ORIGINAL ARTICLE

Long-term administration of quarterly IV ibandronateis effective and well tolerated in postmenopausalosteoporosis: 5-year data from the DIVA studylong-term extension

G. Bianchi & E. Czerwinski & A. Kenwright &A. Burdeska & R. R. Recker & D. Felsenberg

Received: 27 April 2011 /Accepted: 2 September 2011# International Osteoporosis Foundation and National Osteoporosis Foundation 2011

AbstractSummary Long-term bone mineral density (BMD) gains,bone marker levels, and safety of 3 mg quarterly intravenous(IV) ibandronate were studied in this 3-year extension to theDosing IntraVenous Administration (DIVA) trial. QuarterlyIV ibandronate consistently increased lumbar spine bonemineral density measured with dual-energy X-ray absorpti-ometry (DXA-BMD) over 5 years (8.1%) and was welltolerated in women with postmenopausal osteoporosis.Introduction Treatment with IV ibandronate regimens, 2 mgbimonthly and 3 mg quarterly, has been studied for up to5 years in a long-term extension (LTE) to the 2-year DIVA trial.Methods DIVA LTE is an open-label extension to a 2-yearrandomized, double-blind, double-dummy, noninferiority,

phase III study (DIVA core). DIVA LTE involved post-menopausal women who had completed 2 years of DIVAcore, comparing daily oral and IV ibandronate (≥75%adherence with IV ibandronate in year 2 of DIVA). Patientspreviously treated with 2 mg bimonthly or 3 mg quarterlyIV ibandronate continued on the same regimen; patientswho had received 2.5 mg daily oral ibandronate andplacebo IV in DIVA core were switched to IV ibandronate.Results Pooled analysis of 497 intent-to-treat (ITT) patientsreceiving IV ibandronate from DIVA core baseline showedconsistent increases over 5 years in lumbar spine DXA-BMD (8.4% [95% confidence interval (CI)=7.5, 9.3] with2 mg bimonthly and 8.1% [95% CI=7.2, 8.9] with 3 mgquarterly). Three-year data relative to DIVA LTE baseline

Results from this manuscript have previously been presented inabstract form at EULAR 2009 (Bianchi G et al. (2009) Ann RheumDis 68(Suppl 3):494).

Funding sources The DIVA study LTE was funded by F. Hoffmann-LaRoche and GlaxoSmithKline.

G. Bianchi (*)Division of Rheumatology, Azienda Sanitaria Genovese,Genoa, Italye-mail: gerolamo_bianchi@tin.it

E. CzerwinskiKrakow Medical Centre,Krakow, Polande-mail: czerwinski@kcm.pl

A. KenwrightRoche Products Limited,Welwyn Garden City, UKe-mail: andy.kenwright@roche.com

A. BurdeskaF. Hoffmann-La Roche Ltd.,Basel, Switzerlande-mail: alexander.burdeska@roche.com

R. R. ReckerOsteoporosis Research Center, Creighton University,Omaha, NE, USAe-mail: rrecker@creighton.edu

D. FelsenbergCharité—Campus Benjamin Franklin,Freie and Humboldt-Universität Berlin,Berlin, Germanye-mail: dieter.felsenberg@charite.de

Osteoporos IntDOI 10.1007/s00198-011-1793-9

in the full ITT population (756 patients randomized orreallocated from DIVA, including those previously on dailytreatment) showed maintenance of DXA-BMD gains fromDIVA core with further gains in lumbar spine DXA-BMD.These benefits are supported by sustained reductions inmarkers of bone metabolism. No tolerability concerns ornew safety signals were observed.Conclusions Treatment with IV ibandronate 2 mg bimonthlyor 3 mg quarterly is effective and well tolerated for up to5 years in women with postmenopausal osteoporosis.

Keywords Bisphosphonate . Clinical study . Ibandronate .

Intravenous . Long-term treatment . Postmenopausalosteoporosis

Introduction

Bisphosphonates are the mainstay of treatment for post-menopausal osteoporosis, a chronic bone disease associatedwith an increased risk for fractures and associated morbid-ity and mortality [1]. Adherence to medication in postmen-opausal osteoporosis tends to be disappointing, and poorlevels of compliance and persistence have been reportedspecifically with bisphosphonates [2, 3]. Poor patientadherence to daily and weekly dosing of oral bisphospho-nates [4] prompted the development of formulations withmore convenient, extended, between-dose intervals. Twoibandronate studies incorporating patient surveys haveindicated a strong preference by patients for monthly overweekly treatment [5, 6], and the 6-month UK PERsistenceStudy of Ibandronate verSus alendronaTe (PERSIST)showed a 47% improvement with monthly ibandronatecompared with weekly alendronate in the proportion ofpatients persisting with treatment [7]. However, monthlyoral bisphosphonate therapy may not be suitable for allpatients, such as those who experience gastrointestinal (GI)intolerance or those who have difficulty adhering tomonthly oral dosing, indicating a need for an intravenous(IV) formulation of bisphosphonate. IV ibandronate hasgreater bioavailability than the oral formulation (100% vs0.6%) and can be administered less frequently whileproviding the same or higher total annual cumulativeexposure (ACE) to the bone [8].

Results from the 2-year Dosing IntraVenous Adminis-tration (DIVA) study have shown that IV injections ofibandronate are an effective and well-tolerated treatment forpostmenopausal osteoporosis and provide a useful alterna-tive to oral dosing [9, 10]. Both IV ibandronate regimens(2 mg bimonthly and 3 mg quarterly, each providing anACE of 12 mg) gave significantly (P<0.001) greaterincreases in lumbar spine bone mineral density (BMD)measured with dual-energy X-ray absorptiometry (DXA-

BMD) after 2 years of treatment compared with the 2.5-mgdaily oral regimen (ACE=5.5 mg; 6.4% and 6.3% vs 4.8%,respectively) [10]. Based on the results from DIVA, 3 mgquarterly IV ibandronate was licensed for the treatment ofpostmenopausal osteoporosis in the US and Europe.

There are no general recommendations for theoptimum duration of treatment with bisphosphonates;therefore, it is suggested that physicians should beguided by evaluating fracture risk of the patient and thelong-term efficacy and safety of the treatment [11].Long-term effects on BMD have been reported with 10and 7 years of continuous treatment with the oralbisphosphonates alendronate [12] and risedronate [13],respectively, and 6 years of IV zoledronate [14]. Similarly,the Monthly Oral iBandronate In LadiEs (MOBILE) studyhas demonstrated that monthly treatment with 150 mg oralibandronate provides sustained BMD gains and is welltolerated for up to 5 years in women with postmenopausalosteoporosis [15].

The DIVA study was extended for an additional 3 yearsto further characterize the long-term effect on BMD andmarkers of bone turnover and safety of IV ibandronate inpatients with postmenopausal osteoporosis. Here, wepresent data from this long-term extension (LTE), the firststudy to report 5-year data on BMD, markers of boneturnover, and safety for IV ibandronate.

Methods

Study design and participants

DIVA LTE was an international, 3-year, open-label exten-sion to a 2-year randomized, double-blind, double-dummy,noninferiority, phase III study (DIVA core). Participantseligible for DIVA LTE were postmenopausal women whohad completed 2 years of the DIVA core study and were≥75% compliant with the IV regimen in the second year oftreatment. The study was conducted in 39 of the 58 centersoriginally participating in DIVA. Inclusion and exclusioncriteria for the DIVA core study have been publishedpreviously [9]. Women recruited to the DIVA core studywere aged 55–80 years, at least 5 years postmenopausal,and had osteoporosis (mean lumbar spine [L2–L4] DXA-BMD T-score <−2.5 but ≥−5.0). Original randomizationwas to one of four ibandronate treatment arms: 2.5 mg oraldaily plus IV placebo (bimonthly or quarterly); 2 mg IVbimonthly plus daily oral placebo; or 3 mg IV quarterlyplus daily oral placebo.

All patients provided written and informed consent,ethical approval was obtained from the institutional reviewboards of the participating institutions, and the study wascarried out in accordance with the principles of the

Osteoporos Int

Declaration of Helsinki and Good Clinical Practice. DIVAClinicalTrials.gov identifier: NCT00048074.

In DIVA LTE, eligible patients were allocated to one oftwo open-label IV treatment groups: patients previouslytreated with 2 mg bimonthly or 3 mg quarterly IVibandronate in the 2-year DIVA core study continued onthe same regimen for a further 3 years in DIVA LTE;patients who had previously received 2.5 mg daily oralibandronate and IV placebo in the DIVA core wereswitched to IV ibandronate for DIVA LTE (either 2 mgbimonthly or 3 mg quarterly according to their placebo IVtreatment schedule in DIVA core). All patients receivedelemental calcium 500 mg/day and vitamin D 400 IU/day.Each IV injection was given over 15–30 s.

Assessments

The primary variable of the LTE was the relative change (inpercent) in mean lumbar spine (L2–L4) DXA-BMD bycentrally evaluated DXA from the end of the 2-year DIVAcore study (month 24) to completion of the DIVA LTE atmonth 60. The pooled analysis of the relative change inmean lumbar spine DXA-BMD in patients who receivedcontinuous IV (2 mg bimonthly or 3 mg quarterly)treatment for up to 5 years (i.e., pooled data from patientswho received treatment in both the core 2-year DIVA studyand the present DIVA LTE trial) is the focus of the presentmanuscript. DXA-BMD scans were conducted at DIVAcore baseline, month 12, LTE baseline (month 24), andmonths 36, 48, and 60.

The secondary endpoints included relative change (inpercent) in mean total hip DXA-BMD from the end of the2-year DIVA core study (month 24) to completion of theDIVA LTE at month 60 and relative change (in percent) inmedian serum levels of C-terminal peptide of type Icollagen (CTX) assessed from blood samples taken at theend of the dosing interval (immediately before dose) andmeasured from the end of the 2-year DIVA core study(month 24) to completion of the DIVA LTE at month 60.

Exploratory endpoints included (1) the relative (inpercent) and absolute (in grams per square centimeter)change in mean femoral neck and trochanter DXA-BMDfrom the end of the 2-year DIVA study (month 24) tocompletion of the DIVA LTE at month 60; (2) the relative(in percent) and absolute (in grams per square centimeter)change in mean lumbar spine (L2–L4), total hip, femoralneck, and trochanter DXA-BMD from the start of the DIVAcore study to completion of the DIVA LTE in a pooledanalysis of patients who received continuous IV (2 mgbimonthly or 3 mg quarterly) treatment for up to 60 months(5 years); and (3) the relative change in median serumlevels of CTX (sCTX) and serum procollagen type 1 N-terminal propeptide (sP1NP) assessed from blood samples

taken at the end of the dosing interval (immediately beforedose) and measured from the start of the DIVA core studyto completion of the DIVA LTE at month 60 in the pooledpopulation of patients who received continuous IV (2 mgbimonthly or 3 mg quarterly).

For DIVA core and LTE, sCTX levels were analyzedcentrally using the Elecsys S-CTX-I assay (an Electro-ChemiLuminescence ImmunoAssay technique). sP1NPlevels were measured post hoc, based on the fact thatsP1NP demonstrates greater sample stability and lowerintrapatient variability than sCTX, using stored serumsamples from a subset of patients with measurements fromthe DIVA core and LTE studies. Adverse events (AEs) andlaboratory safety parameters were monitored throughout.AEs were graded as mild, moderate, severe, or life-threatening; a serious adverse event (SAE) was anyexperience that suggested a significant hazard, contraindi-cation, side effect, or precaution. Renal function wasmeasured by changes in serum creatinine and blood ureanitrogen (BUN) levels as well as by changes in thecalculated glomerular filtration rate (GFR). Patients withsevere renal failure (calculated GFR<30 mL/min) wereexcluded from the study.

Analysis populations

The primary analysis population for BMD and markers ofbone turnover endpoints was the intent-to-treat (ITT)population. Patients were included in the ITT populationfor the individual analysis of the LTE study if they wererandomized, received at least one dose of trial medication,and had a valid baseline and at least one follow-up DXA-BMD or sCTX measurement. Patients were included in theITT population for the pooled analyses if they met theabove criteria and were randomized to IV ibandronate inDIVA. Patients who switched from oral ibandronate 2.5 mgdaily to one of the IV regimens were excluded from thepooled analysis to show treatment response for thosepatients who received the IV regimen continuously for5 years. Data published previously from the 2-year DIVAcore study are for the per-protocol (PP) population and, assuch, the data for the DIVA LTE are not directlycomparable [10].

The safety population comprised all patients whoreceived at least one dose of study medication and attendedat least one safety follow-up visit. Patients were included inthe safety population for the pooled analyses if they metthese criteria and received IV ibandronate in DIVA.

The PP population was used to summarize sCTX andP1NP for the pooled analysis because the protocol violatorsin an ITT analysis can affect the sCTX results greatly, i.e.,not fasting before dosing. The PP population comprised allpatients in the ITT population who had no major protocol

Osteoporos Int

violations and were at least 75% compliant. For DIVA LTE,participation in the sCTX and P1NP analyses occurred atselected sites only with a planned population number of200 (100 patients from each treatment group). Patientnumbers for these biomarker analyses differ for DIVA coreand DIVA LTE due to the further limited number ofsamples available for those patients who participated in thebone turnover marker assessment in the extension study andwho also received the same continuous IV ibandronatedosing regimen over the whole 5-year study period.

No formal statistical testing was carried out in DIVALTE due to partial randomization and the study notbeing powered to detect statistical differences betweenthe groups. Results are presented as descriptive statis-tical summaries, which include 95% confidence inter-vals (CIs) for relative changes from baseline in eachtreatment arm.

Results

Patient disposition and baseline characteristics

A total of 781 women were enrolled in DIVA LTE andreceived treatment with 2 mg IV ibandronate bimonthly(n=381) or 3 mg IV ibandronate quarterly (n=400). Ofthese, 25 patients were excluded from the ITT population

because they had no valid baseline and/or follow-upinformation for BMD/markers of bone turnover. As such,756 patients were included in the ITT population forDIVA LTE (362 and 394 for 2 and 3 mg groups,respectively). A total of 329 (86%) patients in the 2-mggroup and 359 (90%) patients in the 3-mg groupcompleted DIVA LTE (Fig. 1). Women who continued2 mg ibandronate bimonthly (n=254) or 3 mg ibandronatequarterly (n=264) through both the DIVA core study andthe DIVA LTE were included in the pooled analysis safetypopulation (n=518); the pooled analysis ITT populationcomprised 497 patients (Table 1). Baseline patient char-acteristics were similar between the treatment groups forboth ITT populations. Table 1 shows the baselinecharacteristics for the pooled analysis ITT population.

DXA-BMD and markers of bone turnover

Pooled 5-year analysis

Lumbar spine DXA-BMD increased consistently over5 years of continuous treatment with both doses of IVibandronate (Fig. 2). At month 60, mean lumbar spineDXA-BMD was substantially increased relative to DIVAcore baseline (8.4% [95% CI=7.5, 9.3] and 8.1% [7.2,8.9]) in the 2- and 3-mg treatment groups, respectively(Fig. 2).

Fig. 1 Patient disposition(3-year safety populationfrom DIVA LTE baseline).a refused treatment includes ‘didnot cooperate’ and ‘withdrewconsent’, bother includes‘patient moved to another city,’‘was not cooperative with studyrequirements,’ and ‘DXA-BMDdecreased after 1 year of study’,c other protocol violationincludes ‘patient tookbisphosphonates’

Osteoporos Int

At 12, 24, and 36 months, increases in mean total hipDXA-BMD relative to DIVA core baseline were reported inboth treatment groups (Fig. 3). A plateau was reachedbetween 24 and 36 months, with mean increases at36 months of 3.6% (95% CI=3.2, 4.1) in the 2-mg groupand 3.2% (95% CI=2.6, 3.8) in the 3-mg group. After60 months (5 years), mean increases in total hip DXA-BMD were 3.0% (95% CI=2.4, 3.5) and 2.8% (95% CI=2.1, 3.5) in the 2- and 3-mg groups, respectively.

Mean DXA-BMD gains at the femoral neck and trochanterfollowed a similar pattern to those at the total hip, reaching aplateau at 36 months. Mean femoral neck DXA-BMDincreased by 3.4% (95% CI=2.8, 4.0) at month 36 relativeto DIVA core baseline in the 2-mg group and by 3.0% (95%CI=2.4, 3.7) in the 3-mg group. This gain was maintained to60 months in the 3-mg treatment group. Mean trochanterDXA-BMD increased by 5.5% (95%CI=4.9, 6.2) atmonth 36relative to DIVA core baseline in the 2-mg group and by 5.5%

(95% CI=4.6, 6.4) in the 3-mg group. These gains weregenerally maintained to 60 months in both treatment groups.

In the first 6 months of treatment, median sCTX valuesrapidly decreased by more than 50% from DIVA core baselinein both treatment groups (Fig. 4). During subsequent years oftreatment, median sCTX values remained at approximately40% below baseline, which was generally maintained up to60 months (Fig. 4). Median sP1NP values also showed arapid and pronounced decrease from DIVA core baseline inthe first year (74% and 63% in the 2- and 3-mg treatmentgroups, respectively). Thereafter, sP1NP values appeared toslowly increase, but remained consistently at 30% belowbaseline in both treatment groups up to month 60 (Fig. 5). At0, 24, and 60 months, median sCTX values were 0.49, 0.21,and 0.26 ng/mL and 0.49, 0.22, and 0.28 ng/mL in the 2- and3-mg IV ibandronate groups, respectively. Correspondingvalues for sP1NP were 51.1, 16.7, and 23.4 ng/mL for the 2-mg group and 49.4, 18.6, and 25.6 ng/mL for the 3-mg

Table 1 Baseline demographicsof patients entering DIVALTE (mean values±standarddeviation) from DIVA corebaseline for the pooled analysisITT population, those patientsreceiving continuous IVibandronate only

aValues for sCTX and sP1NPare medians (ranges)bsP1NP was not included inthe original reporting but wasanalyzed post hoc

Ibandronate 2 mg q2months (n=239)

Ibandronate 3 mg q3months (n=258)

Age (years) 66.1±6.1 65.9±6.1

Weight (kg) 64.5±11.2 64.2±9.8

Height (cm) 158.7±6.3 158.6±7.0

Body mass index (kg/m2) 25.6±4.2 25.6±4.1

Lumbar spine (L2–L4) DXA-BMD (g/cm2) 0.75±0.07 0.74±0.08

n=239 n=258

Lumbar spine (L2–L4) DXA-BMD (T-score) −2.96±0.67 −3.07±0.70n=239 n=258

Total hip DXA-BMD (g/cm2) 0.75±0.10 0.74±0.10

n=238 n=256

sCTX (ng/mL)a 0.49 (0.02, 1.67) 0.49 (0.04, 1.23)

n=239 n=256

sP1NP (ng/mL)a, b 49.73 (2.50, 194.30) 49.55 (10.12, 107.30)

n=76 n=76

Fig. 2 Mean relative change (inpercent) with 95% CIs in lumbarspine (L2–L4) DXA-BMD fromDIVA core baseline over 5 yearswith 2 mg IV ibandronate every2 months and 3 mg IV ibandro-nate every 3 months (ITT popu-lation; pooled analysis). Numbersof patients available for evalua-tion at each visit are shown foreach data point and representthose who received IV ibandro-nate only in DIVA core and DIVALTE. aat baseline

Osteoporos Int

group. These data are within the premenopausal range (sCTXpremenopausal range, 0.162–0.436 ng/mL [16]; sP1NPpremenopausal range, 13.8–60.9 ng/mL [17]).

DIVA LTE 3-year analysis

Mean lumbar spine DXA-BMD increased at all time pointsrelative to DIVA LTE baseline (month 24) with a meanincrease in the 2- and 3-mg groups at month 60 of 2.0% (95%CI=1.5, 2.5) and 2.1% (95% CI=1.6, 2.5), respectively(Table 2). Mean total hip DXA-BMD increased at months 36and 48 in the 2-mg group (0.4% [95% CI=0.2, 0.7] and0.05% [95% CI=−0.3, 0.4], respectively) and at 36 monthsin the 3-mg group (0.1% [95% CI=−0.2, 0.4]). However,small and clinically nonrelevant decreases in mean total hipDXA-BMD relative to DIVA core baseline were seen atmonth 60 with the 2-mg dose and at months 48 and 60 withthe 3-mg dose (Table 2). Small increases in DXA-BMD werealso seen at the femoral neck and trochanter at all timepoints, including month 60 (Table 2). Median sCTX andsP1NP levels increased over the 3 years of the LTE relative

to DIVA LTE baseline (i.e., after 2 years of treatment in theDIVA core study had already been completed; Table 2).

Safety and tolerability

Five-year analysis

The safety population comprised 518 patients. Over thecourse of 5 years, the proportions of patients with at leastone AE were comparable between the IV regimens (98%for the 2-mg group and 94% for the 3-mg group). The mostfrequent events in both treatment groups (incidence of atleast 10% in either treatment group) were nasopharyngitis,influenza, back pain, arthralgia, and hypertension (Table 3).Influenza-like symptoms were reported in 8.3% of patientsin the 2-mg group and 4.5% of patients in the 3-mg group.The most common AEs considered to be related to studytreatment were related to the GI system (20% and 14% inthe 2- and 3-mg groups, respectively) and included upperabdominal pain, dyspepsia, constipation, nausea, andgastritis (Table 3). The incidence of influenza-like illness

Fig. 3 Mean relative change (inpercent) with 95% CIs in totalhip BMD from DIVA corebaseline over 5 years with 2 mgIV ibandronate every 2 monthsand 3 mg IV ibandronate every3 months (ITT population;pooled analysis). Numbers ofpatients available for evaluationat each visit are shown for eachdata point and represent thosewho received IV ibandronateonly in DIVA core and DIVALTE. a at baseline

Fig. 4 Median relative change(in percent) with 95% CIs insCTX levels from DIVA corebaseline to year 5 with monthlyibandronate (PP population;pooled analysis). Numbers ofpatients available for evaluation ateach visit are shown for each datapoint and represent those whoreceived IV ibandronate only inDIVA core and DIVA LTE(samples not taken according tothe specified sampling schedulewere excluded from the analyses).a at baseline

Osteoporos Int

considered related to treatment was 6.7% and 4.5% in the2- and 3-mg groups, respectively.

There was no notable increase in the frequency ofAEs compared with the DIVA core study and noimportant differences in frequency and type of AEsbetween the two treatment arms, with no new orunexpected safety signals. In general, the proportion ofpatients reporting at least one SAE (31% and 29% inthe 2- and 3-mg groups, respectively) and the types ofevents reported (including fractures, musculoskeletal andconnective tissue disorders, myocardial infarctions, andpneumonia) were as expected for an elderly populationof patients.

There were no clinically relevant changes over 5 years inhematology or clinical chemistry values during treatment ineither ibandronate group. Creatinine clearance valuesoverall showed no detrimental effect of treatment on renalfunction (mean rate of decrease was 1.5 mL/min/year in

both groups over 5 years of treatment) and no cases ofosteonecrosis of the jaw were reported.

DIVA LTE 3-year analysis

A similar safety profile to the pooled analysis populationwas reported for the safety population of 781 patients overthe 3-year extension period: at least one AE was reported in87% (2 mg group) and 85% (3 mg group) of patients; themost frequent AEs reported were the same as those for thepooled analysis (Table 3), but with lower incidences(≤21%); most AEs were mild to moderate in severity;influenza-like symptoms were reported in 2.6% (2 mggroup) and 0.5% (3 mg group) of patients. Incidences ofSAEs were 24% (2 mg group) and 21% (3 mg group) withonly one SAE considered treatment-related (ulcerativecolitis [2 mg group]); 11 deaths were reported, allconsidered unrelated to study treatment. No cases of

Table 2 After 60 months oftreatment, relative (in percent)changes (mean values±standarddeviation with 95% CIs) inDXA-BMD and bone turnovermarkers from DIVA LTEbaseline (24 months) for thecomplete ITT population

aMedian values with 95% CIs;PP population

Ibandronate 2 mg q2months (n=362)

Ibandronate 3 mg q3months (n=394)

Lumbar spine (L2–L4) DXA-BMD 2.0±4.7 (1.5, 2.5) 2.1±4.5 (1.6, 2.5)

n=314 n=349

Total hip DXA-BMD −0.2±3.4 (−0.5, 0.2) −0.3±4.0 (−0.7, 0.2)n=314 n=349

Femoral neck DXA-BMD 0.2±4.5 (−0.3, 0.7) 0.8±5.8 (0.2, 1.4)

n=314 n=349

Trochanter DXA-BMD 0.6±4.5 (0.1, 1.1) 0.4±5.5 (−0.2, 1.0)n=314 n=349

sCTXa 35.4 (5.9, 50.3) 25.0 (9.2, 54.5)

n=76 n=75

sP1NPa 29.1 (20.2, 40.1) 24.1 (11.9, 41.9)

n=76 n=75

Fig. 5 Median relative change(in percent) with 95% CIs insP1NP levels from DIVA corebaseline to year 5 with monthlyibandronate (PP population;pooled analysis). Numbers ofpatients available for evaluation ateach visit are shown for each datapoint and represent those whoreceived IV ibandronate only inDIVA core and DIVA LTE andtook part in the bone turnovermarker substudy during theextension study (samples nottaken according to the specifiedsampling schedule were excludedfrom the analyses). a at baseline

Osteoporos Int

osteonecrosis of the jaw were reported. Overall rates of allclinical fractures were 11.0% (2 mg group) and 8.5% (3 mggroup).

Discussion

The two IV ibandronate regimens (the licensed 3 mgquarterly and the unlicensed 2 mg bimonthly regimen) eachprovide an ACE of 12 mg but with different dosingfrequencies. In a pooled analysis, “high” ACE ibandronatedoses of at least 10.8 mg (including the licensed doses of150 mg once-monthly oral and 3 mg IV quarterly and theunlicensed dose of 2 mg IV every 2 months) wereassociated with significant reductions in nonvertebralfracture risk (38% and 43%, respectively) compared witha “low” ACE ibandronate dose (ACE of 5.5 mg, including

2.5 mg daily oral) [18]. In the current study, progressiveincreases in lumbar spine DXA-BMD were observed inpatients treated with both IV ibandronate regimens initiallyin DIVA core with further substantial increases observedduring DIVA LTE over a total treatment period of60 months. Notably, in the original DIVA study, DXA-BMD increases with IV ibandronate were superior to thosefor daily oral ibandronate [9, 10], which has been shown toprovide significant DXA-BMD gains over placebo [19].

These findings are consistent with those reportedrecently from the MOBILE LTE in which 60 months ofcontinuous treatment with 150 mg monthly oral ibandro-nate provided a progressive increase in lumbar spine DXA-BMD [15]. The sustained lumbar spine DXA-BMDincrease observed over 60 months predicts sustainedantifracture efficacy with continued treatment, as indicatedby a recent meta-analysis of DIVA and three other phase IIIclinical trials of ibandronate [20] and by longer-term trialsof alendronate and risedronate [12, 13]. The pattern ofmean relative DXA-BMD gain vs baseline observed overthe 5-year observation period, which shows a prominentincrease for the first 1–3 years followed by a more modestincrease or slight reduction (as illustrated in Figs. 2 and 3),is also similar to that seen with these other bisphospho-nates, which have been shown to provide continued long-term antifracture efficacy [12, 13]. In the DIVA LTE study,DXA-BMD gains at the proximal femur (total hip, femoralneck, and trochanter) sites, which were achieved during thefirst 24 months of treatment of the DIVA core study, weregenerally maintained during the 3 years of the DIVA LTEstudy.

Despite their relatively high variability, the use of boneturnover markers, such as sCTX and sP1NP, to monitor theeffectiveness of osteoporosis therapies and to identifypatients at high risk for bone loss, is increasing [21]. Inthe DIVA LTE study, levels of sCTX, which declinedrapidly by ≥50% from baseline in both treatment groupsduring the first 6 months of the DIVA core study, remainedwithin the premenopausal range throughout 60 months ofcontinuous IV treatment. It is expected that maximalinhibition of sCTX would be achieved by the end of thecore study (month 24; DIVA LTE baseline). No furtherdecreases in sCTX were expected during the LTE and theseresults are consistent with those observed for monthly oralibandronate in the MOBILE LTE study (Miller et al.,unpublished). As with sCTX, median sP1NP values fellrapidly during the first year of treatment in both the 2-mgbimonthly and 3-mg quarterly treatment arms. At the end of60 months of continuous treatment, median sP1NP valueswere 57% and 45% lower than at DIVA core baseline, i.e.,within the premenopausal range. Although no furtherreductions in biomarker levels were observed from DIVAcore, biomarker levels remained within the premenopausal

Table 3 Summary of the most common AEs occurring with anincidence of at least 10% and the most common AEs occurring withan incidence of at least 2% considered to be related to treatment over5 years of IV ibandronate treatment

Ibandronate2 mg q2 months(n=254)

Ibandronate3 mg q3 months(n=264)

Number of patients with at leastone AE, n (%)

248 (97.6) 248 (93.9)

Nasopharyngitis 82 (32.3) 59 (22.3)

Back pain 70 (27.6) 85 (32.2)

Arthralgia 65 (25.6) 47 (17.8)

Hypertension 63 (24.8) 66 (25.0)

Influenza 51 (20.1) 47 (17.8)

Hypercholesterolemia 49 (19.3) 30 (11.4)

Osteoarthritis 38 (15.0) 36 (13.6)

Bronchitis 31 (12.2) 28 (10.6)

Urinary tract infection 34 (13.4) 19 (7.2)

Pain in extremity 31 (12.2) 22 (8.3)

Musculoskeletal pain 29 (11.4) 25 (9.5)

Cystitis 28 (11.0) 18 (6.8)

Diarrhea 26 (10.2) 20 (7.6)

Number of patients with at leastone AE, n (%)

111 (43.7) 102 (38.6)

Upper abdominal pain 6 (2.4) 9 (3.4)

Dyspepsia 6 (2.4) 9 (3.4)

Constipation 8 (3.1) 5 (1.9)

Nausea 8 (3.1) 3 (1.1)

Gastritis 5 (2.0) 3 (1.1)

Arthralgia 9 (3.5) 9 (3.4)

Myalgia 15 (5.9) 3 (1.1)

Back pain 7 (2.8) 7 (2.7)

Bone pain 8 (3.1) 2 (0.8)

Influenza-like illness 17 (6.7) 12 (4.5)

Osteoporos Int

range [16, 17] throughout the study and, therefore,remained clinically relevant in terms of responsiveness totreatment [22].

Limitations to the DIVA LTE study include the lackof randomization of the treatment groups, the smallsample size of patients who received IV ibandronatethroughout the 5 years, and the fact that not all eligiblepatients in DIVA core trial continued into the extensionstudy. No formal power calculation was made as thesample size was based on achieving sufficient safetyinformation. Thus, there was no formal statisticalcomparison of BMD, markers of bone turnover, orsafety data between the two treatment arms.

The long-term safety and tolerability profiles weresimilar in the 2- and 3-mg IV groups over 60 months. Asexpected, there were no notable increases in the incidenceof AEs in the LTE and no new or unexpected safety signalscompared with the DIVA core study. Although there is noplacebo control group to compare these safety data with, inDIVA core, the safety profiles of both IV ibandronateregimens were similar to that of oral daily ibandronate,which overall did not differ from the placebo group in thepivotal BONE study [9, 10, 19]. In the DIVA LTE study,reports of influenza-like symptoms were infrequent andoccurred mainly after the first few injections. This isconsistent with other data, suggesting that influenza-likesymptoms in patients treated with IV ibandronate aretypically transitory and self-limiting and do not generallyrecur with subsequent injections [23, 24]. Renal toxicity hasalso been a concern with IV bisphosphonates, and renaltoxicity has been reported with pamidronate and zoledro-nate [25]. For DIVA LTE, neither ibandronate treatmentregimen had a detrimental effect on renal function, asmeasured by serum creatinine levels and BUN; the overallchange in renal function over the course of 60 months wasas expected for this elderly population with postmenopausalosteoporosis [26]. Chronic use of some bisphosphonateshas been linked to rare, but serious AEs [27]. Reports ofosteonecrosis of the jaw have been confined largely to theuse of IV zoledronate and pamidronate at the higher dosesused to treat oncology patients with metastatic or lytic bonelesions [28, 29]. Several publications have demonstratedthat long-term oral alendronate therapy for osteoporosis isassociated with low-energy, insufficiency fractures and thatthese may be the result of oversuppressed bone turnover.Such insufficiency fractures have become apparent in theclinical setting in a general, uncontrolled population likelywith comorbidities and/or receiving concomitant medica-tions, which may be contributory factors [30–40]. Symp-toms suggestive of osteonecrosis of the jaw (e.g., localizedpain and numbness, exposed bone in the oral cavity, andloose teeth) and atypical fractures with ibandronate werenot observed in the DIVA study.

Conclusions

Ibandronate is currently the only licensed bisphosphonateproviding both monthly oral and quarterly IV treatmentoptions for postmenopausal osteoporosis. The approved andmarketed 3 mg quarterly IV ibandronate regimen delivers asimilar ACE to that of the monthly oral formulation andprovides sustained DXA-BMD gains and reduction in boneturnover markers. Quarterly IV ibandronate is an attractivealternative to patients who experience GI intolerance withoral bisphosphonate preparations and/or who may adherebetter to an extended-dosing interval. These results from thefirst trial to examine the effectiveness and safety of long-term IV bisphosphonate treatment support the continueduse of IV ibandronate at the marketed dose.

Acknowledgements Support for third-party writing assistance forthis manuscript was provided by F. Hoffmann-La Roche Ltd.

Conflicts of interest Dr. Bianchi has received honoraria and/orconsulting fees from Abbott, Amgen, Eli Lilly, GlaxoSmithKline,Merck Sharp & Dohme, Novartis, Pfizer, Roche, Schering Plough,and Servier. Dr. Czerwinski has received research grants from Amgen,Eli Lilly, Merck, Novartis, Pfizer, Procter and Gamble, Roche, andServier. Dr. Recker is a paid consultant for Merck, Lilly, Pfizer,Procter and Gamble, Amgen, Roche, Glaxo Smith Kline, and Novartisand has received grant/research support from Merck, Lilly, Wyeth,Procter and Gamble, Amgen, Roche, Glaxo Smith Kline, Novartis,and Sanofi-Aventis through grants to his institution. Dr. Felsenbergserves as a consultant for Amgen, Chugai, GlaxoSmithKline, Lilly,Merck, Novartis, Roche, and Servier. Dr. Burdeska and Dr. Kenwrightare employees of F. Hoffmann-La Roche; Dr Kenwright has someshares in this company.

References

1. NIH Consensus Statement (2000) Osteoporos Prev Diagn Ther 17(1):27–29

2. Cramer JA, Gold DT, Silverman SL, Lewiecki EM (2007) Asystematic review of persistence and compliance with bisphosph-onates for osteoporosis. Osteoporos Int 18:1023–1031

3. Landfeldt E, Ström O, Robbins S, Borgström F (2011) Adherenceto treatment of primary osteoporosis and its association tofractures—the Swedish Adherence Register Analysis (SARA).Osteoporos Int (in press)

4. Reginster J-Y, Rabenda V (2006) Patient preference in themanagement of postmenopausal osteoporosis with bisphospho-nates. Clin Interv Aging 1:415–423

5. Emkey R, Koltun W, Beusterien K, Seidman L, Kivitz A, DevasV, Masanauskaite D (2005) Patient preference for once-monthlyibandronate versus once-weekly alendronate in a randomized,open-label, cross-over trial: the Boniva Alendronate Trial inOsteoporosis (BALTO). Curr Med Res Opin 21:1895–1903

6. Hadji P, Minne H, Pfeifer M, Bourgeois P, Fardellone P, Licata A,Devas V, Masanauskaite D, Barrett-Connor E (2008) Treatmentpreference for monthly oral ibandronate and weekly oral alendr-onate in women with postmenopausal osteoporosis: a randomized,crossover study (BALTO II). Joint Bone Spine 75:303–310

7. Cooper A, Drake J, Brankin E, Investigators PERSIST (2006)Treatment persistence with once-monthly ibandronate and patient

Osteoporos Int

support vs. once-weekly alendronate: results from the PERSISTstudy. Int J Clin Pract 60:896–905

8. Barrett J, Worth E, Bauss F (2004) Ibandronate: a clinicalpharmacological and pharmacokinetic update. J Clin Pharmacol44:951–965

9. Delmas PD, Adami S, Strugala C, Stakkestad JA, Reginster JY,Felsenberg D, Christiansen C, Civitelli R, Drezner MK, ReckerRR, Bolognese M, Hughes C, Masanauskaite D, Ward P,Sambrook P, Reid DM (2006) Intravenous ibandronate injectionsin postmenopausal women with osteoporosis: one-year resultsfrom the dosing intravenous administration study. Arthritis Rheum54:1838–1846

10. Eisman JA, Civitelli R, Adami S, Czerwinski E, Recknor C,Prince R, Reginster JY, Zaidi M, Felsenberg D, Hughes C, MaironN, Masanauskaite D, Reid DM, Delmas PD, Recker RR (2008)Efficacy and tolerability of intravenous ibandronate injections inpostmenopausal osteoporosis: 2-year results from the DIVA study.J Rheumatol 35:488–497

11. Geusens P (2009) Bisphosphonates for postmenopausal osteopo-rosis: determining duration of treatment. Curr Osteoporos Rep7:12–17

12. Bone HG, Hosking D, Devogelaer J-P, Tucci JR, Emkey RD,Tonino RP, Rodriguez-Portales JA, Downs RW, Gupta J, SantoraAC, Liberman UA, Alendronate Phase III Osteoporosis TreatmentStudy Group (2004) Ten years’ experience with alendronate forosteoporosis in postmenopausal women. N Engl J Med 350:1189–1199

13. Mellström DD, Sörensen OH, Goemaere S, Roux C, Johnson TD,Chines AA (2004) Seven years of treatment with risedronate inwomen with postmenopausal osteoporosis. Calcif Tissue Int75:462–468

14. Black D, Reid I, Cauley J, Boonen S, Cosman F, Leung PC,Lakatos P, Man Z, Cummings S, Hue T, Ruzycky M-E, MartinezR, Su G, Bucci-Rechtweg C, Eastell R (2010) The effect of 3versus 6 years of zoledronic acid treatment in osteoporosis: arandomized extension to the HORIZON Pivotal Fracture Trial(PFT). J Bone Miner Res 25(Suppl. 1):1070. Available at http://www.asbmr.org/Meetings/AnnualMeeting/AbstractDetail.aspx?aid=4781a020-dc3a-46be-b49d-f706b5509c4e. Accessed 13January 2011

15. Felsenberg D, Czerwinski E, Stakkestad J, Neate C, MasanauskaiteD, Reginster JY (2009) Efficacy of monthly oral ibandronate ismaintained over 5 years: the MOBILE LTE study. Osteoporos Int 20(suppl1):S15, Abstract OC32

16. Garnero P, Borel O, Delmas PD (2001) Evaluation of a fullyautomated serum assay for C-terminal cross-linking telopeptide oftype I collagen in osteoporosis. Clin Chem 47:694–702

17. Garnero P, Vergnaud P, Hoyle N (2008) Evaluation of a fullyautomated serum assay for total N-terminal propeptide of type Icollagen in postmenopausal osteoporosis. Clin Chem 54:188–196

18. Cranney A, Wells GA, Yetisir E, Adami S, Cooper C, Delmas PD,Miller PD, Papapoulos S, Reginster JY, Sambrook PN, SilvermanS, Siris E, Adachi JD (2009) Ibandronate for the prevention ofnonvertebral fractures: a pooled analysis of individual patient data.Osteoporos Int 20:291–297

19. Chesnut CH, Skag A, Christiansen C, Recker R, Stakkestad JA,Hoiseth A, Felsenberg D, Huss H, Gilbride J, Schimmer RC, DelmasPD, for the Oral Ibandronate Osteoporosis Vertebral Fracture Trial inNorth America and Europe (BONE) (2004) Effects of oralibandronate administered daily or intermittently on fracture risk inpostmenopausal osteoporosis. J Bone Miner Res 19:1241–1249

20. Sebba AL, Emkey RD, Kohles JD, Sambrook PN (2009)Ibandronate dose response is associated with increases in bonemineral density and reductions in clinical fractures: results of ameta-analysis. Bone 44:423–427

21. Brown JP, Albert C, Nassar BA, Adachi JD, Cole D, Davison KS,Dooley KC, Don-Wauchope A, Douville P, Hanley DA, JamalSA, Josse R, Kaiser S, Krahn J, Krause R, Kremer R, Lepage R,Letendre E, Morin S, Ooi DS, Papaioaonnou A, Ste-Marie LG(2009) Bone turnover markers in the management of postmeno-pausal osteoporosis. Clin Biochem 42:929–942

22. Papapoulos SE, Schimmer RC (2007) Changes in bone remodel-ing and antifracture efficacy of intermittent bisphosphonatetherapy: implications from clinical studies with ibandronate. AnnRheum Dis 66:853–858

23. Croom KF, Scott LJ (2006) Intravenous ibandronate: in thetreatment of osteoporosis. Drugs 66:1593–1601

24. Strampel W, Emkey R, Civitelli R (2007) Safety considerationswith bisphosphonates for the treatment of osteoporosis. Drug Saf30:755–763

25. Papapetrou PD (2009) Bisphosphonate-associated adverse events.Hormones 8:96–110

26. Klawansky S, Komaroff E, Cavanaugh PF Jr, Mitchell DY,Gordon MJ, Connelly JE, Ross SD (2003) Relationship betweenage, renal function and bone mineral density in the US population.Osteoporos Int 14:570–576

27. Watts NB, Diab DL (2010) Long-term use of bisphosphonates inosteoporosis. J Clin Endocrinol Metab 95:1555–1565

28. Reid IR (2009) Osteonecrosis of the jaw—who gets it, and why?Bone 44:4–10

29. Wimalawansa SJ (2008) Insight into bisphosphonate-associatedosteomyelitis of the jaw: pathophysiology, mechanisms andclinical management. Expert Opin Drug Saf 7:491–512

30. Sayed-Noor AS, Sjödén GO (2010) Case reports: two femoralinsufficiency fractures after long-term alendronate therapy. ClinOrthop Relat Res 467:1921–1926

31. Odvina CV, Levy S, Rao S, Zerwekh JE, Sudhaker Rao D (2009)Unusual mid-shaft fractures during long term bisphosphonatetherapy. Clin Endocrinol (Oxf) 72:161–168

32. Ing-Lorenzini K, Desmeules J, Plachta O, Suva D, Dayer P, PeterR (2009) Low-energy femoral fractures associated with the long-term use of bisphosphonates: a case series from a Swiss universityhospital. Drug Saf 32:775–785

33. Kwek EB, Goh SK, Koh JS, Png MA, Howe TS (2008) Anemerging pattern of subtrochanteric stress fractures: a long-termcomplication of alendronate therapy? Injury 39:224–231

34. Lenart BA, Lorich DG, Lane JM (2008) Atypical fractures of thefemoral diaphysis in postmenopausal women taking alendronate.N Engl J Med 358:1304–1306

35. Neviaser AS, Lane JM, Lenart BA, Edobor-Osula F, Lorich DG(2008) Low-energy femoral shaft fractures associated withalendronate use. J Orthop Trauma 22:346–350

36. Lee P, van der Wall H, Seibel MJ (2007) Looking beyond lowbone mineral density: multiple insufficiency fractures in a womanwith post-menopausal osteoporosis on alendronate therapy. JEndocrinol Invest 30:590–597

37. Cheung RK, Leung KK, Lee KC, Chow TC (2007) Sequentialnontraumatic femoral shaft fractures in a patient on long-termalendronate. Hong Kong Med J 13:485–489

38. Goh SK, Yang KY, Koh JS, Wong MK, Chua SY, Chua DTC, HoweTS (2007) Subtrochanteric insufficiency fractures in patients onalendronate therapy: a caution. J Bone Joint Surg Br 89:349–353

39. Giusti A, Hamdy NAT, Papapoulos SE (2010) Atypical fracturesof the femur and bisphosphonate therapy. A systematic review ofcase/case series studies. Bone 47:169–180

40. Giusti A, Hamdy NAT, Dekkers OM, Ramautar SR, Dijkstra PDS,Papapoulos SE (2011) Atypical fractures and bisphosphonatetherapy: a cohort study of patients with femoral fracture withradiographic adjudication of fracture site and features. Bone48:966–971

Osteoporos Int