Survival Prognostication in Non-small Cell Lung Cancer

Abstract

Clinical oncology is poised to benefit from continued advances in pharmacogenetics, molecular diagnostics, and targeted therapeutics. The importance of personalized medicine in oncology is reflected in the theme of the 2009 American Society of Clinical Oncology annual meeting: “Personalizing Cancer Care.” In concert with this era of discovery, important advances in non-small cell lung cancer (NSCLC) research have emerged. Molecular signatures of NSCLC have been used to prognosticate survival, 1,2 and rational approaches to targeted therapy have been integrated with chemotherapy in clinical trials with successful outcomes. 3,4 So, it may seem surprising that, despite technological advances in the field, identifying readily available prognostic factors for survival after NSCLC diagnosis continues to hold value. Clinicopathologic factors such as age at diagnosis, gender, stage at presentation, histologic cell type, degree of tumor differentiation, performance status, and smoking status may be technologically mundane, but they continue to have robust prognostic implications for survival after NSCLC diagnosis. Importantly, these factors are easily accessible from the medical record to the busy medical oncologist in clinical practice. Numerous studies have reported on the importance of readily available prognostic factors for survival in NSCLC, sometimes with conflicting results. This places the focus of interpreting prognostic NSCLC studies on data quality. In this issue of the Journal of Thoracic Oncology, Chansky et al. on behalf of the International Association for the Study of Lung Cancer (IASLC) International Staging Committee (ISC) and Participating Institutions performed a retrospective analysis of 9137 surgically managed stage I-III NSCLC cases from the original IASLC database. 5 As first published in the JTO in 2007, the IASLC database represents a landmark achievement in establishing accurate reassessments of tumor, node, metastasis (TNM) descriptors and stage groupings for NSCLC and small cell lung cancer. 6‐11 The purpose of these studies was to inform the recently released 2009 revisions to the Union Internationale Contre le Cancer -7 staging system for lung cancer, which were clearly in need of revision (as best demonstrated by results of the IASLC ISC analyses). This project was the largest of its kind to date, involving analysis of 67,725 NSCLC cases collected worldwide during 1990‐2000, along with an external validation study using data from the United States Surveillance, Epidemiology, and End Results SEER database. The original findings and recommendations from the IASLC ISC have refined the new Union Internationale Contre le Cancer staging system to more accurately define risk groups by stage. In doing so, results from the IASLC ISC will now inform therapeutic management decisions with practice-changing consequences. Findings from the IASLC database have been validated by IASLC as noted, externally validated by our group, 12‐14 and additional studies will likely emerge to test the relevance of these TNM descriptors and stage groupings in various NSCLC tumor databases. Already, IASLC has embarked on an ongoing project to create a prospectively derived NSCLC database that will provide future refinements to the lung cancer staging system and validations of many more clinical, laboratory, and molecular prognostic markers. Previously, the Prognostic Factors Subcommittee of the ISC has reported on the relevant prognostic factors for survival among 12,428 NSCLC cases. 15 The accompanying article by Chansky et al. 5 reveals a detailed analysis of a less heterogenous group of NSCLC cases: stage