Human parathyroid hormone for the treatment of osteoporosis in post-menopausal women

  • Protocol
  • Intervention

Authors


Abstract

This is the protocol for a review and there is no abstract. The objectives are as follows:

To determine the benefit and harm of hPTH in the treatment of postmenopausal osteoporosis.

Background

Description of the condition

Osteoporosis is a chronic disorder of bone remodeling in which bone resorption exceeds bone formation. Reduced bone mass and deteriorated bone structure contribute to an increased fracture risk, primarily of the spine, the hip and the forearm. Osteoporosis predominantly affects postmenopausal women, but also men and patients on long-term corticosteroid therapy. Diagnosis is based on measurement of bone mineral density and defined as more than 2.5 standard deviations below the mean of peak bone mineral density for the same sex. Osteoporosis is a widespread disease with a considerable medical and socioeconomic burden. The US Department of Health and Human Services projects that the disease will have an impact on more than 10 million women by 2020 (Schuiling 2011). Treatment for osteoporosis comprises mostly antiresorptive drugs (e.g. bisphosphonates, denosumab, calcitonin, oestrogen, selective oestrogen receptor modulators). Bone anabolics, such as human parathyroid hormone preparations, are an alternative approach and currently prescribed for osteoporosis treatment and the prevention of osteoporotic fractures (e.g. Briot 2012). Combination therapy (antiresorptive and anabolic drugs) for the management of postmenopausal osteoporosis is also undergoing clinical trials for assessment of cost-effectiveness and safety (Compston 2012). Other treatment options for postmenopausal osteoporosis have also been evaluated in Cochrane reviews, e.g. strontium ranelate (O'Donnell 2004), etidronate (Wells 2008a), risedronate (Wells 2008b) and alendronate (Wells 2008c).

Description of the intervention

Human parathyroid hormone (hPTH) is currently available in two forms for osteoporosis treatment: as a full-length parathyroid hormone (1-84) and teriparatide (1-34), and for daily subcutaneous administration. This is the only approved anabolic treatment for osteoporosis. Recent non-Cochrane reviews of randomised controlled trials have shown that teriparatide (hPTH 1-34) significantly improved bone mineral density at the spine and hip, and significantly reduced the risk of vertebral and non-vertebral fractures in postmenopausal women with prior fractures (Cranney 2006; Han 2012; Silverman 2012). Treatment for 24 months showed improvement of lumbar bone mineral density (Black 2012). Poor adherence is a major problem with this type of treatment in the first year, and an even greater problem in the second, as reported in almost all trials (Baron 2012).

How the intervention might work

Parathyroid hormone acts to stimulate bone turnover, promoting bone formation to a greater extent than bone resorption (e.g. Jilka 2007; Silva 2011). The outcome is an increase of bone mass, in particular of the trabecular thickness and connectivity, but also bone structure and strength (e.g. Black 2012; Ferrari 2012). The anabolic action of intermittent parathyroid hormone administration has better results on the spine than the hip, which has less trabecular bone than the vertebrae; this is due to direct effects on osteoblasts and indirect effects through skeletal growth factors (Jilka 2007; Silva 2011).

Why it is important to do this review

The proposed systematic review will provide the best evidence on the benefit and harm of anabolic therapy for osteoporosis. This is particularly important because osteoporosis is a chronic disease that affects large segments of the population and presents a significant economic burden to society. For example, acute care costs for osteoporosis in Canada have been estimated at CAD 1.2 billion (Tarride 2012). In Germany, an economic simulation model to predict the burden of incident osteoporosis-attributable fractures showed that this burden would sharply increase until 2050, creating heavy demand for hospital and long-term care in the future (Bleibler 2013). Moreover, there is uncertainty regarding the benefit and harm of hPTH as a recent study has demonstrated adverse effects, including increased plasma and urine calcium, arthralgia and headache (Black 2012).

Objectives

To determine the benefit and harm of hPTH in the treatment of postmenopausal osteoporosis.

Methods

Criteria for considering studies for this review

Types of studies

Published randomised controlled trials (RCT) (full reports in a peer-reviewed journal and conference abstracts). Studies published in any language will be considered.

Types of participants

Trials including postmenopausal women (surgical or natural menopause with at least 12 months of continuous amenorrhoea) with diagnosed osteoporosis based on dual x-ray absorptiometry results (lumbar spine, hip or both). Trials including male osteoporosis and corticosteroid osteoporosis will be excluded.

Types of interventions

Trials of hPTH (1-34, 1-84, 1-38) versus placebo or other therapy regimens for osteoporosis (e.g. hormone replacement therapy, bisphosphonates, calcium/vitamin D) will be included. Dosages of hPTH to be included are as follows: 20 and 40 µg/daily for hPTH 1-34 and/or 100 µg/daily by subcutaneous injection for hPTH 1-84. All trial lengths will be included, categorised into short-, moderate- and long-term outcomes (6, 12 and ≥ 24 months, respectively).

Types of outcome measures

Major outcomes
  1. Incident fractures – vertebral

  2. Incident fractures – hip

  3. Incident fractures – wrist

  4. Quality of life

  5. Clinical gastrointestinal symptoms

  6. Withdrawals due to adverse events

  7. Serious adverse events (death, hospitalisation, hypercalcaemia)

Minor outcomes
  1. Incident fractures (other fractures, non-vertebral)

  2. Improvement in skeletal pain sensation (e.g. back pain)

Search methods for identification of studies

Electronic searches

We will search the following electronic databases, unrestricted by date or language:

  1. Cochrane Central Register of Controlled Trials (CENTRAL) (via The Cochrane Library, current issue);

  2. MEDLINE (via OVID 1946 to present);

  3. EMBASE (via OVID 1947 to present);

  4. ISI Web of Science via Thomson Web of Knowledge 1996 to present (for conference proceedings).

The MEDLINE search strategy (see Appendix 1) will combine the subject search with the Cochrane highly sensitive search strategy for identifying reports of randomised controlled trials (as published in Box 6.4c in the Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0, updated March 2011) and will be modified for the other databases (Higgins 2011).

Searching other resources

Handsearch of literature

We will handsearch major bone topic peer-reviewed journals from 1996 to present: Osteoporosis International, Journal of Bone and Mineral Research, Calcified Tissue International and Journal of Clinical Endocrinology and Metabolism.

We will select trials for closer examination if they appear to meet the criteria of being a randomised controlled trial and including a postmenopausal population as participants.

Trial registries

We will search for ongoing studies in the following trial registries.

  • ClinicalTrials.Gov (www.clinicaltrials.gov)

  • International Clinical Trials Registry Platform (http://apps.who.int/trialsearch/), which includes the following registries:

    • Australian New Zealand Clinical Trials Registry

    • ClinicalTrials.gov

    • EU Clinical Trials Register (EU-CTR)

    • ISRCTN

    • Brazilian Clinical Trials Registry (ReBec)

    • Chinese Clinical Trial Registry

    • Clinical Trials Registry – India

    • Clinical Research Information Service - Republic of Korea

    • Cuban Public Registry of Clinical Trials

    • German Clinical Trials Register

    • Iranian Registry of Clinical Trials

    • Japan Primary Registries Network

    • Pan African Clinical Trial Registry

    • Sri Lanka Clinical Trials Registry

    • The Netherlands National Trial Register

Unpublished works

For unpublished but completed studies, we will contact the responsible researcher indicated in the registry. 

Data collection and analysis

Selection of studies

Two members of the review team (VK, AM) will carry out the selection of papers and make decisions about eligibility independently. We will obtain a full copy of all possibly or definitely relevant studies for further assessment. The review authors will independently determine study eligibility and discrepancies will be resolved by discussion and consensus. We will consult a statistician (AJ) in cases of doubt about inclusion or data extraction, as well as with regard to data analysis. We will contact authors for clarification and to obtain additional data in order to perform a systematic review whenever necessary. Studies will be translated into English when necessary.

Data extraction and management

Two members of the review team (VK, AM) will independently extract relevant data from selected studies according to the inclusion criteria. Possible doubts and disagreements will be discussed and, if they cannot be resolved, we will contact the authors of the original articles for clarification.

We will use standard tables for the characteristics of included studies and risk of bias to minimise errors. We will consider abstracts together with the full publication. The review authors will not be blinded to the authors, interventions or results obtained in the included studies. For each study, we will extract the patient and study characteristics, intervention and outcomes data. We will extract the raw data (means and standard deviations for continuous outcomes and number of events and participants for dichotomous outcomes) for the outcomes of interest. We will pilot test predefined data extraction forms and use these, accompanied by a codebook, to collect data. Patient data will also include ethnicity. Disagreements will be resolved by discussion. We will extract both the generic and trade name of the experimental intervention, the type of control used, dosage and frequency, duration of treatment, patient characteristics (average age, postmenopause duration), types of measures used for the outcomes, trial design, trial size, duration of follow-up, type and source of financial support and publication status.

A priori decision rules for the selection of which data to extract in the event of multiple outcome reporting, are the following:

  • if both final values and change from baseline values are reported for the same outcome, we will extract both; present the change scores as one subgroup, and the final values as another subgroup, and then combine the two in an overall analysis;

  • if both unadjusted and adjusted values for the same outcome are reported, we will extract both and include adjusted values in the meta-analysis;

  • if data are analysed based on an intention-to-treat (ITT) sample and another sample (e.g. per-protocol, as treated), we will extract both, regardless of whether the outcomes assess benefits or harms; and use intention-to-treat data in analyses.

  • if multiple time points are used, we will extract data for 6, 12, 24 and longer periods.

When necessary, we will approximate means and measures of dispersion from the figures in the reports. Whenever possible, we will use results from an intention-to-treat analysis.

Assessment of risk of bias in included studies

Two review authors (VK, AJ) will independently assess risk of bias for all included studies using The Cochrane Collaboration's tool for assessing risk of bias tool (Higgins 2011). The tool addresses seven domains: 1) random sequence generation, 2) allocation concealment (selection bias), 3) blinding of participants and researchers (performance bias), 4) blinding of outcome assessment (detection bias), 5) incomplete outcome data (attrition bias), 6) selective reporting (reporting bias) and 7) other sources of bias (Higgins 2011). In the case of lack of important study information, we will contact authors to obtain the information needed, using open-ended questions. To determine the risk of bias of a study, for each criterion we will evaluate the presence of sufficient information and the likelihood of potential bias. We will rate each criterion as low risk of bias, high risk of bias or unclear risk of bias (either lack of information or uncertainty over the potential for bias). In a consensus meeting, disagreements among the review authors will be discussed and resolved. If consensus cannot be reached, a third review author will make the final decision.

Measures of treatment effect

We will perform meta-analysis for benefit and harm outcomes if the data from the studies are clinically and methodologically sufficiently homogeneous. The starting point of all meta-analyses of studies of effectiveness will involve identification of the data type for the outcome measurements.

For dichotomous data, we will express the results of each RCT as risk ratio (RR) with corresponding 95% confidence interval (95% CI).

For continuous data, we will analyse results as mean difference (MD) between the intervention and comparator group with 95% CIs. We will use standardised mean difference (SMD) for outcomes that measure the same construct using different tools.

Unit of analysis issues

For cross-over trials, we will extract data from the first period only. For studies containing more than two intervention groups, to make multiple pair-wise comparisons between all possible pairs of intervention groups, we will adjust the comparison groups so that patients are only counted once in the meta-analysis following the procedure recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Dealing with missing data

For missing summary data, we will impute missing standard deviations from other statistics such as standard errors, confidence intervals or P values, according to the methods recommended in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). Whenever possible, we will contact the original investigators to request missing data. We will explain the possible impact of missing data in the Discussion section of the systematic review.

Assessment of heterogeneity

Prior to meta-analysis, we will first assess studies for clinical homogeneity with respect to type of therapy, control group and outcomes.

We will not combine clinically heterogeneous studies in the analysis but describe them separately. We will assess heterogeneity by visual inspection of forest plots. For studies judged as clinically homogenous, we will test statistical heterogeneity using the Chi2 (Q) test and the I2 statistic (Higgins 2011). A P value of less than 0.10 or an I2 value greater than 50% indicates substantial heterogeneity.

In cases of substantial heterogeneity (ie, I2 > 50% ), we will explore the data further, including subgroup analyses, to try to explain it.

Assessment of reporting biases

To determine whether reporting bias is present, we will look for the trial protocol published by the authors before starting recruitment of the study participants. We will compare trial protocols with subsequent publication(s) to identify any discrepancies in the outcomes reported.

All trials that began enrolment of participants after September 2005 are supposed to be registered in a public trials registry at or before the onset of enrolment in order to be considered for publication in the major medical journals belonging to the International Committee of Medical Journal Editors (ICMJE). Therefore, for studies published after 1 July 2005, we will check the Clinical Trial Register at the International Clinical Trials Registry Platform of the World Health Organization. If we do not find protocols, we will compare the outcomes listed in the methods section of a publication with the reported results. For each study, we will evaluate whether selective reporting of outcomes is present. In the case of suspected reporting bias, we will contact study authors for clarification.

We will make a funnel plot if there are at least 10 studies included in the review to assess the possibility of publication bias. A test for funnel plot asymmetry (small study effects) formally examines whether the association between estimated intervention effects and a measure of study size is greater than might be expected to occur by chance. We will interpret the results from tests for funnel plot asymmetry cautiously. When there is evidence of small study effects, we will consider publication bias as only one of a number of possible explanations. In these circumstances, we will attempt to understand the source of small study effects and consider their implication in sensitivity analyses.

Data synthesis

The aim is to pool results from individual studies statistically for each intervention, to estimate an overall effect. Where studies are sufficiently homogenous that it is clinically meaningful for them to be pooled, we will use meta-analysis with the random-effects model. In the case of significant results we will also calculate the number needed to treat to benefit (NNTB) or harm (NNTH). We will perform analysis using Review Manager 5 (RevMan 2012) and produce forest plots for all analyses.

If the meta-analyses result in statistically significant overall estimates, we will transform these results (pooled estimate of RR, MD or SMD) back into measures which are clinically useful in daily practice, such as the number needed to treat to benefit (NNTB) or harm (NNTH) and the absolute or relative improvement in the original units, when we express the final results of the review. We will translate the results back by multiplying the SMD by the standard deviation from a representative study (Akl 2011).

Subgroup analysis and investigation of heterogeneity

Where sufficient data are available, the following subgroup analyses are planned.

  1. Doses of hPTH (20 and 40 μg/daily for hPTH 1-34 and/or 100 μg/daily for hPTH 1-84)

  2. Patients' age (45 to 65 years versus 66 to 80 years)

  3. Duration of osteoporosis (five years and under versus more than five years)

  4. Previous medication for osteoporosis (e.g. bisphosphonates, yes or no)

All major outcomew will be used for these analyses.

The differences between subgroups will be assessed using the formal test in RevMan.

Sensitivity analysis

We will examine the 'Risk of bias' results for each study. If there are sufficient studies with low overall risk of bias, i.e. estimate low risk of bias on all domains of theThe Cochrane Collaboration's tool for assessing risk of bias tool (Higgins 2011), we will perform meta-analyses on these studies first. If insufficient studies exist with low overall risk of bias, we will determine the number of studies where there is low risk of bias for the following domains: selective reporting, incomplete outcome assessment, unclear allocation concealment, those with subjective outcomes (e.g. pain) or those with unclear or inadequate blinding of outcome assessor. If there are a sufficient number of these studies we will meta-analyse these. We will then perform sensitivity analysis to assess how the results of meta-analyses for all major outcomes might be affected if studies with unclear or high risk of bias in these domains are included.

For studies with missing data, we will perform sensitivity analyses to assess how sensitive the results are to the assumptions made by imputing missing data. If the analysis of heterogeneity finds one or more outlying studies with results that conflict with the rest of the studies, we will perform sensitivity analysis to assess the influence on the results of the meta-analysis.

'Summary of findings' table

We will present the main results of the study in a 'Summary of findings' table, which we will produce using GRADEpro software. This table will provide key information concerning the quality of the evidence, the magnitude of effect of the intervention examined and the sum of available data on the main outcomes. The table will include an overall grading of the evidence related to each of the main outcomes using the GRADE approach, as indicated in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

The important outcomes that could be included in the 'Summary of findings' tables are:

  1. Incident fractures – vertebral

  2. Incident fractures – hip

  3. Incident fractures – wrist

  4. Quality of life

  5. Clinical gastrointestinal symptoms

  6. Withdrawals due to adverse events

  7. Serious adverse events (death, hospitalisation, hypercalcaemia)

Minor outcomes will be included in additional tables:

  1. Incident fractures - other fractures, non-vertebral

  2. Improvement in skeletal pain sensation (e.g. back pain)

The tables will present data separately for the moderate-term outcomes (12 months; main table) and also for the short- and long-term outcomes (6 and ≥ 24 months, respectively) for all interventions and comparators.

In the 'Comments' column of the 'Summary of findings' table, we will provide the absolute per cent difference, the relative per cent change from baseline and the number needed to treat to benefit (NNTB) or harm (NNTH) (the NNT will be provided only when the outcome shows a statistically significant difference).

For dichotomous outcomes, such as serious adverse events, we will calculate the NNT from the control group event rate and the risk ratio using the Visual Rx NNT calculator (Cates 2008). We will calculate the NNT for continuous measures using the Wells calculator (available at the Cochrane Musculoskeletal Group editorial office).

For dichotomous outcomes, we will calculate the absolute risk difference using the risk difference statistic in RevMan and express the result as a percentage. For continuous outcomes, we will calculate the absolute benefit as the improvement in the intervention group minus the improvement in the control group, in the original units.

We will calculate the relative per cent change for dichotomous data as the risk ratio - 1 and express this as a percentage. For continuous outcomes, we will calculate the relative difference in the change from baseline as the absolute benefit divided by the baseline mean of the control group.

Acknowledgements

The authors would like to acknowledge Jessie McGowan for her assistance with search strategies.

Appendices

Appendix 1. Full search strategy

Database: Ovid MEDLINE(R) In-Process & Other Non-Indexed Citations and Ovid MEDLINE(R) <1946 to May 2nd, 2013>

Search strategy:

--------------------------------------------------------------------------------

1 exp OSTEOPOROSIS/ (42121)

2 osteoporo$.tw. (47427)

3 bone loss$.tw. (17896)

4 osteopenia.mp. (6440)

5 bone mass$.mp. (13046)

6 bone mineral densit$.mp. (25011)

7 bone deminerali?ation.mp. (1028)

8 decalcifi$.mp. (4423)

9 deminerali?ed bone.mp. (1223)

10 or/1-9 (92603)

11 exp Postmenopause/ (17698)

12 post-menopause.mp. (349)

13 postmenopausal.tw. (38235)

14 (post menopaus$ or postmenopaus$ or post-menopaus$).tw. (43380)

15 or/11-14 (46268)

16 10 and 15 (13367)

17 exp Osteoporosis, Postmenopausal/ (10318)

18 16 or 17 (17785)

19 exp Parathyroid Hormone/ (24284)

20 parathyroid hormone$.tw. (23647)

21 Teriparatide.tw,rn. (1560)

22 (parathyrin or parathormone).tw. (2216)

23 (hpth or bpth).tw. (1108)

24 or/19-23 (34481)

25 18 and 24 (1487)

26 randomized controlled trial.pt. (347914)

27 controlled clinical trial.pt. (85825)

28 randomized.ab. (265789)

29 placebo.ab. (143593)

30 drug therapy.fs. (1604686)

31 randomly.ab. (193486)

32 trial.ab. (274399)

33 groups.ab. (1247796)

34 or/26-33 (3109513)

35 exp animals/ not humans.sh. (3806380)

36 34 not 35 (2659783)

37 25 and 36 (1003)

What's new

DateEventDescription
8 October 2013New citation required and major changesNew author team to conduct this review. Original protocol updated to reflect latest Cochrane standards.

History

Protocol first published: Issue 2, 2001

DateEventDescription
12 September 2008Amended

Converted to new review format.

CMSG ID: C130-P

Contributions of authors

  •  Conceiving, designing and co-ordinating the review: Vesna Kusec (VK), Ana Marusic (AM), Jonathan Adachi (JA), Gorge Wells (GW), Ana Jeroncic (AJ), Peter Tugwell (PT).

  •  Designing search strategies and undertaking searches: JA, AM, VK.

  •  Screening search results and retrieved papers against inclusion criteria: AM, VK.

  •  Appraising quality of papers: VK, AJ.

  •  Extracting data from papers: VK, AM.

  •  Writing to authors of papers for additional information: VK, AM.

  •  Data management for the review and entering data into RevMan: VK, AM.

  •  Analysis and interpretation of data: VK, AJ, GW, AM, JA, PT.

  • Providing research perspective: VK, AM, JA, PT.

  • Writing the review: VK, AM, JA.

  • Providing general advice on the review: GW, PT.

  • Performing previous work that was the foundation of the current review: JA, GW, PT.

Declarations of interest

The authors declare no conflict of interest.

Sources of support

Internal sources

  • New Source of support, Croatia.

    VK and AM are employed in the public sector (university and public hospital) and will be working on this review in part during their regular work.

External sources

  • No sources of support supplied

Ancillary