Introduction

Spinal cord injury (SCI) is commonly associated with a marked bone loss and increased bone turnover early after SCI, leading to the development of osteoporosis and fractures, especially below the injury level [1–3]. Thus, more than 50 % of patients with complete SCI develop osteoporosis 1 year after injury [1], with a long-term overall prevalence of up to 81 % in these subjects [4]. However, at present, in spite of this well-known common complication, the therapeutic approach of these patients is clearly insufficient with less than 20 % of high-risk patients receiving antiosteoporotic treatment [2]. In addition, although treatment with i.v. or oral bisphosphonates may partly attenuate bone loss after SCI, these drugs have not shown effectiveness in increasing bone mass at sublesional sites, especially in the lower extremities [5–7] where most post-SCI fragility fractures occur [2]. Similarly, osteoanabolic agents such as teriparatide have not demonstrated a positive effect in this specific location [8], thereby indicating the need to improve the therapeutic approach in this clinical complication.

Denosumab, a monoclonal antibody against RANKL, is an essential mediator of osteoclast formation and survival, with a marked antiresorptive effect and proven efficacy in increasing bone mass and preventing fractures in osteoporotic subjects [9], with a notable positive effect on cortical bone, such as the proximal femur and forearm [10]. Thus, the use of denosumab may be of especial interest for treating SCI patients with osteoporosis. Indeed, increased bone tissue RANKL expression has been observed in a rodent model of SCI [11] further suggesting the potential therapeutic effect of denosumab for treating this clinical condition.

Herein, we report our experience in a group of individuals with recent complete SCI with associated osteoporosis treated with denosumab for 12 months.

Patients and methods

Study design and subjects

This is an ongoing prospective observational study with the main goal of analyzing the effect of recent SCI on bone mass and bone metabolism in these individuals. The detailed study design and characteristics of the participants has been previously published [1]. Briefly, the participants were recruited at the Neurorehabilitation Guttmann Institute (from June 2010 to December 2013) and followed at the Metabolic Bone Diseases Unit of the Hospital Clinic. Antiosteoporotic therapy was recommended by the attending physician when participants developed densitometric osteoporosis during follow-up. In subjects with low serum values of 25-hydroxyvitamin D (25OHD) (<20 ng/mL), treatment with vitamin D supplements was recommended. Ethical approval was obtained and all participants provided written informed consent.

Forty-two individuals were included in the study with 26 being followed for 12 months. Herein, we present the data on the effect of antiosteoporotic treatment with denosumab in bone mineral density (BMD) and bone turnover markers (BTMs) evolution in those participants who developed osteoporosis during the follow-up. Thus, 14 participants with SCI-related osteoporosis were treated with denosumab (60 mg every 6 months) for up to 12 months.

Methods

Clinical assessment, biochemical determinations, and BMD measurements were performed at baseline and at 12 months of follow-up. Risk factors for osteoporosis were assessed in all participants as well as the body mass index (BMI), and the level (paraplegia/tetraplegia), type (spastic/flaccid), and severity of SCI according to the ASIA score [12]. Skeletal fractures and adverse events were also recorded according to the results of a questionnaire given to patients at the 6- and 12-month control visits and the medical records shared with other hospitals from our geographic area (Catalonia, Spain).

Biochemical measurements included serum creatinine, calcium, phosphate, and 25OHD levels (Liason DiaSorin). The following BTMs were assessed: bone alkaline phosphatase (bone ALP by IDS, Vitro), propeptide amino-terminal of type I procollagen (PINP by Cobas e411, Roche), and serum carboxy-terminal telopeptide of type I collagen (sCTx by Cobas e411, Roche). The intra- and inter-assay coefficients of variation for each marker were as follows: bone ALP 2.9 and 5.8 %, PINP 2.8 and 4.3 %, sCTx 2 and 5.7 %, respectively.

BMD of the lumbar spine, femoral neck, and total hip was measured by dual X-ray absorptiometry (Lunar Prodigy, Radiation Corporation Madison, WI). Osteoporosis was defined according to the WHO criteria [13]. The coefficients of variations for total femur and lumbar spine were 0.6 and 0.8, respectively.

Quantitative variables are described using means and standard deviations and frequencies, and percentages are reported for qualitative variables. To assess the BMD and BTMs evolution, paired t tests were applied to compare the 12 months and baseline means. Pearson’s correlation coefficient was used to analyze the association between the BMD evolution and BTM changes. All hypothesis tests with a p value lower than 5 % were considered significant.

Results

The clinical characteristics of the participants included in the study are shown in Table 1. Briefly, most participants were young males (mean age 39 years old) with complete motor involvement (ASIA 12A, 1B, 1C) from C4 to T8, all requiring a wheelchair for mobility. Traffic accidents were the most frequently associated cause of SCI. Biochemical determinations including calcium, phosphate, and 25OHD were within normal range previous to initiating treatment (with a mean time-since-injury at this time of 15.2 ± 4 months), whereas BTMs (bone ALP, PINP, sCTx) were increased at this time (Table 1).

Table 1 Clinical, biochemical, and densitometric characteristics of SCI participants at baseline
Full size table

At 12 months, SCI participants treated with denosumab showed a significant increase in BMD at total hip (+2.4 ± 3.6 %, p = 0.042), femoral neck (+3 ± 3.6 %, p = 0.006), and lumbar spine (+7.8 ± 3.7 %, p < 0.001) compared to baseline values (Fig. 1a). In addition, denosumab treatment was associated with a significant decrease in BTMs (bone ALP −42 %, p < 0.001; PINP −58 %, p < 0.001, sCTx −57 %, p = 0.002) at 12 months (Fig. 1b). BMD evolution was not related to BTM changes or to 25OHD serum levels.

Fig. 1
figure 1

Percentage of changes (±standard deviation) in BMD at the proximal femur and lumbar spine (a) and in the BTMs values (b) at 12 months of denosumab treatment. *p < 0.05 compared with baseline values

Full size image

No skeletal fractures were observed during follow-up and no serious treatment-related adverse events were noted during the study. In addition, no local site reactions nor skin infections or respiratory infections were observed. A soft tissue infection after traumatic skin abrasion was observed in one individual, which was resolved with oral antibiotic therapy. Moreover, 9 participants developed a total of 29 urinary tract infections during the 12-month follow-up period of treatment with denosumab. However, all of these individuals had also presented episodes of urinary tract infection during the period prior to initiating treatment with denosumab with no significant difference in the frequency of infections between the two periods of time (before denosumab treatment: 85.7 vs. 64 % during denosumab treatment, p = 0.649). No autonomic dysreflexia was reported during the 12-month follow-up period.

Discussion

This preliminary observational study shows for the first time the efficacy of denosumab in the treatment of SCI-related osteoporosis early after injury. Thus, treatment with denosumab increased lumbar and femoral BMD and decreased BTMs in these participants, suggesting the usefulness of this agent in the treatment of this common clinical complication.

In the present study, treatment with denosumab decreased bone turnover and not only prevented BMD loss after SCI but also significantly increased BMD 1 year after initiating treatment in all the locations evaluated, the lumbar spine and proximal femur. Thus, BMD increased about 3 % at the proximal femur including both the femoral neck and total hip, and ~8 % at the lumbar spine. Although the specific mechanism related to the maintained increase in cortical bone mass areas, such as the proximal femur, with denosumab treatment are not fully understood, recent studies have attributed this finding to both an increase in bone modelling, mostly affecting endocortical and periosteal surfaces, and a decrease in bone remodelling [14]; the latter explaining the marked reduction in BTMs with this treatment. On the other hand, it should be noted that the spontaneous BMD evolution in other locations after SCI such as the lumbar spine may differ. Indeed, maintenance or a slight increase in BMD in this location has been described after SCI, in relation to the continued weight-bearing function in this site in these individuals [15, 16]. Although we cannot totally rule out a spontaneous BMD increase in this specific location, the magnitude of the increase (~8 %) and the absence of radiological artefacts in the DXA scans in these subjects (none had undergone surgical instrumentation in the lumbar spine) further suggest an additional effect of the antiosteoporotic treatment.

Notably, although this is a preliminary study, denosumab increased BMD at the sublesional level in our participants, thereby being the first antiosteoporotic treatment able to improve bone mass in subjects with osteoporosis related to recent SCI to date. Thus, our participants had an early complete SCI with a mean time since SCI of 15 months, when the highest bone loss after SCI is known to be produced [3, 15, 16] and when no previous antiosteoporotic treatment has demonstrated prevention of BMD loss at the sublesional level, especially in the lower extremities. In this sense, several oral or intravenous bisphosphonates, such as alendronate, pamidronate, or zolendronate, or even osteoanabolic agents such as teriparatide have only shown an attenuation of bone mass loss at the lower extremities early after SCI [5–8]. Indeed, preventive effects in BMD loss have only been observed with these agents long-term after SCI when either the magnitude of the BMD loss and/or the increase bone turnover markers is much lower [5, 6].

As expected, treatment with denosumab was associated with a significant decrease in all the bone turnover markers evaluated in our study (bone ALP, PINP, and sCTx) with nearly a 60 % decrease in their values at 1 year of follow-up. However, since a spontaneous decrease in BTMs was observed during the first years after SCI, we cannot totally rule out an influence of the natural evolution of BTMs after injury. Nevertheless, the magnitude of the decrease in these markers was similar to that previously reported in postmenopausal osteoporotic women and men with osteoporosis treated with this agent (~60 % for sCTx and/or PINP) [9, 10, 17], thereby suggesting a similar response in our participants.

Since RANKL is expressed on activated T and B lymphocytes and lymph nodes, a RANKL inhibitor such as denosumab could theoretically increase the risk of infection. Nevertheless, apart from skin infections (cellulitis), the overall incidence of infections in previous studies of patients treated with denosumab for osteoporosis was not increased [9], and similar results were observed in individuals treated with cancer-related complications, in whom the dose given and the periodicity of administration of denosumab are much greater [18]. In our study, one of the concerns about using denosumab for treating osteoporotic SCI subjects was its effect on the possible increased risk of infections, especially in the urinary tract, a well-known clinical complication of SCI [19]. Although this was an observational non-controlled study with a limited number of participants and length of follow-up, which is the main limitation to assess side effects, we did not observe serious adverse events related to this treatment. Thus, no local site reactions nor skin or respiratory infections were observed in our participants. Additionally, although 64 % of participants developed at least one urinary tract infection during the 12-month denosumab treatment, this figure was similar to that observed prior to initiating treatment and was also similar to the reported expected incidence in subjects with complete SCI [19, 20]. Nevertheless, the non-randomized nature of the present study and the small sample size prone to increased risk of infection require controlled randomized clinical trials to further address this issue.

In conclusion, our results indicate that in individuals with SCI-associated osteoporosis treatment with denosumab early after injury increases lumbar and femoral BMD 1 year after initiating therapy. Therefore, this agent might represent a promising therapeutic option for treating this clinical complication. Nonetheless, in view of the present results, further prospective randomized controlled studies analyzing the long-term efficacy of denosumab in a larger number of participants are needed.