DNA damage as assessed by 32P-postlabelling in three rat strains exposed to dietary tamoxifen: the relationship between cell proliferation and liver tumour formation

Unveiling the Pathogenesis of Adenomyosis through Animal Models

Background: Adenomyosis is a common gynecological disorder traditionally viewed as “elusive”. Several excellent review papers have been published fairly recently on its pathogenesis, and several theories have been proposed. However, the falsifiability, explanatory power, and predictivity of these theories are often overlooked. Since adenomyosis can occur spontaneously in rodents and many other species, the animal models may help us unveil the pathogenesis of adenomyosis. This review critically tallies experimentally induced models published so far, with a particular focus on their relevance to epidemiological findings, their possible mechanisms of action, and their explanatory and predictive power. Methods: PubMed was exhaustively searched using the phrase “adenomyosis and animal model”, “adenomyosis and experimental model”, “adenomyosis and mouse”, and “adenomyosis and rat”, and the resultant papers were retrieved, carefully read, and the resultant information distilled. All the retrieved papers were then reviewed in a narrative manner. Results: Among all published animal models of adenomyosis, the mouse model of adenomyosis induced by endometrial–myometrial interface disruption (EMID) seems to satisfy the requirements of falsifiability and has the predictive capability and also Hill’s causality criteria. Other theories only partially satisfy Hill’s criteria of causality. In particular, animal models of adenomyosis induced by hyperestrogenism, hyperprolactinemia, or long-term exposure to progestogens without much epidemiological documentation and adenomyosis is usually not the exclusive uterine pathology consequent to those induction procedures. Regardless, uterine disruption appears to be a necessary but not sufficient condition for causing adenomyosis. Conclusions: EMID is, however, unlikely the sole cause for adenomyosis. Future studies, including animal studies, are warranted to understand how and why in utero and/or prenatal exposure to elevated levels of estrogen or estrogenic compounds increases the risk of developing adenomyosis in adulthood, to elucidate whether prolactin plays any role in its pathogenesis, and to identify sufficient condition(s) that cause adenomyosis.

Tamoxifen-DNA adduct formation in monkey and human reproductive organs

The estrogen analog tamoxifen (TAM), used for adjuvant therapy of breast cancer, induces endometrial and uterine tumors in breast cancer patients. Proliferation stimulus of the uterine endometrium is likely involved in tumor induction, but genotoxicity may also play a role. Formation of TAM-DNA adducts in human tissues has been reported but remains controversial. To address this issue, we examined TAM-DNA adducts in uteri from two species of monkeys, Erythrocebus patas (patas) and Macaca fascicularis (macaque), and in human endometrium and myometrium. Monkeys were given 3-4 months of chronic TAM dosing scaled to be equivalent to the daily human dose. In the uteri, livers and brains from the patas (n = 3), and endometrium from the macaques (n = 4), TAM-DNA adducts were measurable by TAM-DNA chemiluminescence immunoassay. Average TAM-DNA adduct values for the patas uteri (23 adducts/10(8) nucleotides) were similar to those found in endometrium of the macaques (19 adducts/10(8) nucleotides). Endometrium of macaques exposed to both TAM and low-dose estradiol (n = 5) averaged 34 adducts/10(8) nucleotides. To examine TAM-DNA persistence in the patas, females (n = 3) were exposed to TAM for 3 months and to no drug for an additional month, resulting in low or non-detectable TAM-DNA in livers and uteri. Human endometrial and myometrial samples from women receiving (n = 8) and not receiving (n = 8) TAM therapy were also evaluated. Women receiving TAM therapy averaged 10.3 TAM-DNA adducts/10(8) nucleotides, whereas unexposed women showed no detectable TAM-DNA. The data indicate that genotoxicity, in addition to estrogen agonist effects, may contribute to TAM-induced human endometrial cancer.

Hepatocellular carcinoma arising from hepatocellular adenoma in a hepatitis B virus-associated cirrhotic liver

Hepatocellular adenoma (HCA) is a rare, benign proliferation of hepatocytes that occurs mostly in a normal liver and in extreme rare cases, occurs in a cirrhotic liver. Hepatocellular carcinomas (HCC) arising within HCA through malignant transformation is rare. The specific incidence and mechanism of malignant transformation has not been established, but the long term use of oral contraceptives is considered a causative agent. We report a case of HCC arising from HCA detected in a hepatitis B-related cirrhotic liver with serial radiologic images.

Is tamoxifen a genotoxic carcinogen in women?

The anti-oestrogen tamoxifen, which is widely used in the treatment of breast cancer and is also approved for the prevention of this disease, causes an increased incidence of endometrial cancer in women. The ability of tamoxifen to induce endometrial tumours and the underlying carcinogenic mechanisms have been a subject of intense interest over the last approximately 20 years. They are central to the assessment of risks versus benefits for the drug, especially in a chemopreventive context. This review outlines the clinical justification for using tamoxifen as a chemopreventive agent and describes the genotoxic mechanisms considered responsible for tamoxifen-induced tumours in rat liver and how these might relate to women. In rat hepatic tissue, tamoxifen is metabolically activated via alpha-hydroxylation and sulphate conjugation to give a reactive species that binds to DNA predominantly at the N(2)-position of guanine, producing pro-mutagenic lesions. Whether tamoxifen-DNA adducts contribute similarly to the development of cancers in women depends on whether they can be formed in human tissues and the type of specific molecular and cellular responses they induce, if present. This review discusses the current data relating to these issues and highlights areas where further research is needed.

Reduction in tamoxifen-induced CYP3A2 expression and DNA adducts using antisense technology

Tamoxifen (TAM) is widely used in the treatment and prevention of breast cancer. There is clear evidence that cytochrome P450 (CYP) 3A enzymes play an important role in TAM metabolism, resulting in metabolites that lead to formation of TAM-DNA adducts. We have investigated the effect of CYP3A2 antisense (AVI-4472) exposure on CYP3A2 transcription, enzyme activity, translation, and TAM-DNA adducts, in livers of rats administered TAM (50 mg/kg body weight [bw]/day) for 7 days. The study design included administration of 0, 0.5, 2.5, or 12.5 mg AVI-4472/kg bw/day for 8 days, beginning 1 day before TAM exposure. The specific activity of CYP3A2 was increased after TAM administration, and decreased significantly (approximately 70%) in the presence of 12.5 mg AVI-4472. CYP3A2 protein levels, determined by immunoblot analysis, showed a similar pattern. Hepatic TAM-DNA adduct levels were measurable in all TAM-exposed groups. However, when rats were co-treated with 2.5 and 12.5 mg AVI-4472/kg bw/day, statistically significant (approximately 50%) reductions in TAM-DNA adduct levels (2.0-2.8 adducts/10(8) nucleotides) were observed compared to rats treated with TAM alone (5.1 adducts/10(8) nucleotides). Rat toxicology U34 arrays (Affymetrix) were used to investigate the modulation of gene expression patterns on co-administration of TAM with AVI-4472. Results indicated that several CYP genes were down regulated although no significant induction of CYP3A2 was observed in the TAM-exposed rats co-treated with AVI-4472. Overall the data suggest the utility of antisense technology in the redirection of TAM metabolism thereby lowering TAM genotoxicity in rat liver.

The role of metabolic activation in drug-induced hepatotoxicity

The importance of reactive metabolites in the pathogenesis of drug-induced toxicity has been a focus of research interest since pioneering investigations in the 1950s revealed the link between toxic metabolites and chemical carcinogenesis. There is now a great deal of evidence that shows that reactive metabolites are formed from drugs known to cause hepatotoxicity, but how these toxic species initiate and propagate tissue damage is still poorly understood. This review summarizes the evidence for reactive metabolite formation from hepatotoxic drugs, such as acetaminophen, tamoxifen, diclofenac, and troglitazone, and the current hypotheses of how this leads to liver injury. Several hepatic proteins can be modified by reactive metabolites, but this in general equates poorly with the extent of toxicity. Much more important may be the identification of the critical proteins modified by these toxic species and how this alters their function. It is also important to note that the toxicity of reactive metabolites may be mediated by noncovalent binding mechanisms, which may also have profound effects on normal liver physiology. Technological developments in the wake of the genomic revolution now provide unprecedented power to characterize and quantify covalent modification of individual target proteins and their functional consequences; such information should dramatically improve our understanding of drug-induced hepatotoxic reactions.

Hepatic DNA adduct dosimetry in rats fed tamoxifen: a comparison of methods

Liver homogenates from rats fed tamoxifen (TAM) in the diet were shared among four different laboratories. TAM-DNA adducts were assayed by high pressure liquid chromatography-electrospray tandem mass spectrometry (HPLC-ES-MS/MS), TAM-DNA chemiluminescence immunoassay (TAM-DNA CIA), and (32)P-postlabeling with either thin layer ((32)P-P-TLC) or liquid chromatography ((32)P-P-HPLC) separation. In the first study, rats were fed a diet containing 500 p.p.m. TAM for 2 months, and the values for measurements of the (E)-alpha-(deoxyguanosin-N(2)-yl)-tamoxifen (dG-N(2)-TAM) adduct in replicate rat livers varied by 3.5-fold when quantified using 'in house' TAM-DNA standards, or other approaches where appropriate. In the second study, rats were fed 0, 50, 250 or 500 p.p.m. TAM for 2 months, and TAM-DNA values were quantified using both 'in house' approaches as well as a newly synthesized [N-methyl-(3)H]TAM-DNA standard that was shared among all the participating groups. In the second study, the total TAM-DNA adduct values varied by 2-fold, while values for the dG-N(2)-TAM varied by 2.5-fold. Ratios of dG-N(2)-TAM:(E)-alpha-(deoxyguanosin-N(2)-yl)-N-desmethyltamoxifen (dG-N(2)-N-desmethyl-TAM) in the second study were approximately 1:1 over the range of doses examined. The study demonstrated a remarkably good agreement for TAM-DNA adduct measurements among the diverse methods employed.

Neonatal tamoxifen treatment of mice leads to adenomyosis but not uterine cancer

Tamoxifen is contraindicated during pregnancy but many births have been reported in breast cancer patients taking this drug and numbers might be expected to increase with FDA approval of tamoxifen for risk reduction in women at high, risk of breast cancer. The neonatal mouse, exquisitely sensitive to xenobiotic estrogens, has been used to investigate the effects of short-term oral dosing with tamoxifen (1 mg/kg on days 2-5 after birth) on long-term changes in uterine pathology and gene expression. Increased adenomyosis incidence and severity was evident in the tamoxifen-treated mice with increasing age. Uterine weights in treated mice remained lower than the corresponding controls up until 9 months, after which they became greater but during life-time studies (up to 36 months), there was no development of uterine tumours. Pathological examination of uterine tissues showed there to be extensive down-growth of endometrial glands and stroma into thickened, abnormal myometrium that had disorganised fascicles of smooth muscle and increased interstitial collagen deposition. In advanced cases, the endometrial epithelium showed mild degrees of focal hyperplasia and squamous metaplasia but no atypical cytology suggestive of premalignant change. Microarray analysis of uterine RNA taken at 1.5, 3, 6, 9 and 12 months showed from 4500 ESTs, only 12 genes were continuously over-expressed by tamoxifen treatment over this time, while none was continuously down-regulated. Up-regulated genes include those for nerve growth factor (Ngfa), cathepsin B (Ctsb), transforming growth factor beta induced (Tqfbi) and collagens (Colla1, Colla2). Results provide a basis for understanding the mechanism for tamoxifen induced tissue remodelling and the development of adenomyosis.

Gene expression changes induced by estrogen and selective estrogen receptor modulators in primary-cultured human endometrial cells: signals that distinguish the human carcinogen tamoxifen

Tamoxifen has long been the endocrine treatment of choice for women with breast cancer and is now employed for prophylactic use in women at high risk from breast cancer. Other selective estrogen receptor modulators (SERMs), such as raloxifene, mimic some of tamoxifen's beneficial effects and, like tamoxifen, exhibit a complex mixture of organ-specific estrogen agonist and antagonistic properties. However, accompanying the positive effects of tamoxifen has been the emergence of evidence for an increased risk of endometrial cancer associated with its use. A more complete understanding of the mechanism(s) of SERM carcinogenicity and endometrial effects is therefore required. We have sought to compare and characterise the transcript profile of tamoxifen, raloxifene and the agonist estradiol in human endometrial cells. Using primary cultures of human endometria, to best emulate the in vivo responses in a manageable in vitro system, we have shown 230 significant changes in gene expression for epithelial cultures and 83 in stromal cultures, either specific to 17beta-estradiol, tamoxifen or raloxifene, or changed across more than one of the treatments. Considering the transcriptome as a whole, the endometrial responses to raloxifene or tamoxifen were more similar than either drug was to 17beta-estradiol. Treatment of endometrial cultures with tamoxifen resulted in the largest number of gene changes relative to control cultures and a high proportion of genes associated with regulation of gene transcription, cell-cycle control and signal transduction. Tamoxifen-specific changes that might point towards mechanisms for its proliferative response in the endometrium included changes in retinoblastoma and c-myc binding proteins, the APCL, dihydrofolate reductase (DHFR) and E2F1 genes and other transcription factors. Tamoxifen was also found to give rise to the highest number of gene expression changes common to those that characterise malignant endometria. It is anticipated that this study will provide leads for further and more focused investigation into SERM carcinogenicity.

The gene expression of hepatic proteins responsible for DNA repair and cell proliferation in tamoxifen-induced hepatocarcinogenesis

Altered gene expression of the DNA repair- and cell proliferation-associated proteins/enzymes was examined during the process of tamoxifen-induced hepatocarcinogenesis in female Sprague-Dawley rats. When rats were treated by gavage with a single dose of tamoxifen (20 mg/kg body weight) or with the same dose given at 24-h intervals for 2, 12 or 52 weeks, no histopathological change was observed in the liver after 2 weeks. Pathologically altered cell foci and placental form of glutathione-S-transferase (GST-P)-positive foci were observed in the liver after 12 weeks of treatment. Treatment for 52 weeks resulted in the formation of liver hyperplastic nodules that strongly expressed GST-P. During the process of carcinogenesis, changes in hepatic gene expression of DNA repair proteins/enzymes (XPA and XPC, xeroderma pigmentosum complementation groups A and C, respectively; APE, apurinic/apyrimidinic endonuclease) and of cell proliferation-associated proteins (c-myc; PCNA, proliferating cell nuclear antigen; cyclin D1, cyclin B, and p34cdc2) were examined by RT-PCR. The gene expression of XPA and APE was increased by the tamoxifen treatment for 2 or 12 weeks, but no increase was observed after the 52-week treatment. In addition, no significant change in XPC gene expression occurred at any period examined. The gene expression of c-myc, PCNA, and cyclin D1 was increased in a time-dependent fashion up to 12 weeks of treatment, and this increase was maintained up to 52 weeks of treatment. The gene expression of cyclin B and p34cdc2 was increased after the 1-day treatment, reverted to the control level at 2 and 12 weeks of treatment, and was remarkably increased after the 52-week treatment. In the present study, we demonstrate the altered gene expression of various proteins/enzymes involved in DNA repair, cell growth and the cell cycle during the process of tamoxifen-induced hepatocarcinogenesis. We discuss the relationship between the altered gene expression and hepatocarcinogenesis.

Assessing cancer risks from short-term exposures in children

For the vast majority of chemicals that have cancer potency estimates on IRIS, the underlying database is deficient with respect to early-life exposures. This data gap has prevented derivation of cancer potency factors that are relevant to this time period, and so assessments may not fully address children's risks. This article provides a review of juvenile animal bioassay data in comparison to adult animal data for a broad array of carcinogens. This comparison indicates that short-term exposures in early life are likely to yield a greater tumor response than short-term exposures in adults, but similar tumor response when compared to long-term exposures in adults. This evidence is brought into a risk assessment context by proposing an approach that: (1) does not prorate children's exposures over the entire life span or mix them with exposures that occur at other ages; (2) applies the cancer slope factor from adult animal or human epidemiology studies to the children's exposure dose to calculate the cancer risk associated with the early-life period; and (3) adds the cancer risk for young children to that for older children/adults to yield a total lifetime cancer risk. The proposed approach allows for the unique exposure and pharmacokinetic factors associated with young children to be fully weighted in the cancer risk assessment. It is very similar to the approach currently used by U.S. EPA for vinyl chloride. The current analysis finds that the database of early life and adult cancer bioassays supports extension of this approach from vinyl chloride to other carcinogens of diverse mode of action. This approach should be enhanced by early-life data specific to the particular carcinogen under analysis whenever possible.

Delayed effects of tamoxifen in hepatocarcinogenesis-resistant Fischer 344 rats as compared with susceptible strains

The anti-oestrogenic drug tamoxifen has been under investigation as a breast cancer chemopreventive agent for at least a decade. However, its use for this purpose is still debatable since it is able to induce liver tumours in rats via a mechanism involving metabolic activation to a DNA adduct-forming electrophilic intermediate. The metabolic activation and adduct-forming properties of tamoxifen are now well characterized but less is known about its ability to induce hepatic cell proliferation, which is also essential for the carcinogenic process. The effects of tamoxifen on liver weight and cell proliferation were compared in female Fischer 344 (F344), Wistar and Lewis rats given the drug in the diet for up to 26 weeks. The onset and duration of hepatic cell proliferation varied between the strains of rat. In Wistar and Lewis but not F344 rats there was a marked increase in hepatocellular proliferation during the first 4 weeks of tamoxifen administration. In the Wistar strain this was associated with an increase in DNA adduct levels; no such increase was observed in the F344 strain. The onset of the proliferative response was delayed until the 13 week time point in the F344 strain. By the 13 and 26 week time points, cell proliferation in tamoxifen-treated Wistar and Lewis rat liver had returned to normal, but the amount of apoptotic activity in these livers was elevated. This suggests that excess cells generated during the proliferative phase of tamoxifen treatment were being eliminated by apoptosis. In the F344 strain, however, increased proliferative activity was associated with relatively low apoptotic activity at the 26 week time point, suggesting that the delayed proliferative response had yet to be balanced by apoptotic deletion. This is consistent with the fact that tamoxifen-induced hepatocellular tumours develop very late, towards the end of the lifespan, in this strain. The cell proliferative activity of tamoxifen in the Wistar rat liver was compared with that of a non-mutagenic analogue, toremifene. Tamoxifen induced increased cell cycle activity in the livers of rats following gavage dosing at all sampling times (1-12 weeks), whereas toremifene had no effect on the incidence of cycling in hepatic cells, demonstrating that the hepatic cell proliferation is not a general response to anti-oestrogen treatment. These observations suggest that the rate of promotion of liver tumours by tamoxifen is a function of the rate, time of onset and duration of increased cell replication. The susceptibility of rat strains to the hepatocarcinogenic effects of tamoxifen appears to depend upon the balance between initiation via DNA adduct formation, promotion via increased cell proliferation and cell deletion via apoptosis. Our findings suggest that an early proliferative response to tamoxifen is important in this process.

Effects of age and strain on small intestinal and hepatic antioxidant defense enzymes in Wistar and Fisher 344 rats

Age- and strain-associated alterations in intestinal and hepatic antioxidant defense enzymes including superoxide dismutase (SOD), glutathione peroxidase (GSH-PX), and glutathione-S-transferase (GST), and lipid peroxidation were examined in Wistar and F344 rats of both strains aged 2 weeks, 2.5, 10 and 23 months. In the small intestine, activities of SOD and GSH-PX and lipid peroxidation were not affected by age or strain difference. Intestinal GST activity was noticeably increased with age in both strains, but somewhat different pattern of age-related changes occurred between two strains. Wistar rats aged 23 months had a significantly higher intestinal GST activity than corresponding age of F344 rats. In the liver, cytosolic SOD activity was not affected by age and strain, whereas GSH-PX and GST activities and lipid peroxidation were markedly influenced by age or strain difference. In particular, hepatic GSH-PX in Wistar rats resulted in a significant increase after 10 months of age and stayed at this level till 23 months of age we examined. Also, Wistar rats showed a higher lipid peroxidation in the liver of 2.5 months old when compared with corresponding age of F344 rats. However, F344 rats did not show any significant age-dependent changes in GSH-PX and lipid peroxidation. In contrast, the GST activity did show much of an age-associated alteration in both strains. Age-associated change in GST activity of Wistar rats was much greater than that observed in F344 rats, especially late in the lifetime (23 months old). It is concluded from our results that age has profound impact on development of some antioxidant enzymes in the small intestine and liver and also strain-related difference in development of antioxidant defense system was observed at least some time of rat life.

Short-term dosing of alpha-hydroxytamoxifen results in DNA damage but does not lead to liver tumours in female Wistar/Han rats

It is now generally accepted that activation of tamoxifen occurs as a result of metabolism to alpha-hydroxytamoxifen. In this study, alpha-hydroxytamoxifen was given to female Wistar/Han rats (0.103 or 0.0103 mmol/kg, intraperitoneally, daily for 5 days). This resulted in liver DNA damage, determined by (32)P-post-labelling, of 3333 +/- 795 or 343 +/- 68 adducts/10(8) nucleotides, respectively (mean +/- SD, n = 4). Following HPLC separation, the retention times of the major alpha-hydroxytamoxifen DNA adducts were similar to those seen following the administration of tamoxifen. However, after rats were treated with alpha-hydroxytamoxifen (0.103 mmol/kg) for 5 days and the animals kept for up to 13 months, no liver tumours developed (0/7 rats), even with phenobarbital promotion (0/5 rats). GST-P foci were detected in the liver, but only after 13 months was their number or area significantly increased over the corresponding controls. When alpha-hydroxytamoxifen was given to female lambda/lacI transgenic rats (0.103 mmol/kg orally for 10 days) and the animals killed 46 days later, there was an approximate 1.8-fold increase in mutation frequency but no significant increase in G:C to T:A transversions as described after tamoxifen treatment. It is concluded that DNA damage alone, resulting from the short-term administration of alpha-hydroxytamoxifen, is not sufficient to initiate liver tumours even with phenobarbital promotion. As with tamoxifen, long-term exposure may be required to allow promotion and progression of transformed cells.

Anti-oestrogenic drugs and endometrial cancers

Tamoxifen is one of the most effective drugs to be used in the treatment of women with breast cancer and as a chemopreventive agent in women 'at risk' from this disease. Tamoxifen can be regarded as a paradigm for a new range of selective oestrogen receptor modulators that include toremifene, used in the treatment of metastatic breast cancer and raloxifene, presently approved for use in postmenopausal women for the treatment of osteoporosis. Tamoxifen treatment of women leads to a small increase in the incidence of endometrial cancers. It is important to understand the mechanism for this side effect in order to predict the likely human risk for other drugs of this class. Two such mechanisms have been proposed: (1) conversion of the drug to electrophilic metabolites that damage cellular DNA; and (2) an oestrogen agonist action on the uterus, promoting endogenous lesions. In rats, long-term tamoxifen treatment results in liver cancer via a genotoxic mechanism. However, it seems most likely that, in women treated with tamoxifen, endometrial cancer is related to an oestrogen agonist effect of this drug, promoting uterine cell proliferation.

Tamoxifen mutagenesis and carcinogenesis in livers of lambda/lacI transgenic rats: selective influence of phenobarbital promotion

Administration of tamoxifen (TAM) (20 mg/kg per day p.o.) for 6 weeks to female lambda/lacI transgenic rats caused a 4-fold increase in mutation frequency (MF) at the lacI gene locus in the livers of dosed animals compared with controls. After cessation of dosing, the MF showed a further increase with time at 2, 12 and 24 weeks, respectively. Phenobarbital promotion of similarly treated animals resulted in no increase in mutation frequency compared with TAM alone. Treatment with phenobarbital or TAM+phenobarbital resulted in time-dependent increases in liver weight compared with the corresponding controls. There was an increase in cell proliferation in the phenobarbital and TAM+phenobarbital groups, and at 24 weeks in the TAM dosed animals compared with controls. There was also a progressive increase in the number of GST-P expressing foci in the livers of TAM and TAM + phenobarbital rats compared with controls. The induction of cell proliferation and GSTP foci in the rat liver by phenobarbital is consistent with its ability to promote tamoxifen-initiated liver tumours in the rat. If the lacI gene is regarded as being representative of the rat genome in general (albeit that the gene is bacterial) the above observations suggest that promotion by tamoxifen confers selective advantage on mutated genes at loci that contribute to the tumour phenotype and that promotion of rat liver tumours by tamoxifen is not dependent simply upon the enhancement of cellular proliferation.

Chemoprevention of breast cancer by tamoxifen: risks and opportunities

The antiestrogen tamoxifen is widely used in the adjuvant therapy of breast cancers in women and helps to prevent the occurrence of breast tumors in healthy women. However, epidemiological studies have shown tamoxifen treatment to be associated with a 2- to 5-fold increased risk of endometrial cancer. In rats but not in mice, long-term administration of tamoxifen results in an increase in hepatocellular carcinomas. Mechanistically, this occurs through metabolic activation of the drug, mainly by the CYP3A family, to an electrophilic species, that causes DNA damage in target tissues, and subsequently leads to gene mutations. It is controversial whether low levels of DNA damage occur in human uterine tissues, and there is no evidence that this can be causally related to the mechanisms of carcinogenesis. In healthy women, the risk:benefits for the use of tamoxifen is in part related to the risk of developing breast cancer. The results from the carcinogenicity studies in rats do not predict the likelihood that women will develop liver cancer or indeed cancers in other organs. The mechanism of endometrial cancer in women remains unresolved, but the experience with tamoxifen has highlighted the potential problems that need to be addressed in the assessment of future generations of selective estrogen receptor modulators.

Association of tamoxifen biliary excretion rate with prior tamoxifen exposure and increased mdr1b expression

ATPase transporter proteins are commonly found in the hepatocyte canalicular membrane. Some of these, in particular the multidrug resistance (mdr1b) gene, have been previously demonstrated to be inducible genes. In this study, we found that tamoxifen induced expression of the mdr1b gene in the liver up to 40-fold after 14 days' exposure to tamoxifen in the diet at a concentration of 420 ppm. As tamoxifen and its metabolites are primarily excreted into the bile, we investigated if the increased expression of mdr1b in the liver following tamoxifen exposure had any effect on its excretion in rats. We found that the excretion of tamoxifen and its metabolites into bile was increased from 8 +/- 1% to 51 +/- 18% (mean +/- SD) of an administered dose of 180 nmol/kg over a collection period of 3 hr in rats that had received tamoxifen (35 mg/kg) orally for 12 days (plus a 3-day rest) prior to the experiment. These data suggest that prolonged treatment with tamoxifen may result in lower serum and tumour concentrations, due to a self-mediated enhancement of excretion via mdr1b gene-encoded P-glycoprotein. This may have implications for other drugs sharing the same route of excretion and co-administered with tamoxifen.

Characterization of the major DNA adducts in the liver of rats chronically exposed to tamoxifen for 18 months

Our previous study has shown that chronic exposure to tamoxifen (TAM) induced formation of high levels of DNA adducts in the liver, the target tissue of TAM-induced carcinogenesis in rats. One of the major DNA adducts (spot 1), as detected by 32P-postlabeling, accounted for 53% of the total adducts. To characterize this major adduct, the current study has compared spot 1 with two previously identified TAM-DNA adducts, i.e. alpha-TAM-N2-deoxyguanine (alpha-TAM-N2-dG) and alpha-N-desmethyl TAM-N2-deoxyguanine (alpha-N-dmTAM-N2-dG) by various rechromatography methods. It was found that spot 1 was further resolved into two fractions during rechromatography analysis, one fraction co-migrated with the alpha-TAM-N2-dG and the other fraction co-migrated with the alpha-N-dmTAM-N2-dG. These findings have demonstrated that chronic exposure to tamoxifen induced the same major DNA adducts, i.e. alpha-TAM-N2-dG and alpha-N-dmTAM-N2-dG as those detected in acutely exposed rats.

Tamoxifen induces endometrial and vaginal cancer in rats in the absence of endometrial hyperplasia

Tamoxifen was administered orally to neonatal rats on days 2-5 after birth and the subsequent effects on the uterus were characterized, morphometrically, over the following 12 months. Tamoxifen inhibited development of the uterus and glands in the endometrium, indicating a classical oestrogen antagonist action. Between 24 and 35 months after tamoxifen treatment there was a significant increase in the incidence (26%) of uterine adenocarcinomas and a 9% incidence of squamous cell carcinomas of the vagina/cervix in the absence of any oestrogen agonist effect in the uterus. This demonstrates that an oestrogen agonist effect is not an absolute requirement for the carcinogenic effect of tamoxifen in the reproductive tract of the rat. The unopposed oestrogen agonist effect of tamoxifen on the endometrium may not be the only factor involved in the development of endometrial cancers. It is possible that tamoxifen causes these tumours via a genotoxic mechanism similar to that seen in rat liver. However, using (32)P-post-labelling we failed to find evidence of tamoxifen-induced DNA adducts in the uterus. Tamoxifen may affect hormonal imprinting of oestrogen receptor responses in stem cells of the uterus, causing reproductive tract cancers to arise at a later time, in the same way as has been proposed for diethylstilbestrol. If these rodent data extrapolate to humans, then women who are taking tamoxifen as a chemopreventative may have an increased risk of vaginal/cervical cancer, as well as endometrial cancer.

N-demethylation accompanies alpha-hydroxylation in the metabolic activation of tamoxifen in rat liver cells

Previous work has shown that a major route of activation of tamoxifen to DNA-binding products in rat liver cells is via alpha-hydroxylation leading to modification of the N(2)-position of guanine in DNA and to a lesser extent the N(6)-position of adenine. Improved resolution by HPLC has now identified two major adducts in rat liver DNA, one of them the aforementioned tamoxifen-N(2)-guanine adduct and the other the equivalent adduct in which the tamoxifen moiety has lost a methyl group. Treatment of rats or rat hepatocytes with N-desmethyltamoxifen gave rise to the second adduct, whereas treatment with tamoxifen or alpha-hydroxytamoxifen gave rise to both. Furthermore, N,N-didesmethyltamoxifen was found to be responsible for an additional minor DNA adduct formed by tamoxifen, alpha-hydroxytamoxifen and N-desmethyltamoxifen. The involvement of metabolism at the alpha position was confirmed in experiments in which [alpha-D(2)-ethyl]tamoxifen, but not [beta-D(3)-ethyl]tamoxifen, produced reduced levels of DNA adducts. Tamoxifen N-oxide and alpha-hydroxytamoxifen N-oxide also gave rise to DNA adducts in rat liver cells, but the adduct patterns were very similar to those formed by tamoxifen and alpha-hydroxytamoxifen, indicating that the N-oxygen is lost prior to DNA binding. These and earlier results demonstrate that in rat liver cells in vivo and in vitro, Phase I metabolic activation of tamoxifen involves both alpha-hydroxylation and N-demethylation, which is followed by Phase II activation at the alpha-position to form a highly reactive sulphate. Detection of tamoxifen-related DNA adducts by (32)P-postlabelling is achieved with >90% labelling efficiency.

Further characterization of the DNA adducts formed in rat liver after the administration of tamoxifen, N-desmethyltamoxifen or N, N-didesmethyltamoxifen

The present study compares the formation of DNA adducts, determined by (32)P-postlabelling, in the livers of rats given tamoxifen and the N-demethylated metabolites N-desmethyltamoxifen and N, N-didesmethyltamoxifen. Results show that after 4 days treatment (0.11 mmol/kg i.p.), similar levels of DNA damage were seen after treatment with either tamoxifen or N-desmethyltamoxifen [109 +/- 40 (n = 3) and 100 +/- 33 (n = 4) adducts/10(8) nucleotides, respectively], even though the concentration of tamoxifen in the livers of tamoxifen-treated rats was about half that of N-desmethyltamoxifen in the N-desmethyltamoxifen-treated animals (51 +/- 16 and 100 +/- 8 nmol/g, respectively). Administration of N, N-didesmethyltamoxifen to rats resulted in a 5-fold lower level of damage (19 adducts/10(8) nucleotides, n = 2). Following (32)P-postlabelling and HPLC, hepatic DNA from rats treated with tamoxifen and its metabolites showed distinctive patterns of adducts. Treatment of rats with N,N-didesmethyltamoxifen gave a major product that co-eluted with one of the minor adduct peaks seen in the livers of rats given tamoxifen. Following dosing with N-desmethyltamoxifen, the major product co-eluted with one of the main peaks seen following treatment of rats with tamoxifen. This suggests that tamoxifen can be metabolically converted to N-desmethyltamoxifen prior to activation. However, analysis of the (32)P-postlabelled products from the reaction between alpha-acetoxytamoxifen and calf thymus DNA showed two main peaks, the smaller one of which ( approximately 15% of the total) also co-eluted with that attributed to N-desmethyltamoxifen. This indicates that N-desmethyltamoxifen and N,N-didesmethyltamoxifen are activated in a similar manner to tamoxifen leading to a complex mixture of adducts. Since an HPLC system does not exist that can fully separate all these (32)P-postlabelled adducts, care has to be taken when interpreting results and determining the relative importance of individual adducts and the metabolites they are derived from in the carcinogenic process.

Uterotrophic effects of tamoxifen, toremifene, and raloxifene do not predict endometrial cell proliferation in the ovariectomized CD1 mouse

The uterotrophic responses of ovariectomized CD1 mice to tamoxifen, toremifene, and raloxifene have been compared to 17beta-estradiol after a treatment period of 72 h. Uterine and vaginal weight, luminal epithelial thickening, and 5-bromodeoxyuridine (BrdU) labeling index in the endometrial stroma were examined. All three pharmaceuticals, as well as 17beta-estradiol, produced increases in the classic estrogen-dependent variables of uterine and vaginal weights after the 3-day treatment period. Tamoxifen, toremifene, raloxifene, and estradiol all increased luminal epithelial thickness, and increased the BrdU labeling index in the endometrial stroma of the uterus. Although the dose response for the uterotrophic effect and the vaginal weight increases for toremifene differed from tamoxifen and raloxifene, in that there was no dose at which these effects were maximal, the stimulation of BrdU labeling index in the endometrial stroma was dose dependent and very similar for all three, at the clinically relevant doses. Treatment-related hypertrophic effects were estimated by examination of the nuclear profile density in the endometrial stroma. Estradiol and tamoxifen caused a greater hypertrophic effect than toremifene and raloxifene, indicating that factors other than an increase in cell number contribute to the overall uterotrophic effect. This demonstrates that the use of uterine weight to estimate the relative estrogenicity of drugs could give a misleading impression of the response of the uterus to estrogen agonists. Variables, such as increased DNA replication, which may be more important to a subsequent potential carcinogenic process in the uterus, for a particular drug, requires separate evaluation.

Tamoxifen metabolism in rat liver microsomes: identification of a dimeric metabolite derived from free radical intermediates by liquid chromatography/mass spectrometry

Tamoxifen has been shown to be a potent liver carcinogen in rats, and generates covalent DNA adducts. On-line high performance liquid chromatography/electrospray ionisation mass spectrometry (HPLC/ESI-MS) has been used to further study the metabolites of tamoxifen formed by rat liver microsomes in the presence of NADPH with a view to identifying potential reactive metabolites which may be responsible for the formation of DNA adducts, and liver carcinogenesis. A metabolite has been detected with a protonated molecule at m/z 773. The mass of this compound is consistent with a dimer of hydroxylated tamoxifen (m/z 388). Analysis of 4-hydroxytamoxifen incubated with a rat liver microsomal preparation showed the formation of a similar metabolite with an apparent MH+ ion at m/z 773, believed to be a dimer of 4-hydroxytamoxifen formed by a free radical reaction. The retention time for this metabolite from 4-hydroxytamoxifen is identical to that of the tamoxifen metabolite, suggesting that these two compounds are the same. The levels of the dimer were higher when 4-hydroxytamoxifen was used as substrate and, in addition, two isomers were detected. It is proposed that tamoxifen was first converted to arene oxides which react with DNA or to 4-hydroxytamoxifen, either directly or via 3,4-epoxytamoxifen, which then undergoes activation via a free radical reaction to give reactive intermediates which can then react with DNA and protein, or with themselves, to give the dimers (m/z 773).

Alpha-hydroxytamoxifen, a genotoxic metabolite of tamoxifen in the rat: identification and quantification in vivo and in vitro

The metabolic formation of a-hydroxytamoxifen, a reactive metabolite of tamoxifen in rat liver, was characterized and quantified in vitro (hepatic microsomal incubations) and in vivo (bile-duct cannulated animals). This minor metabolite was identified by chromatographic and mass spectral comparisons with the authentic compound. The rates of formation of alpha-hydroxytamoxifen in incubations (30 min) of tamoxifen (25 microM) with liver microsomal preparations from women (pool of six), female CD1 mice or female Sprague-Dawley rats, as quantified by liquid chromatography-mass spectrometry (LC-MS), were 1.15+/-0.03, 0.30+/-0.05 and 2.70+/-0.35 pmol/min/mg protein, respectively. Selective inhibition of microsomal P450 indicated that alpha-hydroxylation was catalysed predominantly by CYP3A in humans. Bile-duct cannulated and anaesthetized female rats and mice given [14C]tamoxifen (43 micromol/kg, i.v.) excreted, respectively, 24 and 21% of the administered radioactivity in bile over 5 and 3.5 h. The major radiolabelled biliary metabolite in rats, characterized by LC-MS after enzymic hydrolysis of conjugates, was the glucuronide of 4-hydroxytamoxifen (10% of dose) and only 0.1% of the dose was recovered as alpha-hydroxytamoxifen. After administration of alpha-hydroxytamoxifen (43 micromol/kg, i.v.) to rats, only 1.19% of the administered compound was recovered from a glucuronide metabolite in bile, indicating a possible 0.84% alpha-hydroxylation of tamoxifen in vivo. There was, however, no indication of the presence in bile of either O-sulphonate or glutathione conjugates derived from alpha-hydroxytamoxifen. This study shows for the first time that alpha-hydroxytamoxifen can be glucuronylated in rat liver. Whereas sulphonation results in electrophilic genotoxic intermediates, glucuronidation may represent a means of detoxifying alpha-hydroxytamoxifen.

New approaches to the quantitation of hypertrophy and hyperplasia in hepatomegaly

A histological method utilizing the optical dissector principle has been developed for determining the contribution of hypertrophy and hyperplasia to the hepatomegaly induced by the peroxisome proliferator gemfibrozil. The optical dissector is a technique derived from the 'new stereology' and has been used to estimate the number of hepatocyte nuclear profiles, that are present in a reference volume of tissue. The overall changes due to hypertrophy and hyperplasia in the rat liver after gemfibrozil treatment, did not reach significance, although the zonal hypertrophy change did. This indicated that although there was a 20% increase in liver weight with treatment, the hepatomegaly was caused by a combination of hypertrophy and hyperplasia, neither of which, on its own, was significantly different from the control values. The distinction of hyperplasia from hypertrophy, using a purely histological method, will be useful in assessing whether treatment related sustained hyperplasia is occurring in the liver.

Species differences in the metabolic activation of tamoxifen into genotoxic derivatives: risk assessment in women

Rats and Rhesus monkeys are compared in their response to tamoxifen treatment, with particular reference to tamoxifen-related liver DNA damage and bioactivation of tamoxifen by isolated microsomes in vitro. Monkeys, treated with tamoxifen, accumulate in their livers a metabolite of tamoxifen, N,N-didesmethyl tamoxifen, with powerful inhibitory activity on cytochrome P450-dependent drug metabolism. The accumulation of this metabolite in the monkeys may limit the cytochrome P450-dependent conversion of tamoxifen into reactive derivatives and, in this way, protect against the formation of DNA adducts. This metabolite is also found in the liver and serum of patients taking tamoxifen, but more work is needed to determine whether inhibition of tamoxifen bioactivation also exists in the human patient in vivo and, if so, to what extent and in which organ.

Tamoxifen inhibits colorectal cancer metastases in the liver: a study in a murine model

Liver metastases account for over 70% of deaths resulting from colorectal carcinoma, with survival rates varying between 6-18 months. At present, surgical resection offers the only hope for a cure, while chemotherapy, focal destructive techniques and selective internal radiation offer palliative care. Tamoxifen, a non-steroidal anti-oestrogen is primarily known for treating oestrogen receptor (ER)-positive breast cancer. Some studies suggest that tamoxifen may have beneficial effects in malignancies other than breast cancer. These inhibitory effects, which have been shown to be independent of the ER, highlight new mechanisms of therapeutic action. Using an intrasplenic animal model we report the efficacy of tamoxifen on experimental liver metastases. In this model, a dimethyl hydrazine-induced colon carcinoma cell suspension is introduced into the portal circulation via the spleen, which results in secondary tumour deposits in the liver in virtually all animals. Tamoxifen was administered at a dose of 1 mg/kg suspended in 1.0% methyl cellulose. The control group received an equal volume of the vehicle. The reagents were administered s.c. on the day of metastases induction and were continued daily over a 4 week period. The effect of tamoxifen on tumour growth was assessed by stereology and bromodeoxyuridine immunohistochemistry at selected time points. Data were assessed by a multiple analysis of variance where P < 0.05 was considered significant. In the control group the volume of metastases increased from 44 +/- 41 mm3 at day 10 to 517 +/- 380 mm3, 1394 +/- 598 mm3 and 2082 +/- 675 mm3 by days 16, 22 and 28, respectively. Daily administration of tamoxifen exerted an inhibitory effect on tumour growth during the first 3 weeks, recording a volume of 421 +/- 299 mm3 by day 22 compared with the control group at that time point (P = 0.00004). The inhibitory effect diminished by the fourth week recording a tumour volume of 1344 +/- 674 mm3 by day 28. Inhibition of tumour growth at day 22 coincides with a reduction of cells in the S phase of the cell cycle. The percentage of brdU-positive nuclear profiles in metastases of tamoxifen-treated mice at 3 weeks was 35.87 +/- 5.60% compared with 48.01 +/- 3.96% in the control group (P = 0.001). These data suggest that tamoxifen has a potent inhibitory action on colorectal liver metastases by exerting an effect on cell proliferation.

Comparisons of the binding of [14C]radiolabelled tamoxifen or toremifene to rat DNA using accelerator mass spectrometry

Tamoxifen, widely used as adjuvant therapy in the treatment of breast cancer, is now undergoing trials as a cancer chemopreventative agent. Previous work has shown an association between 32P-postlabelled adducts in rat liver DNA and the development of liver tumours. With the use of accelerator mass spectrometry, [14C]tamoxifen was shown to bind to liver DNA of female rats in a dose-dependent manner and was linear over 0.1-1 mg/kg, compatible with the therapeutic dose used in women (20 mg/person per day). Radiolabel could also be detected in extrahepatic organs, including reproductive and GI-tract, where levels were about 18 and 46%, respectively those seen in liver. Following enzymatic hydrolysis of liver DNA, normal nucleotides by HPLC showed < 2% incorporation of the [14C]radioactivity while > 80% appeared as non-polar products. In contrast, when animals were given an equivalent dose of [14C]toremifene, binding to DNA was an order of magnitude lower than that seen with tamoxifen and no evidence of non-polar adducted nucleotides following HPLC. However, in vitro, using human, rat or mouse liver microsomal preparations, NADPH-dependent binding of both toremifene and tamoxifen to calf thymus DNA could be demonstrated, suggesting that under favourable circumstances toremifene is capable of undergoing conversion to reactive intermediates.

Identification and mechanism of formation of potentially genotoxic metabolites of tamoxifen: study by LC-MS/MS

On-line high-performance liquid chromatography-electrospray ionization mass spectrometry (HPLC-ESI MS) and tandem mass spectrometry (MS/MS) have been applied to the study of tamoxifen metabolism in liver microsomes and to the identification of potentially genotoxic metabolites. The results showed that the hydroxylated derivatives, including 4-hydroxytamoxifen and alpha-hydroxytamoxifen are detoxication metabolites, while arene oxides, their free radical precursors or metabolic intermediates, are the most probable species involved in DNA-adduct formation.

Analysis of tamoxifen-induced DNA adducts by 32P-postlabelling assay using different chromatographic techniques

DNA isolated from livers of rats receiving tamoxifen was analysed by the 32P-postlabelling method. The postlabelled DNA hydrolysis mixture was analysed both by reversed-phase HPLC with 32P on-line detection and by TLC on polyethyleneimine plates followed by autoradiography. Using the HPLC method, five well separated adduct peaks could be detected, while by the TLC method, two groups of adduct spots were observed. The detection limit of the TLC assay was lower (0.5 adducts/10(10) nucleotides) than that of the HPLC assay (3 adducts/10(10) nucleotides). Thus, the TLC assay is more sensitive but also more laborious. The advantages of the HPLC assay were, in addition to better resolution, the ease of quantification and operation.

Peroxidase activation of 4-hydroxytamoxifen to free radicals detected by EPR spectroscopy

4-Hydroxytamoxifen is a major metabolite of the antiestrogenic drug tamoxifen used in the treatment of women with breast cancer. 4-Hydroxytamoxifen is broken down by a horseradish peroxidase/H2O2 system very much more rapidly than tamoxifen and causes much greater DNA damage determined by 32P-postlabelling. EPR spin trapping of 4-hydroxytamoxifen reaction products in the presence of the free radical trap 5,5-dimethyl-1-pyrroline N-oxide, together with glutathione as a hydrogen donor, resulted in the generation of a species with the characteristics of the glutathione thiyl radical (aN approximately 15.3 G, aH approximately 16.2 G). Support for the creation of thiyl radicals comes from the close to stoichiometric time dependent formation of glutathione disulfide concomitant with the loss of glutathione. Similar results were obtained using 4-hydroxytoremifene but no radical formation or glutathione loss could be detected using 3-hydroxytamoxifen (droloxifene). On-line LC-ESI MS analysis of the incubation products from 4-hydroxytamoxifen has identified three products with a protonated molecular mass of 773, consistent with the formation of dimers of 4-hydroxytamoxifen. The role that radical mechanisms have in the carcinogenic effects of tamoxifen in the endometrium or other target organs of women taking this drug remains to be established.

The role of cell death and cell proliferation in the promotion of rat liver tumours by tamoxifen

Administration of tamoxifen to rats results in liver tumours with a latency time that is dependent on the strain of rat used. Wistar and Lewis rats develop liver tumours more rapidly than Fischer rats. Significant increases in the number of apoptotic hepatocytes were found in the Wistar and Lewis strains of rats after they were fed tamoxifen for up to 6 months, but not in Fischer rats. By 6 months of exposure to tamoxifen there were liver tumours in the Wistar and Lewis rats, but not the Fischers. Sustained elevations of the PCNA labelling index were found in the livers of tamoxifen-treated Wistar and Lewis rats, over the first 6 months of tamoxifen treatment, but not Fischers. It is proposed that sustained cell death by apoptosis may play a role in the mechanism of promotion of tamoxifen-induced liver tumours, by causing liver hyperplasia. To support this concept it has been shown that cyloheximide, which causes apoptosis but not necrosis in the rat liver, causes DNA synthesis and cell division in hepatocytes.

A realistic clinical perspective of tamoxifen and endometrial carcinogenesis

Tamoxifen has been the endocrine treatment of choice for all stages of breast cancer for nearly a decade. Millions of women are currently receiving tamoxifen worldwide, while large-scale randomised trials have been launched aiming to investigate the drug's merit as a preventive agent. However, there are now concerns about tamoxifen's potential carcinogenicity. The goal of this review is to address these concerns, re-evaluate the available data from laboratory biological models and those from clinical reports and put the whole issue into perspective. Our focus is the association between tamoxifen and the increased frequency of endometrial tumours, while key issues, such as the role of duration of tamoxifen therapy, are also addressed. Finally, we discuss the various monitoring strategies for early detection of endometrial lesions and pertinent problems most likely to be encountered by clinicians taking care of patients who are receiving tamoxifen.

Mutational spectra of tamoxifen-induced mutations in the livers of lacI transgenic rats

Tamoxifen, an important drug in breast cancer treatment, causes liver cancer in rats. The standard range of in vitro tests have failed to show that it causes DNA damage, but 32P-postlabelling and DNA-binding studies have shown that tamoxifen forms DNA adducts in rat liver. In 1995 a transgenic rat (Big Blue; Stratagene, La Jolla, CA) became available which harbours the bacterial lacI gene, thereby allowing the in vivo study of tamoxifen mutagenesis. Recently, we [Styles JA et al. (1996): Toxicologist 30; 161] showed that tamoxifen caused on increase in the mutation frequency at the lacI gene in these transgenic rats. In this study, we report on our preliminary analysis of the mutational spectra of 33 control and 38 tamoxifen-induced mutant lacI genes. Plasmid DNA containing the lacI gene was isolated from the mutant phages and its DNA sequence determined. In the control animal group, 81% of the mutant lacI genes were point mutations, whilst in the tamoxifen-treated group, 62% of the mutant lacI genes were point mutations. Of the tamoxifen-induced mutants, 43% were GC-->TA transversions and 70% of point mutations. In the control group, GC-->TA transversions were 19% of all mutations and 24% of point mutations. Thus, compared with control animals, tamoxifen treatment had significantly increased the proportion of GC-->TA transversions.

Chemoprevention of breast cancer by tamoxifen: risks and opportunities

The antioestrogen tamoxifen is of proven efficacy in inhibiting the growth of oestrogen receptor positive breast cancers in women. In rats, long-term dosing leads to the development of hepatocellular tumours. Tamoxifen in this species is a genotoxic carcinogen. Metabolic activation by cytochrome P450-dependent enzymes leads to DNA damage detectable by 32P-postlabelling. Factors important in the development of hepatocellular lesions were the nature and quantity of metabolism and promotion/progression of the DNA lesion by agents such as phenobarbital and cell proliferation. No evidence was found for tamoxifen-induced DNA damage in the livers of 7 women taking this drug therapeutically.