Razvoj modela prehoda pelet skozi želodec za napovedovanje kinetike absorpcije učinkovin : doktorska disertacija

Pišlar, Mitja (author), Mrhar, Aleš (mentor), Bogataj, Marija (comentor)

V splošnem modeli predstavljajo poenostavitve realnih in kompleksnih sistemov, pri čemer modeli vključujejo glavne dejavnike, ki so ključni za razumevanje, pojasnjevanje ter napovedovanje obnašanja takih kompleksnih sistemov. V sklopu doktorske naloge smo razvili model, ki opisuje sproščanje natrijevega diklofenakata (DIK-Na) po zaužitju pelet s podaljšanim sproščanjem in napoveduje absorpcijo diklofenaka v sistemski krvni obtok. Kinetika sproščanja DIK-Na je odvisna od pH medija, zato se sproščanje DIK-Na iz pelet, ki so prisotne v želodcu, bistveno razlikuje od sproščanja DIK-Na iz pelet, ki so prisotne v tankem črevesu. Pri napovedovanju absorpcije diklofenaka je tako pomembno, da upoštevamo kinetiko prehoda pelet skozi želodec. V razvitem modelu napovedovanja absorpcije smo upoštevali individualne profile praznjenja pelet iz želodca, saj se kinetika prehodov pelet skozi želodec v pogojih na tešče med posameznimi osebami razlikuje do te mere, da povprečni prehod ne prikazuje tipičnih prehodov pelet skozi želodec. V literaturi smo dobili nabor 11 individualnih profilov praznjenja pelet iz želodca v pogojih na tešče in za vsak profil smo ocenili čase zadrževanja posameznih frakcij pelet v želodcu. Izvedli smo različne in vitro teste sproščanja, kjer so bile pelete izpostavljene simuliranemu želodčnemu mediju toliko časa, kot so bili ocenjeni časi zadrževanja različnih frakcij pelet v želodcu. Pridobljene profile sproščanja natrijevega diklofenakata smo ustrezno utežili s pripadajočimi frakcijami in jih med seboj sešteli ter dobili napovedane in vivo profile sproščanja za vsak posamezni profil praznjenja pelet iz želodca. Napovedane in vivo profile sproščanja smo primerjali z absorpcijskimi profili diklofenaka pri 24 prostovoljcih, ki so v klinični študiji prejeli pelete s podaljšanim sproščanjem DIK-Na. Dokazali smo, da v primerjavi s konvencionalnimi testi sproščanja, ki jih predpisujejo regulatorni organi, z upoštevanjem individualnih profilov praznjenja pelet iz želodca bolje napovemo individualne in vivo profile sproščanja oziroma absorpcijske profile. Razviti model napovedovanja absorpcije predpostavlja, da so pelete v želodcu izpostavljene kislemu mediju s pH 1,8, zato smo v naslednji raziskavi preverili ali je predpostavka o konstantnem pH kislega medija ustrezna pri napovedovanju sproščanja DIK-Na. Preučevali smo hkraten vpliv časa zadrževanja tablete v simuliranemu želodčnemu mediju (SGF) in vrednosti pH SGF na sproščanje DIK-Na iz HPMC tablet s podaljšanim sproščanjem, saj v primerjavi s peletami s podaljšanim sproščanjem tablete predstavljajo bolj enostaven sistem. Dodatno smo tudi analizirali, kako se odraža ta vpliv na simulirane plazemske koncentracijske profile diklofenaka. Sproščanje DIK-Na v fosfatnem pufru s pH 6,8 je bilo odvisno tako od pH vrednosti SGF kot od časa zadrževanja tablete v SGF. Čas zadrževanja tablet v SGF je imel velik vpliv na sproščanje pri nižjih pH vrednostih SGF (od 1 do 2), kjer so se koncentracije sproščenega DIK-Na v fosfatnem pufru med in vitro testi sproščanja z različnimi časi zadrževanja pelet v SGF bistveno razlikovale. Po drugi strani pa je bil vpliv vrednosti pH večji pri daljših časih zadrževanja tablet v SGF, kjer so se koncentracije DIK-Na v fosfatnem pufru pomembno razlikovale. V primerih, ko so bile tablete le 10 min v SGF, pa so bile hitrosti sproščanja med različnimi pH vrednostmi SGF približno enake. Dokazali smo, da na in vitro sproščanje DIK-Na ter posledično na simulirane plazemske koncentracijske profile diklofenaka poleg časa zadrževanja v SGF vpliva tudi njegova pH vrednost. Preverili smo tudi, ali razviti model napovedovanja absorpcije ustrezno ponazarja čas prehoda pelet skozi tanko črevo, zato smo v literaturi pridobili individualne podatke prehodov tablet skozi tanko črevo pri osebah, ki so zaužili tablete na prazen želodec ter dobili prvi obrok hrane po 4 urah po zaužitju tablete. Analizirali smo čase prehodov tablet, ki ne razpadejo v želodcu ali tankem črevesu, saj v primerjavi s peletami predstavljajo bolj enostaven sistem. V 96 % primerov je bil prehod tablet skozi tanko črevo v območju od 1 do 6 ur. V 40 minutah po zaužitju kosila je približno 39 % tablet prešlo v debelo črevo, kar je tudi v skladu z delovanjem gastro-ileocekalnega refleksa, kjer zaužitje hrane pospeši prehod vsebine iz terminalnega ileuma v debelo črevo. Neparametrična regresija prehodov tablet v debelo črevo v odvisnosti od časa praznjenja pelet iz želodca nakazuje, da je v območju, ko je praznjenje tablet iz želodca 0–2 uri, čas prehoda skozi tanko črevo v povprečju med 3,5 in 4 ure. Dodatno smo analizirali prehode tablet skozi tanko črevo pri osebah, ki so prejele tableto na prazen želodec ter zaužile hrano od 30 min do 2 uri po prejemu tablete. Čas prehoda tablet skozi tanko črevo je bil pri osebah, ki so prejele prvi obrok hrane od 30 min do 2 uri po zaužitju tablete (mediana = 160 min), krajši v primerjavi z osebami, ki so prejele prvi obrok hrane šele 4 ure po zaužitju tablet (mediana = 215 min). Razlika med medianama je bila statistično značilna (p < 0,001). Variabilnost pH vrednosti želodčnega medija ter variabilnost prehodov pelet skozi tanko črevo nista vključeni v razviti model napovedovanja absorpcije, saj bi bil model v tem primeru preveč kompleksen. Razviti model napovedovanja absorpcije zato vključuje le variabilnost prehoda pelet skozi želodec ter ustrezne srednje fiziološke vrednosti pH želodčnega medija ter srednje vrednosti prehodov pelet skozi tanko črevo. Analiza dodatnih dejavnikov nakazuje, da njihova variabilnost vpliva na sproščanje, toda ne vemo, do katere mere se ti dejavniki izrazijo v in vivo sistemu ter kakšen je njihov hkratni vpliv na sproščanje in posledično na absorpcijo učinkovine v in vivo situaciji ob prisotnosti tudi vseh drugih dejavnikov. Razviti model napovedovanja absorpcije temelji le na podlagi 11 individualnih profilov praznjenja pelet iz želodca, ki smo jih dobili v literaturi. Razvili smo dva pristopa, s katerima smo na podlagi absorpcijskih profilov diklofenaka ocenjevali praznjenje pelet iz želodca pri 24 prostovoljcih, ki so v klinični raziskavi prejeli pelete s podaljšanim sproščanjem. Pri prvem pristopu smo z uporabo faktorja podobnosti (f2) med seboj primerjali individualne absorpcijske profile ter napovedane in vivo profile sproščanja. Tako smo dobili frekvenčno porazdelitev nabora 11 individualnih profilov praznjenja pelet iz želodca pri 24 prostovoljcih. Drugi pristop privzame, da je proces sproščanja DIK-Na v tankem črevesu linearen in časovno neodvisen. Tako smo z metodo dekonvolucije neposredno izračunali individualne profile praznjenja pelet iz želodca pri 24 prostovoljcih na podlagi individualnih absorpcijskih profilov ter in vitro testov sproščanja. Oba pristopa ocenjevanja praznjenja pelet iz želodca predstavljata dva različna modela, s katerima lahko ocenjujemo praznjenje pelet iz želodca za posamezno osebo na podlagi individualnih plazemskih koncentracijskih profilov ter in vitro testov sproščanja učinkovine iz pelet. Tako lahko dobimo dodatno informacijo o kinetiki praznjenja pelet iz želodca v populaciji, kar nam omogoča razvoj boljših modelov napovedovanja absorpcije učinkovin ter njihovih plazemskih koncentracijskih profilov.

• Language: Slovenian
• Keywords: zdravilne učinkovine, natrijev diklofenakat, peroralne farmacevtske oblike, pelete, podaljšano sproščanje, krvni obtok, absorpcija, farmakokinetika, prebavila, želodec, pH, matematični modeli, disertacije
• Work type: Doctoral dissertation
• Typology: 2.08 - Doctoral Dissertation
• Organization: FFA - Faculty of Pharmacy
• Place of publishing: Ljubljana
• Publisher: [M. Pišlar]
• Year: 2015
• Number of pages: 133 str.
• PID: 20.500.12556/RUL-143735
• UDC: 615.015:612.32(043.3)
• COBISS.SI-ID: 279280640
• Publication date in RUL: 11.01.2023

Secondary language

• Language: English
• Title: Modeling of pellets gastric emptying for prediction of drug absortion

Abstract

In general, models represent a simplification of the real and complex systems. They include the main factors that are crucial for understanding, explaining, and predicting the behaviour of such complex systems. In the doctoral theses we developed a model describing in vivo dissolution of sodium diclofenac (DIK-Na) and in vivo diclofenac absorption after ingestion of extended release pellets. Dissolution of the DIK-Na depends on the pH of the medium, therefore the DIK-Na release from the pellets that are present in the stomach is significantly different comparing with the DIC-Na release from the pellets that are present in the small intestine. Consequently, it is important to take into account the gastric emptying (GE) kinetics of the pellets when predicting diclofenac absorption. The difference of pellet GE between individuals is significant and the mean profile of pellet GE does not reflect the individual behaviour, hence, GE kinetic of pellets is included in the developed model of diclofenac absorption on the individual level. 11 individual GE profiles of pellets, which were administered under fasting conditions, were obtained from literature. The residence time in stomach was evaluated for each fraction of pellets of the GE profiles. Several in vitro dissolution tests were performed where pellets were retained in the simulated gastric fluid for different time, which was the same as residence time of different fraction of pellets in the stomach. For each pellet gastric emptying profile predicted in vivo dissolution profile was calculated by adequately weighting the in vitro dissolution profiles and adding them together. Predicted in vivo dissolution profiles were compared with the absorption profiles from 24 healthy subjects who had received extended release pellets with DIK-Na in a clinical study. We have shown that individual predicted in vivo dissolution profiles, which take into account individual pellet GE, better predict the in vivo absorption profiles of diclofenac comparing with conventional in vitro dissolution tests that are described by regulatory agencies. One of the assumptions of the developed model of absorption prediction is that pellets in the stomach are exposed to acidic medium with pH 1.8. Hence, a study was conducted where the assumption of a constant pH value of the acidic medium in the stomach was evaluated. In the study we have evaluated the simultaneous effect of retention time of the extended release (HPMC) tablets in the simulated gastric fluid (SGF) and the pH of the SGF on dissolution of DIK-Na. The influence of both effects on simulated plasma concentration profiles of diclofenac was also analysed. HPMC tablets were evaluated in the study as they represent a more easily solved system comparing with pellets. The release of sodium diclofenac was mainly observed after the change of SGF with the phosphate buffer with pH value 6.8. However, the release of DIK-Na was influenced by pH of the SGF and by tablet retention time in the SGF. The influence of the retention time in the SGF was greater at lower pH values (between 1 and 2) of the SGF, as the dissolution profiles of DIK-Na differ significantly between several in vitro dissolution tests with different tablet retention times in SGF. The influence of pH values of the SGF was greater at longer retention times. When tablets were retained 10 min in the SGF, dissolution profiles of DIK-Na did not differ between several in vitro dissolution tests with different pH values of the SGF. It was demonstrated that in addition to the retention time in SGF also the pH value of the SGF influences the vitro dissolution of DIK-Na and subsequently the simulated plasma concentration profiles of diclofenac. Additionally, we have evaluated whether the developed model adequately illustrates the small intestinal transit of the pellets. Therefore, individual small intestinal transit data from subjects who received the first meal at 4 h after tablet administration were obtained in the literature. Small intestinal transit times of non-disintegrating tablets were analyzed also in this case as they represent comparing with the pellets a more plain system. In 96% occasions small intestinal transit times were within 1 and 6 hours. Approximately 39% of tablets transit into the cecum in the 40 min after the meal intake, which is in accordance with gastro-ileocecal reflex, a mechanism that accelerates the transition of the chyme from the ileum into the colon after meal intake. The dependence of colon arrival times on GE times was described by the nonparametric regression, which shows that at GE times 0–2 h small intestinal transit times are on average between 3.5 and 4 h. Furthermore, tablet small intestinal transit times for subject who had ingested food between 30 min and 2 h after tablet administration were analyzed. Small intestinal transit times are shorter for subjects who had ingested the first meal between 30 min and 2 h after tablet administration (median = 160 min) compared to the subjects who had ingested the first meal at 4 h after tablet administration (median = 215 min). The difference between the medians was statistically significant (p < 0.001). The variability of pH value of gastric medium and the variability of pellet transit through the small intestine are not included in the developed model of diclofenac absorption since it would be too complex. Therefore, the developed model of diclofenac absorption includes only the variability of pellet GE and appropriate mean physiological pH values of gastric medium and mean values of small intestinal transit of pellets. Evaluation of additional factors indicates that they influence the diclofenac dissolution. However, it is not known to what extent are these factors manifested in vivo, and what is their impact on diclofenac dissolution and absorption in in vivo situation where also other factors are present. The developed model for absorption prediction is based on 11 individual GE profiles of pellets, which were obtained from literature. In order to estimate individual GE profiles of the pellets in 24 healthy subjects who had received extended release pellets with DIK-Na in a clinical study, two approaches were develop, which are based on absorption profiles of diclofenac. In the first approach the similarity factor (f2) was used in order to compare individual absorption profiles with the predicted in vivo dissolution profiles that were calculated from individual GE profiles. Consequently, we obtained frequency distribution of GE profile for 24 healthy subjects. The second approach presumes that dissolution of DIK-Na in the small intestine is linear and time-invariant process. Therefore, individual GE profiles were directly calculated for each 24 healthy subjects by deconvolution method, which is based on the individual absorption profiles and in vitro dissolution profiles. Both approaches represent two methods that enable estimating GE profiles for each individual based on individual plasma concentration profiles and in vitro dissolution tests of pellets. Consequently, with these methods it is possible to obtain additional information about population GE kinetics of the pellet, which enables us to develop better models for drug absorption and plasma concentration prediction.