| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
EDITORIALS |
From the Departments of Surgery (BNR) and Hematology-Oncology (CR-G), St. Jude Children’s Research Hospital, Memphis, Tennessee.
Correspondence: Address correspondence to: Bhaskar N. Rao, MD, Department of Surgery, St. Jude Children’s Research Hospital, 332 N. Lauderdale St., Memphis, TN 38105; Fax: 901-495-3122; E-mail: bhaskar.rao{at}stjude.org
The optimal treatment of osteosarcoma consists of multiagent chemotherapy and local control measures such as amputation or limb-sparing surgical procedures. Different combinations of platinum compounds, doxorubicin, and high-dose methotrexate have formed the basis of standard chemotherapy regimens that produce cures in 50% to 75% of patients with nonmetastatic disease.1
For over two decades, the treatment of nonmetastatic osteosarcoma has followed the basic guidelines of the T-10 protocol, with current treatment strategies utilizing the lessons learned from it. The T-10 protocol and its variants consist of a multiagent regimen using high-dose methotrexate, doxorubicin, cisplatin, and a combination of bleomycin, cyclophosphamide, and dactinomycin. Using the T-10 guidelines, the 5-year actuarial event-free survival (EFS) may approach 70%.1,2 However, these results are not always reproducible, and multi-institutional American and European studies that followed similar guidelines report EFS rates much lower (45–60%).3–6 One of the major contributions of the T-10 protocol was that histologic response to neoadjuvant chemotherapy is the most important prognostic factor in patients with nonmetastatic disease.2 This observation has dictated recent treatment protocols. Modification of the postoperative treatment, by intensifying cisplatin and doxorubicin or, more recently, with incorporation of new agents such as ifosfamide and etoposide, has not resulted in a significant improvement in the outcome for the group of patients who have poor histologic response (<90% necrosis).2,4,5 Exposure of tumor cells to suboptimal cytotoxic therapy during preoperative treatment may result in the development of chemoresistance and an increase in the propensity for metastatic spread. Here modification of postoperative therapy may not be able to overcome these adverse effects. If this line of reasoning is followed, achieving a rapid, early tumor response should thus be the primary objective. Alternatives to intensification of adjuvant therapy should then focus on improving preoperative chemotherapy. Intensification of preoperative chemotherapy results in a modest increase in the proportion of good histologic responders, but the impact on final outcome is minimal.1 In the early 1980s, Jaffe et al. demonstrated that higher concentrations of cisplatin within a tumor could be achieved when the agent was delivered intra-arterially, without compromising systemic exposure.7 These encouraging results led several groups to use intra-arterial cisplatin in the treatment of patients with nonmetastatic osteosarcoma. Despite the relevance of this new approach, only two groups addressed this question with a randomized study. Investigators at the Rizzoli institute have recently reported their long-term results (Istituto Ortopedico Rizzoli/Osteosarcoma-3 [IOR/OS-3] and IOR/OS-5) in which patients were randomized to intra-arterial or intravenous cisplatin. In both studies, pre- and postsurgery chemotherapy also included doxorubicin and high-dose methotrexate, with the addition of ifosfamide in the latter study. In both studies, the proportion of good responders was higher in the group of patients receiving intra-arterial cisplatin, but the route of administration of cisplatin did not impact outcome.8 A similar study design was the core of the German Cooperative Osteosarcoma Study-86 trial, which also failed to demonstrate an advantage of the use of intra-arterial cisplatin.9 The lack of a proven advantage, and the high cost (and unavoidable risks) of the procedure, led to the abandonment of the use of the intra-arterial route for cisplatin in the treatment of osteosarcoma. In this issue of the Annals of Surgical Oncology, Wilkins et al.10 reopen the debate on the value of intra-arterial administration of cisplatin. In their study, patients with localized osteosarcoma underwent repeat courses of intra-arterial cisplatin and intravenous doxorubicin, and the surgical procedure was delayed until a >90% decrease in the tumor vascularization was achieved. Good histologic responses (>90% necrosis) were obtained in 87% of the patients, and the 10-year EFS and overall survival were 84% and 92%, respectively. These results are among the best ever reported for patients with nonmetastatic osteosarcoma. It is particularly interesting that the majority of patients (41 of 47) were treated with cisplatin and doxorubicin only, a combination that in larger studies has reported to cure no more than 50–60% of patients.3 Therefore, given the small number of patients in this study, these results must be interpreted with great caution.
This study was also built on the premise that achieving a major cytoreduction early in the course of the treatment will impact outcome, and that direct instillation of chemotherapy into the tumor is the best way to do so. What are the reasons that would substantiate such an approach? Intra-arterial administration of chemotherapeutic agents either as isolation perfusion (melanoma) or by direct instillation is common practice in the treatment of malignancies in which it may facilitate the local control (e.g., head and neck, genitourinary cancer, or brain gliomas). In extremity osteosarcoma, however, there is no evidence that this approach improves either local control rates or increases the likelihood of limb preservation.8,11 Nevertheless, early cytoreduction may decrease the risk of acquired chemoresistance and subsequent development of metastatic disease, which is the prime cause of failure in patients with osteosarcoma. If this were the objective, upfront surgery would have provided the same result, that is, early eradication of the large tumor burden. However the timing of surgery has not impacted outcome. In the multi-institutional Pediatric Oncology Group 8651 study, patients were randomized to immediate surgery or to receive preoperative chemotherapy with high-dose methotrexate, cisplatin, and doxorubicin. The 5-year EFS for patients receiving or not receiving preoperative chemotherapy was 61.6% and 69.3%, respectively.12 Similar results have been reported in nonrandomized studies.11
Therefore, how should we interpret the excellent results reported by Wilkins et al. using intra-arterial cisplatin? It is certainly possible that the route of administration of chemotherapy impacts outcome. However, these results are in contradiction with some of the evidences discussed above and with the results of previous randomized trials (the only way of answering this question) that used a similar approach. The major difference in this study and others reported are that Wilkins et al. continued preoperative therapy until they detected a 90% necrosis by imaging studies. The data of this study confirm that the degree of tumor necrosis continues to be one of the most important prognostic indicators; the circle is thus closed and we are back to where it all started. It is also probable that the biological phenotype is what ultimately dictates the outcome. Little is known about the biology of osteosarcoma that may impact outcome. Tumor hyperdiploidy13 or increased expression of p-glycoprotein14 Ki-6715 or HER2/erbB-2,16 for example, have all been shown to correlate with a more aggressive behavior and adverse outcome. However, in comparison with other malignancies in which the biological categorization dictates the treatment, osteosarcoma continues to be regarded as a uniform malignancy. After two decades of modern multimodal approach, little advances have been made in the understanding of the biology of osteosarcoma. The histologic response continues to be the question rather than the answer. The answer maybe is in the biology.
Acknowledgments
Supported in part by grant CA21765 from the National Institutes of Health Cancer Center and by the American Lebanese Syrian and Associated Charities (ALSAC). The authors thank Ms. Flo White for her editorial assistance.
The acknowledgments are available online at www.annalssurgicaloncology.org.
Received for publication April 2, 2003. Accepted for publication April 21, 2003.
REFERENCES
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
