(M1430-11-61) Toward Improving the In Vitro, In Vivo Correlation of Lipid-Based Formulations Using a Simplified Gastro-Intestinal Lipolysis Protocol

Purpose: Lipid-Based Formulations (LBF) are claimed to enhance the oral bioavailability of poorly water-soluble drugs (BSC II/IV) as they can maintain the drug in a solubilized state upon digestion of the formulation. pH-stat lipolysis is the most frequently used in vitro digestion model to evaluate this ability. The in-vitro lipolysis procedure was standardized and validated by the Lipid Formulation Classification System consortium (1,2). Although the method is useful to rank different formulations, it sometimes lacks in vitro in vivo correlation (ivivc). This is mainly due to the absence of gastric lipolysis step into the protocol, although digestion of lipid-based components begins in the stomach and is thought to account for 10–25% of the overall lipolysis process (3). With the recent marketing of a rabbit gastric lipase, the inclusion of this step in our protocol can now be considered. This study aims to establish a robust protocol to evaluate the impact of gastro-intestinal (GI) lipolysis on the LBF performance. Cinnarizine, a BCS II drug, was used as model API. The protocol has been optimised and compared to previous results using recombinant dog gastric lipase (rDGL) (4).

Methods: Preparation of media and enzyme solution: The lipolysis buffer is prepared by dissolving Tris, calcium chloride (CaCl₂), sodium chloride (NaCl) and sodium taurodeoxycholate (NaTDC) in Milli-Q water. The gastric solution is prepared to obtain a final activity of 60 U/mL. The pancreatin solution is composed of 1 g of pancreatin powder dissolved into 5 mL of lipolysis buffer. After 10 minutes of magnetic stirring the solution was centrifuged (Universal centrifuge 320R) at 2800g and 5°C for 10 min. The supernatant was withdrawn to be used in the lipolysis test. In vitro lipolysis test: The experimental setup consisted in a pH stat apparatus (Metrohm® AG, Switzerland), comprising a Titrando 802 propeller stirrer/804 Ti Stand combination, a glass pH electrode (iUnitrode) and one 800 Dosino dosing units coupled to 5 mL autoburette. The apparatus is connected to a PC and operated using Touch Control 900 software. The jacketed vessel was thermostated at 37°C and filled with 30ml of lipolysis medium. Test sample is added to lipolysis buffer and allow to disperse for 10min before addition of 6ml of gastric solution. 4ml of pancreatin solution and 18ml of lipolysis buffer are then added after 30min more. Throughout this test, aliquots of 1 mL were sampled at various time points to determine the apparent solubility of the cinnarizine in the micellar phase by HPLC analysis.

Results: 1-Protocol definition: The gastric and intestinal lipolysis media from protocol 1 come from the compositions described in Food and Function (5) while the media of protocol 2 are prepared according to the compositions determined by the LFCS (table 1). The media of protocol 1 are more difficult to prepare because the reagents are difficult to solubilize, and the pH is problematic to stabilize. 250 mg of Cinnarizine was fully solubilized into 1.25 g of Labrasol® ALF, a water-soluble surfactant (HLB 12). The GI lipolysis of this formulation was hence tested according to both protocols. The data shows similar curve profiles, but there are higher standard deviations for protocol 1 compared to protocol 2 (Figure 1). With a composition very close to LFCS, protocol 2 is the simplest to implement, with similar lipolysis data compared to previous results obtained using rDGL (4). Therefore, protocol 2 was selected for future GI lipolysis trials. 2-Impact of LBF quantity:  A different quantity of LBF was evaluated during the GI lipolysis test since it can impact the micellization and solubilization of the API. In each formulation, 250 mg of cinnarizine was fully solubilised either in 1.25 g or 750mg of Gelucire® 44/14, a water-soluble surfactant with a HLB 14 (figure 2): the higher the quantity of LBF, the better the micellization and solubilisation of the API, with smaller standard deviation.

Conclusion: The implementation of a gastric lipolysis step to simulate the in vitro digestion of LBF is critical to better predict their performance and the fate of poorly water-soluble active ingredients. The protocol 2 is easy to be implemented and demonstrated ability to quantify the impact of the gastric phase before the intestinal phase. This protocol must be further tested with others LBF types and with others API, to see its impact on solubilization of the API. IVIVC studies must also be carried out to evaluate the relevance of this in vitro test.

References: 1. Williams, HD. et al. (2012) Toward the establishment of standardized in vitro tests for lipid-based formulations, Part 2: The effect of bile salt concentration and drug loading on the performance of type I, II, IIIA, IIIB, and IV formulations during in vitro digestion. In: Molecular pharmaceutics, vol. 9, n° 11, p. 3286–3300.
2. Williams, HD. et al. (2012) Toward the Establishment of Standardized In Vitro Tests for Lipid-Based Formulations, Part 1: Method Parameterization and Comparison of In Vitro Digestion Profiles Across a Range of Representative Formulations. In: Journal of pharmaceutical sciences, vol. 101, n° 9, p. 3360–3380.
3. Thomas, N. et al (2012) Characterising lipid lipolysis and its implication in lipid-based formulation development. In: The AAPS journal, vol. 14, n° 4, p. 860–871.
4. Fernandez, S. et al (2009) In Vitro Gastrointestinal Lipolysis of Four Formulations of Piroxicam and Cinnarizine with the Self Emulsifying Excipients Labrasol and Gelucire 44/14. In: Pharmaceutical Research, vol. 26, n° 8, p. 1901–1910.
5. A. Brodkorb, L. Egger, et al (2019), INFOGEST static in vitro simulation of gastrointestinal food digestion, Nat. Protoc. 14 991–1014.


Figure 1: GI lipolysis test according to two different protocols


Figure 2: Comparison of GI lipolysis of cinnarizine with different Gelucire® 44/14 concentration


Table1: Composition of gastric and intestinal lipolysis according to protocols 1 and 2