Abstract
Background
Earlier onset of breast development (thelarche) is associated with increased breast cancer risk. Identifying modifiable factors associated with earlier thelarche may provide an opportunity for breast cancer risk reduction starting early in life, which could especially benefit girls with a greater absolute risk of breast cancer due to family history.
Methods
We assessed associations of maternal pre-pregnancy body mass index (BMI), physical activity during pregnancy, gestational weight gain and daughters’ weight and length at birth with age at thelarche using longitudinal Weibull models in 1031 girls in the Lessons in Epidemiology and Genetics of Adult Cancer from Youth (LEGACY) Girls Study—a prospective cohort of girls, half of whom have a breast cancer family history (BCFH).
Results
Girls whose mothers had a pre-pregnancy BMI of ≥25 and gained ≥30 lbs were 57% more likely to experience earlier thelarche than girls whose mothers had a pre-pregnancy BMI of <25 and gained <30 lbs [hazard ratio (HR) = 1.57, 95% CI: 1.16, 2.12]. This association was not mediated by childhood BMI and was similar in girls with and without a BCFH (BCFH: HR = 1.41, 95% CI: 0.87, 2.27; No BCFH: HR = 1.62, 95% CI: 1.10, 2.40). Daughters of women who reported no recreational physical activity during pregnancy were more likely to experience earlier thelarche compared with daughters of physically active women. Birthweight and birth length were not associated with thelarche.
Conclusion
Earlier thelarche, a breast cancer risk factor, was associated with three potentially modifiable maternal risk factors—pre-pregnancy BMI, gestational weight gain and physical inactivity—in a cohort of girls enriched for BCFH.
Keywords: Puberty, thelarche, breast development, gestational weight gain, pre-pregnancy BMI, physical activity, birthweight, birth length, breast cancer family history
Key Messages.
Identifying potentially modifiable factors associated with early thelarche, a breast cancer risk factor, may provide an opportunity for breast cancer risk reduction starting early in life.
Three modifiable factors—higher maternal pre-pregnancy body mass index (BMI), greater gestational weight gain and lack of maternal physical activity during pregnancy—were associated with earlier thelarche in a prospective pubertal cohort enriched for breast cancer family history.
Daughters of women who were overweight or obese prior to pregnancy and gained ≥30 lbs had the highest risk of early thelarche. This association was similar in girls with and without a family history of breast cancer.
The association between high maternal pre-pregnancy BMI (≥25) and gaining ≥30 lbs during pregnancy and earlier thelarche was not mediated by daughter’s pre-pubertal body size.
Introduction
Early age at onset of breast development (thelarche) has been associated with a >20% increase in breast cancer risk,1,2 independently of the well-established association between earlier menarche and breast cancer.3 Average age at thelarche has declined rapidly over the past 50 years,4 whereas declines in age at menarche have been more modest.5 Although the obesity epidemic contributes to this secular trend, earlier thelarche has also been observed in non-overweight girls,6 suggesting that other environmental factors likely contribute as well. Identifying modifiable factors associated with earlier thelarche has important implications for primary prevention of breast cancer.
Recent studies have observed associations between higher maternal pre-pregnancy body mass index (BMI) and/or higher gestational weight gain (GWG) and earlier age at thelarche in cohorts of girls predominantly at average risk of breast cancer.7–12 Physical activity during pregnancy, a modifiable factor associated with maternal BMI13 and GWG,14 was associated with earlier menarche in daughters independently of pre-pregnancy BMI in one study, but thelarche was not examined.15 Given that recent declines in thelarche have been more rapid than menarche,5 it is important to examine these modifiable factors with thelarche. Further, it is unknown whether these factors are associated with earlier thelarche in girls with a breast cancer family history (BCFH), who are at increased risk of developing breast cancer themselves.16
Our objective was to examine associations of maternal pre-pregnancy BMI, physical inactivity during pregnancy and GWG with age at thelarche in the Lessons in Epidemiology and Genetics of Adult Cancer from Youth (LEGACY) Girls Study—a prospective pubertal cohort enriched for BCFH.17 Since maternal pre-pregnancy BMI and GWG are associated with birthweight,18 we also assessed associations between size at birth and age at thelarche. We examined whether associations were modified by BCFH and mediated by daughters’ pre-pubertal body size.
Methods
Study population
The LEGACY Girls Study is a prospective cohort study of 1040 girls recruited at five study sites (New York, NY; Philadelphia, PA; Salt Lake City, UT; San Francisco Bay Area, CA; Toronto, ON, Canada) between 2011 and 2013 (for more details, see ref. 17) The girls were primarily aged 6–13 years at recruitment and half had a BCFH, defined as breast cancer in a first-degree or second-degree relative reported by the participating mother/guardian at baseline. Additionally, younger siblings were enrolled when they reached age 6 years (N = 28). In this analysis, we included prospective follow-up data through August 2016 for 1031 girls participating with their biological mother. Mothers provided written informed consent for themselves and their daughters, and daughters provided written informed assent according to institutional standards. The study was approved by the institutional review boards of the collaborating institutions.
Data collection
Maternal and prenatal exposures
We calculated pre-pregnancy BMI from maternal reports of height and pre-pregnancy weight at baseline. GWG was recorded as <10, 10–14, 15–19, 20–29, 30–39, 40–49 and ≥50 lbs. We also considered pre-pregnancy BMI and GWG jointly through a combined variable with cut-offs of ≥25 for BMI and ≥30 lbs for GWG. Mothers reported their recreational physical activity level during pregnancy in five categories, from inactive (no walking or regular exercise) to highly active (equivalent to walking ≥3 miles per day).15
Mothers reported their daughters’ weight and length at birth. We calculated ponderal index as weight in kilograms divided by height in metres cubed. We calculated gestational age in weeks from maternal report of the duration of pregnancy in weeks or months, or days born before/after the due date.
We asked mothers to sign a medical release form at baseline to collect growth records from their daughters’ paediatricians. In the subset of girls with birth measures available from medical records, the correlation between maternal report and medical records was high for birthweight (r = 0.9) and moderate for birth length (r = 0.6).
Outcome
Mothers assessed their daughters’ breast development every 6 months using line drawings and descriptions19 of the five Tanner Stages (TS).20 TS 2 indicates the onset of breast development.20 We previously found maternal reports of breast TS ≥2 to be reliable (kappa = 0.73) and valid (sensitivity = 77%, specificity = 94%) in a subset of girls with clinical TS data.21
Covariates
Mothers reported their daughters’ race/ethnicity at baseline, which we categorized as non-Hispanic White, non-Hispanic Black, Hispanic, Asian/Pacific Islander or other/mixed race/ethnicity. Mothers reported their level of education, which we categorized as some college or less, bachelor's degree or graduate degree. We considered BCFH (first-degree or second-degree vs none) as a modifier of the associations between early-life exposures and age at thelarche.
At biannual study visits, trained research staff measured girls’ height and weight at least twice using standardized protocols. We averaged these measures for analysis. We abstracted height and weight data prior to baseline from medical records. We calculated age-specific BMI percentiles based on the 2000 Centers for Disease Control and Prevention (CDC) growth charts.22 We used the earliest measurements between ages 5 and 7 years when available from medical records or the first study visit to assess pre-pubertal BMI in girls with at least one measure in this age range. We used age <8 years as the cut-off to define pre-puberty since <5% of girls had experienced breast TS ≥2 by 8 years of age. We classified girls with a BMI-for-age percentile of ≥85 as overweight.
Statistical analysis
We examined the distribution of early-life exposures and baseline covariates in three analytic samples: all girls participating with their biological mother (N = 1031), girls with pre-pubertal BMI measures available for mediation analyses (N = 619) and girls aged 5–7 years at baseline (N = 259). We conducted sensitivity analyses in girls aged <8 years at baseline in order to restrict the analyses to predominantly prospective thelarche data without excluding participants directly based on breast onset, which can induce bias.23
We examined associations of early-life characteristics with age at thelarche using longitudinal Weibull models with age as the timescale to accommodate left, interval and right censoring. Girls whose mother reported that they had reached TS ≥2 at the first available report were left-censored. Girls whose mothers reported breast TS ≥2 at subsequent visits were interval-censored. The beginning of the interval was the daughter’s age at the last TS1 visit and the end of the interval was the daughter’s age at the first visit where the mother reported TS ≥2. Girls who had yet to experience thelarche were right-censored at the age at last maternal report of TS1. We estimated time ratios (TRs) and hazard ratios (HRs) along with their 95% CIs for each exposure. The TR is interpreted as the ratio of the median age at thelarche for a given exposure level compared with the referent group, whereas the HR is interpreted as the ratio of the rate of transition to breast onset.24 TR <1 and HR of >1 indicate earlier thelarche.
We examined maternal pre-pregnancy BMI, birthweight and birth length continuously and in categories. We also examined combinations of birthweight/birth length in four categories (long/light, long/heavy, short/light and short/heavy) using the median values in the cohort, based on the schema by Adair.25 We adjusted for race/ethnicity and maternal education in multivariable models. We also adjusted for pre-pregnancy BMI in models examining pregnancy physical activity and GWG, and we adjusted for length of gestation in GWG models. We mutually adjusted for weight and length in birth size models and adjusted for pre-pregnancy BMI, GWG and gestational age in weeks. In analyses of girls aged <8 years, we did not adjust for race/ethnicity due to small cell counts for several groups.
We examined mediation of the association of pre-pregnancy BMI and GWG with age at thelarche by pre-pubertal BMI-for-age percentile and the product of BMI-for-age percentile and age at measurement, centred at the mean, by using inverse odds weighting to estimate direct and indirect effects.26 The total effect is the estimated association of pre-pregnancy BMI and GWG with age at thelarche through all mediating pathways. The indirect effect represents the association of pre-pregnancy BMI and GWG with age at thelarche operating through an effect on the daughter’s pre-pubertal body size, whereas the direct effect estimates the association of these exposures with age at thelarche that remains after accounting for the pathway through childhood body size. We also conducted sensitivity analyses excluding girls with a BMI-for-age percentile of ≥85. We examined multiplicative interaction by BCFH through stratification and used cross-product terms to assess statistical significance using the Wald test. We calculated the relative excess risk due to interaction (RERI) to assess additive interaction.27
In sensitivity analyses, we used clinical reports of breast TS ≥2 in 302 girls from two study sites in addition to maternal reports to examine how robust our findings were to potential error in maternal assessments. We also ran analyses restricted to singleton gestations since GWG and fetal growth may differ in multiple-gestation pregnancies.
We used cluster-robust standard errors to account for correlation within families. We conducted these analyses using SAS 9.4 and STATA 15.1.
Results
The distribution of baseline and early-life characteristics was similar across the three analytic samples (Table 1). Mothers of most girls (70%) reported a pre-pregnancy BMI of <25, >50% gained ≥30 lbs during pregnancy and 12% reported no recreational physical activity during pregnancy.
Table 1.
Descriptive characteristics of girls participating with their biological mother in the LEGACY Girls Study cohort
| All girls participating with biological mother |
Subset with BMI measured at <8 years |
Subset aged <8 years at baseline |
|
|---|---|---|---|
| (N = 1031) | (N = 619) | (N = 259) | |
| Early-life characteristics | |||
| Maternal age at birth (years) (mean ± SD) | 32.1 ± 5.4 | 32.5 ± 5.2 | 32.1 ± 5.5 |
| Missing (N, %) | 19 (1.8) | 17 (2.7) | 7 (2.7) |
| Maternal pre-pregnancy BMI, categorized (N, %) | |||
| <18.5 | 47 (4.6) | 28 (4.5) | 8 (3.1) |
| 18.5 to <25 | 676 (65.6) | 415 (67.0) | 162 (62.6) |
| 25 to <30 | 179 (17.4) | 94 (15.2) | 46 (17.8) |
| ≥30 | 96 (9.3) | 59 (9.5) | 31 (12.0) |
| Missing | 33 (3.2) | 23 (3.7) | 12 (4.6) |
| Gestational weight gain (lbs) (N, %) | |||
| <10 | 27 (2.6) | 20 (3.2) | 7 (2.7) |
| 10–14 | 42 (4.1) | 25 (4.0) | 10 (3.9) |
| 15–19 | 86 (8.3) | 54 (8.7) | 17 (6.6) |
| 20–29 | 316 (30.7) | 169 (27.3) | 78 (30.1) |
| 30–39 | 264 (25.6) | 161 (26.0) | 68 (26.3) |
| 40–49 | 145 (14.1) | 87 (14.1) | 34 (13.1) |
| ≥50 | 113 (11.0) | 69 (11.2) | 31 (12.0) |
| Missing | 38 (3.7) | 34 (5.5) | 14 (5.4) |
| Maternal recreational physical activity during pregnancy (N, %) | |||
| Inactive, no walking or other regular exercise | 128 (12.4) | 71 (11.5) | 30 (11.6) |
| Mostly inactive, equivalent to walking about half a mile or less every day | 235 (22.8) | 156 (25.2) | 71 (27.4) |
| Somewhat active, equivalent to walking ∼1 mile every day | 222 (21.5) | 136 (22.0) | 57 (22.0) |
| Active, equivalent to walking ∼2 miles every day | 379 (36.8) | 215 (34.7) | 85 (32.8) |
| Highly active, equivalent to walking ∼3 or more miles every day | 57 (5.5) | 33 (5.3) | 11 (4.3) |
| Missing | 10 (1.0) | 8 (1.3) | 5 (1.9) |
| Type of gestation (N, %) | |||
| Multiple | 45 (4.4) | 34 (5.5) | 13 (5.0) |
| Singleton | 970 (94.1) | 576 (93.1) | 241 (93.1) |
| Missing | 16 (1.6) | 9 (1.5) | 5 (1.9) |
| Gestational age (weeks) (mean ± SD) | 39.0 ± 2.1 | 38.9 ± 2.2 | 38.8 ± 2.2 |
| Missing (N, %) | 18 (1.8) | 14 (2.3) | 9 (3.5) |
| Birthweight (g) (mean ± SD) | 3298.3 ± 583.3 | 3297.8 ± 574.6 | 3287.2 ± 574.6 |
| Missing (N, %) | 13 (1.3) | 8 (1.3) | 4 (1.5) |
| Birth length (cm) (mean ± SD) | 50.5 ± 3.6 | 50.4 ± 3.7 | 50.4 ± 3.8 |
| Missing (N, %) | 133 (12.9) | 73 (11.8) | 30 (11.6) |
| Ponderal index at birth (kg/m3) (mean ± SD) | 25.8 ± 5.8 | 25.8 ± 5.2 | 25.7 ± 5.3 |
| Missing (N, %) | 133 (12.9) | 73 (11.8) | 30 (11.6) |
| Baseline characteristics | |||
| Age at baseline (years) (mean ± SD)a | 10.0 ± 2.4 | 9.2 ± 2.3 | 6.9 ± 0.6 |
| BMI-for-age percentile at baseline, categorized (N, %)a | |||
| ≥85th BMI-for-age percentile | 174 (16.9) | 100 (16.2) | 36 (13.9) |
| <85th BMI-for-age percentile | 806 (78.2) | 503 (81.3) | 212 (81.9) |
| Missing | 51 (5.0) | 16 (2.6) | 11 (4.3) |
| History of breast cancer in a first-degree or second-degree relative (N, %) | |||
| BCFH | 530 (51.4) | 310 (50.1) | 134 (51.7) |
| No BCFH | 501 (48.6) | 309 (49.9) | 125 (48.3) |
| Study site | |||
| Philadelphia | 153 (14.8) | 112 (18.1) | 24 (9.3) |
| New York | 175 (17.0) | 116 (18.7) | 56 (21.6) |
| Utah | 173 (16.8) | 103 (16.6) | 60 (23.2) |
| Ontario | 179 (17.4) | 106 (17.1) | 46 (17.8) |
| San Francisco Bay Area, CA | 351 (34.0) | 182 (29.4) | 73 (28.2) |
| Race/ethnicity | |||
| Non-Hispanic White | 650 (63.1) | 406 (65.6) | 167 (64.5) |
| Non-Hispanic Black | 78 (7.6) | 49 (7.9) | 20 (7.7) |
| Hispanic | 184 (17.9) | 96 (15.5) | 48 (18.5) |
| Asian/Pacific Islander | 88 (8.5) | 52 (8.4) | 20 (7.7) |
| Other or mixed race/ethnicity | 31 (3.0) | 16 (2.6) | 4 (1.5) |
| Maternal education | |||
| Some college, vocational or technical school or less | 287 (27.8) | 147 (23.8) | 75 (29.0) |
| Bachelor’s degree | 373 (36.2) | 226 (36.5) | 93 (35.9) |
| Graduate degree | 346 (33.6) | 232 (37.5) | 85 (32.8) |
| Missing | 25 (2.4) | 14 (2.3) | 6 (2.3) |
BCFH, breast cancer family history; BMI, body mass index; CA, California; LEGACY, Lessons in Epidemiology and Genetics of Adult Cancer from Youth.
Age or BMI-for-age percentile at pilot baseline visit for girls with pilot data (N = 21).
Maternal pre-pregnancy BMI and GWG
Higher pre-pregnancy BMI was associated with earlier thelarche in daughters (HR = 1.14, 95% CI: 1.04, 1.26 per 5 increase in BMI) (Table 2). Daughters of women who gained ≥30 lbs experienced earlier thelarche than daughters of women who gained 20–29 lbs. This association was higher in magnitude in daughters of women who gained >50 lbs (HR = 1.37, 95% CI: 1.01, 1.85), corresponding to development ∼5 months earlier.
Table 2.
Time ratios, hazard ratios and 95% confidence intervals for associations of maternal pre-pregnancy body mass index, maternal recreational physical activity during pregnancy and gestational weight gain with age at thelarche for all girls and girls aged <8 years at baseline in the LEGACY Girls Study
| All girls (N = 1031) |
Girls aged <8 years at baseline (N = 259) |
|||||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Base modela |
Multivariable-adjusted modelb |
Base modela |
Multivariable-adjusted modelc |
|||||||||||
| N | N events | TR (95% CI) | HR (95% CI) | TR (95% CI) | HR (95% CI) | P d | N | N events | TR (95% CI) | HR (95% CI) | TR (95% CI) | HR (95% CI) | P d | |
| Maternal pre-pregnancy BMI | 0.16 | 0.20 | ||||||||||||
| <25 | 714 | 593 | Reference | Reference | Reference | Reference | 165 | 83 | Reference | Reference | Reference | Reference | ||
| 25 to <30 | 178 | 147 | 0.98 (0.94, 1.01) | 1.21 (0.89, 1.64) | 0.99 (0.95, 1.03) | 1.10 (0.81, 1.50) | 46 | 26 | 0.96 (0.91, 1.02) | 1.36 (0.89, 2.07) | 0.98 (0.92, 1.04) | 1.19 (0.76, 1.86) | ||
| ≥30 | 96 | 82 | 0.95 (0.91, 0.98) | 1.55 (1.15, 2.08) | 0.96 (0.92, 1.00) | 1.36 (0.99, 1.86) | 31 | 20 | 0.92 (0.86, 0.99) | 1.80 (1.07, 3.01) | 0.93 (0.85, 1.01) | 1.77 (0.96, 3.26) | ||
| Continuous (per 5) | 988 | 822 | 0.98 (0.97, 0.99) | 1.21 (1.11, 1.32) | 0.98 (0.97, 1.00) | 1.14 (1.04, 1.26) | 0.007 | 242 | 129 | 0.97 (0.95, 0.99) | 1.24 (1.10, 1.41) | 0.98 (0.96, 1.00) | 1.19 (1.03, 1.37) | 0.02 |
| Gestational weight gain (lbs) | 0.25 | 0.08 | ||||||||||||
| <20 | 155 | 130 | 0.98 (0.94, 1.01) | 1.21 (0.92, 1.60) | 0.98 (0.95, 1.02) | 1.15 (0.86, 1.54) | 34 | 20 | 0.92 (0.85, 1.00) | 1.85 (1.02, 3.37) | 0.93 (0.86, 1.01) | 1.72 (0.96, 3.07) | ||
| 20–29 | 315 | 246 | Reference | Reference | Reference | Reference | 77 | 33 | Reference | Reference | Reference | Reference | ||
| 30–39 | 261 | 226 | 0.99 (0.96, 1.02) | 1.10 (0.88, 1.37) | 0.98 (0.95, 1.00) | 1.23 (0.97, 1.55) | 67 | 39 | 0.95 (0.89, 1.01) | 1.43 (0.91, 2.27) | 0.93 (0.87, 0.99) | 1.78 (1.11, 2.84) | ||
| 40–49 | 143 | 118 | 0.98 (0.95, 1.01) | 1.18 (0.91, 1.52) | 0.97 (0.94, 1.01) | 1.24 (0.95, 1.62) | 33 | 17 | 0.93 (0.86, 1.00) | 1.71 (0.99, 2.97) | 0.92 (0.85, 0.99) | 1.93 (1.08, 3.42) | ||
| ≥50 | 109 | 91 | 0.97 (0.93, 1.01) | 1.28 (0.95, 1.72) | 0.96 (0.92, 1.00) | 1.37 (1.01, 1.85) | 29 | 15 | 0.92 (0.83, 1.01) | 1.88 (0.94, 3.78) | 0.92 (0.84, 1.01) | 1.88 (0.94, 3.76) | ||
| Maternal pre-pregnancy BMI and GWG (lbs) | 0.03 | 0.03 | ||||||||||||
| BMI < 25 and GWG <30 | 312 | 246 | Reference | Reference | Reference | Reference | 67 | 24 | Reference | Reference | Reference | Reference | ||
| BMI < 25 and GWG ≥30 | 389 | 334 | 1.00 (0.97, 1.02) | 1.04 (0.86, 1.25) | 0.98 (0.96, 1.01) | 1.14 (0.93, 1.39) | 95 | 56 | 0.92 (0.85, 0.98) | 1.91 (1.15, 3.16) | 0.91 (0.85, 0.97) | 2.06 (1.26, 3.38) | ||
| BMI ≥ 25 and GWG <30 | 149 | 124 | 0.98 (0.94, 1.03) | 1.17 (0.82, 1.66) | 0.99 (0.94, 1.03) | 1.11 (0.77, 1.59) | 43 | 29 | 0.89 (0.82, 0.96) | 2.40 (1.39, 4.16) | 0.91 (0.84, 0.99) | 1.98 (1.10, 3.56) | ||
| BMI ≥ 25 and GWG ≥30 | 118 | 98 | 0.94 (0.90, 0.97) | 1.69 (1.27, 2.26) | 0.95 (0.91, 0.98) | 1.57 (1.16, 2.12) | 32 | 15 | 0.91 (0.83, 1.00) | 1.97 (1.01, 3.84) | 0.91 (0.83, 1.00) | 2.07 (1.02, 4.18) | ||
| Maternal recreational physical activity | 0.47 | 0.11 | ||||||||||||
| Inactive, no walking or other regular exercise | 127 | 107 | 0.96 (0.92, 0.99) | 1.42 (1.07, 1.89) | 0.98 (0.94, 1.01) | 1.20 (0.89, 1.63) | 30 | 18 | 0.91 (0.85, 0.98) | 1.95 (1.17, 3.25) | 0.93 (0.87, 1.00) | 1.70 (1.02, 2.83) | ||
| Mostly inactive, equivalent to walking about half a mile or less every day | 232 | 181 | 1.00 (0.97, 1.03) | 0.98 (0.77, 1.24) | 1.01 (0.98, 1.04) | 0.95 (0.74, 1.22) | 70 | 33 | 1.00 (0.95, 1.06) | 0.98 (0.63, 1.51) | 1.01 (0.95, 1.08) | 0.91 (0.56, 1.47) | ||
| Somewhat active, equivalent to walking ∼1 mile every day | 220 | 185 | 1.01 (0.98, 1.04) | 0.92 (0.75, 1.14) | 1.01 (0.98, 1.04) | 0.92 (0.74, 1.15) | 56 | 33 | 1.02 (0.96, 1.09) | 0.85 (0.53, 1.36) | 1.01 (0.95, 1.07) | 0.94 (0.59, 1.51) | ||
| Active or highly active, equivalent to walking ≥2 miles every day | 433 | 365 | Reference | Reference | Reference | Reference | 93 | 47 | Reference | Reference | Reference | Reference | ||
BMI, body mass index; GWG, gestational weight gain; HR, hazard ratio; LEGACY, Lessons in Epidemiology and Genetics of Adult Cancer from Youth; TR, time ratio.
Adjusted for age as the timescale.
Additionally adjusted for race/ethnicity (non-Hispanic White, non-Hispanic Black, Hispanic, Asian, Other) and maternal education (some college or less, bachelor's degree, graduate degree). Models for maternal physical activity and GWG also adjusted for maternal pre-pregnancy BMI (continuous). Models for GWG and BMI/GWG groups also adjusted for length of gestation (continuous).
Additionally adjusted for maternal education (some college or less, bachelor's degree, graduate degree). Models for maternal physical activity and GWG also adjusted for maternal pre-pregnancy BMI (continuous). Models for GWG and BMI/GWG groups also adjusted for length of gestation (continuous). Race/ethnicity was not included in multivariable model due to small cell counts for several groups.
P-value for exposure of interest from Wald test.
Daughters of women who were overweight or obese prior to pregnancy (BMI of ≥25) and gained ≥30 lbs were almost 60% more likely to experience earlier thelarche than daughters of women who were not overweight and gained <30 lbs (HR = 1.57, 95% CI: 1.16, 2.12), corresponding to a difference of ∼7 months (Table 2). Only having one of these risk factors was not associated with earlier thelarche in daughters. In analyses restricted to girls aged <8 years at baseline, daughters of women who were overweight prior to pregnancy, gained ≥30 lbs or both were two times more likely to experience earlier thelarche compared with daughters of women with neither risk factor (Table 2).
The association of pre-pregnancy BMI of ≥25 and GWG of ≥30 lbs with earlier thelarche was not mediated by daughters’ pre-pubertal body size (Figure 1). The majority of girls (83%) in the cohort were not overweight prior to puberty. When we excluded overweight girls, we observed a borderline statistically significant association with earlier thelarche in daughters of women who were overweight prior to pregnancy and gained ≥30 lbs (HR = 1.55, 95% CI: 1.00, 2.42).
Figure 1.
Mediation of the association between maternal pre-pregnancy body mass index and gestational weight gain and age at thelarche by pre-pubertal body size using inverse odds weighting. BMI, body mass index; GWG, gestational weight gain; HR, hazard ratio. N = 560 girls. Referent group is BMI of <25 and GWG of <30 lbs. Estimates adjusted for age as the timescale, race/ethnicity, maternal education and length of gestation. Mediators included in generating inverse odds weight are BMI-for-age percentile and BMI-for-age percentile times centred age at BMI measure. Indirect effect is calculated by subtracting the estimated log HR for the direct effect from the estimated log HR for the total effect. Standard errors and CIs for direct and indirect effects were generated by bootstrapping with 1000 replications. Hazard ratios are plotted on the log scale.
The positive association between pre-pregnancy BMI of ≥25, GWG of ≥30 lbs and earlier thelarche was similar in girls with and without a BCFH (BCFH: HR = 1.41, 95% CI: 0.87, 2.27; No BCFH: HR = 1.62, 95% CI: 1.10, 2.40) (Figure 2). Daughters of women who were overweight prior to pregnancy and gained <30 lbs were more likely to experience earlier thelarche in girls without a BCFH (HR = 2.03, 95% CI: 1.32, 3.13), but there was no association in girls with a BCFH (HR = 0.67, 95% CI: 0.42, 1.08). This negative interaction was also observed on the additive scale (RERI: −1.50, 95% CI: −2.53, −0.47). There was no interaction on the additive scale for the other two levels of the composite variable (pre-pregnancy BMI of <25, GWG of ≥30 lbs and BCFH, RERI: −0.30, 95% CI: −0.84, 0.24 and pre-pregnancy BMI of ≥25, GWG of ≥30 lbs and BCFH, RERI: 0.11, 95% CI: −0.98, 1.19).
Figure 2.
Hazard ratios for the association between maternal pre-pregnancy body mass index and gestational weight gain and age at thelarche stratified by breast cancer family history. BCFH, breast cancer family history; BMI, body mass index; GWG, gestational weight gain; HR, hazard ratio. Estimates adjusted for age as the timescale, race/ethnicity, maternal education and length of gestation. P for interaction = 0.01. Hazard ratios are plotted on the log scale.
Maternal physical activity during pregnancy
Daughters of women who reported no recreational physical activity during pregnancy experienced earlier thelarche than daughters of physically active women, but the association was attenuated after adjustment for pre-pregnancy BMI, race/ethnicity and maternal education (Table 2). This association was stronger in magnitude in girls aged <8 years (HR = 1.70, 95% CI: 1.02, 2.83), corresponding to thelarche ∼8 months earlier in daughters of women who reported no recreational physical activity.
Birth size
Neither birthweight nor birth length was associated with age at thelarche when modelled separately (age-adjusted HR 1.04, 95% CI: 0.98, 1.12 per 500-g increase in birthweight and HR 1.02, 95% CI: 1.00, 1.04 per 1-cm increase in birth length) or mutually adjusted (Supplementary Table S1, available as Supplementary data at IJE online). Ponderal index at birth was also not associated with age at thelarche.
Sensitivity analyses
In the subset of girls with clinical TS, the association between maternal pre-pregnancy BMI and thelarche as assessed by trained personnel was consistent with the maternal assessments (Supplementary Table S2, available as Supplementary data at IJE online). Associations were similar when we excluded multiple-gestation pregnancies (Supplementary Table S3, available as Supplementary data at IJE online).
Discussion
In the prospective LEGACY cohort enriched for BCFH, three potentially modifiable factors—higher maternal pre-pregnancy BMI, higher GWG and lack of recreational physical activity during pregnancy—were associated with earlier thelarche in daughters. The associations that we observed between pre-pregnancy BMI, GWG and age at thelarche were consistent with previous studies of girls primarily at average risk of breast cancer.7–12
To our knowledge, the association between maternal physical activity during pregnancy and age at thelarche has not been investigated previously. In the Nurses’ Health Study II cohort, there was a modest positive linear relationship between maternal leisure-time physical activity during pregnancy and daughters’ age at menarche, with a 1-month difference in age at menarche between daughters of highly active and inactive women.15 Although we observed earlier thelarche in daughters of women with no recreational physical activity during pregnancy, there was not a trend with increasing levels of physical activity. This lack of trend may mean that any activity helps delay thelarche, but it also may be explained by residual confounding. In the overall cohort, the association with physical inactivity was attenuated after adjustment for pre-pregnancy BMI, race/ethnicity and maternal education. The association was stronger in magnitude in girls aged <8 years at baseline, with minimal attenuation after adjustment for pre-pregnancy BMI and maternal education. This difference could be due to residual confounding in this sub-cohort, but could also be explained by more accurate reporting of age at thelarche in younger girls due to predominantly prospective data. Because we were focused on maternal and prenatal factors, we did not examine daughters’ physical activity in relation to age at thelarche. Daughters of physically inactive women may be more likely to be physically inactive themselves. Athletes experience later menarche than non-athletes,28 but less is known about the effect of moderate levels of physical activity in childhood on pubertal timing.
Our results do not support an independent role for size at birth in relation to thelarche. Mothers recalled birthweight and birth length at baseline and these measures could be subject to error, though the correlation between maternal report and birthweight abstracted from medical records was high in our validation subset. Birth cohorts using prospective measures of birthweight also have not observed an association with thelarche.9,29,30 Higher birthweight infants reached breast TS >2 at earlier ages in the North Carolina Infant Feeding Study,31 but the association was adjusted for weight gain in infancy and early childhood and may reflect the influence of post-natal growth or pubertal tempo, since the referent group included pre-pubertal (TS 1) and early pubertal (TS 2) girls. The correlation for recalled and recorded birth length in our validation subset was modest and 13% of mothers did not report length. Assessments of recumbent length, even when measured by nurses, have poor reliability.32,33 Three previous studies using prospective birth length measures also did not observe an association with thelarche.9,29,34
Maternal pre-pregnancy BMI and GWG are positively associated with daughters’ BMI.35 Overweight girls experience earlier thelarche than non-overweight girls.36,37 Previous studies examining pre-pregnancy BMI and GWG independently in relation to thelarche have observed partial mediation by daughters’ BMI.7,8,10,11 In the Kaiser Permanente Northern California cohort, the joint association between maternal pre-pregnancy obesity, excessive GWG and earlier thelarche was slightly attenuated after adjustment for daughters’ BMI.11 We employed inverse odds weighting to examine mediation by pre-pubertal BMI-for-age percentile, which does not rely on a linear link function or the assumption of no exposure–mediator interaction to estimate direct and indirect effects, provided that other assumptions regarding confounding are met.26 Our results suggest that the association between high maternal pre-pregnancy BMI, high GWG and earlier age at thelarche is not mediated by daughters’ BMI. The majority of girls in our cohort are not overweight, which further supports that maternal pre-pregnancy BMI and GWG are associated with thelarche independently of their associations with childhood body size. Other pathways that could underlie associations between maternal pre-pregnancy BMI, GWG and lack of pregnancy physical activity and daughter’s timing of thelarche may include an epigenetic or developmental programming mechanism.38 For example, in utero exposure to high maternal leptin levels or insulin resistance may predispose daughters to elevated leptin and insulin levels in childhood,39 which may lead to earlier pubertal onset.40,41
Whereas maternal pre-pregnancy overweight or obesity with GWG of <30 lbs was only associated with earlier thelarche in girls without a BCFH, the association between pre-pregnancy overweight or obesity, GWG of ≥30 lbs and earlier thelarche did not differ by BCFH on the multiplicative or additive scale. This suggests that the absolute risk of early thelarche can be modified by changing the early-life environment, even in girls at increased risk of breast cancer due to their family history. Maintaining a healthy weight prior to pregnancy, preventing excessive GWG and engaging in physical activity during pregnancy have many health benefits for mother and child. Raising awareness that these behaviours, which are in line with current recommendations, may delay thelarche both in daughters at population-average risk of breast cancer and those at increased familial risk is an important public health message.
In LEGACY, we assessed thelarche twice a year. Previous studies have primarily assessed thelarche annually,7,42 decreasing precision in outcome assessment. The collection of clinical TS at two study sites allowed us to compare findings using maternal and clinical reports. Although we were underpowered to assess outcomes based only on clinical TS, the direction of the associations was the same and we have previously demonstrated good agreement between clinical and mother-reported TS.21 A limitation of the study is that the categories of GWG we collected did not correspond to the ranges used to define adequate weight gain in the 2009 Institute of Medicine guidelines,43 so we could not examine the clinically relevant categories of inadequate, adequate and excessive weight gain in relation to age at thelarche. Also, since girls were primarily aged 6–13 years at baseline, some had already experienced breast development prior to cohort entry. We included these girls in the analyses with left censoring, but lack of prospective data on these girls could have biased our results towards the null. The inference was similar in analyses restricted to girls aged <8 years at baseline, among whom <5% of girls were left-censored for the outcome. Bias due to retrospective data was limited in these analyses, but the sample size of this subset affected precision and limited adjustment for confounding. Results from our cohort, in which approximately one-third of girls had a mother who completed a graduate degree, may not be generalizable to lower socio-economic status populations that may differ in the distribution of other factors that affect pubertal timing.
Conclusions
Earlier thelarche was associated with three potentially modifiable maternal risk factors—high pre-pregnancy BMI, high GWG and maternal physical inactivity during pregnancy—in a cohort of girls enriched for BCFH. Earlier age at thelarche has been associated with breast cancer risk in both average-risk women1 and women with a first-degree family history,2 suggesting that delaying thelarche may reduce breast cancer risk in adulthood across the familial risk spectrum. Our results suggesting that maintaining a healthy BMI prior to pregnancy, moderate GWG and engaging in physical activity during pregnancy delay thelarche in daughters may therefore have implications for breast cancer risk reduction as well.
Ethics approval
All participating institutions (Columbia University Medical Center, Cancer Prevention Institute of California, Lunenfeld-Tanenbaum Research Institute of Sinai Health System, Huntsman Cancer Institute at University of Utah, Fox Chase Cancer Center, Children’s Hospital of Philadelphia, and the University of Pennsylvania) obtained institutional review board approval to conduct the study. Mothers/guardians provided written informed consent and girls provided assent based on institutional standards.
Supplementary Material
Contributor Information
Mandy Goldberg, Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY, USA.
Jasmine A McDonald, Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY, USA; Herbert Irving Comprehensive Cancer Center, Columbia University Irving Medical Center, New York, NY, USA.
Lauren C Houghton, Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY, USA; Herbert Irving Comprehensive Cancer Center, Columbia University Irving Medical Center, New York, NY, USA.
Irene L Andrulis, Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, Ontario, Canada; Department of Molecular Genetics, University of Toronto, Toronto, Ontario, Canada; Department of Laboratory Medicine and Pathobiology, University of Toronto, Toronto, Ontario, Canada.
Julia A Knight, Lunenfeld-Tanenbaum Research Institute, Sinai Health System, Toronto, Ontario, Canada; Epidemiology Division, Dalla Lana School of Public Health, University of Toronto, Toronto, Ontario, Canada.
Angela R Bradbury, Department of Medicine, Division of Hematology-Oncology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA; Department of Medical Ethics and Health Policy, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
Lisa A Schwartz, Division of Oncology, The Children's Hospital of Philadelphia, Philadelphia, PA, USA; Department of Pediatrics, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA.
Saundra S Buys, Department of Medicine and Huntsman Cancer Institute, University of Utah Health, Salt Lake City, UT, USA.
Caren J Frost, College of Social Work, University of Utah, Salt Lake City, UT, USA.
Mary B Daly, Department of Clinical Genetics, Fox Chase Cancer Center, Philadelphia, PA, USA.
Esther M John, Department of Epidemiology & Population Health, Stanford University School of Medicine, Stanford, CA, USA; Department of Medicine (Oncology), Stanford University School of Medicine, Stanford, CA, USA; Stanford Cancer Institute, Stanford University School of Medicine, Stanford, CA, USA.
Theresa H M Keegan, Center for Oncology Hematology Outcomes Research and Training (COHORT), Division of Hematology and Oncology, University of California, Davis, Sacramento, CA, USA.
Wendy K Chung, Herbert Irving Comprehensive Cancer Center, Columbia University Irving Medical Center, New York, NY, USA; Department of Pediatrics, Columbia University Irving Medical Center, New York, NY, USA; Department of Medicine, Columbia University Irving Medical Center, New York, NY, USA.
Ying Wei, Department of Biostatistics, Columbia University Mailman School of Public Health, New York, NY, USA.
Mary Beth Terry, Department of Epidemiology, Columbia University Mailman School of Public Health, New York, NY, USA; Herbert Irving Comprehensive Cancer Center, Columbia University Irving Medical Center, New York, NY, USA.
Data availability
The data underlying this article will be shared on reasonable request to the corresponding author.
Supplementary data
Supplementary data are available at IJE online.
Author contributions
M.G. and M.B.T. conceptualized the analyses presented in the current study. M.G. conducted and interpreted the data analyses and drafted the initial manuscript. M.B.T. advised the data analyses, interpretation and manuscript writing. I.L.A., A.R.B., S.S.B., M.B.D., E.M.J. and M.B.T. conceptualized the overall parent study. J.A.M., L.C.H., I.L.A., J.A.K., A.R.B., L.A.S., S.S.B., C.J.F., M.B.D., E.M.J., T.H.M.K., W.K.C., Y.W. and M.B.T. participated in the collection and assembly of data, analyses and interpretation, and manuscript writing. All authors critically reviewed and approved the final manuscript.
Funding
The authors acknowledge the funding by the National Cancer Institute at the National Institutes of Health (grants R01CA138638 to E.M.J., R01CA138819 to M.B.D., R01CA138822 to M.B.T., R01CA138844 to I.L.A. and K01CA186943 to J.A.M.), the Breast Cancer Research Foundation (M.B.T.) and the Canadian Breast Cancer Foundation (I.L.A.). I.L.A. holds the Anne and Max Tanenbaum Chair in Molecular Medicine at the Sinai Health System and the University of Toronto.
Conflict of interest
None declared.
References
- 1. Bodicoat DH, Schoemaker MJ, Jones ME. et al. Timing of pubertal stages and breast cancer risk: the Breakthrough Generations Study. Breast Cancer Res 2014;16:R18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Goldberg M, D’Aloisio AA, O’Brien KM, Zhao S, Sandler DP.. Pubertal timing and breast cancer risk in the Sister Study cohort. Breast Cancer Res 2020;22:112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Collaborative Group on Hormonal Factors in Breast Cancer. Menarche, menopause, and breast cancer risk: individual participant meta-analysis, including 118 964 women with breast cancer from 117 epidemiological studies. Lancet Oncol 2012;13:1141–51. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Eckert-Lind C, Busch AS, Petersen JH. et al. Worldwide secular trends in age at pubertal onset assessed by breast development among girls: a systematic review and meta-analysis. JAMA Pediatr 2020;174:e195881. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Lee Y, Styne D.. Influences on the onset and tempo of puberty in human beings and implications for adolescent psychological development. Horm Behav 2013;64:250–61. [DOI] [PubMed] [Google Scholar]
- 6. Aksglaede L, Sørensen K, Petersen JH, Skakkebæk NE, Juul A.. Recent decline in age at breast development: the Copenhagen Puberty Study. Pediatrics 2009;123:e932–39. [DOI] [PubMed] [Google Scholar]
- 7. Lawn RB, Lawlor DA, Fraser A.. Associations between maternal prepregnancy body mass index and gestational weight gain and daughter’s age at menarche: the Avon Longitudinal Study of Parents and Children. Am J Epidemiol 2018;187:677–86. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Brix N, Ernst A, Lauridsen LLB. et al. Maternal pre-pregnancy obesity and timing of puberty in sons and daughters: a population-based cohort study. Int J Epidemiol 2019;48:1684–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Maisonet M, Christensen KY, Rubin C. et al. Role of prenatal characteristics and early growth on pubertal attainment of British girls. Pediatrics 2010;126:e591–600. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Kubo A, Deardorff J, Laurent CA. et al. Associations between maternal obesity and pregnancy hyperglycemia and timing of puberty onset in adolescent girls: a population-based study. Am J Epidemiol 2018;187:1362–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Aghaee S, Laurent CA, Deardorff J. et al. Associations of maternal gestational weight gain and obesity with the timing of pubertal onset in daughters. Am J Epidemiol 2019;188:1262–69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Savinainen SE, Viitasalo A, Sallinen TM, Jääskeläinen JES, Lakka TA.. Child-related and parental predictors for thelarche in a general population of girls: the PANIC study. Pediatr Res 2020;88:676–80. [DOI] [PubMed] [Google Scholar]
- 13. Lindqvist M, Lindkvist M, Eurenius E, Persson M, Ivarsson A, Mogren I.. Leisure time physical activity among pregnant women and its associations with maternal characteristics and pregnancy outcomes. Sex Reprod Healthc 2016;9:14–20. [DOI] [PubMed] [Google Scholar]
- 14. Haakstad LAH, Voldner N, Henriksen T, Bø K.. Physical activity level and weight gain in a cohort of pregnant Norwegian women. Acta Obstet Gynecol Scand 2007;86:559–64. [DOI] [PubMed] [Google Scholar]
- 15. Colbert LH, Graubard BI, Michels KB, Willett WC, Forman MR.. Physical activity during pregnancy and age at menarche of the daughter. Cancer Epidemiol Biomarkers Prev 2008;17:2656–62. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Collaborative Group on Hormonal Factors in Breast Cancer. Familial breast cancer: collaborative reanalysis of individual data from 52 epidemiological studies including 58,209 women with breast cancer and 101,986 women without the disease. Lancet 2001;358:1389–99. [DOI] [PubMed] [Google Scholar]
- 17. John EM, Terry MB, Keegan THM. et al. The LEGACY Girls Study: growth and development in the context of breast cancer family history. Epidemiology 2016;27:438–48. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Frederick IO, Williams MA, Sales AE, Martin DP, Killien M.. Pre-pregnancy body mass index, gestational weight gain, and other maternal characteristics in relation to infant birth weight. Matern Child Health J 2008;12:557–67. [DOI] [PubMed] [Google Scholar]
- 19. Morris NM, Udry JR.. Validation of a self-administered instrument to assess stage of adolescent development. J Youth Adolesc 1980;9:271–80. [DOI] [PubMed] [Google Scholar]
- 20. Marshall WA, Tanner JM.. Variations in pattern of pubertal changes in girls. Arch Dis Child 1969;44:291–303. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Terry MB, Goldberg M, Schechter S. et al. Comparison of clinical, maternal, and self pubertal assessments: implications for health studies. Pediatrics 2016;138:e20154571. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Kuczmarski RJ, Ogden CL, Guo SS. et al. CDC growth charts for the United States: methods and development. Vital Health Stat 11 2000;2002:1–190. [PubMed] [Google Scholar]
- 23. Euling SY, Herman-Giddens ME, Lee PA. et al. Examination of US puberty-timing data from 1940 to 1994 for secular trends: panel findings. Pediatrics 2008;121:S172–91. [DOI] [PubMed] [Google Scholar]
- 24. Hiatt RA, Stewart SL, Hoeft KS. et al. Childhood socioeconomic position and pubertal onset in a cohort of multiethnic girls: implications for breast cancer. Cancer Epidemiol Biomarkers Prev 2017;26:1714–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Adair LS. Size at birth predicts age at menarche. Pediatrics 2001;107:e59. [DOI] [PubMed] [Google Scholar]
- 26. Nguyen QC, Osypuk TL, Schmidt NM, Glymour MM, Tchetgen Tchetgen EJ.. Practical guidance for conducting mediation analysis with multiple mediators using inverse odds ratio weighting. Am J Epidemiol 2015;181:349–56. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. VanderWeele TJ, Knol MJ.. A tutorial on interaction. Epidemiol Methods 2014;3:33–72. [Google Scholar]
- 28. Calthorpe L, Brage S, Ong KK.. Systematic review and meta-analysis of the association between childhood physical activity and age at menarche. Acta Paediatr 2019;108:1008–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Boyne MS, Thame M, Osmond C. et al. Growth, body composition, and the onset of puberty: longitudinal observations in Afro-Caribbean children. J Clin Endocrinol Metab 2010;95:3194–200. [DOI] [PubMed] [Google Scholar]
- 30. Hvidt JJ, Brix N, Ernst A, Braskhøj-Lauridsen LL, Ramlau-Hansen CH.. Size at birth, infant growth, and age at pubertal development in boys and girls. Clin Epidemiol 2019;11:873–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Wang Y, Dinse GE, Rogan WJ.. Birth weight, early weight gain and pubertal maturation: a longitudinal study. Pediatr Obes 2012;7:101–09. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Johnson TS, Engstrom JL, Warda JA, Kabat M, Peters B.. Reliability of length measurements in full-term neonates. J Obstet Gynecol Neonatal Nurs 1998;27:270–76. [DOI] [PubMed] [Google Scholar]
- 33. Johnson TS, Engstrom JL, Haney SL, Mulcrone SL.. Reliability of three length measurement techniques in term infants. Pediatr Nurs 1999;25:13–17. [PubMed] [Google Scholar]
- 34. Papadimitriou A, Kanakis G, Douros K. et al. Constitutional advancement of growth is associated with early puberty in girls. Horm Res Paediatr 2011;76:273–77. [DOI] [PubMed] [Google Scholar]
- 35. Terry MB, Wei Y, Esserman D, McKeague IW, Susser E.. Pre- and postnatal determinants of childhood body size: cohort and sibling analyses. J Dev Orig Health Dis 2011;2:99–111. [DOI] [PubMed] [Google Scholar]
- 36. Biro FM, Greenspan LC, Galvez MP. et al. Onset of breast development in a longitudinal cohort. Pediatrics 2013;132:1019–27. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Terry MB, Keegan THM, Houghton LC. et al. Pubertal development in girls by breast cancer family history: the LEGACY girls cohort. Breast Cancer Res 2017;19:69. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Hanson M, Godfrey KM, Lillycrop KA, Burdge GC, Gluckman PD.. Developmental plasticity and developmental origins of non-communicable disease: theoretical considerations and epigenetic mechanisms. Prog Biophys Mol Biol 2011;106:272–80. [DOI] [PubMed] [Google Scholar]
- 39. Zambrano E, Ibáñez C, Martínez-Samayoa PM, Lomas-Soria C, Durand-Carbajal M, Rodríguez-González GL.. Maternal obesity: lifelong metabolic outcomes for offspring from poor developmental trajectories during the perinatal period. Arch Med Res 2016;47:1–12. [DOI] [PubMed] [Google Scholar]
- 40. Shalitin S, Phillip M.. Role of obesity and leptin in the pubertal process and pubertal growth—a review. Int J Obes Relat Metab Disord 2003;27:869–74. [DOI] [PubMed] [Google Scholar]
- 41. Pinkney J, Streeter A, Hosking J, Mohammod M, Jeffery A, Wilkin T.. Adiposity, chronic inflammation, and the prepubertal decline of sex hormone binding globulin in children: evidence for associations with the timing of puberty (Earlybird 58). J Clin Endocrinol Metab 2014;99:3224–32. [DOI] [PubMed] [Google Scholar]
- 42. Kubo A, Ferrara A, Laurent CA. et al. Associations between maternal pregravid obesity and gestational diabetes and the timing of pubarche in daughters. Am J Epidemiol 2016;184:7–14. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 43. Institute of Medicine and National Research Council. Weight gain during pregnancy: reexamining the guidelines. Washington, DC: National Academies Press, 2009. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article will be shared on reasonable request to the corresponding author.