Rowe, Bryan P. MD*; Boffa, Daniel J. MD†; Wilson, Lynn D. MD, MPH, FASTRO*; Kim, Anthony W. MD†; Detterbeck, Frank C. MD†; Decker, Roy H. MD, PhD*
Introduction: Patients with centrally located lung tumors have been reported to have a higher risk of toxicity when treated with stereotactic body radiotherapy (SBRT) compared with patients with peripheral tumors. The optimal SBRT fractionation schedule for treatment of central tumors is unknown. The primary purpose of this study was to assess toxicity in patients with central lesions treated with SBRT at our institution, the majority of whom were treated with four fractions.
Methods: Forty-seven patients with 51 central lesions, either primary lung cancer or lung metastases, were treated with SBRT at the Department of Therapeutic Radiology, Yale University School of Medicine/Yale Cancer Center from 2007 to 2011. The patients were treated with three to five fractions with the majority of patients receiving 50 Gy in four fractions of 12.5 Gy. Forty of the lesions were located within 2 cm of the proximal tracheobronchial tree whereas 11 were located within 2 cm of other mediastinal structures. Toxicity data were collected and analyzed according to pretreatment and tumor characteristics and dosimetric parameters. Lobar control data were compiled.
Results: With a median follow-up of 11.3 months (range, 4.8–40.8), four patients experienced grade 3 dyspnea and one patient developed hemoptysis that contributed to respiratory failure and subsequent death. Grade 2 toxicity included fatigue (n = 3), dyspnea (n = 3), chest-wall pain (n = 1), and cough (n = 1). Patients with grade 3+ toxicity had larger maximum tumor diameters compared with those patients without grade 3+ toxicity (median diameter 4.3 cm versus 2.9 cm, p = 0.02). There were no detectable significant differences between the two groups with respect to baseline pulmonary function tests, distance to tracheobronchial tree, maximum point dose to the tracheobronchial tree, maximum dose to 5 cc of the tracheobronchial tree, mean lung dose, and volume of lung receiving 5 Gy, 10 Gy, and 20 Gy. There were two patients who experienced local recurrences. The median biological equivalent dose (linear quadratic formula, α/β = 10) for patients with local recurrence was 76 Gy compared with 112.5 Gy for patients without local recurrence (2-tailed t test, p = 0.04). The 2-year actuarial lobar local control for the entire cohort was 94%. The 2-year lobar local-control rate for patients receiving a biological equivalent dose of 100 Gy or more was 100% and for those receiving less than 100 Gy was 80% (log rank, p = 0.02).
Conclusion: SBRT for central lung tumors seems to be safe, although treatment of larger tumors does carry an increased risk of high-grade toxicity. Efforts to decrease the toxicity risk by decreasing the biologically equivalent dose resulted in increased local failure.
For patients with medically inoperable early-stage non–small-cell lung cancer (NSCLC), stereotactic body radiotherapy (SBRT) has emerged as a promising alternative to conventional fractionated radiotherapy. The goal of SBRT is to deliver tumor-ablative doses of radiation by precisely aiming a limited number (typically 1–5) of large radiation fractions at a tumor plus a small margin. SBRT has been used increasingly to treat medically inoperable early-stage NSCLC and pulmonary oligometastases with excellent results in phase I–II prospective trials.1–5 Radiation Therapy Oncology Group (RTOG) 0236, the landmark phase II trial for medically inoperable early-stage NSCLC, demonstrated a 3-year primary tumor control of 97.6% and a 3-year lobar control rate of 90.6% when a dose of 54 Gy was delivered in three fractions of 18 Gy.
Despite these encouraging results, there has been concern that the subset of patients with central lung lesions is at increased risk for high-grade toxicity. The 4-year results of a phase II trial at the University of Indiana showed that 54 Gy delivered in three fractions of 18 Gy resulted in an almost threefold increase in grade 3–5 toxicity for patients with central (tumor within 2 cm of the proximal bronchial tree) versus peripheral tumors (27.3% versus 10.4%, p = 0.088).6 Although this did not reach statistical significance, the data raised enough concern that patients with central lesions were not enrolled in RTOG 0236. In addition, Song et al. 7 reported on nine patients with central tumors treated with SBRT (40–60 Gy in 3 or 4 fractions) of whom three (33%) developed grade 3–5 pulmonary toxicity.
There is intense interest in identifying an optimal dose-fractionation scheme for central lesions, which is not only biologically potent enough to eradicate tumors but also acceptably safe. The limited available published data suggest that SBRT regimens using more than three fractions for central lesions might be better tolerated.8–10 This is being prospectively studied in RTOG 0813, an ongoing phase I/II trial evaluating five-fraction SBRT regimens for early-stage, centrally located NSCLC in medically inoperable patients. Although accrual for RTOG 0813 is nearing completion, results will not be available for some time. The primary purpose of this study was to assess toxicity in patients with central lesions treated with SBRT at our institution, the majority of whom were treated with four fractions.
PATIENTS AND METHODS
We analyzed 47 consecutive patients with 51 central malignant lesions, either primary or recurrent NSCLC, or pulmonary oligometastases, treated with SBRT at the Department of Therapeutic Radiology, Yale University School of Medicine/Yale Cancer Center from September 2007 to May 2011. Patients were offered SBRT for a variety of reasons, including high surgical risk (cardiac disease, poor pretreatment pulmonary function, inadequate predicted postoperative pulmonary function), patient refusal of surgery, and patients with documented metastatic disease for whom a lobectomy was not considered to be justified. A central lesion was defined as a tumor within 2 cm of the proximal bronchial tree (RTOG definition) or within 2 cm of the heart, great vessels, trachea, or other mediastinal structures.8 Pretreatment evaluation included a history and physical examination by a radiation oncologist as well as an experienced thoracic surgeon, baseline bloodwork, and a complete staging evaluation including computed tomography (CT) of the chest and whole body fluorodeoxyglucose (FDG)-positron emission tomography (PET). Patients were discussed prospectively at our multidisciplinary thoracic oncology tumor board. Of the 47 treated patients, 42 had pathologically proven cancer. The remaining five patients who did not undergo biopsy were presumed to have stage I lung cancer. They were discussed at the multidisciplinary tumor conference and the consensus opinion was that the clinical history and radiographic appearance of the lesion were highly suggestive of lung cancer and that the risk of the biopsy procedure was too high. Patients with primary lung cancer received pulmonary function testing and staging of the mediastinum by mediastinoscopy or endobronchial ultrasound-guided biopsy.
Patients were immobilized in a full-length vacuum cushion and underwent a four-dimensional CT scan (2.5-mm slices) with free breathing. The internal target volume was contoured on the average intensity projection using the Advantage Workstation (GE Healthcare, Waukesha, WI) and was modified to include tumor motion throughout all respiratory phases. In rare cases where tumor excursion exceeded 1 cm, abdominal compression was used.
A uniform 7-mm expansion, accounting for both microscopic spread and intrafraction motion, was added to the internal target volume to create the planning target volume (PTV). The heart, lungs, esophagus, proximal tracheobronchial tree (TBT), spinal cord, and brachial plexus were contoured consistent with guidelines provided by RTOG 0236.
Early on in our experience, the majority of patients were treated with seven to 13 nonopposing, noncoplanar 6 MV photon beams that conformed to the PTV, using a multileaf collimator. More recently, we have primarily been using a modified dynamic conformal arc technique that has been previously described.11 Selected patients with tumors particularly close to the brachial plexus or immediately abutting a bronchus or other mediastinal structure were treated using intensity-modulated radiation therapy. Regardless of the technique used, all plans were normalized such that 95% of the PTV was covered by 100% of the prescription dose and 99% of the PTV was covered by at least 90% of the prescription dose. For both forward- and inverse-planned cases, the prescription dose was between 70% and 90% of the maximum point dose. Dose calculation heterogeneity corrections were performed using the anisotropic analytical algorithm. Cone-beam CT image guidance was used for all cases.
Follow-up consisted of a history and physical examination and noncontrast chest CT scan every 3 months for 2 years and every 6 months thereafter. FDG-PET scans were obtained for patients with enlarging lesions on CT or with consolidation that limited evaluation by CT. All hospital records, follow-up notes, and imaging were reviewed and toxicity scored according to National Cancer Institute common terminology criteria for adverse events, version 4.0. Lobar recurrence was defined as a lesion in the treated lobe that was either proven by biopsy to be recurrent/persistent disease, was enlarging on CT compared with baseline, or had increasing FDG uptake on PET. All questionable cases were discussed at tumor board and either verified by biopsy or by consensus of the tumor board. A variety of patient, tumor, and treatment characteristics were collected, including baseline forced expiratory volume in the first second, baseline carbon monoxide diffusing capacity, corrected for hemoglobin, distance from tumor to the proximal TBT, point dose maximum to the TBT (TBTpoint), maximum dose received by 5 cc of the TBT (TBT5cc), mean lung dose (MLD), volume of lung receiving 5 Gy or more (V5), volume of lung receiving 10 Gy or more (V10), and volume of lung receiving 20 Gy or more (V20).
The rate of lobar local control for the entire cohort was estimated using the Kaplan–Meier method. The biological equivalent dose (BED, linear quadratic equation, α/β = 10) delivered to the two patients with local recurrence was compared with the BED delivered to those without recurrence by use of a two-tailed t test. Lobar local-control rates were also calculated for patients receiving a BED of 100 Gy or more and for those receiving less than 100 Gy. These curves were compared using the log rank test. The patient, tumor, and treatment characteristics listed above were compared between patients who did and did not develop grade 3 or higher toxicity by use of a two-tailed t test.
A total of 47 patients with 51 central lesions were evaluated. Patient and tumor characteristics are listed in Table 1 and treatment characteristics are listed in Table 2. Seventy-five percent of the patients received a BED of 100 Gy or more (range, 60–151.2 Gy) and 57% of the patients were treated with 12.5 Gy × 4. Twenty-five percent of the patients were treated with a BED of less than 100 Gy in an effort to limit normal tissue dose. With a median follow-up of 11.3 months (range, 4.8–40.8 months), five patients experienced grade 3 or higher toxicity (Table 3). One patient suffered grade 5 toxicity and it was determined that SBRT likely contributed to his death. This patient was a 75-year-old man with metastatic melanoma with a 5.7-cm dominant pulmonary metastasis abutting the left mainstem bronchus. He developed hemoptysis approximately 10.5 months after completing SBRT. He was hospitalized and eventually developed a collapsed lung and hypoxemia requiring intubation. He ultimately died of respiratory failure.
A total of four patients developed grade 3 dyspnea (shortness of breath at rest) after completion of SBRT. Three of these patients had comorbid chronic obstructive pulmonary disease with supplemental oxygen required at baseline. All three developed an acute increase in their oxygen requirement 2 to 4 months after completion of SBRT. They were briefly hospitalized and treated with bronchodilators and steroid taper with improvement back to their baseline within 1 to 2 weeks. The fourth patient had metastatic melanoma and developed cough and shortness of breath 3 months after treatment. He was treated for presumed radiation pneumonitis and improved with a steroid taper over 4 weeks.
We compared the group of patients with grade 3–5 toxicity versus the group without grade 3–5 toxicity and found no significant differences in baseline forced expiratory volume in the first second (percentage predicted), baseline carbon monoxide diffusing capacity, corrected for hemoglobin (percentage predicted), distance from tumor to the proximal TBT, TBTpoint, TBT5cc, MLD, V5, V10, or V20. Patients who developed grade 3–5 toxicity were found to have significantly larger maximum tumor diameters (Table 4). The patient with grade 5 toxicity had a TBTpoint of 54.2 Gy, which was the fifth highest point dose of the patients examined. His TBT5cc dose was 12.7 Gy, approximately the median.
Two patients had local recurrences a median of 26 months after treatment. The median BED for the patients with local recurrence was 76 Gy (range, 72–80), compared with the patients without local failure who had a median BED of 112.5 Gy (range, 60–151.2; p = 0.04). The 2-year actuarial lobar local-control rate for the entire cohort was 94% (Fig. 1). The 2-year lobar local-control rate for patients receiving a BED of 100 Gy or more was 100% and for those receiving less than 100 Gy was 80% (log rank, p = 0.02).
The optimal dose-fractionation regimen for central lung tumors is not currently known. RTOG 0813, which is investigating five-fraction SBRT regimens for centrally located NSCLC in medically inoperable patients, will provide important prospective data in this regard. Several groups have reported on their retrospective experience treating central lesions with SBRT. Chang et al.8 from MD Anderson Cancer Center reported on 27 patients with central or superiorly located NSCLC treated with 40 to 50 Gy in four fractions. Four patients (28.6%) with recurrent disease but none with stage I disease developed grade 2 pneumonitis. A total of three patients (11.1%) developed grade 2–3 dermatitis and chest-wall pain and one patient with a significant volume of brachial plexus receiving 40 Gy developed a brachial plexopathy. Bral et al.9 from Belgium reported on 17 patients with central lesions treated with 15 Gy × 4 fractions. They found a correlation between tumor location and the development of acute or late grade 3+ pulmonary toxicity, with 2-year lung-toxicity–free survival of 84% for peripheral lesions versus 60% for central lesions. Haasbeek et al.12 from the Netherlands recently reported on 63 patients with tumors in a central hilar location or abutting mediastinal structures, who received 7.5 Gy × 8 fractions. Toxicity was minimal, with only one patient (2%) experiencing grade 3 acute toxicity (chest-wall pain) and four patients (6%) developing grade 3 late toxicity.
In our series, we did have one patient with toxicity from SBRT (bronchopulmonary hemorrhage), which contributed to his death. Of note, a bronchoscopy performed during his hospital admission revealed bronchial necrosis in the left mainstem bronchus, 2 cm below the carina (Fig. 2). The area of necrosis does correspond to the maximum point dose of 54.2 Gy, suggesting that the TBTpoint might be an important dose–volume parameter to prevent bronchial necrosis and hemoptysis. The four patients in our series with TBTpoint doses higher than 54.2 Gy have shown no signs of toxicity, with the longest follow-up among those patients being 15 months.
We also identified four patients who developed grade 3 dyspnea (shortness of breath at rest) 2 to 4 months post-SBRT. It is possible that all four cases represented radiation pneumonitis. Barriger et al.13 have previously reported that V20 and MLD were significant dosimetric parameters related to the development of radiation pneumonitis after SBRT. Our analysis did not reveal a correlation between any of the lung or TBT dose parameters and grade 3+ toxicity, although we did find that patients with grade 3+ toxicity had larger tumors, suggesting an underlying dose–volume correlation to toxicity that was not evident in our analysis.
Our data is consistent with previously published reports suggesting that a BED of 100 Gy is an important cutoff for local control.8,14,15 There have been no local recurrences in our series for patients who were treated to a BED of 100 Gy or more. It is notable that patients treated to a BED of less than 100 Gy in an effort to reduce toxicity had higher local-failure rates. Our primary reason for analyzing local-control outcomes was to compare the relative efficacy of the different dose-fractionation regimens. It is important to note that more than half the patients in this cohort had either metastatic or recurrent disease and therefore had limited survival and were censored from the local-control analysis at the time of their death. It is likely that the local control will be lower with longer follow-up.
Our study is subject to all the usual limitations and biases present in retrospective studies and should, therefore, be interpreted as hypothesis generating. Although we thoroughly reviewed all available records for toxicity information, it is possible that toxicity events, especially grade 1 or 2 events, were missed. For this reason, we did not include grade 1 events in our report. In addition, because the number of grade 3 or higher events was low (n = 5), and our patient population relatively small and heterogeneous, our ability to correlate high-grade toxicity with clinical factors is limited. Last, because late toxicity—such as bronchial stenosis, hemoptysis, or decline in pulmonary function—can manifest beyond the median follow-up time for our cohort, toxicity rates should be interpreted with caution and longer follow-up will be necessary.
Overall, the toxicity with this regimen seems to be acceptable, although caution needs to be exercised when considering SBRT for larger central tumors. Given the reasonable toxicity and increased risk of local failure with attenuated doses in our series, we will be more inclined to treat central lesions to full dosages (BED ≥100 Gy) in the future. Further work is needed to define optimal dose–volume constraints. Long-term follow-up of these patients is obviously needed to confirm the results.
SBRT for central lung tumors seems to be safe, although treatment of larger tumors does carry an increased risk of high-grade toxicity. Efforts to decrease the toxicity risk by decreasing the biologically equivalent dose resulted in increased local failure.
1. Timmerman R, Paulus R, Galvin J, et al. Stereotactic body radiation therapy for inoperable early stage lung cancer. JAMA. 2010;303:1070–1076
2. Rusthoven KE, Kavanagh BD, Cardenes H, et al. Multi-institutional phase I/II trial of stereotactic body radiation therapy for liver metastases. J Clin Oncol. 2009;27:1572–1578
3. Fakiris AJ, McGarry RC, Yiannoutsos CT, et al. Stereotactic body radiation therapy for early-stage non-small-cell lung carcinoma: four-year results of a prospective phase II study. Int J Radiat Oncol Biol Phys. 2009;75:677–682
4. Ricardi U, Filippi AR, Guarneri A, et al. Stereotactic body radiation therapy for early stage non-small cell lung cancer: results of a prospective trial. Lung Cancer. 2010;68:72–77
5. Baumann P, Nyman J, Hoyer M, et al. Outcome in a prospective phase II trial of medically inoperable stage I non-small-cell lung cancer patients treated with stereotactic body radiotherapy. J Clin Oncol. 2009;27:3290–3296
6. Timmerman R, McGarry R, Yiannoutsos C, et al. Excessive toxicity when treating central tumors in a phase II study of stereotactic body radiation therapy for medically inoperable early-stage lung cancer. J Clin Oncol. 2006;24:4833–4839
7. Song SY, Choi W, Shin SS, et al. Fractionated stereotactic body radiation therapy for medically inoperable stage I lung cancer adjacent to central large bronchus. Lung Cancer. 2009;66:89–93
8. Chang JY, Balter PA, Dong L, et al. Stereotactic body radiation therapy in centrally and superiorly located stage I or isolated recurrent non-small-cell lung cancer. Int J Radiat Oncol Biol Phys. 2008;72:967–971
9. Bral S, Gevaert T, Linthout N, et al. Prospective, risk-adapted strategy of stereotactic body radiotherapy for early-stage non-small-cell lung cancer: results of a Phase II trial. Int J Radiat Oncol Biol Phys. 2011;80:1343–1349
10. Lagerwaard FJ, Haasbeek CJ, Smit EF, Slotman BJ, Senan S. Outcomes of risk-adapted fractionated stereotactic radiotherapy for stage I non-small-cell lung cancer. Int J Radiat Oncol Biol Phys. 2008;70:685–692
11. Ross CC, Kim JJ, Chen ZJ, et al. A novel modified dynamic conformal arc technique for treatment of peripheral lung tumors using stereotactic body radiation therapy. Pract Rad Oncol. 2011;1:126–134
12. Haasbeek CJ, Lagerwaard FJ, Slotman BJ, Senan S. Outcomes of stereotactic ablative radiotherapy for centrally located early-stage lung cancer. J Thorac Oncol. 2011;6:2036–2043
13. Barriger RB, Forquer JA, Brabham JG, et al. A dose-volume analysis of radiation pneumonitis in non-small cell lung cancer patients treated with stereotactic body radiation therapy. Int J Radiat Oncol Biol Phys. 2012;82:457–462
14. Onishi H, Shirato H, Nagata Y, et al. Hypofractionated stereotactic radiotherapy (HypoFXSRT) for stage I non-small cell lung cancer: updated results of 257 patients in a Japanese multi-institutional study. J Thorac Oncol. 2007;2(7 Suppl 3):S94–100
15. Olsen JR, Robinson CG, El Naqa I, et al. Dose-response for stereotactic body radiotherapy in early-stage non-small-cell lung cancer. Int J Radiat Oncol Biol Phys. 2011;;81(4):e299–303