Advertisement

Drug holidays from bisphosphonates and denosumab in postmenopausal osteoporosis: EMAS position statement

      Highlights

      • Osteonecrosis of the jaw (ONJ) and atypical femoral fractures (AFF) are very rare adverse events in bisphosphonate users and even more rare in denosumab users.
      • Extending bisphosphonate treatment beyond 3–5 years does not confer additional benefit in low-risk populations.
      • Treatment re-initiation (usually 1–3 years after bisphosphonate withdrawal) depends on risk factors, new fractures and bone mineral density.
      • The evidence regarding denosumab discontinuation is limited but caution is advised, as there may be a “rebound effect” with regard to fractures.

      Abstract

      Background

      Bisphosphonates and denosumab are used extensively in the treatment of postmenopausal osteoporosis. Despite their proven efficacy in the reduction of vertebral and non-vertebral fractures, their optimal duration of use has not been determined. The occurrence of adverse effects, such as osteonecrosis of the jaw (ONJ) and atypical femoral fractures (AFF), has raised the issue of bisphosphonate or denosumab discontinuation (“drug holiday”) after a certain treatment period.

      Aim

      To assess the effect of bisphosphonate and denosumab discontinuation on fracture risk, as well as its possible benefits in reducing the risk of adverse effects.

      Methods

      Systematic review and consensus of expert opinion.

      Results and conclusions

      Discontinuation of bisphosphonates should be considered in all patients who have been treated for more than five years with alendronate, risedronate or zoledronic acid. In view of the limited evidence, no robust recommendations can be made for ibandronate and denosumab. If the patient has not experienced fractures before or during therapy and the fracture risk is low, a “drug holiday” can be recommended. Although there is no solid evidence, 1–2 years for risedronate, 3–5 years for alendronate and 3–6 years for zoledronic acid are suggested. After this time, the patient should be reassessed. If a new fracture is experienced, or fracture risk has increased or BMD remains low (femoral neck T-score ≤−2.5), anti-osteoporotic treatment should be resumed. In the case of denosumab discontinuation, close monitoring is suggested, due to the possibility of rebound fractures.

      Keywords

      1. Introduction

      Bisphosphonates are structural analogues of inorganic pyrophosphate, where the oxygen atom has been substituted by a carbon atom. Differences in the R2 side-chain bound to the carbon atom and the nitrogen group determine their variations in duration of action, bone affinity and anti-fracture efficacy [
      • Russell R.G.
      • Watts N.B.
      • Ebetino F.H.
      • Rogers M.J.
      Mechanisms of action of bisphosphonates: similarities and differences and their potential influence on clinical efficacy.
      ,
      • Khosla S.
      • Bilezikian J.P.
      • Dempster D.W.
      • et al.
      Benefits and risks of bisphosphonate therapy for osteoporosis.
      ]. Bisphosphonates inhibit enzymes involved in osteoclastic activity, and thus suppress bone resorption [
      • Russell R.G.
      • Watts N.B.
      • Ebetino F.H.
      • Rogers M.J.
      Mechanisms of action of bisphosphonates: similarities and differences and their potential influence on clinical efficacy.
      ,
      • Khosla S.
      • Bilezikian J.P.
      • Dempster D.W.
      • et al.
      Benefits and risks of bisphosphonate therapy for osteoporosis.
      ]. The main bisphosphonates are alendronate, risedronate, ibandronate and zoledronic acid, which constitute the first-line therapeutic agents in both postmenopausal and male osteoporosis, as they have well-documented anti-fracture efficacy [
      • Russell R.G.
      • Watts N.B.
      • Ebetino F.H.
      • Rogers M.J.
      Mechanisms of action of bisphosphonates: similarities and differences and their potential influence on clinical efficacy.
      ,
      • Khosla S.
      • Bilezikian J.P.
      • Dempster D.W.
      • et al.
      Benefits and risks of bisphosphonate therapy for osteoporosis.
      ]. Although the different bisphosphonates have not been directly compared, they all reduce the risk of vertebral fractures (38–65% compared with placebo), with the greatest effect seen with zoledronic acid. Excluding ibandronate, bisphosphonates are also effective in reducing the risk of non-vertebral and hip fractures, by 22–31% and 50–55%, respectively, compared with placebo [
      • Murad M.H.
      • Drake M.T.
      • Mullan R.J.
      • Mauck K.F.
      • Stuart L.M.
      • Lane M.A.
      • et al.
      Clinical review. comparative effectiveness of drug treatments to prevent fragility fractures: a systematic review and network meta-analysis.
      ].
      Denosumab, a human monoclonal antibody, an inhibitor of the receptor-activator of nuclear factor kappaB ligand (RANKL), is also a very effective anti-resorptive compound. Its use has been associated with a significant reduction in the risk of vertebral, non-vertebral and hip fractures (by 68%, 20% and 40%, respectively) [
      • Cummings S.R.
      • San Martin J.
      • McClung M.R.
      • Siris E.S.
      • Eastell R.
      • Reid I.R.
      • Delmas P.
      • Zoog H.B.
      • Austin M.
      • Wang A.
      • Kutilek S.
      • Adami S.
      • Zanchetta J.
      • Libanati C.
      • Siddhanti S.
      • Christiansen C.
      FREEDOM trial, denosumab for prevention of fractures in postmenopausal women with osteoporosis.
      ].
      Despite their role in reducing fracture risk, bisphosphonates have been associated with adverse outcomes. The most common include gastrointestinal adverse effects (esophageal ulcer and esophagitis, which have been associated with oral compounds), acute phase reaction (mostly seen after the first infusion of i.v. bisphosphonates) and atrial fibrillation (rarely reported with i.v. bisphosphonates) [
      • Khosla S.
      • Bilezikian J.P.
      • Dempster D.W.
      • et al.
      Benefits and risks of bisphosphonate therapy for osteoporosis.
      ]. Moreover, increased concern has emerged due to major adverse effects, such asosteonecrosis of the jaw (ONJ) and atypical femoral fractures (AFF), the occurrence of which seems to be associated with long-term bisphosphonate use. ONJ is defined as exposed bone in the oral cavity that does not heal within eight weeks of diagnosis, with a history of prior treatment with an anti-resorptive agent (bisphosphonates or denosumab), without a history of craniofacial radiation [
      • Khan A.A.
      • Morrison A.
      • Hanley D.A.
      • Felsenberg D.
      • McCauley L.K.
      • O'Ryan F.J.
      • et al.
      Diagnosis and management of osteonecrosis of the jaw: a systematic review and international consensus.
      ]. ONJ is mostly seen in patients with a history of cancer undergoing treatment with high doses of i.v. bisphosphonates (incidence 1–15%), whereas it is very rarely seen in populations treated for osteoporosis (incidence 0.001–0.04%) [
      • Khan A.A.
      • Morrison A.
      • Hanley D.A.
      • Felsenberg D.
      • McCauley L.K.
      • O'Ryan F.J.
      • et al.
      Diagnosis and management of osteonecrosis of the jaw: a systematic review and international consensus.
      ]. The risk of ONJ increases with the duration of exposure to bisphosphonates; the risk for non-cancer patients increases substantially after five years of use [
      • Barasch A.
      • Cunha-Cruz J.
      • Curro F.A.
      • Hujoel P.
      • Sung A.H.
      • Vena D.
      • et al.
      Risk factors for osteonecrosis of the jaws: a case-control study from the CONDOR dental PBRN.
      ]. AFF are defined as atraumatic or low-trauma fractures located in the subtrochanteric or diaphyseal femoral region. They are usually non-comminuted and have a transverse or short oblique configuration [
      • Shane E.
      • Burr D.
      • Ebeling P.R.
      • Abrahamsen B.
      • et al.
      Atypical subtrochanteric and diaphyseal femoral fractures: a second report of a task force of the American Society for Bone and Mineral Research.
      ]. In general, bisphosphonate use is associated with a two-fold increase in the risk of AFF. This risk increases further when bisphosphonate use exceeds five years [
      • Gedmintas L.
      • Solomon D.H.
      • Kim S.C.
      Bisphosphonates and risk of subtrochanteric, femoral shaft, and atypical femur fracture: a systematic review and meta-analysis.
      ,
      • Lee S.
      • Yin R.V.
      • Hirpara H.
      • et al.
      Increased risk for atypical fractures associated with bisphosphonate use.
      ]. Nevertheless, it must be emphasized that the absolute risk of AFF in patients treated with bisphosphonates is very low (from 3.2 to 50 cases per 100,000 person-years, reaching 113 per 100,000 person-years aftermore than eight years of use) [
      • Meier R.P.H.
      • Perneger T.V.
      • Stern R.
      • Rizzoli R.
      • Peter R.E.
      Increasing occurrence of atypical femoral fractures associated with bisphosphonate use.
      ]. ONJ and, even more rarely, AFF have also been associated with the use of denosumab [
      • Khan A.A.
      • Morrison A.
      • Hanley D.A.
      • Felsenberg D.
      • McCauley L.K.
      • O'Ryan F.J.
      • et al.
      Diagnosis and management of osteonecrosis of the jaw: a systematic review and international consensus.
      ,
      • Papapoulos S.
      • Lippuner K.
      • Roux C.
      • Lin C.J.
      • Kendler D.L.
      • Lewiecki E.M.
      • Bone M.L.
      • et al.
      The effect of 8 or 5 years of denosumab treatment in postmenopausal women with osteoporosis: results from the FREEDOM Extension study.
      ].
      Taking into account the long skeletal retention time of bisphosphonates and the concern about denosumab, a “drug holiday” has been proposed as a means of reducing the risk of both ONJ and AFF. Although other factors may contribute to the occurrence of ONJ and AFF (such as poor oral health, invasive dental surgery, glucocorticoid use, diabetes and anemia for ONJ, and lower limb and hip geometry for AFF) the duration of bisphosphonate and, very probably, denosumab use with consequent long-term suppression of bone turnover seem to play a significant role [
      • Shane E.
      • Burr D.
      • Ebeling P.R.
      • Abrahamsen B.
      • et al.
      Atypical subtrochanteric and diaphyseal femoral fractures: a second report of a task force of the American Society for Bone and Mineral Research.
      ]. A population-based case-control study showed that the risk of AFF significantlydecreases after bisphosphonate withdrawal, irrespective of the total duration of prescription [
      • Schilcher J.
      • Michaëlsson K.
      • Aspenberg P.
      Bisphosphonate use and atypical fractures of the femoral shaft.
      ]. The effect of bisphosphonate or denosumab discontinuation on the risk of ONJ is currently unknown.
      This position statement presents a systematic analysis of the data on the effect of bisphosphonate and denosumab discontinuation on fracture risk, as well as its possible benefits inreducing the risk of adverse effects such as AFF and ONJ. It also provides recommendations on “drug holidays”, specifically (i) the types of patient most appropriate for a drug holiday, (ii) the optimal length of a drug holiday and (iii) the timing of possible re-initiation of treatment.

      2. Methods

      2.1 Trial selection

      The present review included randomized controlled trials (RCTs) which investigated the effect of bisphosphonate or denosumab discontinuation (for each drug separately) on bone mineral density (BMD), bone turnover markers (BTM) and clinical or morphometric vertebral and/or non-vertebral fractures (where available) in postmenopausal women or men aged over 50 years diagnosed with osteoporosis. Studies were excluded if: (i) they were non-randomized, (ii) the bisphosphonate discontinuation period was less than one year, (iii) they had no control group, (iv) they were not conducted in humans, (v) another anti-osteoporosis medication was started after discontinuationof the bisphosphonate, (vi) they were conducted in oncology patients or in those with other metabolic bone disease, such as Paget’s disease of bone, or in patients in receipt of glucocorticoid treatment.

      2.2 Search strategy

      The MEDLINE, Scopus, EMBASE and Cochrane databases were searched for RCTs evaluating the effect of bisphosphonate or denosumab discontinuation on BMD, bone turnover and fractures. The search was carried out up to January 31, 2017. A basic search strategy was developed for MEDLINE and modified accordingly for the other search engines. The search items were: (“alendronate” OR “risedronate” OR “zoledronic acid” OR “zoledronate” OR “denosumab”) AND [“discontinuation” OR “duration” OR “holiday” OR “stop” OR “withdrawal”]. The final set of abstracts was uploaded at the web version of the EndNote software (www.myendnoteweb.com), after removing duplicates. Selection of abstracts, extraction of information from full-text articles andevaluation of risk of bias with the Newcastle-Ottawa Scale (NOS) were done independently by three researchers (PA, SAP and GM). Discrepancies were resolved bydiscussion and if needed with the help of an additional researcher (DGG).

      2.3 Data extraction

      Three independent researchers reviewed all eligible studies. The following data were extracted and recorded: (i) first author, (ii) year of publication, (iii) country in which the study was conducted, (iv) study duration, (v) duration of “drug holiday”, (vi) dose of bisphosphonate or denosumab and (vii) for each group (intervention and comparator) the number of participants, their age, the effect of bisphosphonate or denosumab discontinuation (“drug holiday”) and continuation on lumbar and femoral BMD, bone turnover, and vertebral and non-vertebral fractures.

      2.4 Reporting guidelines

      This systematic review followed the PRISMA (preferred reporting items for systematic reviews and meta-analyses) guidelines [
      • Liberati A.
      • Altman D.G.
      • Tetzlaff J.
      • Mulrow C.
      • Gøtzsche P.C.
      • Ioannidis J.P.
      • et al.
      The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration.
      ].

      3. Data analysis

      3.1 Data presentation

      3.1.1 Alendronate

      The initial search provided 834 results, after excluding duplicates, 13 of which were reviewed by title and abstract (Fig. 1). Of those, eight articles were excluded due to: non-randomized design of the reported study, no full-text availability, concomitant use of another anti-resorptive drug, patients being on glucocorticoids, the study investigating the same population as another study included in the review and the study investigating an oncology population. Finally, five full-text articles were included in the qualitative analysis [
      • Black D.M.
      • Schwartz A.V.
      • Ensrud K.E.
      • et al.
      Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial.; FLEX research group.
      ,
      • Bone H.G.
      • Hosking D.
      • Devogelaer J.P.
      • Tucci J.R.
      • Emkey R.D.
      • Tonino R.P.
      • et al.
      Ten years' experience with alendronate for osteoporosis in postmenopausal women.
      ,
      • Ravn P.
      • Weiss S.R.
      • Rodriguez-Portales J.A.
      • McClung M.R.
      • Wasnich R.D.
      • Gilchrist N.L.J.
      • et al.
      Alendronate in early postmenopausal women: effects on bone mass during long-term treatment and after withdrawal. Alendronate osteoporosis prevention study group.
      ,
      • Bagger Y.
      • Tankó L.B.
      • Alexandersen P.
      • Ravn P.
      • Christiansen C.
      Alendronate has a residual effect on bone mass in postmenopausal Danish women up to 7 years after treatment withdrawal.
      ,
      • Wasnich R.D.
      • Bagger Y.Z.
      • Hosking D.J.
      • McClung M.R.
      • Wu M.
      • Mantz A.M.
      • Yates J.J.
      • et al.
      Early Postmenopausal Intervention Cohort Study Group Changes in bone density and turnover after alendronate or estrogen withdrawal.
      ]. A post-hoc analysis [
      • Schwartz A.V.
      • Bauer D.C.
      • Cummings S.R.
      • et al.
      FLEX Research Group, Efficacy of continued alendronate for fractures in women with and without prevalent vertebral fracture: the FLEX trial.
      ] of the same population from one of these studies [
      • Black D.M.
      • Schwartz A.V.
      • Ensrud K.E.
      • et al.
      Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial.; FLEX research group.
      ] is also discussed.
      Fig. 1
      Fig. 1Flowchart showing the selection strategy of the studies included in the systematic review [
      • Liberati A.
      • Altman D.G.
      • Tetzlaff J.
      • Mulrow C.
      • Gøtzsche P.C.
      • Ioannidis J.P.
      • et al.
      The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration.
      ].
      The hallmark study on an “alendronate holiday” was the Fracture Intervention Trial Long-term Extension (FLEX) study, in which 1099 postmenopausal women treated with alendronate (5 mg/d for two years and then 10 mg/d for another three years) were randomized to continuing treatment with alendronate, 5 mg/d (n = 329) or 10 mg/d (n = 333) or placebo (n = 437) for another five years. The FLEX study showed that discontinuation of alendronate resulted in a significant decline in lumbar (−3.74%), total hip (−2.36%) and neck (−1.94%) BMD compared with treatment extension. Serum levels of C-terminal telopeptide of type 1 collagen (CTX), a marker of bone resorption, as well as of N-propeptide of type 1 collagen (P1NP) and bone-specific alkaline phosphatase, both markers of bone formation, increased by 55.6%, 59.5% and 28.1%, respectively, in the placebo group compared with the alendronate groups (no significant difference between alendronate groups was observed regarding these parameters) [
      • Black D.M.
      • Schwartz A.V.
      • Ensrud K.E.
      • et al.
      Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial.; FLEX research group.
      ]. Alendronate continuation for up to ten years led to a 55% decrease in the risk of clinical vertebral fractures [5.3% with placebo versus 2.4% with alendronate; relative risk (RR) 0.45, 95% confidence interval (CI) 0.24–0.85], but there was no significant difference in the incidence of morphometric vertebral fractures (11.3% with placebo versus 9.8% with alendronate; RR 0.86, 95% CI 0.60–1.22). No difference was observed in clinical or morphometric non-vertebral fractures. Of note, those who discontinued therapy did not return to their pretreatment BMD and BTM values [
      • Black D.M.
      • Schwartz A.V.
      • Ensrud K.E.
      • et al.
      Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial.; FLEX research group.
      ].
      A post-hoc analysis of the FLEX data showed that there was an interaction between femoral neck T-score at baseline and alendronate efficacy. In particular, those with T-scores of ≤−2.5 at the femoral neck demonstrated a significant reduction in non-vertebral fractures with alendronate continuation compared with placebo (RR 0.50, 95% CI 0.26–0.96). No benefit was observed in patients with higher T-scores. No interaction was shown across femoral neck T-scores regarding alendronate efficacy on the risk of vertebral fractures. Neither was an interaction detected with respect to lumbar spine T-scores and alendronate efficacy. Older age was independently associated with a greater risk of fracture (relative hazard ratio 1.54 [95% CI, 1.26–1.85] per 5-year increase) [
      • Schwartz A.V.
      • Bauer D.C.
      • Cummings S.R.
      • et al.
      FLEX Research Group, Efficacy of continued alendronate for fractures in women with and without prevalent vertebral fracture: the FLEX trial.
      ].
      In another study [
      • Bone H.G.
      • Hosking D.
      • Devogelaer J.P.
      • Tucci J.R.
      • Emkey R.D.
      • Tonino R.P.
      • et al.
      Ten years' experience with alendronate for osteoporosis in postmenopausal women.
      ], alendronate discontinuation for five years (after 20 mg/d for two and 5 mg/d for three years, followed by five years of placebo, n = 83) was compared with ten years of continuous treatment at a dose of either 10 mg/d (n = 86) or 5 mg/d (n = 78). Compared with baseline, ten-year treatment with alendronate, for the last five years at doses of 10 mg/d or 5 mg/d, led to 13.7% and 9.3% gains in lumbar BMD, respectively. The discontinuation group maintained a 9.3% increase in lumbar BMD compared with baseline. The respective increases in femoral neck BMD were 5.4%, 2.8% and 1.5%. BTM levels remained lower than their premenopausal values in the alendronate group (10 mg/d), but increased in the discontinuation group, although remaining below the baseline values. There were no differences in the incidence of morphometric vertebral fractures between the three groups, a finding that is in line with the results of the FLEX study [
      • Bone H.G.
      • Hosking D.
      • Devogelaer J.P.
      • Tucci J.R.
      • Emkey R.D.
      • Tonino R.P.
      • et al.
      Ten years' experience with alendronate for osteoporosis in postmenopausal women.
      ].
      In a study of early postmenopausal women with normal bone mass, two years of treatment with alendronate 20 mg/d and discontinuation for three years thereafter was equally protective against postmenopausal bone loss as continuous treatment with alendronate 5 mg/d for five years, in spine, femoral and total body BMD. BTM tended to reverse, but their concentrations remained 40–60% lower than their pre-treatment values [
      • Ravn P.
      • Weiss S.R.
      • Rodriguez-Portales J.A.
      • McClung M.R.
      • Wasnich R.D.
      • Gilchrist N.L.J.
      • et al.
      Alendronate in early postmenopausal women: effects on bone mass during long-term treatment and after withdrawal. Alendronate osteoporosis prevention study group.
      ]. No fracture data were available.
      A Danish study of postmenopausal women (n = 108) [
      • Bagger Y.
      • Tankó L.B.
      • Alexandersen P.
      • Ravn P.
      • Christiansen C.
      Alendronate has a residual effect on bone mass in postmenopausal Danish women up to 7 years after treatment withdrawal.
      ] (lumbar spine BMD from −2 to +2 SD of the premenopausal normal value) randomly allocated participants to alendronate (2.5–20 mg/d) for two, four or six years, followed by no treatment for seven, five or three years, respectively. The residual effects (that is, the total gain in BMD after the end of “drug holiday”) on spine BMD after nine years were 3.8%, 5.9% and 8.6% compared with placebo. However, the rate of bone loss was comparable. CTX and osteocalcin concentrations tended to return to pretreatment values, but were still reduced seven years post-withdrawal.
      In another prospective study in early postmenopausal women [
      • Wasnich R.D.
      • Bagger Y.Z.
      • Hosking D.J.
      • McClung M.R.
      • Wu M.
      • Mantz A.M.
      • Yates J.J.
      • et al.
      Early Postmenopausal Intervention Cohort Study Group Changes in bone density and turnover after alendronate or estrogen withdrawal.
      ], continuous alendronate treatment (5 mg/d) for six years (n = 90) was compared with four (n = 86) and two years (n = 94) of treatment followed by placebo for two or four years, respectively. These three groups were also compared with a group who received only placebo for six years (n = 132). Both alendronate discontinuation groups had greater lumbar and total hip BMD values at six years than the placebo group (+2.9% for the two-year and +5.1% for the four-year treatment groupin lumbar BMD), although lower BMD than the group receiving continuous alendronate treatment for six years. Bone resorption remained suppressed, despite discontinuation for two and four years (−50.6% and −44%, respectively) compared with pre-treatment values. The respective change in the group receiving continuous alendronate therapy was −68%.
      No cases of ONJ or AFF were reported in any of the aforementioned alendronate extension trials.
      Regarding male osteoporosis, no study was found that fulfilled the inclusion criteria of this systematic review.

      3.1.2 Zoledronic acid

      The initial search provided 1165 results, after excluding duplicates, three of which were reviewed by title and abstract (Fig. 1). Of those, one article was excluded due to the non-randomized design of the reported study. The remaining two full-text articles were included in the qualitative analysis [
      • Black D.M.
      • Reid I.R.
      • Boonen S.
      • et al.
      The effect of 3 versus 6 years of zoledronic acid treatment of osteoporosis: a randomized extension to the HORIZON-pivotal fracture trial (PFT).
      ,
      • Black D.M.
      • Reid I.R.
      • Cauley J.A.
      • et al.
      The effect of 6 versus 9 years of zoledronic acid treatment in osteoporosis: a randomized second extension to the HORIZON-pivotal fracture trial (PFT).
      ].
      In the extension of the HORIZON-Pivotal Fracture Trial (PFT), treatment with zoledronic acid (5 mg annually) for a total of up to six years (after completion of three annual infusions) in 616 postmenopausal women was compared with no further treatment for three years (after completion of three annual infusions) (n = 617). Extension treatment reduced the risk of new morphometric vertebral fractures by 49% [3.0% in the group undergoing six years of continuous treatment compared with 6.2% in the discontinuation group (OR 0.51, 95% CI 0.26–0.95)]. No significant difference between the two groups was observed with respect to clinical fractures or morphometric non-vertebral fractures. The total gain in femoral neck BMD compared with pre-treatment values was greater in the extension than in the discontinuation group (4.5% and 3.1%, respectively, p < 0.01). The corresponding gains in lumbar spine BMD were 12.1% and 10.1%, respectively, without significant differences between groups. BTM slightly increased but remained below pretreatment values in the discontinuation group. Of note, no cases of AFF were reported in either treatment group, and only one case of ONJ was recorded in the extension group [
      • Black D.M.
      • Reid I.R.
      • Boonen S.
      • et al.
      The effect of 3 versus 6 years of zoledronic acid treatment of osteoporosis: a randomized extension to the HORIZON-pivotal fracture trial (PFT).
      ]. A post-hoc analysis of the PFT trial, following the same rationale as the FLEX study, showed that patients with a femoral neck or hip T-score >−2.5, no incident fracture and no more than one risk factor may be considered for zoledronic acid withdrawal after three yearly infusions and for up to three years, since the risk of subsequent fracture is low [
      • Cosman F.
      • Cauley J.A.
      • Eastell R.
      • Boonen S.
      • Palermo L.
      • Reid I.R.
      • et al.
      Reassessment of fracture risk in women after 3 years of treatment with zoledronicacid: when is it reasonable to discontinue treatment?.
      ].
      The other study compared nine years of zoledronic acid treatment with six years of treatment and found no benefit for the former in terms of fracture risk, BMD and BTM. In contrast, a small increase in serious and non-serious cardiac arrhythmias was observed in the former group. It must be emphasized that no cases of ONJ or AFF were confirmed in either treatment group [
      • Black D.M.
      • Reid I.R.
      • Cauley J.A.
      • et al.
      The effect of 6 versus 9 years of zoledronic acid treatment in osteoporosis: a randomized second extension to the HORIZON-pivotal fracture trial (PFT).
      ]. This study suggested that no further benefit is expected beyond six years of treatment with zoledronate.
      Regarding male osteoporosis, no study was found that fulfilled the inclusion criteria of this systematic review.

      3.1.3 Risedronate

      The initial search provided 387 results, after excluding duplicates, two of which were reviewed by title and abstract (Fig. 1). Of those, one article was excluded because it reported a study with a non-randomized design. One full-text article was included in the qualitative analysis [
      • Watts N.B.
      • Chines A.
      • Olszynski W.P.
      • et al.
      Fracture risk remains reduced one year after discontinuation of risedronate.
      ].
      In the Vertebral Efficacy with Risedronate Therapy Multi-National (VERT-MN) trial, 398 patients who discontinued treatment for one year after completing three years of risedronate 5 mg/d were compared with 361 patients who received placebo. The risedronate group showed a 46% decreased risk of morphometric vertebral fractures compared with the placebo group (RR 0.54, 95% CI 0.34–0.86). There was no increase in non-vertebral fracture risk in the risedronate group. Femoral neck and lumbar BMD values in the risedronate group remained higher than in the placebo group (2.32% and 2.6%, respectively). BTM increased to pre-treatment values, with no difference from the placebo group [
      • Watts N.B.
      • Chines A.
      • Olszynski W.P.
      • et al.
      Fracture risk remains reduced one year after discontinuation of risedronate.
      ].
      In an extension of the VERT-MN trial, treatment with risedronate 5 mg/d for seven years (risedronate group, n = 31) was compared with placebo for five years and then treatment with risedronate5 mg/d for two years (placebo group, n = 30). Both groups discontinued risedronate for one year. After one year off treatment, lumbar spine and femoral neck BMD was maintained or slightly increased in both groups, whereas total hip and great trochanter BMD decreased. Bone resorption increased to a similar extent in both groups, reaching the levels of the placebo group at five years (before risedronate). The study indicated that the cumulative effect of risedronate is small, perhaps due to its low affinity to bone. No vertebral fractures were reported. There were no reports of ONJ or AFF [
      • Eastell R.
      • Hannon R.A.
      • Wenderoth D.
      • Rodriguez-Moreno J.
      • Sawicki A.
      Effect of stopping risedronate after long-term treatment on bone turnover.
      ].
      Regarding male osteoporosis, no study was found that fulfilled the inclusion criteria of this systematic review.

      3.1.4 Ibandronate

      The initial search provided 192 results, after excluding duplicates (Fig. 1). No study fulfilled the inclusion criteria of this systematic review.

      3.1.5 Denosumab

      The initial search provided 450 results; after excluding duplicates, one was reviewed by title and abstract (Fig. 1). However, this article was excluded, due its retrospective design.

      4. Clinical question 1: do patients need “drug holidays”?

      The idea of a “bisphosphonate holiday” was introduced some time ago and gained considerable popularity overthe years. The rationale of the “holiday” lies in two different concepts: first, after a given period of time, no additional benefit is found following bisphosphonate therapy; and second, prolonged bisphosphonate use may be associated with an increased incidence of adverse events.
      Regarding efficacy, additional benefit was seen only in clinical vertebral fractures when extending alendronate to an additional five years after completing five years of continuous treatment (in comparison with five years of “drug holiday”), without any benefit on morphometric and non-vertebral fractures [
      • Black D.M.
      • Schwartz A.V.
      • Ensrud K.E.
      • et al.
      Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial.; FLEX research group.
      ]. Similarly, extending zoledronate to a further three years after three annual infusions (in comparison with three years of “drug holiday”) reduced the risk of morphometric vertebral fracture, but not of clinical vertebral fracture or either type of non-vertebral fracture [
      • Black D.M.
      • Reid I.R.
      • Boonen S.
      • et al.
      The effect of 3 versus 6 years of zoledronic acid treatment of osteoporosis: a randomized extension to the HORIZON-pivotal fracture trial (PFT).
      ]. However, extending the use of alendronate and zoledronate seems to be beneficial only in patients still athigh fracture risk, particularly in those with a hip T-score <−2.5 and in elderly patients [
      • Schwartz A.V.
      • Bauer D.C.
      • Cummings S.R.
      • et al.
      FLEX Research Group, Efficacy of continued alendronate for fractures in women with and without prevalent vertebral fracture: the FLEX trial.
      ,
      • Cosman F.
      • Cauley J.A.
      • Eastell R.
      • Boonen S.
      • Palermo L.
      • Reid I.R.
      • et al.
      Reassessment of fracture risk in women after 3 years of treatment with zoledronicacid: when is it reasonable to discontinue treatment?.
      ]. Extension of zoledronate treatment to nine years provides no further benefit [
      • Black D.M.
      • Reid I.R.
      • Cauley J.A.
      • et al.
      The effect of 6 versus 9 years of zoledronic acid treatment in osteoporosis: a randomized second extension to the HORIZON-pivotal fracture trial (PFT).
      ].
      The major concerns associated with bisphosphonate and denosumab use are ONJ and AFF. ONJ was initially described in advanced cancer patients receiving high doses of intravenous bisphosphonates [
      • Marx R.E.
      Pamidronate (Aredia) and zoledronate (Zometa) induced avascular necrosis of the jaws: a growing epidemic.
      ]. Soon it was considered a major concern in any patient receiving bisphosphonates by any route. However, evidence from RCTs of a possible association of ONJ with bisphosphonate use is scarce. Indeed, no increased risk was demonstrated in the aforementioned RCTs regarding alendronate, zoledronate and risedronate. It has been calculated that the incidence of ONJ in patients receiving oral bisphosphonates ranges from 1:10,000 to 1:250,000 [
      • Khosla S.
      • Burr D.
      • Cauley J.
      • Dempster D.W.
      • Ebeling P.R.
      • Felsenberg D.
      • et al.
      Bisphosphonate-associated osteonecrosis of the jaw: report of a task force of the American Society for Bone and Mineral Research.
      ].
      An association has also been suggested between long-term bisphosphonate use and AFF [
      • Schilcher J.
      • Michaëlsson K.
      • Aspenberg P.
      Bisphosphonate use and atypical fractures of the femoral shaft.
      ]. Bisphosphonates significantly suppress bone remodeling, which is an essential process of the skeleton to repair microdamage. A very recent study investigated the differences in microdamage in trabecular bone among 1) patients with osteoporosis treated with bisphosphonates, 2) untreated patients with osteoporotic fractures (fracture controls) and 3) healthy individuals without fractures (non-fracture controls). By using synchrotron X-ray micro-CT and image segmentation technology, femoral head bone samples were tested for the density and volume of perforations (i.e. regions of complete breakage in the bone trabeculae, attributed to osteoclastic activity) and microcracks (i.e. microscopic fractures, 30–100 μm in length). Despite the small sample size (n = 16), bisphosphonate-treated patients demonstrated the highest density and volume of microcracks, despite the fact that the density and volume of perforations was comparable to those in healthy individuals. Although the duration of bisphosphonate use was not controlled for, this study suggests that the accumulation of microcracks attributed to bisphosphonate use reduces bone strength and thus may predispose to fractures [
      • Ma S.
      • Goh E.L.
      • Jin A.
      • Bhattacharya R.
      • Boughton O.R.
      • Patel B.
      • Karunaratne A.
      • Vo N.T.
      • Atwood R.
      • Cobb J.P.
      • Hansen U.
      • Abel R.L.
      Long-term effects of bisphosphonate therapy: perforations, microcracks and mechanical properties.
      ].
      Importantly, it has been proposed that ongoing bisphosphonate use is the main risk factor for AFF. It seems to be higher with alendronate than with risedronate [RR 1.9 (95% CI 1.1–3.3)] and about three times higher in women than in men[RR 3.1 (95% CI 1.1–8.4)] [
      • Schilcher J.
      • Koeppen V.
      • Aspenberg P.
      • Michaelsson K.
      Risk of atypical femoral fracture during and after bisphosphonate use.
      ]. The risk rapidly decreases, though, after drug discontinuation (70% per year after the last use of bisphosphonate), regardless of the long-term presence of the drug in the bone [
      • Schilcher J.
      • Michaëlsson K.
      • Aspenberg P.
      Bisphosphonate use and atypical fractures of the femoral shaft.
      ]. A study reviewing data from the US FDA (Food and Drug Administration) Adverse Event Reporting System (FAERS) from 1996 to 2011 reported a proportional reporting ratio (PRR) of 4.51 (95% CI 3.44–5.92) for bisphosphonate use and non-healing femoral fractures [
      • Edwards B.J.
      • Bunta A.D.
      • Lane J.
      • Odvina C.
      • Rao D.S.
      • Raisch D.W.
      • et al.
      Bisphosphonates and nonhealing femoral fractures: analysis of the FDA Adverse Event Reporting System (FAERS) and international safety efforts: a systematic review from the Research on Adverse Drug Events And Reports (RADAR) project.
      ]. A Swedish study has shown that the relative risk of AFF increases with the duration of bisphosphonate use, reaching 126.0 (95% CI 55.1–288.1) after four years of treatment [
      • Schilcher J.
      • Koeppen V.
      • Aspenberg P.
      • Michaelsson K.
      Risk of atypical femoral fracture during and after bisphosphonate use.
      ]. However, another study showed that long-term bisphosphonate use (i.e. for more than four years) is associated with a lower risk of non-traumatic subtrochanteric femoral fractures [
      • Safford M.M.
      • Barasch A.
      • Curtis J.R.
      • Outman R.
      • Saag K.
      Bisphosphonates and hip and nontraumatic subtrochanteric femoral fractures in the veterans health administration.
      ]. Because of the rarity of AFF, it has been calculated that for every 10,000 patients at high risk, 108 hip fractures (from 300 expected) and 750 fractures at other sites are prevented with bisphosphonate treatment, whereas only three subtrochanteric fractures are additionally expected [
      • Rizzoli R.
      • Akesson K.
      • Bouxsein M.
      • Kanis J.A.
      • Napoli N.
      • Papapoulos S.
      • et al.
      Subtrochanteric fractures after long-term treatment with bisphosphonates: a european society on clinical and economic aspects of osteoporosis and osteoarthritis, and international osteoporosis foundation working group report.
      ].
      The safety and efficacy of denosumab treatment extension to eight years have beentested in postmenopausal women. Only five cases of ONJ and one of AFF were reported in the group completing eight years of continuous treatment (n = 1542). BMD continued to increase at all sites and fracture risk remained low [
      • Papapoulos S.
      • Lippuner K.
      • Roux C.
      • Lin C.J.
      • Kendler D.L.
      • Lewiecki E.M.
      • Bone M.L.
      • et al.
      The effect of 8 or 5 years of denosumab treatment in postmenopausal women with osteoporosis: results from the FREEDOM Extension study.
      ].
      Musculoskeletal pain has also been described as a possible adverse effect of bisphosphonate therapy. However, it can develop at any point intreatment, and so prolonged bisphosphonate use is notparticularly associated with an increased risk [
      • Wysowski D.K.
      • Chang J.T.
      Alendronate and risedronate: reports of severe bone, joint, and muscle pain.
      ].
      Atrial fibrillation (AF) is another possible side-effect of bisphosphonate use, mainly suggested by the HORIZON PFT [
      • Black D.M.
      • Delmas P.D.
      • Eastell R.
      • Reid I.R.
      • Boonen S.
      • Cauley J.A.
      • et al.
      Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis.
      ]. Other studies, including large observational trials, have produced conflicting results [
      • Barrett-Connor E.
      • Swern A.S.
      • Hustad C.M.
      • Bone H.G.
      • Liberman U.A.
      • Papapoulos S.
      • et al.
      Alendronate and atrial fibrillation: a meta-analysis of randomized placebo-controlled clinical trials.
      ,
      • Loke Y.K.
      • Jeevanantham V.
      • Singh S.
      Bisphosphonates and atrial fibrillation: systematic review and meta-analysis.
      ,
      • Rhee C.W.
      • Lee J.
      • Oh S.
      • Choi N.K.
      • Park B.J.
      Use of bisphosphonate and risk of atrial fibrillation in older women with osteoporosis.
      ]. In any case, association does not confirm causality, and even if such an association does exist it seems to be independent of the dose and duration of therapy.
      Esophageal or gastric cancer is another major concern about oral bisphosphonate use. However, according to a recent systematic review, no evidence of increased risk of esophageal or gastric cancer was found in men and women receiving bisphosphonate treatment, though evidence from the original studies is conflicting [
      • Wright E.
      • Schofield P.T.
      • Molokhia M.
      Bisphosphonates and evidence for association with esophageal and gastric cancer: a systematic review and meta-analysis.
      ]. In an open-label, prospective, uncontrolled study, risedronate use was associated with significantly decreased ferritin concentration, which in turn was associated with a higher cardiovascular risk [
      • Feldbrin Z.
      • Luckish A.
      • Shargorodsky M.
      Effects of long-term risedronate treatment on serum ferritin levels in postmenopausal women with osteoporosis: the impact of cardiovascular risk factor load.
      ]. However, sufficient evidence to establish a causal link between risedronate and high cardiovascular risk is lacking. Anyhow, it is important to note that additional safety issues were not raised in the clinical trialsofextensionbisphosphonate therapy, thus reinforcing the safety of bisphosphonate treatment [
      • Bone H.G.
      • Hosking D.
      • Devogelaer J.P.
      • Tucci J.R.
      • Emkey R.D.
      • Tonino R.P.
      • et al.
      Ten years' experience with alendronate for osteoporosis in postmenopausal women.
      ,
      • Mellstrom D.D.
      • Sorensen O.H.
      • Goemaere S.
      • Roux C.
      • Johnson T.D.
      • Chines A.A.
      Seven years of treatment with risedronate in women with postmenopausal osteoporosis.
      ].

      5. Clinical question 2: who is a candidate for treatment discontinuation?

      The second important question to be answered concerns the patient population for whom a “drug holiday” is appropriate after long-term treatment with bisphosphonates. The answer to this question should be individualized, considering the long-term efficacy of bisphosphonates, their safety and the fracture risk of the individual patient. Discontinuation of bisphosphonate therapy should be considered for all patients who have been treated for more than five years with alendronate and risedronate or for more than three years with zoledronate. No recommendations can be made for ibandronate [
      • Bauer D.C.
      • Schwartz A.
      • Palermo L.
      • et al.
      Fracture prediction after discontinuation of 4–5 years of alendronate therapy. The FLEX study.
      ]. If the patient has not experienced any fractures before or during therapy and the fracture risk is low, as indicated by afemoral neck T-score ≥−2.5, age <70 years and absence of disease or medication associated with osteoporosis and/or increased fracture risk, a “drug holiday” should be considered [
      • Bauer D.C.
      • Schwartz A.
      • Palermo L.
      • et al.
      Fracture prediction after discontinuation of 4–5 years of alendronate therapy. The FLEX study.
      ,
      • Adler R.A.
      • El-Hajj Fuleihan G.
      • Bauer D.C.
      • Camacho P.M.
      • Clarke B.L.
      • Clines G.A.
      • Compston J.E.
      • Drake M.T.
      • Edwards B.J.
      • Favus M.J.
      • Greenspan S.L.
      • McKinney Jr, R.
      • Pignolo R.J.
      • Sellmeyer D.E.
      Managing osteoporosis in patients on long-term bisphosphonate treatment: report of a Task Force of the American Society for Bone and Mineral Research.
      ]. As the FRAX scores of patients in receipt of treatment have not been extensively assessed [
      • Anagnostis P.
      • Stevenson J.C.
      Bisphosphonate drug holidays − when, why and for how long?.
      ], this tool should not be used in this context.
      The appropriate period of discontinuation is unclear. This should be individualized and the physician should take into consideration the BMD measurement at the time of discontinuation, as well as the patient’s characteristics, such as age, smoking habits and factors that could increase the risk of falls. As a general rule, two to three years is a reasonable discontinuation period. Yearly measurement of BTM might be proposed and resuming treatment could be considered if there is an increase in bone resorption markers (above normal non-menopausal levels) although one-year changes in BTMs were not associated with risk of fracture after discontinuation of alendronate therapy [
      • Bauer D.C.
      • Schwartz A.
      • Palermo L.
      • et al.
      Fracture prediction after discontinuation of 4–5 years of alendronate therapy. The FLEX study.
      ]. If any fracture occurs during this period or another factor emerges that could affect BMD status (e.g. disease or medication), the patient should be reassessed sooner [
      • Adler R.A.
      • El-Hajj Fuleihan G.
      • Bauer D.C.
      • Camacho P.M.
      • Clarke B.L.
      • Clines G.A.
      • Compston J.E.
      • Drake M.T.
      • Edwards B.J.
      • Favus M.J.
      • Greenspan S.L.
      • McKinney Jr, R.
      • Pignolo R.J.
      • Sellmeyer D.E.
      Managing osteoporosis in patients on long-term bisphosphonate treatment: report of a Task Force of the American Society for Bone and Mineral Research.
      ,
      • Anagnostis P.
      • Stevenson J.C.
      Bisphosphonate drug holidays − when, why and for how long?.
      ,
      • Watts N.B.
      • Bilezikian J.P.
      • Camacho P.M.
      • et al.
      American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the diagnosis and treatment of postmenopausal osteoporosis.
      ]. For patients treated with risedronate, a shorter “drug holiday” should be considered, as discontinuation for more than one year may lead to significant loss of protection [
      • Watts N.B.
      • Chines A.
      • Olszynski W.P.
      • et al.
      Fracture risk remains reduced one year after discontinuation of risedronate.
      ,
      • Eastell R.
      • Hannon R.A.
      • Wenderoth D.
      • Rodriguez-Moreno J.
      • Sawicki A.
      Effect of stopping risedronate after long-term treatment on bone turnover.
      ].
      Regarding denosumab discontinuation, some concerns have been raised recently. Some cases of clinical vertebral fractures after its discontinuation have been reported, suggesting a possible rebound effect [
      • Lamy O.
      • Gonzalez-Rodriguez E.
      • Stoll D.
      • Hans D.
      • Aubry-Rozier B.
      Severe rebound-associated vertebral fractures after denosumab discontinuation: nine clinical cases report.
      ]. A recent observational study (n = 82) showed that BMD decreases by 6.7% in the lumbar spine and by 6.6% in thehip, one year after denosumab discontinuation (after having completed eight years of treatment with respective gains of 16.8% and 6.2%, from baseline). Eight patients (9.8%) sustained a fracture (7.3% vertebral, 1.2% in the femoral neck and 1.2% in the radius) during the one-year observation period [
      • McClung M.R.
      • Wagman R.B.
      • Miller P.D.
      • Wang A.
      • Lewiecki E.M.
      Observations following discontinuation of long-term denosumab therapy.
      ]. However, a post-hoc analysis of the Fracture Reduction Evaluation of Denosumab in Osteoporosis Every 6 Months (FREEDOM) study (the hallmark RCT showing fracture efficacy of denosumab in postmenopausal women) [
      • Cummings S.R.
      • San Martin J.
      • McClung M.R.
      • Siris E.S.
      • Eastell R.
      • Reid I.R.
      • Delmas P.
      • Zoog H.B.
      • Austin M.
      • Wang A.
      • Kutilek S.
      • Adami S.
      • Zanchetta J.
      • Libanati C.
      • Siddhanti S.
      • Christiansen C.
      FREEDOM trial, denosumab for prevention of fractures in postmenopausal women with osteoporosis.
      ] did not find a difference in fracture risk between the placebo (n = 470) and denosumab (n = 327) groups during the off-treatment period. It must be noted that the follow-up period was relatively short (median 0.8 years) and 42% of patients in the placebo group and 28% in the denosumab group had initiated other anti-osteoporotic medication after the last dose of the investigational product [
      • Brown J.P.
      • Roux C.
      • Törring O.
      • Ho P.R.
      • Beck Jensen J.E.
      • Gilchrist N.
      • Recknor C.
      • Austin M.
      • Wang A.
      • Grauer A.
      • Wagman R.B.
      Discontinuation of denosumab and associated fracture incidence: analysis from the fracture reduction evaluation of denosumab in osteoporosis every 6 months (FREEDOM) trial.
      ]. The exact pathogenetic mechanisms involved in this rebound phenomenon are unknown. The clinical risk profile and the management of these patients need to be determined. Hypercalcemia has also been reported after denosumab discontinuation [
      • Koldkjær Sølling A.S.
      • Harsløf T.
      • Kaal A.
      • Rejnmark L.
      • Langdahl B.
      Hypercalcemia after discontinuation of long-term denosumab treatment.
      ].

      6. Clinical question 3: who is a candidate for treatment re-initiation?

      Reassessment after bisphosphonate or denosumab discontinuation should include clinical evaluation to check for the presence of new fractures and risk factors for fractures [
      • Adler R.A.
      • El-Hajj Fuleihan G.
      • Bauer D.C.
      • Camacho P.M.
      • Clarke B.L.
      • Clines G.A.
      • Compston J.E.
      • Drake M.T.
      • Edwards B.J.
      • Favus M.J.
      • Greenspan S.L.
      • McKinney Jr, R.
      • Pignolo R.J.
      • Sellmeyer D.E.
      Managing osteoporosis in patients on long-term bisphosphonate treatment: report of a Task Force of the American Society for Bone and Mineral Research.
      ,
      • Watts N.B.
      • Bilezikian J.P.
      • Camacho P.M.
      • et al.
      American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the diagnosis and treatment of postmenopausal osteoporosis.
      ]. BMD measurement by DEXA can be considered at this point. Analysis of the FLEX study data has shown that this is not important for patients who have obtained a significant BMD gain. For example, for a patient with a T-score around −2, DEXA could even be postponed for almost five years, as it is unlikely to change management [
      • McNabb B.
      • Vittinghoff E.
      • Eastell R.
      • Schwartz A.V.
      • Bauer D.C.
      • Ensrud K.
      • Barrett-Connor E.
      • Black D.M.
      A model of BMD changes after alendronate discontinuation to guide post alendronate BMD monitoring.
      ]. Again, the decision should be individualized, taking into consideration each patient’s specific characteristics. Measurement of BTM, especially resorption markers, such as serum CTX, has been also proposed [
      • Khosla S.
      • Bilezikian J.P.
      • Dempster D.W.
      • et al.
      Benefits and risks of bisphosphonate therapy for osteoporosis.
      ]. If any new fracture is present or BMD has decreased (femur neck T-score ≤−2.5), anti-osteoporotic treatment could be resumed [
      • Adler R.A.
      • El-Hajj Fuleihan G.
      • Bauer D.C.
      • Camacho P.M.
      • Clarke B.L.
      • Clines G.A.
      • Compston J.E.
      • Drake M.T.
      • Edwards B.J.
      • Favus M.J.
      • Greenspan S.L.
      • McKinney Jr, R.
      • Pignolo R.J.
      • Sellmeyer D.E.
      Managing osteoporosis in patients on long-term bisphosphonate treatment: report of a Task Force of the American Society for Bone and Mineral Research.
      ,
      • Anagnostis P.
      • Stevenson J.C.
      Bisphosphonate drug holidays − when, why and for how long?.
      ,
      • Leslie W.D.
      • Lix L.M.
      • Johansson H.
      • Oden A.
      • McCloskey E.
      • Kanis J.A.
      Manitoba bone density program. does osteoporosis therapy invalidate FRAX for fracture prediction.
      ,
      • Watts N.B.
      • Bilezikian J.P.
      • Camacho P.M.
      • et al.
      American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the diagnosis and treatment of postmenopausal osteoporosis.
      ]. An increase in bone resorption markers within the reference range of young adults could prompt treatment resumption [
      • Khosla S.
      • Bilezikian J.P.
      • Dempster D.W.
      • et al.
      Benefits and risks of bisphosphonate therapy for osteoporosis.
      ].
      If re-initiation of treatment has been decided, the choices include bisphosphonates, denosumab [
      • Roux C.
      • Hofbauer L.C.
      • Ho P.R.
      • Wark J.D.
      • Zillikens M.C.
      • Fahrleitner-Pammer A.
      • Hawkins F.
      • Micaelo M.
      • Minisola S.
      • Papaioannou N.
      • Stone M.
      • Ferreira I.
      • Siddhanti S.
      • Wagman R.B.
      • Brown J.P.
      Denosumab compared with risedronate in postmenopausal women suboptimally adherent to alendronate therapy: efficacy and safety results from a randomized open-label study.
      ,
      • Recknor C.
      • Czerwinski E.
      • Bone H.G.
      • Bonnick S.L.
      • Binkley N.
      • Palacios S.
      • Moffett A.
      • Siddhanti S.
      • Ferreira I.
      • Ghelani P.
      • Wagman R.B.
      • Hall J.W.
      • Bolognese M.A.
      • Benhamou C.L.
      Denosumab compared with ibandronate in postmenopausal women previously treated with bisphosphonate therapy: a randomized open-label trial.
      ,
      • Lou S.
      • Lv H.
      • Wang G.
      • Li Z.
      • Li M.
      • Zhang L.
      • Tang P.
      The effect of sequential therapy forpostmenopausal women with osteoporosis. A PRISMA-compliant meta-analysis of randomized controlled trials.
      ] (the latter inducing greater BMD gains and BTM suppression than the former) [
      • Roux C.
      • Hofbauer L.C.
      • Ho P.R.
      • Wark J.D.
      • Zillikens M.C.
      • Fahrleitner-Pammer A.
      • Hawkins F.
      • Micaelo M.
      • Minisola S.
      • Papaioannou N.
      • Stone M.
      • Ferreira I.
      • Siddhanti S.
      • Wagman R.B.
      • Brown J.P.
      Denosumab compared with risedronate in postmenopausal women suboptimally adherent to alendronate therapy: efficacy and safety results from a randomized open-label study.
      ,
      • Recknor C.
      • Czerwinski E.
      • Bone H.G.
      • Bonnick S.L.
      • Binkley N.
      • Palacios S.
      • Moffett A.
      • Siddhanti S.
      • Ferreira I.
      • Ghelani P.
      • Wagman R.B.
      • Hall J.W.
      • Bolognese M.A.
      • Benhamou C.L.
      Denosumab compared with ibandronate in postmenopausal women previously treated with bisphosphonate therapy: a randomized open-label trial.
      ], teriparatide [
      • Lou S.
      • Lv H.
      • Wang G.
      • Li Z.
      • Li M.
      • Zhang L.
      • Tang P.
      The effect of sequential therapy forpostmenopausal women with osteoporosis. A PRISMA-compliant meta-analysis of randomized controlled trials.
      ], selective estrogen receptor modulators (SERMs) [
      • Lou S.
      • Lv H.
      • Wang G.
      • Li Z.
      • Li M.
      • Zhang L.
      • Tang P.
      The effect of sequential therapy forpostmenopausal women with osteoporosis. A PRISMA-compliant meta-analysis of randomized controlled trials.
      ], strontium ranelate [
      • Meunier P.J.
      • Roux C.
      • Seeman E.
      • Ortolani S.
      • Badurski J.E.
      • Spector T.D.
      • et al.
      The effects of strontium ranelate on the risk of vertebral fracture in women with postmenopausal osteoporosis.
      ], or menopausal hormone therapy (MHT) [
      • Rossouw J.E.
      • Anderson G.L.
      • Prentice R.L.
      • LaCroix A.Z.
      • Kooperberg C.
      • Stefanick M.L.
      • et al.
      Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial.
      ]. The risk of venous thrombosis should be taken into account when SERMS or oral MHT are considered. Furthermore cardiovascular risk has to be assessed with strontium ranelate use [
      European Medicines Agency Recommendation to Restrict the Use of Protelos/Osseor (strontium Ranelate).
      ]. To date, no trial has investigated the anti-fracture efficacy of re-initiation treatment with bisphosphonates following a “drug holiday”. Furthermore, no bisphosphonate trial has lasted beyond ten years, and switching to other therapies after three to five years of bisphosphonate treatment has not been examined. Therefore, re-initiation of therapy should be individualized, taking into consideration the patient’s profile, the duration of bisphosphonate treatment, and the efficacy and safety characteristics of other anti-osteoporotic agents [
      • Russell R.G.
      • Watts N.B.
      • Ebetino F.H.
      • Rogers M.J.
      Mechanisms of action of bisphosphonates: similarities and differences and their potential influence on clinical efficacy.
      ,
      • Khosla S.
      • Bilezikian J.P.
      • Dempster D.W.
      • et al.
      Benefits and risks of bisphosphonate therapy for osteoporosis.
      ,
      • Roux C.
      • Hofbauer L.C.
      • Ho P.R.
      • Wark J.D.
      • Zillikens M.C.
      • Fahrleitner-Pammer A.
      • Hawkins F.
      • Micaelo M.
      • Minisola S.
      • Papaioannou N.
      • Stone M.
      • Ferreira I.
      • Siddhanti S.
      • Wagman R.B.
      • Brown J.P.
      Denosumab compared with risedronate in postmenopausal women suboptimally adherent to alendronate therapy: efficacy and safety results from a randomized open-label study.
      ,
      • Recknor C.
      • Czerwinski E.
      • Bone H.G.
      • Bonnick S.L.
      • Binkley N.
      • Palacios S.
      • Moffett A.
      • Siddhanti S.
      • Ferreira I.
      • Ghelani P.
      • Wagman R.B.
      • Hall J.W.
      • Bolognese M.A.
      • Benhamou C.L.
      Denosumab compared with ibandronate in postmenopausal women previously treated with bisphosphonate therapy: a randomized open-label trial.
      ,
      • Lou S.
      • Lv H.
      • Wang G.
      • Li Z.
      • Li M.
      • Zhang L.
      • Tang P.
      The effect of sequential therapy forpostmenopausal women with osteoporosis. A PRISMA-compliant meta-analysis of randomized controlled trials.
      ,
      • Meunier P.J.
      • Roux C.
      • Seeman E.
      • Ortolani S.
      • Badurski J.E.
      • Spector T.D.
      • et al.
      The effects of strontium ranelate on the risk of vertebral fracture in women with postmenopausal osteoporosis.
      ,
      • Rossouw J.E.
      • Anderson G.L.
      • Prentice R.L.
      • LaCroix A.Z.
      • Kooperberg C.
      • Stefanick M.L.
      • et al.
      Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial.
      ]. For any patient with osteoporosis, it is important to design a long-term therapy plan, as this disease is a lifelong condition. In accordance with the Committee of Scientific Advisors of the International Osteoporosis Foundation, a sensible order to follow would be replacement of the bisphosphonate by a more potent drug of the same class, then by an injected anti-resorptive agent and then by an anabolic drug [
      • Diez-Perez A.
      • Adachi J.D.
      • Agnusdei D.
      • Bilezikian J.P.
      • Compston J.E.
      • Cummings S.R.
      • Eastell R.
      • Eriksen E.F.
      • Gonzalez-Macias J.
      • Liberman U.A.
      • Wahl D.A.
      • Seeman E.
      • Kanis J.A.
      • Cooper C.
      IOF CSA inadequate responders working group. Treatment failure in osteoporosis.
      ]. Treatment with teriparatide should be followed by an anti-resorptive agent, so that the therapeutic effect is sustained [
      • Lou S.
      • Lv H.
      • Wang G.
      • Li Z.
      • Li M.
      • Zhang L.
      • Tang P.
      The effect of sequential therapy forpostmenopausal women with osteoporosis. A PRISMA-compliant meta-analysis of randomized controlled trials.
      ,
      • Diez-Perez A.
      • Adachi J.D.
      • Agnusdei D.
      • Bilezikian J.P.
      • Compston J.E.
      • Cummings S.R.
      • Eastell R.
      • Eriksen E.F.
      • Gonzalez-Macias J.
      • Liberman U.A.
      • Wahl D.A.
      • Seeman E.
      • Kanis J.A.
      • Cooper C.
      IOF CSA inadequate responders working group. Treatment failure in osteoporosis.
      ].

      7. Conclusions

      After systematically reviewing the existing data regarding the effect of bisphosphonate discontinuation on fracture risk, as well as its possible benefits on reducing the risk of adverse effects, we make the following suggestions regarding bisphosphonate drug holidays:

      7.1 Optimal patient population

      Individualized decision making should take into consideration the long-term efficacy of bisphosphonates, their safety and the fracture risk of the specific patient. In the absence of fractures before or during therapy and for patients with a low fracture risk, as indicated by a femur neck T-score ≥−2.5, age <70 years and absence of diseases or treatments associated with osteoporosis and/or increased fracture risk, a “drug holiday” should be considered.

      7.2 Optimal length of drug holiday

      Similarly, individualized decision making should be encouraged with regard to the optimal length of the drug holiday period. Patients' characteristics, including age and history of falls, should be taken into consideration. Patient preferences should be respected and shared decision making should always be sought. In general, two to three yearsis a reasonable drug holiday, though a shorter period (one year) should be considered in patients taking risedronate. BMD reassessment is strongly suggested, and measurement of BTM may also be helpful. Any patient who experiences a new fracture should be immediately reassessed and therapy re-initiation should be considered.

      7.3 Re-initiation of treatment

      Bisphosphonates, as well as other anti-osteoporotic treatments, including denosumab, teriparatide, SERMs, MHT and strontium ranelate, could be considered after a “drug holiday”. An individualized long-term treatment plan should be constructed for any patient eligible for anti-osteoporotic treatment and it should be reviewed at regular intervals or after a major osteoporotic event (fracture, sudden decrease in BMD).

      8. Summary recommendations

      • Decisions should be individualized, taking into consideration the long-term efficacy of bisphosphonates and denosumab, their safety, and the fracture risk of the specific patient.
      • Discontinuation of bisphosphonatesshould be considered in all patients who have beentreated for more than five years with alendronate or more than three years with risedronate or zoledronic acid.
      • In view of the limited evidence, no robust recommendations can be made for ibandronate and denosumab.
      • Patients should be re-evaluated 1–3 years after bisphosphonate discontinuation. The decision to resume treatment depends on the presence of new fractures, risk factors and possibly bone mineral density.
      • Caution is advised regarding denosumab discontinuation, as there may be a “rebound effect” with regard to fractures.
      • Bisphosphonates as well as other anti-osteoporotic treatments, including denosumab, teriparatide, SERMs, MHT and strontium ranelate could be considered after a “drug holiday”.

      Contributors

      Panagiotis Anagnostis, Stavroula A. Paschou, Gesthimani Mintziori, Irene Lambrinoudaki and Dimitrios G. Goulis prepared the initial draft, which was circulated to all other named authors – EMAS board members – for comment and approval; production was coordinated by Irene Lambrinoudaki and Margaret Rees.

      Conflict of interest

      P. Anagnostis: None declared.
      S. A. Paschou: None declared.
      G. Mintziori: None declared.
      Ceausu: None declared.
      H. Depypere: None declared.
      Lambrinoudaki: None declared.
      A. Mueck: None declared.
      F. R. Pérez-López: None declared.
      M. Rees: None declared.
      L. M. Senturk: None declared.
      T. Simoncini: None declared.
      J. C. Stevenson in the past year has received grants/research support from Abbott, consulting fees from Abbott and Mylan, and speakers’ honoraria from Gedeon Richter and Mylan.
      P. Stute in the past year has received grants/research support from Medinova AG and DrKade/Besins Pharma GmbH, consulting fees from Max Zeller Söhne AG, Madaus GmbH, and speakers honoraria from MSD Merck Sharp & Dohme AG, DrKade/Besins Pharma GmbH and Max Zeller Söhne AG.
      Florence A. Tremollieres: None declared.
      D.G. Goulis: None declared.

      Funding

      No funding was received for the preparation of this position statement.

      Provenance and peer review

      This article is an EMAS position statement and was not externally peer reviewed.

      References

        • Russell R.G.
        • Watts N.B.
        • Ebetino F.H.
        • Rogers M.J.
        Mechanisms of action of bisphosphonates: similarities and differences and their potential influence on clinical efficacy.
        Osteoporos. Int. 2008; 19: 733-759
        • Khosla S.
        • Bilezikian J.P.
        • Dempster D.W.
        • et al.
        Benefits and risks of bisphosphonate therapy for osteoporosis.
        J. Clin. Endocrinol. Metab. 2012; 97: 2272-2282
        • Murad M.H.
        • Drake M.T.
        • Mullan R.J.
        • Mauck K.F.
        • Stuart L.M.
        • Lane M.A.
        • et al.
        Clinical review. comparative effectiveness of drug treatments to prevent fragility fractures: a systematic review and network meta-analysis.
        J. Clin. Endocrinol. Metab. 2012; 97: 1871-1880
        • Cummings S.R.
        • San Martin J.
        • McClung M.R.
        • Siris E.S.
        • Eastell R.
        • Reid I.R.
        • Delmas P.
        • Zoog H.B.
        • Austin M.
        • Wang A.
        • Kutilek S.
        • Adami S.
        • Zanchetta J.
        • Libanati C.
        • Siddhanti S.
        • Christiansen C.
        FREEDOM trial, denosumab for prevention of fractures in postmenopausal women with osteoporosis.
        N. Engl. J. Med. 2009; 361: 756-765
        • Khan A.A.
        • Morrison A.
        • Hanley D.A.
        • Felsenberg D.
        • McCauley L.K.
        • O'Ryan F.J.
        • et al.
        Diagnosis and management of osteonecrosis of the jaw: a systematic review and international consensus.
        Bone Miner. Res. 2015; 30: 3-23
        • Barasch A.
        • Cunha-Cruz J.
        • Curro F.A.
        • Hujoel P.
        • Sung A.H.
        • Vena D.
        • et al.
        Risk factors for osteonecrosis of the jaws: a case-control study from the CONDOR dental PBRN.
        J. Dent. Res. 2011; 90: 439-444
        • Shane E.
        • Burr D.
        • Ebeling P.R.
        • Abrahamsen B.
        • et al.
        Atypical subtrochanteric and diaphyseal femoral fractures: a second report of a task force of the American Society for Bone and Mineral Research.
        J. Bone Miner. Res. 2014; 29: 1-23
        • Gedmintas L.
        • Solomon D.H.
        • Kim S.C.
        Bisphosphonates and risk of subtrochanteric, femoral shaft, and atypical femur fracture: a systematic review and meta-analysis.
        J. Bone Miner. Res. 2013; 28: 1729-1737
        • Lee S.
        • Yin R.V.
        • Hirpara H.
        • et al.
        Increased risk for atypical fractures associated with bisphosphonate use.
        Fam. Pract. 2015; 32: 276-281
        • Meier R.P.H.
        • Perneger T.V.
        • Stern R.
        • Rizzoli R.
        • Peter R.E.
        Increasing occurrence of atypical femoral fractures associated with bisphosphonate use.
        Arch. Intern. Med. 2012; 172: 930-936
        • Papapoulos S.
        • Lippuner K.
        • Roux C.
        • Lin C.J.
        • Kendler D.L.
        • Lewiecki E.M.
        • Bone M.L.
        • et al.
        The effect of 8 or 5 years of denosumab treatment in postmenopausal women with osteoporosis: results from the FREEDOM Extension study.
        Osteopor. Int. 2015; 26: 2773-2783
        • Schilcher J.
        • Michaëlsson K.
        • Aspenberg P.
        Bisphosphonate use and atypical fractures of the femoral shaft.
        N. Engl. J. Med. 2011; 364: 1728-1737
        • Liberati A.
        • Altman D.G.
        • Tetzlaff J.
        • Mulrow C.
        • Gøtzsche P.C.
        • Ioannidis J.P.
        • et al.
        The PRISMA statement for reporting systematic reviews and meta-analyses of studies that evaluate healthcare interventions: explanation and elaboration.
        BMJ. 2009; 339: b2700
        • Black D.M.
        • Schwartz A.V.
        • Ensrud K.E.
        • et al.
        Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial.; FLEX research group.
        JAMA. 2006; 296: 2927-2938
        • Bone H.G.
        • Hosking D.
        • Devogelaer J.P.
        • Tucci J.R.
        • Emkey R.D.
        • Tonino R.P.
        • et al.
        Ten years' experience with alendronate for osteoporosis in postmenopausal women.
        N. Engl. J. Med. 2004; 350: 1189-1199
        • Ravn P.
        • Weiss S.R.
        • Rodriguez-Portales J.A.
        • McClung M.R.
        • Wasnich R.D.
        • Gilchrist N.L.J.
        • et al.
        Alendronate in early postmenopausal women: effects on bone mass during long-term treatment and after withdrawal. Alendronate osteoporosis prevention study group.
        J. Clin. Endocrinol. Metab. 2000; 85: 1492-1497
        • Bagger Y.
        • Tankó L.B.
        • Alexandersen P.
        • Ravn P.
        • Christiansen C.
        Alendronate has a residual effect on bone mass in postmenopausal Danish women up to 7 years after treatment withdrawal.
        Bone. 2003; 33: 301-307
        • Wasnich R.D.
        • Bagger Y.Z.
        • Hosking D.J.
        • McClung M.R.
        • Wu M.
        • Mantz A.M.
        • Yates J.J.
        • et al.
        Early Postmenopausal Intervention Cohort Study Group Changes in bone density and turnover after alendronate or estrogen withdrawal.
        Menopaus. 2004; 11: 622-630
        • Schwartz A.V.
        • Bauer D.C.
        • Cummings S.R.
        • et al.
        FLEX Research Group, Efficacy of continued alendronate for fractures in women with and without prevalent vertebral fracture: the FLEX trial.
        J. Bone Miner. Res. 2010; 25: 976-982
        • Black D.M.
        • Reid I.R.
        • Boonen S.
        • et al.
        The effect of 3 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
        • Black D.M.
        • Reid I.R.
        • Cauley J.A.
        • et al.
        The effect of 6 versus 9 years of zoledronic acid treatment in osteoporosis: a randomized second extension to the HORIZON-pivotal fracture trial (PFT).
        J. Bone Miner. Res. 2015; 30: 934-944
        • Cosman F.
        • Cauley J.A.
        • Eastell R.
        • Boonen S.
        • Palermo L.
        • Reid I.R.
        • et al.
        Reassessment of fracture risk in women after 3 years of treatment with zoledronicacid: when is it reasonable to discontinue treatment?.
        J. Clin. Endocrinol. Metab. 2014; 99: 4546-4554
        • Watts N.B.
        • Chines A.
        • Olszynski W.P.
        • et al.
        Fracture risk remains reduced one year after discontinuation of risedronate.
        Osteopor. Int. 2008; 19: 365-372
        • Eastell R.
        • Hannon R.A.
        • Wenderoth D.
        • Rodriguez-Moreno J.
        • Sawicki A.
        Effect of stopping risedronate after long-term treatment on bone turnover.
        J. Clin. Endocrinol. Metab. 2011; 96: 3367-3373
        • Marx R.E.
        Pamidronate (Aredia) and zoledronate (Zometa) induced avascular necrosis of the jaws: a growing epidemic.
        J. Oral Maxillofac. Surg. 2003; 61: 1115-1117
        • Khosla S.
        • Burr D.
        • Cauley J.
        • Dempster D.W.
        • Ebeling P.R.
        • Felsenberg D.
        • et al.
        Bisphosphonate-associated 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
        • Ma S.
        • Goh E.L.
        • Jin A.
        • Bhattacharya R.
        • Boughton O.R.
        • Patel B.
        • Karunaratne A.
        • Vo N.T.
        • Atwood R.
        • Cobb J.P.
        • Hansen U.
        • Abel R.L.
        Long-term effects of bisphosphonate therapy: perforations, microcracks and mechanical properties.
        Sci. Rep. 2017; 6: 43399
        • Schilcher J.
        • Koeppen V.
        • Aspenberg P.
        • Michaelsson K.
        Risk of atypical femoral fracture during and after bisphosphonate use.
        New Engl. J. Med. 2014; 371: 974-976
        • Edwards B.J.
        • Bunta A.D.
        • Lane J.
        • Odvina C.
        • Rao D.S.
        • Raisch D.W.
        • et al.
        Bisphosphonates and nonhealing femoral fractures: analysis of the FDA Adverse Event Reporting System (FAERS) and international safety efforts: a systematic review from the Research on Adverse Drug Events And Reports (RADAR) project.
        J. Bone Jt. Surg. 2013; 95: 297-307
        • Safford M.M.
        • Barasch A.
        • Curtis J.R.
        • Outman R.
        • Saag K.
        Bisphosphonates and hip and nontraumatic subtrochanteric femoral fractures in the veterans health administration.
        J. Clin. Rheumatol. 2014; 20: 357-362
        • Rizzoli R.
        • Akesson K.
        • Bouxsein M.
        • Kanis J.A.
        • Napoli N.
        • Papapoulos S.
        • et al.
        Subtrochanteric fractures after long-term treatment with bisphosphonates: a european society on clinical and economic aspects of osteoporosis and osteoarthritis, and international osteoporosis foundation working group report.
        Osteoporos. Int. 2011; 22: 373-390
        • Wysowski D.K.
        • Chang J.T.
        Alendronate and risedronate: reports of severe bone, joint, and muscle pain.
        Arch. Int. Med. 2005; 165: 346-347
        • Black D.M.
        • Delmas P.D.
        • Eastell R.
        • Reid I.R.
        • Boonen S.
        • Cauley J.A.
        • et al.
        Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis.
        N. Engl. J. Med. 2007; 356: 1809-1822
        • Barrett-Connor E.
        • Swern A.S.
        • Hustad C.M.
        • Bone H.G.
        • Liberman U.A.
        • Papapoulos S.
        • et al.
        Alendronate and atrial fibrillation: a meta-analysis of randomized placebo-controlled clinical trials.
        Osteoporos. Int. 2012; 23: 233-245
        • Loke Y.K.
        • Jeevanantham V.
        • Singh S.
        Bisphosphonates and atrial fibrillation: systematic review and meta-analysis.
        Drug Saf. 2009; 32: 219-228
        • Rhee C.W.
        • Lee J.
        • Oh S.
        • Choi N.K.
        • Park B.J.
        Use of bisphosphonate and risk of atrial fibrillation in older women with osteoporosis.
        Osteoporos. Int. 2012; 23: 247-254
        • Wright E.
        • Schofield P.T.
        • Molokhia M.
        Bisphosphonates and evidence for association with esophageal and gastric cancer: a systematic review and meta-analysis.
        BMJ Open. 2015; 5: e007133
        • Feldbrin Z.
        • Luckish A.
        • Shargorodsky M.
        Effects of long-term risedronate treatment on serum ferritin levels in postmenopausal women with osteoporosis: the impact of cardiovascular risk factor load.
        Menopause. 2016; 23: 55-59
        • Bone H.G.
        • Hosking D.
        • Devogelaer J.P.
        • Tucci J.R.
        • Emkey R.D.
        • Tonino R.P.
        • et al.
        Ten years' experience with alendronate for osteoporosis in postmenopausal women.
        New Engl. J. Med. 2004; 350: 1189-1199
        • Mellstrom D.D.
        • Sorensen O.H.
        • Goemaere S.
        • Roux C.
        • Johnson T.D.
        • Chines A.A.
        Seven years of treatment with risedronate in women with postmenopausal osteoporosis.
        Calcif. Tissue Int. 2004; 75: 462-468
        • Bauer D.C.
        • Schwartz A.
        • Palermo L.
        • et al.
        Fracture prediction after discontinuation of 4–5 years of alendronate therapy. The FLEX study.
        JAMA Intern. Med. 2014; 174: 1126-1134
        • Adler R.A.
        • El-Hajj Fuleihan G.
        • Bauer D.C.
        • Camacho P.M.
        • Clarke B.L.
        • Clines G.A.
        • Compston J.E.
        • Drake M.T.
        • Edwards B.J.
        • Favus M.J.
        • Greenspan S.L.
        • McKinney Jr, R.
        • Pignolo R.J.
        • Sellmeyer D.E.
        Managing osteoporosis in patients on long-term bisphosphonate treatment: report of a Task Force of the American Society for Bone and Mineral Research.
        J. Bone. Miner. Res. 2016; 31: 1910
        • Anagnostis P.
        • Stevenson J.C.
        Bisphosphonate drug holidays − when, why and for how long?.
        Climacteric. 2015; 18: 32-38
        • Leslie W.D.
        • Lix L.M.
        • Johansson H.
        • Oden A.
        • McCloskey E.
        • Kanis J.A.
        Manitoba bone density program. does osteoporosis therapy invalidate FRAX for fracture prediction.
        J. Bone. Miner. Res. 2012; 27: 1243-1251
        • Watts N.B.
        • Bilezikian J.P.
        • Camacho P.M.
        • et al.
        American Association of Clinical Endocrinologists Medical Guidelines for Clinical Practice for the diagnosis and treatment of postmenopausal osteoporosis.
        Endocr. Pract. 2010; 16: 1-37
        • Lamy O.
        • Gonzalez-Rodriguez E.
        • Stoll D.
        • Hans D.
        • Aubry-Rozier B.
        Severe rebound-associated vertebral fractures after denosumab discontinuation: nine clinical cases report.
        J. Clin. Endocrinol. Metab. 2016; (jc20163170 [Epub ahead of print])https://doi.org/10.1007/s00198-015-3380-y
        • McClung M.R.
        • Wagman R.B.
        • Miller P.D.
        • Wang A.
        • Lewiecki E.M.
        Observations following discontinuation of long-term denosumab therapy.
        Osteoporos. Int. 2017; (Epub ahead of print)https://doi.org/10.1007/s00198-017-3919-1
        • Brown J.P.
        • Roux C.
        • Törring O.
        • Ho P.R.
        • Beck Jensen J.E.
        • Gilchrist N.
        • Recknor C.
        • Austin M.
        • Wang A.
        • Grauer A.
        • Wagman R.B.
        Discontinuation of denosumab and associated fracture incidence: analysis from the fracture reduction evaluation of denosumab in osteoporosis every 6 months (FREEDOM) trial.
        J. Bone Miner. Res. 2013; 28: 746-752
        • Koldkjær Sølling A.S.
        • Harsløf T.
        • Kaal A.
        • Rejnmark L.
        • Langdahl B.
        Hypercalcemia after discontinuation of long-term denosumab treatment.
        Osteoporos. Int. 2016; 27: 2383-2386
        • McNabb B.
        • Vittinghoff E.
        • Eastell R.
        • Schwartz A.V.
        • Bauer D.C.
        • Ensrud K.
        • Barrett-Connor E.
        • Black D.M.
        A model of BMD changes after alendronate discontinuation to guide post alendronate BMD monitoring.
        J. Clin. Endocrinol. Metab. 2014; 99: 4094-4100
        • Roux C.
        • Hofbauer L.C.
        • Ho P.R.
        • Wark J.D.
        • Zillikens M.C.
        • Fahrleitner-Pammer A.
        • Hawkins F.
        • Micaelo M.
        • Minisola S.
        • Papaioannou N.
        • Stone M.
        • Ferreira I.
        • Siddhanti S.
        • Wagman R.B.
        • Brown J.P.
        Denosumab compared with risedronate in postmenopausal women suboptimally adherent to alendronate therapy: efficacy and safety results from a randomized open-label study.
        Bone. 2014; 58: 48-54
        • Recknor C.
        • Czerwinski E.
        • Bone H.G.
        • Bonnick S.L.
        • Binkley N.
        • Palacios S.
        • Moffett A.
        • Siddhanti S.
        • Ferreira I.
        • Ghelani P.
        • Wagman R.B.
        • Hall J.W.
        • Bolognese M.A.
        • Benhamou C.L.
        Denosumab compared with ibandronate in postmenopausal women previously treated with bisphosphonate therapy: a randomized open-label trial.
        Obstet. Gynecol. 2013; 121: 1291-1299
        • Lou S.
        • Lv H.
        • Wang G.
        • Li Z.
        • Li M.
        • Zhang L.
        • Tang P.
        The effect of sequential therapy forpostmenopausal women with osteoporosis. A PRISMA-compliant meta-analysis of randomized controlled trials.
        Medicine (Baltimore). 2016; 95: e5496
        • Meunier P.J.
        • Roux C.
        • Seeman E.
        • Ortolani S.
        • Badurski J.E.
        • Spector T.D.
        • et al.
        The effects of strontium ranelate on the risk of vertebral fracture in women with postmenopausal osteoporosis.
        N. Engl. J. Med. 2004; 350: 459-468
        • Rossouw J.E.
        • Anderson G.L.
        • Prentice R.L.
        • LaCroix A.Z.
        • Kooperberg C.
        • Stefanick M.L.
        • et al.
        Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results From the Women's Health Initiative randomized controlled trial.
        JAMA. 2002; 288: 321-333
      1. European Medicines Agency Recommendation to Restrict the Use of Protelos/Osseor (strontium Ranelate).
        2013 (http://www.ema.europa.eu/ema/index.jsp%3Fcurl=pages/news_and_events/news/2013/04/news_detail_001774.jsp%26mid=WC0b01ac058004d5c1 Accessed 6 April 2017)
        • Diez-Perez A.
        • Adachi J.D.
        • Agnusdei D.
        • Bilezikian J.P.
        • Compston J.E.
        • Cummings S.R.
        • Eastell R.
        • Eriksen E.F.
        • Gonzalez-Macias J.
        • Liberman U.A.
        • Wahl D.A.
        • Seeman E.
        • Kanis J.A.
        • Cooper C.
        IOF CSA inadequate responders working group. Treatment failure in osteoporosis.
        Osteoporos. Int. 2012; 23: 2769-2774

      Linked Article