In Reply: As we noted in our article, the negative trend was not accompanied by an overall deficit of breast cancer cases. The 143 breast cancers in our study occurring from 60 months before to 27 months after AIDS onset in the cohort of 302 834 men and women with HIV infection and AIDS corresponded closely to the expected number (n = 135.3) based on incidence rates for the general population (RR, 1.1; 95% confidence interval [CI], 0.9-1.2).
It is difficult to judge whether 2 cases of breast cancer in a group of 2460 HIV-positive individuals treated at the Beth Israel Deaconess Medical Center is more or less than one should expect. This expected number depends strongly on the sex, age, and race composition of this group and on the observation time these individuals were at risk of breast cancer. However, assuming that their patients were similar to our cohort in terms of sex (16.2% female), age, race, and duration of observation, the expected number of breast cancer cases would only be 1.1 ([2460 × 135.3] ÷ 302 834). Even if one third of the cohort that was followed up by Drs Pantanowitz and Dezube was female, the expected number of breast cancer cases (again assuming an age and race composition and a mean follow-up similar to that of ours) would be 2.3, and the corresponding RR would not be significantly reduced (RR, 0.9; 95% CI, 0.1-3.1).