Abstract
Background: The purpose of this study was to evaluate local control and toxicity in a group of patients treated with stereotactic body radiotherapy (SBRT) for lung metastases (LM) from bone and soft tissue sarcomas. Patients and Methods: From October 2010 to July 2014, patients with LM from sarcomas not suitable for surgery were treated with daily cone-beam computed tomography-guided SBRT. The dose administered ranged from 30 to 60 Gy in 3-8 fractions. Acute and late toxicity were scored according to Common Terminology Criteria for Adverse Events version 4.0. Results: A total of 24 patients with 68 LM from sarcomas were treated with SBRT. The median follow-up after SBRT was 17 months (range=11–51 months). Two-year actuarial lesion local control and overall survival were 85.9% and 66.4%, respectively. No G3 or greater acute and late toxicities were observed. Conclusion: SBRT is a safe and effective treatment for LM from sarcoma and might be used as an alternative option in patients unfit for surgery.
Sarcomas are a rare and heterogeneous group of tumors. They represent approximately 1% of adult cancers and arise from mesenchymal cells. They are divided into two major categories: bone and soft-tissue sarcomas (1). Primary bone sarcomas are less common than soft-tissue sarcomas, accounting for <0.2% of malignant tumors (2). About 50% of patients with sarcoma will experience metastases and the lung is the most frequent site of distant spread (3). Surgical resection is the standard-of-care for these patients, with 5-year survival rates of 25-38% (4-6). The main prognostic factor in patients with lung metastases (LM) is complete resection of all secondary lesions (4-5, 7).
Stereotactic body radiation therapy (SBRT) is now the standard-of-care in patients with early-stage non-small cell lung cancer who are unfit for surgery (8), and has been used for the treatment of LM in many series of patients with different primary tumor types, mainly colon and breast cancer (9). Sarcomas are generally considered highly radioresistent but the radiobiology of very highly hypofractionated treatment, such as SBRT, could hypotethically increase the response of tumor cells to radiation (10). Recently, some authors have reported the results of SBRT in the treatment of LM from bone and soft-tissue sarcomas (11-15). However, all series are retrospective and most of them enrolled fewer than 28 patients. Therefore, evidence for the efficacy of SBRT in such patients are lacking. For these reasons, we conducted this retrospective study evaluating local control and toxicity in a group of consecutive patients treated with SBRT for LM from bone and soft-tissue sarcomas.
Patients and Methods
From October 2010 to July 2014, a group of patients with LM from sarcoma were treated with SBRT. Patients suitable for SBRT were selected after multi-disciplinary evaluation based on prior lung surgery, high risk of postoperative complications, impaired pulmonary reserve, and failure of previous lines of systemic chemotherapy. All patients were informed about treatment modalities and possible adverse events and gave their written consent before radiotherapy. Patients were immobilized in supine position with the help of a vacuum pillow in the Stereotactic Body Frame (SBF©; Elekta, Stockholm, Sweden). Abdominal compression was employed to reduce the movements of the diaphragm. A computed tomographic (CT) scan with 3 mm slice thickness and a slow rotation time (3 seconds) in order to more accurately define the position of the target during respiratory phases was performed as described by Lagerwaard et al. (16). For patients with lesions located in the lower lobes an Active Breath Control device (ABC©; Elekta) was used to reduce the internal margin. In these cases, the CT scan was acquired with a ‘fast’ protocol. The gross tumor volume (GTV) was delineated on each axial CT slice using appropriate pulmonary windowing. GTV and the clinical target volume (CTV) were considered equivalent. The planning target volume (PTV) was CTV plus a 5 mm margin in the trasversal plane and 9 mm margins in the superior-inferior directions. The dose was prescribed to 80% isodose, which had to cover more than 98% of the PTV. The administered dose ranged from 30 to 60 Gy in 3-8 daily fractions. Organ at risk (heart, lungs, esophagus, proximal tracheobronchial tree, spinal cord, brachial plexus and ribs) were delineated on the same CT scan. Dose constraints were the following: Lung: V12Gy<20%, V20Gy<10% volume; Dmax spinal cord: <15 Gy/3 fractions, <20 Gy/5 fractions, <24 Gy/8 fractions; Dmax heart: <24 Gy/3 fractions; <27 Gy/5 fractions; <30 Gy/8 fractions; Dmax esophagus: <24 Gy/3 fractions; <27 Gy/5 fractions; <30 Gy/8 fractions; Dmax brachial plexus: <24 Gy/3 fractions; <27 Gy/5 fractions; <30 Gy/8 fractions; Dmax trachea and proximal bronchial tree: <30 Gy/3 fractions; <32 Gy/5 fractions; <34 Gy/8 fractions; ribs: V30Gy<30 cc /3-5 fractions; V34Gy<30 cc /8 fractions; V50Gy<5 cc (17-20). Treatment was delivered with a linear accelerator using 8-11 fixed coplanar or non-coplanar fields with 10 MV photon energy. A multileaf collimator (leaf width: 4 mm at the isocenter) was used to shape each field. A cone-beam CT was performed daily before each treatment fraction to check the position of the target volume, and online set-up adjustments were obtained by mean of a robotic couch (HEXAPOD-EVO©; Elekta, Stockholm, Sweden).
Routine follow-up included history and physical examination and a contrast-enhanced chest CT scan every 3 months after treatment for the first year then every 6 months. Patients were assessed for toxicity daily during the treatment and every 3 months thereafter. Acute and late toxicity were scored according to Common Terminology Criteria for Adverse Event version 4.0 (21). Local control (LC) was defined as the absence of disease progression in the treated site(s) according to the Response Evaluation Criteria In Solid Tumors (22).
The primary end-point of the study was lesion LC and the secondary endpoint was overall survival, calculated using the Kaplan–Meier method, starting from the end of SBRT treatment (23). Univariate analysis was performed with the log-rank test to investigate the prognostic role of individual variables.
Due to uncertainties about the α/β ratio of sarcomas, we calculated the administered biological equivalent dose (BED) assuming an α/β ratio of both 3 Gy and 10 Gy for all tumors in this study (24, 25).
Results
Twenty-four patients were eligible for this retrospective review. Patients and tumor characteristics are summarized in Table I. The median age was 53 years (range=18-87 years). There were 16 males (66.7%) and 8 females (33.3%). The most common site of primary tumor was the lower extremity (62.5%). Spindle-cell sarcoma (25%) and Ewing Sarcoma (20.8%) were the most frequent histological types. Five patients had LM at diagnosis, whereas the others had developed LM at a median of 21 (range=0-276) months after initial diagnosis. Twenty-two patients underwent at least one thoracic surgery procedure (median of 2.5), and 16 patients had been treated with several lines of chemotherapy. Six patients did not receive any adjuvant or second-line chemotherapy because of comorbidities or because affected by tumors not responding to chemotherapy. Sixty-eight LM were treated with SBRT with a median of one treatment performed for each patient (17 patients: one treatment, 3 patients: two treatments, three patients: 3 treatments, 1 patient: four treatments). The median number of LM treated in each course was one (range=1-4) and the median GTV volume was 5.0 cc (range=0.51-295.7 cc). Figure 1 shows an example of LM from synovial sarcoma treated with SBRT.
The median follow-up after SBRT was 17 months (range=11-51 months). The LC rate at 1 and 2 years were 88.2% and 85.9%, respectively (Figure 2). At the last follow-up, 16 patients were alive and the overall survival rate at 1 and 2 years were 73.1% and 66.4%, respectively (Figure 3).
At univariate analysis, survival was not related to age (p=0.6), gender (p=0.36), prior metastasectomy (p=0.8), number of LM (p=0.73), GTV volume (p=0.6), disease-free interval (p=0.23) or BED (p=0.1). A correlation was observed with GTV volume and LC, GTV ≤5.0 cc being associated with higher LC (p=0.048).
Table II shows univariate summary statistics for LC. The main pattern of failure following SBRT was the appearance of new metastases, since 10 patients presented further secondary lesions after treatment (LM in seven, other sites in three).
The treatment was well-tolerated and no patient experienced acute G3 toxicity. Acute G2 lung toxicity was present in 16.6% of patients and in two patients (8.3%) G1 hemoptysis was observed in the days following SBRT, which receded spontaneously in both cases. One patient presented a G1 rib fracture 19 months after SBRT. No other late complications of the treatment were observed.
Discussion
Metastatic disease has a poor prognosis in cancer, regardless of the histology of the primary tumor. Currently, surgical resection is the cornerstone of management of LM from sarcoma, although there are no randomized trials to compare survival following pulmonary metastasectomy compared to observation. Metastasectomy is associated with 5-year survival rates of 25-38% (4-6). Survival has been related to the number of metastases and to the interval between diagnosis and LM appearance (interval >12 months and <3 metastases) (26).
In recent years, SBRT has been increasingly used in the treatment of patients with early-stage non-small cell lung cancer who are unfit for surgery and is now considered standard-care for these patients due to high rates of LC (8). SBRT has also been utilized in the local treatment of LM and liver metastases from many types of cancer, mainly colon and breast cancer, also providing high LC in these cases and leading to a re-evaluation of the possible role of radiotherapy in patients with oligometastatic disease (9).
There exist only few reports on the role of SBRT in the treatment of metastases from bone or soft-tissue sarcoma. In Table III, some of these studies are summarized. Yu and colleagues reported the results of 58 patients with LM from osteosarcoma. Thirty-one patients underwent metastectomy and the other 27 stereotactic radiotherapy. The results were as follows (surgery vs. stereotactic radiotherapy): 2-year progression-free survival rate: 38.7% vs. 33.3%, 2-year survival: 48.3% vs. 40.7%, median disease-free internval: 8 vs. 5 months and median overall survival of 22 vs. 18 months. However, these differences were not statistically significant (p>0.05). No grade 3 toxicities were reported and grade 1-2 pneumonitis was observed in nine patients (11). Baumann et al. reported the results of 21 patients with 25 LM, treated with SBRT with a median dose of 50 Gy in four (peripheral lesions) or five (central lesions) fractions. One-year LC and OS were 92% and 59%, respectively. The treatment was well-tolerated and only one grade 2 toxicity (rib fracture) was reported (12). Soyfer et al. analyzed 22 patients with 53 LM from soft-tissue sarcoma. The delivered dose was 60 Gy in three fractions in peripheral lesions, while for central lesions or peripheral tumors <1 cm from the chest wall, the dose ranged from 24 Gy in three fractions to 40 Gy in four fractions. According to the diameter of the lesion, the authors divided the lesions into two groups: <10 mm or >10 mm. In the first group (34 lesions <10 mm), 24 lesions achieved a complete response, three a partial response, and seven remained stable. In the second group (18 lesions), five lesions achieved a complete response, five lesions presented progression, and eight lesions were unchanged. Five-year overall survival for the entire group was 50% (13). A study from the Rochester University reported 15 patients with 74 LM who underwent SBRT. The dose delivered was 50 Gy in 5 Gy fractions. Three-year LC was 82%. No patients experienced G3 or more toxicity (14). Stragliotto et al. used SBRT to treat 46 patients with 136 non resectable metastases, mainly in the lung, from primary sarcoma (25). Doses ranged from 10 to 48 Gy in 1-5 fractions prescribed to the 65% isodose. The overall response rate was 88%, with 13 patients surviving more than 36 months. In their experience, there was a trend for an increased local failure rate in patients with larger tumors treated with lower mean doses.
In our series, the overall LC at 2 years was 85.9%, with three patients alive and disease-free at more than 24 months from SBRT. Two-year overall survival was 66.4%; this figure is of course not representative because the treated tumors were highly heterogeneous regarding their histology and natural history. There is, however, a subset of patients with slowly progressive tumors in which SBRT alone can lead to a long survival time with a good control of disease in the chest.
The prescribed dose was generally based on the lesion's size and site, with larger lesions proximal to central structures treated in a greater number of fractions. In patients treated for more than one lesion, the constraints for organs at risk also conditioned the total dose and fraction number, since treatment of multiple sites sometimes made it difficult to contain the dose to lungs or to other structures within the prescribed limits.
Data about radiosensitivity of sarcomas are sparse, although some authors suggest a low alpha/beta ratio, ranging from 0.4 to 5 (25). Furthermore, the heterogeneity of sarcomas, which comprise a wide spectrum of tumors with different proliferation rates and different biological behavior, adds complexity in defining their fractionation sensitivity. In order to circumvent the problem of comparing different total doses and fractionation schemes, we calculated for each fractionation schedule the BED according to the linear quadratic equation (24). We used both an alpha/beta ratio of 3, as proposed by Stragliotto et al. and, due to the uncertainties mentioned before, an alpha/beta ratio of 10 (27). However, in both cases no clear relationship between BED and LC was observed, and the size of the treated lesion remained the most important factor in determining LC.
In our experience, as in those of other authors, toxicity was low. Lung toxicity was observed mainly in patients with larger lesions and was never life-threatening. Rib fracture is a well-known late effect of SBRT for lesions located in proximity to the thoracic wall and can be avoided by conforming the dose distribution with adequate techniques and limiting the volume of rib which receives higher doses (28). The two patients in which hemoptysis was reported both presented metastases from osteosarcoma. The possibility of spontaneous bleeding in LM from osteosarcoma is well known and in CT images it is sometimes possible to observe signs of hemorrhagic alveolitis in the lung surrounding the metastasis (29). However, in both our cases, the symptom receded spontaneously.
SBRT is an attractive and promising alternative to surgery for oligometastatic disease to the lung from sarcoma. The excellent LC and overall survival observed in this and other series confirm that SBRT can confer similar disease-specific and overall survival benefits compared to surgery. This is relevant because 40-80% of patients will experience recurrence in the lungs even after an apparent complete resection of LM. Furthermore, the efficacy of systemic chemotherapy in this setting is limited.
There are several limitations to our and other single-institutional studies, including small sample size, tumor heterogeneity, dose variation, and different selection of patients. However, the safety and efficacy of SBRT consistently reported in literature suggests that it may represent a further therapeutic option in this setting.
Acknowledgements
The Authors thank all the physicians, physicist and radiographers who have been involved in the development of the stereotactic ablative radiotherapy program.
Footnotes
Presented, in part, at the ESTRO forum, Barcelona, Spain, April 24-28, 2015.
Conflicts of Interest
None.
- Received June 2, 2015.
- Revision received July 10, 2015.
- Accepted July 13, 2015.
- Copyright© 2015 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved