Clinical Therapeutics/Volume 37, Number 7, 2015
Review Article
Role of the Pharmacist in Medication Therapy Management Services in Patients With Osteoporosis Megan Murphy-Menezes, PharmD, MSCR College of Pharmacy, University of Florida, Gainesville, Florida ABSTRACT Purpose: This study aimed to review and summarize clinical trial data, updated guidelines, and expert opinions on the recommendations for drug holidays in patients being treated for osteoporosis. Methods: Three unique PubMed searches using the terms osteoporosis drug holiday, osteoporosis discontinuation, and osteoporosis extension produced 149 articles, of which 16 with relevance to postmenopausal osteoporosis and treatment discontinuation were selected. A separate search of 6 national societies with interest in the treatment of osteoporosis produced 2 guideline statements with specific mention of therapy duration. Findings: No pharmacologic therapy for osteoporosis should be considered indefinite. After 3 to 5 years of therapy, a comprehensive patient assessment of individualized risk for fracture should occur. In patients who are at mild to moderate risk for fracture after initial treatment, therapy with a bisphosphonate may be discontinued, whereas patients who remain at high risk for fracture should continue therapy. Although discontinuation with bisphosphonate agents may be appropriate due to mechanism of action and expected residual effects, which may last several years after discontinuation, discontinuation of shorter-acting agents (nonbisphosphonate agents) will result in a rapid reversal of therapeutic benefits and may not be appropriate. Patients who undergo a drug holiday should be monitored and reassessed for reinitiation of therapy. Caution should be employed by readers generalizing results to male patients with osteoporosis as pivotal trial data are limited to postmenopausal women with osteoporosis. Implications: Pharmacists play a crucial role in ensuring that patients achieve optimal therapeutic outcomes. This achievement is partly accomplished through medication-adherence monitoring and education. Most chronic disease states require that patients remain on
July 2015
therapy, even when asymptomatic, for an indefinite period to meet and maintain therapeutic goals. To improve health-related outcomes, some pharmacistmanaged medication therapy management programs are devoted entirely to treatment adherence. However, osteoporosis is a chronic disease state for which indefinite therapy may no longer be recommended due to potential long-term tolerability issues. It is crucial that the pharmacist understand the data and new guidelines addressing the optimal duration of therapy in osteoporotic patients so that they recognize which patients qualify for a drug holiday, do not mistakenly categorize a patient as nonadherent when therapy is stopped, and recommend when therapy should be restarted. This understanding will result in better patient management and health system cost-savings. (Clin Ther. 2015;37:1573–1586) & 2015 Elsevier HS Journals, Inc. All rights reserved. Key words: adherence, discontinuation, drug holiday, FRAX, medication therapy management, osteoporosis.
INTRODUCTION Osteoporosis is a chronic disease state characterized by low bone mass, deteriorated bone architecture, and compromised bone strength that leads to an increased risk for fracture.1 It has been estimated that in the United States, there are 2 million osteoporotic fractures each year and that 50% of white women and 20% of men will have a fracture related to osteoporosis in their lifetime.1 The clinical impact on patients is undeniable, with the mortality at 1 year after hip fracture estimated to be 14% to 58%.2 With Accepted for publication March 11, 2015. http://dx.doi.org/10.1016/j.clinthera.2015.03.023 0149-2918/$ - see front matter & 2015 Elsevier HS Journals, Inc. All rights reserved.
1573
Clinical Therapeutics the aging of the US population, the National Bone Health Alliance has predicted that by 2025, the cost to the health care system of fractures will be US $25 billion/y.3 According to practice guidelines, men and women aged Z50 years should receive pharmacologic treatment if they have a history of a hip or vertebral fracture; a T score r–2.5 at the lumbar spine, total hip, or femoral neck; or a low bone mass (T score –1.0 to –2.5 at the femoral neck or lumbar spine) with a 10-year risk for hip fracture Z3% or a 10-year risk for major osteoporosis-related fracture Z20% based on the US-adapted World Health Organization algorithm.1,4 Antiresorptive medications, most notably the bisphosphonate agents, are the most commonly prescribed class of medication for the treatment of osteoporosis, with a vertebral fracture risk reduction of 40% to 70% and a relative hip fracture risk reduction of 40% to 50%.5–10 Despite the effectiveness of these medications, adherence to treatment with these agents has been challenging, with data showing the 1year persistence rate of weekly bisphosphonate agents to be 35% to 45%.11 The reasons for nonadherence may include denial of disease state, fear of adverse events, cost, and desire for natural remedies. Whatever the reason, in osteoporosis, as in other chronic disease states, much of a pharmacist’s time is spent counseling patients on the importance of medication adherence. As is common in many chronic and progressive disease states, before 2010, the duration of therapy with antiresorptive agents for osteoporosis was regarded as indefinite.5–8,12 However, in 2012, in response to concerns of esophageal cancer, osteonecrosis of the jaw (ONJ), and atypical fracture, 3 rare adverse events associated with long-term therapy with bisphosphonate agents, the US Food and Drug Administration (FDA) released a statement that the optimal duration of bisphosphonate therapy is unknown and the need for continued therapy should be reevaluated at regular intervals.13 Since then, there has been increased interest in determining the appropriate duration of therapy with all antiresorptive agents.11–18 The concept of drug holidays, a prescribed time “off” therapy, has been proposed as a means to individualize treatment in patients with osteoporosis, with the risk and benefit of long-term therapy being weighed for each patient’s unique clinical presentation.1,11,13–15,19–22 As pharmacists are increasingly involved in medication therapy management (MTM),
1574
adherence monitoring, and patient education, it is crucial that they are aware of the new data and guidelines addressing optimal duration of therapy in osteoporotic patients so that holidays are not mistaken for nonadherence.21 Furthermore, pharmacists may have a role in identifying when and whether patients qualify for a holiday as well as qualify for retreatment, all leading to better patient outcomes and health system cost-savings. This article reviews the data associated with antiresorptive agent discontinuation and re-initiation of therapy for osteoporosis.
MATERIALS AND METHODS Three unique PubMed searches using the terms osteoporosis drug holiday, osteoporosis discontinuation, and osteoporosis extension were conducted in September 2014. The searches were limited to clinical trials and reviews published within the previous 10 years. A total of 149 articles were identified in the search. Titles and abstracts of the provided citations were screened and selected only if they were related to osteoporosis, antiresorptive therapy, duration of therapy, discontinuation, and drug holiday. Articles were excluded if they were related to anabolic therapy, were related to products marketed outside of the United States, were published outside of the United States, were related to treatments prescribed for indications other than osteoporosis (eg, cancer or pediatric indications), or were tolerability reviews. A total of 16 articles were analyzed for this review. In addition, the following national society guidelines were reviewed for specific recommendations on drug holidays and therapy durations: the American Association of Clinical Endocrinologists (AACE), American College of Rheumatology, American Society of Bone and Mineral Research, International Society of Clinical Densitometry, National Osteoporosis Foundation (NOF), and the North American Menopausal Society. Recommendations from AACE and NOF were included in this review. The results are presented in the following format: overview of therapeutic classes, followed by overview of individual therapeutic agents, including a review of data from pivotal trials, extensions, and discontinuations as they relate to bone mineral density (BMD), bone turnover markers (BTMs), fracture, and tolerability. Finally, a summary of recommendations regarding drug holidays and therapy durations from the AACE and NOF are included.
Volume 37 Number 7
M. Murphy-Menezes
RESULTS Bisphosphonate Agents Bisphosphonate agents bind to hydroxyapatite crystals and are incorporated into the bone. Once osteoclasts begin to resorb bone, bisphosphonate is released, enters the osteoclast, and results in loss of function and apoptosis. The structure of bisphosphonate agents varies slightly by compound and consists of 2 phosphonic acids, joined by a carbon with 2 side chains, R1 and R2.15 The R1 side chain is involved in binding affinity to bone. Of the current US-marketed bisphosphonate agents, the rank order of affinity is zoledronate 4 alendronate 4 ibandronate 4 risedronate.20 Bisphosphonate agents with lower binding affinity are distributed more widely through the bone and spend less time incorporated in bone. Therefore, a quicker reversibility could be expected with loweraffinity bisphosphonate agents than with higheraffinity bisphosphonate agents. The terminal half-life of bisphosphonate agents has been reported to be within the range of years, with the half-life of alendronate noted to be 10 years.5 The R2 side chain is involved in antiresorptive potency, with rank order of potency zoledronate 4 risedronate 4 ibandronate 4 alendronate.20 Higher potency results in a greater reduction in osteoclast activity, faster apoptosis, and greater suppression of bone turnover.20 Together, the unique R1 and R2 side chains are involved in the length and magnitude of effect seen with each bisphosphonate.
eligible for enrollment.23 Patients were randomized to receive alendronate 5 mg/d for 2 years and 10 mg/d thereafter (n ¼ 3236) or placebo (n ¼ 3223), along with calcium and vitamin D. After 5 years of the FIT study, women who had completed at least 3 years of treatment in the FIT trial and who were being treated with alendronate were eligible for participation in the open-label FLEX trial. Women who had a T score of o–3.5 and a FLEX baseline hip BMD that was less than the FIT hip BMD were excluded. Women enrolled in the extension study were randomized to receive alendronate 5 or 10 mg/d or placebo for an additional 5 years (Figure 1). All patients received 500 mg/d of calcium and 250 IU/d vitamin D. A total of 1099 women were enrolled in the extension. The mean age of FLEX participants was 73 years, 34% had a prevalent vertebral fracture, and 60% had a history of clinical fracture since menopause. Mean T scores were –1.9, –2.2, and –1.3 at the total hip, femoral neck, and lumbar spine, respectively.
BMD For women who were randomized to continue on alendronate, there was a mean total hip BMD decline of –1.02% compared with a –3.38% decrease with placebo, a mean difference of þ2.36% (P o 0.001). The femoral neck and trochanter BMD also displayed losses between alendronate and placebo, with mean loss differences of þ1.94% and þ3.17%, respectively
Alendronate Alendronate has been approved by the FDA for the treatment and prevention of postmenopausal osteoporosis, treatment to increase bone mass in men, and the treatment of glucocorticoid osteoporosis.6 The pivotal Phase III trial FIT (Fracture Intervention Trial)23 reported efficacy in reducing fractures, both clinical and morphometric. Alendronate is available as 5- and 10-mg daily doses and 35- and 70-mg weekly doses, with higher doses approved for the treatment indication.6 To assess long-term efficacy, the FLEX (FIT Long-Term Extension) study16 was conducted to compare the effects of discontinuing alendronate treatment after 5 years versus continuing for 10 years. The primary end point was total hip BMD. The FLEX study was an extension of the pivotal FIT trial.16 In the FIT trial, postmenopausal women aged 55 to 81 years with low femoral neck BMD were
July 2015
Placebo
Placebo Alendronate 5mg or 10mg/day
0
Alendronate 5mg or 10mg/day
5 Year
10
Figure 1. Designs of the FIT (Fracture Intervention Trial)23 and FLEX (FIT Long-Term Extension)16 studies.
1575
Clinical Therapeutics (both, P o 0.001). In contrast, patients randomized to continue on alendronate saw a greater increase in lumbar spine BMD than did the patients receiving placebo, with a mean difference of þ3.74% (P o 0.001) (Table I). In the women who received alendronate in both FIT and FLEX, there was a greater increase in total hip BMD, þ2.41%, compared with that in the subjects who were switched to placebo in FLEX, who saw a decrease of –0.16% (mean difference, þ2.57%; P o 0.001) (data not shown).
vertebral fractures were less in the continued alendronate group compared with those in the placebo group.
Tolerability There were no significant between-group differences in the prevalence of serious adverse events. With regard to adverse events of interest, there were no differences in the rates of upper gastrointestinal tract events, and no cases of ONJ were observed.
Risedronate
BTMs For subjects who continued on alendronate, BTMs— serum C-terminal telopeptide of type 1 collagen (sCTX), serum N-terminal telopeptide (sNTX), both markers of resorption, and serum bone-specific alkaline phosphate (sBSAP), a marker of formation— remained near FLEX baseline values. Patients who received placebo in the extension had gradual increases in mean BTM values over the 5 years, but values were still below FIT baseline levels.
Fracture Post hoc analyses demonstrated that there were no significant differences between treatment groups in all clinical fractures, nonvertebral fractures, or morphometric vertebral fractures. However, women receiving continued alendronate had a significantly lower percentage of clinical vertebral fractures than did women receiving placebo (2.4% vs 5.3%; 95% CI, 0.24–0.85). Although not a statistically significant finding, in a subgroup with lower baseline BMD and a prevalent fracture, the prevalence of nonvertebral and clinical
Risedronate has been approved by the FDA for the treatment and prevention of osteoporosis in postmenopausal women, as treatment to increase bone mass in men with osteoporosis, and as treatment and prevention of glucocorticoid-induced osteoporosis in men and women who are either initiating or continuing systemic glucocorticoid treatment (Z7.5 mg/d of prednisone or equivalent) for chronic diseases.7 Pivotal Phase III trials have reported efficacy in reducing vertebral and nonvertebral fractures.24 Risedronate is available as a daily (5 mg), weekly (35 mg), or monthly (75 mg taken 2 days consecutively, or 150 mg) treatment. To assess the resolution of effect with risedronate therapy, the pivotal Phase III, 3-year fracture study VERT-NA (Vertebral Efficacy with Risedronate Trial–North America)24 was extended for a 1-year observational period. The primary end points of the extension study were the mean percentage changes from baseline in lumbar spine BMD values at months 36 and 48. To be enrolled in the original VERT-NA study, eligible women were at least 5 years postmenopausal,
Table I. Percentage changes from baseline in bone mineral density with alendronate versus placebo. Site
Pooled Alendronate, Mean (SE) (n ¼ 643)
Total hip Femoral neck Trochanter Lumbar spine
–1.02 þ0.46 –0.08 þ5.26
(0.18) (0.24) (0.22) (0.24)
Placebo, Mean (SE) (n ¼ 428) –3.38 –1.48 –3.25 þ1.52
(0.22) (0.30) (0.27) (0.29)
Treatment Difference, Mean (95% CI) þ2.36 þ1.94 þ3.17 þ3.74
(1.81–2.90)* (1.20–2.68)* (2.49–3.84)* (3.03–4.45)*
Data from the FLEX study.16 P o 0.001.
*
1576
Volume 37 Number 7
M. Murphy-Menezes aged o85 years, and had Z2 vertebral fractures or 1 vertebral fracture and a T score o–2. Women were randomized to receive risedronate 2.5 or 5 mg/d or placebo for 3 years. All women received 1000 mg/d calcium and, if baseline serum 25-hydroxyvitamin D was low, vitamin D supplementation. All women who completed the 3-year treatment period (except 121 women who underwent bone biopsy at the end of the 3-year trial) were invited to participate in a 4th, observational year, in which risedronate and placebo were discontinued but calcium and vitamin D were given as described earlier (Figure 2).25 A total of 759 women were enrolled in the 1-year observation phase.25 The mean age was 68 years. The number of prevalent vertebral fractures was not significantly different between the groups who had been treated with risedronate compared with the group who had been treated with placebo. However, as expected at the fourth-year baseline, the group who had previously been treated with risedronate had higher BMD and lower BTM values than did those who received placebo. At baseline of the initial 3-year VERT-NA trial, all baseline characteristics were balanced.
BMD At the end of the 4th year, lumbar spine BMD was decreased by –0.83% in patients previously treated with risedronate. However, BMD remained higher than at the start of the original trial, VERT-NA, and was significantly higher than that in the previous
Placebo
Off-Treatment
Table II. Percentage changes from baseline in bone mineral density with risedronate versus placebo.
Site Lumbar spine Femoral neck Trochanter
Former Risedronate Group, Mean (SE) (n ¼ 296)
Treatment Difference, Mean (95% CI)
–0.83* –1.23* –1.57*
þ2.60 (1.56–3.65)* þ2.32 (1.40–3.25)* þ3.08 (2.06–4.10)*
Data from the risedronate extension study.25 P o 0.001.
*
placebo group at the end of the 4th year, with a between-group difference of þ2.60% (P o 0.001). At the femoral neck, patients previously treated with risedronate had a BMD decrease of –1.23%, which appeared to approach baseline but was still greater than that in the patients previously treated with placebo, with a mean difference of þ2.32% (P o 0.001) (Table II).
BTMs Urinary NTX, a marker of bone resorption, was measured in a subset of women at the end of year 4. In the patients previously treated with risedronate, there was a significant increase from a median of 30.3 to 50.9 nmol bone collagen equivalents per mmol of creatinine (P o 0.05). Although not significantly different from that with placebo, the NTX level remained significantly lower than the baseline value.
Fracture Risedronate 2.5mg or 5mg/day
0
Off-Treatment
3
4
In the 4th year, 6.5% of the original risedronate group experienced a vertebral fracture, while 11.6% of the original placebo group experienced a vertebral fracture (relative risk reduction, 46%; P ¼ 0.009).
Year
Figure 2. Designs of the VERT-NA (Vertebral Efficacy with Risedronate Trial–North America)24 and 1-year extension25 studies.
July 2015
Tolerability In the year of discontinuation, there were no differences in the prevalence of adverse events in the group previously treated with risedronate compared with the group that previously received placebo.
1577
Clinical Therapeutics In 2008, results from a large-scale administrative database studying the clinical implications of bisphosphonate discontinuation were published.5 Data from a US health care organization covering 25 million people were used for identifying a cohort of new bisphosphonate users. The cohort included women 60 to 78 years of age who were new users of alendronate or risedronate from January 1998 to July 2005, had been compliant with bisphosphonate therapy for at least 2 years, and did not have a claim for hip fracture, malignancy, HIV disease, or Paget disease before initiating therapy with a bisphosphonate. Compliance was defined as a medication possession ratio (MPR) of 66% to 100% at 2 years. Discontinuation was defined as discontinuation for any length of time and qualified by duration of discontinuation. The objective of the retrospective study was to determine the association between bisphosphonate discontinuation and hip fracture among persons who had been compliant with bisphosphonate therapy for at least 2 years. Observation of the cohort began at 2 years. From the database, 9063 women initiating treatment with a bisphosphonate were found to be compliant at 2 years and eligible for the discontinuation analysis. In the 6 months leading up to the initiation of osteoporosis medication, 64% of patients had received a BMD test and 51% had a medical diagnosis of osteoporosis. Seventy-seven percent of women initiated therapy with alendronate and 23% initiated therapy with risedronate. Among the cohort of 9063, there were 71 hip fractures identified. In the cohort of patients who had an MPR of 466% at 2 years, the incidence of hip fracture among patients who continued therapy was 4.67% per 1000 patient-years. In the same cohort, in the patients who discontinued therapy, the incidence of hip fracture was 8.43%. Of importance, although not statistically significant, the hip fracture rate in all subgroups (MPR, 466% at 2 and 3 years, 480% at 2 and 3 years) that discontinued was higher than that in the group that continued therapy. Discontinuation of a bisphosphonate for up to 1 year was not associated with a significant increase in the rate of hip fractures. However, in the group with a MPR 466% at 2 years, discontinuation of 1 year or longer was associated with a 2- to 3-fold increase in the relative risk for hip fracture. In addition, among women with greater compliance, MPR 480%, and a longer duration of preceding bisphosphonate therapy, the incidence rate of hip fracture among women
1578
discontinuing therapy was numerically lower than those in women who were less compliant and in women with a shorter treatment duration.
Zoledronate Zoledronate has been FDA-approved for the treatment and prevention of postmenopausal osteoporosis, as treatment to increase bone mass in men with osteoporosis, and for the treatment and prevention of glucocorticoid-induced osteoporosis.9 The pivotal Phase III study of zoledronate, HORIZON-PFT (Health Outcomes and Reduced Incidence with Zoledronic acid Once Yearly Pivotal Fracture Trial),26 reported the efficacy of zoledronate in reducing spine, hip, and other nonvertebral fractures. The approved doses of zoledronate are 5 mg/y for treatment indications and 5 mg every 2 years for preventive indications.9 To determine the effect of zoledronate beyond 3 years, an extension of HORIZON-PFT was conducted.27 The primary end point of the extension was change in femoral neck BMD at year 6 relative to year 3 (baseline of the extension study). In the original HORIZON-PFT study, postmenopausal women between the ages of 65 and 90 years were eligible for inclusion if they had a BMD T score of r2.5 at the femoral neck with or without evidence of existing vertebral fracture, or a T-score r1.5 with radiologic evidence of Z2 mild vertebral fractures or 1 moderate vertebral fracture. Previous use of oral bisphosphonate agents was allowed, with the duration of the washout period dependent on duration of previous use. Exclusion criteria included use of therapy known to affect bone metabolism. Patients with a serum calcium level 42.75 mmol/L or o2.00 mmol/L were ineligible, as were patients with a calculated creatinine clearance of o30.0 mL/min at either of 2 baseline visits or urine dipstick results of 42 for protein, without evidence of contamination or bacteriuria. A total of 7765 osteoporotic women were assigned to receive zoledronate 5 mg IV or inactive vehicle (placebo) given once per year.26 For the extension, women were eligible if they had received 3 infusions in the parent study. Patients were excluded if they had previous major protocol violations, received placebo in the pivotal study, were aged 493 years, or were taking other medications known to affect bone metabolism.27 A total of 1233 patients were enrolled in the extension and were randomized to receive zoledronate for an additional 3 years
Volume 37 Number 7
M. Murphy-Menezes (Z6 group), or placebo for 3 years (Z3P3 group) (Figure 3). All patients received oral calcium 1000 to 1500 mg/d and vitamin D 400 to 1200 IU/d. At baseline of the extension, the mean age of the women was 75.5 years. Mean femoral neck T score was –2.5, and 60% of patients had Z1 prevalent vertebral fracture. The baseline characteristics were not statistically different between the Z6 and Z3P3 groups.
BMD At the end of the sixth year of the extension, there was a statistically significant difference in the mean change in femoral neck BMD, with an increase of þ0.24% in the Z6 group compared with a decrease of –0.80% in the Z3P3 group (mean difference, þ1.04%; P ¼ 0.0009). At the total hip, there was a loss of –0.36 in the Z6 group compared with a loss of –1.58% in the Z3P3 group (mean difference, þ1.22%; P o 0.0001). At the 1/3 radius, the mean difference was þ0.37, which was not statistically significant. The largest difference in loss was seen at the lumbar spine (mean difference, þ2.03%; P ¼ 0.002) (Table III).
BTMs At the end of the 6th year of the extension, mean serum type 1 procollagen N-terminal (P1NP), a marker of formation normally coupled to CTX, increased by þ33% in the Z3P3 group compared with an increase of þ19% in the Z6 group (difference, 14%; P ¼ 0.0001). The absolute difference in ending
Table III. Percentage changes from baseline in bone mineral density with zoledronate 6 months (Z6; n ¼ 616) versus zoledronate 3 months (Z3P3; n ¼ 617). Site Femoral neck Total hip Lumbar spine Distal radius
Treatment Difference, Mean (95% CI) þ1.04 þ1.22 þ2.03 þ0.37
(0.43 to 1.65)* (0.75 to 1.70)† (0.76 to 3.29)‡ (–0.71 to 1.45)§
Data from the HORIZON extension study.27 P o 0.001. † P o 0.0001. ‡ P ¼ 0.002. § P ¼ 0.50.
*
P1NP at year 6 was 3.0 ng/mL (P ¼ 0.001). In the Z3P3 group, P1NP remained statistically significantly less than pretreatment values. The absolute difference in serum CTX between Z6 and Z3P3 at year 6 was 0.02 (P ¼ 0.45).
Fracture Morphometric vertebral fracture risk was significantly lower in the Z6 group compared with that in the Z3P3 group (n ¼ 14 and 30 respectively; OR ¼ 0.51; P ¼ 0.035). However, there were no differences in the prevalence of nonvertebral, hip, or clinical fractures in patients who continued on zoledronate versus those who were switched to placebo.
Placebo
Tolerability Placebo Zolendronate 5mg IV/year
0
Zolendronate 5mg IV/year 3 Year
6
Figure 3. Designs of the HORIZON-PFT (Health Outcomes and Reduced Incidence with Zoledronic acid Once Yearly Pivotal Fracture Trial)26 and 3-year extension27 studies.
July 2015
There was a statistically significantly greater number of patients with an increase from baseline in serum creatinine 40.5 mg/dL in the Z6 group compared with the Z3 group (n ¼ 18 vs 4, respectively; P ¼ 0.002). Atrial fibrillation was slightly more common in the Z6 group, but the difference was not statistically significant (n ¼ 21 vs 13). In addition, there was a numerically greater number of strokes in the Z6 group than in the Z3P3 group, but the difference was not statistically different (n ¼ 19 vs 9). There was 1 case of ONJ in the Z6 group and no cases of atypical femur fracture.
1579
Clinical Therapeutics
RANKL Inhibitor Receptor activator of nuclear factor-κB ligand (RANKL) is a key mediator of osteoclast formation, function, and survival.28 When RANKL binds to the RANK receptor on the surface of osteoclasts, osteoclasts differentiate and begin to actively resorb bone. In the presence of estrogen, RANKL activity is regulated by osteoprotegrin, a soluble protein, which binds to RANKL, preventing binding to the RANK receptor. This system of physiologic checks and balances results in balanced rates of resorption compared with rates of formation. However, when estrogen decreases, excessive amounts of unopposed RANKL are produced, resulting in a disproportionally high level of osteoclast activity and bone resorption. RANKL inhibitors are not incorporated into the bone, and studies have reported nonlinear pharmacokinetic properties, with rapid resolution of effect.29 Currently there is only 1 RANKL inhibitor on the US market.
Denosumab Denosumab is a human monoclonal antibody against RANKL, FDA-approved for the treatment of postmenopausal women with osteoporosis at high risk for fracture and as treatment to increase bone mass in men with osteoporosis.10 Denosumab is given by subcutaneous injection, 60 mg every 6 months, and pivotal data have suggested reductions in vertebral, nonvertebral, and hip fractures.30 An early Phase III, 24-month trial was conducted to determine the effect of denosumab on BMD.32 At the end of 24 months, that study was extended for an additional 24 months to determine the effects of prior administration of denosumab (or placebo) on BMD, BTMs, and tolerability.32 No singular end point was specified as primary, and all changes in BMD, changes in BTMs, and tolerability were exploratory end points. In the original 24-month BMD parent study, women were eligible if they were ambulatory and postmenopausal, had a lumbar spine T score between –1.0 and –2.5, had no underlying medical condition known to result from abnormal bone metabolism, and had no history of fracture after 25 years of age. Exclusion criteria included the use of concurrent medications known to affect bone metabolism, including oral bisphosphonate agents, for at least 3 years cumulatively. If oral bisphosphonate agents were used for 43 months but r3 years, women were eligible if the last dose was administered at least 1 year
1580
before enrollment. Women who used an oral bisphosphonate agent for r3 months were eligible. A total of 332 patients were enrolled in the original 24-month study and were randomized to receive denosumab 60 mg or inactive vehicle (placebo) every 6 months. All participants were instructed to take Z1000 mg/d of calcium and Z400 IU/d vitamin D. Each participant who completed the on-treatment phase, including the last dose of the scheduled regimen of denosumab or placebo, were eligible for participation in the extension study (Figure 4). Two hundred fifty-six patients were enrolled in the extension arm of the study. During the extension phase, no therapy was given in either group. Patients were instructed to continue supplementation with calcium and vitamin D. The mean age of the patients in the extension study was 59 years, with a mean duration of postmenopause of 10 years and a mean lumbar spine T score of –1.61. Overall, baseline demographic and clinical characteristics were balanced at the beginning of the extension study.
BMD At the end of the original 24-month study, patients treated with denosumab had statistically greater gains in BMD at the lumbar spine, total hip, and 1/3 radius than did patients treated with placebo. At the end of the extension phase, 24 months off therapy, BMD was decreased in both groups. Of particular interest, in women previously treated with denosumab, values of BMD at the lumbar spine, total hip, and 1/3 radius declined, with values reaching near-baseline values. However, BMD values in women previously treated with denosumab remained higher than BMD values in
0
Placebo
Off-Treatment
Denosumab 60mg q 6 months
Off-Treatment
2 Year
4
Figure 4. Designs of the denosumab parent study31 and 24-year extension study.32
Volume 37 Number 7
M. Murphy-Menezes women previously treated with placebo. In the patients previously treated with denosumab, most of the decreases in BMD occurred in months 24 to 36, with changes in BMD in months 36 to 48 appearing similar to those seen in the placebo group (data not available).
BTMs On denosumab discontinuation at month 24, there was a rapid reversal in BTM values. Within 3 months, serum CTX increased to above baseline levels, peaked at 30 months, and returned to baseline levels by month 48. Similar results were seen with P1NP, with BTMs increasing to above baseline levels at month 6, reaching peak levels at month 36, and returning to baseline by month 48. In patients previously treated with denosumab, moderate correlations between month-48 sCTX and P1NP measurements and baseline sCTX and P1NP was demonstrated (r ¼ 0.47 and 0.46, respectively; P o 0.001).
Fracture No clinical vertebral fractures were reported. In the original 24-month study, 4% of patients in the placebo group and 1% in the denosumab group experienced centrally confirmed fractures. In the extension, off-treatment phase, 3% of patients previously treated with placebo and 3% previously treated with denosumab experienced centrally confirmed fractures.
Tolerability The prevalence of adverse and serious adverse events was balanced between groups in the offtreatment phase. With regard to adverse events of interest, the prevalence of rash was 0% in the group previously treated with denosumab and 2% in the group previously receiving placebo. There were no reports of ONJ, no difference in neoplasm, and no notable changes in serum calcium. In a separate Phase II clinical trial of denosumab, the effect of re-treatment after a drug-free, discontinuation period was studied.33 In the original Phase II, 4-year, open-label parent study, postmenopausal women aged r80 years with low bone mass were eligible for participation in the study if they had a BMD T score between –1.8 and –4.0 at the lumbar spine or between –1.8 and –3.5 at the total hip or femoral neck. The parent study included 731 patients with a mean age of 62 years, a mean lumbar spine
July 2015
BMD T score of –2.1, and a mean femoral neck BMD T score of –1.8. In the 4-year parent study, a small group of 41 patients was treated with denosumab 30 mg every 3 months for 24 months. After 24 months, patients were discontinued from therapy for 1 year and then re-treated with denosumab 60 mg every 6 months for an additional year (Figure 5). In the 12 months off treatment, BMD declined to near-baseline values at all sites measured—the lumbar spine, total hip, and onethird radius. On retreatment, BMD rebounded within 12 months, back to BMD levels seen at month 24 before discontinuation (data not available).
Estrogen Hormone therapy (HT) with estrogen mimics a woman’s premenopausal physiology and prevents rapid loss of BMD. Estrogen therapy has been approved by the FDA under many brand names for the prevention of osteoporosis.1 The 6-year EPIC (Early Postmenopausal Intervention Cohort) study35 was designed to evaluate the use of alendronate and estrogen–progestin in the prevention of osteoporosis as well as the effects of discontinuation. A total of 1609 postmenopausal women aged 45 to 60 were enrolled. Women with a prior fracture or who had a lumbar spine BMD o0.8 g/cm2 were excluded. Patients were randomized to receive alendronate, estrogen þ medroxyprogesterone, or placebo for various durations, followed by time off therapy (Figure 6). All patients were instructed to take calcium.
BMD/BTMs BMD decreased at all skeletal sites in women who were switched from alendronate to placebo but remained above baseline. The women who had previously been treated with HT for 4 years had large BMD losses. In the group that stopped taking alendronate, BTM values gradually reverted to pretreatment
Denosumab 30mg q 6 month 0
Denosumab 60mg q 6 month
Off-Treatment
2
3
4
Year
Figure 5. Design of the denosumab 4-year parent study.33
1581
Clinical Therapeutics
Placebo
Placebo
Placebo
Alendronate
Alendronate
Alendronate
Alendronate
Alendronate
Placebo
Alendronate
Placebo
Placebo
HT
HT
Off-Treatment
0
4
2
6
Year
Figure 6. Design of the EPIC (Early Postmenopausal Intervention Cohort) study.34
levels but remained below baseline values. In the group that stopped taking HT, there was a rapid increase in NTX to above that in the placebo group.
Tolerability The prevalence of adverse events related to treatment discontinuation was similar between the groups that received alendronate and placebo. More patients treated with HT reported adverse events than did those treated with alendronate.
Cathepsin K Inhibitor Osteoclasts are involved in bone resorption. Bone resorption is a 3-step process.35 First, osteoclasts secrete acid into the resorption lacunae on the bone surface. Second, in the presence of the resultant acidic environment, osteoclasts produce cathepsin K, which is involved in the degradation of organic matrix. Research has shown that inhibition of cathepsin K can slow bone resorption through suppressed osteoclast activity.35
Participants received vitamin D 56,000 IU/wk and calcium supplementation 500 mg/d if their total dose of calcium intake through normal diet and supplementation was o1000 mg/d. On completion of the 24-month original study, patients were eligible to participate in a 1-year extension if they were in good health based on medical history and physical and laboratory evaluations. Two hundred eighty participants entered the extension phase and were randomized to receive either odanacatib 50 mg/wk or placebo (Figure 7). At baseline of the extension (after 2 years of treatment), the mean age was 64 years, and the mean BMD T scores were –2.2 at the lumbar spine and –1.8 at the femoral neck.
BMD In patients who continued with 50 mg of odanacatib, there was a continued gain in BMD. From the end of year 2 to the end of year 3, there were increases of þ2.3% at the lumbar spine, þ2.4% at the total hip, and þ1.6% at the femoral neck. However, there was no significant change in the third year at the 1/3 radius, which produced a loss of –0.4%. In the patients who discontinued odanacatib treatment at 24 months and received placebo in the third year, BMD values were decreased at all sites and returned to levels not different from baseline (Table IV). After 1 year of discontinuation, patients who were first treated with odanacatib followed by placebo had BMD values that remained þ1.4% higher than baseline at the lumbar spine, þ2.3% higher than baseline at the femoral neck, –0.5% lower than baseline at the total hip, and –2.7% lower than baseline at the
Odanacatib Odanacatib is a cathepsin K inhibitor currently being studied for the treatment of postmenopausal osteoporosis, and its use has not been approved by the FDA.36 A Phase II, 24-month, dose-ranging study was extended for an additional year to determine the effects of continued odanacatib treatment versus discontinuation of treatment on BMD, BTMs, and tolerability.35 In the original Phase II, 24-month study, postmenopausal women between the ages of 45 and 85 years were eligible for the study if they had a BMD T score between –2.0 and –3.5.37 In this dose-ranging study, 399 women were randomized to receive weekly doses of odanacatib at 3, 10, 25, or 50 mg or placebo.
1582
Placebo Placebo
Odanacatib 50mg per week
Placebo
Odanacatib 3mg, 10mg, 25mg, or 50mg per week 0
Odanacatib 50mg per week 2
3
Figure 7. Designs of the odanacatib parent37 and 3-year extension35 studies.
Volume 37 Number 7
M. Murphy-Menezes
Table IV. Percentage changes from baseline in bone mineral density with odanacatib versus placebo. Site Lumbar spine Total hip Trochanter Femoral neck 1/3 Radius
Treatment Difference, Mean (95% CI) þ1.4 –0.5 –0.7 þ2.3 –2.7
(–0.8 to 3.6) (–2.6 to 1.6) (–3.8 to 2.5) (0.2 to 4.3) (–5.2 to –0.2)
Data from the odanacatib extension study.36
one-third radius. The rate of bone loss was fastest in the 6 months after discontinuation.
drug holiday may be considered after 4 to 5 years of stability, whereas in patients who have a high risk for fracture, a holiday of 1 to 2 years may be considered after 10 years of treatment. A review of evidence within the guidelines suggested that the appropriate duration of holiday may be different with each therapy. Whereas a holiday of 1 to 2 years may be justified with alendronate due to the high affinity for bone, a holiday of only 1 year may be appropriate with risedronate. The guidelines go on to say that residual effects should not be expected with therapies such as denosumab, raloxifene, and estrogen and therefore a drug holiday may not be appropriate.
National Osteoporosis Foundation In 2014, the NOF updated their treatment guidelines with recommendations similar to those of the AACE regarding durations of therapy and drug holidays.1
BTMs In the patients who continued treatment with odanacatib 50 mg, the changes in markers of resorption from year 2 to 3 were not significant. In the group that was discontinued from treatment at the end of 2 years, rapid increases in bone resorption markers to above baseline were observed at 30 months, with values returning to near-baseline at the end of year 3. With regard to markers of formation, in the group that continued treatment with odanacatib, sBSAP and serum P1NP decreased initially with treatment initiation, followed by a plateau, and then a subsequent increase to above (sBSAP) or near (P1NP) baseline values by the end of year 3.
Tolerability Overall in year 3, the prevalence of adverse events was balanced between the continued-treatment arm versus the treatment-discontinuation arm. However, in year 3, there was an imbalance in the number of events reported and treated as urinary tract infections or cystitis (12 in the active-treatment group, 3 in the placebo group); all were resolved with antibiotic treatment. No differences in the prevalence of skin disorders or in calcium levels were reported.
Review of Guidelines American Association of Clinical Endocrinologists The AACE’s 2010 clinical practice recommendations state that if osteoporosis is considered mild, a
July 2015
DISCUSSION Osteoporosis is a chronic disease state that, if left untreated, can result in fracture, leading to severe disability, loss of mobility, and decreased survival.1,2 Although the effectiveness of products marketed for the treatment of osteoporosis has been shown to decrease the risk for fracture, postmarketing surveillance has demonstrated very rare but serious adverse events.13,17,38 This tolerability concern has led health care professionals and even some patients to question the appropriate duration of therapy.12–15,19–22,39 The evidence reviewed in the present article shows that each therapy has a unique profile that, when matched with a patient’s risk factors, may justify different and patient-specific durations of therapy. Pharmacists are members of the health care team charged with the responsibility of drug therapy outcomes.21 As a result, in recent years there has been an increase in formalized medication therapy management (MTM) services, some of which are dedicated primarily to monitoring and improving compliance and adherence.18,21 As such, it is important that pharmacists understand current treatment guidelines and are aware of the changes that are introduced as a result of new data. As pharmacists perform medication reviews, they must not mistake a patient’s sudden discontinuation of osteoporosis therapy as nonadherence, which could result in inappropriate counseling. With an understanding of a patient’s risk factors and
1583
Clinical Therapeutics previous duration of therapy, a pharmacist can determine whether a drug holiday was likely ordered. In addition, as pharmacists perform medication reviews, they may be presented with the unique opportunity to determine when a holiday may be appropriate by again examining an individual’s previous duration of therapy and risk factors. The potential cost-savings from discontinuing unneeded therapy or from preventing an excessive treatment duration should not be underestimated. A holiday is a break with a specific start and end. A drug holiday with osteoporosis therapy is no different. In fact, the AACE and NOF guidelines recommend that patients be monitored during the holiday period, and if BMD declines substantially, BTM values increase, or fracture occurs, therapy should be restarted.1,4 This need for monitoring patients again presents a unique opportunity for pharmacists providing MTM services. Once a patient’s physician recommends an osteoporosis treatment
holiday, a pharmacist may create a reminder to reevaluate the patient in 1 to 2 years.1,4 In addition to BMD, BTMs, and fracture monitoring, pharmacists can easily reassess a patient’s risk by using the Fracture Risk Assessment Tool (FRAX), an online instrument designed for calculating a patient’s 10-year risk for hip or other major osteoporotic fracture (clinical vertebral, forearm, or proximal humerus) based on individual clinical risk factors.40 Treatment guidelines recommend therapy in patients with a low bone mass (T score of –1.5 to –2.5) and a 10-year risk for hip fracture of Z3% or a 10-year risk for major osteoporotic fracture of Z20%. Although FRAX was originally designed for use in treatment-naive patients, current NOF guidelines state that patients who have been off of osteoporosis medications for 1 to 2 years may be considered treatment naive and therefore may be appropriate for reassessment with FRAX.1 By using the FRAX tool, pharmacists have the opportunity to determine and recommend to treating physicians
Patient is low risk on an oral bisphosphonate x 5 years or Patient is high risk and on an oral bisphosphonate x 10 years
T score ≤ –2.5; Continue Therapy; DEXA q 1–2 years
No history of Nontraumatic Fracture
History of Nontraumatic Fracture or other High Risk Factors
T-score >–2.5; Consider Drug Holiday
T-score >–2.5; Consider Drug Holiday
DEXA q 1–2 years Consider BTM and FRAX
DEXA q1–2 years Consider BTM and FRAX
If experience fracture, BMD declines, and/or FRAX >3% at hip or >20 major OP fracture:
If experience fracture, BMD declines, and/or FRAX >3% at hip or >20 major OP fracture:
Resume Therapy
Resume Therapy
T-score ≤–2.5; Continue Therapy; DEXA q1–2 years
Figure 8. Example of proposed algorithm for selection of candidates for bisphosphonate drug holiday and principles of monitoring.39 This algorithm has been modified from its original form to incorporate recommendations from the 2014 National Osteoporosis Foundation guideline.1
1584
Volume 37 Number 7
M. Murphy-Menezes when extending a holiday may be appropriate or when the holiday should be ended and therapy reinitiated. Pharmacists who are involved in MTM services may benefit from developing treatment algorithms for reference. Figure 8 provides a modified example of such an algorithm that has been proposed by bone health specialists.41
CONCLUSIONS Treatment of chronic disease states, including osteoporosis, requires careful monitoring. The risk versus benefit of a therapy must be continually evaluated to determine the appropriateness of continued treatment. This type of management presents a unique opportunity for pharmacists whose professional responsibility includes managing drug therapy outcomes.
CONFLICTS OF INTEREST The author has indicated that she has no conflicts of interest with regard to the content of this article.
REFERENCES 1. National Osteoporosis Foundation. Clinician’s Guide to Prevention and Treatment of Osteoporosis. Washington, DC: National Osteoporosis Foundation; 2014. 2. Schnell S, Friedman S, Mendelson D, et al. The 1-year mortality of patients treated in a hip fracture program for elders. Geriatr Orthop Surg Rehab. 2010;1:6–14. 3. Burge R, Dawson-Hughes B, Soloman D, et al. Incidence and Economic Burden of Osteoporosis-Related Fractures in the United States, 2005–2025. J Bone Miner Res. 2007;22:465–475. 4. Watts N, Bilezikian J, Camacho P. American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the diagnosis and treatment of postmenopausal osteoporosis. Endocr Pract. 2010;16(Suppl 3): 1–37. 5. Curtis J, Westfall A, Cheng J, et al. Risk of hip fracture after bisphosphonate discontinuation: implications for a drug holiday. Osteoporos Int. 2008;19:1613–1620. 6. Fosamax [package insert]. Whitehouse Station, NJ: Merck; 2014. 7. Actonel [package insert]. Rockaway, NJ: Warner Chilcot; 2013. 8. Boniva [package insert]. South San Francisco, CA: Genentech; 2013. 9. Reclast [package insert]. East Hanover, NJ: Novartis; 2013. 10. Prolia [package insert]. Thousand Oaks, CA: Amgen; 2014.
July 2015
11. Silverman S, Schousboe B, Gold D. Oral bisphosphonate compliance and persistence: A matter of choice? 2011;22:21-26. 12. National Osteoporosis Foundation. Clinician’s Guide to Prevention and Treatment of Osteoporosis. Washington, DC: National Osteoporosis Foundation; 2005. 13. Whitaker M, Guo J, Kehoe T, et al. Bisphosphonate agents for osteoporosis-where do we go from here? N Engl J Med. 2012;366:2051–2053. 14. McClung M. Controversies in osteoporosis management: concerns about bisphosphonate agents and when are “drug holidays” required? Clin Obstet Gynecol. 2013;56: 743–748. 15. Sebba A. Osteoporosis: how long should we treat? Curr Opin Endocrinol Diabetes Obes. 2008;15:502–507. 16. Black D, Schwatrz A, Ensrud K, et al. Effects of continuing or stopping alendronate after 5 years of treatment. The Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial. JAMA. 2006;296:2927–2937. 17. Khosla S, Burr D, Cauley J, et al. Bisphosphonateassociated osteonecrosis of the jaw: report of a task force of the American Society for Bone and Mineral Research. J Bone Miner Res. 2007;22:1479–1491. 18. Stuurman-Bieze AG, Hiddink EG, van Boven JF, Vegter S. Proactive pharmaceutical care interventions decrease patients’ nonadherence to osteoporosis medication. Osteoporos Intl. 2014;25:1807–1812. 19. McClung M, Harris S, Miller P, et al. Bisphosphonate therapy for osteoporosis: benefits risks, and drug holiday. Am J Med. 2013;126:13–20. 20. Watts N, Diab D. Long-term use of bisphosphonate agents in osteoporosis. J Clin Endocrinol Metab. 2010;95: 1555–1565. 21. Figge H. Enabling medication therapy management. US Pharmacist. 2008:33. [Epub ahead of print]. 22. Dunn R, Bird M, Conway S, Stratton M. Use of bisphosphonate agents in older adults: how long is long enough? Consult Pharm. 2013;28:39–57. 23. Cummings S, Black D, Thompson D, et al. Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. JAMA. 1998;280:2077–2082. 24. Harris S, Watts N, Genant H, et al. Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis: a randomized controlled trial. Vertebral Efficacy with Risedronate Therapy (VERT) study group. JAMA. 1999;282:1344–1352. 25. Watts N, Chines A, Olszynski W, et al. Fracture risk remains reduced one year after discontinuation of risedronate. Osteoporos Int. 2008;19:365–372. 26. Black D, Delmas D, Eastell R, et al. Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. N Engl J Med. 2007;356:1809–1822.
1585
Clinical Therapeutics 27. Black D, Reid I, Boonen S, et al. The effect of 3 years versus 6 years of zoledronic acid treatment of osteoporosis: a randomized extension to the HORIZON-Pivotal Fracture Trial (PFT). J Bone Miner Res. 2012;27:243–254. 28. Kearns A, Khosla S, Kostenuik PJ. Receptor activator of nuclear factor kappaB ligand and osteoprotegerin regulation of bone remodeling in health and disease. Endocr Rev. 2008;29:155–192. 29. Bekker PJ, Holloway DL, Rasmussen AS, et al. A single-dose placebocontrolled study of AMG 162, a fully human monoclonal antibody to RANKL, in postmenopausal women. J Bone Miner Res. 2004: 1059–1066. 30. Cummings S, San Martin J, McClung M, et al. Denosumab for prevention of fractures in postmenopausal women with osteoporosis. N Engl J Med. 2009;361:756–765. 31. Bone HG, Bolognese MA, Yuen CK, et al. Effects of denosumab on bone mineral density and bone turnover in postmenopausal women. J Clin Endocrinol Metab. 93:2149– 2157. 32. Bone H, Bolognese M, Yuen C, et al. Effects of denosumab treatment and discontinuation on bone mineral density and bone turnover markers in postmenopausal women with low bone mass. J Clin Endocrinol Metab. 2001;96:972–980. 33. Miller P, Wagman R, Peacock M, et al. Effect of denosumab on bone mineral density and biochemical markers of bone turnover: six-year results of a phase 2 clinical trial. J Clin Endocrinol Metab. 2001;92: 394–402. 34. Wasnich R, Bagger y, Hosking D, et al. Changes in bone density and turnover after alendronate or estrogen withdrawal. Menopause. 2004; 11:622–630. 35. Eisman J, Bone H, Hosking D, et al. Odanacatib in the treatment of postmenopausal women with
1586
low bone mineral density: threeyear continued therapy and resolution of effect. J Bone Miner Res. 2011;26:242–251. 36. Lewiecki M. Odanacatib, a cathepsin K inhibitor for the treatment of osteoporosis and other skeletal disorders associated with excessive bone remodeling. IDrugs. 2009;12: 799–809. 37. Bone HG, McClung MR, Roux C, et al. Odanacatib, a cathepsin-k inhibitor for osteoporosis: a twoyear study in postmenopausal women with low bone density. J Bone Miner Res. 2010;25:937–947. 38. Shane E, Burr D, Ebeling P, et al. Atypical subtrochanteric and diaphyseal femoral fractures: report of a task force of the American Society for Bone and Mineral Research.
J Bone Miner Res. 2010;25:2267– 2294. 39. How long should you take certain osteoporosis drugs? FDA Consumer Health Information. May 2014. Retrieved from http://www. fda.gov/ForConsumers/Consumer Updates/ucm309688.htm. Accessed September, 2014. 40. Kanis JA on behalf of the World Health Organization Scientific Group. Assessment of osteoporosis at the primary health care level. Technical Report. World Health Organization Collaborating Center for Metabolic Bone Diseases. University of Sheffield, UK. 2007. 41. Ro C, Cooper O. Bisphosphonate drug holiday: choosing appropriate candidates. Curr Osteoporos Rep. 2013;11:45–51.
Address correspondence to: Megan Murphy-Menezes, University of Florida College of Pharmacy, PO Box 113195, Gainesville, FL 32611. E-mail: [email protected]fl.edu
Volume 37 Number 7
Review Article
Role of the Pharmacist in Medication Therapy Management Services in Patients With Osteoporosis Megan Murphy-Menezes, PharmD, MSCR College of Pharmacy, University of Florida, Gainesville, Florida ABSTRACT Purpose: This study aimed to review and summarize clinical trial data, updated guidelines, and expert opinions on the recommendations for drug holidays in patients being treated for osteoporosis. Methods: Three unique PubMed searches using the terms osteoporosis drug holiday, osteoporosis discontinuation, and osteoporosis extension produced 149 articles, of which 16 with relevance to postmenopausal osteoporosis and treatment discontinuation were selected. A separate search of 6 national societies with interest in the treatment of osteoporosis produced 2 guideline statements with specific mention of therapy duration. Findings: No pharmacologic therapy for osteoporosis should be considered indefinite. After 3 to 5 years of therapy, a comprehensive patient assessment of individualized risk for fracture should occur. In patients who are at mild to moderate risk for fracture after initial treatment, therapy with a bisphosphonate may be discontinued, whereas patients who remain at high risk for fracture should continue therapy. Although discontinuation with bisphosphonate agents may be appropriate due to mechanism of action and expected residual effects, which may last several years after discontinuation, discontinuation of shorter-acting agents (nonbisphosphonate agents) will result in a rapid reversal of therapeutic benefits and may not be appropriate. Patients who undergo a drug holiday should be monitored and reassessed for reinitiation of therapy. Caution should be employed by readers generalizing results to male patients with osteoporosis as pivotal trial data are limited to postmenopausal women with osteoporosis. Implications: Pharmacists play a crucial role in ensuring that patients achieve optimal therapeutic outcomes. This achievement is partly accomplished through medication-adherence monitoring and education. Most chronic disease states require that patients remain on
July 2015
therapy, even when asymptomatic, for an indefinite period to meet and maintain therapeutic goals. To improve health-related outcomes, some pharmacistmanaged medication therapy management programs are devoted entirely to treatment adherence. However, osteoporosis is a chronic disease state for which indefinite therapy may no longer be recommended due to potential long-term tolerability issues. It is crucial that the pharmacist understand the data and new guidelines addressing the optimal duration of therapy in osteoporotic patients so that they recognize which patients qualify for a drug holiday, do not mistakenly categorize a patient as nonadherent when therapy is stopped, and recommend when therapy should be restarted. This understanding will result in better patient management and health system cost-savings. (Clin Ther. 2015;37:1573–1586) & 2015 Elsevier HS Journals, Inc. All rights reserved. Key words: adherence, discontinuation, drug holiday, FRAX, medication therapy management, osteoporosis.
INTRODUCTION Osteoporosis is a chronic disease state characterized by low bone mass, deteriorated bone architecture, and compromised bone strength that leads to an increased risk for fracture.1 It has been estimated that in the United States, there are 2 million osteoporotic fractures each year and that 50% of white women and 20% of men will have a fracture related to osteoporosis in their lifetime.1 The clinical impact on patients is undeniable, with the mortality at 1 year after hip fracture estimated to be 14% to 58%.2 With Accepted for publication March 11, 2015. http://dx.doi.org/10.1016/j.clinthera.2015.03.023 0149-2918/$ - see front matter & 2015 Elsevier HS Journals, Inc. All rights reserved.
1573
Clinical Therapeutics the aging of the US population, the National Bone Health Alliance has predicted that by 2025, the cost to the health care system of fractures will be US $25 billion/y.3 According to practice guidelines, men and women aged Z50 years should receive pharmacologic treatment if they have a history of a hip or vertebral fracture; a T score r–2.5 at the lumbar spine, total hip, or femoral neck; or a low bone mass (T score –1.0 to –2.5 at the femoral neck or lumbar spine) with a 10-year risk for hip fracture Z3% or a 10-year risk for major osteoporosis-related fracture Z20% based on the US-adapted World Health Organization algorithm.1,4 Antiresorptive medications, most notably the bisphosphonate agents, are the most commonly prescribed class of medication for the treatment of osteoporosis, with a vertebral fracture risk reduction of 40% to 70% and a relative hip fracture risk reduction of 40% to 50%.5–10 Despite the effectiveness of these medications, adherence to treatment with these agents has been challenging, with data showing the 1year persistence rate of weekly bisphosphonate agents to be 35% to 45%.11 The reasons for nonadherence may include denial of disease state, fear of adverse events, cost, and desire for natural remedies. Whatever the reason, in osteoporosis, as in other chronic disease states, much of a pharmacist’s time is spent counseling patients on the importance of medication adherence. As is common in many chronic and progressive disease states, before 2010, the duration of therapy with antiresorptive agents for osteoporosis was regarded as indefinite.5–8,12 However, in 2012, in response to concerns of esophageal cancer, osteonecrosis of the jaw (ONJ), and atypical fracture, 3 rare adverse events associated with long-term therapy with bisphosphonate agents, the US Food and Drug Administration (FDA) released a statement that the optimal duration of bisphosphonate therapy is unknown and the need for continued therapy should be reevaluated at regular intervals.13 Since then, there has been increased interest in determining the appropriate duration of therapy with all antiresorptive agents.11–18 The concept of drug holidays, a prescribed time “off” therapy, has been proposed as a means to individualize treatment in patients with osteoporosis, with the risk and benefit of long-term therapy being weighed for each patient’s unique clinical presentation.1,11,13–15,19–22 As pharmacists are increasingly involved in medication therapy management (MTM),
1574
adherence monitoring, and patient education, it is crucial that they are aware of the new data and guidelines addressing optimal duration of therapy in osteoporotic patients so that holidays are not mistaken for nonadherence.21 Furthermore, pharmacists may have a role in identifying when and whether patients qualify for a holiday as well as qualify for retreatment, all leading to better patient outcomes and health system cost-savings. This article reviews the data associated with antiresorptive agent discontinuation and re-initiation of therapy for osteoporosis.
MATERIALS AND METHODS Three unique PubMed searches using the terms osteoporosis drug holiday, osteoporosis discontinuation, and osteoporosis extension were conducted in September 2014. The searches were limited to clinical trials and reviews published within the previous 10 years. A total of 149 articles were identified in the search. Titles and abstracts of the provided citations were screened and selected only if they were related to osteoporosis, antiresorptive therapy, duration of therapy, discontinuation, and drug holiday. Articles were excluded if they were related to anabolic therapy, were related to products marketed outside of the United States, were published outside of the United States, were related to treatments prescribed for indications other than osteoporosis (eg, cancer or pediatric indications), or were tolerability reviews. A total of 16 articles were analyzed for this review. In addition, the following national society guidelines were reviewed for specific recommendations on drug holidays and therapy durations: the American Association of Clinical Endocrinologists (AACE), American College of Rheumatology, American Society of Bone and Mineral Research, International Society of Clinical Densitometry, National Osteoporosis Foundation (NOF), and the North American Menopausal Society. Recommendations from AACE and NOF were included in this review. The results are presented in the following format: overview of therapeutic classes, followed by overview of individual therapeutic agents, including a review of data from pivotal trials, extensions, and discontinuations as they relate to bone mineral density (BMD), bone turnover markers (BTMs), fracture, and tolerability. Finally, a summary of recommendations regarding drug holidays and therapy durations from the AACE and NOF are included.
Volume 37 Number 7
M. Murphy-Menezes
RESULTS Bisphosphonate Agents Bisphosphonate agents bind to hydroxyapatite crystals and are incorporated into the bone. Once osteoclasts begin to resorb bone, bisphosphonate is released, enters the osteoclast, and results in loss of function and apoptosis. The structure of bisphosphonate agents varies slightly by compound and consists of 2 phosphonic acids, joined by a carbon with 2 side chains, R1 and R2.15 The R1 side chain is involved in binding affinity to bone. Of the current US-marketed bisphosphonate agents, the rank order of affinity is zoledronate 4 alendronate 4 ibandronate 4 risedronate.20 Bisphosphonate agents with lower binding affinity are distributed more widely through the bone and spend less time incorporated in bone. Therefore, a quicker reversibility could be expected with loweraffinity bisphosphonate agents than with higheraffinity bisphosphonate agents. The terminal half-life of bisphosphonate agents has been reported to be within the range of years, with the half-life of alendronate noted to be 10 years.5 The R2 side chain is involved in antiresorptive potency, with rank order of potency zoledronate 4 risedronate 4 ibandronate 4 alendronate.20 Higher potency results in a greater reduction in osteoclast activity, faster apoptosis, and greater suppression of bone turnover.20 Together, the unique R1 and R2 side chains are involved in the length and magnitude of effect seen with each bisphosphonate.
eligible for enrollment.23 Patients were randomized to receive alendronate 5 mg/d for 2 years and 10 mg/d thereafter (n ¼ 3236) or placebo (n ¼ 3223), along with calcium and vitamin D. After 5 years of the FIT study, women who had completed at least 3 years of treatment in the FIT trial and who were being treated with alendronate were eligible for participation in the open-label FLEX trial. Women who had a T score of o–3.5 and a FLEX baseline hip BMD that was less than the FIT hip BMD were excluded. Women enrolled in the extension study were randomized to receive alendronate 5 or 10 mg/d or placebo for an additional 5 years (Figure 1). All patients received 500 mg/d of calcium and 250 IU/d vitamin D. A total of 1099 women were enrolled in the extension. The mean age of FLEX participants was 73 years, 34% had a prevalent vertebral fracture, and 60% had a history of clinical fracture since menopause. Mean T scores were –1.9, –2.2, and –1.3 at the total hip, femoral neck, and lumbar spine, respectively.
BMD For women who were randomized to continue on alendronate, there was a mean total hip BMD decline of –1.02% compared with a –3.38% decrease with placebo, a mean difference of þ2.36% (P o 0.001). The femoral neck and trochanter BMD also displayed losses between alendronate and placebo, with mean loss differences of þ1.94% and þ3.17%, respectively
Alendronate Alendronate has been approved by the FDA for the treatment and prevention of postmenopausal osteoporosis, treatment to increase bone mass in men, and the treatment of glucocorticoid osteoporosis.6 The pivotal Phase III trial FIT (Fracture Intervention Trial)23 reported efficacy in reducing fractures, both clinical and morphometric. Alendronate is available as 5- and 10-mg daily doses and 35- and 70-mg weekly doses, with higher doses approved for the treatment indication.6 To assess long-term efficacy, the FLEX (FIT Long-Term Extension) study16 was conducted to compare the effects of discontinuing alendronate treatment after 5 years versus continuing for 10 years. The primary end point was total hip BMD. The FLEX study was an extension of the pivotal FIT trial.16 In the FIT trial, postmenopausal women aged 55 to 81 years with low femoral neck BMD were
July 2015
Placebo
Placebo Alendronate 5mg or 10mg/day
0
Alendronate 5mg or 10mg/day
5 Year
10
Figure 1. Designs of the FIT (Fracture Intervention Trial)23 and FLEX (FIT Long-Term Extension)16 studies.
1575
Clinical Therapeutics (both, P o 0.001). In contrast, patients randomized to continue on alendronate saw a greater increase in lumbar spine BMD than did the patients receiving placebo, with a mean difference of þ3.74% (P o 0.001) (Table I). In the women who received alendronate in both FIT and FLEX, there was a greater increase in total hip BMD, þ2.41%, compared with that in the subjects who were switched to placebo in FLEX, who saw a decrease of –0.16% (mean difference, þ2.57%; P o 0.001) (data not shown).
vertebral fractures were less in the continued alendronate group compared with those in the placebo group.
Tolerability There were no significant between-group differences in the prevalence of serious adverse events. With regard to adverse events of interest, there were no differences in the rates of upper gastrointestinal tract events, and no cases of ONJ were observed.
Risedronate
BTMs For subjects who continued on alendronate, BTMs— serum C-terminal telopeptide of type 1 collagen (sCTX), serum N-terminal telopeptide (sNTX), both markers of resorption, and serum bone-specific alkaline phosphate (sBSAP), a marker of formation— remained near FLEX baseline values. Patients who received placebo in the extension had gradual increases in mean BTM values over the 5 years, but values were still below FIT baseline levels.
Fracture Post hoc analyses demonstrated that there were no significant differences between treatment groups in all clinical fractures, nonvertebral fractures, or morphometric vertebral fractures. However, women receiving continued alendronate had a significantly lower percentage of clinical vertebral fractures than did women receiving placebo (2.4% vs 5.3%; 95% CI, 0.24–0.85). Although not a statistically significant finding, in a subgroup with lower baseline BMD and a prevalent fracture, the prevalence of nonvertebral and clinical
Risedronate has been approved by the FDA for the treatment and prevention of osteoporosis in postmenopausal women, as treatment to increase bone mass in men with osteoporosis, and as treatment and prevention of glucocorticoid-induced osteoporosis in men and women who are either initiating or continuing systemic glucocorticoid treatment (Z7.5 mg/d of prednisone or equivalent) for chronic diseases.7 Pivotal Phase III trials have reported efficacy in reducing vertebral and nonvertebral fractures.24 Risedronate is available as a daily (5 mg), weekly (35 mg), or monthly (75 mg taken 2 days consecutively, or 150 mg) treatment. To assess the resolution of effect with risedronate therapy, the pivotal Phase III, 3-year fracture study VERT-NA (Vertebral Efficacy with Risedronate Trial–North America)24 was extended for a 1-year observational period. The primary end points of the extension study were the mean percentage changes from baseline in lumbar spine BMD values at months 36 and 48. To be enrolled in the original VERT-NA study, eligible women were at least 5 years postmenopausal,
Table I. Percentage changes from baseline in bone mineral density with alendronate versus placebo. Site
Pooled Alendronate, Mean (SE) (n ¼ 643)
Total hip Femoral neck Trochanter Lumbar spine
–1.02 þ0.46 –0.08 þ5.26
(0.18) (0.24) (0.22) (0.24)
Placebo, Mean (SE) (n ¼ 428) –3.38 –1.48 –3.25 þ1.52
(0.22) (0.30) (0.27) (0.29)
Treatment Difference, Mean (95% CI) þ2.36 þ1.94 þ3.17 þ3.74
(1.81–2.90)* (1.20–2.68)* (2.49–3.84)* (3.03–4.45)*
Data from the FLEX study.16 P o 0.001.
*
1576
Volume 37 Number 7
M. Murphy-Menezes aged o85 years, and had Z2 vertebral fractures or 1 vertebral fracture and a T score o–2. Women were randomized to receive risedronate 2.5 or 5 mg/d or placebo for 3 years. All women received 1000 mg/d calcium and, if baseline serum 25-hydroxyvitamin D was low, vitamin D supplementation. All women who completed the 3-year treatment period (except 121 women who underwent bone biopsy at the end of the 3-year trial) were invited to participate in a 4th, observational year, in which risedronate and placebo were discontinued but calcium and vitamin D were given as described earlier (Figure 2).25 A total of 759 women were enrolled in the 1-year observation phase.25 The mean age was 68 years. The number of prevalent vertebral fractures was not significantly different between the groups who had been treated with risedronate compared with the group who had been treated with placebo. However, as expected at the fourth-year baseline, the group who had previously been treated with risedronate had higher BMD and lower BTM values than did those who received placebo. At baseline of the initial 3-year VERT-NA trial, all baseline characteristics were balanced.
BMD At the end of the 4th year, lumbar spine BMD was decreased by –0.83% in patients previously treated with risedronate. However, BMD remained higher than at the start of the original trial, VERT-NA, and was significantly higher than that in the previous
Placebo
Off-Treatment
Table II. Percentage changes from baseline in bone mineral density with risedronate versus placebo.
Site Lumbar spine Femoral neck Trochanter
Former Risedronate Group, Mean (SE) (n ¼ 296)
Treatment Difference, Mean (95% CI)
–0.83* –1.23* –1.57*
þ2.60 (1.56–3.65)* þ2.32 (1.40–3.25)* þ3.08 (2.06–4.10)*
Data from the risedronate extension study.25 P o 0.001.
*
placebo group at the end of the 4th year, with a between-group difference of þ2.60% (P o 0.001). At the femoral neck, patients previously treated with risedronate had a BMD decrease of –1.23%, which appeared to approach baseline but was still greater than that in the patients previously treated with placebo, with a mean difference of þ2.32% (P o 0.001) (Table II).
BTMs Urinary NTX, a marker of bone resorption, was measured in a subset of women at the end of year 4. In the patients previously treated with risedronate, there was a significant increase from a median of 30.3 to 50.9 nmol bone collagen equivalents per mmol of creatinine (P o 0.05). Although not significantly different from that with placebo, the NTX level remained significantly lower than the baseline value.
Fracture Risedronate 2.5mg or 5mg/day
0
Off-Treatment
3
4
In the 4th year, 6.5% of the original risedronate group experienced a vertebral fracture, while 11.6% of the original placebo group experienced a vertebral fracture (relative risk reduction, 46%; P ¼ 0.009).
Year
Figure 2. Designs of the VERT-NA (Vertebral Efficacy with Risedronate Trial–North America)24 and 1-year extension25 studies.
July 2015
Tolerability In the year of discontinuation, there were no differences in the prevalence of adverse events in the group previously treated with risedronate compared with the group that previously received placebo.
1577
Clinical Therapeutics In 2008, results from a large-scale administrative database studying the clinical implications of bisphosphonate discontinuation were published.5 Data from a US health care organization covering 25 million people were used for identifying a cohort of new bisphosphonate users. The cohort included women 60 to 78 years of age who were new users of alendronate or risedronate from January 1998 to July 2005, had been compliant with bisphosphonate therapy for at least 2 years, and did not have a claim for hip fracture, malignancy, HIV disease, or Paget disease before initiating therapy with a bisphosphonate. Compliance was defined as a medication possession ratio (MPR) of 66% to 100% at 2 years. Discontinuation was defined as discontinuation for any length of time and qualified by duration of discontinuation. The objective of the retrospective study was to determine the association between bisphosphonate discontinuation and hip fracture among persons who had been compliant with bisphosphonate therapy for at least 2 years. Observation of the cohort began at 2 years. From the database, 9063 women initiating treatment with a bisphosphonate were found to be compliant at 2 years and eligible for the discontinuation analysis. In the 6 months leading up to the initiation of osteoporosis medication, 64% of patients had received a BMD test and 51% had a medical diagnosis of osteoporosis. Seventy-seven percent of women initiated therapy with alendronate and 23% initiated therapy with risedronate. Among the cohort of 9063, there were 71 hip fractures identified. In the cohort of patients who had an MPR of 466% at 2 years, the incidence of hip fracture among patients who continued therapy was 4.67% per 1000 patient-years. In the same cohort, in the patients who discontinued therapy, the incidence of hip fracture was 8.43%. Of importance, although not statistically significant, the hip fracture rate in all subgroups (MPR, 466% at 2 and 3 years, 480% at 2 and 3 years) that discontinued was higher than that in the group that continued therapy. Discontinuation of a bisphosphonate for up to 1 year was not associated with a significant increase in the rate of hip fractures. However, in the group with a MPR 466% at 2 years, discontinuation of 1 year or longer was associated with a 2- to 3-fold increase in the relative risk for hip fracture. In addition, among women with greater compliance, MPR 480%, and a longer duration of preceding bisphosphonate therapy, the incidence rate of hip fracture among women
1578
discontinuing therapy was numerically lower than those in women who were less compliant and in women with a shorter treatment duration.
Zoledronate Zoledronate has been FDA-approved for the treatment and prevention of postmenopausal osteoporosis, as treatment to increase bone mass in men with osteoporosis, and for the treatment and prevention of glucocorticoid-induced osteoporosis.9 The pivotal Phase III study of zoledronate, HORIZON-PFT (Health Outcomes and Reduced Incidence with Zoledronic acid Once Yearly Pivotal Fracture Trial),26 reported the efficacy of zoledronate in reducing spine, hip, and other nonvertebral fractures. The approved doses of zoledronate are 5 mg/y for treatment indications and 5 mg every 2 years for preventive indications.9 To determine the effect of zoledronate beyond 3 years, an extension of HORIZON-PFT was conducted.27 The primary end point of the extension was change in femoral neck BMD at year 6 relative to year 3 (baseline of the extension study). In the original HORIZON-PFT study, postmenopausal women between the ages of 65 and 90 years were eligible for inclusion if they had a BMD T score of r2.5 at the femoral neck with or without evidence of existing vertebral fracture, or a T-score r1.5 with radiologic evidence of Z2 mild vertebral fractures or 1 moderate vertebral fracture. Previous use of oral bisphosphonate agents was allowed, with the duration of the washout period dependent on duration of previous use. Exclusion criteria included use of therapy known to affect bone metabolism. Patients with a serum calcium level 42.75 mmol/L or o2.00 mmol/L were ineligible, as were patients with a calculated creatinine clearance of o30.0 mL/min at either of 2 baseline visits or urine dipstick results of 42 for protein, without evidence of contamination or bacteriuria. A total of 7765 osteoporotic women were assigned to receive zoledronate 5 mg IV or inactive vehicle (placebo) given once per year.26 For the extension, women were eligible if they had received 3 infusions in the parent study. Patients were excluded if they had previous major protocol violations, received placebo in the pivotal study, were aged 493 years, or were taking other medications known to affect bone metabolism.27 A total of 1233 patients were enrolled in the extension and were randomized to receive zoledronate for an additional 3 years
Volume 37 Number 7
M. Murphy-Menezes (Z6 group), or placebo for 3 years (Z3P3 group) (Figure 3). All patients received oral calcium 1000 to 1500 mg/d and vitamin D 400 to 1200 IU/d. At baseline of the extension, the mean age of the women was 75.5 years. Mean femoral neck T score was –2.5, and 60% of patients had Z1 prevalent vertebral fracture. The baseline characteristics were not statistically different between the Z6 and Z3P3 groups.
BMD At the end of the sixth year of the extension, there was a statistically significant difference in the mean change in femoral neck BMD, with an increase of þ0.24% in the Z6 group compared with a decrease of –0.80% in the Z3P3 group (mean difference, þ1.04%; P ¼ 0.0009). At the total hip, there was a loss of –0.36 in the Z6 group compared with a loss of –1.58% in the Z3P3 group (mean difference, þ1.22%; P o 0.0001). At the 1/3 radius, the mean difference was þ0.37, which was not statistically significant. The largest difference in loss was seen at the lumbar spine (mean difference, þ2.03%; P ¼ 0.002) (Table III).
BTMs At the end of the 6th year of the extension, mean serum type 1 procollagen N-terminal (P1NP), a marker of formation normally coupled to CTX, increased by þ33% in the Z3P3 group compared with an increase of þ19% in the Z6 group (difference, 14%; P ¼ 0.0001). The absolute difference in ending
Table III. Percentage changes from baseline in bone mineral density with zoledronate 6 months (Z6; n ¼ 616) versus zoledronate 3 months (Z3P3; n ¼ 617). Site Femoral neck Total hip Lumbar spine Distal radius
Treatment Difference, Mean (95% CI) þ1.04 þ1.22 þ2.03 þ0.37
(0.43 to 1.65)* (0.75 to 1.70)† (0.76 to 3.29)‡ (–0.71 to 1.45)§
Data from the HORIZON extension study.27 P o 0.001. † P o 0.0001. ‡ P ¼ 0.002. § P ¼ 0.50.
*
P1NP at year 6 was 3.0 ng/mL (P ¼ 0.001). In the Z3P3 group, P1NP remained statistically significantly less than pretreatment values. The absolute difference in serum CTX between Z6 and Z3P3 at year 6 was 0.02 (P ¼ 0.45).
Fracture Morphometric vertebral fracture risk was significantly lower in the Z6 group compared with that in the Z3P3 group (n ¼ 14 and 30 respectively; OR ¼ 0.51; P ¼ 0.035). However, there were no differences in the prevalence of nonvertebral, hip, or clinical fractures in patients who continued on zoledronate versus those who were switched to placebo.
Placebo
Tolerability Placebo Zolendronate 5mg IV/year
0
Zolendronate 5mg IV/year 3 Year
6
Figure 3. Designs of the HORIZON-PFT (Health Outcomes and Reduced Incidence with Zoledronic acid Once Yearly Pivotal Fracture Trial)26 and 3-year extension27 studies.
July 2015
There was a statistically significantly greater number of patients with an increase from baseline in serum creatinine 40.5 mg/dL in the Z6 group compared with the Z3 group (n ¼ 18 vs 4, respectively; P ¼ 0.002). Atrial fibrillation was slightly more common in the Z6 group, but the difference was not statistically significant (n ¼ 21 vs 13). In addition, there was a numerically greater number of strokes in the Z6 group than in the Z3P3 group, but the difference was not statistically different (n ¼ 19 vs 9). There was 1 case of ONJ in the Z6 group and no cases of atypical femur fracture.
1579
Clinical Therapeutics
RANKL Inhibitor Receptor activator of nuclear factor-κB ligand (RANKL) is a key mediator of osteoclast formation, function, and survival.28 When RANKL binds to the RANK receptor on the surface of osteoclasts, osteoclasts differentiate and begin to actively resorb bone. In the presence of estrogen, RANKL activity is regulated by osteoprotegrin, a soluble protein, which binds to RANKL, preventing binding to the RANK receptor. This system of physiologic checks and balances results in balanced rates of resorption compared with rates of formation. However, when estrogen decreases, excessive amounts of unopposed RANKL are produced, resulting in a disproportionally high level of osteoclast activity and bone resorption. RANKL inhibitors are not incorporated into the bone, and studies have reported nonlinear pharmacokinetic properties, with rapid resolution of effect.29 Currently there is only 1 RANKL inhibitor on the US market.
Denosumab Denosumab is a human monoclonal antibody against RANKL, FDA-approved for the treatment of postmenopausal women with osteoporosis at high risk for fracture and as treatment to increase bone mass in men with osteoporosis.10 Denosumab is given by subcutaneous injection, 60 mg every 6 months, and pivotal data have suggested reductions in vertebral, nonvertebral, and hip fractures.30 An early Phase III, 24-month trial was conducted to determine the effect of denosumab on BMD.32 At the end of 24 months, that study was extended for an additional 24 months to determine the effects of prior administration of denosumab (or placebo) on BMD, BTMs, and tolerability.32 No singular end point was specified as primary, and all changes in BMD, changes in BTMs, and tolerability were exploratory end points. In the original 24-month BMD parent study, women were eligible if they were ambulatory and postmenopausal, had a lumbar spine T score between –1.0 and –2.5, had no underlying medical condition known to result from abnormal bone metabolism, and had no history of fracture after 25 years of age. Exclusion criteria included the use of concurrent medications known to affect bone metabolism, including oral bisphosphonate agents, for at least 3 years cumulatively. If oral bisphosphonate agents were used for 43 months but r3 years, women were eligible if the last dose was administered at least 1 year
1580
before enrollment. Women who used an oral bisphosphonate agent for r3 months were eligible. A total of 332 patients were enrolled in the original 24-month study and were randomized to receive denosumab 60 mg or inactive vehicle (placebo) every 6 months. All participants were instructed to take Z1000 mg/d of calcium and Z400 IU/d vitamin D. Each participant who completed the on-treatment phase, including the last dose of the scheduled regimen of denosumab or placebo, were eligible for participation in the extension study (Figure 4). Two hundred fifty-six patients were enrolled in the extension arm of the study. During the extension phase, no therapy was given in either group. Patients were instructed to continue supplementation with calcium and vitamin D. The mean age of the patients in the extension study was 59 years, with a mean duration of postmenopause of 10 years and a mean lumbar spine T score of –1.61. Overall, baseline demographic and clinical characteristics were balanced at the beginning of the extension study.
BMD At the end of the original 24-month study, patients treated with denosumab had statistically greater gains in BMD at the lumbar spine, total hip, and 1/3 radius than did patients treated with placebo. At the end of the extension phase, 24 months off therapy, BMD was decreased in both groups. Of particular interest, in women previously treated with denosumab, values of BMD at the lumbar spine, total hip, and 1/3 radius declined, with values reaching near-baseline values. However, BMD values in women previously treated with denosumab remained higher than BMD values in
0
Placebo
Off-Treatment
Denosumab 60mg q 6 months
Off-Treatment
2 Year
4
Figure 4. Designs of the denosumab parent study31 and 24-year extension study.32
Volume 37 Number 7
M. Murphy-Menezes women previously treated with placebo. In the patients previously treated with denosumab, most of the decreases in BMD occurred in months 24 to 36, with changes in BMD in months 36 to 48 appearing similar to those seen in the placebo group (data not available).
BTMs On denosumab discontinuation at month 24, there was a rapid reversal in BTM values. Within 3 months, serum CTX increased to above baseline levels, peaked at 30 months, and returned to baseline levels by month 48. Similar results were seen with P1NP, with BTMs increasing to above baseline levels at month 6, reaching peak levels at month 36, and returning to baseline by month 48. In patients previously treated with denosumab, moderate correlations between month-48 sCTX and P1NP measurements and baseline sCTX and P1NP was demonstrated (r ¼ 0.47 and 0.46, respectively; P o 0.001).
Fracture No clinical vertebral fractures were reported. In the original 24-month study, 4% of patients in the placebo group and 1% in the denosumab group experienced centrally confirmed fractures. In the extension, off-treatment phase, 3% of patients previously treated with placebo and 3% previously treated with denosumab experienced centrally confirmed fractures.
Tolerability The prevalence of adverse and serious adverse events was balanced between groups in the offtreatment phase. With regard to adverse events of interest, the prevalence of rash was 0% in the group previously treated with denosumab and 2% in the group previously receiving placebo. There were no reports of ONJ, no difference in neoplasm, and no notable changes in serum calcium. In a separate Phase II clinical trial of denosumab, the effect of re-treatment after a drug-free, discontinuation period was studied.33 In the original Phase II, 4-year, open-label parent study, postmenopausal women aged r80 years with low bone mass were eligible for participation in the study if they had a BMD T score between –1.8 and –4.0 at the lumbar spine or between –1.8 and –3.5 at the total hip or femoral neck. The parent study included 731 patients with a mean age of 62 years, a mean lumbar spine
July 2015
BMD T score of –2.1, and a mean femoral neck BMD T score of –1.8. In the 4-year parent study, a small group of 41 patients was treated with denosumab 30 mg every 3 months for 24 months. After 24 months, patients were discontinued from therapy for 1 year and then re-treated with denosumab 60 mg every 6 months for an additional year (Figure 5). In the 12 months off treatment, BMD declined to near-baseline values at all sites measured—the lumbar spine, total hip, and onethird radius. On retreatment, BMD rebounded within 12 months, back to BMD levels seen at month 24 before discontinuation (data not available).
Estrogen Hormone therapy (HT) with estrogen mimics a woman’s premenopausal physiology and prevents rapid loss of BMD. Estrogen therapy has been approved by the FDA under many brand names for the prevention of osteoporosis.1 The 6-year EPIC (Early Postmenopausal Intervention Cohort) study35 was designed to evaluate the use of alendronate and estrogen–progestin in the prevention of osteoporosis as well as the effects of discontinuation. A total of 1609 postmenopausal women aged 45 to 60 were enrolled. Women with a prior fracture or who had a lumbar spine BMD o0.8 g/cm2 were excluded. Patients were randomized to receive alendronate, estrogen þ medroxyprogesterone, or placebo for various durations, followed by time off therapy (Figure 6). All patients were instructed to take calcium.
BMD/BTMs BMD decreased at all skeletal sites in women who were switched from alendronate to placebo but remained above baseline. The women who had previously been treated with HT for 4 years had large BMD losses. In the group that stopped taking alendronate, BTM values gradually reverted to pretreatment
Denosumab 30mg q 6 month 0
Denosumab 60mg q 6 month
Off-Treatment
2
3
4
Year
Figure 5. Design of the denosumab 4-year parent study.33
1581
Clinical Therapeutics
Placebo
Placebo
Placebo
Alendronate
Alendronate
Alendronate
Alendronate
Alendronate
Placebo
Alendronate
Placebo
Placebo
HT
HT
Off-Treatment
0
4
2
6
Year
Figure 6. Design of the EPIC (Early Postmenopausal Intervention Cohort) study.34
levels but remained below baseline values. In the group that stopped taking HT, there was a rapid increase in NTX to above that in the placebo group.
Tolerability The prevalence of adverse events related to treatment discontinuation was similar between the groups that received alendronate and placebo. More patients treated with HT reported adverse events than did those treated with alendronate.
Cathepsin K Inhibitor Osteoclasts are involved in bone resorption. Bone resorption is a 3-step process.35 First, osteoclasts secrete acid into the resorption lacunae on the bone surface. Second, in the presence of the resultant acidic environment, osteoclasts produce cathepsin K, which is involved in the degradation of organic matrix. Research has shown that inhibition of cathepsin K can slow bone resorption through suppressed osteoclast activity.35
Participants received vitamin D 56,000 IU/wk and calcium supplementation 500 mg/d if their total dose of calcium intake through normal diet and supplementation was o1000 mg/d. On completion of the 24-month original study, patients were eligible to participate in a 1-year extension if they were in good health based on medical history and physical and laboratory evaluations. Two hundred eighty participants entered the extension phase and were randomized to receive either odanacatib 50 mg/wk or placebo (Figure 7). At baseline of the extension (after 2 years of treatment), the mean age was 64 years, and the mean BMD T scores were –2.2 at the lumbar spine and –1.8 at the femoral neck.
BMD In patients who continued with 50 mg of odanacatib, there was a continued gain in BMD. From the end of year 2 to the end of year 3, there were increases of þ2.3% at the lumbar spine, þ2.4% at the total hip, and þ1.6% at the femoral neck. However, there was no significant change in the third year at the 1/3 radius, which produced a loss of –0.4%. In the patients who discontinued odanacatib treatment at 24 months and received placebo in the third year, BMD values were decreased at all sites and returned to levels not different from baseline (Table IV). After 1 year of discontinuation, patients who were first treated with odanacatib followed by placebo had BMD values that remained þ1.4% higher than baseline at the lumbar spine, þ2.3% higher than baseline at the femoral neck, –0.5% lower than baseline at the total hip, and –2.7% lower than baseline at the
Odanacatib Odanacatib is a cathepsin K inhibitor currently being studied for the treatment of postmenopausal osteoporosis, and its use has not been approved by the FDA.36 A Phase II, 24-month, dose-ranging study was extended for an additional year to determine the effects of continued odanacatib treatment versus discontinuation of treatment on BMD, BTMs, and tolerability.35 In the original Phase II, 24-month study, postmenopausal women between the ages of 45 and 85 years were eligible for the study if they had a BMD T score between –2.0 and –3.5.37 In this dose-ranging study, 399 women were randomized to receive weekly doses of odanacatib at 3, 10, 25, or 50 mg or placebo.
1582
Placebo Placebo
Odanacatib 50mg per week
Placebo
Odanacatib 3mg, 10mg, 25mg, or 50mg per week 0
Odanacatib 50mg per week 2
3
Figure 7. Designs of the odanacatib parent37 and 3-year extension35 studies.
Volume 37 Number 7
M. Murphy-Menezes
Table IV. Percentage changes from baseline in bone mineral density with odanacatib versus placebo. Site Lumbar spine Total hip Trochanter Femoral neck 1/3 Radius
Treatment Difference, Mean (95% CI) þ1.4 –0.5 –0.7 þ2.3 –2.7
(–0.8 to 3.6) (–2.6 to 1.6) (–3.8 to 2.5) (0.2 to 4.3) (–5.2 to –0.2)
Data from the odanacatib extension study.36
one-third radius. The rate of bone loss was fastest in the 6 months after discontinuation.
drug holiday may be considered after 4 to 5 years of stability, whereas in patients who have a high risk for fracture, a holiday of 1 to 2 years may be considered after 10 years of treatment. A review of evidence within the guidelines suggested that the appropriate duration of holiday may be different with each therapy. Whereas a holiday of 1 to 2 years may be justified with alendronate due to the high affinity for bone, a holiday of only 1 year may be appropriate with risedronate. The guidelines go on to say that residual effects should not be expected with therapies such as denosumab, raloxifene, and estrogen and therefore a drug holiday may not be appropriate.
National Osteoporosis Foundation In 2014, the NOF updated their treatment guidelines with recommendations similar to those of the AACE regarding durations of therapy and drug holidays.1
BTMs In the patients who continued treatment with odanacatib 50 mg, the changes in markers of resorption from year 2 to 3 were not significant. In the group that was discontinued from treatment at the end of 2 years, rapid increases in bone resorption markers to above baseline were observed at 30 months, with values returning to near-baseline at the end of year 3. With regard to markers of formation, in the group that continued treatment with odanacatib, sBSAP and serum P1NP decreased initially with treatment initiation, followed by a plateau, and then a subsequent increase to above (sBSAP) or near (P1NP) baseline values by the end of year 3.
Tolerability Overall in year 3, the prevalence of adverse events was balanced between the continued-treatment arm versus the treatment-discontinuation arm. However, in year 3, there was an imbalance in the number of events reported and treated as urinary tract infections or cystitis (12 in the active-treatment group, 3 in the placebo group); all were resolved with antibiotic treatment. No differences in the prevalence of skin disorders or in calcium levels were reported.
Review of Guidelines American Association of Clinical Endocrinologists The AACE’s 2010 clinical practice recommendations state that if osteoporosis is considered mild, a
July 2015
DISCUSSION Osteoporosis is a chronic disease state that, if left untreated, can result in fracture, leading to severe disability, loss of mobility, and decreased survival.1,2 Although the effectiveness of products marketed for the treatment of osteoporosis has been shown to decrease the risk for fracture, postmarketing surveillance has demonstrated very rare but serious adverse events.13,17,38 This tolerability concern has led health care professionals and even some patients to question the appropriate duration of therapy.12–15,19–22,39 The evidence reviewed in the present article shows that each therapy has a unique profile that, when matched with a patient’s risk factors, may justify different and patient-specific durations of therapy. Pharmacists are members of the health care team charged with the responsibility of drug therapy outcomes.21 As a result, in recent years there has been an increase in formalized medication therapy management (MTM) services, some of which are dedicated primarily to monitoring and improving compliance and adherence.18,21 As such, it is important that pharmacists understand current treatment guidelines and are aware of the changes that are introduced as a result of new data. As pharmacists perform medication reviews, they must not mistake a patient’s sudden discontinuation of osteoporosis therapy as nonadherence, which could result in inappropriate counseling. With an understanding of a patient’s risk factors and
1583
Clinical Therapeutics previous duration of therapy, a pharmacist can determine whether a drug holiday was likely ordered. In addition, as pharmacists perform medication reviews, they may be presented with the unique opportunity to determine when a holiday may be appropriate by again examining an individual’s previous duration of therapy and risk factors. The potential cost-savings from discontinuing unneeded therapy or from preventing an excessive treatment duration should not be underestimated. A holiday is a break with a specific start and end. A drug holiday with osteoporosis therapy is no different. In fact, the AACE and NOF guidelines recommend that patients be monitored during the holiday period, and if BMD declines substantially, BTM values increase, or fracture occurs, therapy should be restarted.1,4 This need for monitoring patients again presents a unique opportunity for pharmacists providing MTM services. Once a patient’s physician recommends an osteoporosis treatment
holiday, a pharmacist may create a reminder to reevaluate the patient in 1 to 2 years.1,4 In addition to BMD, BTMs, and fracture monitoring, pharmacists can easily reassess a patient’s risk by using the Fracture Risk Assessment Tool (FRAX), an online instrument designed for calculating a patient’s 10-year risk for hip or other major osteoporotic fracture (clinical vertebral, forearm, or proximal humerus) based on individual clinical risk factors.40 Treatment guidelines recommend therapy in patients with a low bone mass (T score of –1.5 to –2.5) and a 10-year risk for hip fracture of Z3% or a 10-year risk for major osteoporotic fracture of Z20%. Although FRAX was originally designed for use in treatment-naive patients, current NOF guidelines state that patients who have been off of osteoporosis medications for 1 to 2 years may be considered treatment naive and therefore may be appropriate for reassessment with FRAX.1 By using the FRAX tool, pharmacists have the opportunity to determine and recommend to treating physicians
Patient is low risk on an oral bisphosphonate x 5 years or Patient is high risk and on an oral bisphosphonate x 10 years
T score ≤ –2.5; Continue Therapy; DEXA q 1–2 years
No history of Nontraumatic Fracture
History of Nontraumatic Fracture or other High Risk Factors
T-score >–2.5; Consider Drug Holiday
T-score >–2.5; Consider Drug Holiday
DEXA q 1–2 years Consider BTM and FRAX
DEXA q1–2 years Consider BTM and FRAX
If experience fracture, BMD declines, and/or FRAX >3% at hip or >20 major OP fracture:
If experience fracture, BMD declines, and/or FRAX >3% at hip or >20 major OP fracture:
Resume Therapy
Resume Therapy
T-score ≤–2.5; Continue Therapy; DEXA q1–2 years
Figure 8. Example of proposed algorithm for selection of candidates for bisphosphonate drug holiday and principles of monitoring.39 This algorithm has been modified from its original form to incorporate recommendations from the 2014 National Osteoporosis Foundation guideline.1
1584
Volume 37 Number 7
M. Murphy-Menezes when extending a holiday may be appropriate or when the holiday should be ended and therapy reinitiated. Pharmacists who are involved in MTM services may benefit from developing treatment algorithms for reference. Figure 8 provides a modified example of such an algorithm that has been proposed by bone health specialists.41
CONCLUSIONS Treatment of chronic disease states, including osteoporosis, requires careful monitoring. The risk versus benefit of a therapy must be continually evaluated to determine the appropriateness of continued treatment. This type of management presents a unique opportunity for pharmacists whose professional responsibility includes managing drug therapy outcomes.
CONFLICTS OF INTEREST The author has indicated that she has no conflicts of interest with regard to the content of this article.
REFERENCES 1. National Osteoporosis Foundation. Clinician’s Guide to Prevention and Treatment of Osteoporosis. Washington, DC: National Osteoporosis Foundation; 2014. 2. Schnell S, Friedman S, Mendelson D, et al. The 1-year mortality of patients treated in a hip fracture program for elders. Geriatr Orthop Surg Rehab. 2010;1:6–14. 3. Burge R, Dawson-Hughes B, Soloman D, et al. Incidence and Economic Burden of Osteoporosis-Related Fractures in the United States, 2005–2025. J Bone Miner Res. 2007;22:465–475. 4. Watts N, Bilezikian J, Camacho P. American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the diagnosis and treatment of postmenopausal osteoporosis. Endocr Pract. 2010;16(Suppl 3): 1–37. 5. Curtis J, Westfall A, Cheng J, et al. Risk of hip fracture after bisphosphonate discontinuation: implications for a drug holiday. Osteoporos Int. 2008;19:1613–1620. 6. Fosamax [package insert]. Whitehouse Station, NJ: Merck; 2014. 7. Actonel [package insert]. Rockaway, NJ: Warner Chilcot; 2013. 8. Boniva [package insert]. South San Francisco, CA: Genentech; 2013. 9. Reclast [package insert]. East Hanover, NJ: Novartis; 2013. 10. Prolia [package insert]. Thousand Oaks, CA: Amgen; 2014.
July 2015
11. Silverman S, Schousboe B, Gold D. Oral bisphosphonate compliance and persistence: A matter of choice? 2011;22:21-26. 12. National Osteoporosis Foundation. Clinician’s Guide to Prevention and Treatment of Osteoporosis. Washington, DC: National Osteoporosis Foundation; 2005. 13. Whitaker M, Guo J, Kehoe T, et al. Bisphosphonate agents for osteoporosis-where do we go from here? N Engl J Med. 2012;366:2051–2053. 14. McClung M. Controversies in osteoporosis management: concerns about bisphosphonate agents and when are “drug holidays” required? Clin Obstet Gynecol. 2013;56: 743–748. 15. Sebba A. Osteoporosis: how long should we treat? Curr Opin Endocrinol Diabetes Obes. 2008;15:502–507. 16. Black D, Schwatrz A, Ensrud K, et al. Effects of continuing or stopping alendronate after 5 years of treatment. The Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial. JAMA. 2006;296:2927–2937. 17. Khosla S, Burr D, Cauley J, et al. Bisphosphonateassociated osteonecrosis of the jaw: report of a task force of the American Society for Bone and Mineral Research. J Bone Miner Res. 2007;22:1479–1491. 18. Stuurman-Bieze AG, Hiddink EG, van Boven JF, Vegter S. Proactive pharmaceutical care interventions decrease patients’ nonadherence to osteoporosis medication. Osteoporos Intl. 2014;25:1807–1812. 19. McClung M, Harris S, Miller P, et al. Bisphosphonate therapy for osteoporosis: benefits risks, and drug holiday. Am J Med. 2013;126:13–20. 20. Watts N, Diab D. Long-term use of bisphosphonate agents in osteoporosis. J Clin Endocrinol Metab. 2010;95: 1555–1565. 21. Figge H. Enabling medication therapy management. US Pharmacist. 2008:33. [Epub ahead of print]. 22. Dunn R, Bird M, Conway S, Stratton M. Use of bisphosphonate agents in older adults: how long is long enough? Consult Pharm. 2013;28:39–57. 23. Cummings S, Black D, Thompson D, et al. Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. JAMA. 1998;280:2077–2082. 24. Harris S, Watts N, Genant H, et al. Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis: a randomized controlled trial. Vertebral Efficacy with Risedronate Therapy (VERT) study group. JAMA. 1999;282:1344–1352. 25. Watts N, Chines A, Olszynski W, et al. Fracture risk remains reduced one year after discontinuation of risedronate. Osteoporos Int. 2008;19:365–372. 26. Black D, Delmas D, Eastell R, et al. Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. N Engl J Med. 2007;356:1809–1822.
1585
Clinical Therapeutics 27. Black D, Reid I, Boonen S, et al. The effect of 3 years versus 6 years of zoledronic acid treatment of osteoporosis: a randomized extension to the HORIZON-Pivotal Fracture Trial (PFT). J Bone Miner Res. 2012;27:243–254. 28. Kearns A, Khosla S, Kostenuik PJ. Receptor activator of nuclear factor kappaB ligand and osteoprotegerin regulation of bone remodeling in health and disease. Endocr Rev. 2008;29:155–192. 29. Bekker PJ, Holloway DL, Rasmussen AS, et al. A single-dose placebocontrolled study of AMG 162, a fully human monoclonal antibody to RANKL, in postmenopausal women. J Bone Miner Res. 2004: 1059–1066. 30. Cummings S, San Martin J, McClung M, et al. Denosumab for prevention of fractures in postmenopausal women with osteoporosis. N Engl J Med. 2009;361:756–765. 31. Bone HG, Bolognese MA, Yuen CK, et al. Effects of denosumab on bone mineral density and bone turnover in postmenopausal women. J Clin Endocrinol Metab. 93:2149– 2157. 32. Bone H, Bolognese M, Yuen C, et al. Effects of denosumab treatment and discontinuation on bone mineral density and bone turnover markers in postmenopausal women with low bone mass. J Clin Endocrinol Metab. 2001;96:972–980. 33. Miller P, Wagman R, Peacock M, et al. Effect of denosumab on bone mineral density and biochemical markers of bone turnover: six-year results of a phase 2 clinical trial. J Clin Endocrinol Metab. 2001;92: 394–402. 34. Wasnich R, Bagger y, Hosking D, et al. Changes in bone density and turnover after alendronate or estrogen withdrawal. Menopause. 2004; 11:622–630. 35. Eisman J, Bone H, Hosking D, et al. Odanacatib in the treatment of postmenopausal women with
1586
low bone mineral density: threeyear continued therapy and resolution of effect. J Bone Miner Res. 2011;26:242–251. 36. Lewiecki M. Odanacatib, a cathepsin K inhibitor for the treatment of osteoporosis and other skeletal disorders associated with excessive bone remodeling. IDrugs. 2009;12: 799–809. 37. Bone HG, McClung MR, Roux C, et al. Odanacatib, a cathepsin-k inhibitor for osteoporosis: a twoyear study in postmenopausal women with low bone density. J Bone Miner Res. 2010;25:937–947. 38. Shane E, Burr D, Ebeling P, et al. Atypical subtrochanteric and diaphyseal femoral fractures: report of a task force of the American Society for Bone and Mineral Research.
J Bone Miner Res. 2010;25:2267– 2294. 39. How long should you take certain osteoporosis drugs? FDA Consumer Health Information. May 2014. Retrieved from http://www. fda.gov/ForConsumers/Consumer Updates/ucm309688.htm. Accessed September, 2014. 40. Kanis JA on behalf of the World Health Organization Scientific Group. Assessment of osteoporosis at the primary health care level. Technical Report. World Health Organization Collaborating Center for Metabolic Bone Diseases. University of Sheffield, UK. 2007. 41. Ro C, Cooper O. Bisphosphonate drug holiday: choosing appropriate candidates. Curr Osteoporos Rep. 2013;11:45–51.
Address correspondence to: Megan Murphy-Menezes, University of Florida College of Pharmacy, PO Box 113195, Gainesville, FL 32611. E-mail: [email protected]fl.edu
Volume 37 Number 7
