Osteoporosis Management

What we know

Data from the FREEDOM trial showed that 3 years of denosumab treatment resulted in increased bone mineral density (BMD) and decreased fractures compared to placebo (68% relative risk [RR] reduction for new vertebral fractures and a 40% RR reduction for hip fractures vs. placebo). ¹ Despite the known benefits of short-term denosumab therapy, longer duration of use remains controversial due to efficacy and safety concerns, including atypical femoral shaft fractures (AFF) and osteonecrosis of the jaw (ONJ).

Study results

The authors used data from the 3-year FREEDOM study, as well as reported data from the first 5 years of its ongoing 7-year extension study. The extension study enrolled 4550 postmenopausal women from the FREEDOM study and provided an additional 5 years of denosumab treatment; 2343 of these women had received 3 years of denosumab, while 2207 had been previously assigned to the placebo arm. Patients were monitored every 6 months with bone turnover markers (BTM) and had DXA at the lumbar spine and hip, or the 1/3 radius in a subset of women. Patients were questioned every 3 months about adverse side effects, and any fracture occurrence was monitored via X-rays (at baseline, and years 2, 3, and 5).

Treatment with continuous denosumab for up to 8 years resulted in sustained reduction in BTM, further increases in bone mineral density at all sites measured (13.1%–18.4% at the spine, 6.2%–8.3% at the hip, and 1.5%–2.3% at the distal 1/3 of the radius), as well as low annualized incidence of fractures (0.9%–1.8%). Furthermore, only eight women developed ONJ, and two suffered an AFF, translating to 4.2 per 10,000 subject-years for ONJ and 1 per 10,000 subject-years for AFF. The incidence of serious adverse effects, such as malignancy, cellulitis, eczema, or other serious infections, remained low throughout the entire study.

What this changes or adds

Denosumab treatment for up to 8 years was associated with persistent reductions of BTMs, continued BMD gains, low fracture incidence, and a favorable benefit/risk profile in an aging population. One important limitation to the study was the absence of an ongoing placebo group. A different publication from the same study showed that the nonvertebral fracture rate reductions in year 4 were most prominent in subjects with persistent low hip BMD. ²

Abrahamsen B, et al. Risk of hip, subtrochanteric, and femoral shaft fractures among mid and long term users of alendronate: nationwide cohort and nested casecontrol study. BMJ 2016;353:i3365.

What we know

BPs remain the most common pharmacologic osteoporosis therapy, with proven efficacy at reducing hip, vertebral, and nonvertebral fractures. ^3,4 Sustained therapeutic effects are seen after cessation of BPs due to their prolonged skeletal half-life, with the exception of an increase in clinical vertebral fractures. ⁵ Continued benefits on discontinuation, along with concerns about risk of AFF with long-term use, have led many to consider a drug holiday after 5 years of treatment in those who do not remain at highest risk for fracture.

Study results

This cohort study of two nested case–control studies used data from the Danish National Health Registry. Previously untreated women and men between the ages of 50 and 94 were followed between 1996 and 2007 after starting alendronate (once weekly 70 mg by mouth). Main outcome measures were AFF (i.e., subtrochanteric femur or femoral shaft, ST/FS) and hip fractures in both the study population as well as population controls; the goal was to determine the influence of fracture incidence on duration of therapy. Up to 3–5 nonfracture controls from the cohort were matched to each fracture case by sex, year of birth, and year of initiation of alendronate treatment. Atypical ST/FS fractures were diagnosed as defined by the American Society for Bone and Mineral Research Task Force criteria, which details fracture history, location, and configuration. ⁶

Out of 61,990 patients treated with alendronate, 6784 (10.9%) hip fractures and 1428 (2.3%) ST/FS fractures occurred. Hip fracture rates dropped from 36.2/1000 patient-years during first year of treatment to 10–15/1000 during years 2 through 12; the ST/FS fracture rates remained stable at 2.7–4.6/1000 patient-years throughout the treatment duration. Those treated beyond 10 years had a 30% lower hip fracture risk over those without treatment, with no further change in risk of ST/FS fractures. Medication adherence was associated with a decreased risk of hip fracture. Those with diabetes mellitus as well as those taking proton pump inhibitors in the last year showed an increased risk for ST/FS fractures with adjusted odds ratios of 1.41 (1.13–1.76) and 1.20 (1.06–1.37), respectively.

What this changes or adds

The drug exposure data assessed here during the two-decade duration distinguish this study from others evaluating therapy benefits versus harm. This study provides reassurance that the total risk of ST/FS fractures did not increase over the first 10 years of high adherence to alendronate, but was accompanied by an adjusted 30% risk reduction in hip fractures. This implies that any increases in ST/FS fractures are offset by the substantial decreases in the risk of hip fractures, supporting 10 years of uninterrupted alendronate therapy as a safe approach to also ensure vertebral fracture risk reduction.

There are limitations to this study. This study included all fractures coded as ST/FS, without verifying these criteria via radiographs, in an attempt to capture all such fractures and avoid underestimating risk. AFF can occur in those on antiresorptive therapies as well as in those without such exposure, with studies suggesting that 1% of all femoral fractures and 3% of all femoral shaft fractures are atypical. ³ Thus, the absolute harm rates reported are likely inflated by inclusion of nonatypical ST/FS fractures. The study population is mostly made up of those of North European descent, and thus, results might not apply to those with other ethnic backgrounds with higher baseline risks of atypical ST/FS fractures (i.e., South Asian heritage). The American Society for Bone and Mineral Research has recently published helpful recommendations on choosing the appropriate treatment duration based on individualized risk assessment. ³

What we know

Women taking MHT in the Women's Health Initiative trials demonstrated lower overall fracture rates compared with those receiving placebo in both arms of the study (conjugated equine estrogen [CEE]/medroxyprogesterone acetate [MPA] and CEE alone), with a 33% decrease in hip fractures and ∼25% decrease in total fractures. ^7,8 However, there has been concern about increased fracture rates after discontinuation of MHT. ⁹

Study results

Watts et al. studied the 15,187 women who were taking MHT (CEE/MPA or CEE alone) or placebo through the intervention phase of the Women's Health Initiative trials and who were not on MHT during the postintervention phase. The incidence of both hip and total fractures was determined through 5 years after MHT discontinuation. The results were adjusted for age, body mass index, ethnicity, time since menopause, smoking, alcohol intake, duration of prior MHT use, physical activity, falls within the past year, prior fracture, calcium and vitamin D intake, bone-active medication use, and fracture before MHT discontinuation. During the 5-year follow-up, hip fractures were infrequent (2.5 per 1000 person-years), with no difference between previous MHT users and placebo groups. However, women previously assigned to CEE alone had a significant decrease in the risk of total fractures compared with those on placebo (hazard ratio, 0.85; 95% CI, 0.73–0.98; p = .03). No significant difference in the incidence of total fractures was noted between those assigned to CEE/MPA versus placebo.

What this changes or adds

This large study suggests that discontinuation of MHT does not seem to result in a rebound increase in fracture rates among postmenopausal women. A limitation of the study is the lack of bone mineral density and BTM data. Nonetheless, it provides valuable information for counseling women who are considering stopping MHT but are concerned about adverse bone health consequences.

What we know

Concerns have been raised regarding a potential association between oral calcium intake and increased risk of cardiovascular disease (CVD), but data are inconsistent. ^{10 –12} It appears that calcium derived from supplements may pose a higher risk for CVD than calcium obtained from dietary sources. ¹³ A proposed mechanism for this source-based risk differential is that large boluses of calcium in the form of supplements (which are nearly impossible to achieve with dietary intake alone) may cause increases in serum calcium levels ¹⁴ and may result in vascular calcifications. This is analogous to the increased risk of nephrolithiasis with the use of calcium supplements compared with the decreased risk associated with dietary intake of calcium. ¹⁵

Study results

Chung, et al. performed a systematic review and meta-analysis of the effects of calcium intake from dietary or supplemental sources, alone or with vitamin D, on CVD risk in healthy adults. Included in the analysis were 4 randomized controlled trials, 1 nested case–control study, and 26 cohort studies. The authors concluded that in healthy adults, calcium intake within the recommended upper intake range (2000–2500 mg/day) is not associated with increased CVD risks. The authors did not make a distinction based on the source of calcium (dietary or supplement). They further noted that in the few trials and cohort studies that did find increased risk with higher calcium intake, risk estimates were small and of doubtful clinical significance.

In the MESA study, Anderson, et al. investigated the relationship between total calcium intake (from dietary sources and supplements) and CAC in 5448 participants, 52% of whom were women. CAC was measured at baseline and again about 10 years later in 2742 participants. Those who were in the highest quintile for total calcium intake (2157.4 mg/day) had a decreased risk of incident CAC over the 10-year follow-up (RR, 0.73; 95% CI, 0.57–0.93) in the fully adjusted model, particularly if achieved without supplement use. Conversely, in patients who obtained their daily calcium mostly from supplemental sources, increased intake was associated with an increased risk of incident CAC in the fully adjusted model and also in the model adjusted for total calcium intake (RR, 1.22; 95% CI, 1.07–1.39).

What this changes or adds

As long as the total oral calcium intake is within the normal recommended allowance limits and is predominantly obtained from dietary sources, it does not seem to increase the risk of CVD. A diet high in calcium is potentially a surrogate for healthy eating habits in general, which may also reduce CVD risk. Although increased use of calcium supplements may be associated with an increase in CAC, it is unclear whether this is clinically significant.

What we know

While multiple medical societies have widely accepted recommendations for initial DXA screening, ^{16 –18} there are uncertainties regarding the role of DXA scans to monitor response to treatment.

Study results

Using a registry of patients from Manitoba, Canada, 6629 women ≥40 years who were initiating any osteoporosis treatment were identified (84.9% were prescribed BPs). Their BMDs from DXA scans were obtained before and after starting treatment (mean interval, 4.5 years). The change in BMD was calculated and categorized as stable, detectable decrease, or detectable increase. The incidence of fractures was obtained from a health service database.

During an average of 9.2 years, 13.7% of the women in the study developed incident fractures. Compared to those with stable BMD, women with a detectable decrease in total hip BMD had an overall increased incidence of fracture (2.9% and 5.5% higher risk in the 5- and 10-year cumulative incidence, respectively). In contrast, women with a detectable increase in total hip BMD had a decreased risk for any fracture (1.3% and 2.6% lower risk after 5 and 10 years, respectively) compared to those with a stable BMD.

What this changes or adds

This study shows that treatment-related increases in total hip BMD, compared to attaining BMD stability, are associated with reduced fracture risk. In addition, decreases in BMD are highly correlated with a greater fracture risk. The follow-up DXA scan has value in not only better understanding the patient's fracture risk but may also help identify those who are not adherent to their osteoporosis therapy.