In recent years, the number of poorly soluble drugs coming out of the drug discovery laboratories has increased significantly and these already constitute nearly 90% of all molecules in the discovery pipeline. Approximately 70% of these drugs are BCS/DCS class II that have good permeability properties but exhibit poor water solubility.¹

To overcome the solubility issues of these new molecular entities, a number of strategies have been used to yield drug formulations with an increased solubility or biological exposure. One such strategy consists of generating amorphous solid dispersions (ASD) by different techniques such as spray drying, spray congealing, jet milling, wet polishing and hot melt extrusion, amongst the most important.

Because solubility is the key parameter driving the formulation work on amorphous solid dispersions, it is very important to have a comprehensive knowledge on dissolution of dosage forms and testing. In this sense, the following sections will describe and clarify the different aspects related to API/dosage form solubility, biorelevant dissolution testing and QC testing for final product release.

Attributes that influence dissolution
The dissolution performance of drugs and drug products is influenced by several active pharmaceutical ingredient (API) molecular and physical attributes as well as dosage form properties. Some of the most important API molecular properties are the molecular weight, melting temperature, pKa, partition coefficient and solubility, which, ultimately, depends on all other properties. With respect to API physical properties, the particle size distribution, solid-state properties (crystallinity vs. amorphicity) and surface characteristics like wettability are well known to influence the dissolution profile of drugs and drug formulations. Dosage form properties such as the formulation type (tablet, capsule, etc.), the release mechanism, type and ratio of excipients used and the manufacturing variables should also be considered.

Another very important parameter that depends on both the solubility and the solid-state properties of the API, is the potential for drugs to generate supersaturated solutions. This property should be taken in consideration when developing high-energy solid dosage forms such as salts, co-crystals or amorphous solid dispersions because there are specific thermodynamic and kinetic aspects that influence the dissolution behavior of these solid forms.

Supersaturation
Supersaturated solutions are highly concentrated solutions of drug where the apparent solubility (kinetic solubility) is above the thermodynamic equilibrium solubility achieved by dissolving the stable crystalline drug. This state is metastable, in the sense that there is no thermodynamic equilibrium and there is an inherent tendency for the drug to crystallize from solution. As previously mentioned, supersaturated solutions can be generated by dissolution of high-energy solids, such as salts, co-crystals or amorphous solid dispersions. Two other strategies can be used to generate supersaturated solutions: a solvent-shift method, that consists of dissolving the drug in a water-miscible solvent in which it has high solubility and dispersing a small aliquot of this solution into aqueous phase to generate a supersaturated system; and a pH-shift method, where the solution pH is changed so rapidly from the high to the low solubility end, that equilibrium is not reached and, therefore, supersaturation is generated.

The phenomenon of supersaturation is still not completely understood and there has always been a substantial knowledge gap in terms of understanding the extent and duration of supersaturation generated by solubility enhancing formulations. Recently, it has been suggested that the maximum theoretical supersaturation is given by the amorphous solubility.²
This correlation has been demonstrated by comparing the theoretical amorphous solubility of drugs with the concentration at which a phenomenon called Liquid-Liquid Phase Separation (LLPS) occurs. This phenomenon occurs at highly supersaturated solutions and the new phase that forms is a sub-micron (colloidal) drug-rich phase with no evidence of crystalline structure. It is therefore suggested that the maximum concentration of drug allowed in solution is governed by the solubility of the amorphous drug. After reaching that concentration, drug precipitates either as a liquid-like phase or a glassy amorphous precipitate.

In theory, determination of the maximum extend of supersaturation is simple; it just requires determination of LLPS concentration. In practice, it is extremely difficult to do that because many drug molecules have a great tendency to crystallize at such high drug concentrations and LLPS concentrations may never be reached as a consequence of fast crystallization kinetics. Polymers are well known to inhibit crystallization and can be used to delay the onset of crystallization thus allowing LLPS to occur at high concentrations.

Supersaturation in formulation development
Supersaturation studies can be very useful when developing a dissolution method because they allow establishing the maximum concentration of drug in solution that can be achieved from dissolution of the dosage form. Knowledge on the LLPS concentration and amorphous solubility is very important to define the upper limit of drug concentration in solution.

Supersaturation studies are also typically used in polymer screening methodologies for amorphous solid dispersion development. Upon establishing the target supersaturation level, the effect of polymers to inhibit crystallization from solution can be studied and those that prove more efficient in preventing crystallization are considered for the selection of the final formulation.

At Hovione, these supersaturation studies are an important part of a three-stage methodology that involves an initial computational screening stage followed by the physical stability evaluation of solid dispersions by DSC.

Biorelevant dissolution
It is important to acknowledge the journey an immediate release (IR) drug takes before reaching the site of absorption. Initially, it will disintegrate into granules, followed by small particle formation—all these forms will begin to dissolve leading to an increased concentration of drug in solution. This will happen both in the stomach, at low pH and also in the upper intestine at high pHs. The free drug substance in solution is now available for absorption. In this way, there are three phenomena controlling the oral absorption of IR drugs: solubility, dissolution and permeation (see Figure 1).

The ability of developing a dissolution-based strategy that allows both the analytical and the formulation scientists to better understand the under-development drug product is of utter importance. As an ultimate goal, the biorelevant dissolution method will allow a decrease in animal studies and also a successful Phase I clinical trial. The biorelevant dissolution method should occur at a pre-clinical phase, during early product development. It separates itself from the traditional pharmacopeial methods, since a 100% API release is not the main goal: either the effective drug substance concentration detection throughout time. For this reason, employing sink conditions should be considered, where three to ten times the volume required for drug substance saturation is used, surpassing the effect of reaching drug saturation. The application of a biorelevant dissolution method should enable the detection of precipitation issues throughout the GI tract, guide in the definition of the most suitable time of drug intake by knowing the effect of food in the solubilization of the drug and also discriminate differences in the manufacturing procedure.

Biorelevant method development
It is important to define the adequate duration for dissolution method, so that it resembles the residence time of the oral drug product in the GI tract. These times, according to Table 1, vary according to a fast or fed situation, and are correspondent to mean residence times, since this is patient dependent.

Accounting for the solubilization effects of bile salts and enzymes should also occur. The use of dissolution media that mimics the presence of these compounds will help understanding the effective drug concentration and enable a correlation with in vivo data. Media like FaSSIF or FeSSIF, corresponding to empty and post-prandial conditions, respectively, allows the analysis of the solubilization effects of bile salts by the presence of sodium taurocholate and enzyme effects by the presence of lecitine and pancreatine.

Compendial dissolution
During development stage, the dissolution method tends to progress depending on its intended purpose/applicability and should be re-assessed when human bioavailability data become available from the clinical formulations.

As a regulatory requirement for dosage forms, dissolution methods should be routinely used for quality control (QC) purposes assisting product release, stability, and ensuring the comparability and consistency between batches. In that sense, with the accumulation of experience, the early biorelevant method should be critically re-evaluated and potentially simplified, giving preference to compendial apparatus. The biorelevant method may not always be practicable, and may or may not be the same as the QC method due to its scope and constraints.

In opposition to common biorelevant dissolution, QC methods require sink conditions and the full release of the drug. Typically, dissolution medium for QC does not apply simulated fasted or fed state gastric/intestinal fluids but, instead, it uses conventional buffers (e.g. acetate, phosphate) for real-world applications. The QC method should be simple, user-friendly, robust and suitable for validation purposes exhibiting low variability and good profile shape. Conditions that are optimal for QC purposes may not be applicable to predict in vivo performance so it may be necessary to use two dissolution tests to meet different objectives such as development needs or regulatory demands.⁵

The development of a dissolution QC method can be initially supported by the dissolution data in different pHs values previously obtained in biorelevant media and at the end there should be an attempt to correlate the biorelevant dissolution test with the QC method in order to the capture the same discriminating trend of both methodologies for critical attributes of the dosage form.

Discriminating ability
The discriminating power of the QC method is an important attribute that should be challenged to ensure that the method is sensitive enough to capture formulation and manufacturing changes.

Based on some experienced case studies of BCS Class II compounds, in-vitro biorelevant dissolution can initially help on capturing the differences in clinical performance generated for two different formulations (e.g. granules from a spray dried intermediate). That discrimination can be achieved using pH shift approaches, taking advantage from the different dissolution profiles previously obtained at separated pH values. A change from pH 1.2 to 6.8, corresponding respectively to the gastric pH (e.g. simulated gastric fluid) and intestinal pH (fasted state simulated intestinal fluid) can led to different dissolution profiles. The decay in the dissolution profile can occur at diverse rates and the differences in the two formulations are pronounced. Based on this previous knowledge, a simpler QC method can be developed applying paddle apparatus and using acidic medium at pH 1.2 with an agitation speed of 50 rpm. Effective homogenization of the suspension in the vessel should be ensured to get almost full API released in about 30 to 45 minutes. The discrimination seen with the pH shift approach needs to be captured also by the QC method which procedure should then be validated.

The discriminative ability of QC method can be illustrated concerning a manufacturing variable, in this specific case, the compressing force during tableting process. For two tablet formulations, differing in the polymer composition of the originator spray dried intermediate, three increasing compression forces were tested. Results revealed the dissolution was slightly slower for middle and high hardness tablets.

Even any slight differences in disintegration time can potentially slightly retard dissolution for the harder tablets. The more friable ones tend to break faster and have a faster dissolution. The compression force itself is a potential critical process parameter because of its possible effect on dissolution. Other manufacturing variables such as lubrication blend time, excipient/API addition order, drying parameters and coating parameters can also be critical to understand dissolution differences between formulations.

The discriminating capability of QC method is also important to capture any possible changes in physical stability of tablets during storage. At six-month timepoint, an atypical decrease in dissolution profile is observed for tablets stored under accelerated conditions in a double polyethylene bag without moisture absorbers inside. When XRPD diffractograms of the originator spray dried intermediate matrix (stored at the same conditions and packaging configuration) were checked, they showed some incidences of crystallinity: API characteristic diffraction peaks are observed together with the halos attributed to the amorphous matrix.

The slower dissolution rate of aged tablet might be related to temperature and humidity conditions used during study. The temperature and moisture during storage also influences the rate of recrystallization from amorphous. This conversion produces a change in the dissolution profile. This dissolution and XRPD behavior was not observed for the same tablet formulation stored in a more conservative packaging configuration (inside PE flasks with moisture absorbers). This example shows the importance of changing the dissolution method to anticipate stability issues and give important information to the need for protecting the drug product in manufacturing/stability perspective.

Final remarks on QC method
The development of QC methods should be facilitated and accelerated by the previous knowledge gained with biorelevant development. Data for different pHs, the intrinsic properties of drug product intermediate, even the type of apparatus to use and the simplification of the medium are all important features to be taken in consideration for QC method development. In any case, final QC method should be challenged using batches produced when deliberate changes in manufacturing process are applied or a design space for critical attributes are defined. Stability issues should also be anticipated and addressed with stress tests. Additional to the historical in vitro data for release and stability, clinical performance of representative batches is also important to define any last update on the method and on the final specifications.

Conclusion
There must be an initial stage where the API properties are well studied and the impact of these properties on solubility and dissolution are assessed. Supersaturation studies are very important in this initial stage to help establishing drug concentrations used during method development and to help selecting the best formulation during polymer screening. The next step should get to know better solubilization strategies and the effective concentration of drug substance based on the most physiologically relevant methods.

Based on the knowledge acquired during the biorelevant screening methodologies, a faster and reliable QC method suitable for daily routine and release purposes should then be developed. 

References

1. Almeida e Sousa, L., Reutzel-Edens, S.M., Stephenson, G.A., Taylor, L.S., Assessment of the Amorphous “Solubility” of a Group of Diverse Drugs Using New Experimental and Theoretical Approaches, Molecular Pharmaceutics, 2015 12 (2), 484-495.
2. Fotaki, N., Vertzoni, M., Biorelevant Dissolution Methods and Their Applications in In Vitro In Vivo Correlations for Oral Formulations; The Open Drug Delivery Journal, 2010, Volume 4.
3. Dressman, J., Krämer, J., Pharmaceutical Dissolution Testing, 1st ed. Taylor & Francis Group, 2005.
4. Paprskářová, A., Možná, P., Oga, E. F., Elhissi, A., Alhnan, M. A., Instrumentation of Flow-Through USP IV Dissolution Apparatus to Assess Poorly Soluble Basic Drug Products: a Technical Note, AAPS PharmSciTech, 2016, 17 (5), 1261-1266
5. Gregory P. Marti, A Rational Approach to Development and Validation of Dissolution Methods, American Pharmaceutical Review, April, 2013.

---

Luís Sousa joined Hovione in 2015, having worked in the R&D Drug Product Development group, where he has been working on analytical development. Luis holds a PhD in Thermal Analysis at the UCL School of Pharmacy, London, UK and a Post-doc at Purdue University, Department of Industrial and Physical Pharmacy, West Lafayette, USA. The postdoc research focused on different solubility enhancing strategies for poorly water soluble drugs.

Mafalda Paiva joined Hovione in 2014 having worked in the R&D Drug Product Development group, where she has been working on analytical method development for new drug products and drug product intermediates, mainly focusing on dissolution and permeability methods. Mafalda has a MSc in Pharmaceutical Sciences and Quality Control, both from Pharmacy Faculty – University of Porto, with background in nanotechnology, pharmaceutical and analytical chemistry.

Pedro Serôdio joined Hovione in 2006, having worked in the R&D Drug Product Development group, where he has been working on the development of new analytical methods for drug product intermediates and oral solid dosage forms. Pedro has ten years of experience in the pharmaceutical quality control focused on analytical development area with strong expertise in HPLC/GC and dissolution methods for poorly-water soluble drugs. Pedro holds a Degree in Technological Chemistry from Faculty of Sciences of University of Lisbon with background in analytical chemistry and chromatographic techniques.