- Email: [email protected]
Carboplatin (C) in Patients (Pts) With Select Stage IB, II, and III Nonsquamous Non-Small Cell Lung Cancer (NS-NSCLC)
Volume 90 Number 5S Supplement 2014
Poster Presentations
the risk for developing a SPLC remains elevated beyond 2 years, particularly in former and current smokers. Therefore, continued surveillance beyond 2 years should be maintained. Author Disclosure: A. Rimner: E. Research Grant; Varian Medical Systems. G. Consultant; Varian Medical Systems, General Electric. V. Adeseye: None. A. Foster: None. K.M. Woo: None. F. Shaikh: None. S. Din: None. Z. Zhang: None. E.D. Yorke: None. R. Gewanter: None. K.E. Rosenzweig: None. J. Huang: None. A.J. Wu: G. Consultant; Pfizer.
121 Predictors of 30-Day Mortality Following Resection of Early-Stage NSCLC: An Analysis of the National Cancer Data Base Early-Stage Non-Small Cell Lung Cancer Z.A. Husain, A.W. Kim, J.B. Yu, R.H. Decker, and C. Corso; Yale University School of Medicine, New Haven, CT Purpose/Objective(s): Studies examining morbidity after lobectomy for early stage NSCLC have shown that patients over the age of 65 have a > 50% incidence of complications, and this risk increases significantly in elderly patients (age 75). Factors that affect postsurgical 30-day mortality (30-DM), however, are less well defined. Materials/Methods: The National Cancer Data Base (NCDB), was used to identify patients 19 years and older with a diagnosis of stage I NSCLC (cT1-2, cN0, cM0) between the years 2003 and 2011. Data from patients who underwent surgical resection with lobectomy or sublobar resection were abstracted. Univariate and multivariate analysis was performed for predictors of 30-DM. Factors examined included: age, sex, year of diagnosis, tumor size, and Charlson-Deyo (CD) morbidity score. Results: A total of 112,216 patients met the criteria. Of these 71,175 or 64% received surgery, and of this group 57,569 (81%) underwent lobectomy and 13,606 (19%) underwent sublobar resection. Pneumonectomy patients were excluded. The median age was 68 years (53% female). Median tumor size (Tsize) was 2.4cm. CD score was 0 in 49% of patients and 1 or higher in 51%. The overall rate of 30-DM was 2.2%. On univariate analysis, younger age, treatment at a research/academic institution, female sex, Tsize 3cm, and Charlson-Deyo (CD) score of 0 were protective against 30-day mortality, while treatment at a community cancer center predicted for worse 30-DM. Extent of surgery was not significant. On multivariate analysis, younger age, CD score of 0, female sex, Tsize 3 cm, and treatment at a comprehensive community cancer center or an academic/research center were protective against 30-DM (P value <.001 for all). While the overall 30-DM rate was low, an interesting picture emerged when age and comorbidity were examined together (Table). In patients younger than 75 years with medical comorbidities, the 30-DM was 1.8%. However, in elderly patients with comorbidities this rate nearly tripled to 4.6% (P<.01). Much of this difference was driven by elderly patients with comorbidities who underwent lobectomy, in whom the 30DM climbed to 5.1% as opposed to 3.1% with sublobar resection (P<.01). Conclusions: The overall 30-DM rate following sublobar or lobar resection was quite low. In patients older than 75 years with medical comorbidities, however, this incidence nearly triples; suggesting improved perioperative care, consideration of more limited resection, or consideration of effective nonsurgical therapies may be indicated. Author Disclosure: Z.A. Husain: G. Consultant; RadOncQuestions.com. A.W. Kim: None. J.B. Yu: None. R.H. Decker: None. C. Corso: None.
Thoracic Abstract 121; Table score, and type of surgery
CD 0 CD 1+
30-DM as a function of age, CD comorbidity
Age
Surgery (all)
Lobectomy
Sublobar
<75 75 <75 75
1.3% 3.3% 1.8%V 4.6%V
1.2% 3.4% 1.8%U 5.1%U,*
1.5% 3.1% 1.8% 3.1%*
Note: V, U, * all represent significant comparisons with P<.05
S15
122 Predictive Significance of Actinin-4 (ACTN4) Gene Expression in Early-Stage Non-Small Cell Lung Cancer (NSCLC) Early-Stage Non-Small Cell Lung Cancer T. Yamada and K. Honda; National Cancer Center Research Institute, Tokyo, Japan Purpose/Objective(s): Even if detected at an early stage, a significant proportion of NSCLCs relapse after curative surgery. Postoperative adjuvant chemotherapy is expected to improve overall outcome. However, as the majority of early-stage NSCLCs do not relapse, only high-risk patients accounting for a small proportion of the total actually benefit from such treatment. In fact, a retrospective subset analysis of stage IB patients in the CALGB 9633 trial revealed that only individuals with a tumor diameter of http://www.practicalradonc.org/content/podcast 4 cm obtained a survival benefit from adjuvant chemotherapy. Amplification of the ACTN4 cell motility gene was recently shown to be associated with an extremely high risk of postsurgical death [hazard ratio (HR) Z 10.53] in patients with stage I lung adenocarcinoma. Therefore we investigated whether expression of ACTN4 was able to predict whether adjuvant vinorelbine and cisplatin would be of benefit to patients with stage-IB and -II NSCLC who were enrolled in the NCIC CTG JBR.10 randomized trial. Materials/Methods: Sixty-two patients in the control arm (observation alone) were divided into those showing high and low expression of the ACTN4 gene using the X-tile algorithm. The same cut-off value was applied to 71 patients who received adjuvant chemotherapy. Results: One hundred eight patients whose tumors showed a low level of ACTN4 expression did not obtain any significant survival benefit from the adjuvant therapy [HR Z 1.01 (95% CI: 0.57-1.77), PZ.979], whereas a clear benefit was evident in 25 patients whose tumors showed a high level of ACTN4 expression [HR Z 0.27 (95% CI: 0.08-0.95), PZ.030]. Cox regression analysis confirmed that ACTN4 gene expression was predictive of the response to adjuvant therapy (Interaction PZ.048). Conclusions: Various multi-gene expression signatures have been generated from microarray data, but their prognostic significance has not always been reproducible among different patient cohorts. Amplification of a single gene, ACTN4, clearly defines a small subset of high-risk patients with stage I lung adenocarcinoma. The prognostic significance of ACTN4 has been reproducibly observed in 3 independent cohorts totaling 1033 patients and exceeds that of conventional TNM staging. The roles of ACTN4 in cancer invasion and metastasis have been well demonstrated in various animals. We conclude that ACTN4 expression has the potential to discriminate patients with early-stage NSCLC who would benefit from adjuvant chemotherapy. Author Disclosure: T. Yamada: None. K. Honda: None.
123 Phase 2 Trial of Preoperative Pemetrexed (P)/Carboplatin (C) in Patients (Pts) With Select Stage IB, II, and III Nonsquamous NonSmall Cell Lung Cancer (NS-NSCLC) Early-Stage Non-Small Cell Lung Cancer D.R. Spigel,1,2 D.M. Waterhouse,1,3 K.C. Shih,1,2 R.V. Boccia,4 V.M. Priego,4 M.J. McCleod,1,5 D.C. Townsend,6 F.K. Kudrik,1,7 R.B. Mitchell,8 H.A. Burris,1,2 J.D. Hainsworth,1,2 and F.A. Greco1,2; 1 Sarah Cannon Research Institute, Nashville, TN, 2Tennessee Oncology, Nashville, TN, 3Ohio Hematology Care, Cincinnati, OH, 4Center for Cancer and Blood Disorders, Bethesda, MD, 5Florida Cancer Specialists, Ft. Myers, FL, 6Medical Oncology Associates of Augusta, Augusta, GA, 7 South Carolina Oncology Associates, Columbia, SC, 8Virginia Cancer Institute, Richmond, VA Purpose/Objective(s): Adjuvant chemotherapy is standard treatment (tx) for operable NSCLC. Potential advantages of preoperative tx include earlier tx of micrometastases and better tx tolerance. Adjuvant P/cisplatin was less toxic and easier to deliver than vinorelbine/cisplatin. This is a phase 2 trial of preoperative P/C in pts with nonsquamous stage I-III NSNSCLC.
S16
International Journal of Radiation Oncology Biology Physics
Materials/Methods: Eligibility included: pts with T2 N0, T1-2 N1, T3 N01, T1-2 N2 and T4 N0-1 (excluding superior sulcus) tumors (AJCC 6th ed) and were candidates for tx followed by resection. Clinical N2 disease was confirmed by mediastinoscopy. Pts received P 500 mg/m2 and C AUC 6 D1 q21 D xr 4 cycles followed by restaging and resection. Pts were assessed every 6 mo for the first 2 yrs following surgery, and every 12 mos thereafter. All pts were evaluable after 2 cycles. The primary objective was to assess 3 yr-survival. Results: Forty-six pts were enrolled; median age 65 yrs (range 42-82), female 61%, smoker 93%, IB/II/III 11%/37%/52%, respectively. Median tx cycles: 4 (range 2-4); 26 pts (57%) completed planned chemotherapy and surgery. Thirteen pts (28%) completed 4 cycles of tx/did not have surgery; 7 pts were unresectable based on size or location of target lesions, 4 pts had progressive disease (PD), 1 pt was too frail for surgery, and 1 pt declined surgery for financial reasons. Seven pts (15%) did not complete 4 cycles of tx; 1 pt had surgery after 2 cycles of chemotherapy, 1 pt had toxicity (decreased kidney function), 1 pt was non-compliant, 1 pt requested to discontinue tx, 1 pt discontinued P during cycle 4 and was deemed a poor surgical candidate, and 2 pts had PD. Best ORR: PR 41% (19 pts), SD 52% (24 pts), and PD 7% (3 pts). 27 pts (59%) were resected and all had pathologic PRs; 2 pts (7%) microscopic (<1 cm)/25 pts (93%) macroscopic ( 1 cm). With a median FU of 16 mos (range 2-41), median time to recurrence, PFS, and OS were: 16.1, 11.6, and 28.6 mos, respectively for stage III pts (stage I and II not reached). Toxicities (all grades [G]) included: fatigue (76%), anemia (76%), thrombocytopenia (67%), neutropenia (61%), and nausea (61%). The most common perioperative morbidities were: anemia (63%), fatigue (59%), and dyspnea (26%). G3/4 toxicity 5% were thrombocytopenia (41%), neutropenia (35%), anemia (15%), fatigue (7%), and dyspnea (7%). Postoperative complications were 2 arrhythmias and 1 air leak. There were no tx-related deaths. Conclusions: This study demonstrates preoperative P/C in nonsquamous stage IB, II, and III NSCLC is well tolerated. However, there was a high rate of pts that did not undergo resection, possibly due to pt selection. A randomized trial of preoperative vs adjuvant P/C would be required to better define the role of this regimen in stage I-III NS-NSCLC. Author Disclosure: D.R. Spigel: None. D.M. Waterhouse: None. K.C. Shih: None. R.V. Boccia: None. V.M. Priego: None. M.J. McCleod: None. D.C. Townsend: None. F.K. Kudrik: None. R.B. Mitchell: None. H.A. Burris: None. J.D. Hainsworth: None. F.A. Greco: None.
poorer in PS, lower in FEV1, higher in CCI, and larger in tumor diameter than those with SLR. Two patients were upstaged to pathological IIIA after SLR because of mediastinal node metastasis. No patients received adjuvant chemotherapy until disease progression. No treatment-related death was observed either in SBRT or in SLR. Before the PSM, 5-year overall survival (OS) was 40.3% and 60.5% for SBRT and SLR, respectively (P<.01). The PSM picked up 53 patients for each treatment group. After the PSM, difference in OS became insignificant (40.4% and 55.6% at 5 years for SBRT and SLR, respectively; PZ.12). Furthermore, cumulative incidence of cause-specific death was similar in the 2 groups (35.5% and 30.3% at 5 years, PZ.43). Although local recurrence tended to be higher in SBRT (28.4% and 14.1%, PZ.06), regional or distant metastasis was not significantly different between the 2 treatment (14.3% and 9.2% for regional, PZ.24; 35.9% and 36.1% for distant, PZ.67). Conclusions: SBRT can be an alternative treatment to SLR for those who cannot tolerate lobectomy because of medical comorbidity. Author Disclosure: Y. Matsuo: None. F. Chen: None. M. Hamaji: None. A. Kawaguchi: None. N. Ueki: None. Y. Nagata: None. M. Sonobe: None. S. Morita: None. H. Date: None. M. Hiraoka: None.
124 Comparison of Long-term Survival Outcomes Between Stereotactic Body Radiation Therapy and Sublobar Resection for Stage I NonSmall Cell Lung Cancer in High-Risk Operable Patients: Propensity Score-Match Analysis Early-Stage Non-Small Cell Lung Cancer Y. Matsuo,1 F. Chen,1 M. Hamaji,1 A. Kawaguchi,1 N. Ueki,1 Y. Nagata,2 M. Sonobe,1 S. Morita,1 H. Date,1 and M. Hiraoka1; 1Kyoto University, Kyoto, Japan, 2Hiroshima University, Hiroshima, Japan Purpose/Objective(s): Stereotactic body radiation therapy (SBRT) is being a standard treatment option for medically inoperable patients with stage I non-small cell lung cancer (NSCLC). However, its role is still unclear in high-risk operable patients who cannot tolerate lobectomy. The purpose of this study is to perform a survival comparison between SBRT and sublobar resection (SLR). Materials/Methods: All patients who underwent SBRT or sublobar resection (SLR) because of medical comorbidity for clinical stage I NSCLC between January 2003 and December 2009 were retrospectively reviewed. Patients with tumor diameter >50 mm or those without histological confirmation were excluded. Propensity score matching (PSM) was performed based on age, sex, performance status (PS), tumor diameter, forced expiratory volume in 1 second (FEV1), and Charlson comorbidity index (CCI). Results: One hundred fifteen patients of SBRT and 65 of SLR were eligible for this study. Median potential follow-up periods for SBRT and SLR were 6.7 and 5.3 years, respectively. Patients with SBRT were elderly,
125 Intrafraction Cone Beam Computed Tomography Imaging Evaluation During Stereotactic Body Radiation Therapy for Lung Tumors and Metastatic Tumors to the Spine Early-Stage Non-Small Cell Lung Cancer Z.S. Fawaz, D. Mathieu, E. Filion, L. Lambert, and L. Masucci; CHUM (University of Montreal Health Centre), Montreal, QC, Canada Purpose/Objective(s): This study aims to assess the displacement and the time relation from pretreatment to mid-treatment CBCT acquisition during SBRT for lung tumors and metastatic tumors to the spine. Materials/Methods: Two hundred fifty-eight SABR fractions were analyzed in total, including 774 translational vectors to evaluate intrafraction displacement: 193 fractions from 50 patients with early-stage non-small cell lung tumors and 65 fractions from 15 patients with vertebral metastatic tumors included in this retrospective study. All patients underwent SBRT at our center; treatment was delivered with volumetric modulated arc therapy with a flattened 6-MV photon beam between April 2012 and June 2013. Precise reproducible patient positioning was routinely obtained with the stereotactic double-vacuum whole body immobilization system. The vertical, longitudinal, and lateral vectors were obtained by using local rigid registration of the vertebra located at the level of the region of interest on pre/mid-treatment CBCT scans. Clinical data was obtained to assess the presence of a correlation with the displacement: age, gender, Karnofsky performance status, and pulmonary function test. Results: For lung tumors, 579 translational vectors were obtained from the 193 fractions: mean vertical, longitudinal and lateral motions were -0.2 mm (SDZ0.9 mm), -0.5 mm (SDZ1.0 mm) and -0.1 mm (SDZ0.9 mm). Maximum absolute vertical, longitudinal, and lateral motions were 3.0 mm, 4.0 mm and 4.0 mm. The mean translational motion vector was 1.4 mm (SDZ0.9 mm). For spine tumors, 195 translational vectors were obtained from the 65 fractions: mean vertical, longitudinal, and lateral motions were 0.0 mm (SDZ0.8 mm), -0.1 mm (SDZ0.6 mm) and 0.1 mm (SDZ0.7 mm). Maximum absolute vertical, longitudinal, and lateral motions were 2.0 mm, 2.0 mm and 2.0 mm. The mean translational motion vector was 0.9 mm (SDZ0.7 mm). Mean time from pretreatment to midtreatment CBCT acquisition (D CBCT) for the 258 SABR fractions was 23 min (SDZ6 min). The translational motion vector did not correlate with the time from pretreatment to mid-treatment CBCT acquisition in lung and spine tumors, and any of the clinical patient characteristics analyzed. Conclusions: The displacement from pretreatment to mid-treatment CBCT during SBRT for lung and spine tumors is minimal, and does not correlate with the imaging time acquisition. Author Disclosure: Z.S. Fawaz: None. D. Mathieu: None. E. Filion: None. L. Lambert: None. L. Masucci: None.
Poster Presentations
the risk for developing a SPLC remains elevated beyond 2 years, particularly in former and current smokers. Therefore, continued surveillance beyond 2 years should be maintained. Author Disclosure: A. Rimner: E. Research Grant; Varian Medical Systems. G. Consultant; Varian Medical Systems, General Electric. V. Adeseye: None. A. Foster: None. K.M. Woo: None. F. Shaikh: None. S. Din: None. Z. Zhang: None. E.D. Yorke: None. R. Gewanter: None. K.E. Rosenzweig: None. J. Huang: None. A.J. Wu: G. Consultant; Pfizer.
121 Predictors of 30-Day Mortality Following Resection of Early-Stage NSCLC: An Analysis of the National Cancer Data Base Early-Stage Non-Small Cell Lung Cancer Z.A. Husain, A.W. Kim, J.B. Yu, R.H. Decker, and C. Corso; Yale University School of Medicine, New Haven, CT Purpose/Objective(s): Studies examining morbidity after lobectomy for early stage NSCLC have shown that patients over the age of 65 have a > 50% incidence of complications, and this risk increases significantly in elderly patients (age 75). Factors that affect postsurgical 30-day mortality (30-DM), however, are less well defined. Materials/Methods: The National Cancer Data Base (NCDB), was used to identify patients 19 years and older with a diagnosis of stage I NSCLC (cT1-2, cN0, cM0) between the years 2003 and 2011. Data from patients who underwent surgical resection with lobectomy or sublobar resection were abstracted. Univariate and multivariate analysis was performed for predictors of 30-DM. Factors examined included: age, sex, year of diagnosis, tumor size, and Charlson-Deyo (CD) morbidity score. Results: A total of 112,216 patients met the criteria. Of these 71,175 or 64% received surgery, and of this group 57,569 (81%) underwent lobectomy and 13,606 (19%) underwent sublobar resection. Pneumonectomy patients were excluded. The median age was 68 years (53% female). Median tumor size (Tsize) was 2.4cm. CD score was 0 in 49% of patients and 1 or higher in 51%. The overall rate of 30-DM was 2.2%. On univariate analysis, younger age, treatment at a research/academic institution, female sex, Tsize 3cm, and Charlson-Deyo (CD) score of 0 were protective against 30-day mortality, while treatment at a community cancer center predicted for worse 30-DM. Extent of surgery was not significant. On multivariate analysis, younger age, CD score of 0, female sex, Tsize 3 cm, and treatment at a comprehensive community cancer center or an academic/research center were protective against 30-DM (P value <.001 for all). While the overall 30-DM rate was low, an interesting picture emerged when age and comorbidity were examined together (Table). In patients younger than 75 years with medical comorbidities, the 30-DM was 1.8%. However, in elderly patients with comorbidities this rate nearly tripled to 4.6% (P<.01). Much of this difference was driven by elderly patients with comorbidities who underwent lobectomy, in whom the 30DM climbed to 5.1% as opposed to 3.1% with sublobar resection (P<.01). Conclusions: The overall 30-DM rate following sublobar or lobar resection was quite low. In patients older than 75 years with medical comorbidities, however, this incidence nearly triples; suggesting improved perioperative care, consideration of more limited resection, or consideration of effective nonsurgical therapies may be indicated. Author Disclosure: Z.A. Husain: G. Consultant; RadOncQuestions.com. A.W. Kim: None. J.B. Yu: None. R.H. Decker: None. C. Corso: None.
Thoracic Abstract 121; Table score, and type of surgery
CD 0 CD 1+
30-DM as a function of age, CD comorbidity
Age
Surgery (all)
Lobectomy
Sublobar
<75 75 <75 75
1.3% 3.3% 1.8%V 4.6%V
1.2% 3.4% 1.8%U 5.1%U,*
1.5% 3.1% 1.8% 3.1%*
Note: V, U, * all represent significant comparisons with P<.05
S15
122 Predictive Significance of Actinin-4 (ACTN4) Gene Expression in Early-Stage Non-Small Cell Lung Cancer (NSCLC) Early-Stage Non-Small Cell Lung Cancer T. Yamada and K. Honda; National Cancer Center Research Institute, Tokyo, Japan Purpose/Objective(s): Even if detected at an early stage, a significant proportion of NSCLCs relapse after curative surgery. Postoperative adjuvant chemotherapy is expected to improve overall outcome. However, as the majority of early-stage NSCLCs do not relapse, only high-risk patients accounting for a small proportion of the total actually benefit from such treatment. In fact, a retrospective subset analysis of stage IB patients in the CALGB 9633 trial revealed that only individuals with a tumor diameter of http://www.practicalradonc.org/content/podcast 4 cm obtained a survival benefit from adjuvant chemotherapy. Amplification of the ACTN4 cell motility gene was recently shown to be associated with an extremely high risk of postsurgical death [hazard ratio (HR) Z 10.53] in patients with stage I lung adenocarcinoma. Therefore we investigated whether expression of ACTN4 was able to predict whether adjuvant vinorelbine and cisplatin would be of benefit to patients with stage-IB and -II NSCLC who were enrolled in the NCIC CTG JBR.10 randomized trial. Materials/Methods: Sixty-two patients in the control arm (observation alone) were divided into those showing high and low expression of the ACTN4 gene using the X-tile algorithm. The same cut-off value was applied to 71 patients who received adjuvant chemotherapy. Results: One hundred eight patients whose tumors showed a low level of ACTN4 expression did not obtain any significant survival benefit from the adjuvant therapy [HR Z 1.01 (95% CI: 0.57-1.77), PZ.979], whereas a clear benefit was evident in 25 patients whose tumors showed a high level of ACTN4 expression [HR Z 0.27 (95% CI: 0.08-0.95), PZ.030]. Cox regression analysis confirmed that ACTN4 gene expression was predictive of the response to adjuvant therapy (Interaction PZ.048). Conclusions: Various multi-gene expression signatures have been generated from microarray data, but their prognostic significance has not always been reproducible among different patient cohorts. Amplification of a single gene, ACTN4, clearly defines a small subset of high-risk patients with stage I lung adenocarcinoma. The prognostic significance of ACTN4 has been reproducibly observed in 3 independent cohorts totaling 1033 patients and exceeds that of conventional TNM staging. The roles of ACTN4 in cancer invasion and metastasis have been well demonstrated in various animals. We conclude that ACTN4 expression has the potential to discriminate patients with early-stage NSCLC who would benefit from adjuvant chemotherapy. Author Disclosure: T. Yamada: None. K. Honda: None.
123 Phase 2 Trial of Preoperative Pemetrexed (P)/Carboplatin (C) in Patients (Pts) With Select Stage IB, II, and III Nonsquamous NonSmall Cell Lung Cancer (NS-NSCLC) Early-Stage Non-Small Cell Lung Cancer D.R. Spigel,1,2 D.M. Waterhouse,1,3 K.C. Shih,1,2 R.V. Boccia,4 V.M. Priego,4 M.J. McCleod,1,5 D.C. Townsend,6 F.K. Kudrik,1,7 R.B. Mitchell,8 H.A. Burris,1,2 J.D. Hainsworth,1,2 and F.A. Greco1,2; 1 Sarah Cannon Research Institute, Nashville, TN, 2Tennessee Oncology, Nashville, TN, 3Ohio Hematology Care, Cincinnati, OH, 4Center for Cancer and Blood Disorders, Bethesda, MD, 5Florida Cancer Specialists, Ft. Myers, FL, 6Medical Oncology Associates of Augusta, Augusta, GA, 7 South Carolina Oncology Associates, Columbia, SC, 8Virginia Cancer Institute, Richmond, VA Purpose/Objective(s): Adjuvant chemotherapy is standard treatment (tx) for operable NSCLC. Potential advantages of preoperative tx include earlier tx of micrometastases and better tx tolerance. Adjuvant P/cisplatin was less toxic and easier to deliver than vinorelbine/cisplatin. This is a phase 2 trial of preoperative P/C in pts with nonsquamous stage I-III NSNSCLC.
S16
International Journal of Radiation Oncology Biology Physics
Materials/Methods: Eligibility included: pts with T2 N0, T1-2 N1, T3 N01, T1-2 N2 and T4 N0-1 (excluding superior sulcus) tumors (AJCC 6th ed) and were candidates for tx followed by resection. Clinical N2 disease was confirmed by mediastinoscopy. Pts received P 500 mg/m2 and C AUC 6 D1 q21 D xr 4 cycles followed by restaging and resection. Pts were assessed every 6 mo for the first 2 yrs following surgery, and every 12 mos thereafter. All pts were evaluable after 2 cycles. The primary objective was to assess 3 yr-survival. Results: Forty-six pts were enrolled; median age 65 yrs (range 42-82), female 61%, smoker 93%, IB/II/III 11%/37%/52%, respectively. Median tx cycles: 4 (range 2-4); 26 pts (57%) completed planned chemotherapy and surgery. Thirteen pts (28%) completed 4 cycles of tx/did not have surgery; 7 pts were unresectable based on size or location of target lesions, 4 pts had progressive disease (PD), 1 pt was too frail for surgery, and 1 pt declined surgery for financial reasons. Seven pts (15%) did not complete 4 cycles of tx; 1 pt had surgery after 2 cycles of chemotherapy, 1 pt had toxicity (decreased kidney function), 1 pt was non-compliant, 1 pt requested to discontinue tx, 1 pt discontinued P during cycle 4 and was deemed a poor surgical candidate, and 2 pts had PD. Best ORR: PR 41% (19 pts), SD 52% (24 pts), and PD 7% (3 pts). 27 pts (59%) were resected and all had pathologic PRs; 2 pts (7%) microscopic (<1 cm)/25 pts (93%) macroscopic ( 1 cm). With a median FU of 16 mos (range 2-41), median time to recurrence, PFS, and OS were: 16.1, 11.6, and 28.6 mos, respectively for stage III pts (stage I and II not reached). Toxicities (all grades [G]) included: fatigue (76%), anemia (76%), thrombocytopenia (67%), neutropenia (61%), and nausea (61%). The most common perioperative morbidities were: anemia (63%), fatigue (59%), and dyspnea (26%). G3/4 toxicity 5% were thrombocytopenia (41%), neutropenia (35%), anemia (15%), fatigue (7%), and dyspnea (7%). Postoperative complications were 2 arrhythmias and 1 air leak. There were no tx-related deaths. Conclusions: This study demonstrates preoperative P/C in nonsquamous stage IB, II, and III NSCLC is well tolerated. However, there was a high rate of pts that did not undergo resection, possibly due to pt selection. A randomized trial of preoperative vs adjuvant P/C would be required to better define the role of this regimen in stage I-III NS-NSCLC. Author Disclosure: D.R. Spigel: None. D.M. Waterhouse: None. K.C. Shih: None. R.V. Boccia: None. V.M. Priego: None. M.J. McCleod: None. D.C. Townsend: None. F.K. Kudrik: None. R.B. Mitchell: None. H.A. Burris: None. J.D. Hainsworth: None. F.A. Greco: None.
poorer in PS, lower in FEV1, higher in CCI, and larger in tumor diameter than those with SLR. Two patients were upstaged to pathological IIIA after SLR because of mediastinal node metastasis. No patients received adjuvant chemotherapy until disease progression. No treatment-related death was observed either in SBRT or in SLR. Before the PSM, 5-year overall survival (OS) was 40.3% and 60.5% for SBRT and SLR, respectively (P<.01). The PSM picked up 53 patients for each treatment group. After the PSM, difference in OS became insignificant (40.4% and 55.6% at 5 years for SBRT and SLR, respectively; PZ.12). Furthermore, cumulative incidence of cause-specific death was similar in the 2 groups (35.5% and 30.3% at 5 years, PZ.43). Although local recurrence tended to be higher in SBRT (28.4% and 14.1%, PZ.06), regional or distant metastasis was not significantly different between the 2 treatment (14.3% and 9.2% for regional, PZ.24; 35.9% and 36.1% for distant, PZ.67). Conclusions: SBRT can be an alternative treatment to SLR for those who cannot tolerate lobectomy because of medical comorbidity. Author Disclosure: Y. Matsuo: None. F. Chen: None. M. Hamaji: None. A. Kawaguchi: None. N. Ueki: None. Y. Nagata: None. M. Sonobe: None. S. Morita: None. H. Date: None. M. Hiraoka: None.
124 Comparison of Long-term Survival Outcomes Between Stereotactic Body Radiation Therapy and Sublobar Resection for Stage I NonSmall Cell Lung Cancer in High-Risk Operable Patients: Propensity Score-Match Analysis Early-Stage Non-Small Cell Lung Cancer Y. Matsuo,1 F. Chen,1 M. Hamaji,1 A. Kawaguchi,1 N. Ueki,1 Y. Nagata,2 M. Sonobe,1 S. Morita,1 H. Date,1 and M. Hiraoka1; 1Kyoto University, Kyoto, Japan, 2Hiroshima University, Hiroshima, Japan Purpose/Objective(s): Stereotactic body radiation therapy (SBRT) is being a standard treatment option for medically inoperable patients with stage I non-small cell lung cancer (NSCLC). However, its role is still unclear in high-risk operable patients who cannot tolerate lobectomy. The purpose of this study is to perform a survival comparison between SBRT and sublobar resection (SLR). Materials/Methods: All patients who underwent SBRT or sublobar resection (SLR) because of medical comorbidity for clinical stage I NSCLC between January 2003 and December 2009 were retrospectively reviewed. Patients with tumor diameter >50 mm or those without histological confirmation were excluded. Propensity score matching (PSM) was performed based on age, sex, performance status (PS), tumor diameter, forced expiratory volume in 1 second (FEV1), and Charlson comorbidity index (CCI). Results: One hundred fifteen patients of SBRT and 65 of SLR were eligible for this study. Median potential follow-up periods for SBRT and SLR were 6.7 and 5.3 years, respectively. Patients with SBRT were elderly,
125 Intrafraction Cone Beam Computed Tomography Imaging Evaluation During Stereotactic Body Radiation Therapy for Lung Tumors and Metastatic Tumors to the Spine Early-Stage Non-Small Cell Lung Cancer Z.S. Fawaz, D. Mathieu, E. Filion, L. Lambert, and L. Masucci; CHUM (University of Montreal Health Centre), Montreal, QC, Canada Purpose/Objective(s): This study aims to assess the displacement and the time relation from pretreatment to mid-treatment CBCT acquisition during SBRT for lung tumors and metastatic tumors to the spine. Materials/Methods: Two hundred fifty-eight SABR fractions were analyzed in total, including 774 translational vectors to evaluate intrafraction displacement: 193 fractions from 50 patients with early-stage non-small cell lung tumors and 65 fractions from 15 patients with vertebral metastatic tumors included in this retrospective study. All patients underwent SBRT at our center; treatment was delivered with volumetric modulated arc therapy with a flattened 6-MV photon beam between April 2012 and June 2013. Precise reproducible patient positioning was routinely obtained with the stereotactic double-vacuum whole body immobilization system. The vertical, longitudinal, and lateral vectors were obtained by using local rigid registration of the vertebra located at the level of the region of interest on pre/mid-treatment CBCT scans. Clinical data was obtained to assess the presence of a correlation with the displacement: age, gender, Karnofsky performance status, and pulmonary function test. Results: For lung tumors, 579 translational vectors were obtained from the 193 fractions: mean vertical, longitudinal and lateral motions were -0.2 mm (SDZ0.9 mm), -0.5 mm (SDZ1.0 mm) and -0.1 mm (SDZ0.9 mm). Maximum absolute vertical, longitudinal, and lateral motions were 3.0 mm, 4.0 mm and 4.0 mm. The mean translational motion vector was 1.4 mm (SDZ0.9 mm). For spine tumors, 195 translational vectors were obtained from the 65 fractions: mean vertical, longitudinal, and lateral motions were 0.0 mm (SDZ0.8 mm), -0.1 mm (SDZ0.6 mm) and 0.1 mm (SDZ0.7 mm). Maximum absolute vertical, longitudinal, and lateral motions were 2.0 mm, 2.0 mm and 2.0 mm. The mean translational motion vector was 0.9 mm (SDZ0.7 mm). Mean time from pretreatment to midtreatment CBCT acquisition (D CBCT) for the 258 SABR fractions was 23 min (SDZ6 min). The translational motion vector did not correlate with the time from pretreatment to mid-treatment CBCT acquisition in lung and spine tumors, and any of the clinical patient characteristics analyzed. Conclusions: The displacement from pretreatment to mid-treatment CBCT during SBRT for lung and spine tumors is minimal, and does not correlate with the imaging time acquisition. Author Disclosure: Z.S. Fawaz: None. D. Mathieu: None. E. Filion: None. L. Lambert: None. L. Masucci: None.