Kidney International, Vol. 63, Supplement 85 (2003), pp. S62–S65

Selective estrogen receptor modulators in chronic renal failure

JOSE R. WEISINGER, ITA PFEFERMAN HEILBERG, EDDY HERNANDEZ, RAUL CARLINI,and EZEQUIEL BELLORIN-FONT

Division of Nephrology, Hospital Universitario de Caracas, Universidad Central de Venezuela, Caracas, Venezuela; andDivision of Nephrology, Universidade Federal de Sao Paulo, Sao Paulo, Brazil

Selective estrogen receptor modulators in chronic renal failure. (BMD) and higher bone resorption markers, comparedBackground. In addition to renal osteodystrophy, post- to similar women with normal menstruation [2]. At the

menopausal women on dialysis could be at risk of osteoporosis. same time, only a few postmenopausal uremic womenHormone replacement therapy (HRT) could have beneficialare being offered hormone replacement therapy (HRT),effects as well as potentially serious risks, especially in uremicindicating that these patients are managed differentlywomen, due to the pharmacokinetics of estradiol in renal fail-

ure. Therapeutic alternatives, such as the selective estrogen than nonuremic women. Apart from the lack of aware-receptor modulators (SERMs), have shown the benefits of ness of the treating physicians, women with end-stageestrogen on bone and serum lipid levels, without its adverse renal failure are usually stoic and rarely complain abouteffects on the breast and endometrium, in nonuremic women.

gynecologic problems that may seem trivial in compari-Methods. Recent data on the effect of the SERM raloxifeneson to their renal disease burden [3–6].in bone and lipid metabolism in osteoporotic postmenopausal

women on dialysis is reviewed. Since the estrogen receptor The adverse effects of hypoestrogenism on bone have(ER) gene has been suggested as a candidate marker for osteo- been widely recognized. Lack of estrogens is associatedporosis, we investigated whether ER polymorphism could have

with increased osteoclastic bone resorption [7–10]. Inpredicted the BMD response to raloxifene.addition, estrogen may affect parathyroid hormone ac-Results. Hemodialyzed women on raloxifene demonstrated

increased trabecular bone mineral density (BMD) and de- tion by increasing PTH mRNA at the parathyroid glandcreased bone resorption markers. Similarly, LDL-cholesterol [11] and modulating PTH action on bone. Clinical evi-values dropped significantly. ER gene polymorphism analysis dence in favor of an effect of estrogen in uremic boneof baseline BMD parameters did not differ between PP/xx or

disease has been recently suggested in a controlled studyPp/Xx groups. Nevertheless, patients on raloxifene with PP/xxof a small group of postmenopausal women on dialysis,genotypes, but not those with Pp/Xx, showed a higher trabecu-

lar BMD after one year on treatment, suggesting that homozy- in which treatment with transdermal estradiol and thegous women for P or x alleles of the ER have a better BMD cyclic addition of norestisterone acetate significantly in-response to raloxifene. creased lumbar spine BMD after one year [12].

Conclusion. Raloxifene and, most likely, other SERMs,Nevertheless, there could be some problems with thecould represent a good alternative to HRT in postmenopausal

use of HRT in uremic women. First, it has been demon-uremic women.strated that women on dialysis, when compared to non-uremic women, showed elevated baseline estradiol levelswhich increased significantly after the ingestion of estra-diol. This could suggest that renal failure may decrease

ESTROGEN AND BONE IN UREMIA estradiol catabolism and affect the pharmacokinetics ofThe term osteoporosis must be applied with caution exogenous estradiol [13]. Second, a recent study on the

in both uremic and patients on dialysis, who have a risks and benefits of estrogen plus progestin in healthycomplex range of metabolic bone disease [1]. Neverthe- postmenopausal women has introduced a word of cau-less, osteoporosis could begin early in time. An impor- tion in HRT therapy [14]. In this randomized controlledtant proportion of young women on dialysis (�50 years primary prevention trial of the Women’s Health Initia-old) are already amenorrheic and relatively hypoestro- tive, in which 16,608 postmenopausal women were fol-genic with significantly lower bone mineral density lowed for 5.2 years, it was demonstrated that the out-

comes were more harmful than beneficial in the estrogenplus progestin group versus the placebo group. AlthoughKey words: osteoporosis, bone mineral density, hemodialysis, meno-the estrogen plus progestin treatment reduced the ob-pause, hormone replacement therapy, selective estrogen receptor mod-

ulator, raloxifene. served hip and vertebral fracture rate and colon cancerby one third, the rate of women experiencing coronary 2003 by the International Society of Nephrology

S-62

Weisinger et al: SERMs in chronic renal failure S-63

heart disease increased by 29%, venous thromboembo-lism was twofold greater, stroke rates were increased by41%, and invasive breast cancer increased by 26%.

SELECTIVE ESTROGEN RECEPTORMODULATORS IN HEALTHY WOMEN

Several alternatives are available for women who can-not, or refuse to, take estrogen. Two potential estrogen-like therapies include raloxifene and tamoxifen. Theseare selective estrogen receptor modulators (SERMs),and they possess tissue-selective estrogen agonist andantagonist properties [15]. In ovariectomized animals,raloxifene preserves bone density, lowers serum total

Fig. 1. Changes in lumbar spine BMD (g/cm2) in postmenopausal hem-cholesterol concentrations, and inhibits aortic choles- odialyzed women given raloxifene 60 mg/day or placebo for one year.

Open bars represent values at baseline and the closed bars after oneterol accumulation, without causing endometrial hyper-year of treatment. *P � 0.01.plasia [16]. The mechanism responsible for the apparent

tissue-selective activity of raloxifene is not completelyunderstood. In vitro experiments suggest that raloxifenehas different effects than estradiol at the estrogen recep- Baseline blood determinations and BMD analysistor, including differential modulation of DNA response were obtained. Blood was drawn every three months,elements [17] and induction of a different conformational and BMD was evaluated after one year. To investigatechange in the transactivation domain of the ligand-bind- whether the ER polymorphism could have predicted the

response of BMD to raloxifene in those postmenopausaling domain [18]. A long-term (40 months) clinical trial,women in chronic hemodialysis, blood was obtained forthe Multiple Outcomes of Raloxifene Evaluation (MORE)genomic DNA extraction from peripheral leukocytes.study, involving 7705 postmenopausal women with os-

The ER gene polymorphic region for both PvuII andteoporosis, found that raloxifene’s antagonistic effect onXbaI, located in part of intron 1 and exon 2, was ampli-the breast appeared to cause a reduction in the risk offied by PCR using the following primers:breast cancer [19]. Raloxifene did not increase the risk

of endometrial cancer, but there was an increased inci-5� – CTG CCA CCC TAT CTG TAT CTT TTC CTAdence of thromboembolic disease, hot flashes, influenza-

TTC TCC – 3�like symptoms, peripheral edema, and leg cramps. In5� – TCT TTC TCT GCC ACC CTG GCG TCG ATTthis same study, most of the women had lumbar spine

ATC TGA – 3�x-rays at baseline and after 36 months of treatment.Among the women receiving 60 mg raloxifene, 6.6% had

After digestion, the products of PvuII and XbaI re-new vertebral fractures, compared with 10.1% in thestriction fragment length polymorphism (RFLP) were

placebo group, but the risk of nonvertebral fractures wassubmitted to agarose gel electrophoresis. The presence

similar. Thus, treatment of uremic women with a SERM, of the restriction site was labeled as p or x alleles (forwith their spectrum of breast and uterus antagonist ef- PvuII and XbaI reactions, respectively), and its absencefects, and action on bone as an estrogen agonist, seems as P or X.a less harmful and more rational therapeutic strategy [20]. Both groups of women were similar regarding age,

time on dialysis, time after menopause, medical manage-ment of renal osteodystrophy, and degree of osteoporo-

SELECTIVE ESTROGEN RECEPTORsis or osteopenia. Blood biochemical parameters, includ-

MODULATORS IN UREMIC WOMEN ing serum creatinine, calcium, PTH, and phosphorus, asWe have recently analyzed results from a prospective, well as sexual hormone levels (total estradiol, follicle

controlled, double-blind study on the effect of raloxifene stimulating hormone, and luteinizing hormone) wereon bone and lipid metabolism on postmenopausal also similar in both groups.women on hemodialysis with severe osteopenia (Her- Bone mineral density at the lumbar spine improvednandez E et al, in press). Selected women were more significantly after one year of treatment with raloxifene,than 50 years old, had at least two years of menopause, with no change in the placebo group (Fig. 1). No signifi-and evidence of severe osteopenia or osteoporosis (T cant changes were observed at the femoral neck. Like-

wise, there was a decrease in markers of bone resorptionscore below �2.0 SD).

Weisinger et al: SERMs in chronic renal failureS-64

in high-risk women with respect to coronary and cerebro-vascular events. The long-term effect of SERMs inchronic renal failure patients remains to be determined.

ACKNOWLEDGMENTS

This study was supported by grant G-97-008808 of the Fondo Nacio-nal de Ciencia, Tecnologıa y Innovacion de Venezuela (FONACIT)and Fundarenal-HUC. The help of the physicians, nurses, and secre-tarial staff of our dialysis units is fully appreciated. We thank SamiraA. Gomes and Larissa G. Ferreira for performing the estrogen receptorgene determinations. We acknowledge the contribution of Drs. JavierSan Martin and Lee Kay Pen, from Ely Lilly and Co. for providingthe study drug.

Fig. 2. Changes in lumbar spine BMD (g/cm2), Z-, and T-scores ac- Reprint requests to Jose R. Weisinger, M.D., Hospital Universitariocording to the estrogen receptor polymorphism (for homozygous PP de Caracas, Division of Nephrology, Apartado Postal 47365, Los Cha-or XX (A) and heterozygous PpXx women (B). Open bars represent

guaramos, Caracas, Venezuela 1040.values at baseline and the closed bars after one year of treatment. *P �E-mail: jweising@telcel.net.ve0.05.

REFERENCES

1. Goodman WG, Coburn JW, Slatopolsky E, Salusky IB: Renal(serum pyridinoline) in the raloxifene group. Blood lev-osteodystrophy in adults and children, in Primer on the Metabolicels of total cholesterol, HDL- and LDL-cholesterol, andBone Diseases and Disorders of Mineral Metabolism (4th ed), ed-serum triglycerides were also determined at baseline and ited by Favus MJ, Philadelphia, Lippincott-Williams & Wilkins,

after one year on drug or placebo. A significant decrease 1999, pp 347–3632. Weisinger JR, Gonzalez L, Alvarez H, et al: Role of persistentin LDL-cholesterol values with no changes in total cho-

amenorrhea in bone mineral metabolism of young hemodialyzedlesterol or triglycerides was observed after six monthswomen. Kidney Int 58:331–335, 2000in the raloxifene group. No side effects, including venous

3. Cochrane R, Regan L: Undetected gynecological disorders inthrombosis, pulmonary thromboembolism, or clotting women with renal disease. Human Reprod 12:667–670, 1997problems with the vascular accesses or dialysis catheters, 4. Holley JL, Schmidt RJ, Bender FH, et al: Gynecological and

reproductive issues in women on dialysis. Am J Kidney Dis 29:685–were observed after one year on raloxifene.690, 1997Genetic factors may influence the rate of change in

5. Lindberg JS, Moe SM: Osteoporosis in end stage renal disease.bone mineral density and the ER gene has been sug- Semin Nephrol 19:115–122, 1999gested as a candidate genetic marker for osteoporosis. 6. Weisinger JR: Role of hormone replacement in the management

of osteoporosis in hemodialyzed women: Perspectives for the fu-For this reason, we decided to evaluate whether theture. Nephrol Dial Transplant 15(Suppl 5):36–37, 2000ER polymorphism could have predicted the response of

7. Pacifici R, Rifas L, McCracken R, et al: Ovarian steroid treatmentBMD to raloxifene. Six genotypes were detected in the blocks a postmenopausal increase in blood monocyte interleukin 1whole sample (PPXX, PPXx, ppxx, Ppxx, PpXx, ppXx). release. Proc Natl Acad Sci USA 86:2398–2402, 1989

8. Jones TH: Interleukin-6, an endocrine cytokine. Clin EndocrinolBaseline lumbar spine or femoral neck BMD parameters40:703–713, 1994did not differ between patients presenting PP or xx geno-

9. Jilka RL, Hangoc G, Girasole G, et al: Increased osteoclasttypes versus Pp/Xx genotypes. However, those patients development after estrogen loss: Mediation by interleukin-6. Sci-homozygous for the P or x (PP or xx) alleles of the ence 257:88–91, 1992

10. Horowitz M: Cytokines and estrogen in bone: Anti-osteoporoticER showed a significantly higher mean BMD, Z-, andeffects. Science 260:626–627, 1993T-scores at the lumbar spine after one year on raloxifene

11. Naveh-Many T, Almogi G, Livni N, Silver J: Estrogen receptors,when compared to baseline (Fig. 2). No significant differ- and biologic response in rat parathyroid tissue and C-cells. J Clinences were observed in lumbar spine BMD, Z-, or Invest 90:2434–2438, 1992

12. Matuszkiewicz-Rowinska SK, Radowicki S, et al: The benefitsT-scores in heterozygous women (Pp/Xx).of hormone replacement therapy in pre-menopausal women withoestrogen deficiency on haemodialysis. Nephrol Dial Transplant14:1238–1243, 1999CONCLUSION

13. Ginsburg ES, Owen WF, Jr, Greenberg L, et al: Estrogen absorp-Although fracture risk reduction data are needed, tion and metabolism in postmenopausal women with end-stage

these preliminary observations on the effect of raloxifene renal disease. J Clin Endocrinol Metab 81:4414–4417, 199614. Writing Group for the Women’s Health Initiative Investiga-on BMD and bone resorption markers in uremic women

tors: Risks and benefits of estrogen plus progestin in healthy post-are interesting. Another important beneficial effect ofmenopausal women: Principal results from the Women’s Healthraloxifene in postmenopausal uremic women could be Initiative Randomized Controlled Trial. JAMA 288:321–333, 2002

the decrease in LDL-cholesterol and the fact that raloxi- 15. Cosman F, Lindsay R: Selective estrogen receptor modulators:Clinical spectrum. Endocr Rev 20:418–434, 1999fene has been shown to be neutral overall, and protective

Weisinger et al: SERMs in chronic renal failure S-65

16. Bjarnason NH, Haarbo J, Byrjalsen I, et al: Raloxifene inhibits 19. Ettinger B, Black DM, Mitlak BH, et al: Reduction of vertebralfracture risk in postmenopausal women with osteoporosis treatedaortic accumulation of cholesterol in ovariectomized, cholesterol-

fed rabbits. Circulation 96:1964–1969, 1997 with raloxifene results from a 3-year: Randomized clinical trial.JAMA 282:637–645, 199917. Yang NN, Venugopalan M, Hardikar S, Glasebrook A: Identi-

fication of an estrogen response element activated by metabolites 20. Riggs BL, Melton J, III: Bone turnover matters: The raloxifenetreatment paradox of dramatic decreases in vertebral fracturesof 17beta-estradiol and raloxifene. Science 273:1222–1225, 1996

18. Khovidhunkit W, Showack DM: Clinical effects of raloxifene without commensurate increases in bone density. J Bone MinerRes 17:11–17, 2002hydrochloride in women. Ann Intern Med 130:431–439, 1999