Does previous bisphosphonate therapy affect responsiveness to PTH?

Two posters presented at ECTS suggest there may be differences in rapidity of response to PTH based on previous bisphosphonate therapy, writes Dr Emma Duncan, an endocrinologist from Queensland. Bisphosphonates are the most common anti-resorptive medications prescribed for osteoporosis.  Bisphosphonates such as alendronate [ALN] and risedronate [RIS] decrease bone turnover (both bone resorption and bone formation) and modestly increase bone mineral density.  In contrast, new anabolic agents such as PTH increase bone turnover as well as markedly increasing bone mass, trabecular thickness and trabecular connectivity.  A major question, therefore, has been whether previous bisphosphonate therapy will affect responsiveness to PTH, and, if so, whether these effects will differ between the different bisphosphonates.  The two posters presented at ECTS suggest that there may be differences in rapidity of response to PTH based on previous bisphosphonate therapy. Bisphosphonates differ in their pharmaceutical properties, causing differences in potency and persistence, which allow clinicians flexibility with treatment options.  For example, retention on bone is longer with alendronate, resulting in greater persistence of effects of alendronate after cessation of therapy 1 .  Risedronate is more potent in inhibiting the osteoclastic enzyme farnesyl pyrophosphate synthase, which may explain observational studies suggesting earlier anti-fracture efficacy of risedronate 2 .  Both drugs, however, have good evidence of their antifracture effect, and the FACT trial has not demonstrated superiority of one agent over another with respect to fracture over time [though notably was not powered for this outcome] 3 . Previous studies have demonstrated that the anabolic response to PTH - as measured by BMD and bone turnover marker changes — is blunted in patients previously treated with alendronate and risedronate when compared with patients previously treated with other anti-resorptives such HRT and SERMS 4,5 . The posters presented at ECTS presented an open-label observational study examining the response to PTH1-34 after at least two years of therapy with ALN or RIS.  Overall, 669 postmenopausal women were screened, enrolling 331 patients, of whom 324 were treated with Teriparatide (TPTD) at a standard dose of 20 mcg/day s/c.  Previous adherence to bisphosphonate therapy was assumed from suppressed urinary NTX (<50nmol/mmol creat).   The ‘completer’ population of 146 in each group was presented, not the ITT data, although a comment was made that the results were similar whether analysed at intention to treat or per protocol.   Screening failure and discontinuation rates were apparently similar between the two groups of patients (data not shown). Higher rates of bone turnover were seen at baseline with patients on RIS compared with patients on ALN.  P1NP was 32 ng/mL in RIS group vs. 24 in ALN group (p<0.01), with similar significant differences also seen for osteocalcin, bone alkaline phosphatase, NTX and CTX (all results significant at p<0.02).  Similar results have been found in randomised controlled comparisons between RIS and ALN 3 .  The REAL study patients on RIS were shown to have a higher rate of corticosteroid usage and a higher rate of rheumatoid arthritis; such disease states are commonly associated with higher bone turnover.  No comment was made in the poster regarding these issues, although BMD, weight, previous fracture, previous form (weekly vs.daily) and duration of bisphosphonate therapy was not significantly different between groups. The primary end point was change in P1NP (a marker of bone formation) at 3 months, with secondary end points of change in markers of bone turnover at 0.5,1,2,3,4,5,6 and 12 months; and BMD changes at 6 and 12 months measured using DXA and QCT.  The data was presented as change compared with baseline, rather than absolute values achieved.  Change from baseline values of P1NP was significantly greater in patients previously treated with RIS than with ALN.  This was evident as early as one month; however, after 6 months there was no significant difference between the two groups with differences continuing to diminish by 12 months.  The absolute value of P1NP was significantly greater in the RIS group compared with ALN at 3 months (221 vs 182 ng/mL) with the time to achieve maximal bone formation achieved earlier (day 215 vs. day 271). The data was stratified according to baseline bone turnover (categorised into 20ng/mL bands).  Significantly greater increases were seen in P1NP at 3 months (the primary endpoint) in the RIS group compared with ALN regardless of baseline bone turnover.  A similar picture was evident with CTX, a marker of bone resorption, with significantly greater increases seen in the RIS group compared with the ALN group as early as 2 weeks.  The data concords with the recently reported EUROFORS trial, showing significantly earlier increases in bone turnover in patients previously treated with RIS compared with ALN by one month, continuing to be significantly greater by 6 months.  The EUROFORS study did not measure bone turnover beyond 6 months, however 4,5 . Areal BMD significantly increased at LS at time 6 and 12 months (by 12 months 5.1% increase in the RIS group; 3.4% increase in the ALN group; both results significantly greater than baseline and RIS significantly greater than ALN p<0.05). At total hip, both groups showed a significant loss of BMD measured by DXA at 6 months (-1.2% with RIS, -1.9% with ALN), with gradual increase in BMD towards baseline by 12 months.  In the RIS group, the results at total hip were not significantly different to baseline at 12 months (-0.3%) whilst patients pre treated with ALN remained significantly lower (-1.7%).  These data also concord with the EUROFORS study, which showed significant losses in the first year of treatment with TPTD at total hip and femoral neck in patients previously treated with bisphosphonates.  This was significant at femoral neck and total hip for ALN, and with total hip with RIS.  However, the EUROFOS study subsequently showed a marked increase in BMD at all sites measured in the second year of therapy, such that by 24 months there were marked increases in total hip and femoral neck BMD compared with baseline for all treated groups, regardless of previous bisphosphonate therapy (this was seen slightly earlier with RIS, with significant increases seen at femoral neck at 18 months).  Volumetric BMD was measured in a subset of patients (112 RIS, 119 ALN), using QCT.  Significant increases were seen in trabecular bone at LS with both groups (approximately 24% increase with RIS, 14% change with ALN, p<0.05).  At total hip, trabecular content increased significantly in both groups, with no significant difference evident between groups (approximately 13% with RIS, 5% with ALN). No fracture difference was seen between the two groups over the time course of the study and incidence of side effects was similar between the two group (hypercalcaemia the most common serious adverse event, at 14.6% with RIS and 13.9% with ALN). Author’s Comment These results are not particularly surprising, given the established differences between the pharmacology of the bisphosphonates; and the results are consistent with similar studies.  However, all these studies have been non-randomised in terms of previous bisphosphonates and there may be baseline differences between the patient groups that are not explored.  It is also not clear that the earlier anabolic response will necessarily result in any clinical difference in fracture rates with time.  Of note, previous studies have shown sustained improvement in fracture rates after cessation of 18 months of TPTD 6 .  In Australia, TPTD is licensed for 18 months’ therapy.  TPTD is not available on the Pharmaceutical Benefits Scheme, and patients are therefore liable for the full cost of the drug, in excess of $10000 for a full course.  Given the expense of PTH, it will be important to maximise the possible response to this anabolic agent.  Choosing a bisphosphonate that does not impair the response to this expensive medication, allowing maximal response for the 18 months of therapy, is thus important.  In the author’s opinion, this is particularly an issue when considering therapy in younger patients with osteoporosis, for whom anabolic therapies may become a realistic and affordable option with time.  These posters suggest that there is a significantly earlier increase in bone turnover in patients previously treated with RIS compared with ALN, and an earlier increase in BMD at both lumbar spine and at femoral neck.  However, this does not necessarily equate to a difference in fracture risk, and both these posters and other publications suggest that the differences abate with time. Written by Dr Emma Duncan, Consultant Endocrinologist, Princess Alexandra Hospital and Senior Lecturer, University of Queensland. The opinions expressed in this article are those of the individual author and not necessarily those of sanofi-aventis.    References Boonen, S. Lindsay, R. et al. ‘Responsiveness of bone turnover markers to teriparatide is greater  and easier following treatment with risedronate compared with alendronate: The OPTAMISE study’ , Poster  SU-P406, ECTS conference 2008....