Smoking and death from prostate cancer

A partir d'une revue systématique de la littérature publiée entre 2000 et 2017 (27 articles, 22 549 patients), cette méta-analyse évalue l'association entre une pratique tabagique, le risque de récidive, le risque de métastase et la mortalité spécifique, chez des patients atteints d'un cancer de la prostate localisé traité par prostatectomie ou par radiothérapie

Résumé en anglais

Heart disease. Stroke. Chronic obstructive pulmonary disease. Lung cancer. Kidney cancer. Bladder cancer. And now prostate cancer?

The adverse effects of smoking are well established and include all of the above diseases. However, only recently has the association between smoking and prostate cancer become more clear. Part of the challenge is that the associations between smoking and many of those diseases are very strong compared with the association of smoking with prostate cancer, which tends to be weaker. Thus, while at first glance associations with prostate cancer may be null, with larger sample sizes, it is increasingly clear that smoking may be associated with prostate cancer. For example, a recent meta-analysis showed that smokers were 24% more likely than nonsmokers to die from prostate cancer.1 This link was suggested to account for more than 10 000 deaths per year from prostate cancer in Europe and North America alone.

In trying to understand the mechanism linking smoking and death from prostate cancer, the 2 major broad classes of mechanisms are nonbiological and biological. Regarding the former, a previous study showed that smokers were less likely to follow up with recommended biopsy schedules, even when they were enrolled in a phase 3 trial that mandated study-related biopsies.2 Even though smoking was unrelated to risk of prostate cancer or aggressive prostate cancer, when accounting for the fact that fewer smokers underwent biopsy, smoking was positively correlated with high-grade prostate cancer. This finding suggests that smokers may not seek early care soon enough, leading to delayed diagnosis and worse disease at diagnosis. This scenario alone could explain a higher risk of death from prostate cancer. Alternatively, as smoking is associated with death from many competing causes, perhaps smokers do not live long enough to die from prostate cancer. This scenario would predict a lower risk of death from prostate cancer among smokers. To what degree the 24% increased risk of death reflects these 2 opposing confounders as well as any true biological link between smoking and prostate cancer is unclear.