Re: Endogenous steroid hormone concentrations and risk of breast cancer among premenopausal women.

JNCI Vol. 99, Issue 5 | March 7, 2007 The case – control study nested within the Nurses’ Health Study II by Eliassen et al. ( 1 ) provides further evidence that circulating sex hormone levels are associated with breast cancer risk. They found that high prediagnostic estrogen and androgen levels were associated with an increased risk of breast cancer in premenopausal women but that there was no association with progesterone levels. Two prospective studies, Hormones and Diet in the Etiology of Breast Tumors [ORDET; (2)] and European Prospective Investigation into Cancer and Nutrition [EPIC; ( 3 )], also showed that a high androgen level was associated with an increased risk of breast cancer but that progesterone level was inversely associated with breast cancer risk. It is diffi cult to investigate associations between breast cancer risk and sex hormones in premenopausal women because hormone levels vary markedly during the menstrual cycle. The approach taken by Eliassen et al. was to separately analyze cryopreserved blood samples that were collected during the luteal and follicular phases of each woman’s menstrual cycle. To better synchronize the timing of blood collection within the cycles, they counted backward from the day on which the woman's next menstrual period started to the bloodsampling day and used the number of days counted to adjust estimates of the relative risk of breast cancer. The ORDET study was designed to assess whether ovarian hyperandrogenism associated with anovulation or luteal in adequacy increased breast cancer risk ( 4 , 5 ); estrogens were not investigated. We controlled for intraindividual hormone variation by collecting blood samples during a narrow window of the menstrual cycle (days 20 – 24) when most women are in the midluteal phase of their menstrual cycle. Women in the ORDET study also gave the date of their next period and information about their menstrual history. When we plotted the mean levels of progesterone, testosterone, luteinizing hormone (LH), and follicle-stimulating hormone (FSH) according the number of days from blood sampling to the start of bleeding in the control subjects, we found that the re sulting curves were similar to the familiar blood concentration curves for these substances in the latter half of the cycle. This fi nding suggested that backward counting is a reliable way of synchronizing when blood was sampled among women who have menstrual cycles of varying length. However, compared with the control subjects, the ORDET case subjects had shorter times from blood sampling to the start of their next menstrual period, shorter cycles, and distinctly lower LH and FSH levels ( 2 ). Because the length of the luteal phase varies less than the length of the Re: Endogenous Steroid Hormone Concentrations and Risk of Breast Cancer Among Premenopausal Women