In vivo absorption behaviour of theophylline from starch-methyl methacrylate matrix tablets in beagle dogs

IN VIVO ABSORPTION BEHAVIOUR OF THEOPHYLLINE FROM STARCH- 1 METHYL METHACRYLATE MATRIX TABLETS IN BEAGLE DOGS 2 3 2F. Fernández-Campos, 1C. Ferrero*, 2H. Colom, 1M.R. Jiménez-Castellanos 4 1Department of Pharmacy and Pharmaceutical Technology, Faculty of Pharmacy, 5 University of Seville, Sevilla, Spain. 6 2Department of Pharmacy and Pharmaceutical Technology, Faculty of Pharmacy, 7 University of Barcelona, Barcelona, Spain. 8 9 10 11 12 13 14 15 16 17 18 19 *Corresponding author: 20 Carmen Ferrero 21 Department of Pharmacy and Pharmaceutical Technology, Faculty of Pharmacy, University 22 of Seville. C/ Prof. García González nº 2, 41002 Sevilla, Spain. 23 Tel.: + 34 954 557218. Fax: + 34 954 556085. E-mail: [email protected] 24 Abstract: 25 This study evaluates in vivo the drug absorption profiles from potato starch-methyl 26 methacrylate matrices * using theophylline as a model drug. Healthy beagle dogs under 27 fasting conditions were used for in vivo studies and plasma samples were analyzed by a 28 fluorescence polarization immunoassay analysis (FPIA method). Non-compartmental and 29 compartmental (population approach) analysis was performed to determine the 30 pharmacokinetic parameters. The principle of superposition was applied to predict multiple 31 dose plasma concentrations from experimental single dose data. An in vitro-in vivo 32 correlation (IVIVC) was also assessed. The sustained absorption kinetics of theophylline 33 from these formulations was demonstrated by comparison with two commercially available 34 oral sustained-release theophylline products (Theo-Dur and Theolair). A one- 35 compartment model with first order kinetics without lag-time best describes the 36 absorption/disposition of theophylline from the formulations. Results revealed a 37 theophylline absorption rate in the order FD-HSMMA ≥ Theo-Dur ≥ OD-CSMMA > 38 Theolair ≥ FD-CSMMA. On the basis of simulated plasma theophylline levels, a twice 39 daily dosage (every 12h) with the FD-CSMMA tablets should be recommended. A Level C 40 IVIVC was found between the in vitro t50% and the in vivo AUC/D, although further 41 optimization of the in vitro dissolution test would be needed to adequately correlate with in 42 vivo data. 43 Key words: Potato starch-methyl methacrylate copolymers; Anhydrous theophylline; 44 Sustained-release matrix tablet; Beagle dog; Pharmacokinetics; IVIVC. 45 * FD-HSMMA: freeze-dried hydroxypropylstarch methyl methacrylate; OD-CSMMA: oven-dried carboxymethylstarch methyl methacrylate; FD-CSMMA: freeze-dried carboxymethylstarch methyl methacrylate. 1. Introduction 46 Theophylline is a methylxanthine derivative widely used for its bronchodilatory, 47 inotropic, central stimulant and diuretic effects both in humans and animals (Mengozzi et 48 al., 1998). A very close relationship has been reported between plasma drug concentrations 49 and its efficacy/safety. The main limitation to therapeutic effectiveness is its low 50 therapeutic index. Therapeutic plasma concentrations are ranging from 5 to 15 µg/mL and 51 plasma concentrations greater than 20 µg/mL may result in adverse effects (Muskó et al., 52 2001). After oral administration as a solution, theophylline is rapidly and completely 53 absorbed. A single dose of 5 mg/kg in adults provides a mean peak serum concentration of 54 ≈ 10 mcg/mL (range 5-15 mcg/mL) at 1-2 hr after the dose (FDA, 2012). This rapid 55 absorption leads to frequent administration to maintain therapeutic drug levels. Sustained- 56 release formulations are desired to maintain plasma concentrations within the therapeutic 57 range during a more lasting period of time, avoiding adverse effects and leading to improve 58 efficacy and to better patient compliance. 59 The most common method of modulating the drug release is to develop polymeric 60 matrix tablets. In this type of systems, judicious selection of release retarding excipients is 61 necessary. Over the last years, a new generation of grafted copolymers combining potato 62 starch derivatives (hydroxypropylstarch -HS- or carboxymethylstarch -CS-) with an acrylic 63 monomer (methyl methacrylate -MMA-) were introduced as matrix-forming excipients for 64 oral sustained-release dosage forms. As described elsewhere (Castellano et al., 1997), these 65 copolymers were synthesised by free radical polymerisation of the monomer (MMA) on the 66 starches using Ce (IV) as an initiator. The products obtained (HSMMA, CSMMA) were 67 either dried in a vacuum oven at 6.67-13.33 hPa and 50 ºC until constant weight (OD 68 copolymers) or freeze-dried (freezing process at -20 ºC for 24h and sublimation process at 69 0.13 hPa and -50ºC until a powdered product was obtained) (FD copolymers). These 70 materials were thoroughly characterized in terms of physico-chemical and technological 71 properties (Bravo-Osuna et al., 2005; Ferrero et al., 1999; Ferrero and Jiménez-Castellanos, 72 2002) and demonstrated their ability to form inert matrix tablets that control drug release by 73 a diffusion mechanism (Ferrero et al., 2003). 74 Several studies have been reported focused on the influence of excipients and 75 technology on bioavailability of sustained release theophylline formulations (Ikegami et al., 76 2006; Miyazaki et al., 2000, 2001; Roshdy et al., 2002; Yu et al., 1996). It is also known 77 that for oral sustained-release dosage forms the release rate is the limiting factor in the 78 absorption process. Therefore it is desirable to use the in vitro data to predict in vivo 79 pharmacokinetic parameters for a rational development and evaluation process of these 80 sustained-release dosage forms. Thus over the past decade, interest has increased on in 81 vitro-in vivo correlations (IVIVC) (FDA, 1997). In this way, as theophylline is a Class I 82 drug according to the Biopharmaceutical Classification System (BCS) (Kimberley et al., 83 2002; Lindenberg et al., 2004) due to its solubility and permeability characteristics, an 84 IVIVC can be expected for slow release formulations of this drug (Roshdy et al., 2002; Yu 85 et al., 1996). However, we must be cautious due to the many physiological factors affecting 86 oral absorption. 87 For the above reasons, a previous study (Ferrero and Jiménez-Castellanos, 2014) 88 assessed the influence of the tablet crushing force, the pH of the dissolution medium and 89 the agitation rate on the in vitro theophylline release kinetics from starch-methyl 90 methacrylate matrix tablets. The results were compared with those of two commercial 91 formulations of theophylline (Theo-Dur® and Theolair®). FD-HSMMA, OD-CSMMA and 92 FD-CSMMA matrix tablets were selected on the basis of their mechanical resistance and 93 their similar release profiles with the marketed products. The aim of the present work was 94 then to investigate if the matrix tablets selected from in vitro release studies confer 95 adequate oral absorption and pharmacokinetic properties, as drug sustained delivery 96 systems. Female beagle dogs were used as an animal model for evaluating theophylline 97 absorption (Cook et al., 1990). Attention has also been focused on the possibility of 98 predicting expected in vivo bioavailability characteristics from dissolution profiles 99 (IVIVC). 100 101 2. Materials and methods 102 103 2.1. Materials 104 Aqueous solution (10 ml) of aminophylline (Eufilina Venosa®, BYK Elmu, Madrid, 105 Spain) corresponding to 175.7 mg of anhydrous theophylline was used as intravenous 106 administration. Five theophylline formulations were selected for oral administration based 107 on the similarity of the in vitro release profiles (Ferrero and Jiménez-Castellanos, 2014): 108 • In vitro pH-independent release: FD-HSMMA matrix tablets (100 mg theophylline) 109 and Theo-Dur® 100 mg (Pharmacia & Upjohn S.A., Barcelona, Spain) as reference 110 product. 111 • In vitro pH-dependent release: OD-CSMMA and FD-CSMMA matrix tablets (175 112 mg theophylline) and Theolair® 175 mg (3M España S.A, Madrid, Spain) as 113 reference product. 114 According to manufacturer’s information and literature review (Munday and 115 Fassihi, 1995; Shangraw, 1988), Theo-Dur® is composed of theophylline sugar pellets 116 coated with lipid materials and cellulose acetate phthalate (CAP) and embedded into a 117 slowly disintegrating waxy type matrix containing additional drug. In contrast, Theolair® is 118 formulated in the form of theophylline tablets containing lactose as soluble excipient and 119 coated with cellulose acetate phthalate (Crombeen and De Blaey, 1983; Shangraw, 1988). 120 The method of preparation of the copolymers tablets is well-described in Ferrero 121 and Jiménez-Castellanos (2014). Briefly, mixtures (500 mg) of copolymer, anhydrous 122 theophylline (as model drug) and stearic acid (as lubricant) were directly compressed 123 (single punch tablet machine Bonals AMT 300, Barcelona, Spain) to obtain flat-faced 124 compacts (12 mm diameter) at a crushing force of 90-100 N. 125 Sodium heparin 5000 UI/ml (Rovi, Madrid, Spain), monoclonal II theophylline 126 (Abbott, Madrid, Spain), calibrators to TDxFLx (Abbott, Madrid, Spain) at 0.0, 2.5, 5.0, 127 10.0, 20.0 and 40.0 g/mL of theophylline, and controls to calibration verification of 128 TDxFLx at 7.0, 12.0 and 26.0 g/mL of theophylline were used as reagents. 129 130 2.2. Animals 131 Six healthy female beagle dogs, weighing 8.5-13 kg were housed individually in 132 controlled conditions (temperature, humidity and light-dark cycles). The dogs did not 133 receive food but had free access to water for 12 h before and after drug administration. The 134 animal experimentation was approved by the Ethical Committee of Animal 135 Experimentation of the Faculty of Veterinary Medicine from Cordoba University (Spain). 136 137 2.3. Experimental design 138 2.3.1. In vivo theophylline absorption 139 Each animal received the three tested sustained-release formulations (FD-HSMMA 140 matrix tablets with 100 mg of theophylline, OD-CSMMA and FD-CSMMA matrix tablets 141 with 175 mg of theophylline), besides Theo-Dur, Theolair and an intravenous solution of 142 theophylline (5 ml of aminophylline solution equivalent to 87.85 mg of anhydrous 143 theophylline) in a Williams’s cross-over design (Jones and Kenward, 1989). A wash-out 144 period of two weeks was allowed between the different treatments. 145 Serial blood samples (5 mL) were collected from the cephalic vein at predetermined 146 time points up to 12h for the intravenous solution and 24h for the oral formulations. All 147 blood samples were taken in heparinized tubes (BD vacutainers® LH 143 IU, New Jersey, 148 USA), and plasma was separated by centrifugation (JP Selecta Cemcom, Abrea, Barcelona, 149 Spain) at 3000 rpm and immediately frozen at -20ºC (Revco, ULT 1786-3-v30, North 150 Carolina, USA) until analysis. 151 152 2.3.2. Theophylline analytical method 153 Theophylline plasma concentrations were determined by a validated fluorescence 154 polarization immunoassay analysis (FPIA) using the TDx/TDxFLx® method (AbbottTM 155 Laboratories, Madrid, Spain) (Jolley et al., 1981). The assay was linear over plasma 156 concentrations ranging from 0.5 to 30 g/mL. The intraday and interday coefficients of 157 variation ranged from 0.48% to 6.42% at the three concentrations tested (7.0, 12.0, 26.0 158 µg/mL). The lower limit of quantification (LLOQ) was established at 0.80  0.07 µg/mL 159 and its variation coefficient was 8.75%. The mean absolute recovery of theophylline was 160 100.6  2.96%. The values obtained are within the limits accepted by FDA (2001) for 161 bioanalytical methods. 162 163 2.4. Data Analysis 164 165 2.4.1. Pharmacokinetic analysis 166 It is known that food could induce absorption changes (“food effect”) from 167 controlled-release formulations of theophylline (Cook et al., 1990; Karim, 1986; Shiu et al., 168 1989). So, plasma levels of theophylline in fasting dogs were plotted against time, and 169 pharmacokinetic parameters were calculated by a non-compartmental method using 170 WinNonlin 5.3 (Pharsight Corporation, Mountain View, California). The area under the 171 plasma concentration vs time curve up to the last sampling time point, AUC0-t, was obtained 172 using the linear and log-linear trapezoidal method. The AUC0-t was extrapolated to infinity 173 (AUC0-∞) by adding the quotient Ct/Kel, where Ct represents the last measured concentration 174 and Kel represents the apparent terminal rate constant. Kel was calculated by the linear 175 regression of the log-transformed concentrations of the drug in the terminal phase. The 176 half-life of the terminal elimination phase was obtained using the relationship t1/2 = 177 0.693/Kel. The maximum value of the plasma concentration (Cmax) and the time of 178 maximum concentration (Tmax) were obtained directly from the data. Mean residence time 179 (MRT) was determined by division of AUMC (area under the first moment curve) by AUC0- 180  . Absolute oral bioavailability (F) was calculated from plasma data using the relationship 181 𝐹 = (𝑑𝑜𝑠𝑒𝐼𝑉×𝐴𝑈𝐶0−∞ 𝑜𝑟𝑎𝑙 𝑑𝑜𝑠𝑒𝑜𝑟𝑎𝑙×𝐴𝑈𝐶0−∞ 𝐼𝑉)×100. 182 A compartmental pharmacokinetic analysis by means of the population approach 183 was also performed. Data from both the intravenous administration and the five oral 184 formulations were simultaneously modeled using NONMEM® 7.2 (Globomax, Rockville, 185 MD) (Bauer, 2011). Graphical diagnostics were assessed using Xpose version 4.2.1 186 (Jonsson and Karlsson, 1999) implemented into R version 2.14.2 and Perl speaks- 187 NONMEM (PsN) version 3.2.4 Tool-kit (Lindbom et al., 2005). The first order conditional 188 estimation method (FOCE) with interaction was used. One and two compartment models 189 with linear elimination were tested in all the cases. First-order kinetics without or with lag- 190 time were tested to describe the absorption profile. The models were parameterized in terms 191 of absorption rate constant (Ka), apparent volume of distribution (Vd), elimination clearance 192 (Cl) and bioavailability (F). Between-animal variability (BAV) evaluated for each 193 pharmacokinetic parameter was modeled exponentially, assuming a log-normal 194 distribution. Additive, proportional and combined (additive + proportional) models were 195 compared to assess the residual error (RE). To statistically distinguish between nested 196 models, the difference in the minimum value of the objective function (MOFV) was used 197 because this difference is approximately 2 distributed. A significance level of p < 0.005 198 that corresponded to a difference in MOFV of 7.879 for 1 degree of freedom was 199 considered. For non-hierarchical models, the most parsimonious model with the lowest 200 objective function according to the Akaike’s Information Criterion (AIC) was chosen 201 (Yamaoka et al., 1978). Once the base model was developed, the effect of the type of 202 formulation on absorption pharmacokinetic parameters (Ka and F) was investigated with 203 NONMEM®. This covariate was tested firstly univariately on each parameter and then by 204 the forward inclusion/backward elimination procedures. Significance levels of 5% 205 (MOFV=-3.841 units) and 0.1% (MOFV=10.8 units) were considered during the 206 forward addition and backward elimination steps. The decrease in MOFV (-2xlog 207 likelihood), parameter precision expressed as relative standard error (RSE%), reductions in 208 BAV associated to parameters, model completion status and visual inspection of goodness- 209 of-fit plots were also considered for model selection. 210 From the final model, simulations were performed based on the final 211 pharmacokinetic estimates using 1000 individuals for each formulation to calculate the 212 CSMMA matrices, a two times daily regimen would provide more fluctuating theophylline 354 concentrations with mean peak concentrations at steady-state higher than 15 but lower than 355 20 g/mL, although toxicity signs seem to appear over 20 g/mL. Hence, we can conclude 356 that, even with a twice daily dosage regimen, FD-CSMMA matrices have acceptable 357 sustained release characteristics, similar to the commercial Theolair tablets (Conard et al., 358 1982). 359 Although the trends in the rate and/or extent of theophylline absorption from the 360 formulations in dogs are expected to be maintained in humans (Cook et al., 1990), an 361 interspecies scaling approach (Gascón et al., 1994) would be needed to predict the best dose 362 regimen in humans from the results obtained in this preclinical research. 363 364 3.2. In vitro-in vivo correlations 365 The in vitro-in vivo correlations for the different formulations were explored by 366 comparing in vivo drug release obtained from deconvolution with the in vitro release data. 367 The in vitro dissolution parameters estimated from the different formulations (Table 368 4) are compared according to their pH-independent (FD-HSMMA vs Theo-Dur) or pH- 369 dependent (OD-CSMMA and FD-CSMMA vs Theolair) character. Theo-Dur show lower 370 percentages dissolved at 4 hours (Q4h) values and higher mean dissolution times (MDT) 371 and t50% values than FD-HSMMA formulation. These results are in agreement with the 372 faster drug release rates reported for FD-HSMMA matrices compared with TheoDur® 373 (Ferrero and Jiménez-Castellanos, 2014). This behaviour was attributed to the different 374 formulation of these systems, as the major part of theophylline in TheoDur® is contained in 375 pellets embedded in the matrix. Although the tendency in Tmax, AUC and F values (Table 1) 376 would indicate also a faster in vivo release/absorption for FD-HSMMA tablets, it could not 377 be confirmed by the population approach as similar Ka values were found for both 378 formulations (Table 3). 379 Concerning pH-dependent formulations, the in vitro results (Table 4) show also 380 lower Q4h values and higher MDT and t50% values for Theolair compared with OD- 381 CSMMA formulation. In contrast, FD-CSMMA formulation shows closer values to 382 Theolair, in agreement with the tendency reported for the drug release rates (Ferrero and 383 Jiménez-Castellanos, 2014). These in vitro results are consistent with the trend observed in 384 the in vivo parameters Tmax, AUC and F (Table 2) and with the highest absorption (Ka and F 385 values) described for OD-CSMMA by the compartmental population approach (Table 3). 386 For the tested matrices, the order in the absorption rate FD-HSMMA > OD- 387 CSMMA > FD-CSMMA is consistent with the tendency reported for the drug release rates 388 (Ferrero and Jiménez-Castellanos, 2014) and could be explained by formulation factors 389 such as polymer nature and tablet porous network. The strongly retarded drug 390 release/absorption of CSMMA and FD matrices could be attributed to the better binding 391 properties of these derivatives. Moreover, OD matrices were characterized by less tortuous 392 pore networks than their homologous freeze-dried, which explains the faster drug 393 release/absorption from those matrices. Hence, the absorption kinetics seems to be 394 determined by the same variables affecting the drug release kinetics. 395 As mentioned in the introduction, theophylline belongs to Class I of BCS, being a 396 good candidate to develop a level A IVIVC when more than two extended release 397 formulations are involved. Such type of IVIVC is generally linear (FDA, 1997) and has 398 been previously reported for theophylline extended release formulations (Ochoa et al., 399 2010). However, in the present study, the percentage of drug absorbed in vivo from 400 TheoDur and Theolair can not be predicted point-to-point from the percentage of in vitro 401 drug released. A more acceptable linear fitting (r2 = 0.93-0.95) is obtained in the case of the 402 three tested matrices (Figure 5). This could be due to: a) the different formulation and 403 method of manufacture of the standard and test systems (Ferrero and Jiménez-Castellanos, 404 2014; Kortejarvi et al., 2002; Nabais et al., 2007); b) the pH-dependent or -independent 405 character of the formulations (Cutler et al., 1997; Ferrero and Jiménez-Castellanos, 2014; 406 Ochoa et al., 2010). The certain degree of curvature exhibited by the profiles is likely a 407 direct result of the difference in release kinetics between in vitro and in vivo. 408 Although we also failed to obtain an acceptable level B correlation for all 409 formulations, a significant level C IVIV correlation (r2 = 0.9411, p < 0.05) can be found 410 between t50% (in vitro) and AUC/D (in vivo) (Figure 6A). The higher deviation of linearity 411 of Theolair® could be a consequence of its pH-dependent release profile (Ferrero and 412 Jiménez-Castellanos, 2014). The fit is better (r2 = 0.9958, p < 0.05) when only the tested 413 matrices are compared (Figure 6B), confirming the importance of the influence of the 414 formulation design. As expected, an increase in the in vitro variable (time for 50% release) 415 is associated with a decrease in the in vivo variable (AUC/D). Due to the therapeutic 416 relevance of this in vivo parameter (directly related to the extent of absorption), this 417 correlation may have and advantage over level B measures (Cutler et al., 1997). 418 Finally, it is interesting to note that in vivo drug absorption rates from all 419 formulations were faster than the in vitro drug release rates, implying that the 420 gastrointestinal physiological conditions, which are more extreme than those in the in vitro 421 dissolution tests, enhanced release and, in turn, the absorption process. So, a most bio- 422 relevant media (with bile salts, enzymes, etc.) would be required in order to totally predict 423 the in vivo release/absorption profile from the in vitro release data. 424 425 CONCLUSIONS 426 In conclusion, potato starch-methyl methacrylate polymers are interesting excipients for 427 sustained drug release in solid oral dosage forms. In addition to the easy manufacture of 428 tablets by direct compression, the results show extended drug release/absorption of the 429 tested formulations in vivo by comparing their pharmacokinetic parameters with the 430 commercially available Theo-Dur® and Theolair® in beagle dogs under fasting conditions. 431 Hence, these materials could be a good alternative to incorporate drug candidates of similar 432 physico-chemical properties to theophylline to maintain therapeutic plasma concentrations 433 during more lasting periods of time without signs of toxicity. FD-CSMMA was the 434 derivative that provided a better control of drug release/absorption process and a twice 435 daily dosage regimen would be recommended on the basis of the simulation studies in 436 dogs. 437 Moreover, the simpler formulation of starch-methyl methacrylate matrices 438 compared with the marketed products allowed establishing stronger relationships between 439 in vitro and in vivo data. The quantitative correlation between t50% in vitro and AUC/D in 440 vivo could be regarded as a first step to predict the extent of absorption from dissolution 441 data. Nevertheless, further investigation will be required on the most bio-relevant media in 442 order to totally predict the in vivo release/absorption profile from the in vitro release data. 443 444 ACKNOWLEDGMENT 445 This work is part of a project (MAT2004-01599) from Spanish Ministry of Education and 446 Science. The authors are extremely grateful to the animal care staff (Cordoba University) 447 for their help during in vivo experiments. 448 449 REFERENCES 450 Alberola, J., Perez, Y., Puigdemont, A., Arboix, M., 1993. Effect of age on theophylline 451 pharmacokinetics in dogs. Am. J. Veter. Res. 54, 1112-1115. 452 453 Bach, J.F., Kukanich, B., Papich, M.G., McKiernan, B.C., 2004. Evaluation of the bioavailability 454 and pharmacokinetics of two extended-release theophylline formulations in dogs. J. Am. Vet. Med. 455 Assoc. 224, 1113-1119. 456 457 Bauer, R.J., 2011. NONMEM Users guide. ICON Development Solutions: Ellicot City, Maryland. 458 459 Bravo-Osuna, I., Ferrero, C., Jiménez-Castellanos, M.R., 2005. Water sorption-desorption 460 behaviour of methyl methacrylate-starch copolymers: effect of hydrophobic graft and drying 461 method. Eur. J. Pharm. Biopharm. 59, 537-548. 462 463 Castellano, I., Gurruchaga, M., Goñi, I., 1997. The influence of drying method on the physical 464 properties of some graft copolymers for drug delivery systems. Carbohyd. Polym. 34, 83-89. 465 466 Conard, G.J., Jeinberg, M.J., Wong, F.N., Mildon, C.A., French, I.W., 1982. Clinical assessment of 467 theophylline absorption from Theolair-SR and two other sustained-release formulations relative to 468 a conventional formulation. Int. J. Pharm. 10, 259-273. 469 470 Cook, C.S., Hauswald, C.L., Grahn, A.Y., Kowalski, K., Karim, A., Koch, R., Schoenhard, G.L., 471 Oppermann, J.A., 1990. Suitability of the dog as an animal model for evaluating theophylline 472 absorption and food effects from different formulations. Int. J. Pharm. 60, 125-132. 473 474 Crombeen, J.P., De Blaey, C.J., 1983. In vitro testing of controlled release preparations: Theolair®, 475 Theograd® and Theolin®. Pharm. Weekblad. 5, 65–69. 476 477 Cutler, D.J., Beyssac, E., Aiache, J-M., 1997. Level B and C in vivo/in vitro correlations: statistical 478 considerations. Int. J. Pharm. 158, 185-193. 479 480 El-Sayed, G., El-Said, Y., Meshali, M., Schwartz, J.B., 1996. Comparative bioavailability of 481 theophylline from tablets with different sustained release kinetics. S.T.P. Pharma Sciences. 6, 398- 482 402. 483 484 FDA’s Guidance for Industry, 1997. Extended Release Oral Dosage Forms: Development, 485 Evaluation and Application of In vitro/In vivo Correlations. Center for Drug Evaluation and 486 Research (CDER), Rockville. 487 http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm 488 489 FDA’s Guidance for Industry, 2001. Bioanalytical Method Validation. Center for Drug Evaluation 490 and Research (CDER), Rockville. 491 http://www.fda.gov/Drugs/GuidanceComplianceRegulatoryInformation/Guidances/default.htm 492 493 FDA, drug information: DailyMed, 2012. Theophylline anhydrous solution. 494 http://dailymed.nlm.nih.gov/dailymed/archives/fdaDrugInfo.cfm?archiveid=56582. 495 496 Ferrero, M.C., Velasco, M.V., Ford, J.L., Rajabi-Siahboomi, A.R., Muñoz, A., Jiménez-Castellanos, 497 M.R., 1999. Determination of the glass transition temperatures of some new methyl methacrylate 498 copolymers using Modulated Temperature Differential Scanning Calorimetry (MTDSC). Pharm. 499 Res. 16, 1464-1469. 500 501 Ferrero, C., Jiménez-Castellanos, M.R., 2002. The influence of carbohydrate nature and drying 502 methods on the compaction properties and pore structure of new methyl methacrylate copolymers. 503 Int. J. Pharm. 248, 157-171. 504 505 Ferrero, C., Bravo, I., Jiménez-Castellanos M.R., 2003. Drug release kinetics and fronts movement 506 studies from methyl methacrylate (MMA) copolymer matrix tablets: effect of copolymer type and 507 matrix porosity. J. Control. Rel. 92, 69-82. 508 509 Ferrero, C., Jiménez-Castellanos, M.R., 2014. In vitro release testing of matrices based on starch- 510 methyl methacrylate copolymers: Effect of tablet crushing force, dissolution medium pH and 511 stirring rate. Int. J. Pharm. 461, 270-279. 512 513 Gascón, A.R., Calvo, B., Hernández, R.M., Domínguez-Gil, A., Pedraz, J.L., 1994. Interspecies 514 scaling of cimetidine-theophylline pharmacokinetic interaction: Interspecies scaling in 515 pharmacokinetic interactions. Pharm. Res. 11, 945-950. 516 517 Hayashi, T., Kanbe, H., Okada, M., Kawase, I., Ikeda, Y., Onuki, Y., Kaneko, T., Sonobe, T., 2007. 518 In vitro and in vivo sustained-release characteristics of theophylline matrix tablets and novel cluster 519 tablets. Int. J. Pharm. 341, 105-113. 520 521 Holford, N., 2005. The Visual Predictive Check – Superiority to Standard Diagnostic (Rorschach) 522 Plots. Abstract 738. Pamplona, Spain, pp. 14. http://www.page-meeting.org/?abstract=738 523 524 Ikegami, K., Tagawa, K., Osawa, T., 2006. Bioavailability and in vivo release behavior of 525 controlled-release multiple-unit theophylline dosage forms in beagle dogs, cynomolgus monkeys, 526 and göttingen minipigs. J. Pharm. Sci. 95, 1888-1895. 527 528 Jolley, M.E., Stroupe, S.D., Schwenzer, K.S., Wang, C.J., Lu-Steffes, M., Hill, H.D., Popelka, S.R., 529 Holen, J.T., Kelso, D.M., 1981. Fluorescence polarization immunoassay. III. An automated system 530 for therapeutic drug determination. Clin. Chem. 27, 1575-1579. 531 532 Jones, B., Kenward, M.G., 1989. Design and Analysis of Cross-Over Trials, second ed. Series: 533 Monographs on Statistics and Applied Probability, Vol. 98. Chapman & Hall, London. pp.408. 534 535 Jonsson, E.N., Karlsson, M.O., 1999. Xpose: an S-PLUS based population 536 pharmacokinetic/pharmacodynamic model building aid for NONMEM. Comput. Meth. Programs 537 Biomed. 58, 51–64. (http://xpose.sourceforge.net) 538 539 Karim, A., 1986. Effects of food on the bioavailability of theophylline from controlled-release 540 products in adults. J. Allergy Clin. Immunol. 78, 695-703. 541 542 Kimberley, A.L., Tolle, S., Sheskey, P.J., 2002. Solubility and permeability of anhydrous 543 theophylline with application to the Biopharmaceutics Classification System. The Dow Chemical 544 Company, Midland MI 48674, August. 545 546 Kortejärvi, H., Mikkola, J., Bäckman, M., Antila, S., Marvola, M., 2002. Development of level A, B 547 and C in vitro-in vivo correlations for modified-release levosimendan capsules. Int. J. Pharm. 241, 548 87-95. 549 550 Kuze, T., Miyazaki, H., Taneike, T., 1988. Theophylline: pharmacokinetics, metabolism and 551 urinary excretion in dogs. Folia Pharmacologica Japonica. 91, 325-334. 552 553 Liaw, J., Rubinstein, A., Robinson, J.R., 1990. Bioavailability study of Theo-dur® tablets in the 554 fasted cannulated dog. Int. J. Pharm. 59, 105-114. 555 556 Lindbom, L., Pihlgren, P., Jonsson, N., 2005. PsN-Toolkit-A collection of computer intensive 557 statistical methods for non-linear mixed effect modeling using NONMEM. Comput. Meth. 558 Programs Biomed. 79, 241–257. (http://psn.sourceforge.net) 559 560 Lindenberg, M., Kopp, S., Dressman, J.B., 2004. Classification of orally administered drugs on the 561 World Health Organization Model list of Essential Medicines according to the biopharmaceutics 562 classification system. Eur. J. Pharm. Biopharm. 58, 265-278. 563 564 Mengozzi, G., Intorre, L., Bertini, S., Giorgi, M., Soldani, G., 1998. Comparative bioavailability of 565 two sustained-release theophylline formulations in the dog. Pharmacol. Res. 38, 481-485. 566 567 Mitenko, P.A., Ogilvie, R.I., 1972. Rapidly achieved plasma concentration plateaus with 568 observations on theophylline kinetics. Clin. Pharmacol.Therap. 13, 329-335. 569 570 Miyazaki, S., Kubo, W., Attwood, D., 2000. Oral sustained delivery of theophylline using in-situ 571 gelation of sodium alginate. J. Control. Rel. 67, 275-80. 572 573 Miyazaki, S., Kawasaki, N., Endo, K., Attwood, D., 2001. Oral sustained delivery of theophylline 574 from thermally reversible xyloglucan gels in rabbits. J. Pharm. Pharmacol. 53, 1185-1191. 575 576 Munday, D.L., Fassihi, A.R., 1995. In vitro–in vivo correlation studies on a novel controlled release 577 theophylline delivery system and on Theo-Dur tablets. Int.J. Pharm. 118, 251–255. 578 579 Muskó, Zs., Pintye-Hódi, K., Gáspár, R., Pintye, J., Szabó-Révész, P., Erős, I., Falkay, G., 2001. 580 Study of in vitro and in vivo dissolution of theophylline from film-coated pellets. Eur. J. Pharm. 581 Biopharm. 51, 143-146. 582 583 Nabais, T., Brouillet, F., Kyriacos, S., Mroueh, M., Amores da Silva, P., Bataille, B., Chebli, C., 584 Cartilier, L, 2007. High amylose carboxymethyl starch matrices for oral sustained drug-release: In 585 vitro and in vivo evaluation. Eur. J. Pharm. Biopharm. 65, 371-378. 586 587 Ochoa, L., Igartua, M., Hernández, R.M., Solinís, M.A., Gascón, A.R., Pedraz, J.L., 2010. In vivo 588 evaluation of two new sustained release formulations elaborated by one-step melt granulation: 589 Level A in vitro-in vivo correlation. Eur. J. Pharm. Biopharm. 75, 232-237. 590 591 Podczeck, F., 1993. Comparison of in vitro dissolution profiles by calculating mean dissolution time 592 (MDT) or mean residence time (MRT). Int. J. Pharm. 97, 93–100. 593 594 Qiu, Y., Chidambaram, N., Flood, K., 1998. Design and evaluation of layered diffusional matrices 595 for zero-order sustained-release. J. Control. Rel. 51, 123-130. 596 597 Roshdy, M.N., Schnaare, R.L., Sugita, E.T., Schwartz, J.B., 2002. The effect of controlled release 598 tablet performance and hydrogel strength on in vitro/in vivo correlation. Pharm. Dev. Technol. 7, 599 155-168. 600 601 A) B) 1 10 100 0 2 4 6 8 10 12 Concentration (µg/ml) Time (h) 0,1 1 10 100 04812 16 20 24 Concentration (µg/ml) Time (h) FD-HSMMA Theo-Dur C) 0,1 1 10 100 04812 16 20 24 Concentration (µg/ml) Time (h) OD-CSMMA FD-CSMMA Theolair Figure 1 Time (h) 0 5 10 15 20 25 Conc (g/mL) 0.1 1 10 THEOPHYLLINE IV PRED THEOPHYLLINE IV P2.5 THEOPHYLLINE IV P97.5 THEOPHYLLINE IV DV Figure 2 Figure 3 Figure 5. Theo-Dur® (reference) and FD-HSMMA (dose = 100 mg) Cmax and Cmin predicted plasma levels of theophylline in steady state. Time (h) 020 40 60 80 100 Concentration (g/mL) 0 2 4 6 8 10 12 14 Theo-Dur FD-HSMMA Time (h) 020 40 60 80 100 Concentration (g/mL) 0 5 10 15 20 FD-CSMMA OD-CSMMA Theo-Lair Max tolerate conc. Figure 4 A) B) C) y = 2,3969x - 22,238 R² = 0,9332 0 20 40 60 80 100 020 40 60 80 100 % absorbed "in vivo" % released "in vitro" FD-HSMMA y = 1,7735x - 19,578 R² = 0,9538 0 20 40 60 80 100 020 40 60 80 100 % absorbed "in vivo" % released "in vitro" OD-CSMMA y = 1,868x - 6,9617 R² = 0,9498 0 20 40 60 80 100 020 40 60 80 100 % absorbed "in vivo" % released "in vitro" FD-CSMMA Figure 5 A) B) y = -0,0389x + 1,0779 R² = 0,9411 0,75 0,8 0,85 0,9 0,95 1 2 3 4 5 6 7 8 AUC/D t50% TheoDur® FD-CSMMA Theolair® FD-HSMMA OD-CSMMA y = -0,0405x + 1,0765 R² = 0,9958 0,75 0,8 0,85 0,9 0,95 1 2 3 4 5 6 7 8 AUC/D t50% FD-CSMMA FD-HSMMA OD-CSMMA Figure 6