Summary.-Follicular-phase (Day 11) plasma prolactin, and plasma and urinary oestrogen levels of 70 nulliparous nuns were compared with those of 80 of their sisters, of whom 62 were parous. The nuns and their nulliparous sisters did not differ significantly in their prolactin and oestrogen levels. No differences in plasma oestrogens or urinary oestriol ratio were found between the parous and the nulliparous women. However, the mean prolactin level of the nuns and their nulliparous sisters was 3500 higher than that of the parous women in the sample taken 13 h after rising (P < 0-0005), and 24% higher (P <0.01) in the 2nd sample taken 2 h later. The elevation was independent of age, weight, and age at menarche. Age at first full-term pregnancy, at least up to the age of 30, and second or subsequent full-term pregnancies had no further effect on prolactin level. This study suggests that the effect of early first full-term pregnancy in lowering breast cancer risk may be mediated, at least in part, by permanently lowering the level of circulating prolactin.ONE OF THE MOST STRIKING breastcancer risk factors is age at first full-term pregnancy (FFTP): women whose FFTP is before the age of 20 have less than half the breast-cancer risk of nulliparous women (MacMahon et al., 1973). Two theories have been proposed to "explain" how this protective effect of early FFTP is brought about. Cairns (1975) theorized that early FFTP would effectively reduce the number of susceptible breast "stem" cells, while Cole et al. (1976) proposed that the protective effect may be due to a change in the "urinary oestriol ratio". There is no evidence available to either support or refute Cairns's hypothesis, but there is some evidence that early FFTP does increase the urinary oestriol ratio (Cole et al., 1976;Trichopoulos et al., 1980).Prolactin may also play a key role in determining risk of breast cancer (Hill et al., 1976;Malarkey et al., 1977;Pike et al., 1977;Welsch & Meites, 1978) and we were struck by the finding of Vekemans & Robyn (1 975a) in a cross-sectional study that prolactin levels decreased after age 25 in women but not in men. If this decrease with age in women is due to pregnancy, this prolactin change may be a critical part of the protective effect of early FFTP.In this paper we describe our results in comparing the plasma prolactin, and plasma and urinary oestrogen levels of nulliparous and parous women aged in California.
METHODSWith the cooperation of a number of Catholic orders in California, we identified Caucasian nuns aged 20-39, who had never been pregnant and who had at least one sister aged 20-39 residing in the continental U.S. We excluded nuns who had had uterine (excluding D&C) or ovarian surgery. We also