Applying Best Evidence to Menopause Management

Applying Best Evidence to Menopause Management Judith Walsh, MD, MPH Division of General Internal Medicine Women s Health Center of Excellence UCSF Overview Natural history of menopause Hormone therapy: Risks and Benefits Menopausal symptoms Current role of hormone therapy for menopausal symptoms Non-hormonal treatment of menopausal symptoms Feminine Forever MENOPAUSE IS NOT A DISEASE Dr. Robert Wilson, 1966 Replacing estrogen is like diabetics replacing insulin Women will be much more pleasant to live with and will not become dull and unattractive. Wyeth-Ayerst funded all expenses Page 1

Menopause Is A Positive Step Natural History of Menopause Gallup poll 1997: Most middle aged American women welcome menopause as a new and fulfilling life stage. Goal: Support women in achieving a successful transition Average age is 51 Predictors of age at menopause Genetics Family history Ethnicity» Earlier in Latino and later in Japanese American compared to Caucasians Smoking: about two years earlier Reproductive history» Never having children and shorter cycle length associated with earlier menopause Vasomotor Symptoms What do you tell her about when they will go away? Minnie Pause is a 53 year old woman who had her last menstrual period 18 months ago. She is still having hot flashes and awakens at least twice a night with them. She is considering taking estrogen but wants to know how much longer this will last. What do you tell her? Average duration is about 2 years and so they should be gone in about 6 months. Average duration is about 4 years They will never go away Page 2

Background Duration of Vasomotor Symptoms Treatment for menopausal symptoms is based on their transitory nature Many clinical guidelines suggest that symptom duration is approximately 2 years Many studies do not follow women more than 2 years Risks and benefits of hormone therapy depend on duration of use Use lowest dose for shortest duration Politi MC et al. Revisiting the duration of vasomotor symptoms of menopause: a meta-analysis. JGIM 2008:23: 1507-13 Objective: to estimate the natural progression of menopausal symptoms Vasomotor symptoms Results Rigorous meta-analysis included 10 studies with over 35,000 participants Clear definition of vasomotor symptoms Assessed prevalence of symptoms and bothersome symptoms Percent of women with symptoms increased in the two years before the final menstrual period (FMP), peaked one year after the FMP and did not return to premenopausal levels until 8 years after the FMP 50% of women had symptoms during the 4 years after FMP 10% of women had symptoms up to 12 years after FMP Page 3

Results: Bothersome Symptoms Duration of Hot Flushes Most prior studies examined populations of older women Newer evidence suggests that these durations might be longer when younger women are included Counseling regarding duration of hot flushes may inform clinical decision making The News Duration of menopausal hot flushes and associated risk factors Freeman, EW et al. Obstetrics and Gynecology, May 2011 Aim: To estimate the duration of moderate-to-severe menopausal hot flushes and identify potential risk factors for hot flush duration. Methods Penn Ovarian Aging Study - 436 premenopausal women, ages 35-47 years (not using hormonal therapy of any type); followed for 13 years Analytic Subsample: 259 women did not report hot flushes at baseline but did experience mod/severe sx during follow-up Measures: validated menopausal symptom list embedded in structured interview, hormones Assessments made at 9-month intervals during study years 0-5, then annually until year 10, every other year until study completion Page 4

Menopausal stages Results Premenopause: regular menstrual cycles 22-35 days long Late premenopause: change of more than 7 days in cycle length Early transition: changes in cycle length of 7 days or more in either direction for 2 consecutive menstrual cycles OR 60 days amenorrhea Late transition: 90 d to 11 months amenorrhea Postmenopause: 12 months or more amenorrhea Soules, et al. Climacteric 2001 Results Hazard Ratio for Likelihood of Hot Flashes Ending Take Home Message Variable HR (95% CI) Menopausal Premenopausal or Late Premenopausal Ref. Early Transition 3.26 (1.78-5.97) Late Transition or Postmenopausal 5.14 (2.70-9.77) Age 39 or younger 0.33 (0.14-0.78) 40-44 0.52 (0.30-0.91) 45-49 Ref. 50 or older 1.00 (0.54-1.87) White Race (Ref: African American) 1.73 (1.11-2.68) BMI 30 or more (Ref: BMI under 30) 1.94 (1.25-3.02) Estradiol (mean) 0.82 (0.64-1.05) In this population-based cohort, median duration of moderate-to-severe hot flushes was 10.2 years! Adding mild hot flushes duration was 11.6 years Younger, thinner, African-American women are likely to have longer hot flush duration Effect of hormone therapy on hot flush duration was not evaluated Clinicians counseling patients about hot flash duration should be mindful that the earlier the hot flashes start, the longer they are likely to last! Page 5

Hormone Therapy Hormone therapy is recommended for prevention of which of the following conditions? What have we learned from the Women s Health Initiative? What questions still remain unanswered? What is the current role of HT? Osteoporosis CHD Colorectal Cancer 1 and 3 None of the above Women s Health initiative WHI OUTCOMES Aim was to determine the health risks and benefits of estrogen/progestin in healthy women 16,608 women aged 50-79 with a uterus Randomized to estrogen/progestin or placebo Primary outcomes CHD (CHD death or non-fatal MI) Invasive breast cancer Writing Group for WHI JAMA 2002 E/P arm stopped at 5.2 years Page 6

OTHER WHI PUBLICATIONS HT and the Brain Effects on the brain Effects on CHD Effects on breast cancer Effects on colorectal cancer Effects on osteoporosis Effects on venous thrombosis Effects on gallbladder disease Effects on incontinence Effects of estrogen alone WHI Memory Study Subgroup of WHI Slight increase in dementia in hormone therapy group» Absolute number of cases was small Greater decrease in modified mini mental status exam in hormone therapy group Rapp, JAMA 2003; Shumaker, JAMA 2003 Increase in stroke in HT group 1.8% vs 1.3% Hazard ratio 1.31 (1.02, 1.68) Wassertheil-Stoller, 2003 HT AND CHD: WHI BREAST CANCER OUTCOMES: WHI Final results of the effects of estrogen and progestin on CHD outcomes HT users had a hazard ratio for CHD higher than non-users HR 1.24 (95% C.I. 1.00, 1.54) Risk was highest in the first year HR 1.81 (95% C.I. 1.09, 3.01) Manson, NEJM 2003 Since the WHI, use of hormone therapy (HT) in the U.S. has decreased Breast cancer incidence has also decreased in the U.S. What is the long term impact of HT on breast cancer? Page 7

Adherent Participants in WHI: Invasive Breast Cancers by Group Figure 2. Incidence of Invasive Breast Cancer in the WHI Clinical Trial Chlebowski RT et al: JAMA 2003; 289: 3243-3253 Chlebowski, R. T. et al. JAMA 2010;304:1684-1692 Copyright restrictions may apply. Breast Cancer Outcomes: WHI OSTEOPOSIS RESULTS: WHI Combined E+P therapy increases the cumulative risk of invasive breast cancer, and diagnosed cancers are more likely to be node-positive This risk becomes evident within about 4.7 years of randomization and is persistent through a total of 11 years of follow-up There were more breast cancer deaths in the combined E+P group Chlebowski JAMA 2010 11.1% of placebo treated women compared with 8.6% of estrogen treated women had a fracture Hazard ratio 0.76 (95% C.I. 0.69, 0.83) Effects did not differ by age, BMI, smoking status, history of falls, family history of osteoporosis, past use of hormone therapy or BMD No global benefit even in high risk women» Cauley, JAMA 2003 Page 8

COLORECTAL CANCER: WHI VTE Outcomes: WHI Reduced rate of colorectal cancer in hormone therapy users Hazard ratio 0.56; 95% C.I. 0.38, 0.81) Invasive cancers in the hormone group had a greater number of positive lymph nodes and were more advanced» Chlebowski NEJM 2004 Risk increased overall in HT users HR 2.06 (95% C.I. 1.57, 2.70) Risk higher in older women HR 4.28 (95% C.I. 2.38, 7.72) for women aged 60-69 HR 7.46 (95% C.I. 4.32, 14.38) for women aged 70-79 Risk higher in obese women HR 3.8 (95% C.I. 2.08, 6.94) for overweight HR 5.61 (95% C.I. 3.12, 10.11) for obese Factor V Leiden increased HT associated risk Cushman, JAMA 2004 Gallbladder Disease: WHI Urinary Incontinence: WHI Increased risk of any gallbladder disease or surgery in hormone therapy users HR 1.67 (95% C.I.. 1.35, 2.06) in estrogen users HR 1.59 (95% C.I. 1.28, 1.97) in estrogen/progestin users Increased risk of cholecystitis, cholelithiasis, and cholecystectomy in both groups NNH 323 for CEE 500 for E plus P Cirillo JAMA 2005 HT has long been thought to relieve the symptoms of urinary incontinence HT increased the incidence of all types of urinary incontinence at 1 year among women who were continent at baseline Among women with UI at baseline, frequency worsened in hormone users Did not assess vaginal estrogens» Hendrix JAMA 2005 Page 9

HT and Quality of Life WHI: Estrogen Alone In the WHI there were small improvements in quality of life among women who had vasomotor symptoms Hormone therapy is associated with improvement in some quality of life measures in women with vasomotor symptoms and may improve sexual function and vitality Hess, 2008; Welton, 2008 10,739 postmenopausal women aged 50-79 with prior hysterectomy Intervention phase ended early after reviewing data through November, 2003 Average follow-up 6.8 years WHI: Estrogen Alone WHI: Outcomes after Stopping CEE Non-significant reduction in breast cancer RH 0.77 (95% C.I. 0.59, 1.01) No reduction in colon cancer RH 1.08 (95% C.I. 0.75, 1.55) Non-significant increase in pulmonary embolism RH 1.34 (95% C.I. 0.87, 2.06) Lower risk of fractures RH 0.70 (95% C.I. 0.63, 0.79) Increased risk of stroke Global index not consistent with harm RH 1.01 (95% C.I. 0.91, 1.12) Follow-up after treatment with CEE CEE use median 5.9 years Mean follow-up 10.7 years No increase or decrease in CHD, DVT, stroke, hip fracture, CRC or total mortality Decreased risk of breast cancer continued» LaCroix, JAMA 2011 Page 10

Adherent Participants In WHI: Invasive Breast Cancers In E Alone Study WHI OUTCOMES (E +P) 23% reduction in BC over 11 years with 6 years of CEE in women without a uterus. Recently updated from all WHI studies Treat 10,000 women for a year 9 strokes 9 P.E.s 12 DVTs 8 invasive breast cancers 5 more lung cancer deaths 6 fewer hip fractures 46 fewer fractures 22 cases of dementia 20 cases of gall bladder disease 872 cases or urinary incontinence Nelson et al. Ann Intern Med 2012:157:104-113 LaCroix, JAMA, 2011 WHI Outcomes (E alone) CLINICAL QUESTIONS Recently updated from all WHI studies Treat 10,000 women for a year 56 fewer fractures 8 fewer invasive breast cancers 2 fewer deaths 11 more strokes 7 more DVTs 33 more cases of gallbaldder disease 1,271 more cases of urinary incontinence What about other estrogen/progestin combinations? What about younger women? Does it matter when you start HT? What about symptomatic treatment? Nelson et al. Ann Inern Med 2012: 157:104-113. Page 11

Transdermal Estrogen Age at Initiation: Background Transdermal estrogen avoids hepatic first pass metabolism Transdermal estrogen may be associated with a lower risk of VTE Canonico, 2007 Transdermal estrogen associated with a lower risk of stroke Renoux BMJ 2010 RCTs and observational studies have found differing results of the effect of HT on CHD Observational studies have generally shown benefit Women in observational studies are generally younger and started HT closer to the time of menopause Mean age of women in the WHI was 63 Hormone Therapy in Younger Women Hormone Therapy in Younger Women Secondary analysis of WHI data Combined the women who received E plus P with the women who received E alone Risk for CHD with HT increased with increasing time since menopause < 10 years: 0.76 (0.50, 1.16) 10-19 years: 1.10 (0.84, 1.45) 20 or more years 1.28 (1.03, 1.58)» P for trend=0.02 Roussow JAMA 2007 Hormone therapy increased risk of stroke regardless of age or years since menopause» HR 1.32 (1.12, 1.56) Non-significant trend for the effects of hormone therapy on total mortality to be more favorable in younger women Page 12

Estrogen Alone: Results by Age at Enrollment Take home message Statistically significant age interactions for CEE group greater safety and benefit for women in 50s potential harm among older women Timing of HT initiation relative to age or menopause onset may influence CHD risk Less underlying atherosclerosis in younger women? Clinicians should counsel women differently based on age and hysterectomy status Unlikely to change current recommendations for HT use Statistically significant age interactions (greater safety and benefit for younger women, potential harm among older women) were shown for: CHD, total MI, colorectal CA, total mortality and the global index. La Croix et al. JAMA 2011 Danish Osteoporosis Study: CV Outcomes 1006 healthy recently menopausal women aged 45-58 Open-label randomization to triphasic estradiol/norethisterone or placebo (hysterectomy: treatment with estradiol alone) for 10 years Outcomes: Composite outcome of death or admission for myocardial infarction or heart failure Individual components of primary endpoint, admission for stroke Schierbeck et al. BMJ 2012; 345:e6409. Results WHI: mean age 63 WHI: 13% with increased lipids WHI: 36% with HTN WHI: 4% with DM Page 13

Primary endpoint and mortality for hormone replacement therapy Take Home Messages The findings from this study substantiate the timing theory but Different type and dose of estrogen used Underpowered to determine effect of Estrogen monotherapy Unclear if primary composite outcome was prespecified Minimal risk associated with therapy but Likely underpowered to detect differences in DVT/PE, CVA, and possibly breast CA Overall: reassured about using HT in symptomatic healthy menopausal women, but would not prescribe for primary prevention Schierbeck L L et al. BMJ 2012;345:bmj.e6409 Menopausal Symptoms: Prevalence Menopausal Symptoms Hot flushes (50% or more)» Often with perspiration Night sweats (50% or more) Sleep disturbance (40-60%) Page 14

OTHER SYMPTOMS Other symptoms happen at the time of menopause but are less clearly related to menopause Mood changes Cognition Changes in sexual function Urinary complaints Treatment of Menopausal Symptoms What do you most commonly recommend for the treatment of severe vasomotor symptoms? HT for Symptomatic Relief Estrogen SSRI Venlafaxine Clonidine Gabapentin Lifestyle changes and/or complementary therapy Any form of estrogen is highly effective Generally can be taken for a few years and gradually stopped A progestin should be added for women with a uterus Therapy can be tailored to a woman s preference Decision should be made SEPARATELY from decision for disease prevention Page 15

Effective Dose Equivalents Lower dose hormone therapy Dose that stops hot flashes in 80% of women 1 mg micronized 17 beta estradiol 50mcg/day transdermal 17 beta-estradiol 0.625 mg conjugated equine estrogens 1.25 mg piperazine estrone sulfate Lower doses have been effective in some trials Estimates of efficacy after 12 weeks 38% placebo 63% low dose estrogen 83% standard dose estrogen Lower doses may take longer for maximal symptom relief 12 weeks vs 4-8 weeks Less bleeding and breast tenderness and may require less progestin» Ettinger, Am J Med 2005 QUESTION QUESTION Estee Jenn is a 60 year old woman who has been on HT for 10 years. You have been trying to encourage her to stop it for a while but she has not wanted to do it. Her best friend has recently developed breast cancer; she has now decided to stop, and wants your advice on the best way to do it. What do you recommend? Taper by decreasing the daily dose over 6-12 months Taper by decreasing the number of days a week HT is used over 6-12 months Just stop Page 16

Discontinuing hormone therapy Progestins Symptoms will recur in up to 25% of women with stopping therapy Unclear if it is best to stop cold turkey or to taper Taper can be by daily dose or number of days per week Taper until mild symptoms Maintain that dose until symptoms resolve Concern about VTE and breast cancer risk High doses Oral megestrol (20-80 mg) has shown efficacy Some weight gain Goodwin J Clin Oncology 2008 High dose depo-mpa (400 mg) is also effective Compared with venlafaxine Loprinzi J Clin Oncol 2006 QUESTION OTHER DRUG TREATMENTS Estee has a resumption of her hot flashes after she stops her estrogen. What pharmacologic alternative do you suggest? Paroxitene Escitalopram Venlafaxine Clonidine Gabapentin SSRIs Venlafaxine Desvenlafaxine Clonidine 50-67% reduction in hot flash frequency with these regimens Placebo effects generally large Generally less than estrogen Page 17

Paroxitene Escitalopram Paroxitene CR led to a significant decrease in hot flash score 62% in 12.5 mg group 65% in 25 mg group 38% in placebo group Avoid in women receiving tamoxifen Decreases active metabolite of tamoxifen Cytochrome P450 CYP2D6 Recent study of the efficacy of escitalopram Reduction in hot flash frequency 55% in escitalopram group 36% in placebo group Effective in African American and Caucasian women Effective regardless of coexisting anxiety or depression» Freeman, JAMA 2011 Venlafaxine Desvenlafaxine Significant reduction in hot flashes 61% vs 27% in placebo (p<0.01) 150 mg no more effective than 75 mg Lopinzi, Lancet 2000 Industry sponsored trial of metabolite of venlafaxine 700 women with severe hot flashes 64% reduction in hot flashes at 12 weeks Vs 51% with placebo Hot flashes less severe in desvenlafaxine group Not currently FDA approved for this indication Speroff, 2008 Page 18

Clonidine and Gabapentin Alternatives for treatment of hot flushes Clonidine Start with 0.1 mg/day transdermal patch 40% reduction in hot flashes Side effects can be limiting Gabapentin 45% reduction in hot flashes vs placebo (29%) 900 mg a day more effective than placebo 300-600 mg at bedtime may help with hot flashes that awaken patients from sleep Treatment Dosage (mg/day) Efficacy (vs placebo) Estrogen 0.625 CEE 80% vs 20-30% Paroxitene 12.5 or 25 62% vs 38% Escitalopram 10 to 20 55% vs 36% Venlafaxine 37.5 37% vs 27% Venlafaxine 75 or 150 61% vs 27% Desvenlafaxine 100 64% vs 51% Clonidine 0.1 38% vs 20% Question Background Estee is tired of medications and would like to try an herbal therapy for treatment of her hot flashes. What treatment do you recommend? Evening primrose Ginseng Dietary soy Wild yam None of the above Soy products have been proposed to provide comparable benefits to estrogen for treating menopausal symptoms, without the risks Epidemiologic studies on Asian women suggest that soy-containing foods are of benefit to the skeleton Limited by short duration, low dose of soy isoflavones, few participants Rapid bone loss occurs during the first 2 years of menopause Page 19

SPARE Study: Soy isoflavones Results SPARE study (Soy Phytoestrogens As Replacement Estrogen) parallel group, placebo controlled, double blind trial 248 women (early menopause) were randomly assigned to receive 200mg soy isoflavones or placebo Measures: BMD change Menopausal symptoms Vaginal Maturation Lipids and thyrotropin Assessed at baseline, 12 mo, & 24 mo No significant differences between soy and placebo group: Spinal BMD Total hip BMD Femoral neck BMD NTx (N-telopeptide type I bone collagen) Subgroup analyses by: race, BMI, estradiol, 25-OH Vit D Women with low 25-OH vit D at baseline soy group had lower decrease in spinal BMD than placebo Rate of hot flashes actually increased in soy isoflavone group (48% vs 32%: p=0.02) Summary Of Herbal Evidence Lifestyle and Complementary Modalities Evidence: no benefit Red Clover Dong quai Ginseng Evening primrose Wild yam Vitamin E Acupuncture Evidence mixed Soy Black cohosh No data Chasteberry Licorice Cooling body temperature Exercise Avoiding hot and spicy foods Relaxing therapies Yoga, massage, mediation, slow breathing, baths Mind-body therapies Relaxation, biofeedback, paced respiration, hypnosis NAMS Some evidence that weight loss is beneficial Page 20

Hot Flash: Self Help - Be Cool! Relaxation Keep house cool Avoid caffeine, alcohol, spicy foods, hot drinks Dress in layers, cotton Light bed linens Use fan, cool drinks Exercise Relaxation therapy 73% reduction HF Diaphragmatic breathing Simple deep breathing when sense HF > 50% reduction HF Wijma, 1997 Freedman 1995 Recommendations Guidelines for Hormone Therapy Use USPSTF: Harmful effects are likely to exceed the chronic disease prevention benefits in most women ACOG, AHA, and Canadian Task Force recommend against use of HT for prevention of chronic disease NAMS 2012: When alternative therapies not appropriate, extended use of HT appropriate for women at high risk of fracture Page 21

NAMS 2012 Recommendations Summary Focuses on emerging differences between ET and EPT as varying ages and time intervals since menopause Individualization in decision to use HT :consider individual health, personal risk factors and quality of life priorities ET has a more favorable risk benefit profile which allows for more flexibility in duration of use EPT associated with an increased risk of breast cancer incidence and mortality after 3-5 years Average duration of menopausal symptoms is approximately 4 years but seems to be longer in younger women Estrogen either alone or with a progestin is not recommended for chronic disease prevention in postmenopausal women Risks and benefits may differ in older and younger women Premature menopause: HT until median age of natural menopause and then reassess Menopause 2012:257-71 Summary Estrogen works best for menopausal symptoms Use lowest dose for shortest duration Best method for discontinuation is not known Start with lifestyle modifications and nonprescription remedies Other drug alternatives include venlafaxine, desvenlafaxine, paroxitene, escitalopram, gabapentin, and clonidine Questions? Page 22