INTRODUCTION — Osteosarcomas are primary malignant tumors of bone that are characterized by the production of osteoid or immature bone by the malignant cells. Osteosarcomas are uncommon; approximately 750 to 900 cases are diagnosed each year in the United States, of which 400 are in children and adolescents under the age of 20 [1,2]. (See "Osteosarcoma: Epidemiology, pathology, clinical presentation, and diagnosis".)
This topic review will cover the use of adjuvant and neoadjuvant chemotherapy and radiation therapy (RT) in the management of osteosarcoma. The same principles apply to other primary bone tumors, such as fibrosarcoma and undifferentiated high-grade pleomorphic sarcoma (previously referred to as malignant fibrous histiocytoma of bone), and they are treated similarly [3,4]. On the other hand, primary bone angiosarcomas do not behave clinically like other primary bone tumors, and these patients are treated according to the principles of soft tissue sarcoma rather than osteosarcoma. (See "Overview of multimodality treatment for primary soft tissue sarcoma of the extremities and superficial trunk" and "Adjuvant and neoadjuvant chemotherapy for soft tissue sarcoma of the extremities".)
The surgical management of patients with primary bone tumors, the clinical features, epidemiology, diagnosis, pathology, and pathogenesis of osteosarcoma, the management of osteosarcomas arising in the head and neck, the management of chondrosarcomas in general and of chordomas and chondrosarcomas arising in the skull base, and the treatment of Ewing sarcoma are addressed separately. (See "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management" and "Osteosarcoma: Epidemiology, pathology, clinical presentation, and diagnosis" and "Head and neck sarcomas", section on 'Osteosarcoma' and "Chondrosarcoma" and "Chordoma and chondrosarcoma of the skull base" and "Treatment of Ewing sarcoma".)
Rarely, osteosarcomas occur in soft tissue. There are conflicting data on their management as soft tissue sarcomas or osteosarcoma of bone [5-7]. These tumors are treated as soft tissue sarcomas at some institutions, with surgery and radiation alone or surgery, radiation, and soft tissue-based chemotherapy. At other institutions, they are treated as osteogenic sarcomas, with surgery, radiation, and doxorubicin plus cisplatin with or without methotrexate. The choice of chemotherapy is best determined by the availability of clinical trials and, otherwise, is decided on a patient-by-patient basis, balancing the risks and benefits of chemotherapy for this high-risk diagnosis.
PROGNOSIS AND EVOLUTION OF TREATMENT — Chemotherapy is a standard component of most osteosarcoma treatment, both in children and in adults. The choice of regimen and the optimal timing (ie, preoperative versus postoperative) are controversial; however, many centers preferentially utilize preoperative chemotherapy, particularly if a limb-sparing procedure is being contemplated for an extremity osteosarcoma. (See 'Neoadjuvant versus adjuvant chemotherapy' below and "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management".)
The survival of patients with malignant bone sarcomas improved dramatically after the introduction of systemic chemotherapy. Chemotherapy can eradicate tumor deposits if it is initiated when disease burden is low. Prior to the use of systemic therapy for conventional osteosarcoma, 80 to 90 percent of patients developed metastases, despite achieving local tumor control, and died of their disease. It was subsequently demonstrated that the majority of patients had subclinical metastatic disease present at the time of diagnosis [8-10]. The benefits of chemotherapy are best illustrated by a systematic review of the literature, which showed that long-term survival after local tumor control without chemotherapy was only 16 percent (95% CI 9-23 percent) [11]. In contrast, the addition of systemic chemotherapy with three or more drugs provided a five-year overall survival rate of 70 percent.
Despite an active search for new agents, the standards of care for neoadjuvant or adjuvant therapy have not changed significantly. Nonetheless, with multimodality therapy, approximately two-thirds of children, adolescents, and adults under the age of 40 with nonmetastatic extremity osteosarcomas will be long-term survivors and presumably cured of their disease. Retrospective studies have demonstrated worse prognosis in patients with a primary tumor in the pelvis or axial skeleton, age greater than or equal to 40, metastatic disease at diagnosis, or a limb osteosarcoma involving more than one third of the bone or located in the proximal bone [12,13].
Long-term survival can be expected in less than 20 percent of all other patients who present with or develop overt metastatic disease; however, up to 50 percent of those with limited pulmonary metastases may be cured with multimodality therapy. In most (but not all [14]) series, prognosis is worse for adults than for children [15-18].
A nomogram predicting metastasis-free and overall survival for patients undergoing neoadjuvant chemotherapy and surgery for nonmetastatic disease has been developed and validated [19].
OVERVIEW OF PRIMARY MANAGEMENT — Surgery and systemic chemotherapy are the mainstays of treatment for patients with nonmetastatic osteosarcoma.
Local therapy
●Surgery is a standard of care for management of osteogenic sarcoma. For most patients, limb-sparing surgery is as effective as is amputation. As is outlined below, chemotherapy is usually given prior to surgery (neoadjuvant therapy) [20]. (See "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management".)
●The most challenging primary sites for local control include the sacrum and base of skull. In some cases, if patients decline surgery or if surgical morbidity relative to a patient's comorbidities does not permit surgery, definitive radiation can be an option. Intensity-modulated radiation therapy is the standard of care in this scenario, while proton beam irradiation can be an option, particularly for younger patients, to reduce the radiation dose to noninvolved normal tissues [21]. (See "Radiation therapy techniques in cancer treatment", section on 'Intensity-modulated radiation therapy' and "Radiation therapy techniques in cancer treatment", section on 'Proton beam'.)
●There is no role for adjuvant radiation therapy, except in patients with incompletely resected osteosarcomas and possibly in the rare patient with a small cell osteosarcoma. (See 'Radiation therapy' below.)
Chemotherapy
●Historically, more than 80 percent of patients with osteosarcoma treated with surgery alone developed metastatic disease, despite achieving local control. It is presumed that subclinical metastases are present at diagnosis in the majority of patients. The survival of patients with malignant bone sarcomas has improved dramatically over the past 50 years, largely owing to chemotherapeutic advances. Adjuvant chemotherapy significantly improves survival compared with surgery alone and is a standard component of therapy for both children and adults.
●The optimal timing of chemotherapy (ie, preoperative versus postoperative) has not been established. There is no definite survival benefit for neoadjuvant as compared with adjuvant chemotherapy [20], but many centers preferentially utilize preoperative chemotherapy, particularly if a limb-sparing procedure is being contemplated for an extremity osteosarcoma. An important point is that chemotherapy is not a substitute for sound surgical judgment when assessing the need for amputation. It is reasonable to proceed to immediate surgery followed by adjuvant chemotherapy if the nature of the resection would not necessarily be influenced by a good response to chemotherapy. (See 'Neoadjuvant chemotherapy' below.)
●Response to neoadjuvant chemotherapy is a major prognostic factor, but there is no evidence that outcomes in poor histologic responders to neoadjuvant chemotherapy can be improved by altering the postoperative chemotherapy regimen [22]. (See 'The option to change systemic therapy after neoadjuvant chemotherapy and surgery' below.)
●The optimal regimen has not been established; however, the available evidence supports the benefit of a three-drug as compared with a two-drug regimen, particularly for children and younger adults [11].
For children and adolescents, we recommend the methotrexate plus doxorubicin and cisplatin (MAP) regimen as was used in the control arm of the American Osteosarcoma Study Group (AOST) 0331 (EURAMOS-1) protocol (table 1) [23]. If preoperative therapy is chosen, we administer 10 weeks of chemotherapy prior to surgery and continue postoperative chemotherapy for 29 weeks, beginning one week postoperatively. Where available, mifamurtide (muramyl tripeptide phosphatidylethanolamine [MTP-PE]) can be used in addition to MAP chemotherapy. (See 'MAP (methotrexate, doxorubicin, and cisplatin) as a standard regimen' below and 'Mifamurtide' below.)
For adult patients up to age 40 years, we endeavor to use the same regimen as used in children and adolescents. For patients 40 years and older, in whom the biology of the tumor may be somewhat different, we generally employ doxorubicin and cisplatin only, although a methotrexate-containing regimen is also a reasonable approach. (See 'Chemotherapy in adults' below.)
PRINCIPLES OF CHEMOTHERAPY
Neoadjuvant chemotherapy — Initially, postoperative chemotherapy was used, and five-year survival rates rose from less than 20 percent to between 40 and 60 percent in the 1970s [24]. Two subsequent randomized studies conducted in the 1980s demonstrated significant relapse-free and overall survival benefits for adjuvant chemotherapy that persisted over time [25-28], although the trials were limited in size, and the survival benefits were modest. The chemotherapy regimens used in these studies included high-dose methotrexate (HDMTX) plus doxorubicin, bleomycin, cyclophosphamide, and dactinomycin, and either vincristine [25] or cisplatin [26,27].
The concept of induction or neoadjuvant chemotherapy arose in concert with the evolving use of limb-sparing surgery. Because of the time needed for fabrication of custom metallic endoprostheses, chemotherapy was often given while awaiting definitive surgery [29]. The ability to function as an in vivo drug trial to determine the drug sensitivity of an individual tumor and to customize postoperative therapy is one of the most compelling rationales for neoadjuvant chemotherapy.
Responsiveness of an osteosarcoma to neoadjuvant chemotherapy is a major determinant of clinical outcome for most histologic subtypes [30-37]. Five-year survival rates for patients with an extremity sarcoma and a "good" response to chemotherapy (as defined by 90 percent or more necrosis in the surgical specimen) are significantly higher than for those with a lesser response (71 to 80 versus 45 to 60 percent, respectively) [33,35,36,38] (see 'The option to change systemic therapy after neoadjuvant chemotherapy and surgery' below). However, the demonstration of an inverse relationship between the predictive value of tumor necrosis and the intensity of induction therapy in one report has led to questions about the true value of histologic response as a prognostic marker [39].
Certain subtypes, such as chondroblastic osteosarcoma, have lower rates of necrosis, yet there are no differences in outcome between good and poor responders to neoadjuvant chemotherapy [36]. For all others, histologic response to chemotherapy is correlated with prognosis (table 2) [33,36]. In one large series in which 1058 patients received presurgical chemotherapy for osteosarcoma over a 20-year period the likelihood of a "good" response (>90 percent necrosis) was significantly higher for fibroblastic and telangiectatic osteosarcomas (83 and 80 percent, respectively) as compared with chondroblastic (43 percent); they were intermediate for osteoblastic osteosarcomas (58 percent) [33]. Five-year survival rates paralleled the quality of the response to induction chemotherapy (83, 75, 62, and 60 percent for fibroblastic, telangiectatic, osteoblastic, and chondroblastic osteosarcomas, respectively).
Response to chemotherapy is usually defined by a histologic appearance in the resected specimen (picture 1). A grading system for assessing the effect of preoperative chemotherapy on the tumor was developed at Memorial Sloan Kettering Cancer Center (MSKCC) and is in widespread use (table 3) [24].
The option to change systemic therapy after neoadjuvant chemotherapy and surgery — As discussed in more detail above, patients with a near-complete absence of viable tumor cells in the resection specimen after neoadjuvant therapy tend to do well when the same therapy is continued after surgery [24,30-32,40]. Attempts to improve outcomes with additional therapy, such as with interferon alfa, have not improved overall survival [41]. (See 'Addition of ifosfamide-based therapy: The EURAMOS-1 trial' below.)
However, outcomes are worse when the tumor contains 10 percent or more residual viable cells after neoadjuvant chemotherapy.
Because of this, some have hypothesized that a change in the chemotherapeutic regimen might be beneficial. A strategy of altering the postoperative chemotherapy regimen based upon the response to the neoadjuvant regimen was pioneered in the T10 protocol at MSKCC [24,40]. This study suggested that altering postoperative chemotherapy for poor responders is beneficial, a result confirmed by at least one other group [42,43].
However, this strategy of intensifying chemotherapy in response to less-than-optimal histologic response is no longer commonly used for the following reasons:
●With more mature follow-up, the initial beneficial results from the T10 protocol were not sustained [44].
●Major cooperative groups (eg, Children's Cancer Group, German Society for Pediatric Oncology, Scandinavian Sarcoma Group) have been unable to confirm that altering the postoperative chemotherapy regimen in patients with a less-than-complete response to presurgical chemotherapy improves their overall outcome [45-49].
●The international multigroup EURAMOS-1 trial, which examined the benefit of switching to ifosfamide and etoposide (I/E) versus continuing chemotherapy with methotrexate plus doxorubicin and cisplatin (MAP) for poor histologic responders after neoadjuvant MAP chemotherapy, failed to demonstrate a benefit to switching to I/E [22]. (See 'Addition of ifosfamide-based therapy: The EURAMOS-1 trial' below.)
For patients treated outside of a research protocol who received neoadjuvant chemotherapy, we recommend against using the histologic response to guide selection of the chemotherapy regimen to be given after surgery. Even in this less favorable setting, patients who receive chemotherapy have a better outcome than those who do not. These data imply a difference in the sensitivity of metastatic versus primary disease.
Neoadjuvant versus adjuvant chemotherapy — Positive outcomes after neoadjuvant chemotherapy ultimately led to a randomized clinical study conducted between 1986 and 1993 by the Pediatric Oncology Group (POG) trial 8651 that compared immediate surgery and postoperative chemotherapy versus 10 weeks of the same chemotherapy regimen followed by surgery in 100 patients under the age of 30 with nonmetastatic high-grade osteosarcoma [20]. Chemotherapy consisted of alternating courses of HDMTX with leucovorin rescue, cisplatin plus doxorubicin, and bleomycin, cyclophosphamide, and dactinomycin (BCD). The five-year relapse-free survival rates were similar between the two groups (65 versus 61 percent for adjuvant and neoadjuvant therapy, respectively), as was the limb-salvage rate (55 and 50 percent for immediate and delayed surgery, respectively).
The study was criticized for the relatively low rate of limb-sparing surgery in both groups (by modern standards) and the inclusion of BCD as a component of the regimen. The contribution of BCD to the therapeutic efficacy of this regimen is unclear, while it can clearly contribute to long-term bleomycin-related pulmonary toxicity (see "Bleomycin-induced lung injury"). Nevertheless, this trial established that survival was similarly improved by either presurgical or postsurgical chemotherapy and established a benchmark outcome for future studies (five-year event-free survival [EFS] 65 percent).
Due to its success in killing cancer cells (although actual tumor shrinkage during treatment is not common [50], particularly with chondroblastic osteosarcomas), neoadjuvant chemotherapy has evolved to a method of increasing the proportion of patients who are suitable candidates for limb-salvage surgery. The majority of limb-sparing surgical procedures for extremity osteosarcomas are now performed at institutions using presurgical chemotherapy.
While it is clear that the number of patients with osteosarcoma who are deemed suitable candidates for limb-sparing surgery has increased in parallel with the increasing use of presurgical chemotherapy, this finding may reflect improvements in reconstructive techniques and the increased experience and confidence of tumor surgeons rather than a benefit attributable to the use of induction chemotherapy. A major concern in this regard is the possibility that inexperienced surgeons may expand selection criteria to accommodate inappropriate candidates who might be better served by amputation. Neoadjuvant chemotherapy is never a substitute for sound surgical principles.
Adjuvant chemotherapy for osteosarcomas arising in the head and neck is addressed in detail elsewhere. (See "Head and neck sarcomas", section on 'Chemotherapy'.)
Choice of systemic therapy
MAP (methotrexate, doxorubicin, and cisplatin) as a standard regimen — There is no worldwide consensus on a standard chemotherapy approach for osteosarcoma. The development of adjuvant chemotherapy has been largely empiric, with the majority of regimens incorporating doxorubicin and cisplatin with or without HDMTX (6 to 12 g/m2 with leucovorin rescue), called the MAP regimen (table 1) [26,27,43,47,51-54]. (See "Therapeutic use and toxicity of high-dose methotrexate".)
Carboplatin versus cisplatin — For patients who demonstrate intolerance of HDMTX and for institutions that cannot provide pharmacokinetic monitoring for HDMTX, the combination of carboplatin, ifosfamide, and doxorubicin appears to be a reasonable alternative. Investigators at St. Jude Children's Research Hospital have demonstrated good outcomes (five-year EFS and overall survival rates 66 and 75 percent) with a non-methotrexate-containing chemotherapy regimen consisting of carboplatin plus ifosfamide and doxorubicin [54]. However, the increased doses of alkylating agents may possibly increase the risk of second malignancies (leukemia).
Addition of ifosfamide-based therapy: The EURAMOS-1 trial — The activity of I/E in metastatic osteosarcoma brought the issue of modifying postsurgical treatment based upon response to induction therapy to the forefront [42,43,55,56]. A meta-analysis concluded that adding ifosfamide to neoadjuvant MAP chemotherapy did not increase the histologic response rate, five-year EFS, or overall survival [57]. However, the high level of antitumor activity of I/E in patients with measurable metastatic osteosarcoma (66 percent in one trial [55]) also prompted the investigation of adding postoperative I/E to MAP in patients with resectable disease who have a poor initial histologic response to standard chemotherapy.
An international multigroup trial (AOST 0331, the EURAMOS-1 trial) was developed to test prospectively the benefit of altering postoperative chemotherapy based upon the response to initial chemotherapy. (See 'The option to change systemic therapy after neoadjuvant chemotherapy and surgery' above.)
In this trial, patients who had a poor histological response (ie, ≥10 percent viable tumor) to standard induction chemotherapy (two or more cycles of MAP) were randomly assigned to MAP with or without I/E after resection, while those with a good histologic response after induction chemotherapy were randomly assigned to continued MAP with or without the addition of pegylated interferon alfa-2b as maintenance therapy for two years. The following results are available from the EURAMOS-1 trial:
●There was no benefit for the addition of interferon in good responders [41].
●The 618 patients with a poor histological response after preoperative MAP chemotherapy were randomized to receive either continued MAP chemotherapy alone or MAP plus I/E (MAPIE) [22]. At a median follow-up of 62 months, there was no benefit to the addition of I/E in terms of EFS, the primary endpoint (HR 0.98, 95% CI 0.78-1.23). Because hazards were not proportional over time, EFS was estimated using the restricted mean survival time approach, which measures the mean time to a first event (local recurrence, new or progressive metastases, second malignancy, death, or a combination) six years after randomization. This revealed that the time to first event in MAP versus MAPIE patients was not significant (44.1 versus 43.3 months). The three-year EFS for MAP and MAPIE were 55 and 53 percent, respectively. Patients receiving MAPIE had significantly more febrile neutropenia without documented infection (73 versus 50 percent) and more grade 4 nonhematologic toxicity than those receiving MAP (24 versus 12 percent).
These results do not support the addition of I/E to postoperative MAP chemotherapy in patients who have a poor response to induction MAP chemotherapy. While the results with chemotherapy as initiated are not as good in a patient with a poor response to chemotherapy as in one with a good response to chemotherapy, outcomes are overall better with chemotherapy than without it, arguing for completion of the regimen that had been started. Better methods are needed to prospectively identify chemoresistant tumors at diagnosis. Regardless of the regimen chosen, based on available evidence, in patients with localized disease of an extremity, we suggest resuming chemotherapy within 21 days after definitive surgery, when feasible [58].
Role of high-dose methotrexate — The role of high-dose methotrexate (HDMTX) in chemotherapy for osteosarcoma has been questioned, particularly in adults:
●Nearly all of the trials purporting to show benefit for chemotherapy regimens that include HDMTX are phase II trials that have been conducted predominantly in children, although many have enrolled patients up to age 40 [11].
●Neither of the two randomized studies comparing HDMTX plus doxorubicin and cisplatin versus doxorubicin/cisplatin alone have shown an advantage to three-drug therapy [48,53]. However, one of these studies was criticized because the outcome in the group receiving HDMTX, cisplatin, and doxorubicin was particularly poor by modern standards (41 percent five-year disease-free survival compared with 57 percent with doxorubicin/cisplatin alone) [48].
●A single randomized trial comparing higher as compared with intermediate doses of methotrexate did not show a survival advantage for high-dose therapy [59]. However, a benefit for HDMTX is supported by at least one phase II trial demonstrating a superior outcome with high-dose as compared with intermediate-dose methotrexate in the context of a multiagent chemotherapy regimen [34]. Furthermore, many studies have shown a correlation between peak serum levels of methotrexate, tumor response, and outcome [46,60-62]. Thus, it is possible that determining a benefit for HDMTX has been compromised by the use of insufficient doses [63] or administration schedules.
●Additional information comes from a literature-based systematic review of chemotherapy trials for localized high-grade osteosarcoma [11]. In an analysis of 18 trials, almost all conducted in patients under the age of 40, treatment with three or more drugs (including MAP) was associated with a significant improvement in both EFS (HR 0.701, 95% CI 0.615-0.799) and overall survival (HR 0.792, 95% CI 0.677-0.926) compared with a two-drug regimen. However, only two of the trials included in this analysis were randomized trials comparing MAP versus a two-drug regimen.
The role of HDMTX in chemotherapy for osteosarcoma requires further study [64]. Until further information is available, we and others still consider a methotrexate-containing regimen to represent a standard approach for children and adults age 40 or younger.
The optimal methotrexate-containing regimen is not established. However, we consider the control arm of the American Osteosarcoma Study Group (AOST) 0331 (EURAMOS-1) trial (MAP chemotherapy) (table 1) to represent a reasonable and appropriate standard treatment [22]. At least some data suggest that substitution of carboplatin for cisplatin is associated with inferior long-term outcomes, particularly for patients with metastatic disease at diagnosis [65].
Older adults are more commonly offered the combination of doxorubicin and cisplatin, although MAP chemotherapy (table 1) is also a reasonable standard of care in this population, recognizing that adults do not clear MTX as efficiently as children. The principal practical issue in older adults is that slow clearance of MTX can delay the administration of a subsequent cycle of doxorubicin-cisplatin, thus reducing dose intensity and negatively impacting outcome. (See 'Chemotherapy in adults' below.)
Practical tips on administration of HDMTX, including the rationale for leucovorin rescue, and prevention and management of prolonged high plasma methotrexate levels, is discussed in detail elsewhere. (See "Therapeutic use and toxicity of high-dose methotrexate".)
Mifamurtide — The benefit of ifosfamide in conjunction with the liposomal formulation of the immune stimulant muramyl tripeptide phosphatidylethanolamine (MTP-PE; mifamurtide) was evaluated in a randomized phase III trial (INT-0133) of 677 patients (age up to 30 years) with nonmetastatic osteosarcoma [66]. All patients received MAP and were randomly assigned in a two-by-two factorial design to receive or not receive ifosfamide and then to receive or not receive liposome encapsulated mifamurtide. At median follow-up of approximately eight years, the addition of mifamurtide improved overall survival (78 versus 70 percent at six years; HR 0.71, 95% CI 0.52-0.96) [66].
However, the results of this trial are complex for several reasons. Importantly, the data were initially analyzed and then reanalyzed when more complete survival data could be assembled, leading to two separate publications of the same dataset with somewhat different results. The initial published report stated that there was an interaction between ifosfamide and mifamurtide that prevented the results from being analyzed as originally intended [67]. In addition, the improvement in EFS (67 versus 61 percent) with mifamurtide did not reach statistical significance [66], which is somewhat unusual in an oncology study in which the intervention leads to a statistically significant improvement in overall survival; such findings are occasionally seen in immuno-oncology trials, in particular with agents involving vaccines.
As a result of these data, further studies are necessary prior to including mifamurtide as a standard treatment for osteosarcoma [68]. Although mifamurtide has received regulatory approval in Europe and some other countries due to the overall survival benefit seen in one phase III trial [66], further development of this agent (and drug approval in the United States) may prove difficult given a "not approvable" letter for use of mifamurtide in the adjuvant setting from the US Food and Drug Administration (FDA) resulting from the presentation of the initial data, not the follow-up dataset, to the FDA.
Chemotherapy in adults — Intensive treatment with chemotherapy and resection is warranted in adults since osteosarcoma is a potentially curable tumor [16,69]. Similar to children, the lack of a near-complete response to neoadjuvant chemotherapy predicts a poor prognosis [69].
In many (but not all [70]) series, adults, especially older adults, have a worse prognosis than do children with osteosarcoma. This was shown in a population-based series from the Surveillance, Epidemiology, and End Results (SEER) database of the National Cancer Institute [17]. Of the 3482 cases of osteosarcoma reported between 1973 and 2004, there were 1855 cases in the 0 to 24 age group, 974 cases in adults 25 to 59 years of age, and 653 in adults 60 to ≥85 years of age. Five-year survival rates in the three age groups were 62, 59, and 24 percent, respectively. When broken down by decade of age, five-year survival rates for adults in their 50s, in their middle to late 60s, and 80 to 84 years were 50, 17, and 11 percent, respectively.
The optimal chemotherapy regimen for adults (at least those over the age of 40) is not established; almost all of the trials included in the systematic review discussed above were limited to individuals age 40 or younger [11]. Only one trial enrolled older adults up to the age of 65 (range 14 to 62; median 42 years), and it failed to show a benefit for the addition of HDMTX to cisplatin and doxorubicin [53].
Older adults are most often offered doxorubicin plus cisplatin (in fit patients, doxorubicin 25 mg/m2 per day on days 1 to 3, cisplatin 100 mg/m2 on day 1 every three weeks for six cycles) (table 4) [3,48,53]. However, the dose and tolerability of high-dose cisplatin and the role of HDMTX remain unanswered questions. A methotrexate-containing regimen is a reasonable standard of care in this population, if patients can tolerate it. Options include a five-week cycle of cisplatin (100 mg/m2 day 1) and doxorubicin (25 mg/m2 days 1 to 3), followed by two weekly doses of HDMTX (6 to 12 g/m2 with leucovorin rescue), with three cycles administered preoperatively and three postoperatively [66], or the control arm of the AOST 0331 protocol (MAP chemotherapy) (table 1) [23], recognizing that dose reductions may be necessary in older adults. The slower clearance of MTX in adults, and mucositis or other toxicity frequently make the sequential weekly administrations of MTX unfeasible in older adults in particular. Monitoring renal function before and during treatment is paramount in such patients to minimize the risk of irreversible renal failure. (See "Treatment protocols for soft tissue and bone sarcoma".)
RADIATION THERAPY — In contrast to Ewing sarcoma, conventional osteosarcoma is believed to be relatively resistant to radiation therapy (RT), although the small cell variant may be more radiosensitive [71]. Primary RT is usually inadequate to achieve local control, particularly for bulky tumors; surgery is preferred, if possible. (See "Treatment of Ewing sarcoma", section on 'Local treatment' and "Bone sarcomas: Preoperative evaluation, histologic classification, and principles of surgical management".)
With the improvements in RT techniques over time, there has been renewed interest in the use of RT for patients whose tumors respond to chemotherapy and in whom surgery would be debilitating. One report described a five-year local control rate of 56 percent among 31 patients with nonmetastatic extremity osteosarcoma who refused surgery and were instead treated with RT (median dose 60 Gy) [72]. The five-year metastasis-free survival rate was 91 percent, and there were no local failures among the 11 patients who responded well to chemotherapy (ie, had both a radiographic and biochemical response with normalization of serum alkaline phosphatase). Among patients who achieved local control, 86 percent had "excellent" limb function. However, this single study does not represent sufficient data to replace surgery with RT in patients with resectable tumors. RT also cannot substitute for inadequate surgery.
For patients with tumors in challenging axial locations (skull base, spine, sacrum), RT may be a local control option when surgery is not performed. One series reported local control in 72 percent of such cases at five years using proton beam-based irradiation to a mean dose of 68.4 Gy [73]. The carbon ion facility in Chiba, Japan reported a local control of 62 percent in 78 patients who had medically inoperable osteosarcoma of the trunk and who received treatment with carbon ion RT between 1996 and 2009 [74].
Adjuvant radiation therapy — Prophylactic whole-lung radiation has been used in an attempt to improve outcomes following surgery for nonmetastatic localized disease; however, it is not effective in the absence of systemic chemotherapy [75-77]. Furthermore, in patients treated with effective surgery and chemotherapy, adjuvant radiation does not improve survival and increases the risk for secondary tumors; it should be evaluated as an option only in the setting of an unresectable or incompletely resected primary tumor [21].
The benefit of RT for local control may need to be readdressed in view of the greater utilization of radiation techniques such as intensity-modulated RT, proton beam irradiation [73], and carbon ion RT [74].
PATIENTS WITH METASTATIC DISEASE AT DIAGNOSIS — Patients who present with overtly metastatic osteosarcoma have a poor prognosis; long-term survival rates with standard chemotherapy and surgery range from 10 to 50 percent [14,37,78-80]. This is in contrast to patients with apparently localized disease at presentation, two-thirds of whom will achieve long-term survival with appropriate therapy.
Analogous to the situation with primary osteosarcomas, the ability to control all foci of macroscopic disease is an essential element for successful treatment [35,81,82]. The minority of patients with overt metastatic disease who achieve long-term survival and are presumably cured have usually been treated with a combination of surgery, chemotherapy, and sometimes radiation therapy (RT) [14,80,83-85].
The location of the metastases is of prognostic importance. Patients with only pulmonary disease appear to have a chance of long-term event-free survival (EFS) on the order of 20 to 30 percent [84,86]. In contrast, most series report a dismal prognosis for patients with bone metastases [65,87,88], with the exception of a single study that added etoposide and high-dose ifosfamide to standard chemotherapy [55]. (See 'Choice of chemotherapy' below.)
Although it might be argued that pulmonary lesions are more susceptible to chemotherapy than are bone metastases, there is also a tendency to surgically pursue pulmonary lesions more aggressively. In some studies, long-term survival correlates with the number of pulmonary nodules at diagnosis and the ability to successfully resect those that persist after chemotherapy [14,55,84,85,89-91]. (See "Surgical resection of pulmonary metastases: Outcomes by histology" and "Surgical resection of pulmonary metastases: Benefits, indications, preoperative evaluation, and techniques".)
As noted previously, patients with bone metastases do less well than patients with lung-only metastatic disease. There are some long-term survivors among patients in whom bone disease has been treated aggressively (usually with chemotherapy in addition to complete resection, sometimes with irradiation) [55,79,88,92], but there are no prospective data on the question.
Choice of chemotherapy — Optimal management for patients who present with metastatic osteosarcoma has not been defined by randomized clinical trials, and thus, there is no single standard approach. The most active drugs in patients with measurable disease (high-dose methotrexate [HDMTX], doxorubicin, cisplatin, ifosfamide) have single-agent response rates between 20 and 40 percent [65,87,89,93-95]. Response rates are higher with multiagent regimens, but a lower proportion of patients treated for metastatic disease show a good histological response to neoadjuvant chemotherapy as compared with those with apparently localized disease [60,83,96]. This suggests an underlying difference in the biological behavior.
In an effort to improve outcomes, the Children's Oncology Group (COG) and others have utilized a strategy of applying novel agents to patients with newly diagnosed metastatic disease prior to standard therapy (termed the "therapeutic window" approach) [55,65,85,89].
Using this approach, the Pediatric Oncology Group (POG; one of the predecessors of COG) identified the combination of ifosfamide/etoposide as effective induction therapy, particularly for those with metastatic bone disease. In one report, 43 patients under the age of 30 with measurable metastatic osteosarcoma at diagnosis (28 lung only, 12 with metastatic bone involvement with or without lung metastases) received two three-week courses of etoposide and high-dose ifosfamide (17.5 g/m2 per cycle) with hematopoietic growth factor support, followed by surgery and an additional 34 weeks of "continuation therapy" [55]. Postoperative therapy consisted of ten courses of HDMTX with leucovorin rescue, four courses of doxorubicin/cisplatin, one course of doxorubicin alone, and three additional courses of etoposide plus lower-dose ifosfamide [55].
Treatment-related toxicity was prominent and included 83 percent grade 4 neutropenia, 24 percent sepsis, 29 percent grade 4 thrombocytopenia, and five patients with Fanconi syndrome. (See "Etiology and diagnosis of distal (type 1) and proximal (type 2) renal tubular acidosis".)
However, the overall response rate was 59 percent, and it was 80 percent for patients with synchronous bone metastases. The projected two-year progression-free survival rates were 34 and 58 percent for patients with lung-only and bone involvement, respectively. These results are superior to any other reports of patients with metastatic disease at diagnosis, particularly bone metastases.
Nevertheless, few patients with metastatic osteosarcoma are cured, and new therapeutic approaches are needed.
Dose-intensive chemotherapy with peripheral blood stem cell support is ineffective in metastatic osteosarcoma [97-99]. For patients who present with overt metastatic disease, participation in research trials should be encouraged. (See 'Investigational approaches' below.)
POSTTREATMENT SURVEILLANCE — Patients with osteosarcoma who have completed definitive therapy should undergo surveillance for disease recurrence. The optimal surveillance schedule is not established, and there is variability between clinical guidelines.
We perform a physical examination, a complete blood count, chest imaging, and local imaging of the primary site (both using the same imaging studies used during initial evaluation) every three months for years 1 and 2, every four months for year 3, every six months for years 4 and 5, and annually thereafter [100,101]. These surveillance recommendations are consistent with guidelines from the National Comprehensive Cancer Network (NCCN). (See "Osteosarcoma: Epidemiology, pathology, clinical presentation, and diagnosis", section on 'Diagnostic evaluation' and "Osteosarcoma: Epidemiology, pathology, clinical presentation, and diagnosis", section on 'Evaluation for systemic disease'.)
The Children’s Oncology Group (COG) also provides guidance on imaging surveillance for local and distant disease recurrence [101].
●Local surveillance of the primary tumor site – Plain radiographs and magnetic resonance imaging (MRI) with and without contrast are obtained for local surveillance of the primary tumor site. For patients who have completed limb salvage surgery, MRI is not required for surveillance. However, for those with new symptoms or other imaging findings concerning for disease recurrence, MRI can be performed using contemporary imaging techniques that reduce metallic artifact.
●Surveillance for distant metastases – Patients with completely treated pulmonary metastases may undergo non-contrast enhanced CT of the chest for years 1 through 5 followed by plain chest radiograph (CXR) for years 6 through 10.
Patients with symptoms or imaging findings concerning for extrapulmonary disease may undergo whole body fluorodeoxyglucose (FDG) PET-CT or PET-MRI. Routine whole-body surveillance is not recommended.
Following primary therapy, long-term follow-up of patients should continue indefinitely because of treatment-related toxicity (including secondary malignancy). Most recurrences are observed within ten years of treatment completion. However, late relapses are possible and may occur as late as 20 years after treatment completion [102].
Recommendations for long-term follow-up are available from the COG Survivorship Guidelines. For individuals treated with anthracycline-containing chemotherapy, the recommendation includes an echocardiogram or equilibrium multigated blood pool imaging (MUGA scan) at the first long-term follow-up visit, then periodical reassessment of cardiac function, with the frequency of testing dependent on age at initial treatment, the total anthracycline dose, and whether chest irradiation was also administered. Orthopedic complications of therapy are common, especially with the aging of prostheses that are placed in adolescent or young adult populations.
TREATMENT OF RECURRENT DISEASE — Patients with a disease recurrence after resection alone can often be salvaged with additional surgery and chemotherapy, although their long-term survival is inferior to that of patients who received conventional multiagent chemotherapy in conjunction with surgery previously [103,104].
Treatment of relapse in patients who have already received adjuvant and/or neoadjuvant chemotherapy is a more difficult situation. Such patients usually have received most of the effective drugs, and presumably their tumors are more chemotherapy resistant than those that have never been exposed to antineoplastic agents [105].
Nevertheless, salvage is still possible and is more likely in patients with a longer relapse-free interval. In a large database of 565 osteosarcoma patients who relapsed after being treated on one of three different neoadjuvant chemotherapy protocols within the European Osteosarcoma Intergroup, five-year survival post-relapse in those whose disease recurred after two years versus within two years of randomization was 35 versus 14 percent, respectively [106]. Other favorable prognostic factors in recurrent osteosarcoma include no more than one or two pulmonary nodules, the presence of unilateral pulmonary involvement, lack of pleural disruption, and achieving a second surgical remission [38,103,107-110].
The predominant site of disease relapse is in the lung, but the increasing use of adjuvant/neoadjuvant chemotherapy for localized osteosarcoma may have changed the pattern of relapse. Patients whose disease recurs following adjuvant chemotherapy tend to relapse later and with fewer pulmonary lesions than those treated with surgery alone; they also have a higher chance of relapsing at distant bony sites [105,111-113]. However, others have observed no change in relapse pattern with the increasing use of upfront intensive chemotherapy [114]. Although rare, the possibility of metachronous, sometimes multiple osteosarcomas must be in the differential diagnosis of a patient who has what appears to be isolated bone relapse without pulmonary involvement [115].
Therapeutic approach — The therapeutic approach to recurrent disease and the likelihood of prolonged survival are related to the location, extent, and timing of relapse [103,108,116]. Attempted resection is probably the most appropriate initial treatment for patients who relapse late (ie, beyond one year after initial therapy) with a small number of pulmonary nodules that do not invade the pleura and can be completely resected [82,117-124].
Potentially resectable disease — Complete resection of all metastatic sites is a prerequisite for long-term survival. The importance of resection can be illustrated by a series of 249 consecutive patients who developed a second or subsequent recurrence after modern combined modality treatment of osteosarcoma [116]. While only 1 of 205 patients with recurrence survived past five years without surgical remission, the five-year survival and event-free survival (EFS) rates were 32 and 18 percent for 119 second, 26 and 0 percent for 45 third, 28 and 13 percent for 20 fourth, and 53 and 0 percent for five fifth recurrences, respectively, in which a surgical remission was achieved.
Some patients with isolated pulmonary metastases that develop more than one year after original treatment will be long-term survivors with just surgical removal of their lung nodules [82,117,125]. Multiple procedures may be necessary, although the odds of long-term survival decrease with each subsequent relapse.
Patients who relapse early (ie, within six months of surgery) seem to fare poorly, even if their disease is amenable to complete resection [82,107,118,126]. This is most likely because of drug resistance that develops during chemotherapy. Although aggressive use of surgery, chemotherapy, and at times, radiation therapy (RT) aimed at eradicating all sites of disease can lead to prolonged survival in some of these patients [105,107,119,127-130], it is difficult to be overly optimistic about the chances of cure after early relapse, especially relapse involving nonpulmonary sites or unresectable pulmonary disease [30,107,118,120].
A number of investigators advocate the use of postthoracotomy chemotherapy (and even lung irradiation) to destroy presumed residual microscopic deposits after surgical treatment of overt metastatic disease [127]. The contribution of such therapy to long-term outcomes has not been examined in a controlled study, and its routine use remains to be defined. However, in our view, it is reasonable to administer such treatment to patients who undergo resection of more than three lesions appearing within 6 to 12 months of surgery, assuming at least some of the traditionally active agents have not been examined in the patient; the intent of such treatment is delay of the next recurrence rather than improving the cure rate.
Patients with unresectable disease — Patients who relapse with unresectable metastatic disease are most often incurable and should be evaluated for clinical trials and/or palliative therapy (eg, chemotherapy or short-course RT) [131].
For selected patients with a limited volume of unresectable disease, particularly those who have had no prior exposure to systemic chemotherapy, administering chemotherapy before resection is an option. While chemotherapy rarely produces a complete response at metastatic sites, some patients with inoperable metastases (including the rare patient with bone metastases [132]) may respond sufficiently to permit complete resection at a later date. While it has not been shown that chemotherapy given before or after resection of metastatic lung disease improves patient survival in comparison with patients who have had resection of metastases alone [90], there are no randomized trials that directly address this issue.
Choice of chemotherapy regimen for metastatic disease — Most patients with a disease recurrence will already have received standard chemotherapy with high-dose methotrexate (HDMTX), doxorubicin, and cisplatin (MAP). Combinations of etoposide and ifosfamide appear to be more active than either agent alone [85,133,134] and are commonly used, without or with carboplatin [135-137]. In general, response rates are lower than in the setting of newly diagnosed disease. However, although ifosfamide clearly shows a higher response rate in newly diagnosed patients, objective responses have been reported in patients whose disease relapses after they have received ifosfamide-containing adjuvant therapy [85,135,138].
In a phase II trial, 14 of 42 children with recurrent osteosarcoma (33 percent) responded to etoposide (100 mg/m2 daily for three days) and lower-dose ifosfamide (2 g/m2 daily for three days) [138]. With the expectation that adding granulocyte-colony stimulating factor (G-CSF) would permit dose intensification, a phase I trial of etoposide and escalating doses of ifosfamide in patients with recurrent osteosarcoma was completed; in this study, 6 of 13 patients exhibited a partial response [55,139].
Other options include high-dose ifosfamide alone [140], cyclophosphamide plus etoposide [141-143], or a gemcitabine-based regimen, such as gemcitabine plus docetaxel [144-147]. Clinical trials, when available, are appropriate as well.
Samarium — Samarium-153 lexidronam is a bone-seeking radiopharmaceutical that may provide pain palliation for patients with an unresectable local recurrence or skeletal metastases [148,149]. However, as a radioactive agent that is taken up by bone, samarium-153 can be toxic to bone marrow and can leave patients transfusion dependent. Thus, marrow reserve must be carefully evaluated before its use. It is unclear if there is any benefit to the use of samarium in patients with lung-predominant metastatic disease. (See "Radiation therapy for the management of painful bone metastases", section on 'Bone-targeted radioisotopes'.)
INVESTIGATIONAL APPROACHES — Other treatment strategies remain under investigation in patients with osteosarcoma. Many systemic agents have been investigated in this disease, but have limited activity.
Multitargeted kinase inhibitors — Data suggest that small molecule tyrosine kinase inhibitors that target VEGFR and other kinases, such as regorafenib, cabozantinib, sorafenib, and lenvatinib (in combination with etoposide and ifosfamide) have some limited activity in advanced osteosarcoma. Further data are still needed, given the relatively short time to disease progression when these agents are administered as subsequent therapy.
●Regorafenib – Regorafenib may be used as an off-label option for progressive (recurrent, relapsed, refractory) or metastatic osteosarcoma. Activity for regorafenib has been demonstrated in several randomized phase II trials [150,151]:
•In one placebo-controlled, randomized phase II trial of 38 patients with progressive osteosarcoma, 17 of 26 patients (65 percent) treated with regorafenib had no disease progression at eight weeks, compared with none of the 12 patients (0 percent) on placebo. Adverse events were similar to those seen previously for regorafenib.
•A separate randomized, placebo-controlled phase II trial (SARC024) of 42 patients with progressive osteosarcoma [151] was stopped early, based on data from the prior trial [150]. Compared to placebo, regorafenib improved progression-free survival (PFS; median 3.6 versus 1.7 months) but not overall survival (OS), presumably because some patients receiving placebo crossed over to regorafenib at the time of disease progression.
●Cabozantinib – In a phase II trial (CABONE), which included 42 patients with recurrent measurable osteosarcoma treated with cabozantinib, partial responses were seen in five patients (12 percent), and PFS at four months was 71 percent [152]. The PFS reported in this trial is better than that historically observed in similar patient populations and thus suggests activity of cabozantinib in recurrent osteosarcoma [153].
●Sorafenib – Sorafenib may have activity as a subsequent-line agent in osteosarcoma. As an example, in one phase II trial of 35 patients with relapsed or unresectable osteosarcoma treated with sorafenib, PFS at four months was 46 percent [154], which compares favorably with historical data [153].
●Lenvatinib – In a nonrandomized phase I/II trial (ITCC-050) of 35 patients with relapsed or refractory osteosarcoma treated with lenvatinib in combination with etoposide and ifosfamide, objective responses were seen in 3 of 32 evaluable patients (9 percent) and PFS at four months was 51 percent. [155-157]
Other agents — Other systemic agents have demonstrated limited activity in osteosarcoma, such as mTOR inhibitor combinations [155-157], eribulin [158], bisphosphonates [159], checkpoint inhibitor immunotherapy such as pembrolizumab [160], Bacille Calmette-Guerin (BCG) plus interferon [161,162], aerosolized granulocyte-macrophage colony stimulating factor (GM-CSF) [163], and antidisialoganglioside (GD2) antibodies [164,165], among others.
SOCIETY GUIDELINE LINKS — Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Bone sarcomas".)
INFORMATION FOR PATIENTS — UpToDate offers two types of patient education materials, "The Basics" and "Beyond the Basics." The Basics patient education pieces are written in plain language, at the 5th to 6th grade reading level, and they answer the four or five key questions a patient might have about a given condition. These articles are best for patients who want a general overview and who prefer short, easy-to-read materials. Beyond the Basics patient education pieces are longer, more sophisticated, and more detailed. These articles are written at the 10th to 12th grade reading level and are best for patients who want in-depth information and are comfortable with some medical jargon.
Here are the patient education articles that are relevant to this topic. We encourage you to print or e-mail these topics to your patients. (You can also locate patient education articles on a variety of subjects by searching on "patient info" and the keyword(s) of interest.)
●Basics topic (see "Patient education: Bone cancer (The Basics)")
SUMMARY AND RECOMMENDATIONS
●Prognosis of localized osteosarcoma – The survival of patients with malignant bone sarcomas has improved dramatically over the past 30 years, largely owing to chemotherapeutic advances. Adjuvant chemotherapy significantly improves survival compared with surgery alone and is a standard component of therapy for both children and adults. (See 'Prognosis and evolution of treatment' above.)
●Neoadjuvant versus adjuvant chemotherapy – For patients with localized disease, the optimal timing of chemotherapy (ie, preoperative versus postoperative) has not been established. There is no definite survival benefit for neoadjuvant as compared with adjuvant chemotherapy, but many centers preferentially utilize preoperative chemotherapy, particularly if a limb-sparing procedure is being contemplated for an extremity osteosarcoma. (See 'Neoadjuvant versus adjuvant chemotherapy' above.)
●Choice of initial therapy for localized disease – For children, adolescents, and young adults up to age 40 with localized disease, we suggest a methotrexate-containing chemotherapy regimen over a two-drug non-methotrexate-containing regimen (Grade 2B), although there is no worldwide consensus on the optimal choice of initial systemic therapy. (See 'MAP (methotrexate, doxorubicin, and cisplatin) as a standard regimen' above.)
•The optimal methotrexate-containing regimen has not been established. For children, adolescents, and young adults up to age 40 with localized osteosarcoma, the control arm of the American Osteosarcoma Study Group (AOST) 0331 (EURAMOS-1) trial (high-dose methotrexate [HDMTX] plus doxorubicin and cisplatin (table 1)) [22] is a reasonable and appropriate choice for standard therapy. If preoperative therapy is chosen, we administer 10 weeks of chemotherapy prior to surgery and continue postoperative chemotherapy to complete a total of 29 weeks, starting one week postoperatively. (See 'MAP (methotrexate, doxorubicin, and cisplatin) as a standard regimen' above.)
•For most patients, we suggest against a four-drug, as compared with a three-drug, regimen (Grade 2B). However, in countries where mifamurtide is available, mifamurtide in conjunction with methotrexate, doxorubicin, and cisplatin chemotherapy may be used. (See 'Addition of ifosfamide-based therapy: The EURAMOS-1 trial' above and 'Mifamurtide' above.)
•For older adults, the benefit of HDMTX is unproven, and we suggest cisplatin plus doxorubicin alone (Grade 2B). An HDMTX-containing regimen is a reasonable alternative in younger patients with adequate renal function. (See 'Chemotherapy in adults' above and "Treatment protocols for soft tissue and bone sarcoma", section on 'Doxorubicin plus cisplatin'.)
•Practical tips on administration of HDMTX, including therapeutic drug monitoring, the rationale for leucovorin rescue, and prevention and management of prolonged high plasma methotrexate levels, are discussed separately. (See "Therapeutic use and toxicity of high-dose methotrexate".)
•For patients who demonstrate intolerance of HDMTX and for institutions that cannot provide pharmacokinetic monitoring for HDMTX, the combination of carboplatin, ifosfamide, and doxorubicin appears to be a reasonable alternative. However, the increased doses of alkylating agents may possibly increase the risk of second malignancies (leukemia). (See 'Carboplatin versus cisplatin' above.)
●Impact of histologic response to neoadjuvant chemotherapy – Response to neoadjuvant chemotherapy is a major prognostic factor, but there is no evidence that outcomes in poor histologic responders to neoadjuvant chemotherapy can be improved by altering the postoperative chemotherapy regimen. For patients treated off protocol, we recommend no change in the chemotherapeutic regimen after surgery in patients who have a less-than-complete response to the initial neoadjuvant chemotherapy (Grade 1B). (See 'The option to change systemic therapy after neoadjuvant chemotherapy and surgery' above.)
●Is there a role for adjuvant radiation therapy? – There is no role for adjuvant radiation therapy (RT) in patients with completely resected osteosarcoma, and for most patients, we recommend not pursuing this approach because of the risk for late toxicity (Grade 1B). We restrict the use of adjuvant RT to those patients with incompletely resected sarcomas and the rare patient with a small cell osteosarcoma. (See 'Adjuvant radiation therapy' above.)
RT is the primary modality for local control only for patients who decline surgery or for whom there is no effective surgical option. (See 'Radiation therapy' above.)
Prognosis of metastatic osteosarcoma at diagnosis – Patients who present with overtly metastatic osteosarcoma have a poor prognosis, although long-term survival is possible in up to 50 percent of those with isolated pulmonary metastases who are treated with combined systemic and surgical therapy. (See 'Patients with metastatic disease at diagnosis' above and "Surgical resection of pulmonary metastases: Benefits, indications, preoperative evaluation, and techniques" and "Surgical resection of pulmonary metastases: Outcomes by histology".)
●Choice of initial therapy for metastatic disease at diagnosis – For children and adolescents with metastatic osteosarcoma at the time of diagnosis who do not have a surgical option, there is no single standard approach to treatment, and these patients should be encouraged to enroll in clinical trials testing new therapies. If protocol enrollment is not available or if patients are ineligible, treatment with the control arm of the AOST 0331 protocol (HDMTX plus doxorubicin and cisplatin chemotherapy alone (table 1)) or as per AOST 06P1 without zoledronic acid (HDMTX plus doxorubicin, cisplatin, ifosfamide, and etoposide) is a reasonable and appropriate choice for standard therapy. (See 'Choice of chemotherapy' above.)
●Posttreatment surveillance – Patients should undergo posttreatment surveillance for local and distant disease recurrence. (See 'Posttreatment surveillance' above.)
●Recurrent, metastatic, potentially resectable disease – The predominant site of disease relapse following initially successful treatment is the lung. Attempted resection is probably the most appropriate initial treatment for patients who relapse beyond one year after initial therapy with a small number of pulmonary nodules that do not invade the pleura and that can be completely resected. Although some of these patients may be cured with resection alone, most receive a combination of surgery plus chemotherapy. (See "Surgical resection of pulmonary metastases: Outcomes by histology" and "Surgical resection of pulmonary metastases: Benefits, indications, preoperative evaluation, and techniques" and 'Potentially resectable disease' above.)
●Recurrent metastatic, unresectable disease – Patients who relapse with unresectable metastatic disease are incurable and should be evaluated for palliative therapy (eg, RT, chemotherapy). (See 'Patients with unresectable disease' above.)
However, in selected patients with a limited volume of unresectable disease, particularly those who have no prior exposure to systemic chemotherapy, administering chemotherapy before resection is an option. While chemotherapy rarely produces a complete response at metastatic sites, some patients with inoperable metastases may respond sufficiently to permit complete resection at a later date. (See 'Therapeutic approach' above.)
●Choice of systemic therapy for recurrent metastatic disease – The choice of chemotherapy depends upon prior treatment. Clinical trial enrollment is encouraged, where available. (See 'Choice of chemotherapy regimen for metastatic disease' above.)
•Most patients will have already received HDMTX, doxorubicin, and cisplatin. For these patients, we suggest treatment in a clinical trial, when feasible.
•For patients not enrolled in clinical trials, we typically use etoposide plus ifosfamide, with or without carboplatin. If chemotherapy has not been previously administered, we suggest the same regimens as are used for primary management.
•Investigational agents such as regorafenib, cabozantinib, sorafenib, and lenvatinib plus etoposide and ifosfamide have some limited activity. (See 'Investigational approaches' above.)
ACKNOWLEDGMENT — The UpToDate editorial staff acknowledges Thomas F DeLaney, MD, who contributed to earlier versions of this topic review.
The UpToDate editorial staff acknowledges Allen Goorin, MD, now deceased, who contributed to earlier versions of this topic review.
72 : Neoadjuvant chemotherapy and local radiotherapy for high-grade osteosarcoma of the extremities.
94 : Phase II study of high-dose methotrexate in patients with unresectable metastatic osteosarcoma.
Do you want to add Medilib to your home screen?