A Population-Based Analysis of Mediastinal Staging for Non-Small Cell Lung Cancer

描述（由申请人提供）：非小细胞肺癌（NSCLC）是美国第二常见的癌症，也是癌症相关死亡的主要原因。一般来说，那些没有肿瘤扩散到纵隔淋巴结的患者进行根治性肺切除术，而那些有纵隔淋巴结疾病的患者则进行姑息性放化疗。非侵入性和/或侵入性诊断方式用于对纵隔进行分期。由于任何单一诊断模式的诊断准确性有限，最佳的纵隔分期可能需要使用多种诊断模式。然而，一种基于证据的分期方法尚未建立，实践指南也没有提供一致的建议。整个社区如何利用分期方式的组合是未知的。由于新的分期技术、科学的不确定性和护理的差异，分期方法的变化可能分别随着时间、地理位置和社会人口因素而发生。在使用分期方式组合的变化从未被描述过。分期可能通过适当和不适当的治疗分配影响结果。纵隔淋巴结分期过低的患者不必要地暴露于肺切除术的风险和后果，而那些纵隔淋巴结分期过高的患者则被剥夺了潜在的治愈性治疗。分期和生存之间的联系从未被报道过。该项目的总体目标是利用与医疗保险索赔相关的大型癌症登记处，对1998年至2002年间诊断为非小细胞肺癌的患者进行纵隔分期特征分析。我们的目标是1)描述与使用不同的纵隔分期方式组合相关的时间趋势、区域差异和患者因素；2)探讨不同的纵隔分期方式组合与生存率的关系。泊松回归将用于评估正电子发射断层扫描（PET）随时间的使用频率。逻辑回归将用于评估地理区域和/或患者特征是否预测计算机断层扫描、PET和纵隔镜联合使用。最后，Cox多元回归将用于对广泛纵隔分期与生存之间的关系进行探索性分析。这项工作的发现将有助于证明前瞻性临床试验更适合为不同分期策略提供无偏倚的风险调整生存率估计。从这项工作中出现的大量证据也对质量改进倡议和将标准化分期纳入治疗的临床试验具有启示意义。

DESCRIPTION (provided by applicant): Non-small cell lung cancer (NSCLC) is the second most common cancer and the leading cause of cancer- related deaths in this country. In general, those without spread of cancer to mediastinal lymph nodes undergo curative lung resection, and those with mediastinal lymph node disease undergo palliative chemoradiation therapy. Non-invasive and/or invasive diagnostic modalities are used to stage the mediastinum. Because of the limited diagnostic accuracy of any single modality, optimal mediastinal staging likely requires the use of more than one diagnostic modality. However, an evidence-based approach to staging has yet to be established and practice guidelines do not provide consistent recommendations. How the community at large utilizes combinations of staging modalities is unknown. Variation in the approach to staging may occur over time, geographic location, and sociodemographic factors as a result of new staging technology, scientific uncertainty, and disparities in care, respectively. Variation in the use of combinations of staging modalities has never been described. Staging likely affects outcomes through the appropriate and inappropriate allocation of treatment. Patients with understaged mediastinal lymph nodes are unnecessarily exposed to the risks and consequences of a lung resection, whereas those with overstaged mediastinal lymph nodes are denied a potentially curative therapy. An association between staging and survival has never been reported. The overall goal of this project is to characterize mediastinal staging in a cohort of patients diagnosed with NSCLC between 1998 and 2002 using a large cancer registry linked to Medicare claims. We aim to 1) describe temporal trends, regional variation, and patient factors associated with the use of different combinations of mediastinal staging modalities; and 2) explore the association between different combinations of mediastinal staging modalities and survival. Poisson regression will be used to evaluate the frequency of use of positron emission tomography (PET) over time. Logistic regression will be used to evaluate whether geographic area and/or patient characteristics predict the use of computed tomography, PET, and mediastinoscopy in combination. Finally, Cox multivariate regression will be used to perform an exploratory analysis of an association between the use of extensive mediastinal staging and survival. Findings from this work will help justify prospective clinical trials better suited to provide unbiased estimates of risk-adjusted survival for differing staging strategies. The body of evidence emerging from this line of work also has implications for quality improvement initiatives and incorporating standardized staging into clinical trials of therapy.