Ann Thorac Surg 2007;83:358
© 2007 The Society of Thoracic Surgeons
Correspondence
Has Smoking Status Only a Prognostic Value in Patients With Stage I Pulmonary Adenocarcinomas?
Peyman Sardari Nia, MD,
Paul Van Schil, MD
Department of Thoracic and Vascular Surgery, University Hospital Antwerp, Wilrijkstraat 10, Edegem, B-2650 Belgium
(Email: peyman.sardari.nia{at}uza.be).
To the Editor:
Despite the indisputable link between smoking and increased risk for lung cancer, the inclusion of this factor in survival analyses has been relatively scarce. In recent years valuable evidence has been put forward demonstrating that smoking is one of the most significant prognostic factors in lung cancer. This is important because smoking status is one of the few prognostic factors in direct control of the patient.
We have read with great interest the study published by Yoshino and colleagues [1] evaluating the prognostic value of smoking status in 999 patients operated on for nonsmall cell lung cancer. The principal conclusion of this study was that smoking status is a significant prognostic factor for stage I adenocarcinoma.
We have some concerns regarding the results published by Yoshino and colleagues [1]. In their study, patients were divided into smokers and nonsmokers, and into four subgroups according to the number of smoked pack-years. The smokers’ subgroup included both current smokers and former smokers who had stopped smoking within 10 years before surgery. Therefore, former smokers who had stopped smoking more than 10 years ago were included in the nonsmokers group. We have recently shown that recent quitters (patients who stop smoking between the diagnosis and resection of lung cancer), former smokers (who have stopped prior to having lung cancer develop), and nonsmokers (who never smoked) have a better prognosis in multivariable analyses for disease-free and overall survival than current smokers (who were smokers at the time of resection) [2]. The mixture of smoking subgroups in the study of Yoshino and colleagues [1] probably undermines the correct interpretation of the reported results as it is not clear how many patients were actually active current smokers (worse prognostic group in our study) at the time of surgery. Moreover, current smoking at the time of operation has been shown to be consistently associated with worse prognosis in many studies [2–6].
In the study of Yoshino and colleagues [1], most of the analyses are univariate and performed in subgroups. These could easily give rise to misleading results, as they are not adjusted for different confounders in multivariable analysis [7, 8].
For example, Yoshino and colleagues [1] found a larger proportion of stage I disease among nonsmokers in adenocarcinoma, and also found that nonsmokers had a better prognosis, but only in stage I adenocarcinoma. They subsequently performed a limited multivariable analysis (without taking, for example, age into the model) in patients with stage I adenocarcinoma and confirmed their results. The finding that smoking status has only a prognostic value in adenocarcinoma in stage I could result from uneven distribution of stage of disease and other clinicopathologic factors among the subgroups. Therefore the multivariable model building must include all disease stages and clinicopathologic factors. To answer the question whether the prognostic value of smoking status is different within the different stages or different histiotypes, or both, the authors could have simply included two interaction factors, namely "stage ‘x’ smoking status" and "histiotype ‘x’ smoking status."
Due to the mixture of smoking subgroups and incomplete statistical model building in the study of Yoshino and colleagues [1], the prognostic importance of smoking status is probably underestimated in their patient population.
 |
References
|
|---|
- Yoshino I, Kawano D, Oba T, Yamazaki K, Kometani T, Maehara Y. Smoking status as a prognostic factor in patients with stage I pulmonary adenocarcinoma Ann Thorac Surg 2006;81:1189-1193.[Abstract/Free Full Text]
- Sardari Nia P, Weyler J, Colpaert C, Vermeulen P, Van Marck E, Van Schil P. Prognostic value of smoking status in operated non-small cell lung cancer Lung Cancer 2005;47:351-359.[Medline]
- Sobue T, Suzuki T, Fujimoto I, et al. Prognostic factors for surgically treated lung adenocarcinoma patients, with special reference to smoking habit Jpn J Cancer Res 1991;82:33-39.[Medline]
- Martins SJ, Pereira JR. Clinical factors and prognosis in non-small cell lung cancer Am J Clin Oncol 1999;22:453-457.[Medline]
- Tammemagi CM, Neslund-Dudas C, Simoff M, Kvale P. Smoking and lung cancer survival: the role of comorbidity and treatment Chest 2004;125:27-37.
- Myrdal G, Lambe M, Gustafsson G, et al. Survival in primary lung cancer potentially cured by operation: influence of tumor stage and clinical characteristics Ann Thorac Surg 2003;75:356-363.[Abstract/Free Full Text]
- Clark TG, Bradburn MJ, Love SB, Altman DG. Survival analysis: basic concepts and first analyses Br J Cancer 2003;89:232-238.[Medline]
- Bradburn MJ, Clark TG, Love SB, Altman DG. Survival analysis part II: multivariate data analysis: an introduction to concepts and methods Br J Cancer 2003;89:431-436.[Medline]
Top
References