Combination therapy: the Holy Grail for the treatment of postmenopausal osteoporosis?

Abstract

Agents for the treatment of osteoporosis are divided into broad categories according to whether the predominant effect is to inhibit bone resorption (antiresorptive drugs) or stimulate bone formation (osteo-anabolic drugs). However, due to the coupling of these two components of bone remodeling, drugs that inhibit bone resorption generally also reduce bone formation, and those that stimulate bone formation also increase bone resorption. Since these synchronous changes may limit the beneficial effects of treatment, researchers have undertaken a search for combinations of antiresorptive and osteo-anabolic drugs given concurrently, sequentially, intermittently, or cyclically that could partially or totally uncouple bone resorption and bone formation. This offers the potential for greater increases in bone mineral density (BMD), restoration of lost structural elements, and perhaps greater reduction in fracture risk than monotherapy with currently approved drugs. While some methods of combining drugs have been shown to enhance BMD response and perhaps extend the duration of osteo-anabolic effects compared to monotherapy, none have been proven to provide greater reduction of fracture risk. Upon completion of a course of osteo-anabolic therapy with daily subcutaneous parathyroid hormone, antiresorptive therapy must be started in order to prevent subsequent loss of BMD.

Key words::

• Bone density
• Combination
• Fracture
• Osteoporosis
• Treatment

Imagine a ‘cure’ for osteoporosis. Bone mineral density (BMD) would go from very low to normal. Structural degradation and lost microarchitectural elements of the skeleton would be restored to normal. Bone strength would improve to what it was long ago and fracture risk would become what it was at the age of 30 years. These are the dreams of scientific visionaries searching for the Holy Grail of osteoporosis treatment. But what could be done to achieve this goal, and is it really possible?

Recent lines of investigation have focused on combinations of existing drugs and development of novel therapeutic agents that might ‘uncouple’ bone resorption and bone formation – the yin and the yang of bone remodeling. The process of bone remodeling serves two major purposes: calcium homeostasis, whereby mineral fluxes in and out of bone maintain a very tightly controlled concentration of calcium in extracellular fluid, and a structural function, where continual repair of microdamage, a natural consequence of repetitive mechanical loading, adapts the skeleton to the demands of its mechanical environment1,2. In a skeletal steady state, as in healthy premenopausal women with stable BMD, bone resorption ≈ bone formation. However, in estrogen deficient postmenopausal women, there is a high bone remodeling rate with bone resorption > bone formation, resulting in a net loss of bone and deterioration of skeletal microarchitecture over time. This weakens the bone and increases the risk of low trauma fractures. Currently approved medications for the treatment of osteoporosis are divided into two broad categories according to the predominant effect on bone remodeling. Antiresorptive (anti-catabolic) agents (e.g., bisphosphonates, denosumab, raloxifene, and salmon calcitonin) reduce bone resorption. Osteo-anabolic (bone-forming) agents, such as teriparatide (parathyroid hormone [1‐34]) and PTH(1‐84), increase bone formation. Drugs in both categories have been shown to reduce fracture risk, despite having opposite effects on bone remodeling3. However, due to coupling of bone resorption and formation, drugs that inhibit bone resorption also inhibit bone formation, thereby mitigating the potential benefit of the antiresorptive effect, and those that increase bone formation also increase bone resorption, mitigating the osteo-anabolic effect. If a treatment could be developed that partially or totally uncoupled bone resorption and formation, there is the potential for greater increases in BMD, restoration of degraded structural elements (e.g., cortical thickness, trabecular microarchitecture), improvement of bone strength, and perhaps greater anti-fracture efficacy compared to monotherapy with drugs that are currently available. The achievement of this elusive goal could provide the ultimate treatment for osteoporosis and potentially a cure. Some combinations of osteo-anabolic and antiresorptive drugs, as well as several individual drugs, have shown promise in recent clinical trials. Combinations of two antiresorptive drugs have been shown to have small additive effects on BMD and/or suppression of bone turnover markers (BTMs)4–7, although there is no proven benefit for fracture risk reduction and a possibility of harm due to excessive suppression of bone turnover; this type of combination therapy will not be discussed here. Consideration of what is known about various methods of combining drugs (Table 1) provides insights into potential strategies for using drugs that are currently available and a peek at what might be the future of osteoporosis therapy.

Table 1.  Combination therapy for postmenopausal osteoporosis. This table represents a summary of general concepts in the use of combinations of drugs. It is not intended to be complete, as many variations of combinations have been reported in studies that differ in their design and analysis. There are no data sufficiently robust to conclude that any combination reduces fracture risk more than monotherapy.

Type of combination	Examples	Comments
Antiresorptive followed by anabolic	ALN ↠ PTH15–17	Prior treatment with ALN attenuates response to PTH in some studies. This effect is diminished or may not occur with prior RIS or RLX treatment.
	RIS  ↠ PTH16
	RLX ↠ PTH17
Antiresorptive with concurrent anabolic	ALN + PTH8–10,17	Concurrent treatment with ALN attenuates response to PTH in some studies, while other antiresorptive agents may have neutral or beneficial effects with PTH.
	RLX + PTH11,17
	HRT + PTH12
	ZOL + PTH14
Anabolic followed by antiresorptive	PTH ↠ ALN18–20	Standard of care is to follow PTH with an antiresorptive drug to maintain or enhance the benefit achieved. Failure to do so results in a rapid decline in lumbar spine BMD.
	PTH ↠ RIS20
Antiresorptive with cyclic anabolic	ALN with PTH 3 months on and 3 months off22	Cyclic PTH with ongoing ALN may provide BMD increases similar to daily PTH alone with lower cost and greater patient convenience.
Two concurrent antiresorptive agents	HRT + ALN5	Small additive effects on BMD reported. Potential risk of excessive suppression of bone turnover.
	HRT + RIS6
	ALN + RLX4
Two concurrent anabolic agents	No studies	No data.

PTH = daily subcutaneous parathyroid hormone, either teriparatide [PTH(1-34)] or PTH(1-84); ALN = alendronate; RIS = risedronate; ZOL = zoledronic acid; RLX = raloxifene; HRT = hormone replacement therapy or estrogen therapy; Anabolic = osteo-anabolic; BMD = bone mineral density.

Concurrent therapy with an antiresorptive and osteo-anabolic drug

The ‘ParaThyroid Hormone and alendronate’ (PaTH) study was a prospective, randomized, double-blind study in 238 postmenopausal women with low BMD assigned to receive PTH(1‐84), alendronate, or both8. At 12 months, the lumbar spine BMD response, whether measured by dual-energy X-ray absorptiometry (DXA) or quantitative computed tomography (QCT), was similar in all three groups. The total hip BMD response by DXA was greater in the combination group compared to the PTH(1-84) group, and the lumbar spine trabecular BMD by QCT was greater in the PTH(1-84) group than the combination group. Overall, it was concluded that there was no synergy in combining alendronate with PTH(1-84) and that alendronate may reduce the anabolic effects of PTH(1-84). A similar study was conducted in 93 postmenopausal women randomized to receive teriparatide, alendronate, or both9. Here, however, teriparatide was started at month 6 and subjects were followed for 30 months. Lumbar spine, total hip, and femoral neck BMD by DXA, as well as lumbar spine trabecular BMD by QCT, increased more with teriparatide than with alendronate alone or the combination of the two. In another study of 83 men with osteoporosis, subjects were randomized assigned to receive PTH(1-34), alendronate, or both10. Alendronate was found to impair the BMD response (measured by DXA at the lumbar spine or femoral neck and by QCT of the trabecular compartment of the lumbar spine) to PTH(1-34). The increase in serum alkaline phosphatase, a marker of bone formation, was greater with PTH(1-34) alone than with combination therapy. These data suggest that concurrent use of alendronate reduces the anabolic effects of teriparatide.

In contrast to the studies of concurrent alendronate and PTH, studies using less potent antiresorptive agents concurrently with PTH suggest a different type of response. In a 6 month study of 137 women with postmenopausal osteoporosis, subjects were randomized to receive teriparatide alone or teriparatide plus raloxifene11. It was found that the increase in bone resorption markers was significantly less and total hip BMD by DXA was significantly increased with combination therapy compared to teriparatide alone. This suggests that concurrent treatment with raloxifene may enhance the bone forming effects of teriparatide. Further study is needed to validate these findings. The response to PTH(1-34) given for 3 years was evaluated in 17 women with postmenopausal osteoporosis who were on estrogen replacement therapy that was continued for the duration of treatment with the PTH(1-34)12. The anabolic response to PTH was robust, suggesting that estrogen may not reduce the effects of PTH. This is supported by data in a larger study of 52 women with osteoporosis who had been on hormone replacement therapy (HRT) for at least 2 years and were then randomized to receive ongoing HRT alone or HRT combined with PTH(1-34) for 3 years13. In this study, there was a BMD increase and reduction in morphometric vertebral fracture risk in those treated with HRT + PTH compared to those on HRT alone. The fracture risk data in this study should be interpreted with caution due to the small number of events (a total of 7–14 fractures in both groups depending on the definition of incident fracture).

Studies showing a blunting of the anabolic effect of PTH with concurrent use of alendronate may not apply to all bisphosphonates. In a recent 1-year, partially double-blinded, randomized study, 412 women with postmenopausal osteoporosis were assigned to receive a single intravenous infusion of zoledronic acid, daily subcutaneous teriparatide, or both14. It was found that teriparatide increased spine BMD more than zoledronic acid and zoledronic acid increased hip BMD more than teriparatide, while combination therapy provided the largest and most rapid response when both skeletal sites were considered. With combination therapy there was an attenuation of the increase in BTMs compared to teriparatide alone and a different pattern: an initial increase in a marker of bone formation, followed by a dip, and then a progressive increase, while with the marker of bone resorption there was an initial dip followed by a progressive increase. The implication is that combination therapy with zoledronic acid and teriparatide may have a role in the treatment of some patients at high risk for hip fractures.

Sequential therapy with an antiresorptive drug followed by an osteo-anabolic drug

Patients treated with an osteo-anabolic agent have commonly been treated previously with an antiresorptive drug. It is therefore instructive to know whether this has any influence on the efficacy of osteo-anabolic therapy, and whether it is preferable to stop the antiresorptive drug or continue it when starting the osteo-anabolic drug. The ‘Anabolic After Antiresorptive’ (AAA) study was a prospective, open label, nonrandomized 18 month trial evaluating the effects of teriparatide on BMD and BTMs in 59 postmenopausal women previously treated with raloxifene or alendronate15. It was found subjects in both groups responded to teriparatide, although the increase in BMD and BTMs with teriparatide was delayed and attenuated with prior alendronate treatment, while the response following raloxifene treatment was similar to what has been seen in treatment naïve patients in other trials. The ‘Open-label study to determine how Prior Therapy with Alendronate or risedronate in postMenopausal women with osteoporosis InfluenceS the clinical Effectiveness of teriparatide’ (OPTAMISE) reported findings in a modified intent-to-treat population of 317 subjects treated with teriparatide for 12 months16. In subjects previously treated with risedronate, the BMD and BTM responses were greater than with alendronate. Taken together, these studies suggest that there are differences in the response to teriparatide according to the type of previous treatment, perhaps due to differences in their antiresorptive potency, affinity for hydroxyapatite, and offset of effect. Whether these differences are clinically important is unknown.

A randomized open label trial in postmenopausal women previously treated with either alendronate or raloxifene for at least 18 months evaluated the effects of switching to teriparatide alone or adding teriparatide to ongoing antiresorptive therapy, with a follow-up period of 18 months17. Regardless of whether prior treatment was with alendronate or raloxifene, greater increases in BTMs were achieved by switching to teriparatide, whereas greater BMD increases were achieved by adding teriparatide. The safety data for adding teriparatide appeared to be favorable, although it is unclear whether there is any clinical benefit in adding compared to switching.

Sequential therapy with an osteo-anabolic drug followed by an antiresorptive drug

There is evidence from several studies that the beneficial effects of PTH will dissipate if not followed by an antiresorptive drug. In a pre-specified second year of the PaTH study, 119 women with postmenopausal osteoporosis who received 1 year of treatment with PTH(1-84) were randomized to receive either alendronate or placebo for the following year18. BMD increased or remained stable for the alendronate group, while BMD rapidly decreased in the placebo group. Similar results were found in another study of PTH(1-84) followed by alendronate in women with postmenopausal osteoporosis19 and in a study of PTH(1-34) followed by a bisphosphonate (alendronate, risedronate, or etidronate) in men with osteoporosis20. These findings strongly suggest that PTH treatment should be followed by antiresorptive treatment to consolidate or enhance the gains in BMD with PTH alone.

Intermittent therapy with an osteo-anabolic drug with or without an antiresorptive drug

Several studies have evaluated various ways of cycling osteo-anabolic therapy, sometimes including antiresorptive therapy as a component of the regimen. When 21 treatment-naïve men and women were given a 2 year course of therapy with daily subcutaneous teriparatide followed by a 1 year drug-free period and then a third year of daily subcutaneous teriparatide, there was a substantial improvement in lumbar spine BMD and increase in BTMs with the first course of teriparatide, with the greatest changes observed in the first 6 months21. There was a decrease in BMD and BTMs in the drug free year and an attenuated BMD and BTM response to the third year of teriparatide treatment, with restoration of BMD to the level achieved after the initial 2 years of teriparatide treatment but not exceeding it. This study validated previous findings of a rapid loss of BMD when teriparatide therapy is not followed by antiresorptive therapy, and did not show benefit with a third year of teriparatide beyond restoring BMD that had been lost the previous year. Another study of 15 months duration (n = 126) assessed the effects of cycling PTH(1-34) with a regimen of 3 months on and 3 months off in osteoporotic women taking alendronate, compared with continuous concurrent PTH(1-34) and alendronate, and alendronate only22. All of the women had taken alendronate for at least 1 year prior to the start of the study. Both cyclic and daily regimens of PTH(1-34) increased lumbar spine BMD and bone formation markers. This study showed that patients with ongoing alendronate therapy are capable of responding to osteo-anabolic therapy; the response to cyclic PTH(1-34) was similar to daily PTH(1-34) but with a lower cost and less inconvenience to the subjects. Of the 126 patients in this study, 32 who had been treated with either cyclic or daily teriparatide for 15 months followed by 12 months of alendronate alone were enrolled in a retreatment study and received an additional 15 months of daily subcutaneous teriparatide23. Retreatment resulted in an increase in lumbar spine BMD and an increase in bone formation markers similar to what was seen in the original course of teriparatide, suggesting that some patients at high risk for fracture may benefit from retreatment with teriparatide.

Monotherapy that may uncouple bone resorption and bone formation

Strontium ranelate, a drug that is approved for the treatment of postmenopausal osteoporosis in Europe and some other world regions, may have both antiresorptive and osteo-anabolic properties24. There is evidence suggesting that the investigational drug odanacatib, a cathepsin K inhibitor, may partially uncouple bone resorption and formation, with less of a reduction in bone formation than other antiresorptive drugs25. Inhibition of sclerostin with an investigational monoclonal antibody, AMG 785, appears to increase bone formation and reduce bone resorption26.

There is still much to be learned about optimal use of combinations of antiresorptive and osteo-anabolic drugs. Although a cure for osteoporosis remains a worthwhile goal, we are not there yet. Monotherapy remains the current standard of care. There is no regulatory approval for the use of concurrent drugs for the treatment of osteoporosis. The most important message for clinicians is that treatment with teriparatide or PTH(1-84), agents that can at least partially restore lost bone microarchitecture, must be followed by antiresorptive therapy in order to maintain or enhance the benefit achieved. Alendronate taken before or concurrently with PTH may attenuate or delay its skeletal effects. Risedronate taken before teriparatide does not appear to reduce the osteo-anabolic effects to the same extent as alendronate. Zoledronic acid given at the same time as starting teriparatide may offer some benefit beyond either drug given alone. A mild antiresorptive drug, such as estrogen or raloxifene, taken before or concurrently with an osteo-anabolic drug does not appear to diminish, and may enhance, the skeletal effects. Virtually all of the studies with combination therapy evaluated changes in BMD and BTMs, with no convincing evidence of greater fracture risk reduction with any combination. Investigational drugs that partially or totally uncouple bone resorption and formation offer the potential of benefit beyond currently available agents. More study is needed to continue the quest for the Holy Grail of osteoporosis therapy.

Transparency

Declaration of financial/other relationships

The author has received grant/research support from Amgen, Merck, Eli Lilly, Novartis, Warner Chilcott, and Genentech. He has served as a consultant, advisory board member, speakers bureau participant, or given presentations at sponsored speaking events for Amgen, Eli Lilly, Novartis, and Genentech. The author has not received funding for preparation or writing of this manuscript.