Solubility in buffers at physiological pHs is an essential property of a compound that determines the feasibility of delivering it into the body via oral or parenteral routes. There is no absolute scale upon which to judge a compound’s solubility as a predictor of development potential. The assessment involves a number of factors, including:

• what is the intended dose?
• what is the dose route?
• what salt forms will be used?
• what is the physical form of the dosing formulation?

Typically a solid tablet or powder-filled capsule is the preferred dose form for oral administration. Liquids or suspensions may be preferred for certain indications, such as drugs intended for pediatric use. Development of such dose forms usually occurs in the clinical stages of development. For most preclinical purposes a solution or a suspension of the compound is preferred.

In order for a compound to be absorbed, it must be dissolved in the physiological medium surrounding it. Simple aqueous buffers are frequently used as surrogates for this medium. If a compound is soluble in artificial gastric fluid at pH 2 and in phosphate-buffered saline at pH 7.4, it is very likely to be soluble in the gastro-intestinal tract. However, several drugs with good oral absorption do not meet these criteria. Therefore, more complex dissolution media, such as artificial intestinal fluids simulating the fed or fasted state, are frequently preferred as more physiologically relevant media for measuring solubility.

Stability of a compound in physiological media is also an important determinant of its development potential. Compounds that undergo significant degradation during short-term exposure to physiological buffers are unlikely to reach the intended therapeutic target in vivo after oral or intravenous administration. On the other hand, many drugs are administered as precursors (prodrugs) of the therapeutically active compound and depend on timely conversion to the active compound for their therapeutic effect. In those cases, the stability must be such as to allow efficient administration to the body, but not prevent conversion to the active moiety in a timely fashion. The kinetics of NCE breakdown or prodrug conversion can be studied in vitro in buffer systems or tissue extracts that mimic the physiological processes expected in vivo.

pK_a and LogP are more fundamental measurements from which LogD at any pH may be calculated. They can be determined from potentiometric titration curves produced with various volume fractions of organic solvent. The measurements can be technically challenging although the introduction of automated equipment and non-linear curve fitting algorithms has made the determination of these parameters less tedious than it was in the past. However, such methods are confined to certain relatively well-behaved compounds and are by no means universal. Nevertheless, under appropriate conditions determination of LogP and pK_a provides significantly more information about an NCE’s behavior over a broad range of pHs.

LogD is considered one of the most useful basic measurements that can be made. It is usually measured by adding the compound in a small amount of polar organic solvent, such as DMSO or methanol, to an organic phase, typically octanol, and then extracting the octanol with buffer. The phases are mixed thoroughly, allowed to separate, then the compound concentration in each phase is determined. The ratio of the organic phase concentration to the buffer phase concentration is the partition coefficient. The base 10 logarithm of this ratio is know as LogD. Thus positive LogD values imply greater partitioning into the organic phase, while negative LogD values imply greater partitioning into buffer. The range of this type of assay typically runs from LogD values of about –3 to about +5, depending on analytical sensitivity. The precision of these measurements is typically about +/- 0.5 log units.

LogD is a pH-dependent value for compounds containing ionizable groups. The un-ionized form of the compound is considered to be the form that partitions into the organic phase. Thus the partition coefficient will depend on equilibria between the ionized and un-ionized forms of the NCE in the buffer phase and the relative solubility of the un-ionized form of the compound in the organic phase and the buffer. Typically, LogD values are measured at physiologically relevant pHs, such as 7.4 for blood and 6.5 for the intestinal lumen. LogDs between about 2 and 4 are considered favorable predictors of good NCE absorption and distribution. Extremes of LogD are to be avoided except in unusual circumstances, such as compounds intended for dermal applications [high LogDs imply better skin penetration] or treatment of gut luminal infections [low Log Ds imply poor intestinal permeation and, therefore, better retention of the drug in the gut lumen].

It is true that the ADME properties of a compound depend upon its physicochemical properties. Another way of saying this is that it is often very difficult to get acceptable intestinal absorption with an insoluble compound.