Improving cancer clinical trial design and clinical practice: the case for tobacco use in cancer patients

Abstract

Substantial data support the causative nature of tobacco on the development of several cancers [1]. Unfortunately, most cancer studies do not include tobacco use as a variable to consider in the interpretation of outcomes. Increasing evidence demonstrates that tobacco use during cancer treatment is associated with poor outcomes including decreased treatment response, increased treatment-related toxicity, increased noncancer comorbidity, increased risk of second primary cancer, decreased quality of life and increased mortality from cancerand noncancer-related causes [2–7]. The effects of smoking affect both tobacco-related cancers (e.g., head/neck and lung cancer), as well as traditionally non-tobacco-related cancers (e.g., prostate cancer). In fact, data suggest that in cancers with long expected survival, such as prostate cancer, smoking may be a dominant risk factor for overall mortality due to increased risk of cardiovascular disease and death from other tobacco-related disease [8]. A history of ‘ever smoking’ decreases survival in several disease sites [2], but ‘current smoking’ at diagnosis is associated with increased overall and disease-specific mortality compared with those who have never smoked, former smokers (patients who quit ≥12 months prior to diagnosis) and recent quitters (patients who quit during the 12 months prior to diagnosis) [2]. These data suggest that a smoking history is not a fixed variable that affects outcome, but rather suggest that the effects of smoking may be reversible. There are limited studies on the effects of smoking cessation after diagnosis on cancer outcomes; however, available studies indicate that smoking cessation may reverse some of the effects of smoking on cancer outcomes [9–11]. To precisely understand the effects of tobacco on cancer patient outcomes, accurate assessments of current and former tobacco use are necessary. Unfortunately, the overwhelming majority of published literature that reports on the effects of smoking on cancer relies upon nonstandardized tobacco assessments and/or reviews of clinician-based assessments obtained in the medical record [9]. Recommendations for the systematic collection of tobacco use in cancer clinical trials have been proposed and include smoking history, current smoking status and amount, nicotine dependence, readiness to stop smoking, other tobacco use and exposure to secondhand smoke [12]. Cancer patients often report high tobacco quit rates at cancer diagnosis or the beginning of cancer treatment but later show high relapse rates [13]. Due to these unique trends in tobacco use among cancer patients, tobacco status should be collected at cancer diagnosis and treatment intake, as well as throughout treatment and follow up. Failure to assess tobacco use at all patient encounters may miss clinically meaningful changes in tobacco-use status that can dramatically influence cancer treatment response and outcomes. Thus, prospective assessment of tobacco use at multiple time points throughout cancer clinical trials can substantially enhance understanding of the precise impact of tobacco use on cancer treatment outcomes.