- Email: [email protected]
Fluvastatin does not prevent the acute-phase response to intravenous zoledronic acid in post-menopausal women
Bone 49 (2011) 140–145
Contents lists available at ScienceDirect
Bone j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b o n e
Fluvastatin does not prevent the acute-phase response to intravenous zoledronic acid in post-menopausal women Keith Thompson a,⁎, Fran Keech a, David J. McLernon b, Kumar Vinod c, Robin J. May d, William G. Simpson e, Michael J. Rogers a, David M. Reid a a
Division of Applied Medicine, University of Aberdeen, UK Division of Applied Health Sciences, University of Aberdeen, UK Rheumatology Department, NHS Grampian, Aberdeen, UK d Novartis Pharmaceuticals UK Ltd, Frimley, Camberley, Surrey, UK e Clinical Biochemistry, NHS Grampian, Aberdeen, UK b c
a r t i c l e
i n f o
Article history: Received 24 August 2010 Revised 26 October 2010 Accepted 26 October 2010 Available online 31 October 2010 Edited by: David Burr Keywords: Zoledronic acid Gamma,delta T cell Fluvastatin Acute-phase response Bisphosphonates
a b s t r a c t The acute-phase response (APR) to aminobisphosphonates is triggered by activation of γδ T cells, resulting in pro-inflammatory cytokine release. Statins prevent aminobisphosphonate-induced γδ T cell activation in vitro, raising the possibility that statins might prevent the APR in vivo. The objective of this study was to determine whether fluvastatin prevents the APR to zoledronic acid in post-menopausal women. A doubleblind, randomised, placebo-controlled study was conducted in 60 healthy, post-menopausal, female volunteers (mean age 60.6 ± 4.0). Volunteers received 5 mg zoledronic acid by intravenous infusion, and either three times 40 mg fluvastatin (0 hr, 24 hr and 48 hr), 40 mg fluvastatin (0 hr) plus placebo (24 hr and 48 hr), or placebo (0 hr, 24 hr and 48 hr), orally. Post-infusion symptoms were assessed by questionnaire. Changes in γδ T cell levels, pro-inflammatory cytokines (TNFα, IFNγ, IL-6) and C-reactive protein (CRP) were measured in peripheral blood at various time-points post-infusion. Zoledronic acid administration triggered increased serum levels of TNFα, IFNγ, IL-6 and CRP in ≥ 70% of study volunteers, whilst characteristic APR symptoms were observed in N 50% of participants. Zoledronic acid also induced a transient fall in circulating Vγ9Vδ2 T cell levels at 48 hr, consistent with Vγ9Vδ2 T cell activation. Concurrent fluvastatin administration did not prevent zoledronic acid-induced cytokine release, alter circulating Vγ9Vδ2 T cell levels, nor diminish the frequency or severity of APR symptoms. In conclusion, intravenous zoledronic acid induced proinflammatory cytokine release and APR symptoms in the majority of study participants, which was not prevented by co-administration of fluvastatin. This article is part of a Special Issue entitled Bisphosphonates. © 2010 Elsevier Inc. All rights reserved.
Introduction Bisphosphonates (BPs) are currently the most common treatment for a variety of disorders characterised by excessive osteoclastic bone resorption, such as post-menopausal osteoporosis [1], Paget's disease of bone [2] and tumour-associated osteolysis [3]. A common sideeffect of intravenous administration of nitrogen-containing BP (N-BP), such as pamidronate or zoledronate (ZOL), is the development of a transient flu-like syndrome called the acute-phase response (APR). An APR typically occurs in 10–50% of patients receiving their first infusion [4–6] and is characterised by symptoms such as pyrexia
⁎ Corresponding author. Bone and Musculoskeletal Research Programme, Division of Applied Medicine (Musculoskeletal & Genetics Section), Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD, Scotland, UK. Fax: +44 1224 559533. E-mail address: [email protected] (K. Thompson). 8756-3282/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2010.10.177
and musculoskeletal aches and pains. These symptoms are frequently associated with increased circulating levels of pro-inflammatory cytokines such as IFNγ, TNFα and IL-6 [5,7–9]. Whilst the exact molecular mechanism underlying this phenomenon is not fully understood, it is becoming clear that activation of a specific subset of T cells, so-called γδ T cells, plays a central role. γδ T cells were first identified as possible initiators of the APR to N-BPs by Kunzmann et al., who observed marked increases in the number of circulating γδ T cells in PAM-treated multiple myeloma patients up to 28 days post-infusion, which correlated with the severity of the APR [10]. Subsequent studies revealed that N-BPs specifically activate the major subset of γδ T cells in peripheral blood, Vγ9Vδ2 T cells [11]. N-BP-induced activation of Vγ9Vδ2 T cells in peripheral blood mononuclear cell cultures in vitro results in production of IFNγ, TNFα and IL-6 [5,12,13], which closely mirrors cytokine production triggered by N-BP administration in vivo. This suggests that strategies to minimise or prevent Vγ9Vδ2 T cell activation by N-BPs in vivo may prevent the APR.
K. Thompson et al. / Bone 49 (2011) 140–145
We and others have previously shown that activation of Vγ9Vδ2 T cells by N-BPs occurs via an indirect mechanism [12,14], requiring intracellular uptake of N-BP to inhibit its molecular target, farnesyl diphosphate (FPP) synthase [15–18]. Recently, we proposed that peripheral blood monocytes play a crucial role in Vγ9Vδ2 T cell activation due to selective internalisation of the N-BP by highly endocytic CD14+ monocytes [19]. We demonstrated that, following a clinically relevant pulse of ZOL, inhibition of FPP synthase in monocytes causes the accumulation of the enzyme substrates, isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), both of which are agonists of the Vγ9Vδ2-T cell receptor (TCR) [20]. Consistent with a central role for IPP and DMAPP accumulation in triggering Vγ9Vδ2 T cell activation, we and others have previously shown that the stimulatory effects of N-BPs on Vγ9Vδ2 T cell activation and proliferation in peripheral blood mononuclear cell cultures can be prevented by simultaneous treatment with a statin [12–14]. These widely-used cholesterol-lowering drugs inhibit HMGCoA reductase, an enzyme upstream of FPP synthase, and thus prevent N-BP-induced IPP/DMAPP accumulation. We therefore investigated in this study whether co-administration of a statin could prevent ZOLinduced cytokine release and flu-like symptoms of the APR in healthy post-menopausal women.
141
Siemens reagents. Serum levels of type I collagen C-telopeptide (CTx) were measured using a β-Crosslaps/serum kit and an Elecsys 2010 Immunoassay Analyser (Roche). Proportion of Vγ9Vδ2 T cells in the circulation Peripheral blood samples were collected at baseline (0 hr), 48 hr, 4 weeks and 8 weeks, using EDTA as anti-coagulant. Peripheral blood mononuclear cells (PBMCs) were isolated by density-gradient separation using Lymphoprep reagent (Axis-Shield). The number of Vγ9Vδ2 T cells in peripheral blood, relative to the total T cell population, was then determined by immunostaining using anti-human Vδ2-FITC (Beckman Coulter) and anti-CD3-PerCP (BD Biosciences) antibodies, followed by flow cytometric analysis. Data acquisition and analysis was performed on a BD FACSCalibur flow cytometer using CellQuestPro software. Assessment of acute-phase reaction symptoms Symptoms characteristic of the APR were assessed in 6 categories (joint aches; muscle aches; bone aches and pains; nausea; headache; flu-like symptoms) by questionnaire using a 4-point scale (1 = no symptoms; 2 = mild; 3 = moderate; 4 = severe). Questionnaires were completed at 8 hr, 24 hr, 48 hr and 72 hr post-infusion of ZOL.
Materials and methods
Statistical analysis
Study population and design
The study was designed with the primary end-point being the effectiveness of FLU to inhibit the APR based on follow-up measurements of CRP. In assessing the power of the study it was assumed that at least 50% of subjects would get a N2-fold rise in their peak CRP values (CRP rising from a mean of 10 ± 4 mg/dl at baseline to 30 ± 12 mg/dl) and that pre-treatment with FLU would prevent this elevation. Based on these two assumptions, and a drop-out rate of 10% during follow-up, we aimed to recruit 60 women (20 subjects per group), in order to give 90% power with an alpha of less than 0.05. All statistical analyses were performed using SPSS software and SAS v9.1 (SAS Institute). A fixed-effects repeated measures analysis was conducted to test for statistical differences between the three treatment groups. Where a significant interaction was found between treatment group and time, differences between the treatment groups at each time point were investigated separately. To adjust for multiple comparisons, p b 0.01 was taken to be statistically significant. Measurements that were not normally distributed (CRP, TNFα, IFNγ, IL-6, Vγ9Vδ2 T cells and cholesterol) were log-transformed and geometric means (95% CI) were calculated from the model. Two patients were excluded from the analysis due to abnormal baseline values of CRP and TNFα, respectively.
All research was conducted at Aberdeen Royal Infirmary, Aberdeen, UK. Ethical approval for this study was granted by the Grampian Local Research Ethics Committee. A total of 61 healthy women aged 50–70, who were N12 months post-menopause and had a T-score ≥2.5 (i.e. non-osteoporotic), were recruited into the double-blind, placebocontrolled study. Participants were BP-naïve and had not received hormone replacement therapy within 6 months of inclusion. Subjects receiving statin therapy were not excluded, but their therapy was suspended for 7 days prior to the ZOL infusion, and recommenced 7 days post-infusion. One participant withdrew from the study postrandomisation but before treatment commenced, resulting in 60 study participants. Subjects were randomly assigned to 3 treatment groups (n = 20). All study participants received a single 5 mg zoledronic acid intravenous infusion over 15 min. In addition, one group (designated 3xFLU) received three times 40 mg oral fluvastatin (FLU) (at 0 hr, 24 hr and 48 hr); a further group (designated FLU+ PLAC) received 40 mg oral FLU (at 0 hr) plus oral placebo (at 24 hr and 48 hr); the final group (designated PLAC) received oral placebo (at 0 hr, 24 hr and 48 hr). The first dose of FLU or placebo was administered 30 min prior to infusion of ZOL, to allow peak plasma concentrations of FLU during the ZOL administration, in line with the pharmacokinetic properties of FLU [21]. Study participants were allowed to self-administer acetaminophen to treat symptoms such as headache and pyrexia as required, and were encouraged to increase their fluid intake for 24 hr following the infusion. Measurement of pro-inflammatory cytokines, C-reactive protein, cholesterol, and type I collagen C-telopeptide Serum samples were obtained at baseline (0 hr), and at 4 hr, 8 hr, 24 hr, 48 hr, 4 weeks, 8 weeks and 12 weeks after infusion of ZOL, and stored at −80 °C until analysis. Levels of pro-inflammatory cytokines were determined using Quantikine ELISA kits (IFNγ & IL-6) or a Quantikine Highly Sensitive ELISA kit (TNFα) (all R&D Systems). Serum Hi-Sensitivity C-reactive protein (CRP) was determined using a standard nephelometric immunoassay on a Behring nephelometer with Behring reagents. Total serum cholesterol was determined using a standard colorimetric method on a Siemens auto-analyser with
Results and discussion Patient information and determination of FLU treatment protocol Sixty female volunteers completed the study. The mean age was 60.6 ± 4.0 years. All volunteers were post-menopausal and presented with a T-score of ≥2.5. The subject disposition is shown in Fig. 1. The proposed timing and dosage of FLU for the study was determined based on our findings in an in vitro model of the APR to N-BPs (Suppl. Fig. 1) using IFNγ release as a measure of N-BP-induced γδ T cell activation in PBMC cultures [12]. This showed that treatment of PBMCs with 1 μΜ FLU for 2 hr (which mimics the Cmax following a typical 40 mg dose [21]) was effective in preventing γδ T cell activation induced by a pulse of 1 μM ZOL for 2 hr (which mimics the Cmax following a typical iv dose of ZOL [22]). FLU pre-treatment period (3 hr, 1 hr, 0 hr) did not influence the effectiveness of FLU in our in vitro assay system (Suppl. Fig. 1), indicating that FLU does not need to be present long before ZOL to prevent γδ T cell activation. For
142
K. Thompson et al. / Bone 49 (2011) 140–145
Fig. 1. Study flow diagram.
this reason, we administered FLU at a dose (40 mg) and a time (30 min prior to the ZOL infusion) that would be anticipated to give a sufficiently high Cmax (~1 μM [21]) known to inhibit ZOL-induced Vγ9Vδ2 T cell activation in vitro, and that would be maintained for the duration of the peak ZOL concentration in peripheral blood to ensure inhibition of HMG-CoA reductase and prevention of ZOL-induced IPP accumulation that typically occurs within 1-3 hr following ZOL exposure in vitro [14]. Changes in cytokine levels following ZOL ZOL administration alone (PLAC group) induced a marked elevation in mean serum CRP, which first showed a statistically significant increase at 24 hr (p b 0.001), and was further increased at
48 hr (p b 0.001: Fig. 2). CRP values had returned to baseline levels at 4 weeks post-infusion. This increase in CRP was accompanied by increased levels of the pro-inflammatory cytokines IL-6, TNFα and IFNγ. Serum IL-6 levels increased rapidly, with a ~ 3-fold increase detected within the first 4 hr following ZOL infusion (p b 0.001), peaked at 24 hr (p b 0.001), and had returned towards baseline values at 48 hr. In contrast, serum TNFα or IFNγ were not elevated at 4 or 8 hr following ZOL infusion, but were significantly increased at 24 hr (p b 0.001), with TNFα levels remaining elevated at 48 hr (p b 0.001). All cytokines had returned to baseline levels at 4 weeks post-ZOL infusion. Previous findings from in vitro studies have shown that monocytes are directly targeted by ZOL whilst activation of Vγ9Vδ2 T cells occurs secondary to the internalisation of ZOL by monocytes and accumulation of IPP in these cells [19]. The increase in serum IL-6 in
Fig. 2. Effect of fluvastatin on serum levels of pro-inflammatory cytokines following ZOL administration in healthy post-menopausal women. Following treatment, serum levels of pro-inflammatory cytokines (A) TNFα, (B) IFNγ, and (C) IL-6 were determined by ELISA. (D) Serum Hi-Sensitivity C-reactive protein (CRP) was determined by nephelometric immunoassay. Data shown are the geometric mean (95% CI) of 20 patients/treatment group. (Symbols denote p b 0.001 vs baseline (0 hr) values: *PLAC group; #FLU + PLAC group; † 3xFLU group).
K. Thompson et al. / Bone 49 (2011) 140–145
143
response to ZOL infusion ahead of the increases in TNFα and IFNγ therefore suggests that monocytes are the predominant producers of IL-6 in vivo in response to ZOL, whereas activated Vγ9Vδ2 T cells are the major source of IFNγ and TNFα. In support of this, the N-BP alendronate has been previously shown to augment LPS-induced IL-6 release from macrophages, although alendronate alone had no stimulatory effect [23]. However, a previous in vitro study suggested that, in a subset of healthy donors, N-BP-treated Vγ9Vδ2 T cells themselves are the source of IL-6 [13]. Thus, which cell types secrete which cytokines in response to ZOL in vivo remains to be clarified.
Table 2 Total number of mild, moderate and severe symptoms reported in 6 symptom categories per treatment group (20 participants per group).
Relationship between changes in cytokine levels and development of APR symptoms following ZOL
although there was a trend towards decreased incidence of nausea and increased incidence of headaches in the FLU + PLAC and 3xFLU groups, compared to the ZOL alone (PLAC) group. The mean severity of APR symptoms was also not different between either of the FLU treatment groups and the PLAC (ZOL alone) group at either 24 or 48 hr (Fig. 3). At 72 hr post-infusion, mean symptom severity scores had decreased markedly in all treatment groups, consistent with the transient nature of the APR. The lack of effect of FLU for preventing ZOL-induced cytokine release or APR symptoms was not due to differences between treatment groups in self-administration of antipyretic or NSAID medication, since similar numbers of study participants took rescue medication in both PLAC and FLU + PLAC groups (both 12/20), whilst this proportion was increased in the 3xFLU group (16/20).
Analysis of cytokine increases in individual study participants revealed that at least 70% of study participants treated with ZOL alone (PLAC group) had an increase in pro-inflammatory cytokine and CRP levels of ≥2 S.D. over mean baseline values (Table 1). IFNγ and IL-6 were increased in 90% and CRP in 80% of study participants treated with ZOL alone. However, whilst the majority of participants experienced significant cytokine increases, this did not always result in reported APR symptoms, since only 55% (11 out of 20) of participants receiving ZOL alone experienced flu-like symptoms to various degrees within 72 hr of the ZOL infusion (Table 2). Correlating individual cytokine increases with APR symptoms revealed that no single cytokine had greater prognostic power (8 out of 11 showed increased TNFα; 9 out of 11 showed increased IFNγ, IL-6 and CRP; Table 3). Furthermore, most participants that reported no APR symptoms demonstrated elevated cytokine levels (6 out of 9 showed increased TNFα; 9 out of 9 showed increased IFNγ, IL-6 and CRP), indicating that marked increases in the release of pro-inflammatory cytokines do not always trigger the development of APR symptoms. This suggests a more complex relationship between increased cytokine levels and triggering of flu-like symptoms than previously suggested [9]. Recent reports have shown that regulatory T cells may limit responses of Vγ9Vδ2 T cells to phosphoantigen stimulation in vitro [24,25]. However, the majority of our study participants experienced ZOL-induced elevations in circulating cytokines, suggesting that Vγ9Vδ2 T cell responsiveness is not impaired, either via regulatory T cells or through other regulatory pathways, in these subjects. Further studies will be required to fully elucidate the relationship between N-BP-induced cytokine release and APR symptoms in vivo. Effect of FLU on changes in cytokine levels and development of APR symptoms following ZOL FLU administration, either single or triple-dosing, did not have any significant inhibitory effect on ZOL-induced increases in mean CRP, TNFα, IFNγ or IL-6 (Fig. 2), although there was a non-significant trend for FLU to decrease serum IL-6 levels at 24 hr (PLAC: 12.79 (7.77, 21.06) pg/ml; FLU + PLAC: 9.82 (5.77, 16.69) pg/ml; 3xFLU: 7.88 (4.84, 12.81) pg/ml). Furthermore, no difference was found in the number of study participants who experienced significant increases in serum levels of TNFα, IFNγ, IL-6 or CRP in either the FLU + PLAC or the 3xFLU treatment groups, as compared to the PLAC group (Table 1). Consistent with this, FLU administration (either single or triple dose) did not decrease the reported incidence of APR symptoms (Table 2),
Table 1 Number of study participants (out of 20) with elevated cytokine levels (N 2× standard deviations above baseline) within 48 hr of ZOL infusion. Treatment group
TNFα
IFNγ
IL-6
CRP
PLAC FLU + PLAC 3xFLU
14 15 10
18 17 16
18 16 17
16 15 16
Treatment Joint Muscle Bone aches Nausea Headache Flu-like Total subjects group aches aches and pains symptoms with symptoms PLAC FLU+ PLAC 3xFLU
10 9
10 10
8 7
9 2
10 12
11 9
11 13
7
13
11
4
16
11
13
Effects of ZOL and FLU on circulating Vγ9Vδ2 T cell levels ZOL treatment induced a non-significant, ~ 30% decrease in circulating levels of Vγ9Vδ2 T cells (relative to the total T cell population) in peripheral blood at 48 hr, which recovered to baseline levels by 4 weeks (Fig. 4). The decrease in circulating levels of Vγ9Vδ2 T cells is suggestive of Vγ9Vδ2 T cell activation and subsequent extravasation of activated T cells into peripheral lymphoid tissues. Transient lymphopenia following intravenous N-BP administration has previously been reported [4,7,26,27], which indicates that the pro-inflammatory cascade of cytokines triggered by γδ T cells promotes a more general extravasation of lymphocytes from the peripheral circulation. However, the studies by Kunzmann et al. that first implicated γδ T cells in the APR to N-BPs reported increased circulating levels of γδ T cells, which were maintained for up to 4 weeks post-N-BP infusion [11]. Similarly, ZOL (particularly in combination with IL-2 administration) has been reported to result in sustained increases in circulating Vγ9Vδ2 T cells in patients with hormone-refractory prostate cancer [28] and lymphoid malignancies [29]. This discrepancy is suggestive of (as yet poorly characterised) differences in Vγ9Vδ2 T cell reactivity between cancer patients and the healthy population in our study. No correlation was found between experienced symptoms (and/or cytokine increases) and circulating levels of Vγ9Vδ2 T cells, either at baseline, or at 48 hr when APR symptoms were most prevalent (data not shown). Co-administration of FLU, either a single dose (FLU+ PLAC group) or three consecutive doses (3xFLU group), did not prevent the ZOL-induced decrease in Vγ9Vδ2
Table 3 Frequency of cytokine increases in participants reporting acute-phase reaction symptoms per treatment group. Treatment group
TNFα
PLAC FLU + PLAC 3xFLU
8/11 10/13 6/13
a
IFNγ
a
9/11 12/13 7/13
IL-6a
CRP
a
9/11 10/13 12/13
9/11 12/13 11/13
a The first value represents the number of participants with acute-phase reaction symptoms that showed an increase in cytokine level (N 2 S.D. above baseline), followed by the total number of participants experiencing symptoms.
144
K. Thompson et al. / Bone 49 (2011) 140–145
for inhibiting bone resorption in our study population. FLU treatment, using either treatment regimen, had no significant effect on the ZOLinduced decrease in serum CTx levels at any of the 3 time-points investigated (48 hr, 4 or 12 weeks; Suppl. Fig. 3). Safety
Fig. 3. Effect of concurrent fluvastatin administration on severity of ZOL-induced acute-phase reaction symptoms. Following treatment, acute-phase reaction symptoms (joint aches; muscle aches; bone aches and pains; nausea; headache; flu-like symptoms) were assessed using a 4-point scale (1=no symptoms; 2=mild; 3=moderate; 4=severe) at 8 hr, 24 hr, 48 hr and 72 hr post-ZOL infusion. A symptom score of 6 indicates no symptoms, whereas a score of 24 indicates severe symptoms in all 6 categories. Data shown are the mean reported symptom scores of 20 patients/treatment group±S.E.M.
T cell levels at 48 hr, nor was there any effect of FLU treatment on the recovery of circulating Vγ9Vδ2 T cell levels at 4 weeks (Fig. 4). This is consistent with the lack of effect of FLU on the increased serum cytokine and CRP levels in response to ZOL. Effects of ZOL and FLU on serum cholesterol levels ZOL administration resulted in a transient decrease in total serum cholesterol of ~ 10% at 48 hr. Cholesterol levels recovered to baseline values by 4 weeks (Suppl. Fig. 2). Co-administration of either a single dose of FLU (FLU + PLAC group), or three consecutive doses of FLU (3xFLU group) did not significantly decrease total serum cholesterol levels greater than ZOL treatment alone at 48 hr, although there was a trend towards a greater decrease in serum cholesterol in the triple dose FLU group. Effects of ZOL and FLU on bone resorption ZOL treatment resulted in a profound and sustained decrease in bone resorption, as assessed by serum CTx (Suppl. Fig. 3). The inhibitory effect on bone resorption was rapid, with maximal inhibition (~95% reduction in serum CTx) observed after 48 hr, and was sustained (N80% reduction) at 12 weeks following ZOL infusion. There was no significant correlation between severity of APR symptoms and inhibition of CTx release (data not shown), suggesting that APR symptom severity was not indicative of effectiveness of ZOL
Apart from the APR symptoms reported, the therapies tested were generally well-tolerated. Six serious adverse events were reported including 2 hospitalisations due to non-study related problems. There were 3 cases of short-lasting uveitis/iritis (one bi-lateral): 1 in PLAC group, 2 in 3xFLU group. One patient who had uveitis also developed gum swelling during the study period, which settled with conservative saline mouthwash therapy. Conclusions We investigated whether FLU could prevent the most common adverse event associated with intravenous ZOL administration, the development of an APR, in a placebo-controlled, double-blind study in a healthy cohort of post-menopausal women. Our results show that, despite the inhibitory effect of FLU on N-BP-induced Vγ9Vδ2 T cell activation in vitro (Suppl. Fig. 1 & [13]), co-administration of FLU at the dosing regimens used did not prevent ZOL-induced Vγ9Vδ2 T cell activation, pro-inflammatory cytokine release, or APR symptoms in our study population. The lack of effect of FLU on incidence or severity of APR symptoms is consistent with a recent report detailing the lack of efficacy of atorvastatin for preventing ZOL-induced APR symptoms (increased serum CRP, musculoskeletal pain, and fever) in a small cohort of 12 children, although a non-significant trend was observed for decreased pain scores and diminished use of rescue medication (oxycodone and acetaminophen) in response to co-administration of atorvastatin [30]. Furthermore, whilst major advances in elucidating the mechanisms involved in Vγ9Vδ2 T cell activation by N-BPs have been made in recent years through in vitro studies [12–14], the findings of our study highlight that the exact mechanism underlying the APR to N-BPs such as ZOL remains to be clarified. In particular, the relationship between cytokine increases and the manifestation of symptoms awaits more thorough characterisation. Supplementary materials related to this article can be found online at doi:10.1016/j.bone.2010.10.177. Acknowledgments We gratefully acknowledge the technical assistance provided by Mrs Denise Tosh, Mrs Kelly Deans and Mrs Linda Duncan. We would also like to thank Dr. Anke Roelofs for critical comments on the manuscript. The study was funded by an unrestricted educational grant by Novartis. DMR and KT are grateful to Arthritis Research UK for continued infrastructure support. DMR has received additional research funding from Novartis, participates in Advisory Boards for the company and is a member of their speakers panel. MJR has received research funding from Novartis, Roche, The Alliance for Better Bone Health, and Merck. References
Fig. 4. Effect of fluvastatin on ZOL-induced changes in circulating levels of Vγ9Vδ2 T cells. The proportion of Vγ9Vδ2 T cells present in the CD3+ (T cell) population of peripheral blood mononuclear cells was determined by immunostaining with anti-human Vδ2-TCRFITC and anti-human CD3-PerCP antibodies and flow cytometric analysis. Data acquisition and analysis were performed using a Becton Dickinson FACSCalibur flow cytometer and CellQuest Pro software on the lymphocyte-gated population. Data shown are the geometric mean (95% CI) of 20 patients/treatment group.
[1] Kanis JA, Burlet N, Cooper C, Delmas PD, Reginster JY, Borgstrom F, et al. European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO). European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 2008;19:399–428. [2] Ralston SH, Langston AL, Reid IR. Pathogenesis and management of Paget's disease of bone. Lancet 2008;372:155–63. [3] Berenson JR, Rajdev L, Broder M. Treatment strategies for skeletal complications of cancer. Cancer Biol Ther 2006;5:1074–7. [4] Adami S, Bhalla AK, Dorizzi R, Montesanti F, Rosini S, Salvagno G, et al. The acutephase response after bisphosphonate administration. Calcif Tissue Int 1987;41: 326–31.
K. Thompson et al. / Bone 49 (2011) 140–145 [5] Thiebaud D, Sauty A, Burckhardt P, Leuenberger P, Sitzler L, Green JR, et al. An in vitro and in vivo study of cytokines in the acute-phase response associated with bisphosphonates. Calcif Tissue Int 1997;61:386–92. [6] Strampel W, Emkey R, Civitelli R. Safety considerations with bisphosphonates for the treatment of osteoporosis. Drug Saf 2007;30:755–63. [7] Sauty A, Pecherstorfer M, Zimmer-Roth I, Fioroni P, Juillerat L, Markert M, et al. Interleukin-6 and tumor necrosis factor alpha levels after bisphosphonates treatment in vitro and in patients with malignancy. Bone 1996;18: 133–9. [8] Schweitzer DH, Oostendorp-Van de Ruit M, Van der Pluijm G, Löwik CWGM, Papapoulos SE. Interleukin-6 and the acute phase response during treatment of patients with Paget's disease with the nitrogen-containing bisphosphonate dimethylaminohydroxypropylidene bisphosphonate. J Bone Miner Res 1995;10:956–62. [9] Dicuonzo G, Vincenzi B, Santini D, Avvisati G, Rocci L, Battistoni F, et al. Fever after zoledronic acid administration is due to increase in TNF-alpha and IL-6. J Interferon Cytokine Res 2003;23:649–54. [10] Kunzmann V, Bauer E, Wilhelm M. Gamma/delta T-cell stimulation by pamidronate. N Engl J Med 1999;340:737–8. [11] Kunzmann V, Bauer E, Feurle J, Weissinger F, Tony HP, Wilhelm M. Stimulation of gammadelta T cells by aminobisphosphonates and induction of antiplasma cell activity in multiple myeloma. Blood 2000;96:384–92. [12] Thompson K, Rogers MJ. Statins prevent bisphosphonate-induced gamma, delta-T-cell proliferation and activation in vitro. J Bone Miner Res 2004;19: 278–88. [13] Hewitt RE, Lissina A, Green AE, Slay ES, Price DA, Sewell AK. The bisphosphonate acute phase response: rapid and copious production of proinflammatory cytokines by peripheral blood γδ T cells in response to aminobisphosphonates is inhibited by statins. Clin Exp Immunol 2005;139:101–11. [14] Gober HJ, Kistowska M, Angman L, Jenö P, Mori L, De Libero G. Human T cell receptor gammadelta cells recognize endogenous mevalonate metabolites in tumor cells. J Exp Med 2003;197:163–8. [15] Van Beek E, Pieterman E, Cohen L, Lowik C, Papapoulos S. Farnesyl pyrophosphate synthase is the molecular target of nitrogen-containing bisphosphonates. Biochem Biophys Res Commun 1999;264:108–11. [16] Bergstrom JD, Bostedor RG, Masarachia PJ, Reszka AA, Rodan G. Alendronate is a specific, nanomolar inhibitor of farnesyl diphosphate synthase. Arch Biochem Biophys 2000;373:231–41. [17] Dunford JE, Thompson K, Coxon FP, Luckman SP, Hahn FM, Poulter CD, et al. Structure-activity relationships for inhibition of farnesyl diphosphate synthase in vitro and inhibition of bone resorption in vivo by nitrogen-containing bisphosphonates. J Pharmacol Exp Ther 2001;296:235–42.
145
[18] Kavanagh KL, Guo K, Dunford JE, Wu X, Knapp S, Ebetino FH, et al. The molecular mechanism of nitrogen-containing bisphosphonates as anti-osteoporosis drugs: crystal structure and inhibition of farnesyl pyrophosphate synthase. Proc Natl Acad Sci USA 2006;103:7829–34. [19] Roelofs AJ, Jauhiainen M, Mönkkönen H, Rogers MJ, Mönkkönen J, Thompson K. Peripheral blood monocytes are responsible for gammadelta T cell activation induced by zoledronic acid through accumulation of IPP/DMAPP. Br J Haematol 2009;144:245–50. [20] Tanaka Y, Morita CT, Tanaka Y, Nieves E, Brenner MB, Bloom BR. Natural and synthetic non-peptide antigens recognized by human gamma delta T cells. Nature 1995;375:155–8. [21] Corsini A, Bellosta S, Baetta R, Fumagalli R, Paoletti R, Bernini F. New insights into the pharmacodynamic and pharmacokinetic properties of statins. Pharmacol Ther 1999;84:413–28. [22] Chen T, Berenson J, Vescio R, Swift R, Gilchick A, Goodin S, et al. Pharmacokinetics and pharmacodynamics of zoledronic acid in cancer patients with bone metastases. J Clin Pharmacol 2002;42:1228–36. [23] Makkonen N, Salminen A, Rogers MJ, Frith JC, Urtti A, Azhayeva E, et al. Contrasting effects of alendronate and clodronate on RAW 264 macrophages: the role of a bisphosphonate metabolite. Eur J Pharm Sci 1999;8:109–18. [24] Mahan CS, Thomas JJ, Boom WH, Rojas RE. CD4+ CD25(high) Foxp3+ regulatory T cells downregulate human Vdelta2+ T-lymphocyte function triggered by antiCD3 or phosphoantigen. Immunology 2009;127:398–407. [25] Kunzmann V, Kimmel B, Herrmann T, Einsele H, Wilhelm M. Inhibition of phosphoantigen-mediated gammadelta T-cell proliferation by CD4+ CD25+ FoxP3+ regulatory T cells. Immunology 2009;126:256–67. [26] Harinck HIJ, Papapoulos SE, Blanksma HJ, Moolenaar AJ, Vermeij P, Bijvoet OLM. Paget's disease of bone: early and late responses to three different modes of treatment with aminohydroxypropylidene bisphosphonate (APD). Br Med J 1987;295:1301–5. [27] Pecherstorfer M, Jilch R, Sauty A, Horn E, Keck AV, Zimmer-Roth I, et al. Effect of first treatment with aminobisphosphonates pamidronate and ibandronate on circulating lymphocyte subpopulations. J Bone Miner Res 2000;15:147–54. [28] Dieli F, Vermijlen D, Fulfaro F, Caccamo N, Meraviglia S, Cicero G, et al. Targeting human gamma}delta T cells with zoledronate and interleukin-2 for immunotherapy of hormone-refractory prostate cancer. Cancer Res 2007;67:7450–7. [29] Wilhelm M, Kunzmann V, Eckstein S, Reimer P, Weissinger F, Ruediger T, et al. Gammadelta T cells for immune therapy of patients with lymphoid malignancies. Blood 2003;102:200–6. [30] Srivastava T, Haney CJ, Alon US. Atorvastatin may have no effect on acute phase reaction in children after intravenous bisphosphonate infusion. J Bone Miner Res 2009;24:334–7.
Contents lists available at ScienceDirect
Bone j o u r n a l h o m e p a g e : w w w. e l s e v i e r. c o m / l o c a t e / b o n e
Fluvastatin does not prevent the acute-phase response to intravenous zoledronic acid in post-menopausal women Keith Thompson a,⁎, Fran Keech a, David J. McLernon b, Kumar Vinod c, Robin J. May d, William G. Simpson e, Michael J. Rogers a, David M. Reid a a
Division of Applied Medicine, University of Aberdeen, UK Division of Applied Health Sciences, University of Aberdeen, UK Rheumatology Department, NHS Grampian, Aberdeen, UK d Novartis Pharmaceuticals UK Ltd, Frimley, Camberley, Surrey, UK e Clinical Biochemistry, NHS Grampian, Aberdeen, UK b c
a r t i c l e
i n f o
Article history: Received 24 August 2010 Revised 26 October 2010 Accepted 26 October 2010 Available online 31 October 2010 Edited by: David Burr Keywords: Zoledronic acid Gamma,delta T cell Fluvastatin Acute-phase response Bisphosphonates
a b s t r a c t The acute-phase response (APR) to aminobisphosphonates is triggered by activation of γδ T cells, resulting in pro-inflammatory cytokine release. Statins prevent aminobisphosphonate-induced γδ T cell activation in vitro, raising the possibility that statins might prevent the APR in vivo. The objective of this study was to determine whether fluvastatin prevents the APR to zoledronic acid in post-menopausal women. A doubleblind, randomised, placebo-controlled study was conducted in 60 healthy, post-menopausal, female volunteers (mean age 60.6 ± 4.0). Volunteers received 5 mg zoledronic acid by intravenous infusion, and either three times 40 mg fluvastatin (0 hr, 24 hr and 48 hr), 40 mg fluvastatin (0 hr) plus placebo (24 hr and 48 hr), or placebo (0 hr, 24 hr and 48 hr), orally. Post-infusion symptoms were assessed by questionnaire. Changes in γδ T cell levels, pro-inflammatory cytokines (TNFα, IFNγ, IL-6) and C-reactive protein (CRP) were measured in peripheral blood at various time-points post-infusion. Zoledronic acid administration triggered increased serum levels of TNFα, IFNγ, IL-6 and CRP in ≥ 70% of study volunteers, whilst characteristic APR symptoms were observed in N 50% of participants. Zoledronic acid also induced a transient fall in circulating Vγ9Vδ2 T cell levels at 48 hr, consistent with Vγ9Vδ2 T cell activation. Concurrent fluvastatin administration did not prevent zoledronic acid-induced cytokine release, alter circulating Vγ9Vδ2 T cell levels, nor diminish the frequency or severity of APR symptoms. In conclusion, intravenous zoledronic acid induced proinflammatory cytokine release and APR symptoms in the majority of study participants, which was not prevented by co-administration of fluvastatin. This article is part of a Special Issue entitled Bisphosphonates. © 2010 Elsevier Inc. All rights reserved.
Introduction Bisphosphonates (BPs) are currently the most common treatment for a variety of disorders characterised by excessive osteoclastic bone resorption, such as post-menopausal osteoporosis [1], Paget's disease of bone [2] and tumour-associated osteolysis [3]. A common sideeffect of intravenous administration of nitrogen-containing BP (N-BP), such as pamidronate or zoledronate (ZOL), is the development of a transient flu-like syndrome called the acute-phase response (APR). An APR typically occurs in 10–50% of patients receiving their first infusion [4–6] and is characterised by symptoms such as pyrexia
⁎ Corresponding author. Bone and Musculoskeletal Research Programme, Division of Applied Medicine (Musculoskeletal & Genetics Section), Institute of Medical Sciences, University of Aberdeen, Foresterhill, Aberdeen, AB25 2ZD, Scotland, UK. Fax: +44 1224 559533. E-mail address: [email protected] (K. Thompson). 8756-3282/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.bone.2010.10.177
and musculoskeletal aches and pains. These symptoms are frequently associated with increased circulating levels of pro-inflammatory cytokines such as IFNγ, TNFα and IL-6 [5,7–9]. Whilst the exact molecular mechanism underlying this phenomenon is not fully understood, it is becoming clear that activation of a specific subset of T cells, so-called γδ T cells, plays a central role. γδ T cells were first identified as possible initiators of the APR to N-BPs by Kunzmann et al., who observed marked increases in the number of circulating γδ T cells in PAM-treated multiple myeloma patients up to 28 days post-infusion, which correlated with the severity of the APR [10]. Subsequent studies revealed that N-BPs specifically activate the major subset of γδ T cells in peripheral blood, Vγ9Vδ2 T cells [11]. N-BP-induced activation of Vγ9Vδ2 T cells in peripheral blood mononuclear cell cultures in vitro results in production of IFNγ, TNFα and IL-6 [5,12,13], which closely mirrors cytokine production triggered by N-BP administration in vivo. This suggests that strategies to minimise or prevent Vγ9Vδ2 T cell activation by N-BPs in vivo may prevent the APR.
K. Thompson et al. / Bone 49 (2011) 140–145
We and others have previously shown that activation of Vγ9Vδ2 T cells by N-BPs occurs via an indirect mechanism [12,14], requiring intracellular uptake of N-BP to inhibit its molecular target, farnesyl diphosphate (FPP) synthase [15–18]. Recently, we proposed that peripheral blood monocytes play a crucial role in Vγ9Vδ2 T cell activation due to selective internalisation of the N-BP by highly endocytic CD14+ monocytes [19]. We demonstrated that, following a clinically relevant pulse of ZOL, inhibition of FPP synthase in monocytes causes the accumulation of the enzyme substrates, isopentenyl diphosphate (IPP) and dimethylallyl diphosphate (DMAPP), both of which are agonists of the Vγ9Vδ2-T cell receptor (TCR) [20]. Consistent with a central role for IPP and DMAPP accumulation in triggering Vγ9Vδ2 T cell activation, we and others have previously shown that the stimulatory effects of N-BPs on Vγ9Vδ2 T cell activation and proliferation in peripheral blood mononuclear cell cultures can be prevented by simultaneous treatment with a statin [12–14]. These widely-used cholesterol-lowering drugs inhibit HMGCoA reductase, an enzyme upstream of FPP synthase, and thus prevent N-BP-induced IPP/DMAPP accumulation. We therefore investigated in this study whether co-administration of a statin could prevent ZOLinduced cytokine release and flu-like symptoms of the APR in healthy post-menopausal women.
141
Siemens reagents. Serum levels of type I collagen C-telopeptide (CTx) were measured using a β-Crosslaps/serum kit and an Elecsys 2010 Immunoassay Analyser (Roche). Proportion of Vγ9Vδ2 T cells in the circulation Peripheral blood samples were collected at baseline (0 hr), 48 hr, 4 weeks and 8 weeks, using EDTA as anti-coagulant. Peripheral blood mononuclear cells (PBMCs) were isolated by density-gradient separation using Lymphoprep reagent (Axis-Shield). The number of Vγ9Vδ2 T cells in peripheral blood, relative to the total T cell population, was then determined by immunostaining using anti-human Vδ2-FITC (Beckman Coulter) and anti-CD3-PerCP (BD Biosciences) antibodies, followed by flow cytometric analysis. Data acquisition and analysis was performed on a BD FACSCalibur flow cytometer using CellQuestPro software. Assessment of acute-phase reaction symptoms Symptoms characteristic of the APR were assessed in 6 categories (joint aches; muscle aches; bone aches and pains; nausea; headache; flu-like symptoms) by questionnaire using a 4-point scale (1 = no symptoms; 2 = mild; 3 = moderate; 4 = severe). Questionnaires were completed at 8 hr, 24 hr, 48 hr and 72 hr post-infusion of ZOL.
Materials and methods
Statistical analysis
Study population and design
The study was designed with the primary end-point being the effectiveness of FLU to inhibit the APR based on follow-up measurements of CRP. In assessing the power of the study it was assumed that at least 50% of subjects would get a N2-fold rise in their peak CRP values (CRP rising from a mean of 10 ± 4 mg/dl at baseline to 30 ± 12 mg/dl) and that pre-treatment with FLU would prevent this elevation. Based on these two assumptions, and a drop-out rate of 10% during follow-up, we aimed to recruit 60 women (20 subjects per group), in order to give 90% power with an alpha of less than 0.05. All statistical analyses were performed using SPSS software and SAS v9.1 (SAS Institute). A fixed-effects repeated measures analysis was conducted to test for statistical differences between the three treatment groups. Where a significant interaction was found between treatment group and time, differences between the treatment groups at each time point were investigated separately. To adjust for multiple comparisons, p b 0.01 was taken to be statistically significant. Measurements that were not normally distributed (CRP, TNFα, IFNγ, IL-6, Vγ9Vδ2 T cells and cholesterol) were log-transformed and geometric means (95% CI) were calculated from the model. Two patients were excluded from the analysis due to abnormal baseline values of CRP and TNFα, respectively.
All research was conducted at Aberdeen Royal Infirmary, Aberdeen, UK. Ethical approval for this study was granted by the Grampian Local Research Ethics Committee. A total of 61 healthy women aged 50–70, who were N12 months post-menopause and had a T-score ≥2.5 (i.e. non-osteoporotic), were recruited into the double-blind, placebocontrolled study. Participants were BP-naïve and had not received hormone replacement therapy within 6 months of inclusion. Subjects receiving statin therapy were not excluded, but their therapy was suspended for 7 days prior to the ZOL infusion, and recommenced 7 days post-infusion. One participant withdrew from the study postrandomisation but before treatment commenced, resulting in 60 study participants. Subjects were randomly assigned to 3 treatment groups (n = 20). All study participants received a single 5 mg zoledronic acid intravenous infusion over 15 min. In addition, one group (designated 3xFLU) received three times 40 mg oral fluvastatin (FLU) (at 0 hr, 24 hr and 48 hr); a further group (designated FLU+ PLAC) received 40 mg oral FLU (at 0 hr) plus oral placebo (at 24 hr and 48 hr); the final group (designated PLAC) received oral placebo (at 0 hr, 24 hr and 48 hr). The first dose of FLU or placebo was administered 30 min prior to infusion of ZOL, to allow peak plasma concentrations of FLU during the ZOL administration, in line with the pharmacokinetic properties of FLU [21]. Study participants were allowed to self-administer acetaminophen to treat symptoms such as headache and pyrexia as required, and were encouraged to increase their fluid intake for 24 hr following the infusion. Measurement of pro-inflammatory cytokines, C-reactive protein, cholesterol, and type I collagen C-telopeptide Serum samples were obtained at baseline (0 hr), and at 4 hr, 8 hr, 24 hr, 48 hr, 4 weeks, 8 weeks and 12 weeks after infusion of ZOL, and stored at −80 °C until analysis. Levels of pro-inflammatory cytokines were determined using Quantikine ELISA kits (IFNγ & IL-6) or a Quantikine Highly Sensitive ELISA kit (TNFα) (all R&D Systems). Serum Hi-Sensitivity C-reactive protein (CRP) was determined using a standard nephelometric immunoassay on a Behring nephelometer with Behring reagents. Total serum cholesterol was determined using a standard colorimetric method on a Siemens auto-analyser with
Results and discussion Patient information and determination of FLU treatment protocol Sixty female volunteers completed the study. The mean age was 60.6 ± 4.0 years. All volunteers were post-menopausal and presented with a T-score of ≥2.5. The subject disposition is shown in Fig. 1. The proposed timing and dosage of FLU for the study was determined based on our findings in an in vitro model of the APR to N-BPs (Suppl. Fig. 1) using IFNγ release as a measure of N-BP-induced γδ T cell activation in PBMC cultures [12]. This showed that treatment of PBMCs with 1 μΜ FLU for 2 hr (which mimics the Cmax following a typical 40 mg dose [21]) was effective in preventing γδ T cell activation induced by a pulse of 1 μM ZOL for 2 hr (which mimics the Cmax following a typical iv dose of ZOL [22]). FLU pre-treatment period (3 hr, 1 hr, 0 hr) did not influence the effectiveness of FLU in our in vitro assay system (Suppl. Fig. 1), indicating that FLU does not need to be present long before ZOL to prevent γδ T cell activation. For
142
K. Thompson et al. / Bone 49 (2011) 140–145
Fig. 1. Study flow diagram.
this reason, we administered FLU at a dose (40 mg) and a time (30 min prior to the ZOL infusion) that would be anticipated to give a sufficiently high Cmax (~1 μM [21]) known to inhibit ZOL-induced Vγ9Vδ2 T cell activation in vitro, and that would be maintained for the duration of the peak ZOL concentration in peripheral blood to ensure inhibition of HMG-CoA reductase and prevention of ZOL-induced IPP accumulation that typically occurs within 1-3 hr following ZOL exposure in vitro [14]. Changes in cytokine levels following ZOL ZOL administration alone (PLAC group) induced a marked elevation in mean serum CRP, which first showed a statistically significant increase at 24 hr (p b 0.001), and was further increased at
48 hr (p b 0.001: Fig. 2). CRP values had returned to baseline levels at 4 weeks post-infusion. This increase in CRP was accompanied by increased levels of the pro-inflammatory cytokines IL-6, TNFα and IFNγ. Serum IL-6 levels increased rapidly, with a ~ 3-fold increase detected within the first 4 hr following ZOL infusion (p b 0.001), peaked at 24 hr (p b 0.001), and had returned towards baseline values at 48 hr. In contrast, serum TNFα or IFNγ were not elevated at 4 or 8 hr following ZOL infusion, but were significantly increased at 24 hr (p b 0.001), with TNFα levels remaining elevated at 48 hr (p b 0.001). All cytokines had returned to baseline levels at 4 weeks post-ZOL infusion. Previous findings from in vitro studies have shown that monocytes are directly targeted by ZOL whilst activation of Vγ9Vδ2 T cells occurs secondary to the internalisation of ZOL by monocytes and accumulation of IPP in these cells [19]. The increase in serum IL-6 in
Fig. 2. Effect of fluvastatin on serum levels of pro-inflammatory cytokines following ZOL administration in healthy post-menopausal women. Following treatment, serum levels of pro-inflammatory cytokines (A) TNFα, (B) IFNγ, and (C) IL-6 were determined by ELISA. (D) Serum Hi-Sensitivity C-reactive protein (CRP) was determined by nephelometric immunoassay. Data shown are the geometric mean (95% CI) of 20 patients/treatment group. (Symbols denote p b 0.001 vs baseline (0 hr) values: *PLAC group; #FLU + PLAC group; † 3xFLU group).
K. Thompson et al. / Bone 49 (2011) 140–145
143
response to ZOL infusion ahead of the increases in TNFα and IFNγ therefore suggests that monocytes are the predominant producers of IL-6 in vivo in response to ZOL, whereas activated Vγ9Vδ2 T cells are the major source of IFNγ and TNFα. In support of this, the N-BP alendronate has been previously shown to augment LPS-induced IL-6 release from macrophages, although alendronate alone had no stimulatory effect [23]. However, a previous in vitro study suggested that, in a subset of healthy donors, N-BP-treated Vγ9Vδ2 T cells themselves are the source of IL-6 [13]. Thus, which cell types secrete which cytokines in response to ZOL in vivo remains to be clarified.
Table 2 Total number of mild, moderate and severe symptoms reported in 6 symptom categories per treatment group (20 participants per group).
Relationship between changes in cytokine levels and development of APR symptoms following ZOL
although there was a trend towards decreased incidence of nausea and increased incidence of headaches in the FLU + PLAC and 3xFLU groups, compared to the ZOL alone (PLAC) group. The mean severity of APR symptoms was also not different between either of the FLU treatment groups and the PLAC (ZOL alone) group at either 24 or 48 hr (Fig. 3). At 72 hr post-infusion, mean symptom severity scores had decreased markedly in all treatment groups, consistent with the transient nature of the APR. The lack of effect of FLU for preventing ZOL-induced cytokine release or APR symptoms was not due to differences between treatment groups in self-administration of antipyretic or NSAID medication, since similar numbers of study participants took rescue medication in both PLAC and FLU + PLAC groups (both 12/20), whilst this proportion was increased in the 3xFLU group (16/20).
Analysis of cytokine increases in individual study participants revealed that at least 70% of study participants treated with ZOL alone (PLAC group) had an increase in pro-inflammatory cytokine and CRP levels of ≥2 S.D. over mean baseline values (Table 1). IFNγ and IL-6 were increased in 90% and CRP in 80% of study participants treated with ZOL alone. However, whilst the majority of participants experienced significant cytokine increases, this did not always result in reported APR symptoms, since only 55% (11 out of 20) of participants receiving ZOL alone experienced flu-like symptoms to various degrees within 72 hr of the ZOL infusion (Table 2). Correlating individual cytokine increases with APR symptoms revealed that no single cytokine had greater prognostic power (8 out of 11 showed increased TNFα; 9 out of 11 showed increased IFNγ, IL-6 and CRP; Table 3). Furthermore, most participants that reported no APR symptoms demonstrated elevated cytokine levels (6 out of 9 showed increased TNFα; 9 out of 9 showed increased IFNγ, IL-6 and CRP), indicating that marked increases in the release of pro-inflammatory cytokines do not always trigger the development of APR symptoms. This suggests a more complex relationship between increased cytokine levels and triggering of flu-like symptoms than previously suggested [9]. Recent reports have shown that regulatory T cells may limit responses of Vγ9Vδ2 T cells to phosphoantigen stimulation in vitro [24,25]. However, the majority of our study participants experienced ZOL-induced elevations in circulating cytokines, suggesting that Vγ9Vδ2 T cell responsiveness is not impaired, either via regulatory T cells or through other regulatory pathways, in these subjects. Further studies will be required to fully elucidate the relationship between N-BP-induced cytokine release and APR symptoms in vivo. Effect of FLU on changes in cytokine levels and development of APR symptoms following ZOL FLU administration, either single or triple-dosing, did not have any significant inhibitory effect on ZOL-induced increases in mean CRP, TNFα, IFNγ or IL-6 (Fig. 2), although there was a non-significant trend for FLU to decrease serum IL-6 levels at 24 hr (PLAC: 12.79 (7.77, 21.06) pg/ml; FLU + PLAC: 9.82 (5.77, 16.69) pg/ml; 3xFLU: 7.88 (4.84, 12.81) pg/ml). Furthermore, no difference was found in the number of study participants who experienced significant increases in serum levels of TNFα, IFNγ, IL-6 or CRP in either the FLU + PLAC or the 3xFLU treatment groups, as compared to the PLAC group (Table 1). Consistent with this, FLU administration (either single or triple dose) did not decrease the reported incidence of APR symptoms (Table 2),
Table 1 Number of study participants (out of 20) with elevated cytokine levels (N 2× standard deviations above baseline) within 48 hr of ZOL infusion. Treatment group
TNFα
IFNγ
IL-6
CRP
PLAC FLU + PLAC 3xFLU
14 15 10
18 17 16
18 16 17
16 15 16
Treatment Joint Muscle Bone aches Nausea Headache Flu-like Total subjects group aches aches and pains symptoms with symptoms PLAC FLU+ PLAC 3xFLU
10 9
10 10
8 7
9 2
10 12
11 9
11 13
7
13
11
4
16
11
13
Effects of ZOL and FLU on circulating Vγ9Vδ2 T cell levels ZOL treatment induced a non-significant, ~ 30% decrease in circulating levels of Vγ9Vδ2 T cells (relative to the total T cell population) in peripheral blood at 48 hr, which recovered to baseline levels by 4 weeks (Fig. 4). The decrease in circulating levels of Vγ9Vδ2 T cells is suggestive of Vγ9Vδ2 T cell activation and subsequent extravasation of activated T cells into peripheral lymphoid tissues. Transient lymphopenia following intravenous N-BP administration has previously been reported [4,7,26,27], which indicates that the pro-inflammatory cascade of cytokines triggered by γδ T cells promotes a more general extravasation of lymphocytes from the peripheral circulation. However, the studies by Kunzmann et al. that first implicated γδ T cells in the APR to N-BPs reported increased circulating levels of γδ T cells, which were maintained for up to 4 weeks post-N-BP infusion [11]. Similarly, ZOL (particularly in combination with IL-2 administration) has been reported to result in sustained increases in circulating Vγ9Vδ2 T cells in patients with hormone-refractory prostate cancer [28] and lymphoid malignancies [29]. This discrepancy is suggestive of (as yet poorly characterised) differences in Vγ9Vδ2 T cell reactivity between cancer patients and the healthy population in our study. No correlation was found between experienced symptoms (and/or cytokine increases) and circulating levels of Vγ9Vδ2 T cells, either at baseline, or at 48 hr when APR symptoms were most prevalent (data not shown). Co-administration of FLU, either a single dose (FLU+ PLAC group) or three consecutive doses (3xFLU group), did not prevent the ZOL-induced decrease in Vγ9Vδ2
Table 3 Frequency of cytokine increases in participants reporting acute-phase reaction symptoms per treatment group. Treatment group
TNFα
PLAC FLU + PLAC 3xFLU
8/11 10/13 6/13
a
IFNγ
a
9/11 12/13 7/13
IL-6a
CRP
a
9/11 10/13 12/13
9/11 12/13 11/13
a The first value represents the number of participants with acute-phase reaction symptoms that showed an increase in cytokine level (N 2 S.D. above baseline), followed by the total number of participants experiencing symptoms.
144
K. Thompson et al. / Bone 49 (2011) 140–145
for inhibiting bone resorption in our study population. FLU treatment, using either treatment regimen, had no significant effect on the ZOLinduced decrease in serum CTx levels at any of the 3 time-points investigated (48 hr, 4 or 12 weeks; Suppl. Fig. 3). Safety
Fig. 3. Effect of concurrent fluvastatin administration on severity of ZOL-induced acute-phase reaction symptoms. Following treatment, acute-phase reaction symptoms (joint aches; muscle aches; bone aches and pains; nausea; headache; flu-like symptoms) were assessed using a 4-point scale (1=no symptoms; 2=mild; 3=moderate; 4=severe) at 8 hr, 24 hr, 48 hr and 72 hr post-ZOL infusion. A symptom score of 6 indicates no symptoms, whereas a score of 24 indicates severe symptoms in all 6 categories. Data shown are the mean reported symptom scores of 20 patients/treatment group±S.E.M.
T cell levels at 48 hr, nor was there any effect of FLU treatment on the recovery of circulating Vγ9Vδ2 T cell levels at 4 weeks (Fig. 4). This is consistent with the lack of effect of FLU on the increased serum cytokine and CRP levels in response to ZOL. Effects of ZOL and FLU on serum cholesterol levels ZOL administration resulted in a transient decrease in total serum cholesterol of ~ 10% at 48 hr. Cholesterol levels recovered to baseline values by 4 weeks (Suppl. Fig. 2). Co-administration of either a single dose of FLU (FLU + PLAC group), or three consecutive doses of FLU (3xFLU group) did not significantly decrease total serum cholesterol levels greater than ZOL treatment alone at 48 hr, although there was a trend towards a greater decrease in serum cholesterol in the triple dose FLU group. Effects of ZOL and FLU on bone resorption ZOL treatment resulted in a profound and sustained decrease in bone resorption, as assessed by serum CTx (Suppl. Fig. 3). The inhibitory effect on bone resorption was rapid, with maximal inhibition (~95% reduction in serum CTx) observed after 48 hr, and was sustained (N80% reduction) at 12 weeks following ZOL infusion. There was no significant correlation between severity of APR symptoms and inhibition of CTx release (data not shown), suggesting that APR symptom severity was not indicative of effectiveness of ZOL
Apart from the APR symptoms reported, the therapies tested were generally well-tolerated. Six serious adverse events were reported including 2 hospitalisations due to non-study related problems. There were 3 cases of short-lasting uveitis/iritis (one bi-lateral): 1 in PLAC group, 2 in 3xFLU group. One patient who had uveitis also developed gum swelling during the study period, which settled with conservative saline mouthwash therapy. Conclusions We investigated whether FLU could prevent the most common adverse event associated with intravenous ZOL administration, the development of an APR, in a placebo-controlled, double-blind study in a healthy cohort of post-menopausal women. Our results show that, despite the inhibitory effect of FLU on N-BP-induced Vγ9Vδ2 T cell activation in vitro (Suppl. Fig. 1 & [13]), co-administration of FLU at the dosing regimens used did not prevent ZOL-induced Vγ9Vδ2 T cell activation, pro-inflammatory cytokine release, or APR symptoms in our study population. The lack of effect of FLU on incidence or severity of APR symptoms is consistent with a recent report detailing the lack of efficacy of atorvastatin for preventing ZOL-induced APR symptoms (increased serum CRP, musculoskeletal pain, and fever) in a small cohort of 12 children, although a non-significant trend was observed for decreased pain scores and diminished use of rescue medication (oxycodone and acetaminophen) in response to co-administration of atorvastatin [30]. Furthermore, whilst major advances in elucidating the mechanisms involved in Vγ9Vδ2 T cell activation by N-BPs have been made in recent years through in vitro studies [12–14], the findings of our study highlight that the exact mechanism underlying the APR to N-BPs such as ZOL remains to be clarified. In particular, the relationship between cytokine increases and the manifestation of symptoms awaits more thorough characterisation. Supplementary materials related to this article can be found online at doi:10.1016/j.bone.2010.10.177. Acknowledgments We gratefully acknowledge the technical assistance provided by Mrs Denise Tosh, Mrs Kelly Deans and Mrs Linda Duncan. We would also like to thank Dr. Anke Roelofs for critical comments on the manuscript. The study was funded by an unrestricted educational grant by Novartis. DMR and KT are grateful to Arthritis Research UK for continued infrastructure support. DMR has received additional research funding from Novartis, participates in Advisory Boards for the company and is a member of their speakers panel. MJR has received research funding from Novartis, Roche, The Alliance for Better Bone Health, and Merck. References
Fig. 4. Effect of fluvastatin on ZOL-induced changes in circulating levels of Vγ9Vδ2 T cells. The proportion of Vγ9Vδ2 T cells present in the CD3+ (T cell) population of peripheral blood mononuclear cells was determined by immunostaining with anti-human Vδ2-TCRFITC and anti-human CD3-PerCP antibodies and flow cytometric analysis. Data acquisition and analysis were performed using a Becton Dickinson FACSCalibur flow cytometer and CellQuest Pro software on the lymphocyte-gated population. Data shown are the geometric mean (95% CI) of 20 patients/treatment group.
[1] Kanis JA, Burlet N, Cooper C, Delmas PD, Reginster JY, Borgstrom F, et al. European Society for Clinical and Economic Aspects of Osteoporosis and Osteoarthritis (ESCEO). European guidance for the diagnosis and management of osteoporosis in postmenopausal women. Osteoporos Int 2008;19:399–428. [2] Ralston SH, Langston AL, Reid IR. Pathogenesis and management of Paget's disease of bone. Lancet 2008;372:155–63. [3] Berenson JR, Rajdev L, Broder M. Treatment strategies for skeletal complications of cancer. Cancer Biol Ther 2006;5:1074–7. [4] Adami S, Bhalla AK, Dorizzi R, Montesanti F, Rosini S, Salvagno G, et al. The acutephase response after bisphosphonate administration. Calcif Tissue Int 1987;41: 326–31.
K. Thompson et al. / Bone 49 (2011) 140–145 [5] Thiebaud D, Sauty A, Burckhardt P, Leuenberger P, Sitzler L, Green JR, et al. An in vitro and in vivo study of cytokines in the acute-phase response associated with bisphosphonates. Calcif Tissue Int 1997;61:386–92. [6] Strampel W, Emkey R, Civitelli R. Safety considerations with bisphosphonates for the treatment of osteoporosis. Drug Saf 2007;30:755–63. [7] Sauty A, Pecherstorfer M, Zimmer-Roth I, Fioroni P, Juillerat L, Markert M, et al. Interleukin-6 and tumor necrosis factor alpha levels after bisphosphonates treatment in vitro and in patients with malignancy. Bone 1996;18: 133–9. [8] Schweitzer DH, Oostendorp-Van de Ruit M, Van der Pluijm G, Löwik CWGM, Papapoulos SE. Interleukin-6 and the acute phase response during treatment of patients with Paget's disease with the nitrogen-containing bisphosphonate dimethylaminohydroxypropylidene bisphosphonate. J Bone Miner Res 1995;10:956–62. [9] Dicuonzo G, Vincenzi B, Santini D, Avvisati G, Rocci L, Battistoni F, et al. Fever after zoledronic acid administration is due to increase in TNF-alpha and IL-6. J Interferon Cytokine Res 2003;23:649–54. [10] Kunzmann V, Bauer E, Wilhelm M. Gamma/delta T-cell stimulation by pamidronate. N Engl J Med 1999;340:737–8. [11] Kunzmann V, Bauer E, Feurle J, Weissinger F, Tony HP, Wilhelm M. Stimulation of gammadelta T cells by aminobisphosphonates and induction of antiplasma cell activity in multiple myeloma. Blood 2000;96:384–92. [12] Thompson K, Rogers MJ. Statins prevent bisphosphonate-induced gamma, delta-T-cell proliferation and activation in vitro. J Bone Miner Res 2004;19: 278–88. [13] Hewitt RE, Lissina A, Green AE, Slay ES, Price DA, Sewell AK. The bisphosphonate acute phase response: rapid and copious production of proinflammatory cytokines by peripheral blood γδ T cells in response to aminobisphosphonates is inhibited by statins. Clin Exp Immunol 2005;139:101–11. [14] Gober HJ, Kistowska M, Angman L, Jenö P, Mori L, De Libero G. Human T cell receptor gammadelta cells recognize endogenous mevalonate metabolites in tumor cells. J Exp Med 2003;197:163–8. [15] Van Beek E, Pieterman E, Cohen L, Lowik C, Papapoulos S. Farnesyl pyrophosphate synthase is the molecular target of nitrogen-containing bisphosphonates. Biochem Biophys Res Commun 1999;264:108–11. [16] Bergstrom JD, Bostedor RG, Masarachia PJ, Reszka AA, Rodan G. Alendronate is a specific, nanomolar inhibitor of farnesyl diphosphate synthase. Arch Biochem Biophys 2000;373:231–41. [17] Dunford JE, Thompson K, Coxon FP, Luckman SP, Hahn FM, Poulter CD, et al. Structure-activity relationships for inhibition of farnesyl diphosphate synthase in vitro and inhibition of bone resorption in vivo by nitrogen-containing bisphosphonates. J Pharmacol Exp Ther 2001;296:235–42.
145
[18] Kavanagh KL, Guo K, Dunford JE, Wu X, Knapp S, Ebetino FH, et al. The molecular mechanism of nitrogen-containing bisphosphonates as anti-osteoporosis drugs: crystal structure and inhibition of farnesyl pyrophosphate synthase. Proc Natl Acad Sci USA 2006;103:7829–34. [19] Roelofs AJ, Jauhiainen M, Mönkkönen H, Rogers MJ, Mönkkönen J, Thompson K. Peripheral blood monocytes are responsible for gammadelta T cell activation induced by zoledronic acid through accumulation of IPP/DMAPP. Br J Haematol 2009;144:245–50. [20] Tanaka Y, Morita CT, Tanaka Y, Nieves E, Brenner MB, Bloom BR. Natural and synthetic non-peptide antigens recognized by human gamma delta T cells. Nature 1995;375:155–8. [21] Corsini A, Bellosta S, Baetta R, Fumagalli R, Paoletti R, Bernini F. New insights into the pharmacodynamic and pharmacokinetic properties of statins. Pharmacol Ther 1999;84:413–28. [22] Chen T, Berenson J, Vescio R, Swift R, Gilchick A, Goodin S, et al. Pharmacokinetics and pharmacodynamics of zoledronic acid in cancer patients with bone metastases. J Clin Pharmacol 2002;42:1228–36. [23] Makkonen N, Salminen A, Rogers MJ, Frith JC, Urtti A, Azhayeva E, et al. Contrasting effects of alendronate and clodronate on RAW 264 macrophages: the role of a bisphosphonate metabolite. Eur J Pharm Sci 1999;8:109–18. [24] Mahan CS, Thomas JJ, Boom WH, Rojas RE. CD4+ CD25(high) Foxp3+ regulatory T cells downregulate human Vdelta2+ T-lymphocyte function triggered by antiCD3 or phosphoantigen. Immunology 2009;127:398–407. [25] Kunzmann V, Kimmel B, Herrmann T, Einsele H, Wilhelm M. Inhibition of phosphoantigen-mediated gammadelta T-cell proliferation by CD4+ CD25+ FoxP3+ regulatory T cells. Immunology 2009;126:256–67. [26] Harinck HIJ, Papapoulos SE, Blanksma HJ, Moolenaar AJ, Vermeij P, Bijvoet OLM. Paget's disease of bone: early and late responses to three different modes of treatment with aminohydroxypropylidene bisphosphonate (APD). Br Med J 1987;295:1301–5. [27] Pecherstorfer M, Jilch R, Sauty A, Horn E, Keck AV, Zimmer-Roth I, et al. Effect of first treatment with aminobisphosphonates pamidronate and ibandronate on circulating lymphocyte subpopulations. J Bone Miner Res 2000;15:147–54. [28] Dieli F, Vermijlen D, Fulfaro F, Caccamo N, Meraviglia S, Cicero G, et al. Targeting human gamma}delta T cells with zoledronate and interleukin-2 for immunotherapy of hormone-refractory prostate cancer. Cancer Res 2007;67:7450–7. [29] Wilhelm M, Kunzmann V, Eckstein S, Reimer P, Weissinger F, Ruediger T, et al. Gammadelta T cells for immune therapy of patients with lymphoid malignancies. Blood 2003;102:200–6. [30] Srivastava T, Haney CJ, Alon US. Atorvastatin may have no effect on acute phase reaction in children after intravenous bisphosphonate infusion. J Bone Miner Res 2009;24:334–7.