Enabling Lipid Formulations That Harness Supersaturation and Drug Absorption in the GI Tract

Enabling Lipid Formulations That Harness Supersaturation and Drug Absorption in the GI Tract Colin W Pouton Monash Institute of Pharmaceutical Sciences colin.pouton@monash.edu AAPS, October 2015 monash.edu

Enabling Lipid Formulations That Harness Supersaturation and Drug Absorption in the GI Tract Generation of supersaturation during digestion and dispersion Lessons learned from digestion testing Polymeric precipitation inhibitors Future perspectives 2

Lipid Based Drug Delivery Systems basic concepts digestion bile D D disp D ppt Aim is to avoid drug precipitation (D ppt ) during dispersion and/or digestion Dissolution rate from crystalline solid drug is likely to be limiting for poorly water-soluble drugs fast slow fast D s Porter et al., Adv. Drug. Del. Rev. (2008) 60, 673 3

Generation of supersaturation during digestion and dispersion in the gastrointestinal tract Although the digestion of formulations and interaction with bile can provide a matrix for solubilization of drug, in practice many formulations can lead to drug precipitation. Precipitation can take place during dispersion or digestion Supersaturation will drive more rapid absorption as well. If this happens before precipitation then it may be a benefit. 4

LBDDS May Also Promote Supersaturation Drug solubility in GIT fluids plus (digested) lipids, surfactants in LBDDS Drug solubility in GIT fluids Solubilisation Effect Supersaturated LBDDS Supersaturation Effect Dispersed and digested LBDDS Conventional IR solid dose form 5

Lessons learned from digestion testing The Lipid Formulation Classification System (LFCS) Consortium was formed in 2010 (www.lfcsconsortium.org) During 2011-2014 the Consortium funded studies on dispersion and digestion testing and performed some IVIVC studies in dogs which have advanced our understanding - but there is a lot more research to be done 6

Introducing the LFCS Consortium A non-profit organization that sponsors and conducts research on lipid-based drug delivery systems (LBDDS) for the oral administration of poorly soluble drugs (www.lfcsconsortium.org) The Consortium is in its third year of operation Industrial Full Members Capsugel Sanofi Aventis Academic Members Monash University University of Copenhagen University of Marseille Associate Members Actelion BMS Gattefossé Merck-Serono NicOx Roche www.lfcsconsortium.org 7

What do lipid formulations consist of? LIPIDS/OILS Soybean oil, corn oil, Miglyol, Maisine TM 35-1, Capmul Include for: - Dissolving highly lipophilic drugs. - Maintaining solubilization post-dispersion. - Harnessing natural digestion, absorption/lymphatic pathways. COSOLVENTS Ethanol, PEG 400, propylene glycol etc. NON IONIC SURFACTANT(S) Tween 85 (low HLB), Tween 80, Cremophor EL (higher HLB) Include for : - Emulsification of the oil component - Minimizing loss of solubilization on dispersion/digestion Include for: - Higher drug loading capacity - Better dispersibility Drug is usually dissolved, and therefore, molecularly dispersed in the formulation. Commercial examples: Neoral, Agenerase, Fortovase etc. Further reading: Pouton and Porter 2008 Advanced Drug Delivery Reviews 60, 625-637 8

Current Lipid Formulation Classification System Formulations classified according to composition Typical composition (%) Type I Type II Type IIIA Type IIIB Type IV Triglycerides or mixture of glycerides Water insoluble surfactant Water soluble surfactant 100 40-80 40-80 <20 - - 20-60 - - 0-20 - - 20-60 20-50 30-80 Cosolvent - - 0-40 20-50 0-50 Pouton, Eur J Pharm Sci 29: 278-287 (2006) 9

The 8 lipid formulations investigated by the LFCS Type I Type II Type IIIA Type IIIB Type IV Lipids: Long-chain: Corn oil and Maisine TM 35-1 (in a 1:1 ratio) Medium-chain : Captex 300 and Capmul MCM EP (in a 1:1 ratio) Lipophilic surfactant: Tween 85 Hydrophilic surfactant: Cremophor EL Cosolvent: Transcutol HP Variation in : Composition Drug Loading Capacity Dispersibility Digestibility www.lfcsconsortium.org 10

Metrohm titration apparatus for in vitro digestion testing Dosing units Overhead stirrer ph-stat titrator to maintain constant ph during digestion. Rate and total titrant added informs of LBF digestibility. ph probe Reaction vessel www.lfcsconsortium.org 11

Drug load (saturation) very important for MC Type II-MC % danazol 100 75 50 25 0 20 40 60 80 90 Solubilised drug (aqueous phase) Precipitated drug Drug loading (% sat sol in formulation) Increasing drug load from 20-90% of saturated solubility in formulation, significantly decreases % solubilised post digestion www.lfcsconsortium.org Williams et al Mol Pharmaceutics (2012)

Drug load (saturation) very important for MC % danazol 100 75 50 25 0 Type II-MC Type IIIA-MC Type IIIB-MC Type IV Solubilised drug (aqueous phase) Precipitated drug 20 40 60 80 90 20 40 60 80 90 20 40 60 80 90 20 40 60 80 90 Drug loading (% sat sol in formulation) Degree of precipitation most significant in least lipophilic formulations (ie more cosolvent, more hydrophilic surfactants) www.lfcsconsortium.org Williams et al Mol Pharmaceutics (2012)

Drug load (saturation) less critical in LC formulations Type II-MC Type IIIA-MC Type IIIB-MC Type IV Type IIIA-LC 100 % danazol 75 50 25 0 20 40 60 80 90 20 40 60 80 90 20 40 60 80 90 20 40 60 80 90 20 40 60 80 90 Drug loading (% sat sol in formulation) Long chain lipid containing formulations resist ppt more favourably and are less sensitive to drug load Note: Type IIIA-LC formulations are incompletely digested in the fixed volume available (low density oil phase is shown in yellow) www.lfcsconsortium.org Williams et al Mol Pharmaceutics (2012)

Drug load (saturation) less critical in LC formulations Type II-MC Type IIIA-MC Type IIIB-MC Type IV Type IIIA-LC 100 % danazol 75 50 25 0 20 40 60 80 90 20 40 60 80 90 20 40 60 80 90 20 40 60 80 90 20 40 60 80 90 Drug loading (% sat sol in formulation) Why do LC formulations resist ppt more effectively? - Generate intestinal colloids with intrinsically higher drug solubility? - Stabilise supersaturated drug concentrations? www.lfcsconsortium.org Williams et al Mol Pharmaceutics (2012)

Towards a classification based on performance A-type B-type C-type D-type Performance test Dispersion Digestion (fasted) Digestion (stressed) Grading ( / ) based on level of drug precipitation Decreasing performance A traditional Type IIIA-MC formulation might be classified B-type at high drug loading but demonstrate A-type performance at low drug loading. Sub-classification of the A-type might be required e.g. LBFs showing A-type properties across a wide-range of drug loadings might be called a A*-type 16

Quantifying supersaturation in lipid based formulations [drug] aq Maximum [drug] aq (100% solubilised) Supersaturation ratio at time, t [drug] aq = Maximum supersaturation ratio (SR max ) Equilibrium drug solubility in drug-free AP DIGEST t time The greater the maximum supersaturation ratio, the greater the driving force to precipitation dependent on drug loading (ie saturation level) Does maximum (initial) supersaturation pressure dictate precipitation profiles?

Fenofibrate solubilized (mg/ml) In vitro digestion testing 4.0 Disp Digestion IIIA-LC 3.0 2.0 1.0 0.0 IIIA-MC IV IIIB-MC 0 20 40 60 Time (min) In vitro digestion data at a fixed 125 mg fenofibrate dose www.lfcsconsortium.org 18

Does SR MAX predict the likelihood of drug precipitation? SR MAX % non-precipitated dose < 40% 40-70% >70% 2.5 7.5 www.lfcsconsortium.org 19

Saturation study: the trends based on supersaturation High BS Fasted Highly diluted Min 40% 60% 80% 20% 40% 60% 80% 40% 60% 80% II-MC 30 3.1 4.7 6.2 1.5 3.1 4.6 6.2 7.6 11.5 15.3 60 3.1 4.7 6.2 1.5 3.1 4.6 6.2 7.6 11.5 15.3 IIIA-MC 30 2.7 4.0 5.3 1.4 2.8 4.3 5.7 5.0 7.6 10.1 60 2.7 4.0 5.3 1.4 2.8 4.3 5.7 5.0 7.6 10.1 IIIB-MC 30 4.5 6.8 9.1 2.0 3.9 5.9 7.8 3.6 6.0 8.0 60 4.5 6.8 9.1 2.0 3.9 5.9 7.8 3.6 6.0 8.0 IV 30 7.5 11.2 15.0 2.9 5.7 8.6 11.5 5.7 8.5 11.4 60 7.5 11.2 15.0 2.9 5.7 8.6 11.5 5.7 8.5 11.4 % non-precipitated dose < 40% 40-70% >70% www.lfcsconsortium.org 20

Tolf. acid solubilized in the aqueous phase (%) Danazol solubilized in the aqueous phase (%) Fenofibrate solubilized in the aqueous phase (%) SR M predicts drug fate during in vitro digestion of LBFs: 100 danazol 100 fenofibrate 75 75 50 Threshold SR M ~2.5 50 Threshold SR M ~3.0 25 25 0 0 5 10 SR M 100 75 50 25 (increasing supersaturation pressure) tolfenamic acid Threshold SR M ~3.0 0 0 5 10 15 SR M SR M threshold identified in each case. Above this supersaturation threshold, formulation performance is variable/poorer due to precipitation. 0 0 5 10 15 SR M Williams et al Mol Pharmaceutics (2012) in press www.lfcsconsortium.org 22

Profiling danazol precipitation during in vitro digestion Initiation of digestion 120 % Drug in aqueous phase 100 80 60 40 20 F1 F2 F3 F4 Oil Aq Pellet 0 0 10 20 30 40 50 60 Time (min) Cremophor EL On digestion, solubilising capacity lost to varying degrees 2 3 4 1 SBO/maisine Ethanol Cuine et al, Pharm Res 24, 748-757, 2007 23

Comparison of in vitro digestion vs in vivo exposure In vitro dispersion and digestion In vivo bioavailability in beagle dogs % Drug in aqueous phase 120 100 80 60 40 20 0 0 10 20 30 40 50 60 F1 F2 F3 F4 Danazol plasma concentration (ng/ml) 140 120 100 80 60 40 20 0 F1 F4 0 2 4 6 8 10 F2 F3 Time (min) Time (hr) In vivo exposure after oral administration to beagle dogs Lack of in vitro precipitation correlated well with enhanced in vivo exposure Use of digestion tests increasingly popular but some variation in methods across laboratories Cuine et al, Pharm Res 24, 748-757, 2007 24

In vivo results summary Fenofibric acid Plasma Conc. [ng/ml] Decreasing AUC 12000 10000 8000 6000 4000 2000 C max (µg/ml) AUC 0- (µg.h/ml) T max (hours) Type IIIA-LC 7.4 ± 1.6 34.0 ± 4.4 1.3 ± 0.3 Type IIIA-MC 4.8 ± 1.0 29.1 ± 2.9 1.4 ± 0.5 Type IV 6.8 ± 3.4 26.1 ± 6.6 0.9 ± 0.1 Type IIIB-MC 5.9 ± 1.3 25.3 ± 3.2 1.4 ± 0.3 no significant differences were observed across the four LFCS formulation types 0 0 2 4 6 8 Time [h] www.lfcsconsortium.org 25

In vivo AUC (µg*h/ml) In vitro-in vivo correlations In vitro summary: Fenofibrate solubilized (mg/ml) 4.0 3.0 2.0 1.0 0.0 Disp Digestion IIIA-LC IIIA-MC IV IIIB-MC 0 20 40 60 Calculate AUC and plot vs. in vivo AUC 50 40 30 20 10 0 IIIA-LC IV IIIA-MC IIIB-MC R² = 0.9182 0 50 100 150 200 250 300 Time (min) In vitro AUC (mg*min/ml) In vitro performance data at a fixed 125 mg fenofibrate dose Differences in in vivo exposure were comparatively small compared to differences in vitro www.lfcsconsortium.org 26

Polymeric precipitation inhibitors HPMC and HPMC-AS are polymeric excipients that have been used to produce amorphous spray-dried dispersion formulations of drugs. These polymers inhibit the precipitation of some drugs after dissolution of the formulations In recent years similar approaches to precipitation inhibition have been explored with lipid systems but this research is in its infancy. 27

Effect of HPMC on precipitation of danazol during digestion 600 Danazol in AP [ug/ml] 500 400 300 200 100 0.125% HPMC 0.025% HPMC 0.0125% HPMC No polymer 0-20 0 20 40 60 Time [min] HPMC effectively reduced ppt from MC SEDDS in conc dependent manner Conc required ~ 10 fold higher than in screening assay 28

% danazol solubilized It is at the SR M threshold when: Polymer precipitation inhibitors have the greatest effect: % Solubilised 100 80 60 40 20 Formulations: Captex /Capmul / Cremophor EL/EtOH % Solubilised add HPMC to the LBF 100 80 60 40 20 0 0 2 4 6 8 10 SF max S M 0 0 2 4 6 8 10 SF max HPMC inhibits precipitation at the threshold, but exhibits a lower utility at high supersaturations (i.e., >5). S M is shown during dispersion (triangles) or digestion (circles) from formulations containing danazol at either 40% (orange) or 80% (black) solubility in formulation. Purple formulations are identical but also contain 5% HPMC Anby et al Mol. Pharm. (2012) 9, 2063-2079 29

Future perspectives For weak electrolyte drugs the production of novel salts in the form ionic liquids is an interesting new option We need to develop a better understanding of the consequences of supersaturation in the gastrointestinal tract. This will require laboratory models that incorporate absorption. In the long-term IVIVC studies with a range of drugs will be required to build a database to help formulators predict in vivo performance. 30

Example Ionic liquid forms of CIN CIN octadecylsulphate T m 78-81ºC CIN decylsulphate Viscous liquid at RT CIN triflimide T m 38-43ºC CIN oleate T m 93-98ºC CIN free base T m 118-120ºC Sahbaz et al Mol Pharm (2015) 12, 1980-1991 31

CIN solubility, mg/g (free base equivalent) Solubility of CIN ionic liquids in two lipid formulations 400 350 300 LC-SEDDS MC-SEDDS >300 mg/g >300 mg/g LC-SEDDS: SBO/Maisine/Crem EL/EtOH 15/15/60/10 250 200 150 MC-SEDDS: MCT/Capmul/Crem EL/EtOH, 15/15/60/10 100 50 0 CIN Cin OS Cin ST Cin OL CIN DS Cin LS Cin TF Ionic liquid forms of CIN Solubility of CIN.IL forms substantially higher (up to 10-fold), some (decyl sulphate and triflimide) essentially miscible Sahbaz et al Mol Pharm (2015) 12, 1980-1991 32

C in n a riz in e p la s m a c o n c e n tra tio n (n g /m l) In vivo performance of CIN.DS containing lipid formulations 3 0 0 0 CIN.DS, LC SEDDS (125 mg/kg) 2 0 0 0 1 0 0 0 CIN.FB, LC SEDDS (35 mg/kg) CIN.FB, aq. suspension (125 mg/kg) 0 0 5 1 0 1 5 2 0 2 5 T im e (h ) CIN.DS IL allowed >3.5 increase in CIN dose Despite increased CIN/formulation ratio, absorption maintained Sahbaz et al Mol Pharm (2015) 12, 1980-1991 33

In situ rat loop model combined with in vitro digestion testing better IVIVC Matt Crum 34

The LFCS Consortium Industrial Full Members Capsugel Sanofi Aventis Academic Members Monash University University of Copenhagen University of Marseille Associate Members Actelion BMS Gattefossé Merck-Serono NicOx Roche

Acknowledgements University of Bath (1982-2001) Mark Wakerly Debbie Challis Linda Solomon Naser Hasan Rajaa Al Sukhun Monash University (2001-present) Jean Cuine Prof Chris Porter Kazi Mohsin Prof Bill Charman Mette Anby Dallas Warren Ravi Devraj Hywel Williams Orlagh Feeney Matt Crum R P Scherer Ltd (collaboration on digestion experiments 1991-1994) Cardinal Health (support for molecular dynamics modeling 2002-2004) Abbott Laboratories (lipid formulations 2002-2007) Michelle Long Capsugel (lipid formulations, LFCS and IVIVC 2003-present) Hassan Benameur, Jan Vertommen, Keith Hutchison, Hywel Williams 36