Placental Weight and Breast Cancer Risk in Young Women: A Registry-Based Cohort Study from Norway

Pregnancy has a dual effect on breast cancer risk indicated by a short-term risk increase followed by a lifelong risk reduction (1, 2). The short-term increase in risk lasts for 5 to 10 years (1, 2) and has been attributed to the promoting effects that pregnancy hormones such as estrogens, progesterone, prolactin, and insulin-like growth factor-I may have on initiated breast cancer cells or prevalent but undetected tumors (3). This hypothesis is supported by indications that women with a family history of breast cancer, who may be more likely to develop breast cancer at a young age, seem more susceptible to the short-term adverse effects of pregnancy than other women (4). Some investigators have found that the increased risk after a pregnancy may be more pronounced among uniparous women (1, 3), whereas others have found a similar risk increase also among women with more than one birth (2).

Breast cancers diagnosed within the first 5 to 10 years following a pregnancy constitute a small proportion of all breast cancer cases (5). Still, the personal and public health impact is considerable as these tumors affect young mothers and typically have a poorer prognosis than breast cancer in older women (5). A deeper understanding of factors that influence the risk of breast cancer in young women is needed as it may provide clues to improve prevention and treatment.

Whether hormone levels during pregnancy are related to subsequent breast cancer risk remains unclear (6–8). During pregnancy, the placenta is the main source or regulator of many of the potentially implicated hormones (9). Although placental weight may be a crude indicator of maternal hormone levels in pregnancy (10–13), 2 studies have previously reported strong associations of placental weight with later breast cancer risk (14, 15). However, more research is needed to clarify whether women who develop large placentas in their pregnancies are at increased risk of breast cancer.

On the basis of the hypothesis that pregnancy hormones may promote the growth of prevalent but undetected tumors, it seems to be most relevant to study the most recent pregnancy before breast cancer diagnosis. Using data from a prospective and nationwide registry-based cohort of Norwegian women, we have investigated whether a high placental weight in a woman's most recent pregnancy is associated with increased breast cancer risk during the first years after the pregnancy.

Data were obtained from the Medical Birth Registry of Norway and the Cancer Registry of Norway. In the Medical Birth Registry, information on all births since 1967 is collected through a standardized form completed by midwives and doctors. Since 1999, information on placental weight has also been included, and therefore, the study population is restricted to women registered with births since 1999. Thus, a total of 348,671 women met the following inclusion criteria: information on date of birth available for all births, singleton births only, and at least one birth between January 1, 1999, and December 31, 2008. We excluded 87 women who were diagnosed with breast cancer before the start of follow-up and 10,533 women (3.0%) with missing information on placental weight and/or length of pregnancy in all their pregnancies. The remaining 338,051 women were followed from the date of their first registered birth after January 1, 1999, until the date of breast cancer diagnosis, death, emigration, or December 31, 2008, whichever came first. Among these 338,051 women, 216,223 gave their first birth after 1998.

Placental weight has been registered since 1999. The placenta is weighed according to standardized routines. Immediately after delivery and independent of delivery mode, the placenta is placed in a bowl with membranes and umbilical cord attached. The bowl is weighed within one hour after delivery. The weight of the bowl is subtracted, and the placental weight is reported in grams.

From the Medical Birth Registry, we also obtained information on the following factors: age at first birth, number of previous pregnancies, prepregnancy diabetes mellitus (type I, type II, or unspecified), gestational diabetes (blood glucose levels after glucose tolerance test ≥ 7.8 mmol/L), hypertension (blood pressure ≥ 140/90 mm Hg after 20 weeks of gestation, measured repeatedly), preeclampsia (hypertension with proteinuria of at least 1+ on a semiquantitative dipstick after 20 weeks of gestation, measured repeatedly), offspring's birth weight in grams, length in centimeters, and duration of pregnancy in completed weeks (based on ultrasound examination in 95.9% of the pregnancies and on the first day of the last menstrual period in 4.1%). The diagnoses of diabetes mellitus and hypertensive disorders were registered by the use of separate check boxes in the standardized form.

The unique identification number of Norwegian citizens was used to link information from the Medical Birth Registry to information on cancer occurrence from the Cancer Registry of Norway, which also includes information on vital status and emigration, as recorded by the Population Registry at Statistics Norway. Invasive breast cancer was registered according to the International Classification of Diseases, 7th edition (ICD-7, code 170). Among women in this study, all breast cancer diagnoses were verified through histologic (99.4%) or cytologic (0.6%) examination. Reporting of new diagnoses of cancer to the Cancer Registry is mandatory by law, and follow-up for breast cancer is considered practically complete (16).

The study was approved by the regional committee for medical research ethics.

The association of placental weight with breast cancer incidence was evaluated using Cox proportional hazards models to estimate hazard ratios (HRs) with 95% confidence intervals (CI). Placental weight was categorized as <500, 500–599, 600–699, 700–799, and ≥800 grams, and the lowest category (<500 g) was used as the reference in the statistical analyses.

We studied the association of placental weight in the most recent pregnancy with breast cancer risk, treating placental weight as a time-dependent variable. Thus, women contributed person-time to one or more categories of placental weight, depending on the placental weight in the most recent pregnancy. To investigate whether the association could be modified by the number of previous pregnancies, we conducted analyses stratified by parity (uniparous, biparous, and triparous). In these analyses, follow-up of women with a given parity (1, 2, or 3) were censored whenever a subsequent birth occurred.

To account for the possibility of measurement error in placental weight, we also investigated whether mean placental weight among a woman's pregnancies was associated with breast cancer risk. After each delivery, mean placental weight was calculated and included as a time-dependent variable.

To facilitate comparison with previous studies (15), we also investigated whether placental weight in a woman's first pregnancy was associated with breast cancer risk. This analysis was restricted to the 216,223 women who had given their first birth after 1998, and differed from the previously described analysis among uniparous women, as there was no censoring at subsequent births.

Analyses of trend were conducted by scoring each category of placental weight with the median value of placental weight within that category. The categories of placental weight were then treated as a continuous variable, and HRs with 95% CI per 100-gram increase in placental weight were estimated.

For all analyses, attained age was used as the time scale. We adjusted for maternal birth year (before 1965, 1965–1969, 1970–1974, 1975–1979, and 1980 or later), age at first birth (<20, 20–24, 25–29, 30–34, and ≥35 years), duration of pregnancy (<37, 37, 38, 39, 40, 41, and ≥42 weeks), number of pregnancies, hypertensive disease in pregnancy (including preeclampsia), and diabetes mellitus (pre pregnancy or gestational). In the analyses of placental weight in the most recent pregnancy and in the analyses of mean placental weight, the number of pregnancies, duration of pregnancy, hypertensive disease in pregnancy, and diabetes mellitus were treated as time-dependent variables. The number of pregnancies was categorized as 1, 2, 3, and ≥4. Hypertensive disease in pregnancy and diabetes mellitus were treated as cumulative variables and categorized as ever/never. In the analysis of placental weight in the first pregnancy, only the number of subsequent pregnancies was treated as a time-dependent variable (1, 2, and ≥3).

Birth weight and birth length were considered as consequences of placental function and therefore as potential colliders (17). Therefore, these variables were not included as covariates in the final analyses. Nonetheless, the causal relation between these measures and placental weight is unknown, and we therefore conducted a separate analysis to investigate whether adjustment for birth weight (<2,500, 2,500–2,999, 3,000–3,499, 3,500–3,999, 4,000–4,499, and ≥4,500 grams) and length (<47, 47–48, 49–50, 51–52, 53, and ≥54 cm) could influence the association of placental weight with breast cancer risk. In a separate analysis, we also assessed whether a further categorization of placental weights above 800 grams (800–899, 900–999, and ≥1,000 grams) would influence the association with breast cancer risk.

Proportionality between hazards was checked by comparing log minus log plots of survival and by conducting tests based on Schoenfeld residuals. Assumptions were met for all the previously described models.

All analyses were conducted using Stata version 11.1 MP for Windows (Stata Corp.).

A total of 338,051 women without breast cancer at baseline were followed for a total of 1895,111 person years. Median time of follow-up was 6.0 years (interquartile range, 3.0–8.3 years). During follow-up, 648 women were diagnosed with invasive breast cancer. Mean age at diagnosis was 38.4 years (standard deviation, 5.3 years). Among the 338,051 women, 57.3% contributed with one pregnancy, whereas the remaining 42.7% contributed with between 2 and 7 pregnancies. Characteristics of the study participants are shown in Table 1, and characteristics of the included pregnancies are shown in Table 2.

Breast cancer risk among women in different categories of placental weight is presented in Table 3. Placental weight in a woman's most recent pregnancy was not associated with breast cancer risk: the HR per 100-gram increase in placental weight was 1.03 (95% CI, 0.96–1.10). The risk among women with placental weight in the highest category (≥800 grams) in their most recent pregnancy was similar (HR, 1.07; 95% CI, 0.77–1.49) to that of women with placental weight in the reference (<500 grams) category. For mean placental weight, the estimated associations were slightly stronger than that for placental weight in the most recent pregnancy. In women with mean placental weight in the highest compared to the lowest category, the HR was 1.16 (95% CI, 0.82–1.63), and per 100 grams increase in mean placental weight, the HR was 1.04 (95% CI, 0.97–1.11). Among women who gave their first birth after 1998, placental weight in the first pregnancy was not associated with breast cancer risk (HR, 1.02; 95% CI, 0.91–1.13, per 100-gram increase in placental weight).

Table 4 shows associations of placental weight in the most recent pregnancy with risk of breast cancer, stratified by parity. Among uniparous and biparous women, there was no association of placental weight with breast cancer risk (HR per 100-gram increase: 0.98; 95% CI, 0.85–1.13 and 1.02; 95% CI, 0.92–1.12, respectively). Among triparous women, there was a positive but moderate association (HR per 100-gram increase: 1.08; 95% CI, 0.96–1.21). The statistical power in these analyses was limited, with only 11 cases in the reference group, and should therefore be interpreted with caution.

Crude analysis (adjusted for age only) provided similar results as the fully adjusted analysis, and additional adjustment for birth weight or birth length had no material influence on the results. In a separate analysis, treating placental weights above 800 grams in distinct categories, there was no indication that a particularly high placental weight was associated with increased risk of breast cancer (results not shown).

In this prospective study of 338,051 women, placental weight was not associated with breast cancer risk. Associations were similar for placental weight in the most recent pregnancy, for mean placental weight across pregnancies, and for placental weight in the first pregnancy. No clear associations were detected among uniparous, biparous, or triparous women.

We could not confirm the results from previous studies on placental weight and breast cancer risk. In a small study from the United States of both primiparous and multiparous women, it was reported that low placental weight was associated with reduced risk for breast cancer (14). In a large, registry-based study from Sweden, there was a consistent positive association of placental weight in a woman's first pregnancy with later breast cancer risk (15). Compared with those studies, the women in our study were followed up for a shorter period, and the cases were younger at the time of breast cancer diagnosis. However, in the 2 previous studies, there was no evidence that the associations differed by duration of follow-up (14, 15). Therefore, it seems unlikely that the lack of association in our study can be explained by shorter follow-up alone.

In our study, there was no information on the women's body mass index, which could have confounded the association of placental weight with breast cancer risk. A high body mass index is associated with reduced risk of breast cancer among premenopausal women (18), and body mass index is positively associated with placental weight (19). A confounding effect of body mass index could therefore mask a positive association. However, adjustment for preeclampsia or hypertension in pregnancy and diabetes mellitus, both of which are strongly associated with obesity (20), did not influence our estimates. Furthermore, there was no substantial confounding by body mass index in the Swedish study (15). Therefore, our results are not likely to be strongly influenced by the lack of adjustment for body mass index.

Our study encompassed all women who gave birth in Norway during the study period, and information on incident breast cancer was collected prospectively. Therefore, bias in selection or information recall is unlikely. Misclassification of breast cancer diagnosis is most likely negligible, as reporting of cancer to the Cancer Registry of Norway is mandatory by law, and all breast tumors among women in this study were based on pathology reports. Although attempts are made to standardize measurements of placental weight, some inaccuracy is likely to occur, since placental weight depends on factors such as mode of delivery, removal of blood clots, and time to cord clamping (21). However, misclassification of placental weight is likely to be non-differential with respect to subsequent breast cancer diagnosis, and bias the association towards the null. Using placental weight in the most recent pregnancy and using mean placental weight yielded similar results, but nonetheless, we cannot exclude the possibility that misclassification of placental weight could have masked a possible positive association with breast cancer risk.

It has been reported that the correlation of placental weight is high with maternal pregnancy levels of insulin-like growth factor-I (13) and moderate with maternal levels of sex steroids (10–12). The lack of a clear association of placental weight with breast cancer risk in this study may suggest that differences in placental weight may not reflect differences in pregnancy hormone levels that are relevant for breast cancer risk during the first years following pregnancy.

In conclusion, placental weight was not related to the risk of early breast cancer in this study. More research is needed to clarify whether placental weight is associated with breast cancer risk, and if so, for which age groups it may be of importance.

No potential conflicts on interest were disclosed.

The authors thank the Cancer Registry of Norway and the Medical Birth Registry of Norway for providing the data.

The study was financially supported by a grant from the Norwegian Research Council.

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.