Chest
Volume 124, Issue 5, November 2003, Pages 1828-1833
Journal home page for Chest

Clinical Investigations
LUNG CANCER
Tumor Size Predicts Survival Within Stage IA Non-Small Cell Lung Cancer*

https://doi.org/10.1378/chest.124.5.1828Get rights and content

Study objectives

The basic premise of CT screening is that size is an important determinant of survival in lung cancer. We sought to examine this hypothesis within stage IA non-small cell lung cancer (NSCLC).

Methods

A retrospective analysis of all patients with pathologically confirmed stage IA NSCLC resected from 1991 to 2001 was conducted. All but seven patients underwent anatomic lung resection and mediastinal lymph node dissection. Kaplan-Meier survival analysis was performed to estimate the 5-year overall and disease-specific survival probability stratified by tumor size. The influence of age, gender, histology, and tumor size on survival was also analyzed using a Cox proportional hazards regression model.

Results

There were 244 patients (mean age, 66.7 years; 45.1% were men). Lobectomy was performed in 229 patients, segmentectomy in 8 patients, and wedge resection in 7 patients. Operative mortality was 0.4%. Histologic breakdown was as follows: adenocarcinoma (59.4%), squamous (18.9%), bronchoalveolar (15.2%), large cell (4.5%), and poorly differentiated (2.0%). The median follow-up time for all patients was 2.6 years. The 5-year survival probability for all patients was 71.1% (95% confidence interval [CI], 63.6 to 78.6%). For 161 patients with tumor sizes ≤ 2.0 cm, the 5-year survival probability was 77.2% (95% CI, 68.6 to 85.8%) in comparison with 60.3% (95% CI, 46.7 to 73.8%) in 83 patients with tumor size > 2.0 cm (p = 0.03 by log-rank test). The overall 5-year disease-specific survival was 74.9% (95% CI, 67.6 to 82.2%). Disease-specific survival was 81.4% (95% CI, 73.3 to 89.4%) for patients with tumors ≤ 2.0 cm and 63.4% (95% CI, 49.6 to 77.1%) for patients with tumors > 2.0 cm.

Conclusions

These data suggest that size within stage IA is an important predictor of survival and that further substaging should be considered.

Section snippets

Patients and Methods

We conducted a retrospective chart review of all patients with NSCLC surgically treated at our institution between January 1990 and December 2001. All patients with completely resected pathologic stage IA NSCLC were identified. Hospital and office records were examined for demographic and pathologic data including age, gender, histology, and tumor size. Tumor size was categorized around the value of 2 cm.

Univariate associations between tumor size category and other prognostic factors were

Results

Patient characteristics are presented in Table 1. Two hundred forty-four patients were studied. Twenty-two patients had their tumor detected as part of an ongoing CT screening program. All patients underwent a complete mediastinal lymph node dissection at the time of thoracotomy. Median follow-up for all patients was 31 months. Tumor size distribution is illustrated in Figure 1. The overall 5-year survival was 71.1% (95% confidence interval [CI], 63.6 to 78.6%) [Fig 2]. Forty-nine patients had

Discussion

The current lung cancer staging system clearly recognizes a difference in survival between tumors < 3 cm and those > 3 cm. This distinction is well supported by several retrospective studies23456 that have shown a survival advantage for stage I patients with T1 tumors vs T2 tumors. However, there is conflicting evidence to support the impact of size as a determinant of outcome when only tumors < 3 cm in size are considered. This issue has important implications for lung cancer screening as well

Cited by (148)

View all citing articles on Scopus
View full text