Table 2.
Nitrogenous bisphosphonate use and risk of second primary contralateral breast cancer*
| Use following diagnosis with estrogen receptor–positive breast cancer | Any nitrogenous bisphosphonate |
Alendronate |
||||||
| No. of case subjects | No. of control subjects | OR (95% CI) | P | No. of case subjects | No. of control subjects | OR (95% CI) | P | |
| Ever use | ||||||||
| Never | 308 | 532 | 1.00 (referent) | 307 | 526 | 1.00 (referent) | ||
| Short-term (<6 mo) | 15 | 20 | 1.34 (0.67 to 2.70) | .41 | 12 | 16 | 1.28 (0.59 to 2.74) | .53 |
| Ever (≥6 mo) | 17 | 65 | 0.53 (0.30 to 0.93) | .03 | 13 | 50 | 0.51 (0.27 to 0.98) | .04 |
| Duration of use among ever users | ||||||||
| ≥12 mo | 11 | 55 | 0.38 (0.19 to 0.75) | .005 | 9 | 42 | 0.40 (0.19 to 0.85) | .02 |
| ≥24 mo | 6 | 31 | 0.36 (0.14 to 0.88) | .03 | 4 | 26 | 0.27 (0.09 to 0.79) | .02 |
| Per year† | 16 | 63 | 0.74 (0.58 to 0.95) | .02 | 13 | 48 | 0.73 (0.55 to 0.95) | .02 |
| Recency of use among ever users‡ | ||||||||
| Former use | 6 | 15 | 0.76 (0.29 to 2.03) | .59 | 5 | 9 | 1.04 (0.33 to 3.32) | .94 |
| Current use | ||||||||
| ≥6 mo duration | 10 | 48 | 0.41 (0.20 to 0.84) | .02 | 8 | 39 | 0.39 (0.18 to 0.88) | .02 |
| ≥12 mo duration | 7 | 43 | 0.31 (0.13 to 0.72) | .006 | 6 | 34 | 0.33 (0.13 to 0.82) | .02 |
| ≥24 mo duration | 3 | 24 | 0.21 (0.06 to 0.73) | .01 | 2 | 21 | 0.16 (0.04 to 0.71) | .02 |
| Per year† | 10 | 47 | 0.71 (0.52 to 0.96) | .03 | 8 | 38 | 0.71 (0.52 to 0.97) | .03 |
ORs and 95% CIs were estimated using conditional logistic regression to account for matching factors (age and year of diagnosis with first primary breast cancer, Surveillance, Epidemiology and End Results historic stage of the first primary breast cancer, county of residence, and race/ethnicity). All models were additionally adjusted for body mass index at first breast cancer diagnosis. The alendronate-specific analyses exclude users of other types of bisphosphonates. P values (two-sided) were obtained using the Wald test. CI = confidence interval; OR = odds ratio.
The relative change associated with a 1-year increase in duration of bisphosphonate use among ever and current users, extrapolated from the relative change associated with a 1-month increase. P value reflects the Ptrend.
Former use was defined as last use more than 6 months before the reference date, and current use was defined as use within 6 months before the reference date. For each matched case–control set, the reference date was defined at the date on which the case subject was diagnosed with a second primary contralateral breast cancer.