195-miller

An International Evaluation of the Cancer-Prevention An International Evaluation of the Cancer-Preventive Potential
of Nine Retinoids
Anthony B. Miller1,2, Paul Nettesheim1,3, Bernard W.Stewart4
Abstract
The International Agency for Research on Cancer (IARC) convened a working Group of experts in March 1999 to
evaluate the cancer preventive potential of nine retinoids and to compile the fourth volume of the IARC Handbooks
of Cancer Prevention. The handbook provides a comprehensive review of the relevant information in the published
scientific literature through March 1999 on the potential role of all-trans-retinoic acid, 13-cis-retinoic acid, 9-cis-
retinoic acid, all-trans N-(4-hydroxyphenyl)retinamide, etretinate, acitretin, N-ethylretinamide, targretin and LGD
1550 in cancer prevention. Of these, the data suggest that all-trans-retinoic acid, 13-cis-retinoic acid and N-
ethylretinamide are not suitable for chemoprevention of cancer in humans either because they are too toxic, may
enhance cancer occurrence or are ineffective. In contrast, 9-cis-retinoic acid, etretinate and acitretin show some
promise, but more data are required, while all- trans N-(4-hydroxyphenyl)retinamide is quite promising. Targretin
and LGD 1550 are of interest, based on theoretical grounds, but there are no significant human and little experimental
data as yet.
Asian Pacific J Cancer Prev, 1, 195-202
Key words: biomarkers, chemoprevention, evaluation, retinoids
Introduction
Remarks’ in which features common to anti-inflammatoryagents and relevant to their impact on cancer are delineated.
In 1997 the International Agency for Research on Cancer Volume 2 concerned ‘Carotenoids’ (IARC, 1998a), while (IARC) initiated its series ‘IARC Handbooks of Cancer Volume 3 was restricted to Vitamin A (IARC, 1998b), a Prevention’ with publication of a volume in which the cancer scenario which anticipated subsequent evaluation of certain chemopreventive effects of non-steroidal anti-inflammatory retinoids which have been employed in the context of drugs were subject to evaluation (IARC, 1997). The Handbook series has much in common with the IARC The retinoids are a class of compounds structurally related Monographs on ‘The Evaluation of Carcinogenic Risks to to Vitamin A. In the last 30 years, more than 2,500 retinoids Humans’ which first appeared in 1972. The first volume of have been synthesized and biologically tested, with the the new series contained four sections, each presenting a objective of identifying those with an enhanced therapeutic systematic summation of published data for a specific anti- ratio. Within the first generation of retinoids, all trans-retinoic inflammatory agent (aspirin, sulindac, piroxicam and acid (tretinoin) and 13-cis-retinoic acid (isotretinoin) were indomethacin) and culminating in an evaluation of evidence identified. The second generation included the aromatic for cancer chemopreventive activity (IARC, 1997). This retinoids etretinate and acitretin with an enhanced therapeutic volume (in common with all in the series) also contained a ratio and the third generation the poly-aromatic retinoids with Preamble in which the organization and principles of the or without polar end groups. The majority of the retinoids evaluation process are described, together with ‘General have been studied for their cancer-preventive activity in 1 Unit of Chemoprevention, International Agency for Research on Cancer, 150 Cours Albert Thomas, 69372 Lyon, Cedex 08, France2 Division of Clinical Epidemiology, Deutsches Krebsforschungszentrum, Im Neuenheimer Feld 280, D-69120 Heidelberg, Germany3 National Institute for Environmental Health Sciences, P.O.Box 12233, Research Triangle Park, NC 27709, USA 4 Cancer ControlProgram, South East Health, Locked Bag 88, Randwick NSW 2031, Australia Asian Pacific Journal of Cancer Prevention, Vol 1, 2000 195
experimental models, and some in clinical trials in humans.
On 24-30 March, 1999, a working group of international The relationship of toxicity of an agent to its beneficial experts met in Lyon to consider the existing evidence on the effects (the risk/benefit or therapeutic ratio referred to earlier) cancer-preventive activity of nine retinoids. The resulting is critical with regard to chemoprevention because the handbook details the evidence considered, and summarizes subjects considered for this approach to cancer prevention the evaluations made by the working group, and should be are healthy, and it is likely (with the exception of carriers of referred to for relevant citations relating to the statements the rare dominant cancer-susceptibility genes) that their probability of not developing cancer is substantially greater The retinoids subject to evaluation by the working group than the risk that they will develop it. These considerations are not so prominent for agents considered for use among patients at risk for second primary tumors, as patients may be prepared to tolerate side-effects in this situation, regarding this as an extension of their therapy. Toxic effects also may be less problematic when retinoids are used in dermatology, a role for which at least one of the agents considered in this handbook has become almost standard therapy (e.g. 13-cis- retinoic acid in acne). Concerns over toxicity are also likely to be less problematic when the agent is used as therapy for a malignancy (e.g. all-trans-retinoic acid in acutepromyelocytic leukemia). But this is not so for most It was recognized that in view of the many retinoids which chemoprevention applications, and therefore particular might be considered for evaluation, the designation of nine attention to the toxicity of these compounds was given during would necessitate omitting a number of compounds including the evaluation by the working group.
those at various stages of development, or which have been The influence of retinoids on plasma retinol has relevant studied and are no longer under clinical consideration. Some clinical implications as regard to toxicity and it might also of the newer retinoids are already in preliminary human study.
be relevant for the preventive effect. all-trans-Retinoic acid Those eventually found to have a better risk/benefit ratio and all-trans N-(4-hydroxyphenyl)retinamide were first could eventually find a role for use in humans.
shown to inhibit the concentrations of endogenous retinol in Evaluation of chemopreventive effects in humans is humans and in rats. The same effect has been found to occur impossible by conventional observational analytic with 13-cis-retinoic acid, 9-cis-retinoic acid and N- epidemiology studies (case-control and cohort) if an agent ethylretinamide. Endogenous retinol level reduction has been is not in use in the general population, as is the case for all related with the high binding affinity of retinoids bearing the agents to be considered in this volume. Therefore the modifications in the area of the retinol hydroxyl end group human evidence that was considered by the working group with retinol-binding protein, the specific protein that was derived almost exclusively from randomized trials, either with development of cancer as the endpoint, or utilizing somebiomarker of neoplasia. Thus almost uniquely to date in this Background to retinoid action
series, with the exception of sulindac in Volume 1 of theHandbook series (IARC, 1997), the human data arose largely Retinoids are involved in signal transduction. Most of their as a consequence of calculated and orderly usage, a effects are believed to be mediated through two families of characteristic normally restricted to experimental studies.
retinoid receptors, retinoic acid (RARs a, b and g) and retinoid This availability of clinical trial data was a considerable X (RXRa, b and g) receptors. These receptors belong to the advantage, as it meant that bias and confounding were largely superfamily of nuclear receptors (NR), comprising such taken care of in the design of the trial. However, it has one diverse receptors as those for steroids and thyroid hormones, important disadvantage: conclusions must necessarily be retinoids and vitamin D3, which are present in vertebrates, restricted to the population studied in the trial, and arthropods, and nematodes. The members of this superfamily extrapolation to other groups would be tenuous at best, and act both as ligand-modulated transcriptional activators and/ potentially in serious error. Further, many of the clinical or suppressors, while for a large group of so-called “orphan” trials considered involved individuals already diagnosed with nuclear receptors no ligands exist or have not yet been found.
one neoplasm, and hence known to be at substantial risk of a Nuclear receptors may have acquired ligand-binding ability second primary tumor, either at the same or a different site during evolution, suggesting that the ancestral NR was an but with similar etiology. In considering these trials, the orphan. Control of gene expression by retinoid receptors, working group found that often the evidence for like all nuclear receptors, results from both direct modulations chemopreventive efficacy was derived from a secondary data of the activity of cognate gene programs, the mutual analysis (chemoprevention of the relevant cancer site had interference with the activity of other signalling pathways not been the primary hypothesis of the trial), and thus the and regulatory events that occur at the post-transcriptional findings require independent confirmation, which has not level (e.g., mRNA and/or protein stabilization or 196 Asian Pacific Journal of Cancer Prevention, Vol 1, 2000
An International Evaluation of the Cancer-Prevention nitrosodiethylamine-induced liver carcinogenesis, and in one Retinoid receptors regulate complex physiological events study in rats, it was ineffective against N-methyl-N- that trigger key steps during development, control nitrosourea-induced mammary carcinogenesis.
maintenance of homeostasis, and induce or inhibit cellular In vitro, all-trans-retinoic acid inhibited the transformation proliferation and differentiation, and cell death. Importantly, of normal cells by carcinogens and of immortalized cells by retinoid receptors display a strong differentiative and anti- viral oncogenes. all-trans-Retinoic acid inhibited cell proliferative activity. Each of the subtypes of retinoid proliferation in monolayer cultures and modulated the receptors includes three isotypes designated a, b, and g, differentiation of a large number of immortalized, localized to chromosomes 17q21, 3p24,and 12q13, transformed and tumorigenic cell types derived from trachea, respectively. The RXRa, RXRb, and RXRg genes have been skin and cervical epithelia. all-trans-Retinoic acid also mapped to chromosome 9q34.3, 6p21.3 and 1q22-23, suppressed the anchorage-independent growth of a variety respectively. The RARs bind both all-trans-retinoic acid and of tumor cell lines and abnormal squamous differentiation 9-cis-retinoic acid, whereas the RXRs bind only 9-cis-retinoic in immortalized, transformed and arrested cells in the G1 acid. These receptors also bind a variety of synthetic retinoids, some of which exhibit RAR or RXR selectivity or Inhibitory effects of all-trans-retinoic acid against preferentially bind to specific RAR isotypes.
carcinogen-induced genotoxicity were most often observed Retinoids induce cellular differentiation or suppress when agents were used that required bioactivation. Inhibition proliferation in many malignantly transformed cell lines of carcinogen-induced neoplastic transformation in vitro including epithelial cancers, melanoma, neuroblastoma and occurs when all-trans-retinoic acid is added after carcinogen leukaemia, and germ cell, bone and breast cancers. The most exposure. Studies in animals and humans after topical plausible mechanism by which retinoids may affect various application of all-trans-retinoic acid have demonstrated aspects of carcinogenesis (both early and late events) is alterations in enzymes that mediate carcinogen metabolism.
modulation of genes whose products regulate cell growth, Most studies of the effects of all-trans-retinoic acid on differentiation and apoptosis. Indeed, some genes involved carcinogenesis indicate that inhibition of the post-initiation in cell cycle control such as p21WAF1/Cip1 and cyclin D1 or in stage is the main mechanism of its putative preventive effects.
differentiation are regulated by retinoid-activated receptors.
The mechanisms of action may be related to growth Immediate evidence that changes in retinoid receptors may inhibition, induction of differentiation and/or apoptosis.
be associated with carcinogenesis is based on the It was concluded by the IARC working group that there chromosomal translocation of RAR b in acute pro-myelocytic is inadequate evidence that all-trans-retinoic acid has leukemia and loss of RAR b expression in various malignant cancer-preventive activity in humans, and inadequate evidence that it has cancer-preventive activity inexperimental animals (IARC, 1999). In addition, all-trans- Summary of the literature and results of the
retinoic acid therapy gives rise to significant toxicity, and evaluations
is an established teratogen in experimental animals.
1. all-trans-Retinoic acid
2. 13-cis-Retinoic acid
No studies have been reported of the use of all-trans-retinoic Secondary analyses of the results of one randomized trial acid for the prevention of invasive cancer in humans. The of the use of 13-cis-retinoic acid as adjuvant therapy for results of one randomized, controlled trial indicate that cancers of the head and neck indicated a statistically topically applied all-trans-retinoic acid is effective in significant reduction in the incidence of second primary reversing moderate dysplasia of the uterine cervix (cervical tumours of the upper aerodigestive tract. A study of use of intra-epithelial neoplasia-II) but not against more severe 13-cis-retinoic acid at high doses and in a group at inherited dysplastic lesions. It was reported to be efficacious against high risk, with no controls, suggested that this compound is actinic keratosis of the skin in one randomized trial but not effective in preventing basal- and squamous-cell cancers of in another at a lower dose. Two further trials suggest that the skin. Two randomized controlled trials among patients topically applied all-trans-retinoic acid is effective against at lower risk involving lower doses of 13-cis-retinoic acid have shown no evidence of preventive efficacy.
The preventive efficacy of all-trans-retinoic acid was High doses of 13-cis-retinoic acid were shown to be evaluated in experimental models of skin, liver and mammary effective against oral leukoplakia in two randomized trials, gland carcinogenesis. The results of several experiments in one with controls receiving placebo and the other receiving mice indicated that all-trans-retinoic acid was effective b-carotene. One controlled trial showed no effect of 13-cis- against two-stage skin carcinogenesis when 7,12- retinoic acid in reducing cytological changes in the bronchi.
dimethylbenz[a]anthracene was used as the initiator, whereas Studies of molecular markers suggested that 13-cis-retinoic it enhanced skin carcinogenesis induced by this carcinogen acid increases expression of human retinoic acid receptor b, alone or with ultraviolet radiation. One study in mice but the relevance of these findings to cancer-preventive indicated that all-trans-retinoic acid enhanced N- Asian Pacific Journal of Cancer Prevention, Vol 1, 2000 197
A single intervention study showed a decrease in 3. 9-cis-Retinoic acid
micronucleus formation in cells of the buccal cavity inpatients, some of whom were smokers, who had been No data were available to the Working Group on putative treated with 13-cis-retinoic acid for 12 months.
cancer preventive activity in humans.
13-cis-Retinoic acid is a confirmed human teratogen. The The cancer preventive efficacy of 9-cis-retinoic acid was potential developmental toxicity associated with maternal evaluated in two animal studies on carcinogen-induced therapy with this retinoid depends on the dose, the stage of mammary carcinogenesis, one on prostate carcinogenesis and gestation, the duration of treatment and the route of one on colon carcinogenesis in rats. 9-cis-Retinoic acid prevented mammary and prostate tumors but not colon The preventive efficacy of 13-cis-retinoic acid has been tumors; however, it reduced the numbers of aberrant crypt evaluated in two-stage skin carcinogenesis models in mice and in urinary bladder carcinogenesis models in mice and In general, the in vitro effects of 9-cis-retinoic acid in vitro rats. 13-cis-Retinoic acid was effective in most studies with were similar to those of all-trans-retinoic acid, in that both both models. It was ineffective in models of tracheal, salivary inhibited cell proliferation and induced differentiation and gland, esophageal and renal carcinogenesis.
apoptosis in some cell lines; however, the 9-cis isomer was In vitro, 13-cis-retinoic acid inhibited proliferation in more potent than the all-trans isomer in several cell systems.
numerous cell lines. 13-cis-Retinoic acid inhibited growth 9-cis-retinoic acid caused growth inhibition in normal, in both monolayers of adherent cell cultures and in semi- immortalized and malignant cell lines, often but not always solid medium (anchorage-independent growth). 13-cis- in G or G . Induction of differentiation and apoptosis were Retinoic acid also induced cell differentiation in transformed seen in several types of cells. The cells that were sensitive to cells and triggered apoptosis in a few cell lines. In most cell 9-cis-retinoic acid responded to concentrations that are lines, the response to 13-cis-retinoic acid was similar to that achieved in plasma with standard pharmacological doses.
The ability of 9-cis-retinoic acid to inhibit carcinogen- The ability of 13-cis-retinoic acid to inhibit genetic and induced genotoxicity has not been studied in vitro or in vivo; related effects in cell cultures has been examined in a limited however, two studies suggest that it might reduce carcinogen number of studies, and these have yielded mixed results. In induced DNA damage by altering the activity of some two studies, a reduction in the frequency of chromosomal cytochrome P450 isozymes both in vitro and in vivo.
damage was seen in human lymphocytes exposed to radical- 9-cis-Retinoic acid suppresses cell proliferation and generating agents (bleomycin and X-irradiation) when they increases differentiation and apoptosis. The mechanisms by were pretreated with 13-cis-retinoic acid; in contrast, a third which proliferation is inhibited may involve antagonism of study showed an increase in the frequency of diepoxybutane- AP-1, decreased concentrations of cyclins, increased amounts induced sister chromatid exchanges and chromosomal of cyclin-dependent kinase inhibitor and inhibition of damage in human lymphocytes treated concurrently with the growth-stimulating signalling pathways. Induction of apoptosis and differentiation also appear to contribute to the Orally administered 13-cis-retinoic acid inhibited the putative cancer-preventive effect of 9-cis-retinoic acid.
induction of micronucleated cells in the bone marrow of The working group concluded that there is inadequate animals treated with benzo[a]pyrene and reduced the binding evidence that 9-cis-retinoic acid has cancer-preventive of this carcinogen to DNA in the liver, stomach and lung, activity in humans (IARC, 1999). There is limited evidence but not the kidney. Although the mechanism of this protective that 9-cis-retinoic acid has cancer-preventive activity in effect is unknown, it might be related to alterations in experimental animals (IARC, 1999). This evaluation was microsomal enzyme activity which has been shown to occur based on the observation of inhibitory effects in two studies in both liver and skin of mammals treated with 13-cis-retinoic of mammary carcinogenesis and one study of prostate carcinogenesis in rats. 9-cis-Retinoic acid is a teratogen in The working group concluded that there is limited evidence that 13-cis-retinoic acid has cancer-preventive activity inhumans (IARC, 1999). This evaluation was based on its 4. all-trans N-(4-hydroxyphenyl)retinamide
effectiveness against oral leukoplakia, and preliminaryevidence for prevention of second primary cancers of the In a preliminary report of a large randomized trial of use aerodigestive tract. There is also limited evidence that 13- of all-trans N-(4-hydroxyphenyl)retinamide, equivocal cis-retinoic acid has cancer-preventive activity in results were obtained with regard to the development of new experimental animals (IARC, 1999). This evaluation was contralateral tumors among women previously treated for based on the observation of inhibitory effects in most but early breast cancer. There were fewer new cancers among not all studies with models of skin and urinary bladder treated pre-menopausal women but more cancers among treated post-menopausal women. A decrease in the risk for However, 13-cis-retinoic acid has a relatively low ovarian cancer was reported among all treated women in this therapeutic ratio of efficacy to toxicity, and is an established Two studies, only one of which was randomized, of 198 Asian Pacific Journal of Cancer Prevention, Vol 1, 2000
An International Evaluation of the Cancer-Prevention intermediate end-points, suggested an effect of all-trans N- experimental animals (IARC, 1999). This evaluation was (4-hydroxyphenyl)retinamide against oral leukoplakia. A based on the observation of inhibitory effects in models of possible effect on ploidy in urinary bladder cells has also mammary carcinogenesis in mice and rats and its effectiveness in a limited number of studies against prostate The chemopreventive efficacy of all-trans N-(4- and colon carcinogenesis and lymphomagenesis. all-trans hydroxyphenyl)retinamide has been evaluated in animal N-(4-Hydroxyphenyl)retinamide does not have significant models of mammary gland, prostate, lung, skin, urinary toxicity in humans with the dose schedule normally used.
bladder, and colon carcinogenesis and lymphomagenesis. It Therefore, all-trans N-(4-hydroxyphenyl)retinamide shows was effective in reducing tumour incidence or multiplicity promise as a cancer preventive agent in humans.
in 11 of 12 studies of mammary carcinogenesis in mice orrats. The results of such studies depend critically on the 5. Etretinate
experimental conditions, including the strain and age of theanimals, their diet and the dose of both carcinogen and Etretinate was evaluated in six randomized trials for retinoid. It was effective in one study in a model of urinary efficacy in preventing the recurrence of superficial tumours bladder carcinogenesis in mice and ineffective in another of the urinary bladder. None showed unequivocal evidence and effective in one study of prostate carcinogenesis but not of an effect of treatment; efficacy was suggested in analyses in two others. It was ineffective in one study of lung of some end-points. Etretinate was not effective in preventing carcinogenesis in mice. It was effective in one study of second primary tumours in subjects with head-and-neck carcinogenesis of the colon and in two studies of cancer when compared with those given placebo. In two lymphomagenesis in mice. In one skin carcinogenesis study reports of the same study without a separate control group, in mice, all-trans N-(4-hydroxyphenyl)retinamide was etretinate was reported to reduce an index of metaplasia in ineffective, or enhanced skin tumor development.
bronchial biopsy samples from heavy smokers. In a In-vitro studies suggest that all-trans N-(4- randomized trial involving 150 subjects, however, etretinate hydroxyphenyl)retinamide can affect carcinogenesis at showed no efficacy in reducing atypia in sputum samples several levels: it inhibited the transformation of cultured cells when compared with placebo. In one study with no controls and of tissue in organ culture; it inhibited the proliferation of in which etretinate was given orally at a high dose or orally a variety of tumour cell lines; and it induced apoptosis but it at a moderate dose plus topical application as a paste, regression of leukoplakia of the mouth was reported, more There are insufficient data to conclude whether all-trans notably when topical application was added. In one study N-(4-hydroxyphenyl)retinamide can reduce the genotoxic with no controls, oral treatment with etretinate appeared to effects of carcinogens in vitro or in vivo. Indications that it reduce the severity of actinic keratotic and keratocanthoma alters the metabolism of carcinogens and thus may affect lesions of the skin. In two double-blind cross-over trials DNA damage are provided by a study showing alterations to involving patients with actinic keratosis, improvement in cytochrome p450 mRNA levels in cell cultures exposed to terms of the number and size of lesions was reported in the retinoid, and a study in which phase I and phase II patients treated orally with etretinate.
enzymes were shown to be altered in the liver of animals fed The cancer-preventive efficacy of etretinate has been this compound. The altered metabolism was associated in assessed in mouse, rat and rabbit models of carcinogenesis vivo with a reduction in the binding to tissue DNA of a and in relation to virus-induced tumours. It was ineffective carcinogen known to be metabolized by these enzymes.
in inhibiting UV-induced skin carcinogenesis in mice, but Few reports indicate any activity of all-trans N-(4- inhibited chemically-induced skin tumours in mice and hydroxyphenyl)retinamide at the initiation stage of rabbits. In one study in mice, etretinate reduced the size of carcinogenesis and most suggest it acts on tumour promotion.
skin papillomas. It was effective in various models of The mechanisms that may account for the cancer-preventive digestive tract carcinogenesis in mice and rats. In single effects of this retinoid appear to be associated with its ability studies, etretinate was ineffective in preventing either to inhibit cell proliferation by increasing the amount of a leukaemia or lung tumours but was effective in preventing cyclin-dependent kinase inhibitor and by down-regulating urinary bladder carcinogenesis in rats. It was effective in a cyclin D1 and inducing apoptosis which has been extensively model of benign tumours induced in mice by Shope papilloma studied. Its limited effects on differentiation raise doubts as virus and in models of malignant tumours in hamsters and to whether this is a mechanism for cancer prevention. The chickens induced by Rous sarcoma virus. In some high concentrations required to induce apoptosis in vitro experimental models, etretinate enhanced the tumorigenic restrict extrapolation of these studies to infer relevance to Etretinate has been shown to modify differentiation in The working group concluded that there is inadequate several models in vitro: in tracheas of hamsters, squamous evidence that all-trans-N-(4-hydroxyphenyl)retinamide has metaplasia induced by vitamin A deficiency was reversed.
cancer preventive activity in humans (IARC, 1999).
In respiratory tracts exposed to carcinogens, etretinate However, there is sufficient evidence that all-trans N-(4- inhibited loss of mucus secretion and ciliary action. In many hydroxyphenyl)retinamide has cancer preventive activity in studies with human and animal keratinocytes, etretinate Asian Pacific Journal of Cancer Prevention, Vol 1, 2000 199
causes changes in differentiation similar to those seen after may be associated with modifications in the pattern of keratin treatment with all-trans-retinoic acid. In contrast to all-trans- expression and membrane glycosylation. The only retinoic acid, etretinate did not induce differentiation in mechanism that has been studied in relation to the anti- promyelocytic leukemic cell lines. Proliferation was inhibited proliferative activity of acitretin is inhibition of ornithine in murine and human melanoma cell lines, in lymphoblastoid decarboxylase activity, which is more marked in lines and in normal keratinocytes. Because of differences in hyperproliferative states. The effects of acitretin on immune the experimental protocols, it is not clear whether etretinate function have not been studied extensively. It stimulated the is selectively active against tumour cells. In all cases, it was production of interleukin-1 both in vitro and in vivo, which less active than all-trans-retinoic acid. Etretinate has been might result in activation of lymphoid cells. An effect of studied in many in-vitro models of immune function, but no acitretin, which might contribute to cancer-preventive activity, is inhibition of angiogenesis which has been There have been no detailed studies of the mechanism of action of etretinate. Its ability to inhibit the induction of The working group concluded that there is inadequate ornithine decarboxylase in keratinocytes after treatment evidence that acitretin has cancer-preventive activity in with phytohaemagglutinin suggests that, like all-trans- humans and there is inadequate evidence that acitretin has retinoic acid, it acts in the promotional phase of cancer-preventive activity in experimental animals (IARC, 1999). However, acitretin is a derivative of etretinate, and The working group concluded that there is inadequate therefore probably has similar cancer preventive efficacy to evidence that etretinate has cancer preventive activity in etretinate. Further, its toxicity is less than etretinate, though humans (IARC, 1999). There is limited evidence that it is a potent teratogen in experimental animals.
etretinate has cancer preventive activity in experimentalanimals (IARC, 1999). This evaluation was based on the 7. all-trans-N-Ethylretinamide
observation of inhibitory effects in studies with models ofskin cancer, and in single studies with models of digestive No data were available to the Working Group on the tract and urinary bladder carcinogenesis and in three putative cancer preventive activity of all-trans-N- models of virus-induced tumors. However, etretinate is toxic, and a human teratogen, and in most countries is no The cancer preventive efficacy of all-trans-N- ethylretinamide has been evaluated in models of respiratorytract and pancreas carcinogenesis in hamsters, of liver 6. Acitretin
carcinogenesis, in mice, of urinary bladder carcinogenesisin mice and rats and of colon carcinogenesis in rats. Tumour The active form of etretinate is acitretin. In one trial with incidence was enhanced in the trachea and pancreas of 44 renal transplant patients, acitretin reduced the frequency hamsters, in the liver in mice, and in one study, in the urinary of occurrence of squamous-cell cancers of the skin when bladder in rats. all-trans-N-Ethylretinamide had cancer compared with placebo. In the same trial, the prevalence of preventive effects in some studies of urinary bladder keratotic skin lesions was also reduced by acitretin. When carcinogenesis in mice and rats but was ineffective in models treatment was stopped, the numbers of cancers and keratotic of colon carcinogenesis. all-trans-N-Ethylretinamide inhibited carcinogen-induced neoplastic transformation at In single experimental studies, acitretin reduced the concentrations similar to those at which all-trans-retinoic acid incidence of spontaneous and chemically induced liver had this effect. In the hamster trachea it was less potent than tumors in mice and rats in conjunction with reductions in all-trans-retinoic acid in reversing squamous metaplasia; when tested in chick skin for an equivalent endpoint, its Acitretin has been tested for its ability to inhibit activity was similar to all-trans-retinoic acid.
proliferation or to induce differentiation of tumour and normal The working group concluded that there is inadequate cells in vitro. Acitretin was more active than etretinate, and evidence that all-trans-N-ethylretinamide has cancer both were less active than all-trans-retinoic acid and 13-cis- preventive activity in humans, and there is evidence retinoic acid. Acitretin had an anti-proliferative effect on some suggesting lack of cancer preventive activity in experimental but not all tumour cell lines that were tested. Several studies animals (IARC, 1999). all-trans-N-Ethylretinamide was with epidermal cells showed that the effects of acitretin never approved for use in humans and is no longer being depended on the culture conditions and/ or the proliferation rate. It did not induce differentiation of leukemic cells invitro. In normal epidermal cells, it decreased cornified 8. Targretin
envelope formation and modified the pattern of keratin. Instudies of lymphocyte proliferation in vitro, the effects No data were available to the Working Group on the depended on the concentration of acitretin and on the mitogen putative cancer preventive activity of targretin in humans.
In a single study of three months duration, targretin was The differentiating effects of acitretin on epidermal cells effective in preventing mammary cancer induced by N- 200 Asian Pacific Journal of Cancer Prevention, Vol 1, 2000
An International Evaluation of the Cancer-Prevention methyl-N-nitrosourea in rats. In two models of differentiation biochemical and molecular mechanisms of retinoid action, in human cells in vitro, targretin, which preferentially binds raise the hope that new synthetic retinoids can be synthesized to the RXRs, was less active than ligands binding to retinoic that may be useful in chemoprevention of cancer. Such acid receptors; however, in both models supra-additive compounds may target specific molecules, and may result in activity was seen when the cells were treated simultaneously far fewer unwanted side-effects than those in use up till now.
with targretin and the ligands binding to retinoic acid In this regard the new retinoid receptor subtype specific The working group concluded that there is inadequateevidence that targretin has cancer-preventive activity in Acknowledgement
humans and inadequate evidence that targretin has cancer-preventive activity in experimental animals (IARC, 1999).
The Foundation for Promotion of Cancer Research, Japan,is gratefully acknowledged for its generous support of the 9. LGD 1550
meeting of the Working Group, and the production of Volume4 of the IARC Handbooks of Cancer Prevention.
No data were available to the Working Group on thepotential cancer preventive activity of LGD 1550 in humans References
or in experimental animals. LGD 1550 inhibited proliferationin human breast cancer cells expressing retinoic acid receptor- IARC (1997). IARC handbooks of cancer prevention Volume 1: a, but not in cells that did not express this receptor.
Non-steroidal anti inflammatory drugs. Lyon, International The working group concluded that there is inadequate Agency for Research on Cancer, 202 pp.
evidence that LGD 1550 has cancer-preventive activity in IARC (1998a). IARC handbooks of cancer prevention Volume 2: humans and in experimental animals (IARC, 1999).
Carotenoids. Lyon, International Agency for Research on Cancer, 326 pp.
Discussion
IARC (1998b). IARC handbooks of cancer prevention Volume 3: Vitamin A. Lyon, International Agency for Research on Cancer, 261 pp.
Of the retinoids examined, none were considered by the IARC (1999). IARC handbooks of cancer prevention Volume 4: Working Group as exhibiting sufficient evidence of cancer Retinoids. Lyon, International Agency for Research on Cancer, preventive activity in humans (IARC, 1999). Working groups have so far concluded that for chemoprevention suchcategorization requires findings from randomized controlled Notes: A.B.Miller was Senior Epidemiologist, International trials with malignant disease as an endpoint, as was the case Agency for Research on Cancer, Lyon and Acting Chief, Unit for aspirin in the first evaluation (IARC, 1997). Though such of Chemoprevention, at the time of the working group an outcome is theoretically possible, the limitations of meeting reported in this paper; P. Nettesheim was visiting organizing such trials are now widely appreciated. It seems Scientist in the Unit of Chemoprevention during the same likely that reliance will increasingly be placed on time period, and B.W. Stewart was Chair of the Working ‘intermediate effect’ biomarkers for the purpose of comparing retinoids or similar agents thought to have chemopreventiveactivity.
Appendix: List of Participants. Retinoids Meeting One of the compounds considered, N-hydro-xyphenylretinamide, is characterized by a considerable body J.S. Bertram, Cancer Research Center of Hawaii, USA; W.S.
of essentially positive experimental findings but relatively Blaner, Columbia University, New York, USA; R.A.
little human data. The working group considered that the Chandraratna, Allergan, Irvine, CA, USA; C. Chomienne, experimental findings provide a positive encouragement to Hôpital Saint Louis, Paris, France; J.A. Crowell, National examining chemopreventive activity of this compound in Cancer Institute, Bethesda MD, USA; M.I. Dawson, humans (IARC, 1999). Nonetheless, this scenario highlighted Molecular Medicine Research Institute, Mountain View, CA, a specific aspect of such experimental studies. By comparison USA; F. Formelli, Istituto Nazionale per lo Studio e la Cura with the massive amount of data which underpins dei Tumori, Milan, Italy; E.R. Greenberg (Vice-Chairman), extrapolation of carcinogenesis findings from experimental Dartmouth-Hitchcock Medical Center, Lebanon, NH, USA animals to humans, the amount of analogous data in relation ; H. Gronemeyer, Institut de Génétique et de Biologie Moléculaire et Cellulaire, Illkirch, France; D.L. Hill, The biology of retinoid action, specifically in relation to University of Alabama at Birmingham, AL, USA; R.J.
RARs and RXRs, has been subject to intense molecular Kavlock, Environmental Protection Agency, NC, USA; R.
analysis. However, despite the advances that have been made Lotan, M.D. Anderson Cancer Center, Houston, TX, USA; in this field, there were few instances when such ‘mechanistic’ M. Maden, King’s College, London, England; R.G. Mehta, data significantly influenced evaluation of cancer University of Illinois at Chicago, IL; T.E. Moon, Ligand chemopreventive activity exhibited by individual retinoids.
Pharmaceuticals Inc., San Diego, CA, USA; H. Nau, School Nevertheless, recent advances in the understanding of the of Veterinary Medicine Hannover, Germany; P. Nettesheim, Asian Pacific Journal of Cancer Prevention, Vol 1, 2000 201
National Institute of Environmental Health Sciences, NC,USA; J.A. Olson, Iowa State University, Ames, IA, USA;M.P. Rosin, British Columbia Cancer Agency, Vancouver,BC, Canada; B.W. Stewart (Chairman), Eastern Sydney AreaHealth Service, Randwick, NSW, Australia; C.C. Willhite,State of California, Berkeley, CA, USA; R.A. Woutersen,TNO-Nutrition and Food Research Institute, AJ Zeist, TheNetherlands.
Personal Profile: Anthony Miller
Dr Miller qualified in Medicine from the University ofCambridge, England in 1955 and subsequently specializedin Internal Medicine. He became a Member of the RoyalCollege of Physicians of London, England in 1964, and waselected a Fellow in 1987. In 1973 he became a Fellow of theRoyal College of Physicians of Canada, in 1977 a Fellow ofthe Faculty of Public Health Medicine of the United Kingdomand in 1985 a Fellow of the American College ofEpidemiology.
In 1962, he joined the British Medical Research Councilas a member of the scientific staff of the Tuberculosis andChest Diseases Unit. While with the Medical ResearchCouncil he was responsible for clinical trials andepidemiological studies in lung cancer and pulmonarytuberculosis.
In April 1971, he went to Canada to set up the EpidemiologyUnit of the National Cancer Institute of Canada. In 1986 hebecame full-time in the Department of Preventive Medicineand Biostatistics of the University of Toronto where he wasDirector of the MSc/PHd Specialization in Epidemiologyfrom 1986-91 and Chairman of the Department 1992-96. In1996 he became Emeritus, and since then has served as aspecial expert in the Early Detection Branch, US NationalCancer Institute and Senior Epidemiologist in theInternational Agency for Research on Cancer in Lyon. Hewas acting Chief of Chemoprevention in the Agency in theyears 1998-99. He was seconded to the Deutsches Krebsforschungszentrum in 1999 and is currently head of the Division of Clinical Epidemiology. His current researchprogramme continues his interests in diet and cancer, screening for cancer, and the application of knowledge to cancercontrol.
Dr Miller has served on numerous scientific advisory committees in Canada and the United States as well as Internationally.
He has over 300 scientific publications to his credit.
202 Asian Pacific Journal of Cancer Prevention, Vol 1, 2000

Source: http://www.apocpcontrol.net/paper_file/issue_abs/Volume1_No3/Anthony%20Miller.pdf