The impact of extrusion process parameters on drug recovery and the dissolution performance of solid dispersions of Ritonavir and AFFINISOL HPMC HME
Hot melt extrusion (HME) is a versatile, continuous and solvent-free process technology that has gained significant popularity in the pharmaceutical industry in recent years. The high number of poorly soluble drugs in both development pipelines and commercial products has directed formulators to utilise HME to manufacture amorphous solid dispersions (ASDs); the high energy state of the amorphous active and stabilising properties of the polymeric excipients provide significant increases in apparent solubility without sacrificing permeability — potentially resulting in great improvements in bioavailability.1–3
Hydroxypropyl methylcellulose (HPMC) is an amorphous, water-soluble polymer used in both immediate-release and controlled-release applications. Although HPMC has been shown to be highly successful at inhibiting recrystallisation and increasing bioavailability when used as the polymeric stabiliser in ASDs, it has been disadvantaged in hot melt extrusion owing to a high glass transition temperature (Tg), high melt viscosity and a significant colour change at elevated temperatures, requiring unique formulation work to overcome processing difficulties.4–6 Recently, The Dow Chemical Company introduced a new grade of HPMC, AFFINISOL HPMC HME, designed for hot melt extrusion with a notably lower Tg, reduced melt viscosity and reduced colour change at elevated temperatures.7
AFFINISOL HPMC HME can be successfully processed at a wide range of conditions into binary solid dispersions, with the resulting formulations providing increased solubility of the model compounds.8 Ritonavir (RTV) is a poorly soluble drug used in the treatment of HIV-infection (Figure 1). Previous studies have confirmed that RTV, when formulated into an ASD, can greatly improve the drug’s solubility and a commercial ASD is currently available.6,9 However, RTV displays thermal instability above its melt temperature, making formulation by extrusion challenging. In HME formulation, it may be necessary to include additives that enable process temperature reductions to ensure the stability of each component during extrusion. However, this can create undesirable formulation complexity and such additives may not be required if the operating design space is adequately explored. Thus, it is important to understand the impact of key process variables on drug degradation and product performance to fully optimise a system and ensure robust production.
The primary variables controlled by the operator include barrel temperature, which will dominate the extent of thermal exposure the active experiences, screw speed, which changes the residence time distribution and can generate viscous/frictional heating, and feed rate, which will dictate the residence time of the material within a twin screw extruder. In the present study, solid dispersions of RTV and AFFINISOL HPMC HME 15LV (AFF) were prepared by HME. The impact of the process variables (screw speed, feed rate and temperature) on drug degradation — as well as drug release rate — were explored using a factor screening Design of Experiments (DoE).
Ritonavir was purchased from Indo Overseas Trading Agencies-India. AFFINISOL HPMC HME 15LV was obtained from The Dow Chemical Company (Midland, MI, USA). The acetonitrile, potassium dihydrogen orthophosphate, methanol and acetonitrile used in the study were analytical grades procured from Orion Scientific Suppliers Pvt (Mumbai, India).
Hot melt extrusion: The formulation comprised 33% RTV blended with 66% AFF. All extrusions were performed on a Thermo Fisher Pharma 11 (Darmstadt, Germany) twin screw extruder and the formulations were fed into the extruder via a gravimetric feeder. The screw kneading elements used have 30, 60 and 90° configurations. Process variables were adjusted according to the DoE outlined below. The obtained extrudates were ground and sieved through a 16 mesh sieve.
Design of Experiments (DoE): The drug polymer blends were extruded at different processing conditions to explore the variables of feed rate, screw speed and temperature. The factor screening DoE used for the study was designed using JMP software. The order of the trials was randomised but, for clarity, the runs have been reordered by temperature in Table I.
Differential scanning calorimetry (DSC): DSC (TA Instruments, DE, USA) was used to study the thermal behaviour of RTV and the extrudates. The DSC experiments were run in a dry nitrogen atmosphere at a flow rate of 50 mL/min. The samples were weighed into an aluminium pan, crimped and heated at a ramp rate of 10 °C/min from 25–225 °C. A temperature modulation of 0.5 °C with a frequency of 40 seconds was applied. Data analysis was performed in TA Instruments Universal Analysis.
Powder X-ray diffraction: X-ray diffraction was performed on a PANalytical Empyrean (Almelo, the Netherlands) X-ray diffractometer. The X-ray was applied at a voltage of 40 kV and a current intensity of 20 mA. The samples were analysed across a 2θ range of 5–50° with a step size of 0.0020. The samples were placed in a zero background sample holder and incorporated on a spinner stage.
High performance liquid chromatography: The RTV drug content of the extrudates was evaluated using HPLC (Agilent 1260). The mobile phase was composed of acetonitrile and phosphate buffer (adjusted to pH of 4.0) in a ratio of 55:45 at a flow rate of 1 mL/min. The column used was an Eclipse plus C8 4.6 x 150 mm (5 µm) and RTV detection was determined at 246 nm.
Dissolution studies: In vitro dissolution studies were conducted using USP dissolution apparatus II while maintaining sink conditions to determine whether processing conditions impacted drug release rates. A quantity equivalent to 100 mg of RTV from extruded samples was weighed and filled into hard gelatin capsules. Dissolution of RTV from the capsules was performed in triplicate in 0.1 N HCl with the temperature maintained at 37 °C at a paddle speed of 75 RPM. The samples were collected at time points of 10, 20, 30, 45, 60 and 120 minutes and analysed by HPLC after filtering the samples through a 0.45 µm PVDF filter.
Hot melt extrusion: Ritonavir and AFFINISOL HPMC HME 15 cP were successfully extruded at all conditions of the DoE, confirming the broad processing window of AFF previously observed.8 No processing challenges were observed during the trials despite the low temperature and high feed rate of some runs. This suggests RTV has a plasticising effect on the polymer upon mixing. When extruded at temperatures of 130, 150 and 170 °C, the extrudates were clear and transparent, which indicated homogenous mixing and amorphisation of the drug with AFF. The colour of the extrudates was dependent upon the processing temperatures and feed rate. For example, the extrudate obtained at 130 °C, 250 g/h at 100 RPM (B. No. 2) was light yellow compared with extrudate obtained at 170 °C, 100 g/h at 100 RPM (B. No. 6), which was comparatively yellow in colour (Figure 2). This is because of the synergistic effects of high temperature and a longer residence time of B. No. 6 inside the extruder. The difference in colour does not impact the release rate from the resulting ASD, as will be discussed in Part II.