Teriparatide (recombinant human parathyroid hormone 1-34) in postmenopausal women with osteoporosis: systematic review

ABSTRACT CONTEXT AND OBJECTIVE: Osteoporosis is defined as a disease characterized by low bone mass and deterioration of the bone tissue microarchitecture. Teriparatide stimulates the formation and action of osteoblasts, which are responsible for bone formation, thus promoting bone tissue increase. The aim was to assess the effectiveness and safety of teriparatide for treating postmenopausal osteoporosis. METHODS: A systematic review was conducted using the Cochrane Collaboration methodology. RESULTS: 1) Teriparatide 20 µg or 40 µg versus placebo: there was a benefit from teriparatide, considering the following outcomes: reduction in the number of new vertebral and non-vertebral fractures, and increased whole-body, lumbar and femoral bone mineral density. 2) Teriparatide 40 µg versus alendronate 10 mg/day for 14 months: there was no statistical difference regarding the incidence of new vertebral or non-vertebral fractures, although in the group that received teriparatide there was greater bone mineral density increase in the whole body, lumbar column and femur. 3) Estrogen plus teriparatide 25 µg versus estrogen: there was a benefit, considering the following outcomes: reduction in the number of new vertebral fractures, and increased whole-body, lumbar and femoral bone mineral density after three years. CONCLUSION: When teriparatide is intermittently administered in low doses, it reduces the incidence of vertebral fractures (67%) and non-vertebral fractures (38%) and increases bone mineral density in the lumbar column and femur. There is a need for studies with longer observation in order to allow conclusions regarding the safety and duration of the therapeutic effects.


INTRODUCTION
Osteoporosis is a disease that is characterized by low bone mass and deterioration of the bone tissue microarchitecture, which cause bone fragility and increased risk of fractures. 1t is defi ned by bone densitometry as bone mineral density values lower than 2.5 standard deviations below the reference values. 2 Fractures caused by low-level impact and increased mortality are the main outcomes from osteoporosis.The causes of the increased mortality are principally pulmonary thromboembolism, infections and systemic surgical and postoperative complications. 3,4The results from epidemiological studies show that there is a mortality rate of 14 to 34% over the fi rst year following a hip fracture. 3,4 50-year-old woman presents a risk of osteoporotic fracture during her remaining lifetime of 17.5% for the femoral neck, 15.6% for the vertebrae, 16% for the distal radius, and around 40% in any other location in the skeleton. 5The etiology of approximately 70% of hip fractures has been attributed to osteoporosis, with mortality ranging from 12 to 34%. 6ractures most frequently occur in the vertebral column, hip or wrists, and they can occur even in the absence of falls.Between 50 and 100% of adults who have had one fracture will have a second one. 3,4,7ip fractures may cause incapacity or death.Such patients present two to three times greater chance of dying than do individuals who have not had fractures. 8After suffering a hip fracture, only 30 to 40% of such individuals return to their usual activities. 8n the United States, approximately 10% of these patients become incapacitated following hip fractures, and 19% of them are institutionalized. 9In 2002, the expenditure on patients with osteoporosis exceeded 17 billion dollars, taking into account hospitalizations, home visits, physiotherapy and nursing. 10evention and treatment are undertaken through dietary counseling, exposure to the sun, exercises, prevention of falls, calcium and vitamin D supplementation and pharmacological treatment.2][13] Calcium supplementation (1000-1500 mg per day) and vitamin D supplementation reduce the risk of vertebral fracture, but not the risk of non-vertebral fracture.They increase the bone density in the lumbar column by 1.5 to 2%. 1,5,62][13] Bisphosphonates inhibit bone reabsorption by inhibiting osteoclasts.They increase bone mineral density by 10% and reduce the risk of vertebral and non-vertebral fracture by 40 to 50%. 1,5,6Calcitonin is an anti-reabsorption medication that reduces vertebral fracture by 20%, but not non-vertebral fracture, and it increases the bone mineral density in the vertebral column and femur. 1,5,6There is evidence that the use of calcitonin decreases the pain following vertebral fracture. 1,5,6edication for selective estrogen receptor modulation (raloxifene) presents an estrogenlike effect on bones and fat metabolism and an antagonist effect on the breast and uterus.It leads to a slight increase in bone mineral density and a reduction in vertebral fractures, but not in non-vertebral fractures.Patients may present adverse effects consisting of thromboembolic events, with a relative risk of 3.29. 1,5,6Hormone replacement therapy increases the bone mineral density and reduces vertebral fractures by 34% and non-vertebral fractures by 27%, but has the adverse effects of increasing the risks of breast cancer and thromboembolic events.It must therefore not be indicated for treating osteoporosis. 1,5,6eriparatide, the recombinant 1-34 fragment of human parathyroid hormone (rhPTH1-34) differs from the other therapeutic options for treating osteoporosis, such as bisphosphonates or calcitonin, which act by reducing bone reabsorption.Instead, it stimulates the formation and action of osteoblasts, which are the cells responsible for bone formation, thereby promoting increases in bone tissue. 14hen teriparatide is intermittently administered in low doses, it presents an anabolic effect on bones, because it stimulates osteoblasts.This medication has been contraindicated for patients who are at risk of developing osteosarcoma, such as those with unexplained high alkaline phosphatase levels, those with Paget's disease, or those who have undergone irradiation on the bones.Teriparatide has also been contraindicated for patients with hyperparathyroidism or hypercalcemia. 1,5,6

OBJECTIVE
The aim was to determine the effectiveness and the safety of teriparatide for treating postmenopausal osteoporosis.

MATERIALS AND METHODS
This systematic review of the literature was developed in accordance with the methodology of the Cochrane Collaboration and conducted at the Brazilian Cochrane Center, in the Universidade Federal de São Paulo -Escola Paulista de Medicina (Unifesp-EPM), on request from the Brazilian Ministry of Health.
The review only included randomized controlled clinical trials that evaluated the use of teriparatide for treating women with a diagnosis of natural or surgical postmenopausal osteoporosis.All clinical trials comparing teriparatide against placebo, hormone replacement, bisphosphonates, calcitonin or calcium in association with vitamin D could be included.
The outcomes evaluated were: a) percentage changes in bone mineral density, as quantified by dual-energy X-ray densitometry apparatus or by quantitative computed tomography of the femoral neck, femoral trochanter, hip, lumbar column or forearm; b) number of vertebral and non-vertebral fractures; c) toxicity, as measured by hypercalcemia, gastrointestinal symptoms or abandonment of treatment due to adverse effects.

Search strategy for identifying studies
The search strategy included the following databases: Medical Literature Analysis and Retrieval System Online (Medline) (1996  The terms used in the databases were: osteoporosis, postmenopausal, parathyroid hormone (PTH) and teriparatide (Chart 1).Clinical trials were selected if they fulfilled the criteria that they were randomized controlled trials on postmenopausal populations and that they evaluated bone mineral density and fractures (Figure 1).

Data extraction and methodological quality assessment
The search strategy identified the relevant articles.Each of these articles was reviewed by two independent reviewers.All the data were extracted by these two reviewers.Details relating to the population, treatment periods and demographic baseline were extracted independently.A third reviewer was consulted to help in resolving differences.The quality of each trial was evaluated independently by the two reviewers, using the validated quality  assessment tool that was published by Jadad et al. 15 in 1996.

Statistical analysis and presentation of the results
The statistical analysis was carried out using the Manager program (RevMan, 2000), and in accordance with the Cochrane Collaboration Handbook. 16For dichotomous variables, the odds ratio (OR) method was used, with 95% confidence intervals (random effect model).When there was a statistical difference, the number needed to treat (NNT) or the number needed to harm (NNH) was calculated.For continuous variables, the weighted mean difference was calculated (random effect model) with the corresponding 95% confidence interval.If necessary, the original data were transformed into a logarithmic basis to obtain better distribution, or into scales that presented similar properties (the data on this scale would be the input for meta-analysis).In addition, if necessary, the continuous variables were subdivided for dichotomous analysis.
To analyze the sensitivity, the following strategy using the Review Manager (RevMan, 2000) was proposed: 16 a) reanalysis of the data using reasonable variation of values for lost data: when dichotomous variables were extracted, it was assumed that participants lost from the experimental group presented unsuccessful treatment and that losses from the control group presented improvement; b) reanalysis of the data using reasonable variation of the results from the studies, when there was some uncertainty in the results; c) reanalysis of the data using different statistical methods; d) statistical heterogeneity: it was planned that this would be evaluated in the studies by inspection of the graphical presentation (a dispersion graph in which the study weight or sample size was put on the yaxis, versus the risk ratio on the x-axis), and by the heterogeneity test (chi-squared test with n degrees of freedom, in which n was the number of studies that contributed data, minus one).
The studies included in the review are summarized in Table 1 and the studies excluded, together with the reason for exclusion, are identified in Table 2. [17][18][19][20][21][22][23][24][25][26][27][28][29][30][31] The studies included were: -Neer et al. 32     Different intervention eral density in postmenopausal women with osteoporosis.This was a paper originating from a double-blind randomized clinical trial with 24 months of follow-up.It included 1637 women who had been postmenopausal for at least five years and who had suffered at least one moderate or two slight non-traumatic fractures.An additional inclusion criterion for women who had had fewer than two fractures was presentation of hip or lumbar column bone mineral density of at least one standard deviation below the mean.They were administered teriparatide 20 µg or 40 µg daily, or placebo.The outcomes were: 1) new fractures; 2) bone mineral density of the lumbar column, proximal femur, radius and whole body measured on the Lunar, Hologic or Norland densitometry apparatus; and 3) adverse effects.The loss during follow-up was 311 patients.This study was performed in the United States with pharmaceutical industry sponsorship.

The skeletal response to teriparatide is largely independent of age initial bone mineral density, and prevalent vertebral fracture in postmenopausal women with osteoporosis. Effect of parathyroid hormone (1-34) on fracture and bone mineral density in postmenopausal women with osteoporosis.
This paper originated from the same clinical trial as described in the preceding paper, but assessing different outcomes, since the patients included were subdivided into three groups according to age group (less than 65 years old, from 65 to 75 years old, and more than 75 years old).-Body et al. 34

RESULTS
For the purposes of statistical analysis, the comparisons were made according to the control group: 1) placebo; 2) alendronate; 3) hormone replacement; and 4) other interventions.

Teriparatide versus placebo
This megatrial, 32,33 which included 1637 postmenopausal women, gave rise to two papers (Neer et al., 32 and Marcus et al., 33 ) that assessed different outcomes and have therefore been included separately in this review.
There was a statistically significant benefit from using teriparatide  33 demonstrated that the reductions in the risk of new fractures in the teriparatide group were similar, independent of age (p = 0.558).
Likewise, the increase in bone mineral density did not show dependency on the baseline bone mineral density (p = 0.615).
The relative risks of new fractures were similar between the subgroups and there was no association with the baseline bone mineral density (p = 0.817).Furthermore, there was no statistically significant difference between the two doses of teriparatide, in relation to reduction in the risk of new fractures (p = 0.649).

Teriparatide versus alendronate
There was no statistical difference in relation to the incidence of new vertebral or non-vertebral fractures (RR 0.3; CI 0.09 to 1.05). 34

Teriparatide plus estrogen versus estrogen
There was a statistically significant benefit from using estrogen in association with teriparatide, considering the following outcome: 35 reduction in the number of new vertebral fractures (RR 0.15; CI 0.03 to 0.73).The data relating to bone mineral density could not be extracted for analysis but, by the end of the three years of the study, there had been increases of 13% in lumbar bone mineral density, 2.7% in femoral bone mineral density and 7.8% in whole-body bone mineral density in the teriparatide/estrogen group.There was practically no increase in the group that received estrogen alone.

Others: teriparatide plus nafarelin versus nafarelin
The study that compared teriparatide with placebo (nafarelin versus nafarelin plus teriparatide 40 µg) for preventing bone mass loss among women with medication-induced menopause (nafarelin gonadotrophic hormone) 36 showed that, after six months of treatment, there was a bone mass increase of 3.4% in the teriparatide group and a bone mass loss of 3.5% in the placebo group, considering the lumbar column and whole body.However, there was a slight reduction in the bone mineral density of the femoral neck in both groups.

DISCUSSION
The results from this systematic review have shown that the use of teriparatide reduced the number of vertebral fractures (67%) and non-vertebral fractures (38%) and increased the bone mineral density of the lumbar column and femur, over the period when it was used.However, it is still not possible to assess the duration of these beneficial effects after the treatment was stopped.
The adverse effects attributed to teriparatide did not stop the patients from continuing with the treatment.However, some important points must be made with regard to the use and safety of this medication.Previous studies have demonstrated increased incidence of osteosarcoma in rodents treated with PTH, while there have not been any reports of cases in monkeys. 6The studies among humans have not had sufficient duration to find definitive evidence in this respect.Therefore, this medication must still be contraindicated for patients who are at risk of developing osteosarcoma, such as those with unexplained high alkaline phosphatase levels, those with Paget's disease, or those who have undergone irradiation on the bones.Moreover, it must be contraindicated for patients with hyperparathyroidism or hypercalcemia.
With regard to the ideal dose for teriparatide, the study by Neer et al., which was initially envisaged to last for three years but was halted after 19 months of use of the medication, showed that there were no differences between the therapeutic effects from 20 µg and 40 µg. 32Thus, a dose of 20 µg for a maximum period of 24 months was established. 17ecause of its high cost (approximately 14,000 dollars for two years 1,5,6 ) and considering that there are other effective interventions that are economically more viable for treating osteoporosis, teriparatide must only be prescribed after the other therapeutic options have been exhausted.

CONCLUSIONS
Teriparatide administered intermittently at doses of 20 µg or 40 µg, in comparison with placebo, reduced the incidence of new vertebral and non-vertebral fractures and improved the bone mineral density of the whole body and lumbar column, without apparent correlation with any serious adverse effects.Teriparatide (40 µg) was superior to alendronate (10 mg/day), considering bone mass increases for the whole body, lumbar column and femur (intertrochanteric and Ward's triangle regions), but there was no difference regarding the incidence of new fractures.

Table 2 .
Summary of the clinical trials excluded