Table 4.
Association of height genetic score and breast cancer risk in 22 588 participants in CIMBA, per 10-cm increase in genetically predicted height
| Breast cancer group | N/events | HR (95% CI) | P* | Heterogeneity (I2) |
|---|---|---|---|---|
| Height-GS† | ||||
| All participants (confounding adjustment sequentially) | ||||
| Unadjusted | 22 588/11 451 | 1.11 (1.00 to 1.23) | .05 | – |
| Adjusted for principal components | 22 588/11 451 | 1.04 (0.93 to 1.17) | .48 | – |
| Additionally adjusted for country | 22 588/11 451 | 1.03 (0.92 to 1.16) | .57 | – |
| Additionally adjusted for birth cohort | 22 588/11 451 | 1.04 (0.92 to 1.16) | .56 | – |
| Additionally adjusted for mutation status | 22 588/11 451 | 1.04 (0.93 to 1.17) | .45 | – |
| Additionally adjusted for menopausal status | 22 588/11 451 | 1.04 (0.93 to 1.17) | .47 | |
| By mutation status‡ | ||||
| BRCA1 carrier | 14 676/7360 | 1.03 (0.91 to 1.18) | .62 | – |
| BRCA2 carrier | 7912/4091 | 1.07 (0.87 to 1.32) | .50 | – |
| Pinteraction | .95 | |||
| By menopausal status§ | ||||
| Premenopausal | 22 588/7410 | 1.09 (0.96 to 1.24) | .20 | – |
| Postmenopausal | 8459/3926 | 0.95 (0.79 to 1.13) | .55 | – |
| Pinteraction | .18 | |||
| Meta-analysis method§ | ||||
| All participants | 22 588/11451 | 1.05 (0.93 to 1.19) | .42 | 17.0% |
| BRCA1 carrier | 14 676/7360 | 1.04 (0.90 to 1.20) | .57 | 11.8% |
| BRCA2 carrier | 7912/4091 | 1.09 (0.87 to 1.36) | .45 | 6.6% |
| Pinteraction | .75 | |||
| Two-stage residual inclusion method | ||||
| All participants | 7657/3653 | 1.09 (0.93 to 1.27) | .27 | – |
| BRCA1 carrier | 4502/2154 | 1.16 (0.96 to 1.40) | .13 | – |
| BRCA2 carrier | 3155/1499 | 1.05 (0.80 to 1.37) | .74 | – |
P values were calculated using weighted Cox models. All P values are two-sided. CIMBA = Consortium of Investigators of Modifiers of BRCA1/2; H-GS = height genetic score; HR = hazard ratio; CI = confidence interval.
H-GS combining 586 height-associated single-nucleotide polymorphisms (SNPs).
Adjusted for principal components, birth cohort, country of enrollment, and menopausal status.
Adjusted for principal components, mutation status, birth cohort, and country of enrollment.
Hazard ratios were calculated using inverse-variance meta-analysis and rescaled to the corresponding units by calculating the height measurements per z score among controls. Effect estimates for breast cancer for each SNP were calculated from weighted Cox model adjusting for principal components, birth cohort, country of enrollment, menopausal status, and mutation status.