Can Placenta Factors Explain Race Patterns of Breast Cancer?

Institution: Public Health Institute
Investigator(s): Barbara Cohn, Ph.D., MPH, MCP -
Award Cycle: 2002 (Cycle VIII) Grant #: 8IB-0193 Award: $84,502
Award Type: IDEA
Research Priorities
Etiology and Prevention>Etiology: the role of environment and lifestyle
Community Impact of Breast Cancer>Disparities: eliminating the unequal burden of breast cancer



Initial Award Abstract (2002)
Asian and Hispanic women experience a striking, reduced incidence of breast cancer in California compared to either Caucasian or African American women. African American women also experience higher rates of mortality and later stage at diagnosis. Understanding race/ethnicity differences in breast cancer incidence offers promise for primary breast cancer prevention and breast cancer mortality reduction for all women. To date, Asian and Hispanic origin, and the novel protective placental factors we described for whites in a recent publication are among the largest protective factors identified. It is critical to understand how they are related.

As for breast cancer, there are striking differences in pregnancy outcomes by race and ethnicity including birth weight. Higher birth weights have reportedly increased breast cancer risk, and birth weights of mothers and daughters are also highly correlated. A key to progress in breast cancer prevention will lie in reconciling critical paradoxes. Race/ethnicity patterns for birth weight (lowest for African Americans and Asians, higher for Hispanics and non-Hispanic whites) do not match patterns of breast cancer in California (lowest in Asians and Hispanics).

There are three aims for this project:

Aim #1 - Test the hypothesis that race/ethnicity differences in the distribution of novel protective placental factors are consistent with patterns of breast cancer incidence in California:
Lower incidence for Asians and Hispanics than for whites and African Americans
Higher incidence at younger ages among African Americans compared to whites.

Aim # 2 - Test the hypothesis that novel protective placental factors for maternal breast cancer, identified for White women, have similar protective associations for African American women.

Aim #3 - Test the hypothesis that black/white differences in protective placental factors explain later stage at diagnosis among African Americans.

We propose to carry out these aims by analysis of existing data from a unique California population study, The Child Health and Development Studies, in which fifteen thousand women participated during their pregnancies between 1959 and 1967. At that time, extensive data were collected on the progress of their pregnancies, including standardized examination of the placentas from their pregnancies. They have been followed for breast cancer using the California Cancer Registry. This is one of the only resources in the United States to examine detailed characteristics of pregnancy recorded decades before cancer developed. We will compare characteristics that relate to the placenta, the organ that regulates the baby's growth and the production of hormones responsible for profound changes in pregnancy, for Asian (Chinese, Japanese, and Korean descent), Hispanic, African American, and non-Hispanic white women to determine whether differences in placental function could explain race ethnicity differences in cancer incidence. For African Americans where sample size is sufficient, we will examine whether protective placental factors are less common and explain larger tumor size and poorer prognosis at diagnosis.

In very recent work, we identified placental factors (lower placental weight and diameter, higher increases in blood pressure, and the presence of an abnormal placental condition called maternal floor infarction) during pregnancy as strong protective factors for breast cancer 40 years later. These risk factors are newly described and have never been studied in African Americans (because sample sizes were not large enough until now, as cancer cases have accumulated in our study population over the last 5 years). Nor have the distributions of these risk factors been described for Asians and Hispanics. An editorial that accompanied our publication suggested that uncovering underlying biological factors that explain the protective relations we observed could lead to prevention strategies. Demonstrating race and ethnicity differences in these factors will provide further clues. For example, there is a possibility that tumor suppressor genes have relevance both to the growth of the placenta (which is like a tumor) as well as to breast cancer, and that these genes vary by race/ethnicity.


Final Report (2005)
Higher birth weights have been reported to increase breast cancer risk. However, race/ethnicity patterns for birth weight (lowest for African Americans and Asians, higher for Hispanics and non-Hispanic Caucasians) do not match patterns of breast cancer in California (lowest in Asians and Hispanics). A key to progress in breast cancer prevention will lie in reconciling critical paradoxes like this one. Unfortunately there are few long-term prospective study populations where both breast cancer, and the characteristics of pregnancies that occurred decades before diagnosis can be studied.

The Child Health and Development Studies have followed women and their families beginning in early pregnancy for more than forty years. Using this unique data, we recently discovered that a number of placental characteristics that may regulate fetal growth are associated with a substantially lower risk of subsequent breast cancer in whites. In this grant we examined these placental characteristics for other race/ethnicity groups to determine whether they could explain race/ethnicity differences in breast cancer incidence and also stage at diagnosis.

Findings 1: After considering placental weight and birth weight together, we discovered that race/ethnicity patterns matched those for breast cancer: Asian (primarily of Chinese and Japanese descent) and Hispanic women (primarily of Mexican descent) had the lowest placental weight for a given birth weight, matching their low incidence of breast cancer, while non-Hispanic Caucasians (whites) and African Americans had higher placental weights for a given birth weight.

Impact 1: We used this clue to refine our understanding about how to study placental size in relation to breast cancer incidence, which is a primary goal of studying race/ethnicity differences. Our work in this new area was not fully funded by this grant, however our preliminary analyses show that the ratio of birth weight to placenta weight may be an even stronger predictor of breast cancer incidence in all women than placental size or birth weight alone. Therefore this grant has already had an important impact on the field by suggesting a new risk factor, which may lead to a better understanding of cause and therefore prevention.

Findings 2: Smaller placental diameter, smaller placental weight and maternal floor infarction (a placental abnormality) were all protective factors among African-American women as already found in white women. The absence of a protective blood pressure association among African-American women was an exception, and could be a result of a higher prevalence of preexisting hypertension among African-Americans. Blood pressure change during pregnancy may not have been a specific marker for placental function in our African American population.

Impact 2: We interpret the consistent findings for most placental risk factors as support for the hypothesis that placental function, or an underlying determinant of placental function is a universal risk factor for breast cancer. This has clear implications for the priority of further research into natural protective mechanisms since this work will most likely be relevant to African American and white women.

Findings 3: African American cases were more likely to have a later stage at diagnosis (regional or distant spread) than Caucasians (44%, 24/55 versus 34%, 64/186), while Asian cases were less likely to have a later stage at diagnosis (13%, 2/15). The difference between Asian and African American cases was statistically significant at p=0.04, indicating that there was only a 4% probability that this difference occurred due to chance alone, despite a small sample. The lower risk of later stage disease among Asian women was partly explained by a lower placenta weight for a given birth weight. The excess late-stage disease in African-Americans compared to Caucasians was not explained by placental factors. A high placental weight, for a given birth weight doubled the probability that the case would be diagnosed at a later stage.

Impact 3: To our knowledge, this study is the first to link placental factors to stage at diagnosis of breast cancer. These results suggest that placental growth, or a related regulatory factor may be involved in tumor progression or detection. This should open a field where biologists can search for an explanation. Clearly it is possible that this new line of research could lead to new prevention or treatment strategies.

Overall Impact: This study is the first to suggest that underlying differences in pregnancy factors could explain the striking protection from breast cancer observed for Asian and Hispanic women. We expect these results to lead to clinical and laboratory investigations that have the goal of identifying growth factors and endocrine factors related to the birth weight to placental weight ratio. Growth and endocrine factors may be responsible for the lower risk of breast cancer in Asian and Hispanic women, despite their very different pre-pregnancy weight, birth weights (Hispanics have considerably larger babies than Chinese and Japanese women in our study), and genetic background. The underlying mechanisms responsible for race/ethnicity differences in placental weight by birth weight may involve the same mechanisms that lead to breast cancer. Understanding how race/ethnicity is related to protective placental characteristics offers promise for primary breast cancer prevention for all women, based on naturally occurring protection during pregnancy.

Supplementary analyses supported by this grant:

A new protective risk factor—Recent reports establishing links between placenta factors and asthma led us to explore these associations in our own data. We found that women who had a doctor’s visit for asthma during pregnancy had higher rates of pre-eclampsia and systolic blood pressure change compared to women without asthma symptoms. When we examined the relationship between asthma and breast cancer, we found a surprising and strongly protective association – women with asthma during pregnancy were 67% less likely to be diagnosed with subsequent breast cancer. Moreover, women who visited a doctor for chronic respiratory disease experienced a 3-fold higher risk of breast cancer than women without respiratory symptoms, indicating that the asthma association is not likely to be an artifact of more frequent use of health care services.

The protective asthma association and chronic respiratory association could not be explained by known breast cancer risk factors including age at first pregnancy, race/ethnicity, parity, body mass and placenta factors (placental size and shape, blood pressure change and weight change during pregnancy). The asthma and respiratory disease associations were independent of use of steroids as a treatment medication, indicating that the breast cancer associations were not a result of steroid medication, but more likely directly related to the conditions. Asthma during pregnancy could represent a state of increased immunity, which protects against developing tumor cells while chronic respiratory infection could represent a state of decreased immunity, which allows tumor cells to go undetected and spread. Investigating mechanisms responsible for the exacerbation of asthma and respiratory infection during pregnancy may ultimately lead to insights about how pregnancy affects breast cancer. The opposing effects of asthma and respiratory infection on breast cancer risk suggest that the mechanisms are complex and specific.

Impact: A new protective factor opens an opportunity for new understanding about cause or breast cancer. The contribution of Immune function to breast cancer is not yet understood. By identifying a protective effect for asthma in pregnancy it may be possible to narrow the search for a specific immune mechanism.


Symposium Abstract (2003)
Much evidence supports the hypothesis that pregnancy influences a woman’s subsequent risk of breast cancer. In our own data, we have previously found pregnancy characteristics that are strongly associated with a reduced risk of breast cancer. These findings suggest that immune and endocrine mechanisms activated during pregnancy may be linked to cancer. Pregnancy characteristics, which appear to alter cancer risk, may provide clues about how cancer develops.

In 1959 the Child Health and Development Studies (CHDS) began enrolling pregnant women who visited the San Francisco East Bay Kaiser Permanente obstetric clinic. Nearly every pregnant woman who was a Kaiser member was enrolled until September 1966. We selected the first, singleton pregnancy per mother resulting in a live-born infant or stillborn with a gestation greater than 19 weeks for our analysis.

Subject’s medical records were abstracted in detail, beginning six months before the woman's last menstrual period (LMP) and extending through the end of her pregnancy. Physician diagnoses for asthma or for respiratory symptoms where adrenalin or ephedrine were prescribed formed the basis for a measure of asthma or asthma-like episodes. We also created a measure for physician diagnoses of chronic respiratory conditions, including chronic bronchitis, laryngitis, nasopharyngitis, pharyngitis, rhinitis, tracheitis or sinusitis.

Routine surveillance of the CHDS cohort is accomplished by linking our files to the California Department of Motor Vehicles, the California Department of Vital Statistics and the California Cancer Registry. This information allows us to ascertain cancer cases and compute the population-at-risk for cancer. Our analysis sample includes 449 breast cancer cases among 11,856 women with information on age at study entry, age at first pregnancy, race and parity.

Women who had a physician visit for asthma during pregnancy experienced a 70% reduction in risk of subsequent breast cancer. However, women with a physician visit for a chronic respiratory infection showed a 3-fold increase in risk. Both findings were statistically significant, meaning that the probability of their chance occurrence was less than 5%. Women who saw a doctor more frequently during pregnancy might have been predisposed to get mammograms for early diagnosis of breast cancer, however, if so, we should not have observed opposite associations for these two conditions.


Symposium Abstract (2003)
Higher birthweights have been reported to increase breast cancer risk, and birthweights of mothers and daughters are highly correlated. However, race/ ethnicity patterns for birthweight (lowest for African Americans and Asians, higher for Hispanics and non-Hispanic whites) do not match patterns of breast cancer in California (lowest in Asians and Hispanics). A key to progress in breast cancer prevention will lie in reconciling critical paradoxes like this one. In our data, we recently discovered that placental characteristics are associated with a substantially lower risk of subsequent breast cancer in whites. We decided to examine these characteristics for other race/ethnicity groups to determine whether they could explain race/ethnicity differences in breast cancer incidence.

The Child Health and Development Studies (CHDS) enrolled pregnant women between 1959 and 1967 who were members of the San Francisco East Bay Kaiser Permanente Health plan. We ascertain cancer cases and compute population at risk for cancer by routinely linking our files with the California Cancer Registry, the California Department of Motor Vehicles and the California Department of Vital Statistics. For this analysis, we selected the first, singleton pregnancy per mother resulting in a live-born infant or stillborn with a gestation greater than 27 weeks, and with information on placental measurements, age at entry, age at first pregnancy, parity and birth weight. The study sample included 1,629 African Americans, 282 Asians, 217 Hispanics and 4,780 non-Hispanic whites.

Birth weight patterns by race/ethnicity in the CHDS were the same as those reported by other researchers: Asians had the lowest birthweights, followed by African Americans, then Hispanics and non-Hispanic whites who have almost overlapping distributions. We also found the same breast cancer incidence patterns in our cohort as observed for California between 1994-1998: Asians and Hispanics had the lowest rates, while African Americans and non-Hispanic whites had the highest rates of breast cancer.

When we examined distributions of placental weight by race/ethnicity, we observed a slightly different profile: Asians had the lowest placental weights and non-Hispanic whites had the highest, with intermediate and overlapping distributions for African Americans and Hispanics. However, after considering placental weight and birthweight together, we discovered that race/ethnicity patterns were virtually identical to those for breast cancer: Asian and Hispanic women had the lowest placental weight for a given birthweight, matching their low incidence of breast cancer, while non-Hispanic whites and African Americans had higher placental weights for a given birthweight.

The underlying mechanisms responsible for race/ethnicity differences in placental weight by birthweight may involve the same mechanisms that lead to breast cancer. Understanding how race/ ethnicity is related to protective placental characteristics offers promise for primary breast cancer prevention.