PTH-related peptide (PTHrP) was first brought to medical attention as a cause of humoral hypercalcemia of malignancy. It was subsequently demonstrated to have key physiological actions in the skeleton as well as other systems [30]. Despite the fact that both PTH and PTHrP act at a common receptor, the proteins are products of different genes. While there is limited overall sequence homology between the two peptides, the N-terminal regions are intensely homologous. This N-terminal sequence homology helps to explain the shared interaction with a common PTH–PTHrP receptor. In the normal state, PTHrP has evolved to regulate local tissue functions, as opposed to the systemic hormonal role of PTH [31]. Similar to PTH in primary hyperparathyroidism, PTHrP is catabolic to bone when administered continuously, and therefore intermittent administration has been studied. Intermittent administration of PTHrP increases bone mass in rodents with varying potency in relationship to PTH [32].

Initial short-term studies with PTHrP(1-36) by Horwitz et al. [33, 34] showed early effects that appeared to favor a rather exclusive stimulation of bone formation. Results from a phase II study [35] in 105 postmenopausal women aged 45–75 years (n = 35 in each group) are now available. Subjects were randomized to daily treatment with PTHrP(1-36) 400 μg, PTHrP(1-36) 600 μg, or PTH(1-34) 20 μg for 3 months. The primary outcome measures were bone turnover markers, with secondary outcome measures of BMD and safety. PTH(1-34) and PTHrP(1-36) stimulated bone formation early (day 15), although by day 90 PTH(1-34) increased bone formation markers two to fourfold greater than PTH(1-36) at the 600 or 400 μg doses, respectively (p < 0.05). As expected, the increase in bone resorption occurred later (day 60 for PTH(1-34) and day 90 for both PTH(1-36) groups), and was not as dynamic. The increase in bone resorption at day 90 was threefold greater for the PTH(1-34) arm than either of the PTH(1-36) dosage groups (p < 0.05), which were not different from each other. At 3 months, PTH(1-34) and PTHrP(1-36) at both doses significantly increased BMD by about 2 % at the lumbar spine. There were small but significant increases in hip BMD in the PTHrP(1-36) groups but not in the PTH(1-34) group. There were no significant differences in BMD at the forearm. Adverse effects were similar between the PTH(1-34) and PTHrP(1-36) groups, with the exception of more frequent episodes of mild hypercalcemia in the PTHrP(1-36) group.

A PTHrP analog is being developed by Radius Health, Inc. [36]. The first 22 residues are identical to PTHrP but thereafter, from amino acids 22 to 34, the molecule contains several replacement residues designed to optimize its osteoanabolic potential. The results of a multicenter, randomized, double-blind, placebo-controlled phase II investigation of safety and effects on BMD and bone markers were recently presented [36]. Postmenopausal women with osteoporosis, aged 55–85 years, were randomized to placebo, BA058 20, 40, or 80 μg, or teriparatide 20 μg for 24 weeks. After completion of the 24-week study, subjects were then eligible to participate in a 24-week extension study. One hundred and eighty-four of 221 patients completed 6 months of treatment. The mean percent change in lumbar spine BMD at 24 weeks was 1.6 % with placebo, 2.9 %, 5.2 %, and 6.7 % with BA058 20 μg, 40 μg, and 80 μg, respectively, and 5.5 % with teriparatide (p < 0.001 compared to placebo for BA058 40 and 80 μg and teriparatide). Further dose-dependent increases in lumbar spine BMD were noted during the extension (n = 55), with a mean percent change at 48 weeks of 0.7 % with placebo, 5.1 %, 9.8 %, and 12.9 % with BA058 20 μg, 40 μg, and 80 μg, respectively, and 8.6 % with teriparatide. The investigators also found dose-dependent increases in total hip BMD. At 24 weeks, changes were noted in serum and urine bone turnover markers from baseline (p ≤ 0.05) for BA058 40 and 80 μg and for teriparatide. BA058 was well tolerated, with an adverse event profile comparable with blinded placebo. BA058 is the focus of a phase III placebo-controlled, 18-month international study in postmenopausal women [24]. The primary endpoint is the incidence of new vertebral fractures at a dose of 80 μg in comparison to placebo. A transdermal microneedle technology is also under development for BA058 and in phase I and II trials (www.​radiuspharm.​com).

In theory, the combination of an antiresorptive and osteoanabolic agent offers the potential for increased efficacy over monotherapy with either drug class given their differing mechanisms of action. If bone resorption is inhibited by an antiresorptive while bone formation is stimulated by an osteoanabolic agent, combination therapy might give better results than therapy with either agent alone. Despite the intuitive attraction of this reasoning, important data to the contrary have been provided by Black et al. [25] and Finkelstein et al. [37]. These two groups independently conducted trials using a form of PTH alone, alendronate alone, or the combination of the PTH formulation and alendronate. Black et al. studied postmenopausal women treated with 100 μg of PTH(1-84) and Finkelstein et al. studied men given 40 μg of teriparatide. PTH monotherapy was associated with greater densitometric gains than with combination therapy or alendronate alone at the lumbar spine with both dual energy X-ray absorptiometry (DXA) and quantitative computed tomography. Combination therapy was not different from alendronate alone. Bone turnover markers followed the expected course for monotherapy with anabolic or antiresorptive agents. For combination therapy, however, bone markers followed the course of alendronate, not PTH, with reductions in bone formation and bone resorption markers. The findings from these two trials suggest that the impaired response to combination therapy was due to the dominating effects of alendronate on bone remodeling dynamics when both drugs are used in combination.

The results of these combination therapy studies with alendronate led to the concept that an antiresorptive agent that did not impair the anabolic actions of PTH to increase bone formation while mitigating its effects on bone resorption may be a more effective approach to combination therapy. Deal et al. [38] studied the effects of raloxifene, a less potent antiresorptive agent than the bisphosphonates, in combination with teriparatide over 6 months. Their results support the idea that combination therapy might be advantageous with a mild antiresorptive drug but are not conclusive because of the short duration of the study. As a further test of this hypothesis, Walker et al. [39] investigated the combination of teriparatide and risedronate, a bisphosphonate with less potent effects on bone turnover than alendronate or zoledronic acid. Men with low BMD were randomized to receive risedronate 35 mg weekly plus daily injected placebo, teriparatide 20 μg subcutaneously daily plus weekly oral placebo, or risedronate plus teriparatide (combination) for 18 months. At study conclusion, all three treatment arms significantly increased lumbar spine BMD, the primary endpoint, but there were no between-group differences. In contrast, at the total hip, BMD increased to a greater extent in the combination group (3.86 ± 9.2 %) vs. teriparatide (0.29 ± 8.0 %) or risedronate alone (0.82 ± 8.0 %; p < 0.05 for both). Bone turnover markers in the combination group paralleled the teriparatide alone arm, supporting the idea that an antiresorptive that does not have a profound effect on bone turnover might permit salutary effects of combination therapy with an osteoanabolic agent. The results of this proof-of-concept study are favorable but require further investigation.

Cosman et al. [40] studied the use of a single dose of zoledronic acid in combination with daily teriparatide. This approach was based upon animal studies of Gasser et al. [41] in which a single dose of zoledronic acid led to greater improvements in BMD in rats treated simultaneously with teriparatide than those treated with bisphosphonate therapy alone. With the combination of zoledronic acid and teriparatide, BMD increased after 6 months at the spine and hip to a greater degree than in either monotherapy arm. With combination therapy, there was a rapid, but only transient, reduction in bone turnover markers. At the lumbar spine, there were significantly greater changes after 6 months with combination therapy but by the end of the study at 12 months, there were no differences between combination therapy and teriparatide alone (7.3 % vs. 7.3 %; p = NS). At the hip, similar results were observed with the combination therapy arm showing a greater enhancement of BMD than zoledronic acid alone, but by the 12-month endpoint of the study, differences were no longer appreciated (2.3 % vs. 2.2 %; p = NS). However, if one considered the lumbar spine and hip sites together as a composite endpoint, only combination therapy provided improvement in BMD that was greater than either zoledronic acid or teriparatide alone.

The first section of this chapter is devoted to denosumab, a potent inhibitor of RANKL. Several interesting properties have surfaced over continued investigation of this therapy. The distal 1/3 radius, a cortical site, increases and remains above baseline for 8 years. The second observation is that PTH levels rise after denosumab administration by approximately twofold and remain above baseline for about 3 of the 6-month interval between doses [5, 42, 43]. To relate these observations to each other requires appreciation of the fact that PTH requires RANKL for its catabolic actions [44–46]. As a RANKL inhibitor, denosumab blocks this catabolic pathway. The increase in PTH associated with denosumab, therefore, may preferentially exploit the anabolic Wnt signaling pathway [47, 48]. The increase in cortical bone density (distal 1/3 radius) is compatible with this hypothesis. Seeman et al. [49] have provided additional evidence for this view when they showed that the higher PTH in connection with denosumab is associated with lower cortical porosity. In the control arm of this study, alendronate administration was associated with increases in PTH but there was a positive relationship with higher PTH levels associated with greater porosity. These observations led to the hypothesis that denosumab and teriparatide in combination may be more beneficial than the combination of teriparatide with other antiresorptives. The Denosumab and Teriparatide Administration (DATA) Study [50] has in fact shown a densitometric benefit to combination therapy in postmenopausal women. At 12 months, lumbar spine BMD increased more in the denosumab and teriparatide combination group (n = 30; 9.1 %) than the teriparatide alone group (n = 31; 6.2 %, p = 0.014) or denosumab alone (n = 33; 5.5 %, p = 0.0005). Femoral neck BMD also increased to a greater extent in the combination group (4.2 %) than the teriparatide alone (0.8 %, p = 0.0007) or denosumab alone arms (2.1 %, p = 0.0238), with similar findings at the total hip site. There was an increase of 2.6 % in BMD at the distal 1/3 radius in the combination therapy arm and 1.7 % in the denosumab alone arm without between-group differences, although both groups differed when compared to teriparatide alone, which demonstrated a 1.8 % decline in BMD. As expected, teriparatide alone increased bone turnover markers and denosumab alone decreased bone turnover markers. While the combination therapy group decreased the bone resorption marker CTX to a similar extent as the denosumab alone group, bone formation decreased more gradually and to a lesser extent.

It should be noted that these and other combination studies have not been designed with fracture outcome as a definitive endpoint. These trials have only targeted surrogate endpoints such as BMD and bone turnover markers, and have also been shorter than typical definitive fracture trials. One must therefore exercise caution in the interpretation of these data. In addition, the use of two pharmacologic therapies simultaneously carries with it the potential for more adverse events than the use of a single therapy as well as added expense.

Cathepsin K, a member of the papain family of cysteine proteases, is highly expressed by activated osteoclasts, and promotes degradation of type I collagen (Fig. 12.1) [51]. It plays an important role in the process of bone resorption through its actions to remove collagen, an initial step in the creation of the bone remodeling unit [51]. Clues to the therapeutic potential of inhibiting the actions of cathepsin K came from studies of a human genetic disorder, pycnodysostosis, a disorder of excess bone due to an inactivating point mutation in the gene encoding cathepsin K [52, 53]. Several oral compounds have been designed to inhibit cathepsin K [54]. Preclinical studies of cathepsin K inhibitors noted that the drug reduces bone resorption without suppressing bone formation to an appreciable degree [55–57]. Different from the bisphosphonates, cathepsin K inhibitors appear to permit certain functions of the osteoclast such as signaling to the osteoblast while preventing its classical actions to excavate bone.