Purpose: To examine the effect of baseline forced expiratory volume in 1 second (FEV1) and diffusion capacity for carbon monoxide (DLco) on posttreatment survival and pulmonary function decrease after stereotactic body radiotherapy (SBRT) for patients with early-stage non–small-cell lung cancer (NSCLC). Methods and Materials: Seventy medically inoperable patients with Stage I NSCLC were treated with definitive SBRT to a dose of 6,000 (Stage IA) or 6,600 cGy (Stage IB), given in three equal fractions. Baseline and serial posttreatment pulmonary function data were collected. Results: Median age was 70.5 years, and median follow-up was 2.17 years. Median pretreatment FEV1 and DLco were 1.05 L and 10.06 mg/min/mm Hg, respectively. There was no significant decrease in survival in patients with baseline FEV1 and DLco less than the median value and less than the lowest quartile, whereas patients with values greater than the highest quartile of baseline FEV1 had significantly inferior survival. There was no significant effect of pretreatment FEV1 or DLco on posttreatment levels. There was a statistically significant decrease in DLco of 1.11 mg/min/mm Hg/y. Conclusions: Poor baseline pulmonary function did not predict decreased survival or pulmonary function after treatment. A statistically significant decrease in DLco after treatment was seen, similar to decreases seen in studies delivering standard thoracic radiotherapy. We conclude that low pretreatment FEV1 and/or DLco alone should not be used to exclude patients with NSCLC from treatment with SBRT. � 2008 Elsevier Inc. Stereotactic body radiotherapy, Non–small-cell lung cancer, Pulmonary function, Toxicity. INTRODUCTION The present standard of care for patients with American Joint Committee on Cancer Stage I non–small-cell lung cancer (NSCLC) (1) is surgical resection. Primary surgical therapy in patients with Stage I NSCLC results in a favorable 5-year overall-survival (OS) rate of 60–70% (2–5). Unfortu- nately, significant cardiovascular and pulmonary comorbid- ities often are present in this population, and many patients are medically inoperable. At some institutions, more than 60% of patients with early-stage lung cancer are medically inoperable or refuse surgical intervention (6). Traditional nonsurgical options for patients with early-stage NSCLC include definitive radiotherapy and observation. Although conventional radiotherapy to doses of 50–70 Gy is potentially curative, with 5-year OS rates in the range of 15–30% (7–9), these results are clearly inferior to surgery. Observation likewise portends a poor outcome. Even in this population with significant competing risks for death, more than half the untreated patients will die of lung cancer (6). Stereotactic body radiotherapy (SBRT) is an emerging technique for the treatment of patients with Stage I NSCLC. In a recent large multi-institutional series from Japan, patients Reprint requests to: Ronald McGarry, M.D., Ph.D., University of Supported by Grant No. 5R21CA097721-02 from the US CLINICAL INVESTIGATION BASELINE PULMONARY FUNCTION A DECLINE IN PULMONARY FUNCTION O STEREOTACTIC BODY RADIOTHERAP NON–SMALL-CEL MARK HENDERSON, M.D.,* RONALD MCGARRY, M ACHILLES FAKIRIS, M.D.,* DAVID HOO AND ROBERT TIM *Department of Radiation Oncology and zDivision of Biostatis yDepartment of Radiation Medicine, University of Kentucky, Lexing Center, Indianapolis, IN; and kDepartment of Radiation Oncology, doi:10.1016/j.ijro Kentucky, Department of Radiation Medicine, Lexington, KY 40536. Tel: (859) 323-6880; Fax: (859) 257-7483; E-mail: ronald.
[email protected] Presented in abstract form as an oral presentation at the 48th Annual Meeting of the American Society of Therapeutic Radiol- ogy and Oncology (ASTRO), November 5–9, 2006, Philadel- phia, PA. 40 Lung S A PREDICTOR FOR SURVIVAL AND VER TIME IN PATIENTS UNDERGOING Y FOR THE TREATMENT OF STAGE I L LUNG CANCER .D., PH.D.,y CONSTANTIN YIANNOUTSOS, PH.D.,z PES, M.D.,* MARK WILLIAMS, M.D.,x MERMAN, M.D.k tics, Indiana University School of Medicine, Indianapolis, IN; ton, KY; xPulmonary Division, Richard L. Roudebush VA Medical The University of Texas Southwestern Medical Center, Dallas, TX Int. J. Radiation Oncology Biol. Phys., Vol. 72, No. 2, pp. 404–409, 2008 Copyright � 2008 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/08/$–see front matter bp.2007.12.051 National Institutes of Health. ClinicalTrials.gov identifier: NCT00246181. Conflict of interest: none. Acknowledgments—We thank Kathy Tudor and Jill DeLuca for invaluable help in data collection. Received June 6, 2007, and in revised form Dec 16, 2007. Accepted for publication Dec 18, 2007. 4 a cohort of medically inoperable patients with NSCLC treated with SBRT. METHODS AND MATERIALS Patients Full eligibility requirements were published previously (17). To be eligible, patients had to be 18 years or older, have American Joint Committee on Cancer Stage IA or IB NSCLC, be technically resect- able but deemed medically inoperable by an experienced thoracic surgeon, have Karnofsky performance status of 60 or higher, and sign study-specific informed consent paperwork. All patients were Table 1. Patien Characteristics All patients (N = 70) FEV1 < bottom quartile (N = 16/67)* Sex Men 34 (48.6) 4 (25.0) Women 36 (51.4) 12 (75.0) Race/ethnicity Caucasian 62 (88.6) 13 (81.3) Other 8 (11.4) 3 (18.7) Stage T1 34 (48.6) 7 (43.7) T2 36 (51.4) 9 (56.3) Reason for inoperabilityz Cardiac morbidity N = 17 0 (0.0) Pulmonary morbidity N = 51 15 (100.0) Age at first treatment (y) 70.5 (66.0–77.0) 68.0 (59.0–71.5) Planning treatment volume (cm3) 39.1 (25.6–69.2) 50.6 (26.0–90.8) Abbreviations: FEV1 = forced expiratory volume in 1 second, DLco Sex, race, stage, reason for inoperability expressed as number (pe type indicates statistically significant comparison at the 5% a level (Fi variables). * Three subjects were missing baseline FEV1 observations.y Seven subjects were missing baseline DLco observations. z Reason for inoperability was available for only 68 patients. NSCLC. Patients with Stage IB received 66 Gy in three 22-Gy frac- tions. The three fractions were separated by a minimum of 2 and a maximum of 8 days, with most fractions delivered 2–3 days apart. This protocol was reviewed and approved by the IU Purdue Univer- sity Indianapolis and Clarian Institutional Review Boards. Follow-up Pulmonary function tests (including spirometry, lung volumes, DLco, and arterial blood gas) were performed before treatment. The same studies were repeated at 6 weeks after radiation, then ev- ery 3 months for the first 2 years, every 6 months for Years 3 and 4, and then annually. Per protocol, a significant change in pulmonary function values was defined as a greater than 15% change. t characteristics FEV1 > top quartile (N = 17/67) DLco < bottom quartile (N = 17/63)y DLco > top quartile (N = 16/63) 10 (58.8) 7 (41.2) 10 (62.5) 7 (41.2) 10 (58.8) 6 (37.5) 16 (94.1) 14 (82.4) 16 (100.0) 1 (5.9) 3 (17.6) 0 (0.0) 7 (41.2) 10 (58.8) 8 (50) 10 (58.8) 7 (41.2) 8 (50.0) 9 (56.3) 0 (0.0) 11 (73.) 7 (43.7) 16 (100.0) 4 (26.7) 72.0 (65.0–78.0) 70.0 (63.0–74.0) 73.5 (69.5–78.0) 38.3 (26.1–59.9) 31.1 (24.5–57.1) 33.9 (14.6–50.9) = diffusion capacity for carbon monoxide. rcent); age and PTV expressed as median (interquartile range). Bold sher’s exact test, categorical variables; Kruskal-Wallis test, continuous with Stage I NSCLC treated with SBRT had a 5-year OS rate of 47%; medically operable patients in this series had a 5-year OS rate of approximately 80% (10). Total doses in the range of 48–60 Gy in four to 10 fractions are routinely used in Japan (11–13). A Phase I dose-escalation trial at Indiana University (IU) established 60 Gy in three fractions as safe (14, 15), and subsequent IU and Radiation Therapy Oncol- ogy Group Phase II trials adopted similar dose and fraction- ation schemes. Some investigators questioned the wisdom of such large fraction sizes in a patient population with extremely poor baseline pulmonary function. This is a reason- able question because SBRT is an ablative procedure, and surgical data indicated that the predicted diffusion capacity for carbon monoxide (DLco) after resection was a predictor of postoperative complications (16). However, few data exist about the effect of SBRT on pulmonary function (14). The IU Phase II trial of SBRT in medically inoperable patients with Stage I NSCLC mandated detailed serial pulmonary function testing in all patients. The purpose of this report is to determine whether pretreatment values of forced expiratory volume in 1 minute (FEV1) or DLco predict for survival or pulmonary function decrease after treatment in staged by using positron emission tomography (PET) and computed tomography (CT) scans (preferably combined PET/CT). Any medi- astinal lymphadenopathy seen on PET or CT had to be biopsy neg- ative for the patient to be eligible for treatment on study. The primary tumor had to be 7 cm or less in greatest dimension. Radiation therapy A detailed description of our SBRT technique was published else- where (14). Briefly, patients were positioned in a Stereotactic Body Frame (Elekta, Stockholm, Sweden) using a vacuum pillow to make the immobilization reproducible. Prepatellar and sternal positioning marks were permanently applied. Abdominal pressure was applied until the motion of the diaphragm was decreased to 1.0 cm or less, as seen on fluoroscopy. The CT simulation was performed, and gross tumor volume was defined as visible tumor (excluding at- electasis). Planning target volume consisted of gross tumor volume with a 0.5-cm radial and 1.0-cm superior-inferior margin. Three-di- mensional conformal radiotherapy was used, with 7–10 noncopla- nar compensated beams. Treatment was delivered with 6 (and, less frequently, 15) MV photons using the x, y, and z coordinates provided by the stereotactic frame to target the planning target vol- ume. Dose was prescribed to the 80% isodose line. A dose of 60 Gy in three 20-Gy fractions was delivered to patients with Stage IA Changes in pulmonary function after SBRT d M. HENDERSON et al. 405 Statistics Estimates of subject survival distribution were obtained by using the Kaplan-Meier method. Survival distributions were compared be- tween groups by means of the log-rank test. Associations between categorical variables were assessed using Fisher’s exact test. Com- parisons between continuous variables were performed using the Kruskal-Wallis test. Two random-effect longitudinal models (18) were fit to the posttreatment FEV1 and DLco data, with the pretreat- ment measurement and weekly trend as the predictors of posttreat- ment FEV1 and DLco levels. A line was fit to represent FEV1 or DLco levels over time. A patient-specific intercept and slope of the line was allowed (the ‘‘random’’ component of the random-effect model). All analyses were performed using SAS System, version 9.1 (SAS Institute, Cary, NC). Baseline patient characteristics were evaluated by using Cox univariate and multivariate analyses to determine their effect on patient survival and time to first Grade 2 or greater pulmonary toxicity. p values# 0.05 are reported as sta- tistically significant in all cases. RESULTS From January 2002 to September 2004, a total of 70 patients were enrolled in the IU Phase II protocol. Median age at treatment was 70 years (range, 51–86 years). At the time of this analysis, median follow-up was 2.17 years (range, 0.12–3.62 years). Median pretreatment FEV1 was 1.05 L (range, 0.29–2.12 L), and median pretreatment DLco was 10.06 ml/min/mm Hg (range, 3.50–23.05 ml/min/mm Hg). Patient demographics are listed in Table 1. In the top quartile of pretreatment FEV1 and DLco, significantly more patients had cardiac comorbidity as the reason for inoperabil- ity than in the bottom three quartiles (Fisher’s exact test, p = 0.002 and p < 0.001 for FEV1 and DLco, respectively). The bottom quartile of pretreatment FEV1 contained fewer men and had a younger age at first treatment than patients in the highest quartile of pretreatment FEV1 (p < 0.05). Median FEV1 at baseline in patients who were inoperable because of pulmonary comorbidities was 0.85 L, whereas it was 1.54 L in those inoperable because of cardiac comorbidities (p < 0.001). Baseline DLco for patients who were inoperable because of pulmonary comorbidities was 9.25 ml/min/mm Hg, whereas it was 10.57 ml/min/mm Hg for those inopera- Fig. 1. Kaplan-Meier survival curve of patients with values greater (dashed) and less than (solid) the median for pretreatment forced 406 I. J. Radiation Oncology d Biology d Physics expiratory volume in 1 second (FEV1; p = 0.659). F/U = follow-up. ble because of cardiac comorbidities (p < 0.001). Additional patient demographics and data for pretreatment pulmonary function tests were published separately (17). To facilitate interpretation of the data, patients were divided into quartiles based on baseline FEV1 and DLco values. Patients with pretreatment FEV1 less than the median value had a median survival of 83 weeks vs. 82.1 weeks for patients with pretreatment FEV1 greater than the median (Fig. 1; p = 0.659). Patients in the bottom quartile of pretreat- ment FEV1 (1.39 L) had statistically significantly decreased survival vs. those with values less than the highest quartile (45.9 vs. 95.0 weeks, respectively; Fig. 3; p = 0.049). Pretreatment FEV1 was strongly related to posttreatment FEV1 levels (Table 2; p < 0.001). There was a small decrease over time that was not statistically significant. Pretreatment FEV1 was not significantly associated with posttreatment DLco levels (data not shown; p = 0.237). Patients with pretreatment DLco less than the median (10.06 ml/min/mm Hg) had a median survival of 80.3 weeks, whereas those with DLco greater than the median value had Fig. 2. Kaplan-Meier survival curve of patients with values greater (dashed) and less than (solid) the bottom quartile of pretreatment forced expiratory volume in 1 second (FEV1; p = 0.379). F/U = follow-up. Fig. 3. Kaplan-Meier survival curve of patients with values greater (dashed) and less than (solid) the highest quartile for pretreatment forced expiratory volume in 1 second (FEV1; p = 0.049). Volume 72, Number 2, 2008 F/U = follow-up. a median survival of 73.4 weeks (Fig. 4; p = 0.789). Patients with a pretreatment DLco in the lowest quartile (13.67 ml/min/mm Hg) had a median survival of 53.6 weeks vs. 93.0 weeks for those with DLco less than the top quartile (Fig. 6; p = 0.339). Pre- treatment DLco was linearly related to posttreatment DLco levels (Table 2; p < 0.001). There was a statistically signifi- cant decrease of 1.11 ml/min/mm Hg/y (Table 2; p < 0.001). On Cox univariate analysis, baseline DLco, baseline FEV1, cardiac morbidity, and tumor stage (T stage) were evaluated for their effect on patient survival. None of these characteris- tics statistically significantly predicted for patient survival after treatment. However, on multivariate analysis, baseline FEV1 significantly predicted for posttreatment survival (Table 3). Multivariate analysis indicates that for each 1-L increase in baseline FEV1, the patient has an approximately three times greater risk of death after treatment. Cox analysis was also used to determine whether baseline DLco or baseline FEV1 predicted for time to the development of the first Grade 2 or higher pulmonary toxicity (scores were recorded for pneumonitis, pneumonia, and ‘‘other pulmonary toxicity’’). Neither baseline FEV1 (p = 0.192) nor baseline Table 2. Effect of pretreatment FEV1 and DLco on posttreatment values Outcome Predictor Mean � SE p* Follow-up FEV1 Baseline FEV1 0.895 � 0.034 Patients who were inoperable because of cardiac comorbid- ities had numerically inferior survival to other patients (median survival, 92.9 vs. 146.1 weeks, respectively). How- ever, neither univariate nor multivariate analysis showed sta- tistically significantly lower survival than in patients inoperable because of pulmonary morbidity. This may be caused by the small number of events. Specific cardiac co- morbidities were not recorded as part of this trial; therefore, no further investigation of reasons for increased mortality in these patients was possible. A dose–volume analysis of pa- tients with cardiac comorbidities who received SBRT might be able to determine whether there is an interaction of SBRT with the patient’s underlying heart disease. It seems more likely that patients with poor cardiovascular health are at increased risk of death at baseline. Although few data are available in the literature regarding pulmonary function after SBRT for patients with early-stage lung cancer, several studies examined the effects of radio- therapy and chemoradiotherapy on pulmonary function in pa- tients with NSCLC treated with standard radiotherapy techniques. Few of these studies had long follow-up, which is not surprising considering the natural history of NSCLC. Of studies with pulmonary function data from 1 year or more after treatment, FEV1 was shown to remain relatively stable over time after treatment (after a brief decrease and re- REFER 1. Greene FL, Page DL, Fleming ID, et al. AJCC Cancer staging manual. 6th ed. New York: Springer-Verlag; 2002. 2. Harpole DH Jr., Herndon JE II, Young WG Jr., et al. Stage I non-small cell lung cancer. A multivariate analysis of treat- ment methods and patterns of recurrence. Cancer 1995;76: 787–796. 3. Martini N, Bains MS, Burt ME, et al. Incidence of local recurrence and second primary tumors in resected stage I lung cancer. J Thorac Cardiovasc Surg 1995;109:120–129. 4. Mountain CF. A new international staging system for lung cancer. Chest 1986;89(Suppl.):225S–233S. 5. Williams DE, Pairolero PC, Davis CS, et al. Survival of patients surgically treated for stage I lung cancer. J Thorac Cardiovasc Surg 1981;82:70–76. 6. McGarry RC, Song G, des Rosiers P, et al. Observation-only management of early stage, medically inoperable lung cancer: Poor outcome. Chest 2002;121:1155–1158. approximately 11% at 1 year. Limited data are available to predict the clinical significance of radiation toxicity based on pulmonary function testing after treatment. With a median follow-up of 7 months, a small retrospective study by Gopal et al. (26) indicated that the rate of Grade 2 or higher pulmo- nary toxicity (scored using Radiation Therapy Oncology Group criteria for chronic pulmonary toxicity [24]) is greatest after decreases in DLco greater than 30% of baseline. Addi- tional follow-up will be needed to better describe the decrease in DLco over time and determine whether it proves to be clinically significant in our study population. It would be interesting to compare the decrease in DLco over time in our study population with an age- and comorbidity-matched population of patients who have not received thoracic radiotherapy. CONCLUSIONS Neither very low baseline FEV1 nor very low baseline DLco predict for decreased survival or more abrupt posttreat- ment decrease in pulmonary function. The DLco may be the most useful pulmonary function test to follow up radiation lung injury in future studies. Patients with early-stage NSCLC should not be excluded from treatment with SBRT based on very low baseline FEV1 or DLco. ENCES 7. Gauden S, Ramsay J, Tripcony L. The curative treatment by radiotherapy alone of stage I non-small cell carcinoma of the lung. Chest 1995;108:1278–1282. 8. Hayakawa K, Mitsuhashi N, Furuta M, et al. High-dose radiation therapy for inoperable non-small cell lung cancer without mediastinal involvement (clinical stage N0, N1). Strahlenther Onkol 1996;172:489–495. 9. Sibley GS. Radiotherapy for patients with medically inoperable stage I nonsmall cell lung carcinoma: Smaller volumes and higher doses—A review. Cancer 1998;82:433–438. 10. Onishi H, Araki T, Shirato H, et al. Stereotactic hypofractio- nated high-dose irradiation for stage I non-small cell lung carcinoma: Clinical outcomes in 245 subjects in a Japanese multiinstitutional study. Cancer 2004;101:1623–1631. 11. Fukumoto S, Shirato H, Shimzu S, et al. Small-volume image-guided radiotherapy using hypofractionated, coplanar, turn to baseline) (19–21). The DLco was found to more often be significantly decreased after treatment with radiotherapy. Our analysis of patients with low baseline DLco showed no significant decrease in survival compared with patients with greater baseline DLco. Pretreatment DLco was linearly related to posttreatment DLco. There was a statistically significant decrease in posttreatment DLco of 1.11 ml/min/mm Hg/y in our SBRT patient population. Population-based studies showed that the decrease in DLco in 70-year-old patients not receiving thoracic radiotherapy was 0.394–0.63 ml/ min/mm Hg/y (22, 23). There are few published series with long follow-up that detail decreases in DLco over time after thoracic radiotherapy. Available data would indicate that decreases of 10–27% could be expected at 1 year (19, 20, 24, 25). The decrease in our study population would be Table 3. Cox multivariate analysis of effects of baseline patient characteristics on survival Characteristics p Hazard ratio Baseline FEV1 0.014 3.26* Baseline DLco 0.364 0.94 Cardiac morbidity 0.579 0.72 Tumor stage 0.109 2.07 Abbreviations: FEV1 = forced expiratory volume in 1 second, DLco = diffusion capacity for carbon monoxide. Bold type indicates statistically significant comparison at the 5% a level (Fisher’s exact test, categorical variables; Kruskal-Wallis test, continuous variables). * Increasing baseline FEV1 was associated significantly with increased risk of death. 408 I. J. Radiation Oncology d Biology d Physics Volume 72, Number 2, 2008 and noncoplanar multiple fields for patients with inoperable stage I nonsmall cell lung carcinomas. Cancer 2002;95: 1546–1553. 12. Nagata Y, Takayama K, Matsuo Y, et al. Clinical outcomes of a phase I/II study of 48 Gy of stereotactic body radiotherapy in 4 fractions for primary lung cancer using a stereotactic body frame. Int J Radiat Oncol Biol Phys 2005;63:1427–1431. 13. Uematsu M, Shioda A, Suda A, et al. Computed tomography- guided frameless stereotactic radiotherapy for stage I non-small cell lung cancer: A 5-year experience. Int J Radiat Oncol Biol Phys 2001;51:666–670. 14. Timmerman R, Papiez L, McGarry R, et al. Extracranial stereotactic radioablation: Results of a phase I study in medically inoperable stage I non-small cell lung cancer. Chest 2003;124:1946–1955. 15. McGarry RC, Papiez L, Williams M, et al. Stereotactic body radiation therapy of early-stage non-small-cell lung carcinoma: Phase I study. Int J Radiat Oncol Biol Phys 2005;63: 1010–1015. 16. Takeda S, Funakoshi Y, Kadota Y, et al. Fall in diffusing capacity associated with induction therapy for lung cancer: A predictor of postoperative complication? Ann Thorac Surg 2006;82:232–236. 17. Timmerman R, McGarry R, Yiannoutsos C, et al. Exces- sive 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. 18. Laird NM, Ware JH. Random-effects models for longitudinal data. Biometrics 1982;38:963–974. 19. Abratt RP, Willcox PA. The effect of irradiation on lung function and perfusion in patients with lung cancer. Int J Radiat Oncol Biol Phys 1995;31:915–919. 20. Choi NC, Kanarek DJ. Toxicity of thoracic radiotherapy on pulmonary function in lung cancer. Lung Cancer 1994; 10(Suppl. 1):S219–S230. 21. Mehta V. Radiation pneumonitis and pulmonary fibrosis in non-small-cell lung cancer: Pulmonary function, prediction, and prevention. Int J Radiat Oncol Biol Phys 2005;63:5–24. 22. Sherrill DL, Enright PL, Kaltenborn WT, et al. Predictors of longitudinal change in diffusing capacity over 8 years. Am J Respir Crit Care Med 1999;160:1883–1887. 23. Viegi G, Sherrill DL, Carrozzi L, et al. An 8-year follow-up of carbon monoxide diffusing capacity in a general population sample of northern Italy. Chest 2001;120:74–80. 24. Gopal R, Starkschall G, Tucker SL, et al. Effects of radiother- apy and chemotherapy on lung function in patients with non-small-cell lung cancer. Int J Radiat Oncol Biol Phys 2003;56:114–120. 25. Miller KL, Zhou SM, Barrier RC Jr., et al. Long-term changes in pulmonary function tests after definitive radiotherapy for lung cancer. Int J Radiat Oncol Biol Phys 2003;56:611–615. 26. Gopal R, Tucker SL, Komaki R, et al. The relationship between local dose and loss of function for irradiated lung. Int J Radiat Oncol Biol Phys 2003;56:106–113. Changes in pulmonary function after SBRT d M. HENDERSON et al. 409 Baseline Pulmonary Function as a Predictor for Survival and Decline in Pulmonary Function Over Time in Patients Undergoing Stereotactic Body Radiotherapy for the Treatment of Stage I Non-Small-Cell Lu ng Cancer Introduction Methods and Materials Patients Radiation therapy Follow-up Statistics Results Discussion Conclusions References