Uterine Leiomyomas

Uterine leiomyomas (fibroids) are the leading indication for hysterectomy in the United States. Despite the morbidity and high medical costs associated with fibroids, there has been little epidemiologic study of this condition. Uterine leiomyomas are histologically identifiable as benign smooth muscle tumors with varying amounts of associated fibrous tissue. Many women have more than one uterine leiomyoma, but each appears to be clonally distinct. MED12 mutations that appear to cause "gain of function" are found in the majority of tumors and are likely triggers for tumorigenesis. Several specific cytogenetic changes have also been identified in tumor tissue, but most tumors show no chromosomal abnormalities. These benign tumors are hormone-dependent. They develop after puberty and regress after menopause. Both estrogen and progesterone are considered important stimulants, or at least permissive factors for tumor growth.

To address the research needs in this field we designed four studies, the NIEHS Uterine Fibroid Study, the Fibroid Growth Study, and the Postpartum Uterine Regression Study, and a fourth, the major current focus of our fibroid group. In 2010 2012 we enrolled nearly 1700 African American women, aged 23-34, in the Study of Environment, Lifestyle & Fibroids (SELF), based in the Detroit, Michigan area with collaboration from Henry Ford Health System. Any prior diagnosis of fibroids was an exclusion criterion. The Participants were screened for fibroids with ultrasound at enrollment (detection limit of lesion = 0.5 cm diameter). There were three subsequent clinic visits at approximately 20-month intervals to monitor fibroids by ultrasound examinations to identify onset time. Fibroids detected at enrollment (newly detected, not previously clinically diagnosed), as well as those that develop during the study, were followed in the same manner to assess fibroid growth. We collected risk factor and symptom data at enrollment and at each 20-month visit for five years.

The study was designed to collect a broad spectrum of exposure data including recognized risk factors for fibroids (age of menarche, pregnancy history, alcohol use, body mass index) and potential risk factors for which there has been only suggestive data. Three primary hypotheses will be tested. (1) Vitamin D deficiency is a risk factor for fibroids, (2) Reproductive tract infections are risk factors for fibroids, and (3) A higher proportion of African ancestry is associated with increased fibroid risk (African ancestry measured by informative SNPs known to have very different frequencies between Europeans and Africans). We have completed participant-visits for the 3rd and final follow-up. Our participation at the 5-year follow-up was 90% of those initially enrolled. We continue in our efforts to develop longitudinal data files for the prospective fibroid incidence and growth analyses, We have developed survivorship methods we will use to investigate exposure effects on fibroid incidence, and have modified the mixed model regression analyses used in the Fibroid Growth Study to assess exposure effects on fibroid growth.

Using the prospective ultrasound data on fibroid development, we have the first age-specific measures of ultrasound-detected fibroid incidence (new fibroid detected in a recently documented fibroid-free uterus). We find an increase in incidence with age, and the frequency of incidence before age 30 supports the previously reported earlier onset of fibroids in Blacks compared to Whites based on estimates modeled from cross-sectional fibroid prevalence data. An ongoing analysis of age and fibroid growth indicates that unlike incidence, growth decreases with age, even after adjusting for the demonstrated variation in growth by fibroid size (smaller fibroids have more rapid growth on average than larger fibroids).

In an examination of oral contraceptive (OC) use and cumulative incidence of fibroid at 40 months, those who used OCs had reduced incidence compared to those who had used OCs, the great majority of the cohort. However among women who had used OCs there was no increased fibroid incidence for those who used longer rather than shorter times or for those who used recently compared to early in their lives. Infact, the risk of fibroid incidence were somewhat elevated for recent an long-term users among the OC users. An survivorship analysis approach using the full follow-up data will be used to further address this important question.

We have completed a detailed investigation of use of the injectable contraceptive, Depo Provera, and fibroid development. Consistent with our previously published cross-sectional findings, there was an inverse association between Depo Provers use and fibroid development. DMPA use for at least 10 months within 2 years of a study visit was associated with a 40% lower fibroid incidence (adjusted hazard ratio 0.6, 95% Confidence Interval (CI) 0.3, 1.0) and 45% lower fibroid growth (95% CI -54, -33) compared to other women. Recent use of DMPA was also associated with increased likelihood of fibroid shrinkage (adjusted relative risk (aRR) 2.1, 95% CI 1.5, 3.0) and increased fibroid loss (aRR 1.5, 95% CI 1.1, 2.2). There is also a clear indication of lower fibroid growth rates and more fibroid loss among users. These beneficial side-effects of a commonly used contraceptive has the potential to limit fibroid progression and delay (and perhaps even eliminate) the need for invasive surgeries such as hysterectomy. We are currently examining use of the hormonal IUD (another progestin exposure) and its association with fibroid development.

We have completed analyses of several reproductive tract exposures (serologically documented Chlamydia and genital herpes infections, as well as bacterial vaginosis as measured by bacterial assessment of vaginal swab samples). None were found to increase fibroid development, suggesting that this long-standing hypothesis is unlikely to explain the increased fibroid burden seen in Black women compared to White women in the U.S.

Early life infant formula feeding with it's known high phytoestrogen exposure was found to be associated with increased fibroid incidence. This supports the prior predictions based on laboratory studies with the primary phytoestrogen, genistein.

Our findings from our analysis of vitamin D and prospectively monitored fibroid development has just been accepted for publication. Despite the majority of our study sample having insufficient vitamin D, we were able to detect a nearly 10% decrease in fibroid growth and the suggestion of both more fibroid loss and reduced incidence.

We completed genotyping for our study sample, and have begun examining both ancestry and mitochondrial haplotypes in relation to fibroid development.

Three research groups have extramural funding for add-on studies with SELF. Lauren Wise, at Boston University, along with Ganesa Wegienka at Henry Ford Health are investigating endocrine disrupting chemical exposures and fibroid development. They have measured several endocrine disruptors, but the concentrations observed showed no convincing evidence of increase fibroid incidence. Anissa Vines at the University of North Carolina at Chapel Hill is examining psychosocial factors, including perceived racism, and fibroid development.

I have also continued my collaborations with Ebbie Stewart in the U-Compare study to examine pregnancies following fibroid treatments and with Erica Marsh on her studies of fibroids in Hispanic women.