Study objectives: Staging and classification in lung
cancer are important for both patient management and clinical research.
Results of survival after resection in patients with primary non-small
cell lung cancer (NSCLC) are analyzed in order to validate recent
refinements of the staging system.
Retrospective study; period from 1970 to 1992; follow-up ≥ 5
Patients: A total of 2,361 previously untreated
patients who underwent resection for stage I, II, or IIIA primary
Measurements: Survival was estimated from the
date of operation using the Kaplan-Meier survival analysis method.
Deaths within 30 days of operation were excluded. Survival comparisons
of different surgical-pathologic TNM classification (based on
pathologic examination of resected specimens) as well as further
discriminative factors were analyzed by log-rank test.
Results: Postoperative death occurred in 3.9% of patients.
For survival analyses, 2,263 patients were included. The overall 5-year
survival was 937/2,263 (41.4%). Five-year survival in stage IA was
255/404 (63%); in stage IB, 367/797 (46%); in stage IIA, 43/83
(52%); in stage IIB, 210/642 (33%); and in stage IIIA, 63/337 (19%).
No significant difference in survival was demonstrated between stages
IB and IIA. Until 4 years after surgery, age at operation did not
influence survival; after 5 years, patients > 65 years old had a
significantly lower survival.
Conclusion: The TNM
staging system accurately reflects the prognosis in primary NSCLC, but
some stage definitions can be discussed. Despite the fact that the
staging system is built on clinical data, the present analysis, which
includes postsurgical data, confirms the similar survival of patients
with T2N0M0 and T1N1M0. These results also stress the use of two
separate substages, especially because these patients are offered
surgery when possible.