Journal of the Royal Society of Medicine Volume 72 November 1979
797
Editorials
-~ Hormone replacement therapy Not long ago it was fashionable to argue whether treatment of postmenopausal women with oestrogens could cause cancer of the uterus. The
correspondence columns of the New England Journal of Medicine, the Lancet and the British Medical Journal reverberated with letters from diverse workers who criticized any work (or workers) that attempted to show that the association of endometrial cancer and unopposed oestrogen therapy was anything more than fortuitous. As the tempo of debate increased, even writers of editorials on the subject became objects of criticism. Fortunately the debate can now cease - the data are in, they are hard and they make sense. Eight papers published in the last four years have shown that treatment of postmenopausal women with unopposed oestrogens is associated with an increased risk of developing cancer of the uterus, and that the risk is related to the dose and the duration of treatment (for references see Hammond et al. 1979, Jick et al. 1979). The several studies have produced remarkably similar risk ratios, risk ratios that have survived independent re-evaluation of the original histopathological diagnoses (Gordon et al. 1977). These estimates of the carcinogenic risk of oestrogen treatment were based on retrospective studies performed in the United States. Prospective clinical trials in this country have shown that the incidence of endometrial hyperplasia (the most severe form of which is considered to be a precursor of endometrial cancer) is directly related to the dose and duration of unopposed oestrogen therapy (Whitehead, McQueen et al. 1978). The data also make sense as we are beginning to develop an understanding of what underlies these events at the molecular level. While there is no room here for an extended discussion of the mechanism(s) of action of oestrogen, suffice it to say that ultimately the hormone acts by entering the cell nucleus and attaching itself on to the DNA and causing it to direct the synthesis of certain specific proteins. The synthesis of these proteins results in the changes we associate with the action of oestrogen. To enter the nucleus, oestrogens must first be bound and then transported by a specific protein that is found in the cytoplasm, the 0141-0768/79/1 10797-02/$01.00/0
~~~~~~~ jsh
oestrogen receptor protein. The amount of oestrogen that can enter the cell nucleus depends on the number of oestrogen receptors, their affinity for the various oestrogens and also on the stability of the complex formed with the particular oestrogen. Oestradiol 17,B, the oestrogen secreted by the ovarian follicles of premenopausal women, is the most potent; oestrone is a weak oestrogen because it has a low affinity for the oestrogen receptor protein. Oestriol is even weaker because the complex it forms with the protein is retained in the nucleus for so short a time that the response it provokes is quite evanescent (Anderson et al.
1975). One of the advances in this field has been the realization that interconversion of steroids, and particularly of oestrone and oestradiol, occurs in several tissues. For instance, as described on page 839 by Hutton and his colleagues, when oestradiol is ingested the gut itself oxidizes it to oestrone, so plasma concentrations of oestrone and oestradiol are similar whether the women are treated with a preparation based on oestrone or on oestradiol. Curiously the vaginal epithelium lacks the necessary enzymes to effect this conversion and so when oestradiol is put directly into the vagina, plasma oestradiol concentrations are much higher than they are when it is given by mouth (Rigg et al. 1978). This apparently obscure corner of biochemistry has important clinical consequences because it means that contraindications to systemic oestrogen therapy are now contraindications to local therapy too (Whitehead, Minardi et al. 1978). Oestradiol can also be converted in endometrial cells to the less potent oestrogen, oestrone. The rate of conversion is increased by progesterone, a hormone which, inter alia, induces activity of the enzyme that is responsible - oestradiol 171 dehydrogenase (Gurpide et al. 1976). In premenopausal women, activity of this enzyme is low in the follicular phase of the cycle, so the uterine endometrium grows at this stage (the proliferative phase). After ovulation the activity of the enzyme is stimulated by the progesterone secreted by the corpus luteum. Despite high blood levels of oestradiol during the luteal phase, the main oestrogen within the endometrial cell is therefore oestrone and, as this is only a weak oestrogen, continued proliferation of the endometrium does not occur. If progesterone is not secreted, as occurs © 1979 The Royal Society of Medicine
798
Journal of the Royal Society of Medicine Volume 72 November 1979
during anovular menstrual cycles, endometrial hyperplasia may develop. In women with persistently anovular cycles, for instance those with the polycystic ovary syndrome, symptomatic endometrial hyperplasia and eventually endometrial cancer become significant risks (Lucas 1974). It is likely that, in addition to stimulation of oestradiol 17P dehydrogenase, progesterone directly inhibits DNA directed synthesis of oestrogen receptor proteins. Progesterone thus has two actions that limit the impact of oestrogen on the endometrium. It is a reasonable prediction from these studies of molecular biology that the addition of progesterone to treatment with oestrogen of postmenopausal women will reduce the incidence of endometrial hyperplasia and cancer. Studies from the United Kingdom (Thom et al. 1979) and from the United States (Hammond et al. 1979) bear out these predictions. The results of Thom and her colleagues show that the risks of oestrogen to the uterus can be abolished by concurrent treatment with progestogens. The smallest dose of oestrogen that will abolish climacteric symptoms, given cyclically, with the addition of a progestogen for 7 to 10 days of the cycle, is now the best way of administering postmenopausal oestrogen therapy. What problems remain? Well, the obvious one is the anxiety that having abolished the risk of oestrogen to the uterus, we will replace it with the systemic risks of progestogens. Careful metabolic surveillance is therefore necessary. Should natural or synthetic oestrogens be used? At present no one can be certain because in those studies in which comparisons were made the doses, in terms of oestrogenic potency, were not comparable. Who should receive prolonged oestrogen therapy to avoid postmenopausal osteoporosis? Presumably the answer is, those who lose bone fastest. It is not yet certain, however, how to identify this group on a large enough scale to make selection a practicable
possibility. There is no doubt, however, that the greatest innovation that can be anticipated in the next few years will be the development of a nonsteroidal, nonoestrogenic treatment for postmenopausal flushing attacks. Most women develop this symptom, indeed it often comes on before the menopause and its course is quite unpredictable. Some women continue to have attacks well into their seventies. As Hutton et al. show (p 838), flushes do not seem to be related to specific hormone concentrations. The development of a simple treatment that does not involve giving oestrogens would represent a gigantic breakthrough for many climacteric women.
H S Jacobs Reader in Gynaecological Endocrinology St Mary's Hospital Medical School London W2 IPG 0141-0768/79/1 10798-03/$01.00/0
References Anderson J N, Peck E J & Clark J H (1975) Endocrinology 96, 160 Gordon J, Reagan J W, Finkle W D & Ziel H K (1977) New England Journal of Medicine 297, 570 Gurpide E, Gusberg S P & Tseng L (1976) Journal of Steroid Biochemistry 7, 891 Hammond C B, Jelovsek F R, Lee K L, Creasman W T & Parker R T (1979) American Journal of Obstetrics and Gynecology 133, 537 Jick H, Watkins R N, Hunter J R, Dinan B J, Madsen S, Rothman K J & Walker A M (1979) New England Journal ofMedicine 300, 218 Lucas W E (1974) Obstetrical and Gynecological Survey 29, 507 Rigg L A, Hermann H & Yen S S C (1978) New England Journal ofMedicine 298, 195 Thom M, White P, WilLiams R M, Sturdee D W, Paterson M E L, Wade-Evans T & Studd J W W (1979) Lancet ii, 455 Whitebead M I, McQueen J, Minardi J & Campbell S (1978) In: The Role of Estrogen/Progestogen in the Management of the Menopause. Ed. I D Cooke. MTP Press, Lancaster; pp 121-133 Whitehead M I, Minardi J, Kitcbin Y & Sharples M J (1978) In: The Role of Estrogen/Progestogen in the Management of the Menopause. Ed. I D Cooke. MTP Press, Lancaster; pp 63-71
Central venous feeding Weight loss to the layman is one of the most obvious signs of chronic disease and it is curious that it has needed the challenge of the demanding, expensive and potentially dangerous technique of intravenous feeding to reawaken most clinicians' interest in it. Malnutrition appears to be common on hospital wards even in developed countries (Bistrian et al. 1976, Hill et al. 1977) and can usually be effectively countered cheaply and simply. Patients should be tempted to eat by serving them appetizing food, food supplements can be provided between meals or liquid diets can be infused intestinally through soft fine bore tubes. We should treat malnutrition because, as well as specific deficiency syndromes, it results in poor wound healing, diminished immunity and a weak apathetic immobile patient. There can no longer be any doubt that intravenous feeding is effective in maintaining or improving nutrition (Dudrick et al. 1968, 1969) and patients can be fed satisfactorily by this route for periods up to at least five years (Jeejeebhoy et al. 1976). Effective feeding, particularly in the long term, depends upon providing all the patient's nutrient requirements, since diets deficient in individual nutrients (phosphate or potassium, for example) may be poorly utilized (Rudman et al. 1974). The amino acid compositions of most of the ©) 1979 The Royal Society of Medicine
797
Editorials
-~ Hormone replacement therapy Not long ago it was fashionable to argue whether treatment of postmenopausal women with oestrogens could cause cancer of the uterus. The
correspondence columns of the New England Journal of Medicine, the Lancet and the British Medical Journal reverberated with letters from diverse workers who criticized any work (or workers) that attempted to show that the association of endometrial cancer and unopposed oestrogen therapy was anything more than fortuitous. As the tempo of debate increased, even writers of editorials on the subject became objects of criticism. Fortunately the debate can now cease - the data are in, they are hard and they make sense. Eight papers published in the last four years have shown that treatment of postmenopausal women with unopposed oestrogens is associated with an increased risk of developing cancer of the uterus, and that the risk is related to the dose and the duration of treatment (for references see Hammond et al. 1979, Jick et al. 1979). The several studies have produced remarkably similar risk ratios, risk ratios that have survived independent re-evaluation of the original histopathological diagnoses (Gordon et al. 1977). These estimates of the carcinogenic risk of oestrogen treatment were based on retrospective studies performed in the United States. Prospective clinical trials in this country have shown that the incidence of endometrial hyperplasia (the most severe form of which is considered to be a precursor of endometrial cancer) is directly related to the dose and duration of unopposed oestrogen therapy (Whitehead, McQueen et al. 1978). The data also make sense as we are beginning to develop an understanding of what underlies these events at the molecular level. While there is no room here for an extended discussion of the mechanism(s) of action of oestrogen, suffice it to say that ultimately the hormone acts by entering the cell nucleus and attaching itself on to the DNA and causing it to direct the synthesis of certain specific proteins. The synthesis of these proteins results in the changes we associate with the action of oestrogen. To enter the nucleus, oestrogens must first be bound and then transported by a specific protein that is found in the cytoplasm, the 0141-0768/79/1 10797-02/$01.00/0
~~~~~~~ jsh
oestrogen receptor protein. The amount of oestrogen that can enter the cell nucleus depends on the number of oestrogen receptors, their affinity for the various oestrogens and also on the stability of the complex formed with the particular oestrogen. Oestradiol 17,B, the oestrogen secreted by the ovarian follicles of premenopausal women, is the most potent; oestrone is a weak oestrogen because it has a low affinity for the oestrogen receptor protein. Oestriol is even weaker because the complex it forms with the protein is retained in the nucleus for so short a time that the response it provokes is quite evanescent (Anderson et al.
1975). One of the advances in this field has been the realization that interconversion of steroids, and particularly of oestrone and oestradiol, occurs in several tissues. For instance, as described on page 839 by Hutton and his colleagues, when oestradiol is ingested the gut itself oxidizes it to oestrone, so plasma concentrations of oestrone and oestradiol are similar whether the women are treated with a preparation based on oestrone or on oestradiol. Curiously the vaginal epithelium lacks the necessary enzymes to effect this conversion and so when oestradiol is put directly into the vagina, plasma oestradiol concentrations are much higher than they are when it is given by mouth (Rigg et al. 1978). This apparently obscure corner of biochemistry has important clinical consequences because it means that contraindications to systemic oestrogen therapy are now contraindications to local therapy too (Whitehead, Minardi et al. 1978). Oestradiol can also be converted in endometrial cells to the less potent oestrogen, oestrone. The rate of conversion is increased by progesterone, a hormone which, inter alia, induces activity of the enzyme that is responsible - oestradiol 171 dehydrogenase (Gurpide et al. 1976). In premenopausal women, activity of this enzyme is low in the follicular phase of the cycle, so the uterine endometrium grows at this stage (the proliferative phase). After ovulation the activity of the enzyme is stimulated by the progesterone secreted by the corpus luteum. Despite high blood levels of oestradiol during the luteal phase, the main oestrogen within the endometrial cell is therefore oestrone and, as this is only a weak oestrogen, continued proliferation of the endometrium does not occur. If progesterone is not secreted, as occurs © 1979 The Royal Society of Medicine
798
Journal of the Royal Society of Medicine Volume 72 November 1979
during anovular menstrual cycles, endometrial hyperplasia may develop. In women with persistently anovular cycles, for instance those with the polycystic ovary syndrome, symptomatic endometrial hyperplasia and eventually endometrial cancer become significant risks (Lucas 1974). It is likely that, in addition to stimulation of oestradiol 17P dehydrogenase, progesterone directly inhibits DNA directed synthesis of oestrogen receptor proteins. Progesterone thus has two actions that limit the impact of oestrogen on the endometrium. It is a reasonable prediction from these studies of molecular biology that the addition of progesterone to treatment with oestrogen of postmenopausal women will reduce the incidence of endometrial hyperplasia and cancer. Studies from the United Kingdom (Thom et al. 1979) and from the United States (Hammond et al. 1979) bear out these predictions. The results of Thom and her colleagues show that the risks of oestrogen to the uterus can be abolished by concurrent treatment with progestogens. The smallest dose of oestrogen that will abolish climacteric symptoms, given cyclically, with the addition of a progestogen for 7 to 10 days of the cycle, is now the best way of administering postmenopausal oestrogen therapy. What problems remain? Well, the obvious one is the anxiety that having abolished the risk of oestrogen to the uterus, we will replace it with the systemic risks of progestogens. Careful metabolic surveillance is therefore necessary. Should natural or synthetic oestrogens be used? At present no one can be certain because in those studies in which comparisons were made the doses, in terms of oestrogenic potency, were not comparable. Who should receive prolonged oestrogen therapy to avoid postmenopausal osteoporosis? Presumably the answer is, those who lose bone fastest. It is not yet certain, however, how to identify this group on a large enough scale to make selection a practicable
possibility. There is no doubt, however, that the greatest innovation that can be anticipated in the next few years will be the development of a nonsteroidal, nonoestrogenic treatment for postmenopausal flushing attacks. Most women develop this symptom, indeed it often comes on before the menopause and its course is quite unpredictable. Some women continue to have attacks well into their seventies. As Hutton et al. show (p 838), flushes do not seem to be related to specific hormone concentrations. The development of a simple treatment that does not involve giving oestrogens would represent a gigantic breakthrough for many climacteric women.
H S Jacobs Reader in Gynaecological Endocrinology St Mary's Hospital Medical School London W2 IPG 0141-0768/79/1 10798-03/$01.00/0
References Anderson J N, Peck E J & Clark J H (1975) Endocrinology 96, 160 Gordon J, Reagan J W, Finkle W D & Ziel H K (1977) New England Journal of Medicine 297, 570 Gurpide E, Gusberg S P & Tseng L (1976) Journal of Steroid Biochemistry 7, 891 Hammond C B, Jelovsek F R, Lee K L, Creasman W T & Parker R T (1979) American Journal of Obstetrics and Gynecology 133, 537 Jick H, Watkins R N, Hunter J R, Dinan B J, Madsen S, Rothman K J & Walker A M (1979) New England Journal ofMedicine 300, 218 Lucas W E (1974) Obstetrical and Gynecological Survey 29, 507 Rigg L A, Hermann H & Yen S S C (1978) New England Journal ofMedicine 298, 195 Thom M, White P, WilLiams R M, Sturdee D W, Paterson M E L, Wade-Evans T & Studd J W W (1979) Lancet ii, 455 Whitebead M I, McQueen J, Minardi J & Campbell S (1978) In: The Role of Estrogen/Progestogen in the Management of the Menopause. Ed. I D Cooke. MTP Press, Lancaster; pp 121-133 Whitehead M I, Minardi J, Kitcbin Y & Sharples M J (1978) In: The Role of Estrogen/Progestogen in the Management of the Menopause. Ed. I D Cooke. MTP Press, Lancaster; pp 63-71
Central venous feeding Weight loss to the layman is one of the most obvious signs of chronic disease and it is curious that it has needed the challenge of the demanding, expensive and potentially dangerous technique of intravenous feeding to reawaken most clinicians' interest in it. Malnutrition appears to be common on hospital wards even in developed countries (Bistrian et al. 1976, Hill et al. 1977) and can usually be effectively countered cheaply and simply. Patients should be tempted to eat by serving them appetizing food, food supplements can be provided between meals or liquid diets can be infused intestinally through soft fine bore tubes. We should treat malnutrition because, as well as specific deficiency syndromes, it results in poor wound healing, diminished immunity and a weak apathetic immobile patient. There can no longer be any doubt that intravenous feeding is effective in maintaining or improving nutrition (Dudrick et al. 1968, 1969) and patients can be fed satisfactorily by this route for periods up to at least five years (Jeejeebhoy et al. 1976). Effective feeding, particularly in the long term, depends upon providing all the patient's nutrient requirements, since diets deficient in individual nutrients (phosphate or potassium, for example) may be poorly utilized (Rudman et al. 1974). The amino acid compositions of most of the ©) 1979 The Royal Society of Medicine
