Disintegration and Esophagealmail.jrnlappliedresearch.com/articles/Vol5Iss2/Epstein.pdf ·...

The Journal of Applied Research • Vol. 5, No. 2, 2005 253

Disintegration and EsophagealIrritation Profiles of AlendronateFormulations: Implications forClinical Safety and Efficacy

*Mount Sinai School of Medicine, New York, NY, USA and University of Pennsylvania,Philadelphia, PA, USA †Biomedical Research Institute-LUC, Diepenbeek, Belgium and Department of Rheumatology,University Hospital, Maastricht, The Netherlands ‡Merck Research Laboratories, West Point, PA, USA §Howard Hughes Medical Institute, University of Pennsylvania, Philadelphia, PA, USAIIMerck Frosst, Montreal, Canada

excipients, are typically derived fromsmall, single-dose, bioavailability stud-ies. While these studies provide infor-mation on systemic effects, they do notaddress the risk of local esophageal irri-tation and may, therefore, inadequatelycharacterize safety profiles.

Objective: To compare the esophagealirritation potential of NFA preparationsto that of the innovator medication,Fosamax.

Methods: Two preclinical models of irri-tation were used. In the first, 24 rabbitswere randomized to a single, subcuta-neous injection of saline, 10.6 mg ofFosamax in saline, or 10.6 mg NFA insaline. Blinded measurements of skinthickness (a measure of inflammation),wet weight, and histopathology of injec-tion site tissues were performed. In the

KEY WO R D S : a l e n d r o n a t e, e s o p h a g i t i s,t o l e r a b i l i t y, s a f e t y

ABSTRACTBackground: Fosamax (alendronatesodium, Merck & Co., Inc., WhitehouseStation, NJ, USA) is an effective oralbisphosphonate widely used to treat andprevent osteoporosis, with a safety andtolerability profile similar to placebo inclinical trials. It has been evaluated inclinical trials with over 20,000 partici-pants and up to 10 years duration. Oralbisphosphonates have been associatedwith esophagitis, which involves eventsthat occur prior to absorption anddepends on factors such as the frequen-cy of administration, dose, and formula-tion. Data on non- Fosamax alendronate(NFA) preparations, which contain aform of alendronate with differing

Solomon Epstein, MD*

Piet Geusens, MD†

John E. Fisher, PhD‡

Susan L. Hill, DVM‡

Sophie Roy, PhD‡

Gideon Rodan, MD, PhD§

Nagaraja Muniappa, MVSc, PhD‡

Gordon K.Wo l l e n b e r g, DV M , P h D‡

Laurence Handt, DVM‡

Nancy Kelly, DVM, MScII

Chi-Chung Chan, PhDII

Alfred A. Reszka, PhD‡

Srinivasa Prahalada, BVSc, PhD‡

Epstein-vol5no2 6/1/05 8:20 PM Page 253

Vol. 5, No. 2, 2005 • The Journal of Applied Research254

second study, 16 dogs were anesthetizedand a placebo tablet, a 10-mg Fosamaxtablet, or a 10-mg NFA tablet wasplaced by endoscopy in the caudal thirdof the esophagus for 1 hour, followed bya saline rinse, daily for 5 days. After thefinal dose, the dogs were sacrificed andesophageal morphology was examined.

R e s u l t s: In the rabbit injection study, t h eN FA suspension elicited a significantlygreater irritant response than an equiva-lent suspension of Fo s a m a x . The meanwet weight increase at injection site tis-sues was 70% greater (P < 0.01) for NFAthan Fo s a m a x . Fosamax treatmentinduced smaller, predominantly foreign-body granuloma type lesions, while NFAinduced larger, encapsulated cy s t i clesions containing the injected material,consistent with a post-inflammatoryp r o c e s s. In the esophagus study, all 4 dogs(100%) treated with NFA for 5 daysexhibited marked ulcerative esophagitis,whereas only 1 of the 5 dogs treated withFosamax (20%) had marked ulceration;the remaining 4 dogs had more moderateesophageal changes than those observedin dogs treated with NFA .

Conclusions: Generic drugs are expectedto have similar efficacy and safety toinnovator drugs; however, the greaterirritant responses of NFA in rabbits anddogs suggest that important differencesmay exist between the effects ofFosamax and NFA preparations in theclinical setting. These findings, alongwith other data demonstrating differ-ences in the disintegration/dissolutionprofiles of NFA preparations relative toFosamax, suggest that bioavailabilitystudies may not be adequate for mean-ingful assessment of the safety and effi-cacy of NFA or other bisphosphonatepreparations.

INTRODUCTIONFosamax (alendronate sodium, Merck &

Co., Inc., Whitehouse Station, NJ, USA)is a nitrogen-containing oral bisphos-phonate (N-BP) that exhibits a potentand selective inhibitory effect on osteo-clast-mediated bone resorption. It hasbeen evaluated in clinical trials withmore than 20,000 participants and up to10 years’ duration and is approvedworldwide for the treatment of post-menopausal osteoporosis.1-3 Fosamax isalso an established therapy for the pre-vention of osteoporosis.4 It is approvedin all guidelines for the treatment ofosteoporosis in postmenopausal women,in men, and in patients with glucocorti-coid-induced osteoporosis.5 Treatmentwith Fosamax has been shown to reducebone turnover and increase bone densi-ty, producing decreases in the incidenceof osteoporotic fractures, including thoseat the spine and hip, of approximately50%.6,7 Fosamax was initially availableas a 10-mg daily tablet and subsequentlydeveloped as a 70-mg once-weekly for-mulation. The daily and weekly doseswere determined to be therapeuticallyequivalent in a large randomized con-trolled clinical trial.8,9

Appropriate dosing is important toensure both efficacy and safety for bis-phosphonates. The product should betaken with a full glass of water and thepatient should remain upright at least 30minutes before the first food or bever-age of the day.10 Cases of esophagitishave been reported in clinical use, pri-marily associated with poor adherenceto the dosing instructions.11-13 This is ageneral characteristic of oral bisphos-phonates, as described in the productlabels for each bisphosphonate on themarket. The safety and tolerability pro-file of Fosamax is consistently similar toplacebo in each of several large clinicaltrials1-4,6,14 and to other approved N-BPsevaluated in head-to-head clinical trialsin which the only gastrointestinal exclu-sion criterion was conditions associatedwith delayed esophageal emptying,15

Epstein-vol5no2 6/1/05 8:20 PM 54


suggesting it is generally safe whentaken as directed. By contrast, non-Fosamax alendronate (NFA) prepara-tions (marketed in some countries) havenot been evaluated extensively, and onestudy reported substantial differences inthe disintegration/dissolution profiles ofthese preparations relative to innovatorFosamax.16 Those findings suggest thatimportant differences may exist betweenNFA preparations and Fosamax withregard to bioavailability and pharmaco-kinetics, and that the products may notnecessarily provide the expected similarsafety and efficacy for reasons describedpreviously.16,17

The purpose of the present studywas to extend the earlier studies bycomparing the upper gastrointestinalsafety profile of NFA tablets to that ofinnovator Fosamax in 2 animal modelsof local irritation, and to consider theclinical implications of these datatogether with previous findings.

METHODSFosamax and NFA were compared in 2studies: a rabbit skin model of local irri-tation, and a dog esophageal irritationmodel.

Rabbit Skin Model Of Local Irritation Skin thickening is a measure of the irri-tant/inflammatory response that can befollowed over time in living animals. Themodel described here has been usedpreviously as a surrogate model ofesophageal irritation.18 By measuringskin thickness, the dynamics of local irri-tation can be monitored over prolongedperiods without requiring euthanasia oflarge numbers of animals to obtain dataat multiple time points. At study termi-nation, the local irritant effects can beverified via postnecropsy analyses of theaffected tissue. This includes measure-ment of the wet weights of the inflamedtissue and histopathology analyses.

Animals, Dosing, and Skin-ThicknessMeasurements. Twenty-four approxi-mately 3.5-kg New Zealand White rab-bits (Charles River BreedingLaboratories, Saint Constant, Quebec,Canada) were randomized, 8 per group,to one of 3 treatment groups (vehicle,Fosamax, or NFA), balanced by weight.Animals were given ad libitum access towater and standard rabbit chow. A n i m a l swere shaved in the mid-scapular regionand a 3-cm circle drawn to define theinjection site. Baseline skin thicknessmeasurements were taken in the morn-ing using a digital caliper to lightly pincha fold of skin at the prepared injections i t e. In the afternoon, injection sites weredisinfected with 70% isopropanol andtopical anesthetic was applied.

For dosing suspensions, 22.5 mL ofsterile isotonic saline were stirred with15 tablets of either Fosamax or NFA(alendronate sodium; NOVO-Alendronate, Novopharm, Toronto,Canada) for more than 30 minutes priorto injecting to allow complete dissolu-tion of alendronate. The calculatedamounts of alendronate in 1.6 mL ofdosing suspensions were 10.61 mg forFosamax and 10.68 mg for NFA, sug-gesting 99.3% accuracy in administra-tion. For each dosing, 1.6 mL of vehicle,Fosamax, or NFA suspension was with-drawn from an actively stirring prepara-tion into a sterile, disposable 3-ccsyringe using an attached sterile 18-gauge needle, and this was immediatelyadministered at the center of themarked site. Blinded skin thicknessmeasurements were performed onMondays, Wednesdays, and Fridays,beginning on the second day after dos-ing and continuing for 40 days after thebaseline measurement.

At study termination all animalswere euthanized following administra-tion of a sedative (Ketamine/Xylazine,50/10 mg/kg IM) using intravenous sodi-



um pentobarbital (120 mg/kg). Injectionsites were dissected out within approxi-mately 5 minutes of euthanasia, andspecimens were immediately immersedin 10% neutral-buffered formalin forfixation. Data from one animal in theFosamax group that developed a bacter-ial infection at the injection site wereexcluded from all analyses.

Wet Weights and Histology. After 10days of fixation, injection site tissueswere rinsed with, and then transferredinto, 70% ethanol. Specimens were thentrimmed to include only the markedsites (vehicle) and/or the thickenedregions extending beyond the markedsite, and weighed in a blinded manner.

For histology, one transverse cut wasmade through the maximally thickened(or center) subregion of each injectionsite. Halved specimens were paraffin-embedded, and 5-µm sections were thenprepared for microscopy by standardmethods. One section from each samplewas stained with hematoxylin-eosin anda second was stained with MassonsTrichrome to permit characterization offibrotic tissues. Stained sections wereblindly evaluated by a veterinary pathol-ogist (S.L.H.) for necrosis, granulomaformation, cyst formation, mineraliza-tion, intralesional vacuolation,macrophage infiltration, and presence ofmultinucleated giant cells, polymor-phonuclear leukocytes (heterophils),lymphocytes, plasma cells, and foreignmaterial.

Statistical Analysis. Repeated measuresanalysis of variance (ANOVA) was per-formed on skin thickness data usingSAS software version 8.0.1. (SASCorporation, Cary, NC, USA). Otherstatistical analyses of skin thickness andwet weight were by Fisher’s ProtectedLeast Significant Difference (PLSD)and ANOVA using StatView for

Windows software, version 5.0.1. (SASCorporation, Cary NC, USA).

Dog Esophageal Irritation Study The dog model used in the present studyhas been described previously, havingbeen developed to examine the mecha-nism of bisphosphonate-inducedesophageal irritation as a function of pHand frequency of exposure.19 In brief,when dogs are anesthetized, theesophageal sphincter closes, enabling theplacement of tablets or solutions in thecaudal esophagus for fixed periods oftime to assess the effect on the esopha-gus. Animals were sacrificed after thefinal exposure and the esophagus wasevaluated for gross and histologicchanges.

Two similarly designed, comparativeexperiments were conducted: in the first,dogs were randomized to receive either10-mg NFA tablets (n = 4; alendronatesodium, Teva Pharmaceutical IndustriesLtd., Petah-Tikva, Israel) or placeboFosamax tablets (n = 3). In the secondstudy, dogs were randomized to eitherplacebo Fosamax tablets (n = 4) or 10-mg Fosamax tablets (n = 5).

Female beagle dogs (MarshallFarms, USA, Inc., North Rose, NewYork, USA), aged 51 to 54 weeks andweighing 5.7 kg to 7.1 kg, were fastedovernight prior to dosing and received350 g standard canine diet for approxi-mately 2 hours each day following dos-ing and recovery. Dogs were assigned totreatment groups based on a weight-bal-anced random allocation scheme andwere treated for 1 hour once daily for 5days.

Medication Administration. Generalanesthesia was induced with thiopentalsodium (2.5%) at a dose of approxi-mately 15 mg/kg to 25 mg/kg, an endo-tracheal tube was inserted, andanesthesia was maintained throughout



the procedure using isoflurane gas. AnOlympus gastrointestinal endoscope wasused to place the placebo or NFA prod-uct tablet in a similar location duringeach daily administration; the endoscope(shielded to prevent the tablet frombecoming moist) was inserted to a dis-tance of 25 cm to 30 cm down theesophagus of all dogs. The forceps wereadvanced to a distance of 4.0 cm fromthe endoscope tip, and the tablet wasdropped onto the esophageal mucosa. Ifnecessary, air was suctioned from theesophagus to ensure optimal contact ofthe tablet with the esophageal mucosa.The endoscope was then removed.

After 1 hour, the endoscope wasagain advanced into the esophagus untilthe remnants of the tablet were visual-ized, and 5 mL of sterile water wereflushed from the tubing to dislodge thetablet remnants from their original loca-tion in the esophagus. If necessary, airwas suctioned from the esophagus andthe scope was then removed. Isofluranewas discontinued, and the animals wereallowed to recover from anesthesia.

The animals were examined dailyfor mortality and drug-related clinicalsigns. Body weight measurements wereperformed twice during a 2-week accli-mation period and again on the day ofrandomization. During the treatmentperiod, animals were weighed on day 3and day 5. Food consumption was meas-ured daily during the 5-day pretreat-ment period and from day 1 to day 4.

Necropsy and Histomorphology. Threesamples (one from the cranial third, onefrom the middle third, and one from thecaudal third) of the esophagus, whichincluded any grossly observed changes,were fixed in 10% neutral buffered for-malin. Histology processing and histo-morphological evaluation of tissuesamples were conducted by MerckResearch Laboratories (West Point, PA,USA). Paraffin-embedded and hema-

toxylin- and eosin-stained sections of theesophagus from all sampled regionswere prepared by routine proceduresand examined using light microscopy.

RESULTSRabbit Skin Model of EsophagealIrritationSkin Thickness and Wet WeightMeasurements. Mean skin thickness wasnumerically greater in the NFA groupthan in the Fosamax group for all timepoints after baseline. Statistically signifi-cant differences (P < 0.05) betweenmean skin thicknesses in the Fosamaxversus the NFA groups were observed atdays 5, 8, 10, 22, and 36. Repeated meas-ures ANOVA revealed a statistically sig-nificant overall difference between thetwo groups (P = 0.034). In the repeatedmeasures analyses, comparison of salinevehicle versus NFA yielded a P-value of< 0.0001 and vehicle versus Fosamaxyielded a P-value of 0.017.

At the end of the study, the increasein mean wet weight of injection site skinspecimens dissected from animals at sac-rifice was 70% greater (P < 0.01) forNFA-treated animals relative to thosethat had received Fosamax (Figure 1).

Histopathology. Analysis of injectionsite tissue specimens indicated differ-ences in the character of tissue-level irri-tant responses elicited by suspensions ofFosamax versus those of the NFA for-mulation. Whereas Fosamax treatmentwas associated with a predominantlyforeign-body granuloma type response,NFA treatment produced a larger cysticnodule (Figure 2). In the Fosamax treat-ment group, nodules were smaller andwere characterized by a thick rim ofmultinucleated giant cells mixed withactivated macrophages and a centralaccumulation of foreign material (tabletsuspension material). In the NFA group,the lesions were larger and devoid ofsignificant inflammatory cell infiltrate,



which may be indicative of a postinflam-matory process, and tablet suspensionmaterial was encapsulated by fibrous tis-sue. Both treatments elicited adipocytenecrosis in basal and lateral regions ofnodules. Control animals exhibited nolesion formation.

The Fosamax diagnosis was subcutis,foreign body granuloma, multifocal tocoalescing, chronic, moderate to severe,with centralized foreign material andadjacent fat necrosis. The NFA diagnosiswas subcutis, fibrous cyst, focal, chronic,moderate to severe, with centralized for-eign material and adjacent fat necrosis.

Dog Esophageal Irritation Study There were no treatment-relatedchanges in body weight, food consump-tion, or behavior, and no dogs died.There was no treatment-related gross orhistologic evidence of tissue damage inthe cranial or mid-esophagus in anysample, confirming that damage was notcaused by the procedure.

At necropsy, all 4 dogs exposeddaily for 5 days to NFA tablets showed

multiple, raised, and discolored foci inthe caudal esophageal mucosa. Thesegross changes were present in the regionwhere the tablets had been placed andcorresponded histomorphologically toareas of marked ulcerative esophagitisin all four dogs (Table 1). The histomor-phologic changes in dogs exposed toNFA were characterized by necrosis ofstratified squamous epithelium liningthe lumen of the esophagus and partialto complete loss of mucosal lining(ulceration) (Figure 3). The inflammato-ry response associated with the ulcera-tion consisted of large numbers ofneutrophils within the necrotic and dis-rupted epithelial lining and within theunderlying submucosa. Marked submu-cosal edema was also present. Early re-epithelialization, consistent with an earlyreparative process, was observed nearthe margins of the esophageal injury. Incontrast, most of the changes observedin dogs exposed to Fosamax 10 mg dailyfor 5 days were more moderate thanthose observed in dogs exposed to 5days of 10-mg NFA (Figure 3); only 1 of

Figure 1. Mean wet weight of formalin-fixed rabbit skin injection site tissues. *P < 0.01 versusFosomax.



the 5 dogs treated with 10 mg Fosamaxfor 5 days had marked caudal esophaguschanges (Table 1).

DISCUSSIONEsophagitis is a potentially serious andimportant local adverse event associatedwith the use of bisphosphonates. Whilethe incidence of cases has been reducedwith proper dosing instructions, casesare still reported and the risk remains,especially in real-world use. One wouldexpect NFA and Fosamax to producesimilar degrees of irritation in the pre-

clinical models of esophageal irritationused here, because both formulationscontained similar amounts of alen-dronate. However, the present findingsdemonstrate a greater irritant responseto NFA compared with Fosamax in rab-bit and dog models used in this study.Subcutaneous injection of rabbits withNFA suspension elicited significantlygreater irritation than was produced byan equivalent dose of Fosamax, as meas-ured by skin thickness and wet weight.Similarly, esophageal exposure to daily10-mg NFA tablets for 5 days caused

Figure 2. Representative photomicrographsof rabbit skin injection sites 40 days after asingle subcutaneous dose of vehicle (toppanel), Fosamax (middle panel), or non-Fosamax alendronate (NFA) (bottom panel).F indicates foreign material; M, macrophageand giant cell infiltrate; C, margins of fibrouscyst.

Figure 3. Photomicrographs of the caudaldog esophagus at sites exposed to placebo(top panel), Fosamax (middle panel), or non-Fosamax alendronate (NFA) (bottom panel)for 1 hour per day for 5 days; these imagesare representative of normal, less severe, andmore severe esophageal tissue changes,respectively.



greater irritation in dogs than that pro-duced by an equivalent dose and dura-tion of Fosamax tablets. These resultsprovide evidence for potential increasedesophageal irritation with NFA in clini-cal practice, and resulting increase innumber of events or severity (eg, moreperforations).

Localized Skin Responses in RabbitsSkin irritation elicited by N-BPs in rab-bits has been modeled as a surrogate forgastrointestinal tract irritation followingoral dosing in humans, since these tis-sues are both known to be affected bylocal exposure to bisphosphonates.20

Although differences exist in the cellularprocesses underlying esophageal versussubcutaneous inflammatory responses,irritant effects of subcutaneous exposureto N-BPs may index analogous reactionsin the esophageal lining. Recent studiessuggest that skin and upper gastroin-testinal tract irritation produced by N-BPs is mechanistically linked to thebone-sparing effects of these com-pounds, via inhibition of the same cho-lesterol synthesis metabolic pathway.21

Thus, one would expect similar degreesof tissue irritation using equivalentamounts of the active compound, alen-

dronate. However, the magnitude of skinthickening and wet tissue weightincrease elicited by NFA was significant-ly greater than that produced byFosamax. The larger differences in wetweights compared to skin thickness sug-gest that the measurements of skinthickness underestimated the differencein irritation response between the 2groups. Furthermore, histopathologyanalysis revealed qualitative differencesin tissue-level inflammatory responsesbetween NFA and Fosamax. Theseresults suggest that aspects of tablet for-mulation other than those related to theactive component (alendronate) mayincrease the tissue irritation potential ofNFA tablets. While not definitive, thegreater irritant effects at subcutaneousinjection sites in rabbits suggest thatNFA may have greater potential tocause esophageal irritation in humans,compared to Fosamax.

Esophageal Irritation in DogsThe dog model used in the present studywas developed to assess the mechanismof esophageal irritation with alen-dronate.19 Prior studies using this modelhave demonstrated that repeated, dailyexposure to an alendronate solution at

Table 1. Caudal Esophagus Histologic Findings Following 5 Days of Repeated EsophagealExposure*

Placebo Non- Placebo Fosamax(n = 3) Fosamax (n = 4) (n = 5)

Histologic findings Alendronate(n = 4)

No irritation 3 0 4 0

Erosive esophagitisSlight 0 0 0 1Moderate 0 0 0 2

Ulcerative esophagitisModerate 0 0 0 1Marked 0 4 0 1

*n values represent number of dogs.



pH 2.0 (similar to the pH of gastricjuice) resulted in esophagitis, butesophagitis was not observed with simi-lar dosing at higher pH (3 and 5).19

The present findings demonstratedmarked ulcerative esophagitis in all dogsexposed to NFA, whereas more moder-ate changes were observed in most dogsexposed to Fosamax. As with the resultsobserved in the rabbit model ofesophageal irritation, these findings raiseconcern about the potential foresophagitis with clinical use of NFApreparations. Moreover, this potentialeffect would likely be more severe ifreflux of acidic gastric contents into thecaudal esophagus were to occur duringdosing in the clinical setting.

Clinical ImplicationsIn addition to the present findingsdemonstrating that NFA is more irritat-ing in preclinical models of esophagealirritation, other factors may influencethe clinical efficacy and safety profilesand should be considered.

Variability in Disintegration Profiles.Prior research has shown notable differ-ences in the physical properties of someNFA preparations compared withFosamax that may impact both safetyand efficacy.16 Some NFA tablets disinte-grated 2- to 10-fold faster than Fosamax,whereas other NFA tablets disintegratedat least 5-fold slower (Figure 4).16

Further evaluation of one slow-disinte-grating NFA product demonstrated thatthe release of alendronate into solutionlagged behind that from Fosamax.16

Delayed disintegration couldincrease the chances of irritation if thetablet or fragments of the tablet becomestuck or reflux into the esophagus. Forinstance, a delayed-release 30-mg rise-dronate tablet for the treatment ofPaget’s disease was abandoned in favorof an immediate-release formulationdue to reports of esophagitis.22 On the

other hand, if tablets disintegrate tooquickly, particles may be retained in themouth or esophagus, leading to irrita-tion. An even higher incidence of pillesophagitis would be expected inpatients with motility disorders thatdelay esophageal transit or patients withesophageal reflux disease who do notadhere to dosing instructions.Furthermore, a difference in disintegra-tion times for NFA preparations relativeto Fosamax could impair absorption andreduce efficacy.16 Disintegration toosoon could leave alendronate in themouth or esophagus where it may notbe absorbed. Conversely, slow disinte-gration would increase the likelihoodthat the tablet would come into contactwith ingested food or liquids before ade-quate absorption can occur.

Other Factors. Other factors (includingtablet size, density, shape and coating,granulation and crystallization charac-teristics, and type of excipients) mayinfluence the esophageal safety andbioavailability of NFA preparationsindependent of their effects on disinte-gration time. Some of these formulationfactors may be responsible for the dif-ferences in irritation observed in thecurrent study. There are notable differ-ences in excipients between Fosamaxand various NFA preparations, whichmay alter disintegration profiles andother aspects including safety and effica-cy. The tendency of products to adhereto the esophageal mucosa can be modi-fied by shape and formulation, includingcoating material.23 For instance, gel cap-sules have longer esophageal transittimes relative to tablets due to a greatertendency for adherence.23-25 Severeupper gastrointestinal adverse eventswere reported in a previous trial ofpamidronate gel capsules;26 thus, recentbisphosphonate development programshave avoided the use of gel capsules.

In addition to formulation, manufac-



Figure 4. Disintegration times in water at 37°C of different preparations of non- Fosamax alen-dronate (NFA) that disintegrated faster (top panel) or slower (bottom panel) than Fosamax. Errorbars represent the range of individual values measured.



turing processes can affect granulationand crystallization characteristics;27 anyof these factors may influence the extentof drug absorption and thus might influ-ence the acceptable interval betweenadministration and subsequent foodintake. This is very important with bis-phosphonates since absorption dependson pH level and local salts in the gutand is almost completely abolished inthe presence of food.28 The manufactur-er of Fosamax does not sell alendronatefor further processing or distribution byother entities and, to our knowledge,there is no publicly available informa-tion regarding the manufacturingsource(s), processes, or amounts of alen-dronate in most NFA preparations.16,17

Indeed, it was previously reported thatat least one NFA preparation did notcontain the amount of alendronate stat-ed on the product label.29

Bioequivalence: What ConstitutesProof?In some countries, NFA preparationshave sought, and in some instancesobtained, approval for marketing basedexclusively on data from small, single-dose bioavailability studies relative toFosamax. However, such studies havemajor limitations and may not accurate-ly indicate the clinical safety and efficacyof NFA preparations. For example, somehave measured bioavailability using longfasting times of up to 5 hours after dos-ing with NFA. Since food and beveragesstrongly interfere with absorption ofalendronate,28 long fasting times couldincrease the amount of alendronateabsorbed, and comparisons to previousstudies of Fosamax that used shorterfasting times could be misleading. Also,unlike many other drugs, the pharmaco-logical action and relevant exposurelevel of bisphosphonates is not deter-mined by the levels circulating in theblood, but instead by the cumulativeamount deposited on the bone surface,

which will increase with repeated dosingand cannot be directly measured inhumans. In addition, bioavailability oforal bisphosphonates is highly variable,and small studies could easily fail todetect clinically important differences inbioavailability of 20% or more, due toType II statistical errors. Moreover, thepotential for esophagitis with oral bis-phosphonates involves processes thatoccur prior to absorption and is relatedto the frequency of dosing and other fac-tors. Thus, due to these unique charac-teristics, even small differences in tabletdisintegration rates, types of excipients,and formulation characteristics betweena test and reference bisphosphonatemight result in important differences insafety and efficacy that would not bedetected in small, single-dose bioavail-ability studies.

An example of the potential risks ofapproval and use of such drugs in theabsence of direct safety information wasobserved in 2001, when some patientswere switched from Fosamax to an NFApreparation (which had been approvedon the claim of similar bioavailability)and many began to experience signifi-cant gastrointestinal adverse experi-ences, including 1 case of esophagealrupture.

Thus, given the unusual characteris-tics of bisphosphonates and the influ-ence of seemingly minor differences informulation, it appears that simplebioavailability studies may not be ade-quate for predicting the clinical efficacyand safety profiles. Given the problemswith using bioavailability data alone,large clinical trials of each new formula-tion may be required to assess its safetybefore NFA products are prescribed.

CONCLUSIONIn the present study, NFA was more irri-tating than Fosamax in two preclinicalmodels of esophageal irritation. Thesedata, together with previous findings



demonstrating differences in the disinte-gration/dissolution profiles of NFApreparations relative to Fosamax, sug-gest that important differences may existbetween Fosamax and NFA prepara-tions regarding clinical safety and effica-cy. Additional studies, including largeclinical trials, may be necessary to ade-quately evaluate the clinical efficacy andsafety of NFA products.

ACKNOWLEDGMENTSThe authors thank Christine McCrarySisk for assistance with preparation ofthis manuscript.

REFERENCES1. Liberman UA, Weiss SR, Broll J, et al. Effect

of oral alendronate on bone mineral densityand the incidence of fractures in post-menopausal osteoporosis. The AlendronatePhase III Osteoporosis Treatment StudyGroup. N Engl J Med. 1995;333:1437-1443.

2. Black DM, Cummings SR, Karpf DB, et al.Randomised trial of effect of alendronate onrisk of fracture in women with existing verte-bral fractures. Fracture Intervention TrialResearch Group. Lancet. 1996;348:1535-1541.

3. Bone HG, Hosking D, Devogelaer J-P, et al.Ten years’ experience with alendronate forosteoporosis in postmenopausal women. NEngl J Med. 2004;350:1189-1199.

4. Hosking DJ, Chilvers CED, Christiansen C, etal, for the Early Postmenopausal InterventionCohort Study Group. Prevention of bone losswith alendronate in postmenopausal womenunder 60 years of age. New Engl J Med.1998;338:485-492.

5. Geusens P. Review of guidelines for testingand treatment of osteoporosis. CurrOsteoporos Rep. 2003;1:59-65.

6. Cranney A, Guyatt G, Griffith L, Wells G,Tugwell P, Rosen C. Summary of meta-analy-sis of therapies for postmenopausal osteo-porosis. Endocrine Reviews. 2002;23:570-578.

7. Wehren LE, Hosking D, Hochberg MC.Putting evidence-based medicine into clinicalpractice: comparing anti-resorptive agents forthe treatment of osteoporosis. Curr Med ResOpin. 2004;20:525-531.

8. Schnitzer T, Bone HG, Crepaldi G, et al.Therapeutic equivalence of alendronate 70mg once-weekly and alendronate 10 mg dailyin the treatment of osteoporosis. Aging.2000;12:1-12.

9. Rizzoli R, Greenspan SL, Bone G 3rd, et al,for the Alendronate Once-Weekly StudyGroup. Two-year results of once-weeklyadministration of alendronate 70 mg for thetreatment of postmenopausal osteoporosis. JBone Miner Res. 2002;17:1988-1996.

10. Fosamax Prescribing Information. Physicians’Desk Reference. 55th ed. Montvale, NJ:Medical Economics; 2001:1930-1936.

11. Cryer R, Bauer DC. Oral bisphosphonatesand upper gastrointestinal tract problems:what is the evidence? Mayo Clin Proc.2002;77:1031-1043.

12. Watts N, Freedholm D, Daifotis A. The clini-cal tolerability profile of alendronate. Int JClin Pract. 1999;101(suppl):51-61.

13. de Groen PC, Lubbe DF, Hirsch LJ, et al.Esophagitis associated with the use of alen-dronate. N Engl J Med. 1996;335:1016-1021.

14. Greenspan S, Field-Munves E, Tonino R, etal. Tolerability of once-weekly alendronate inpatients with osteoporosis: a randomized,double-blind, placebo-controlled study. MayoClin Proc. 2002;77:1044-1052.

15. Rosen CJ, Hochberg M, Bonnick S, et al.Treatment with once-weekly alendronate 70mg compared to once-weekly risedronate 35mg in women with postmenopausal osteo-porosis: a randomized, double-blind study. JBone Miner Res. 2005;20:141-151.

16. Epstein S, Cryer B, Ragi S, et al.Disintegration/dissolution profiles of copiesof Fosamax (alendronate). Curr Med ResOpin. 2003;19:781-789.

17. Epstein S, Ragi S. Comment on disintegra-tion/dissolution profiles of copies of Fosamax(alendronate) [letter]. Curr Med Res Opin.2004;20:1035-1037.

18. Reszka AA, Fisher JE, Rodan GA.Mechanism-based local irritation by nitrogen-bisphosphonates [abstract SC22 T]. Bone.2001;28;5 (suppl):S95.

19. Peter CP, Handt LK, Smith SM. Esophagealirritation due to alendronate sodium tablets:possible mechanisms. Dig Dis Sci.1998;43:1998-2002.

20. Fisher JE, Rodan GA, Reszka AA. Evidencefor mechanism-based irritant effects of nitro-gen-bisphosphonates on soft tissues. J BoneMiner Res. 2001;16:S218.

21. Reszka A, Halasy-Nagy JM, Rodan GA.Nitrogen-bisphosphonates block retinoblas-toma phosphorylation and cell growth byinhibiting the cholesterol biosynthetic path-way in a keratinocyte model esophageal irri-tation. Mol Pharmacol. 2001;59(2):193-202.



22. Siris ES, Chines AA, Altman RD, et al.Risedronate in the treatment of Paget’s dis-ease of bone: an open label, multicenterstudy. J Bone Miner Res. 1998;13:1032-1037.

23. Marvola M, Vahervuo K, Sothmann A,Marttila E, Rajaniemi M. Development of amethod for study of the tendency of drugproducts to adhere to the esophagus. J PharmSci. 1982; 71:975-977.

24. Channer KS, Virjee JP. The effect of formula-tion on oesophageal transit. J PharmPharmacol. 1985;37:126-129.

25. Perkins AC, Wilson CG, Blackshaw PE, et al.Impaired oesophageal transit of capsule ver-sus tablet formulations in the elderly. Gut.1994;35:1363-1367.

26. Lufkin EG, Argueta R, Whitaker MD, et al.Pamidronate: an unrecognized problem ingastrointestinal tolerability. Osteoporos Int.1994;4:320-2.

27. Videau J-Y. Making medicines safe. BullWorld Health Org. 2001;79:87.

28. Gertz BJ, Holland SD, Kline WF, et al.Studies of the oral bioavailability of alen-dronate. Clin Pharmacol Ther. 1995;58:288-298.

29. Rägo L. Ensuring Access To Drug ProductsThat Are of Acceptable Quality. World HealthOrganization Pilot Procurement Quality andSourcing (“pre-qualification”) project. [WorldHealth Organization Web site]. Available at:http://www.who.int/medicines/organization/par/briefing/9prequalification.ppt. AccessedSeptember 4, 2003.


Disintegration and Esophagealmail.jrnlappliedresearch.com/articles/Vol5Iss2/Epstein.pdf ·...

Documents

Transcript of Disintegration and Esophagealmail.jrnlappliedresearch.com/articles/Vol5Iss2/Epstein.pdf ·...