Tobacco Documents Online

3 exposure. Thus analyses involving the high-risk controls might underestimate any true relationship of ETS to breast cancer. Patients undergoing breast biopsy may in any case not be representative of women in general for various reasons, regardless of whether the biopsy reveals a cancer or not. Overall, the paper must be regarded as inconclusive. (I have also attached to this review a copy of an abstract by Innes and ByersZ reporting results of a study claiming that smoking in pregnancy increases the risk of early onset breast cancer. I would need to see a full paper to review the adequacy of this claim.) P N Lee 17.1.2001 Reference List 1. Wartenberg D, Calle EE, Thun MJ, Heath CWJr, Lally C, Woodruff T. Passive smoking exposure and female breast cancer mortality. JNCI 2000;92:1666-73. 2. Innes K, Byers T. Smoking during pregnancy and breast cancer risk in young women [Abstract]. Am J Epidemio12000;151:S92.

I(~ ~ 2° 1 IInOSy REVIEW 1137 CONFIDENTIAL Subject ref 8b "Breast cancer, passive and active cigarette smoking and N-acetyltransferase 2 genotype" R J Delfmo et al Pharmacogenetics (2000), 10, 461-469 Following a spate of papers suggesting a possible relationship of ETS with breast cancer (see reviews 868, 1032, 1103, 1104, 1114), I recently reviewed (in review 1133) a paper by Wartenberg et al' which reported, based on the large American Cancer Society CPS-II prospective study, finding no significant association between breast cancer and ETS. In that paper the authors pointed out a significant heterogeneity between relative risk estimates for case- control studies and cohort studies, where meta-analysis gave figures of 1.8 (95% CI 1.4-2.5) and 1.1 (0.9-1.4) respectively. I suggested that one reason for this might be the strong possibility of recall bias in case-control studies, which of course is not relevant to prospective studies. The present paper is of interest in that it describes the results of a case-control study, conducted in Orange County California, in which recall bias was not an issue as data were collected on smoking, ETS exposure at home and other risk factors from subjects before they knew their case-control status. The study involved some 391 women over 39 years of age who had a moderate to high clinical suspicion of a breast carcinoma or there existed sufficient clinical suspicion to warrant a diagnostic biopsy. Following biopsy the women were divided into two main groups: Cases - 113 women with mammary carcinoma (30 of which were carcinoma in situ) Controls - 278 women with benign breast disease

2 For some analyses two subgroups of controls were considered: 148 High risk controls(Hyperplasia with or without atypia, complex fibroadenoma) 107 Low risk controls (Normal breast or no hyperplasia) The main results of the study concerning active and passive smoking and risk of breast cancer are presented in Table 3. The authors note that there was no significant increase in breast cancer risk in active former or current smokers as compared with the comparison group of women with no active or passive exposure. Nor was there any significant increase in risk associated with passive smoking status among never smokers, or with duration or quantity of amount smoked. The conclusions held true for all three control groups (high, low or total) and when women with carcinoma in situ were removed from the cases. N-acetyltransferase 2 (NAT2) enzyme activity, believed to play a role in the activation of tobacco smoke carcinogens, was also studied but no relationship with breast cancer risk was seen. Nor was any interaction seen between NAT2 and either active or passive smoking in their relationship with breast cancer risk. Although the results of the study are generally null, and many relative risk estimates for active smoking are below 1(see Table 3) it should be noted that the estimates for ETS in never smokers are all above 1, 1.50 (95% CI 0.79-2.87) based on all controls, 1.86 (0.81-4.27) based on low risk controls and 1.20 (0.58-2.50) based on high risk controls. Thus the results do not actually distinguish between either the meta-analysis estimate of 1.8 for case-control studies or that of 1.1 for prospective studies. It should also be noted that the validity of the study to some extent depends on the assumption that benign breast disease is unrelated to ETS or smoking. If ETS does cause malignant breast tumours, and hyperplasia and fibroadenoma are on a pathway leading to malignancy, one would not expect these pre-malignant lesions to be unassociated with ETS