largenr El: Am) Radiol 1950; 17:197.209.

42, Domes JM. Dent FA: BrJ Cancer 1950; 4:212-222.

43. Fmnkel MP, Biskis BC Jmnkins rB Science 1966; 151:698.701.

44. Soehr,er RL, Dnsocho,o,stc, L: Nature 1969; 224:191.192.

45. Macton DL. Malsngren BA: Science 1968; 162:1279.1281.

46. Finkel MP. Reilly CA Jr. Biskis DO: Front Radiog Titer Oncol 1975: 10:18.

47. Martland HS: JAMA 1929; 92:466,473.

48. Cede 5:3 B CoIl Sing Edinh 1955: 1:79-Ill.

49. Toupa A, Tanaka T, Marl T, eta!: I Redear Ret (Tokyo) 1963; 4:126-145.

50. Boyd iT, Lariglands ,~3. Maccabe JJ: Br Mcd) 1968; 2:517-523.

51. Spins J, Mays CW: Health Phys 1970:19:713.729.

52 Rosenblatt LS, etal: Health Phys 1971; 21(63:869.875.

53. Li Fr, Cassady JR. Jafle N: Cancer 1975: 35:1230.

54. Zaclsariades N, Patrinou C. Benetas 5, Ct al: 3 Oral Maxillofac Stag 1985: 43( 4) : 297. 299.

55. Cahan WG. Wcx,dard HQ, Hig;nl,axham NL, etal: Cancer 1948: 1:3.29.

56. Huvjo AG, W,xalatd HQ, Cahan WO, etal: Cancer 1985:55:1244.1255.

57. Tucker MA, D'Angmo 03, Booce JO, eta!: N Engl 3 Med 1987: 317:588.593.

58. Oordman MA, McMastet ill. Drools AL, eral: Cancer 1978; 42:603.610,

59. Adamson Ii. Beostrons LA, EIendic S. eta!: Ann Eodoctnol 1980: 94:517-522.

60. Bech.Nielsen 5, Haskins ME, RejfJ5, etal: I Nail Cancer Inst 1978:60:349.353.

61. Jacobton SA: Am) C1n Radial 1966: 50:622.623.

62. Tlalma BA:] Nail Cancer Inst 1979:61:743.748.

63. Huvoo AG. Higinb.icharn NL, Millet TR: 3 Bone Joint Sing 1972: 54A:1047-1056.

64. Schso-artt DT, Alpert M: Am] Med Scm 1964: 247:1.20.

65. Brodeur GM. Caces 1. W:lliams DL, era!: Cancer 1980: 46:1197.1201.

66- Looter WRit: Cancer Chemo Rep 1975: 5:151.161.

67. Ikemoto K, Yamamoto T: GANN 1972: 63:141.

68. Soehner RL. Fujinaga 5, Dnocho.wski L: Co,nparaove Leukemia Research 1969. Kaigee, Betel, 1970, pp 593-599.

69. Diamandopoulos 01:1 NazI Cancer Inst 1973: 50:1347.1365.

70. R,auo P, Murphy JR. Tytlet WH: JAMA 1912: 59:1793-1794,

71. Tyrlet WH: 3 Esp Med 19)3; 17:466.481.

72. Misdarp W, Hart AAM: 3 Nail Cancer Inst 1979: 62:537-545.

73, Sabin FR, Doan CA, Forknet CE:] Eap Med 1932: 56:267.289.

74. Nielson SW: Maligna,,t Bo',e Tumors. Berlin, Springer-Verlag, 1976. pp 3-16.

75. Bed ER: Tumors ml Domestic Animal,. is1 2. Berkeley, University ofCalifon,a. 1978, pp 67.71.

76. Lw 5K. Dorlhnan HD. l'atnaik AK:] Small Anim Pract 1974: 15:141.156.

77. Weichselbaorn R, Epotein J, Little JB: In Vitro 1976: 12:833.836

78. McAllister RM. Gardner MB. Greene AE, etal: Cairn 1971:27:397-402.

79. Thurman GB. Maps CW, Taylor ON, etal: Cancer Des 1973: 33:1604-1607.

80. Mays CW, Dougherty rE, Taylor (iN, et al: Delayed Effects of Bone-Seeking Bathonaarlides Salt lake City. University of Utah Press, 1969. pp 387.408.

81. Thunnan GB, Mays CW. Taylor ON. eta!: Growth 1971: 35:119.125,

82. Taylor DM. Bensted JPM: Delayed Effects of Bone Seeking Ilathomodlides. Salt Lake City. University of Utah Press, 1969, pp 357.370.

83. Cobb LM: Br] Cancer 1970: 24:294-299.

84. Lisco H. F,nkel Ml', Bnaes AM: Radiology 1947; 49:361.363.

85. Bass JM:] Parka1 Bacterial 1936; 43:267-276.

86. Janet JM, Higgins GM. Herrick JF: I Bonejoini Stag 1954: 56B:543-S5l.

87. lanes JM, Higgins GM, Herrick if:) Bone 10Mg Siarg 1956: 38A:809-8!6.

88. Janes JM, Herrick iF, Higgins GM: Maa Clin Bose 1959: 34:47-49.

89. FinEd MP, Jenkins PB, ToRe J, er al: Radiology 1966; 87:333.339.

90. I.evy JA, Hartley JW, Rowe WP. eral:) Nail Cotter foist 1973: 51:525-539.

91. Fsiedlaendet OE, Mitchell MS: Cancer 1975: 36:1631-1639.

92. Noon 0, Stanch 0. Mehraein P. a al: Arch Vend 1981; 69:239-25).

93. Fmnkel MP, Biskis BC Farrell C: Arch Bothol 1967: 64:425-428.

94. Finkel MP. Biskis DO. Farrell C: Radiology 969; 92:1546-1552.

95. Fmnkel MP. Biskis BC Farrell C: P.nc Nail Acad Sci 1968: 60:1223.1230.

96. Finkel MP, ResEts DO: Pnsg in E~er Tumor Ret 1968: 10:72-Ill.

97. Rester W, et aI: Skeletal Radial 1984: 11(33:216-219.

98. Coventry MB, Dahlin Ir:] Bcme hint Sung 1957: 39A:741-758.

99. Bubis JJ: Osteosarco,.sa: New Trends in Dio8,,osis and Treanoenc New York. Alan R. Liar. 982, p 6.

100. Codman EA: Stag Oynecol Olsstei 1922; 34:335-343.

101. C,xlo,an EA: MR 1925: 13:105.126.

102. DeSanros LA, Murray JA, Aysla AG: Cancer 1979: 43:735-744.

103. Stonuby N, Ackennan M: Acta Cytol (Baltimore) 1973~ 17:166-172.

104. Marknez-Tello 9, Navao.PalaciessJj: Cancer 1982: 50:949-961.

105. Katenkaoip D. Stiller D. Waldooann 0: Verb Arch (Radial Anaa) 1978: 381:49.6!.

106. Reddick RI,, Miclselitch HJ. Levine AM. et sI: Cancer 1980; 45:64-71.

107. Dablin DC, Unni KK: Am] Sung Barkal 1977: 1:61-72.

108. Dahlin EC, Unni KK, Mactuna T: Cancer 1977; 30:1598-1516.

109. Schajowic: F: Bone Tumors in Children. Littleton, MA. PSG Publishing Co. 1979, pp 23-60.

110. Matsuno T. Unni KK. McLeod RA. et al: Cancer 1976; 38:2538-2547.

111. Van der Heul RO, von Rönnen JR: ; Bone Joint Sun; 1967; 49A:415-439.

112. Farr GH. Huvos AG: J Bone Joint Surg 1972; 54A:1205-1216.

113. Unni KK, Dahlin DC. Beabout JW. et al: Cancer 1976; 37:2476-2485.

1 14. Cohen P: Eur ) Cancer 1978; 14:995-1004.

115. Bleyer WA, Haas JE, Feigl P, et al: J Bone Joint Sur* 1982; 64B:2 33-238.

116. Sim FH, Unni KK, Beabout JW. et al: J Bone Joint Sur« 1979; 61A:207-215.

117. Stein JJ: AJR 1975; 123:607-613.

118. Levine E. Lee KR. Neff JR, et al: Radiology 1979; 131:431-437.

119. Kean DM. Worthington BS, Preston BJ, et al: BrJ Radiol 1983; 56:355-364

120. Caceres E. Zaharia M, Valdivia S. et al: ImJ Rad Oncol Biol Phys 1984; 10:35-39.

121. Spittle MF. Heal J. Harmer C, et al: Clin Radiol 1968; 19:400.

122. Swaney JJ. Cangir A: J ftdiatr 1973; 82:165.

123. Muhm JR. Pritchard DJ: Proc Am Assoc Cancer Res 1980; 21:148.

124. Papanicolaou N, Kozakewich H. Treves S, et al: J Nucí Med 1982; 23:7.

125. Chew FS, Hudson TM: AJR 1982; 139:49-54.

126. Goldstein H, McNeil BJ. Zufall E, et al: Radiologi 1980; 135:177-180.

127. TealesCD, Brower AC. Williamson BR: Clin Nucí Med 1977; 2:298-302.

128. McNeil BJ: Semi Nucí Med 1978; 8:336-345.

129. McKenna R). Schwinn CP. Soung KY, et al: J Bone Joint Surg 1966; 48A:l-26.

130. Jeffree GM. Price CHG: J Bone Joint Surg 1965; 47B:120.

131. Levine AM. Rosenberg SA: Cancer 1979; 44:2291-2293.

132. Mosende C, Gutierrez M, Caparros B, et al: Cancer 1977; 40:2779-2786.

133. Thorpe WP. Reilly JJ. Rosenberg SA: Cancer 1979; 43:2178-2181.

134. Brostrom LA: Acta Orthop Scand (Suppl) 1979; 183:9-39.

135. Sim FH, IvinsJC. Pritchard DJ: Cancer Treat Rep 1978; 62:205-211.

136. Rosen G: Pediatr Ann 1978; 7:518-530.

137. Brostrom LA. Harris M. Simon M, et al: J Bone Joint Surg 1979; 61B:209-212.

138. Enneking WF. Kagan A: Cancer 1975; 36:2192-2205.

139. Enneking WF, Kagan A: Cancer 1978; 41:1526-1537.

140. Lewis RJ, Lutz MJ: Cancer 1974; 33:371-375.

141. Miller TR: Surg Clin North Am 1974; 54:913.

142. Van Nes CP-. J Bone Jon« Surg 1950. 32B.t2-2Q.

143. Jacobs PA: CÜn Orlhop Surg 1984; 188:217-222.

144. Dahlin DC: Cancer Treat Rep 1978; 62:189-192.

145. Eilber FR, Townsend C. Morton DL: Clin Orthop 1975; 111:94-100.

146. Jaffe N, Watts H. Frei E, et al: Recent Res Cancer Res 1977; 62:114-122.

147. Rosen G, Murphy ML, Huvos AG. et al: Cancer 1976; 37:1-11.

148. Hoover HC, Ketcham AS, Millar RC, et al: Cancer 1978; 41:2475-2480.

149. Rosen G. Marcove RC. Caparros B, et al: Cancer 1979; 43:2163-2177.

150. Benjamin RS: JAMA 1985; 254:1795-1796.

151. Marcove RC: Bull NV Acod Med 1979; 55:744-750.

152. Winkler K, Beron G. Kotz R. et al: J Clin Oncol 1984; 2:617-624.

153. Jaffe N: Clmicai Cancer Brie/s 1983; 5:3-5. 13.

154. Jaffe N, Watts H. Fellows KE. et al: Cancer Treat Rep 1978; 62:217-223.

155. Marcove RC: Cancer Treat Rep 1978; 62:225-231.

156. Sweetnam R: Clin Orthop ReI Res 1975; 111:57-64.

157. lvins JC, Pritchard DJ: Recent Res Cancer Res 1976; 54:221-230.

158. Eilber FR, Grant T, Morton DL: Cancer Treat Rep 1978; 62:213-216.

159. Marcove RC: Can J Surg 1977; 20:521-528.

160. Miller TR: Can J Surg 1977; 20:513-517.

161. Link MP. Goorin AM. Miser AW. et al: N Engl) Med 1986; 314:1600-1606.

162. Jaffe N, Robertson R. Takaue Y. et al: Cancer 1985; 56:461-466.

163. Jaffe N: ftdiarr Clin North Am 1985; 32:801-810.

164. Edmonson JH, Green SJ. lvins JC, et al: J CIm Oncol 1984; 2:152-156.

165. Morgan E. Baum E. Bleyer WA. et al: Cancer Treat Rep 1984: 68:661-664.

166. Risen G. Caporros B. Groshen S: Cancer Invest 1984; 2:181-192.

167. Gilchrist GS. lvins JC, Ritts RE, et al: Cancer Treat Rep 1978; 62:289-294.

168. Etcubanas E, Wilbur JR: Cancer Treat Rep 1978; 62:283-287.

169. Tudway RC: J Bone Joint Surg 1961; 43B.-63-67.

170. Woodard HQ. Coley BL: AJR 1947; 57:464-471.

171. Sweetnan R, Knowelden J. SeddonJ: Br Med J 1971; 2:363-367.

172. Telander RL, Pairolero PC, Pritchard DJ. et al: Surgery 1978; 84:335-341.

173. Breur K. Cohen P, Schweisguth O. et al: Eur J Cancer 1978; 14:461-471.

174. Phillips TL. Scheline GE: Radiology 1969; 92:1537-1545.

175. Beck JC. Wara WM. Bovili EG. et al: Radiology 1976; 120:163-165.

176. Suit H. Lindberg R: AJR 1968; 102:27-37

177. Ruh GT. Ivins JC, Childs DS. et al: Cancer 1976; 38:939-942.

178. Carter SK: J Clin Oncol 1984; 2:147-148.

179. Jaffe N, Abelson HT, Frei E III: J Clin Oncol 1983; 1:251-254.

180. Eilber F. Guiliano A. Eckardt J. et al: J Clin Oncol 1987; 6:21-26.

181. Masuda S: Tukuoka Acta Medica 1980; 71:600.

182. Ito H. Murakami K. Yanagawa T. et al: Cancer 1980; 46:1562-1565.

183. Strander H: Blut 1977; 35:277.

184. Ritts RE, Pritchard DJ. Gilchrist GS, et al: Immunotherapy of Cancer: Present Status ?/ Trials in Man. New York. Raven Press, 1978, pp 293-298.

185. Strander H: Tex Rep Biol Med 1977; 35:429.

186. Neff JR. Enneking WF: J Bone Joint Surg 1972; 54A: 1 367-1 397.

187. Jaffe N, Frei III E. Watts H, et al: Cancer Treat Rep 1978; 62:259-264.

188. Jaffe N. Goorin A. Link M, et al: Cancer Treat Reports 1981; 65:99-106.

189. Jaffe N, Traggis D, Cassady JR. et al: /nt J Rad Oncol Biol Phys 1977; 2:261-266.

190. Sutow WW, Sullivan MP, Fembach DJ, et al: Cancer 1975; 36:1598-1602.

191. Sutow WW, Gehan EA, Dyment PG. et al: Cancer Treat Rep 1978; 62:265-269.

192. Jasmin C: Adjuvant Therapy of Cancer II. New York. Crune & Stratton, 1979, pp 375-382.

193. Cortes EP, Holland JF. Glidewell O: Cancer Treat Rep 1978; 62:271-277.

194. Cortes EP, Holland JF. Wang JJ, et al: N Engl] Med 1974; 291:998-1000.

195. Bullimore JA: Cancer Treat Rep 1981; 65(suppl 1):113-U7.

196. Lange B, Kramer S. Gregg JR. et al: Am J Clin Oncol 1982; 5:3-8.

197. Krailo M, Ertel I. Makley J. et al: Med ftdiatr Oncol 1987; 15:69-77.

198. Jasmin C: Recent Results Cancer Res 1982; 80:108-114.

199. Mavligit GM, Benjamin R, Patt YZ, et al: Cancer 1981; 48:1-4.

200. Kamalakar P. Freeman A, Higby D. et al: Cancer Treat Rep 1977; 61:835-839.

201. Vietti T. Nitschke R, Starling K, et al: Cancer Treat Rep 1977; 63:611-614.

202. Ochs )). Freeman Al. Douglass HO. et al: Cancer Treat Rep 1978; 62:239-245.

203. Rosen G, Nirenberg A, Juergens H. et al: Proc Am Soc Clin Oncol 1979; 20:363 (#C-299 Abstr).

204. Baum E. Gaynon P, Greenberg L. et al: Cancer Treat Rep 1979; 63:1621-1627.

205. Gasparini M, Rouesse J, Van-Oosterom A, et al: Cancer Treat Rep 1985; 69:211-213.

206. Freeman A, Ettinger L. Brecher M: Cancer Treat Rep 1979: 63:1615-1620.

207. Ettinger LJ, Douglas HQ Mindel ER, et al: J Clin Oncol 1986; 4:353-362.

208. Mills R, Mauere L. Foslier R. et al: Cancer Treat Rep 1977; 61:477-479.

209. Jaffe N, Kcifer R. Robertson R, et al: Cancer 1987; 59:1577-1581.

210. Jürgens H. Beron G, Winkler K: J Cancer Res Clin Oncol 1983: 106(suppl):14-18.

211. Miser J. KinsellaT. TricheT,etal:ProcAmSocClinOncoll987;6:218(Abstr#859).

212. Rosen G, Caparros B. Huvos AG, et al: Cancer 1982; 49:1221-1230.

213. Rosen G. Caparros B. Huvos AG, et al: Cancer 1982; 49:1221-1230.

214. Smith J. Heelan RT, Huvos AG. et al: Radiology 1982; 143:355-360.

215. Rosen R: Pediatr CIm North Am 1976; 23:183-213.

216. Rosen G: Recent Results in Cancer Research 103. Berlin-Heidelberg, Springer- Verlag, 1986, pp 148-157.

217. Rosen G, Nirenberg A: Cancer Treat Rep 1982; 66:1687-1697.

218. Baca G. Picei P. Gherlinzoni F, et al: Chemoterapia 1986; 5:140-143.

219. Jaffe N, Knapp J, Chuang VP. et al: Cancer 1983. 51:402-407.

220. Epelman S. Petrilli AS, Franco EL: Proc Am Soc CIm Oncol 1987; 6:223 (Abstr).

221. Shirkhoda A, Jaffe N. Wallace S. et al: AJR 1985; 144:95-99.

222. Jaffe N, Robertson R, Ayala A, et al: J Clin Oncol 1985; 3:1101-1104.

223. Stephens FO, Tettersall MH. Marsden N. et al: Cancer 1987; 60:724-735.

224. Shimizu H. Jaffe N. Kleinerman E: J Neurol Oncol 1986; 4:165-167.

225. Sundaresan N. Galicich JH. Lane JM: J Neumsurg 1984; 60:282-286.

226. Das L. Farooki ZQ. Hakami MN. et al: Med Pediatr Oncol 1983; 11:164-166.

227. Ayres R, Curry NS, Gordon L, et al: Urol Radiol 1985; 7:39-41.

228. Spanos PK, Payne WS. lvins JC, et al: J Bone Joint Surg 1976; 58A:624-628.

229. Martini N, Huvos AG, Mike KV. et al: Ann Thorac Surg 1971; 12:271-280.

230. Giuliano AE. Feis S, Eilber FR: Cancer 1984; 54:2160-2164.

231. Derstappen T. Roessner A, Müller KM. et al: J Cancer Res Clin Oncol 1987; 113:241-248.

232. Marcove RC. Martini N. Rosen G: Clin Orthop 1975; 111:65-70.

233. Jaffe N, Frei E IH, Traggis D. et al: Cancer 1977; 39:45-50.

234. Giritsky AS. Etcubanas E. Mark JBD: J Thorac Cardioiusc Surg 1978; 73:354-362.

235. Huang MN, Edgerton F. Takita H. et al: AmJ Surg 1978; 135:804-806.

236. Rosenberg SA, Flye MW. Conkle D. et al: Cancer Treat Rep 1979. 63:753-756.

237. Marcove RC. Rosen G: Cancer 1980; 45:3040-3044.

238. Meyer WH, Schell MJ. Kumar AP. et al: Cancer 1987; 59:374-379.

239. Goorin AM. Delorey MJ, Lack EE. et al: J CIm Oncol Y9&4·, 2:425-431.

240. Schaller RT Jr. Haas J. Schaller J. et al: J Pediatr Surg 1982; 17:546-550.

241. Moersch R. Clagett O: Surgery 1961; 50:579-585.

242. Takita H. Edgerton F, Karakousis C. et al: Surg Gynecol Obstet 1981; 152:191-194.

243. Feldman PS. Kyriakos M: J Thorac Cordiovasc Surg 1972; 64:784-799.

244. Thomford N. Woolner L. Clagett OT: J Thorac Cardiouasc Surg 1965; 19:357-363.

245. Regal AM, Reese P, Antkowiak J, et al: Cancer 1985; 55:1334-1339.

246. Meng RL, Jensik RS, Kittle F, et al: J Thorac Cardiotasc Surg 1980; 80:1-7.

247. Shah A. Exelby PR, Rao B. et al: J Ped Surg 1977; 12:983-990.

248. Huang MN. Takita H. Douglass HO: J Surg Oncol 1978; 10:179-182.

249. Jaffe N: CanJ Surg 1977; 20:537-539.

250. Boyle M. Tebbi CK. Mindell ER. et al: Med ftdiotr Oncol 1982; 10:301-312.

251. Tebbi CK, Malion JC: J Psychosoc Oncol. Winter. 1987.

TABLE 1

Single Agent Useful In Therapy of Osteosarcoma

10.3928/0090-4481-19880401-08

Sign up to receive

Journal E-contents