Discovery Formulation

Formulation Screening

In a previous role, I operated a discovery formulation screen to support Pharmacology, the main goal of which was the avoidance of unpleasant surprises for the Pharmacologists immediately prior to their experiments. The number of formulations screened was minimised and the procedure standardised as far as possible in the interests of speed and also to conserve valuable material especially bearing in mind that I was dealing with Medicinal Chemistry batches typically 20-100mg in size.  The standard procedure required 3-4mg solid. The ideal formulation was a solution but homogeneous suspensions were also acceptable for oral and ip dosing. A simple (visual) short term (1-2h) assessment of the physical stability of the formulations was included as a standard part of the protocol. A further option was a serial dilution test at low (gastric) and neutral pH. By default, none of this actually involved any analytical chemistry.

However, from time to time, projects outside of the scope of the standard protocol arose which necessitated making analytical (LC) measurements and the rest of this article is a discussion of the approaches needed to deal with some of the problems presented by formulation analysis.

Formulation Stability

Pharmacologists frequently request analytical work to support their dosing protocol. This may be to support repeated use of a dose form over a working day necessitating verification that the formulation is stable at room temperature for up to 8 hours. Both the chemical stability of the test compound and the physical stability of the formulation needs to be assessed. For a repeat dosing study, the Pharmacologist may wish to prepare a bulk formulation and then sub-aliquot into sufficient individual tubes for the duration of the study and store these frozen. Hence, a stability study would be needed to support this protocol by confirming compound and formulation stability for the relevant length of time at -20°C and for one freeze-thaw cycle. Freeze-thawing issues can include precipitation of solutions that subsequently won’t re-dissolve and loss of good suspension properties. Storage at 4°C can be a better option.

Formulation Solubility

Solubility measurement is often just an extension of formulation development in cases where simply observing full solution by eye is not reliable.

Although the high concentrations used in formulations in general allows for visual assessment of solubility a lot of the time, this is not always possible and, at lower concentrations, some instrumental analysis will be required to guide formulation development. As a rule of thumb, this might be for concentrations below 0.5mg/ml.

The measurement of the soluble fraction of drug in suspensions can yield interesting information for DMPK (absorption) interpretation. There may be significant differences in the soluble level of a drug between different suspension formulations that can only be revealed by analysis, e.g. of the filtrate (saturated solution) by LC-UV. Care should be taken in choosing a suitable filter type especially to ensure it has sufficient capacity and is not blocked by what can be a very heavy suspension.

The techniques described in my previous article on solubility are equally applicable to measuring solubility in formulations.

Analysis of Dosing Solutions

Qualification of actual dose is obviously good practice and generally straight forward by LC.

Due to the concentrations often encountered, dilution of the formulation may be required.  This can be done in blank vehicle but an organic solvent is often better. I would screen methanol, acetonitrile, acetonitrile:H2O [50:50] and DMSO as the diluent. It may be that insolubility of components of the vehicle being used will dictate the choice of solvent. Appropriate calibration standards should be analysed at the nominal target analyte concentration in the diluted sample, preferably from multiple independent weighings.

Analysis of Dosing Suspensions

The analysis of suspensions presents far more of a challenge. Homogeneity of the suspension is a key issue.

To measure the total drug concentration in a suspension it is necessary to first dissolve (and dilute) an aliquot. The same organic solvents that were suggested above can be screened. For the actual analysis, it is essential that multiple samples are taken (at least 3, preferably 6). The Coefficient of Variation (CV) on the peak area is a good measure of homogeneity.  It is possible for this to be as low as 1% for very homogeneous suspensions but usually it will be much higher.

A large degree of caution needs to be exercised when interpreting the results of analysis of residual dosing suspensions post biological studies. How reflective is the remaining suspension of the bulk of the formulation used in the study? It could feasibly be more concentrated depending on the properties of the suspension and how well mixed it was during the study. Also, how much time has elapsed between the study and the analysis and how has the sample been stored in the meantime? Has it been stored at -20°C for example or transported in dry ice? What effect might that have had? These are all likely to be largely unanswered questions but these factors could have a significant bearing on the analytical results.

Again the soluble fraction can be measured after filtration as discussed above.

Analysis of Cassette Formulations

Cassette formulations are frequently used in early DMPK screening. As many as 10 different compounds may be present in a single formulation. The foregoing discussion on the LC analysis of formulations of single compounds applies equally to cassettes but, of course, method development is more of a challenge and it is important to leave sufficient time to achieve an adequate separation. The methods are also likely to be longer but this is unlikely to be an issue.

Further Information

If any of the issues discussed in these articles are of interest to you, please feel free to contact me directly on for further information.