78P - Comparison of chemoradiotherapy treatment strategies in stage III non-small cell lung cancer among elderly patients from multiple data sources

Date 07 May 2017
Event ELCC 2017
Session Poster Display Session
Topics Thoracic malignancies
Surgical oncology
Non-small-cell lung cancer
Radiation oncology
Presenter Herbert Pang
Citation Annals of Oncology (2017) 28 (suppl_2): ii24-ii27. 10.1093/annonc/mdx092
Authors H. Pang1, T. Stinchcombe2, P. Cheng1, A.W. Lee3, V.H. Lee3, E.E. Vokes4, X. Wang5
  • 1School Of Public Health, Li Ka Shing Faculty Of Medicine, The University of Hong Kong, NA - Hong Kong/CN
  • 2Duke University Medical Center, Durham/US
  • 3Department Of Clinical Oncology, The University of Hong Kong, NA - Hong Kong/CN
  • 4Department Of Medicine, The University of Chicago Medical Centre, Chicago/US
  • 5Department Of Biostatistics And Bioinformatics, Duke University Medical Center, Durham/US



Comparative effectiveness research can benefit from combining data from multiple sources. This analysis integrally evaluated the survival benefits of combined modality therapies (CMTs) in locally advanced non-small cell lung cancer (NSCLC) based on individual patient data from US and Chinese populations.


Two patient populations were included. SEER-Medicare cohort consists of 65 years or higher who were diagnosed with stage III (IIIA/IIIB) NSCLC from 2006 to 2010 and received combined modality treatment. Queen Mary Hospital cohort from Hong Kong for aged 65 or higher who were diagnosed with stage III (IIIA/IIIB) NSCLC from 2007 to 2016. The four CMTs of interest were concurrent (CMT-ONLY), sequential chemotherapy followed radiation (CMT-SEQ), induction followed by concurrent (CMT-IND) and concurrent followed by consolidation (CMT-CON) chemoradiation. The primary endpoint for was overall survival. Patients hospitalized for neutropenia were ascertained through inpatient claim with the claim occurring within 130 days after the first chemotherapy. Stepwise multivariable regression models and propensity score adjusted models were used to control confounding variables. We also compared the findings with our previous study based on clinical trials data.


2682 locally advanced NSCLC patients were included. For CMT-ONLY, CMT-SEQ, CMT-IND, and CMT-CON respectively, their corresponding median overall survivals were 13.0 (95% CI 12.0-14.0), 13.0 (95% CI 12.0-16.0), 17.0 (95% CI 15.0-20.0), and 15.4 (95% CI 14.0-17.0) months. CMT-IND and CMT-CON had survival benefit over CMT-ONLY and CMT-SEQ. Patients receiving CMT-SEQ were relatively healthier patients in this study based on the Charlson comorbidity weights. 7.9%, 2.8%, 8.6%, and 9.5% were hospitalized for neutropenia for CMT-ONLY, CMT-SEQ, CMT-IND, and CMT-CON, respectively. CMT-SEQ had a lower hospitalization for neutropenia rate than the other CMTs.


The findings on efficacy and toxicity are quite consistent with previously reported studies based on clinical trials or observational databases.

Clinical trial identification

Legal entity responsible for the study

Duke University




All authors have declared no conflicts of interest.