The purpose of this work was to determine the hormone dependence of ma
mmary tumors induced in Sprague-Dawley rats by three intraperitoneal i
njections of N-nitroso-N-methylurea at 50, 80, and 110 days of age. Tw
o experimental designs were carried out: (a) Ten days before the first
NMU injection, 130 rats were divided into 13 batches and randomly ass
igned to the following treatments: control, ovariectomy (OVX), tamoxif
en (TAM), bromocriptine (BROM), haloperidol (HAL), estradiol (E(2)), p
rogesterone (Pg), OVX + BROM, TAM + BROM, OVX + HAL, TAM + HAL, OVX TAM, and E(2) + BROM. After 150 days off treatment the following growt
h parameters were determined: latency period (LP), mean tumor number p
er rat (n/t) and tumor incidence (TI). LP was significantly increased
(p < 0.05) only by Pg and TAM + BROM. The n/t was significantly decrea
sed (p < 0.05) by all treatments except HAL. TI was significantly redu
ced by OVX, TAM, BROM, and their combinations. (b) Rats bearing ip-NMU
-induced mammary tumors were divided into 7 batches and assigned to th
e following treatments: control, OVX, TAM, BROM, HAL, OVX + BROM, and
TAM + BROM. Tumor growth was assessed up to 60 days of treatment; only
OVX, TAM and their combination with BROM were able to produce tumor r
egression. These results support the essential role of E(2) and prolac
tin in the promotion stage of carcinogenesis. However, for established
tumors, growth becomes more independent from hormone influence, in pa
rticular from prolactin deprivation. We conclude that this model seems
suitable for studying the mechanisms under-lying the evasion of hormo
nal control of tumor growth.