The authors summarized the outcomes of patients with unresectable osteosarcoma of the trunk who received carbon ion radiotherapy (CIRT).
The authors summarized the outcomes of patients with unresectable osteosarcoma of the trunk who received carbon ion radiotherapy (CIRT).
The authors performed a retrospective analysis of 78 patients who had medically inoperable osteosarcoma of the trunk and received treatment with CIRT between 1996 and 2009. Tumor sites included the pelvis in 61 patients, the spine and paraspinal region in 15 patients, and other sites in 2 patients. The median applied CIRT dose was 70.4 Gray equivalent (GyE) in a total of 16 fixed fractions over 4 weeks.
The minimum duration of follow-up for survivors was 14 months. Forty-eight patients remained alive. The 5-year overall survival rate was 33%, and the local control rate was 62%. Thirty-eight patients who had a clinical target volume <500 cm3 had a 5-year overall survival rate of 46% and a 5-year local control rate of 88%. Except for 3 patients who experienced severe skin/soft tissue complications requiring skin grafts, no other severe toxicities were observed. Of 9 patients who were continuously disease free for >5 years, 8 were able to walk with or without the help of a cane, and 6 were free from pain killers.
CIRT appeared to be a safe and effective modality for the management of unresectable osteosarcoma of the trunk, providing good local control and offering a survival advantage and good long-term functional results without unacceptable morbidity. Cancer 2012. © 2012 American Cancer Society.
Osteosarcoma of the trunk is rare, representing 6.3% of all osteosarcomas.1 Surgery has been the mainstay of local therapy for this disease; however, resection is sometimes difficult because of the location and extension of the tumor. Even when it is possible, tumor resection often damages nerves and muscles and causes severe functional impairment.2 Thus, osteosarcoma of the trunk is one of the most challenging tumors to treat. In the Cooperative Osteosarcoma Study Group (COSS) study of 1702 patients with high-grade osteosarcoma of the extremities and trunk, the 5-year overall survival rate for all patients was 65.3%, whereas the survival rate among 107 patients who had tumors of the trunk was only 34.2%.1 Other studies have reported 5-year overall survival rates of 15% to 38% for patients with osteosarcoma of the trunk.3-12
Chemotherapy is another mainstay of treatment for osteosarcoma. Adjuvant chemotherapy has improved prognosis,13, 14 and tumor response to preoperative chemotherapy has been reported as a significant prognostic factor.1 However, it is impossible to control osteosarcoma without effective local treatment, even if the response to chemotherapy is good. Because there has been no effective local treatment except surgery, the prognosis for patients with unresectable osteosarcoma invariably has been dismal.
Osteosarcoma is relatively radioresistant, and few reports have described effective local control using photon radiotherapy for osteosarcoma of the trunk.6, 7, 15 Carbon ion radiotherapy (CIRT) has better dose distribution to the tumor compared with photon radiotherapy. Carbon ion beams emit only a low dose of radiation after penetrating the body and deliver their maximum dose at the end of their range, beyond which the dose drops sharply (the Bragg peak). This pattern of irradiation facilitates the delivery of an optimal dose to the tumor while exposing critical organs surrounding the tumor to lower doses. Another property that distinguishes carbon ion beams from photons is their high biologic effectiveness. Carbon ion beams deliver a larger mean energy per unit length of their trajectory (ie, higher linear energy transfer [LET]) to body tissues than photon and proton beams. Furthermore, the LET of carbon ion beams increases steadily from the initial value at the entrance point, reaching the maximum value at the end of its range. These advantageous treatment profiles for carbon ion beams likely are responsible for their good results in the treatment of bone and soft tissue sarcomas.16-18
Since 1996, we have been involved in clinical trials of CIRT for medically inoperable bone and soft tissue sarcomas at the National Institute of Radiological Sciences (NIRS), Chiba, Japan. The first phase I/II dose-finding clinical trial was implemented between June 1996 and February 2000 and was followed by a phase II fixed-dose clinical trial conducted between April 2000 and November 2003, after which we continued using the protocol, then approved by the Ministry of Health, Labor, and Welfare as “highly advanced medical technology,” in medical practice.16-18 Thus, as of February 2011, 500 patients have been enrolled on the protocol. The objective of this study was to evaluate the effectiveness and safety of CIRT for patients with unresectable osteosarcoma of the trunk who were treated on phase I/II and phase II clinical trials.
The patients who met all of the following eligibility criteria were registered: histologic confirmation by the central pathologist, tumors judged medically inoperable by referring surgeons, grossly measurable tumors ≤15 cm in greatest diameter, an Eastern Cooperative Oncology Group performance status of 0 to 2, no distant metastasis at initial referral for treatment, no prior radiation therapy at the same site (excluding radiation-associated sarcoma), no prior chemotherapy within 4 weeks before CIRT, no infection at the tumor site, and no intravascular tumor embolism. The details of these trials were reported previously.16, 17 All patients signed an informed consent form that was approved by the local institutional review board.
The specific CIRT technique used at NIRS has been described in previous publications.19-21 Carbon ion beams generated by the Heavy Ion Medical Accelerator in Chiba (HIMAC) have accelerated energies of 290 millions of electron volts per nucleon (MeV/n), 350 MeV/n, and 400 MeV/n, which have a range of a depth between 15 cm and 25 cm in water. For modulation of the Bragg peak to conform to a target volume, the beam lines for treatment are equipped with a pair of wobbler magnets, beam scatterers, ridge filters, multileaf collimators, and a compensation bolus. The ridge filter is designed to produce biologically equal effects along the spread-out Bragg peak (SOBP). Energies of 350 MeV/n and 400 MeV/n are used mainly for treatment of osteosarcoma of the trunk.
Patients were immobilized in customized cradles with a low-temperature thermoplastic sheet. Respiratory gating of both the planning computed tomography (CT) acquisition and therapy was performed when indicated.19 Three-dimensional treatment planning of CIRT was performed using the HIPLAN software program (NIRS, Chiba, Japan).20, 21 The clinical target volume (CTV) encompassed the gross tumor volume (GTV) and enhanced area by contrast medium surrounding the GTV on magnetic resonance imaging (MRI) and CT studies. The area that was considered subclinical disease also was added to the CTV. The planning target volume (PTV) included the CTV plus a 5-mm margin for positioning errors but depended on the distance from critical organs like the spinal cord and the intestine. The CTV was covered by at least 90% of the prescribed dose (Fig. 1). The boost irradiation was not usually used. We did not adopt a planed combination of surgery and pre/postsurgical carbon ion radiotherapy.
The dose was expressed as the Gray (Gy) equivalent (GyE) (the carbon physical dose [Gy] × relative biologic effectiveness [RBE]). The RBE was evaluated by both radiobiologic and physical studies.21 CIRT was performed once daily, 4 days per week (Tuesday to Friday), for a total of 16 fixed fractions over 4 weeks. Patients received 2 to 8 irregularly shaped ports (median, 3 ports). One port was treated in each session. At every treatment session, the patient's position was verified with a computer-aided online positioning system.
A phase I/II dose-escalation trial with a total dose ranging from 52.8 GyE to 73.6 GyE in 16 fractions over 4 weeks was carried out for 64 lesions in 57 patients with bone and soft tissue sarcomas during the period from June 1996 to February 2000. After the trial, the recommended dose was fixed at 70.4 GyE in 16 fractions.16 A fixed-dose phase II trial was then initiated in April 2000. In this study group, 3 patients received a total dose of 52.8 GyE, 3 received a total dose of 57.6 GyE, 8 received a total dose of 64.0 GyE, 57 received a total dose of 70.4 GyE, and 7 received a total dose of 73.6 GyE.
Patients were monitored with CT and MRI studies at least every 6 months after CIRT. When the patients could not come to our hospital, we estimated their condition using imaging films taken at local hospitals and medical reports from local physicians. Local control was defined as no increase in tumor volume observed on 2 consecutive MRI or CT studies. Recurrence developing >2 cm away from the PTV was defined as skip failure. The follow-up periods started from the initiation of CIRT. Acute and late toxicities attributable to radiotherapy were scored using the National Cancer Institute Common Toxicity Criteria, version 3.0, and the Radiation Therapy Oncology Group/European Organization for Research and Treatment of Cancer late radiation morbidity scoring scheme.
Overall survival and local control rates were calculated by using the Kaplan-Meier method with JMP software (version 7.0; SAS Institute, Cary, NC). Univariate analysis was conducted using the log-rank test, and a Cox proportional hazards model was used for multivariate analysis. Statistical significance was set at P < .05.
Between June 1996 and July 2009, 78 patients (49 men and 29 women) with unresectable osteosarcoma of the trunk received CIRT. Patient characteristics are summarized in Table 1. The median age at the time of initial referral was 41 years (range, 11-83 years). The tumor was located in the pelvis in 61 patients, in the spine or paraspinal region in 15 patients, and in other regions in 2 patients (mediastinum and chest wall). Seventy-four tumors were primary, and 4 were metastatic. Tumor subtypes were osteoblastic in 36 patients, chondroblastic in 16 patients, fibroblastic in 14 patients, and other or unclassified in 12 patients. The median maximum tumor size was 10 cm (range, 2-18 cm), and the median CTV was 510 cm3 (range, 60-2299 cm3). Eleven patients had undergone previous surgical resection, including 9 patients who underwent surgical resection and developed local recurrence and 2 patients who had residual tumors from insufficient surgical resection. Sixty-one patients had received prior chemotherapy. Eight patients underwent previous surgical resection and received chemotherapy. Fourteen patients received no prior treatment. Three patients had radiation-associated osteosarcoma, including 1 patient who received radiotherapy for uterine cervical cancer 16 years ago, 1 patient who received radiotherapy for prostatic cancer 7 years ago, and 1 patient who received radiotherapy for plasmacytoma of the pubis 16 years ago.
|Characteristic||No. of Patients||Median (Range)|
|Age, y||41 (11-83)|
|Spine or paraspine||15|
|Others or unclassified||12|
|Maximum tumor size, cm||10 (2-18)|
|Clinical target volume, cm3||510 (60-2299)|
|ALP, IU/L||347 (153-3729)|
|CRP, mg/dL||0.21 (0.0-10.9)|
|Total dose, GyE||70.4 (52.8-73.6)|
The minimum duration of follow-up for survivors was 14 months. The median follow-up time was 42 months (range, 14-166 months) for the 30 survivors and 24 months (range, 2-166 months) for all 78 patients. Thirty patients (38%) were alive, 22 patients (28%) remained continuously disease free at last follow-up, and 48 patients (62%) died. Forty-five patients died of their disease, and 3 died of other causes (1 from acute myocardial infarction at 110 months, 1 from a brain hemorrhage at 23 months, and 1 from pulmonary infarction 2 months after CIRT). The median survival time for all 78 patients was 28 months (range, 2-166 months), and the 2-year and 5-year overall survival rates were 58% and 33%, respectively (Fig. 2A). The 2-year and 5-year disease-specific survival rates were 60% and 34%, respectively; the 2-year and 5-year progression-free survival rates were 34% and 23%, respectively; and the 2-year and 5-year local control rates were 73% and 62%, respectively (Fig. 2B). Twenty-one patients (27%) developed local recurrences. The median time to diagnosis of local recurrence was 15 months (range, 4-96 months), and the median CTV at the time of initial treatment among patients with locally recurrent tumors was 838 cm3 (range, 273-1677 cm3). Six of those patients received salvage treatment with repeated CIRT. Distant metastases were observed in 41 patients (53%), and the most frequent site of metastasis was the lung (28 patients).
In univariate and multivariate analysis, performance status, CTV, alkaline phosphatase (ALP) levels, and C-reactive protein (CRP) levels before the start of CIRT were significant prognostic factors for overall survival; and performance status and CTV were significant prognostic factors for local control (Tables 2, 3). Thirty-eight patients with a CTV <500 cm3 had a 5-year overall survival rate of 46% and a 5-year local control rate of 88%, whereas 40 patients with a CTV ≥500 cm3 had a 5-year overall survival rate of 19% and a 5-year local control rate of 31% (Fig. 3).
|Clinical Factor||No. of Patients||5-Year-OS, %||P||5-Year-LC, %||P|
|Others or unclassified||42||21||66|
|Clinical target volume, cm3|
|≥Upper limit (335 IU/L)||39||25||54|
|≥Upper limit (0.3mg/dL)||31||11||71|
|Total dose, GyE|
|Clinical Factor||HR||95% CI||P||HR||95% CI||P|
|CTV ≥ 500 cm3||1.94||1.08-3.60||.02a||5.85||2.03-21.4||.0004a|
|ALP ≥ upper limit||1.85||1.00-3.41||.04a||1.33||0.54-3.42||.52|
|CRP ≥ upper limit||2.10||1.12-3.93||.01a||1.05||0.37-2.75||.91|
All 78 patients were able to complete the planned CIRT without interruption. There were no fatal toxicities during follow-up after CIRT. Grade 3 acute skin reactions were observed in 3 patients, grade 3 late skin/soft tissue reactions were observed in 4 patients, and grade 4 late skin/soft tissue reactions requiring skin grafts occurred in 3 patients. The majority of patients presented with functional deficits of various degrees, depending on the location and extent of the tumor before CIRT. Four patients had permanent neurologic complications for which radiotherapy was believed to be the sole cause. Two patients experienced bone fractures requiring surgery, including 1 patient with osteosarcoma of the ileum who experienced a femoral neck fracture 13 months after CIRT and 1 patient with lumbar spinal osteosarcoma who developed a compression fracture 6 months after CIRT.
Of all 78 patients, 12 patients survived for >5 years. Nine patients were alive and remained continuously disease free, and 3 patients died after 5 years. One patient died of other disease, 1 developed a local recurrence 53 months after CIRT and died 66 months after CIRT, and another developed a local recurrence and distant metastases 98 months after CIRT and died 101 months after CIRT. Long-term functional evaluations after CIRT were performed on the 9 patients who remained continuously disease free. Of those 9 patients, 6 had tumors located in the pelvis, and 3 had tumors located in the spine or paraspine. Eight of 9 patients were able to walk, although some required the help of a cane. One patient with acetabular osteosarcoma used a wheelchair in daily life, but she also retained the ability to walk short distances. Six patients were free from pain killers, 1 used nonsteroidal anti-inflammatory drugs, and 2 who used narcotics before CIRT still used them. All but 2 patients aged >65 years were able to work or attend school.
The 5-year overall survival rates for patients with osteosarcoma of the trunk who received resection were 22% to 44%.3-11 In the COSS study of 1702 patients with osteosarcoma, an axial tumor site was a poor prognostic factor for response to chemotherapy, overall survival, and event-free survival.1 Among 107 patients who had axial tumors in that study, 23 patients (21%) experienced a first surgical remission, and 66 patients (61%) experienced local recurrence.1 Ozaki et al reported that patients with osteosarcoma of the spine and pelvis who underwent intralesional or no surgery had significantly poorer overall survival compared with patients who underwent wide or marginal resection.6, 7 The 5-year overall survival rates for patients with osteosarcoma of the trunk who did not undergo surgical resection were 0% to 30%.3-10
Osteosarcoma is known for its relative resistance to photon radiotherapy, such as x-ray radiotherapy, and there are few reports on the effectiveness of photon therapy. Especially in patients with axial osteosarcoma, the dose constraints to normal tissues surrounding the tumor, like intestine and spinal cord, make it difficult to apply high doses to the tumor. Only a few reports have examined charged particle therapies for osteosarcoma of the trunk. DeLaney et al reported the results from preoperative and postoperative photon/proton radiotherapy with or without radical resection for 41 patients with osteosarcoma, including 16 with axial osteosarcoma.22 Patients with axial osteosarcoma received a median applied dose of 66 cobalt GyE (CGE) with a preoperative radiotherapy dose of 20 Gy and postoperative radiotherapy. Patients who underwent either gross or subtotal resection had greater rates of overall survival and local control compared with those who underwent biopsy only at 5 years (73.9% vs 25% and 77.7% vs 40%, respectively). Ten of 41 patients (24%) experienced significant late complications related to radiotherapy that required hospitalization or surgery, including 1 of 8 patients (13%) with spinal osteosarcoma and 3 of 7 patients (43%) with pelvic osteosarcoma.22 After that report, Ciernik et al presented updated data on proton-based radiotherapy for unresectable or incompletely resected osteosarcoma.23 In that study, the median applied dose was 68.4 Gy higher than that in the previous report. Among the 55 patients who were analyzed, there were 12 patients with unresected disease, 19 patients with partially resected disease, and 24 patients with grossly resected disease who had positive margins. The median initial target volume, which included the areas at risk for subclinical disease beyond the original gross tumor, was 213 mL. The 5-year local control rate was 72%, and the 5-year overall survival rate was 67%. The study indicated that applying a higher dose was essential to attaining better local control. In our study, we did not adopt planed surgery and pre/postsurgical CIRT, and the median tumor volume was much greater than that in the study by Ciernik et al. It is difficult to determine the impact of irradiation dose on the local control rate, because there was less diversity of applied doses in our study. However, in our previous report on a dose-escalation study in patients with bone and soft tissue sarcoma who received CIRT, there was a significant difference in the local control rate achieved with a total dose ≤57.6 GyE and that achieved with a total dose ≥64.0 GyE.16
In 2010 at the Heidelberg Ion Therapy Center, a phase I/II nonrandomized trial to determine the safety and efficacy of CIRT in patients with osteosarcoma was initiated and is currently ongoing.24 For patients aged ≥6 years with unresectable osteosarcoma, a total applied dose of 60 to 66 CGE with a proton radiotherapy of 45 Gy and a carbon ion boost of 15 to 21 GyE will be administered. Changes in 2-deoxy-2(18F)fluoro-D-glucose–positron emission tomography (FDG-PET) imaging before and after therapy will be investigated prospectively. The primary objective of the trial is to determine the feasibility and toxicities of CIRT.
Compared with results from previous studies, in our study, the 5-year overall survival rate was 33%, and the local control rate was 62%. Particularly in patients with a CTV <500 cm,3 the 5-year overall survival and local control rates were superior to those produced by surgery (Fig. 3). Large tumor volume has been identified as a predictor of a poor outcome.1, 3, 5-7, 11, 25 In univariate and multivariate analyses, the performance status and the CTV before CIRT were identified as prognostic factors for overall survival and local control, and ALP and CRP levels before CIRT were identified as significant prognostic factors for overall survival. Bramer et al reported that high postchemotherapy ALP levels were correlated with poor survival and response to chemotherapy in patients with localized osteosarcoma.26 In our study, the group with higher ALP levels had worse survival, which may indicate the tumor response to chemotherapy before CIRT. Funovics et al reported that a high preoperative CRP level without infection was an independent predictor of survival in patients with osteosarcoma.27 In our current study, the group with higher CRP levels had significantly shorter survival than the group with normal CRP levels.
Fifteen patients aged >60 years completed CIRT. Among them, 7 patients were alive at the last follow-up. Longhi et al reported results from 43 patients aged ≥65 years with osteosarcoma.25 In that study, 5-year overall survival rate for all patients was 22%, and the rate for 11 patients who did not undergo surgery was 0%. In our study, the 5-year overall survival and local control rates for 15 patients were 34% and 84%, respectively. These results indicate that CIRT should be a good option for elderly patients who are deemed intractable to surgery.
The development of local recurrence after surgery is associated with a poor prognosis.28 The reported 5-year overall survival rate after recurrence was 18% to 23%.29, 30 In our study, 11 patients had local recurrence or residual tumors after undergoing previous surgical resection. For those patients, the 5-year survival rate after recurrence was 48%. These results indicate that CIRT can successfully treat some patients with local recurrence.
In this study, 2 patients survived for >5 years and then developed local recurrence. Even if treated patients remain continuously disease free, long-term follow-up will be essential. Three patients experienced severe, late skin/soft tissue complications requiring skin grafts. Yanagi et al used a dose-surface histogram (DSH) to analyze the late skin reaction of patients with bone and soft tissue sarcoma after CIRT.31 In that analysis, the area in skin surface irradiated with over 60GyE on DSH (S60) > 20cm2 was the most important factor influencing the development of late skin reaction. However, it was possible to prevent the reaction by aiming from 3 ports and modifying the irradiation method to reduce the dose delivered to the skin. The incidence of severe skin reactions in patients who received a total dose of 70.4 GyE was within the acceptable level for the past several years.
In conclusion, data from the current study indicate that CIRT has efficacy against unresectable osteosarcoma of the trunk and that toxicity from CIRT is acceptable. Although more patients must be monitored over a longer period to clearly establish the effectiveness of CIRT, CIRT will be a mainstay for unresectable osteosarcoma of the trunk and could be an alternative to surgery. Even for elderly patients and patients with postoperative recurrent tumors, CIRT will represent an alternative to surgery.
This study was supported by the Research Project with Heavy Ions at the National Institute of Radiological Sciences-Heavy Ion Medical Accelerator in Chiba (NIRS-HIMAC).
CONFLICT OF INTEREST DISCLOSURES
The authors made no disclosures.