Optimizacija izdelave filmsko obloženih pelet v zvrtinčenih plasteh s pomočjo in-line tehnik : doktorska disertacije

Hudovornik, Grega (author), Vrečer, Franc (mentor)

S povečevanjem zahtevnosti in kompleksnosti postopkov izdelave zdravil in zviševanjem regulatornih zahtev pri registracijah novih izdelkov se je v zadnjih letih začel bliskovit razvoj tehnik, ki omogočajo vrednotenje fizikalno-kemijskih parametrov farmacevtskih izdelkov med procesom izdelave. Te tehnike nam omogočajo pridobivanje večjega števila podatkov v realnem času, kar pomeni boljše poznavanje in hkrati boljši nadzor nad procesom izdelave kompleksnih farmacevtskih oblik, med katere sodijo tudi pelete. Proces izdelave pelet, ki s povečevanjem števila zdravil s prirejenim sproščanjem pridobivajo pomen, je zahteven in dolgotrajen, zato je popolno obvladovanje oblaganja pelet ključnega pomena. Med tehnike, ki se največkrat uporabljajo za spremljanje tehnoloških procesov, zagotovo sodi bližnja infrardeča spektroskopija (NIR). Uporabljamo jo v različnih tehnoloških fazah za identifikacijo materialov, določevanje končne točke mešanja, merjenje količine vode med sušenjem ali oblaganjem ter tudi za spremljanje nanosa zdravilnih učinkovin in funkcionalnih oblog pri oblaganju pelet. Pri tem je izjemnega pomena tudi informacija o velikosti pelet, saj se ta med procesom oblaganja z nanosom obloge spreminja. Tehnika prostorskega filtra (SFT) omogoča spremljanje rasti pelet v oblagalni napravi in v realnem času, kar bistveno pripomore k boljšemu nadzoru prirasta velikosti pelet med oblaganjem. V uvodu doktorskega dela smo se v prvem delu osredotočili na pelete, proces oblaganja in na proces tvorbe filma. Te procese smo v nalogi preučevali s pomočjo tehnik za »in-line« spremljanje procesov. Pri pregledu literature smo ugotovili, da se objave na področju tvorbe filma in procesa temperiranja osredotočajo na preučevanje formulacijskih in procesnih parametrov temperiranja. Večina avtorjev poudarja pomen ključnih procesnih parametrov (čas, temperatura, vlaga), vendar manjka celovit pristop, ki bi ocenil pomen medsebojnega vpliva vseh treh parametrov. V drugem delu uvodnega poglavja smo predstavili procesno-analizno tehnologijo (PAT), ki je skupni izraz za tehnike, ki se uporabljajo za spremljanje procesov v realnem času. Število raziskav uporabnosti »in-line« tehnik na različnih področjih farmacevtske tehnologije je v zadnjih letih zelo narastlo. Pregledali smo objave na področju spremljanja kakovostnih parametrov pelet z NIR in ugotovili, da so potrebne še dodatne raziskave v smeri robustnosti modelov pri pogojih, ki so izven območja meritev, uporabljenih pri kalibraciji modelov. Pri pregledu objavljene literature na področju spremljanja procesov v zvrtinčenih plasteh s pomočjo »in-line« tehnik za merjenje velikosti delcev smo ugotovili, da je večina objav na področju granuliranja. Področje oblaganja pelet pa je kljub velikemu potencialu tehnik za neposredno merjenje velikosti delcev in zgodnjo odkrivanje neželenih pojavov, kot sta aglomeracija in krušenje pelet, še relativno neraziskano. Prav tako v literaturi nismo zasledili neposredne primerjave med dvema najpogosteje uporabljenima tehnikama za merjenje velikosti delcev, in sicer tehniko merjenja odboja fokusirane laserske svetlobe (FBRM) in tehniko modificiranega prostorskega filtra (SFT). Obe merita tetivne porazdelitve velikosti delcev, vendar pri tem uporabljata nekoliko drugačen način merjenja in različne algoritme pri izračunu vrednosti, kar lahko vodi v drugačno odzivnost pri spremljanju procesov. V prvem delu doktorskega dela (poglavje 1) smo s pomočjo faktorskega načrta eksperimentov sistematično preverili vpliv treh ključnih procesnih parametrov temperiranja na hitrost sproščanja modelne zdravilne učinkovine. Kot sekundarni parameter smo spremljali še izgubo pri sušenju pelet (LOD). Ugotovili smo, da čas in temperatura temperiranja pozitivno vplivata na koalescenco polimernih delcev in s tem znižata hitrost sproščanja zdravilne učinkovine. Vendar je ta vpliv odvisen od vlažnosti zraka med temperiranjem. Pozitiven učinek smo ugotovili pri vlažnosti zraka do 50 %, pri višjih vlažnostih zraka pa pri dolgem času temperiranja in pri visokih temperaturah pride do zvišanja hitrosti sproščanja modelne zdravilne učinkovine. Z nadaljnjimi študijami smo ugotovili, da pri teh pogojih pride do difuzije zdravilne učinkovine iz obloge z učinkovino v filmsko oblogo. Zaradi povišane vlažnosti zraka pride do absorpcije vlage v pelete, kar omogoči difuzijo zdravilne učinkovine v filmsko oblogo. Učinkovina se pri tem vgradi v strukturo filmske obloge, pri stiku pelet z medijem za raztapljanjem pa se nato raztopi. Ob tem nastanejo v oblogi prazni kanali, ki spremenijo permeabilnost obloge. To se nato odrazi v hitrejšem sproščanju zdravilne učinkovine. V dodatku k poglavju 1 smo preverili, ali lahko z NIR spremljamo proces temperiranja v vzorcih z izrazito različno vsebnostjo vlage v peletah. Čeprav ima voda v spektru NIR zelo široke absorpcijske trakove, kar moti določevanje ostalih fizikalno-kemijskih sprememb v vzorcih, lahko to tehniko uporabljamo za predviden namen, torej za vrednotenje vpliva procesa temperiranja na hitrost sproščanja uporabljene zdravilne učinkovine iz pelet. V drugem delu naloge (poglavje 2) smo se osredotočili na študij procesa oblaganja pelet. Pri tem smo uporabili tehniki NIR in SFT. Iz meritev NIR smo uspešno izdelali model za napoved vsebnosti vlage v peletah, ki smo ga povezali z meritvami izgube pri sušenju. V nadaljevanju smo preverili njegovo robustnost ter ugotovili, da je model manj natančen izven kalibracijskega območja in občutljiv za večje spremembe v sestavi obloge. Pri manjših spremembah v sestavi pa smo vseeno dosegli zadovoljivo napovedno moč modela. Pri tehniki SFT smo rezultate meritev preverili s statično slikovno analizo. Primerjava rezultatov je pokazala, da je SFT dobro orodje za spremljanje rasti in krušenja pelet, slabše rezultate pa smo dosegli pri zaznavanju aglomeracije. To smo pripisali tudi pogojem oblaganja in lastnostim obloge, saj meritve kažejo, da so se med oblaganjem tvorili šibko povezani agregati pelet, ki med procesom oblaganja razpadejo. Kljub temu jih je sonda zaznala kot aglomerate. V zadnjem delu raziskave pa smo povezali meritve obeh »in-line« sond in podatke sonde SFT uporabili za kalibracijo modela NIR za napovedovanje debeline pelet. Dobljeni model je imel zadovoljivo natančnost, kar kaže, da lahko meritve NIR uporabimo tudi za napovedovanje debeline obloge, ki je pri peletah s prirejenim sproščanjem ključnega pomena za doseganje ustrezne kakovosti izdelka. V tretjem delu (poglavje 3) smo izvedli primerjavo dveh najbolj uporabljenih tehnik za spremljanje velikosti delcev v realnem času. Kljub nekoliko drugačnemu načinu merjenja tehnika merjenja odboja fokusirane laserske svetlobe (FBRM) in tehnika modificiranega prostorskega filtra (SFT) merita tetivne dolžine delcev in iz njih izračunavata distribucije tetivnih dolžin. Razlike v načinu merjenja pa tudi v načinu čiščenja merilnega dela sonde bi lahko prispevale k razlikam v odzivnosti na spremembe v velikosti delcev. Kot modelni sistem smo uporabili granuliranje s talinami, ki je v primerjavi z oblaganjem pelet proces, kjer je sprememba velikosti hitrejša in bolj izrazita kot pri oblaganju. Rezultati so pokazali, da sta obe sondi izmerili primerljive trende evolucije rasti granul, pričakovano pa smo zaznali manjše razlike v absolutnih vrednostih izmerjenih velikosti granul. Rezultati in zaključki doktorskega dela bodo prispevali k boljšemu razumevanju uporabnosti »in-line« tehnik za spremljanje procesov v zvrtinčenih plasteh, še posebej oblaganja pelet. Z uporabo razvitih modelov NIR ter merjenjem velikosti delcev bistveno prispevamo k boljšemu razumevanju procesov, hkrati pa so dodatne informacije, ki jih pridobimo v realnem času, poleg procesnih parametrov ključnega pomena za vodenje in optimizacijo procesa oblaganja pelet.

• Language: Slovenian
• Keywords: farmacevtska tehnologija, farmacevtske oblike, filmsko obložene tablete, oblaganje, tehnološki postopki
• Work type: Dissertation
• Typology: 2.08 - Doctoral Dissertation
• Organization: FFA - Faculty of Pharmacy
• Place of publishing: Ljubljana
• Publisher: [G. Hudovornik]
• Year: 2015
• Number of pages: 177 str.
• PID: 20.500.12556/RUL-143752
• UDC: 615.453.5(043.3)
• COBISS.SI-ID: 282794752
• Publication date in RUL: 11.01.2023

Secondary language

• Language: English
• Title: Optimization of fluid bed pellet film coating process using in-line techniques

Abstract

Increasing complexity of the drug manufacturing process on one hand and increased demands raised by regulatory agencies in the past years on the other hand led to a development of new techniques for real time monitoring of pharmaceutical processes. These techniques are able to acquire a vast amount of data during the process which increases the knowledge about the process and thus enables better process control. By increasing the number of modified release formulations the pellet coating process is also being used for many new medicines coming to the market. Since the pellet coating process is a demanding task the control over the process is essential. One of the most frequently used techniques used for real time process monitoring is near-infrared spectroscopy (NIR). NIR is used to monitor different quality attributes in the most solid dosage forms manufacturing steps ranging from material identification, mixing endpoint determination, moisture level monitoring during the drying or coating and also for quantitative monitoring of API layering or pellet film coating process. Another key characteristic of pellets is particle size which is changing during the coating process. The Spatial filtering technique (SFT) is based on modified spatial filter and enables the particle chord lengths measurements during the coating process which contributes to better process control during the coating process. The first part of the introduction to the thesis is focused on the pellets as a dosage form, the pellet coating process and the film formation process. These are the process that were studied in the present thesis by the in-line techniques. By reviewing the available literature sources on the film formation and the curing process we concluded that most of the authors emphasize the impact of the key curing process parameters (Time, Temperature and Relative Air Humidity) but we did not find any publication dealing with the complete overview of key process parameters and their interactions. The second part of the introduction is focused on the process analytical technology (PAT) which is a common expression used for real time process monitoring techniques. The number of published research articles on the use of in-line techniques has risen in the past years. By studying the available literature on implementation of NIR for pellet coating monitoring we concluded that additional research is needed to investigate the robustness of the models when changing conditions which were not taken into account when calibrating the models are used. By studying the available literature on the particle size measurement in fluid bed processes we concluded that most of the publications are focused on the fluid bed granulation although the particle size is also one of the key characteristics in the pellet coating process. Only few publications deal with the particle size measurements and agglomeration/attrition detection during the pellet coating process. Most of the publications on particle size measurements deal with Focused Beam Reflectance Measurement (FBRM) and SFT techniques which both measure the particle chord length yet no direct comparison of both techniques is published to date. The first part of thesis (Chapter 1) focuses on the pellet curing process. The Design of Experiments (DoE) was used to systematically study the influence of the three key curing process variables on the released amount of the model API. Loss on drying (LOD) was also measured as a secondary parameter during the trials. We concluded that time and temperature have positive impact on the coalescence of polymer particles. However, the impact of time and temperature is strongly dependent on the relative air humidity used during the curing process. When higher air RH was used (above 50%) a rise in the released amount of API was observed. Additional research has shown that active pharmaceutical ingredient (API) diffusion into the prolonged released coating has taken place with high air RH. We concluded that high air RH causes the moisture uptake by the pellets which enables API diffusion into the film coating. The API is then included in the structure of the coating. When subjecting the pellets to the dissolution media the API dissolves forming the channels in the coating that change the permeability of the coating and increases the drug release. The supplement to the Chapter 1 focuses on the study of the ability of NIR to monitor the curing process in the samples with different moisture levels. Water has wide and high absorption bands in the NIR spectrum which often interferes with the determination of other characteristics. We concluded that in spite of the interfering water absorption bands a model for quantitative prediction of API release can be built. However, the modelling of API release in the samples where API diffusion took place was not successful. The second part of the thesis (Chapter 2) focuses on monitoring of the pellet coating process by NIR and SFT. We successfully modelled the NIR measures to predict the moisture content which was correlated to loss on drying. Further on we tested the robustness of the model to changes and found out that the model was less accurate outside the moisture range that was used for calibration but maintained sufficient accuracy when minor changes in composition were implemented. SFT measurements were correlated to static image analysis. The comparison of the measurements confirmed the usefulness of SFT to monitor pellet coating process and detection of pellet abrasion. However, the probe failed to identify the agglomeration. This fact can be attributed to the coating conditions used and the coating properties. The results of the research show that pellet aggregates were formed during the coating and these were faulty detected as agglomerates by the probe. In the last part of the research on pellet coating we developed a NIR model for pellet coating thickness prediction by using SFT measurements. The model was accurate enough to state that only NIR can be used to monitor the moisture content and coating thickness during the coating process. The third part (Chapter 3) of the thesis focused on the comparison of the two most frequently used techniques for in-line particle size measurements. Despite differences in the measuring principle both FBRM and SFT measure particle chord lengths and chord length distributions. The differences in the measuring principle as well as the different approaches used to clean the probes could attribute to the different response of the probes. The fluid bed hot melt granulation was used for comparison since the particle growth is much faster than in pellet coating process. The results have shown that both probes detect similar granule size growth trends but smaller differences in the absolute values which were measured were detected. The results and conclusions presented in the thesis will attribute to better understanding of the in-line techniques used for fluid bed process monitoring, especially the pellet coating process. Using the NIR models and in-line particle size measurement a better process understanding is achieved which is the basis for control and optimization of pellet coating process.