Combined oral contraceptives containing chlormadinone acetate and breast cancer: results of a case-control study

Postmarketing withdrawal of human medicinal products because of adverse reactions in animals: a systematic review and analysis

PURPOSE

We have identified human medicinal products for which animal data were used as evidence for withdrawal, determined whether the adverse reactions were reported in humans, established whether confirmatory human studies were conducted, and explored the withdrawal patterns over time.

METHODS

We searched the World Health Organization's Consolidated List of [Medicinal] Products, drug regulatory authorities' websites, PubMed, Google Scholar, and selected textbooks to identify medicinal products withdrawn from 1950 to June 2016. We included medicinal products for which animal data were specifically reported as a reason for withdrawal. We used a checklist adapted from the International Agency for Research on Cancer criteria to rate the evidence.

RESULTS

In 37 cases, evidence from animals was the reason given for withdrawal between 1963 and 2000. Evidence of carcinogenicity was cited in 23 cases (62%). Limited evidence for harms occasioned withdrawal in over 80% of cases. In 11 cases (30%), the adverse drug reactions were subsequently reported in humans. In 5 instances (14%), formal studies were conducted in humans. The median interval to withdrawal following reports of adverse reactions was 2 years (IQR = 1-9 y).

CONCLUSIONS

Regulatory authorities and drug manufacturers are likely to withdraw medicinal products quickly from the market when animal experiments suggest increased risks of cancers or congenital malformations. Human studies are seldom conducted when harms are suspected in animals. Future research should explore better methods of extrapolating harms data from animal research to humans.

Progestogens and membrane-initiated effects on the proliferation of human breast cancer cells

OBJECTIVES

Evidence is accumulating that progestogens may play a crucial role in the development of breast cancer under contraception and hormone therapy in reproductive and menopausal women. Progesterone receptor membrane component 1 (PGRMC1) expressed in breast cancer may be important in tumorigenesis and thus may increase breast cancer risk. The aim of this project was to investigate the influence of progesterone and nine synthetic progestins on MCF-7 breast cancer cells overexpressing PGRMC1.

METHODS

MCF-7 cells were stably transfected with PGRMC1 expression plasmid (WT-12). To test the effects of progestogerone (P) and the synthetic progestins chlormadinone acetate (CMA), desogestrel (DSG), drospirenone (DRSP), dydrogesterone (DYD), levonorgestrel (LNG), medroxyprogesterone acetate (MPA), nomegestrol (NOM) and norethisterone (NET) on cell proliferation, MCF-7 and WT-12 cells were stimulated with different concentrations (0.01-1 µmol/l).

RESULTS

In MCF-7 cells, DRSP, DSG, DYD, LNG and NET increased the proliferation at 1 µmol/l, the effect being highest for NET with about 20%. In WT-12 cells, the same progestins, but additionally MPA, showed a significant increase, which was much higher (30-245%) than in MCF-7 cells. Here again, NET showed the highest proliferative effect. No effect was found for CMA, NOM and P.

CONCLUSION

Some synthetic progestins trigger a proliferative response of PGRMC1-overexpressed MCF-7 cancer cells. The effect of progestogens on breast cancer tumorigenesis may clearly depend on the specific pharmacology of the various synthetic progestins.

The effect of progesterone, testosterone and synthetic progestogens on growth factor- and estradiol-treated human cancerous and benign breast cells

OBJECTIVE

The effects of progesterone (P), testosterone (T), chlormadinone acetate (CMA), medroxyprogesterone acetate (MPA), norethisterone (NET), levonorgestrel (LNG), dienogest (DNG), gestodene (GSD) and 3-ketodesogestrel (KDG) were investigated in normal human breast epithelial MCF10A cells. In addition, the effects of these steroids were tested in estrogen and progesterone receptor positive HCC1500 human primary breast cancer cells.

STUDY DESIGN

MCF10A cells were incubated with each progestogen at 1 microM and 100 nM for 7 days with growth factors epidermal growth factor (EGF), fibroblast growth factor (FGF) and insulin-like growth factor (IGF-I) (GFs, each 1 pM), and HCC1500 cells with GFs and/or estradiol (E2) 100 pM. Cell proliferation rate was measured by ATP-assay and cell death by photometric enzyme immunoassay. Ratios of cell death:proliferation were calculated.

RESULTS

MPA and CMA with GFs induced proliferation of MCF10A cells. P, T, NET, LNG, DNG, GSD and KDG had no significant effect. In HCC1500 cells, MPA and CMA with GFs had an inhibitory effect compared to GFs alone. NET, LNG, DNG, GSD, KDG and T enhanced the proliferative effect of GFs. P had no significant effect. No progestogen could further enhance the stimulatory effect of E2 on HCC1500 cells; all but KDG inhibited it. MPA, GSD, T, CMA and NET had an anti-proliferative effect on the mitotic GF and E2 combination. P, LNG, DNG and KDG had no significant effect.

CONCLUSIONS

Estrogens and mitogenic growth factors from stromal breast tissue are significant in growth-regulation of breast cells and may alter responses to progestogens. Certain progestogens are able to induce proliferation of or inhibit growth of benign or malignant human breast epithelial cells independently of the effects of growth factors and E2; therefore, choice of progestogen for hormone therapy may be important in terms of influencing possible breast cancer risk.

Characterization of the stimulatory effect of medroxyprogesterone acetate and chlormadinone acetate on growth factor treated normal human breast epithelial cells

OBJECTIVE

Evidence is increasing that adding progestogens to hormone replacement therapy may be more harmful than beneficial, however it is debatable whether all progestogens act equally on breast cells. Mitogenic growth factors from stromal breast tissue are important in growth-regulation of breast cells, and may modify responses to progestogens. We investigated the effect of two C-21 derivatives, medroxyprogesterone acetate (MPA) and chlormadinone acetate (CMA) on growth-factor treated normal breast epithelial cells and tried to explore the underlying mechanisms of proliferation.

METHOD

MCF10A (human epithelial, estrogen- and progesterone-receptor negative normal breast cells) were incubated with MPA or CMA at 0.1 and 1 microM for 7 days with the growth factors (GFs) EGF, bFGF and IGF-I at 1pM. The same combinations, as well as growth factors alone, were also incubated with the proliferation inhibitors PD98059 and LY294002 at 1 microM for 4 days. Cell proliferation rate was measured by the ATP-assay.

RESULTS

MPA 0.1 and 1 microM, and CMA 1 microM in combination with GFs both significantly increased cell proliferation rate, with MPA having the greatest effect. MPA- and CMA-induced proliferation of GF stimulated cells was blocked by both PD98059 (selective inhibitor of MAP kinases) and LY294002 (phosphatidylinositol 3-kinase inhibitor); GF stimulated cells could not be significantly reduced by any of the inhibitors used.

CONCLUSION

MPA and CMA have a stimulatory effect on benign growth factor stimulated MCF10A cells, possibly via activation of MAP kinase and subsequent substrates and activation of PI3-kinase. GF induced proliferation appear to be mediated by pathways other than those investigated here. Growth factors and progestogens therefore have an additive, synergistic effect on cell proliferation, eliciting their effects via different pathways.

Prognostic value of proliferating cell nuclear antigen and c-erbB-2 compared with conventional histopathological factors in breast cancer

Expression of proliferating cell nuclear antigen (PCNA) and c-erbB-2 oncoprotein has been assessed in 471 women with breast cancer to evaluate their prognostic value as compared to conventional histopathological factors. In univariate analysis, high PCNA expression (> or = 20%) predicted a significantly worse survival in lymph-node-negative tumors (univariate P = 0.031). However, the effect disappeared in multivariate analysis and the histological grade remained the only independent factor for this group. Despite its close correlation to histological grade (P < 0.001), PCNA expression discriminated subsets with different survival within the heterogeneous group of moderately differentiated tumors (univariate P = 0.073, multivariate P = 0.075). PCNA expression was not found to be a significant prognostic factor in lymph-node-positive tumors, thus it was of limited value for breast cancer patients as a whole. c-erbB-2 protein overexpression was associated with a worse survival (univariate P = 0.019, multivariate P = 0.057) for the entire group of patients. The effect was mainly attributed to the significance of c-erbB-2 as an independent factor in lymph-node-positive (up to three nodes, multivariate P = 0.04; four or more nodes: multivariate P = 0.017) and large tumors (> 2 cm: multivariate P = 0.002). c-erbB-2 was without significance in lymph-node-negative patients. Though both factors might amplify the prognostic information for distinct patient subsets they do not achieve the strong prognostic value of conventional histopathological features in breast cancer.

C-erbB-2 overexpression in primary breast cancer: independent prognostic factor in patients at high risk

The prognostic value of c-erbB-2 protein overexpression has been evaluated in 463 patients with operable breast cancer after a median follow-up of 66 months. Overexpression was observed in 99/463 (21%) of the breast tumors. It showed significant positive correlation to histological grade (p < 0.0001) and tumor size (p < 0.02). A relationship of borderline significance was observed between c-erbB-2 protein overexpression and negative or low estrogen receptor (ER) content. No significant correlation was found to lymph node involvement or proliferating tumor cell fraction as determined by the proliferating cell nuclear antigen (PCNA). After a median follow-up of 66 months (range 6 to 109 months), the overall survival of all patients amounted to 63%. Multivariate analysis revealed lymph node involvement, tumor size, histological grade, histological type, c-erbB-2 protein overexpression, progesterone receptor (PR) content, and oral contraceptive use as independent prognostic factors. In an univariate analysis, the overall survival amounted to 72% and 38% of tumor patients with negative and positive c-erbB-2 protein overexpression, respectively. The most significant finding is that c-erbB-2 overexpression has been recognized as an independent predictive factor in subsets of tumor patients who would be expected to have a generally poor prognosis, such as those indicating axillary lymph node involvement, large tumor size (> 2 cm), and PR negativity.

Oral contraceptive use and the prognosis of breast cancer

In 471 breast cancer patients the influence of a positive history of oral contraceptive (OC) use on survival was investigated. 297 (63%) patients used OCs during any period of their life and 92 (20%) used them still at the time of diagnosis. Sixty months after diagnosis OC users had a significantly increased overall survival (p = 0.037). Survival rates amounted to 79.5% and 70.3% for OC users and non-users, respectively. The effect persisted after adjustment for other prognostic factors and was mainly attributed to women who had taken OCs four years or longer (p = 0.025). Comparing the survival after a 56 months median follow-up dependent on duration of OC use (never, 1-48 months, > or = 49 months) in subgroups of prognostic factors, the most significant influence on survival was observed among long-term users with tumors more than 2 cm in diameter (p = 0.005), with axillary node-positive tumors (1-3 nodes, p = 0.055/ > or = 4 nodes, p = 0.019), and with tumors of low estrogen receptor (p = 0.015) or progesterone receptor content (p = 0.04). The difference in survival between OC users and non-users cannot be explained by the distribution of prognostic factors investigated (histological type, histological grade, tumor size, lymph node involvement, hormonal receptor content). OC users had an even higher percentage of poorly differentiated tumors (p = 0.003). These results suggest an effect of OC use on tumor biology during the preclinical phase of the disease.

Accuracy of recall of use of an intrauterine device

Within a case-control study of the relationship between oral contraceptives and breast cancer, information on other contraceptive practices, including use of an intrauterine device (IUD) was obtained through interviews of study subjects and from the records of gynaecologists. Information from these two sources was compared for 239 women who had ever used an IUD. The results showed that agreement on total duration, number of IUD episodes, and time since first and time since last IUD use was excellent. As approximately 75% of IUDs used were unknown, agreement on brands could not be investigated. Studies investigating the effects of individual types of IUDs should use physician records as an additional source of information.