Cigarette Smoking and Prostate Cancer Risk : Negative Results of the Seoul Male Cancer Cohort Study

Prostate cancer is the most frequently diagnosed cancer among men in developed countries and has the second highest incidence worldwide (Jemal et al., 2011). Migrant studies demonstrating an increase in the incidence of prostate cancer among men who move from lowto high-incidence countries (Cook et al., 1999) suggest that environmental factors may play a role in the development of prostate cancer. Smoking is a modifiable environmental factor that may increase the risk of prostate cancer because cigarettes contain several known carcinogens and have been reported to increase circulating levels of androgens in men (Pour, 1919; Dai et al., 1988). Huncharek et al. (2010) argued an association between smoking and the incidence of prostate cancer in a meta-analysis of 24 prospective cohort studies. However, one of their results showed no increase in the risk of cancer in current smokers. Previous observational studies of the link between smoking and prostate cancer in humans have yielded inconsistent results, and the incidence of prostate cancer varies widely according to race and ethnicity (Huncharek et al., 2010; Mordukhovich et al., 2011); thus, the present study used a population-based, prospective cohort design to evaluate the relationship between smoking habits and the risk of prostate cancer in Korean men.


Introduction
Prostate cancer is the most frequently diagnosed cancer among men in developed countries and has the second highest incidence worldwide (Jemal et al., 2011).Migrant studies demonstrating an increase in the incidence of prostate cancer among men who move from low-to high-incidence countries (Cook et al., 1999) suggest that environmental factors may play a role in the development of prostate cancer.
Smoking is a modifiable environmental factor that may increase the risk of prostate cancer because cigarettes contain several known carcinogens and have been reported to increase circulating levels of androgens in men (Pour, 1919;Dai et al., 1988).Huncharek et al. (2010) argued an association between smoking and the incidence of prostate cancer in a meta-analysis of 24 prospective cohort studies.However, one of their results showed no increase in the risk of cancer in current smokers.
Previous observational studies of the link between smoking and prostate cancer in humans have yielded inconsistent results, and the incidence of prostate cancer varies widely according to race and ethnicity (Huncharek et al., 2010;Mordukhovich et al., 2011); thus, the present study used a population-based, prospective cohort design to evaluate the relationship between smoking habits and the risk of prostate cancer in Korean men.

Study cohort
The source population was the participants of the Seoul Male Cancer Cohort (SMCC) study, which has been reported elsewhere (Bae et al., 2002;2007;2013).The cohort was constructed in 1992 and 1993 for investigating the association between exposure to modifiable lifestyle factors and the risk of major cancers in Korean men.The age distribution of this cohort is similar to that of the Korean population.After excluding subjects with less than 1-year follow-up, 14,450 men were included in the present study.
A 15-page confidential questionnaire including questions on smoking habits was used to collect information about smoking status (never, past, current), duration of smoking (years), and daily amount of smoking (cigarettes/day).The reproducibility and validity of the questionnaire was evaluated in a subset of participants (Kim et al., 1996).
Follow-up was conducted over the 16-year period between January 1, 1993, andDecember 31, 2008.The incidence of all cases of cancer, including prostate cancer, occurring during the 16-year follow-up period were identified using the Seoul Regional Cancer Registry (SRCR) database, a population-based cancer registry (Parkin et al., 2002); the Korea Central Cancer Registry

Statistical methods
Accumulated person-years were calculated by determining the number of days from the initiation of follow-up, January 1, 1993, until the date of cancer diagnosis, death from other causes, or the end of the follow-up period, December 31, 2008, after which the number of days was converted into years.The ageadjusted relative risk (aRR) was calculated using the Cox proportional hazards regression.Confidence intervals were obtained using the Wald method, and all reported p values are two-sided.The chi-squared test for trends was used to evaluate linear trends.Analyses were conducted using Stata software ver.12 (StataCorp, 2013) (http://www.stata.com/).

Ethics statement
The study protocol was approved by the Institutional Review Board of the Seoul National University College of Medicine (IRB No. C-1193-146-357).

Results
Table 1 shows the distribution of participants by age and smoking status at entry into the study and the accumulated person-years of follow-up.During the total 207,326 person-years of follow-up, 87 newly diagnosed cases of prostate cancer were identified in the 14,450 study participants.Because the risk of prostate cancer was strongly age-dependent, the aRR were calculated using Cox's proportional hazard model.
Table 2 shows the estimated aRRs and 95% CIs according to smoking history at entry.The risk of prostate cancer was slightly lower for past (aRR, 0.60) and current smokers (aRR, 0.70) compared to never-smokers, although the difference was not statistically significant.We found no significant effect of duration of smoking, amount of daily smoking, or total cigarette index on the aRR of prostate cancer, with the exception of 21-30 years of smoking duration.However, no significant trend with regard to the smoking duration was observed (p for

Discussion
Our results suggest that smoking may not be a relevant risk factor for prostate cancer in Korean men.This finding is supported by a previous study (Levi and La Vecchia, 2001) that reported that minor long-term changes in prostate cancer mortality rates follow a habitual change in cigarette smoking over time in various countries, whereas mortality rates from lung and other tobacco-related neoplasms change substantially.Furthermore, Leitzmann and Rohrmann (2012) argued that the decreased risk of prostate cancer among smokers may be explained by lower PSA levels in smokers than in non-smokers (Singer et al., 2008), and a lower likelihood that smokers will undergo regular prostate cancer screening examinations compared to non-smokers.
We explored the disparity between the results of our 16-year follow-up study and those of the meta-analysis conducted by Huncharek et al. (2010) by selecting six cohort studies that reported RRs with cancer incidences after at least a 16-year follow up in Table 1 of that previous study.As shown Table 3, the results of five of the six cohorts were consistent with our findings.Thus, results of long-term follow-up studies seem to be more valid given that the incidence of prostate cancer is age related.
Despite the prospective nature of our study, two limitations should be considered when interpreting the results.The first is related to the effect of changes in participant smoking habits during the follow-up period.For example, quitting smoking may have resulted in an underestimation of the RR (Bae et al., 2013).The second is that we did not assess patients according to disease severity.Although current smoking has been reported to be associated with the risk of fatal prostate cancer (Zu and Giovannucci, 2009), we were unable to assess disease severity because a Gleason score was available for only 25% of the prostate cancer patients (Mondul et al., 2011).
In conclusion, cigarette smoking was not found to be a risk factor for prostate cancer in Korean men, indicating that etiological factors other than smoking underlie the incidence of prostate cancer.In Korea, the annual change in incidence rates of prostate and breast cancer were 12.6% and 6.0% between 1999 and 2010, respectively (KCCR, 2013).This suggests that the increasing trend in the incidence of prostate and breast cancer in Korea may be the result of a Westernized diet (Coffey, 2001).Thus, a study of the relationship between dietary factors, particularly fat intake, and prostate cancer is warranted.