Pharmaceutical Applications of microCalorimetry (II)

Stability Testing



When a promising new compound enters into the early development phase, studies are carried out to determine properties of the drug under various chemical and physical conditions as part of the search for a formulation suitable for use in clinical trials.   Measuring both the chemical and physical stability of the drug candidate is an important part of the early phase development process since the shelf life of the compound will ultimately depend on the stability of the compound in the formulation.   The information gathered in early phase stability screening tests is useful to late phase development scientists responsible for the development of a final product formulation.

The most common method for characterization of stability for a new drug candidate involves preparing solutions under various “stressing” conditions such as low and high pH, and in the presence of oxidizing agents such as hydrogen peroxide1.  Solid samples are stressed at various humidity levels, and temperatures.   The solutions are analyzed via HPLC to determine the levels of degradation over time, and the solid samples are analyzed by methods such as powder x-ray diffraction to determine if significant physical changes occurred.  The solids are also dissolved and analyzed for chemical stability using reliable techniques such as HPLC and spectroscopy.

Drug candidates, by design, are generally stable materials.  Therefore, the stability testing mentioned above is generally carried out at relatively high temperatures to accelerate degradation to the point that analysis can occur within a reasonable amount of time after the studies are initiated.   While this can result in errors if exact kinetics is the goal, due to non-Arrhenius reaction mechanisms for example, the general practical goal of these stability studies is to screen the compounds for relative reactivity, not to obtain exact reaction kinetics.  For example, to determine if the drug candidate is susceptible to hydrolysis in acidic conditions, basic conditions, or both.

Numerous authors have proposed using microcalorimetry as a method to characterize the stability of pharmaceutical compounds2,3,4.  Advantages such as the ability to monitor reactions at lower temperatures, due to the high sensitivity of microcalorimetry, and the ability to monitor both chemical and physical processes are proposed and examples given. 

It is true that modern isothermal microcalorimeters are capable of measuring extremely small heat flows.  Based on assumptions about molecular weight, enthalpy, and reaction mechanism Angberg5 et al stated it is possible to detect degradation rates of only .68% per year.  However, the actual signal that can be reliably measured depends on the rate of reaction and the heat of reaction.  In a practical lab setting, the ability to measure very slow rates my require that a relatively large sample size be used, often in the range of 1 or more grams.

In the early phases of development such large qualities of the drug candidate are generally not available.  HPLC studies require just a few milligrams of material for the various stressing conditions, making it unlikely that microcalorimetry can be expected to be seen as practical alternative, even when the advantage of being able to carry out reactions at or near room temperature is taken into account.

However, even in the cases where small amounts of sample are available, microcalorimetry can be a very useful tool for screening relative reactivity, if higher temperatures, such as those used with traditional HPLC based methods, are employed.  The microcalorimetry technique requires no method development, as solutions, suspensions,  or solids are placed directly into the instrument for measurements.  Relative reactivity data can be obtained without the need to, for example, optimize an HPLC gradient method, or filter insoluble reaction products that can form as the drug candidate degrades.  

The microcalorimetry data obtained can be used to help optimize the protocol for traditional stability screening of the drug candidate.  For example, the microcalorimetry results can yield approximate relative reactivity rates under the various stressing conditions, which can be useful for planning how high of temperatures, and how much time would be required before enough degradation takes place to obtain reliable HPLC results.  In some cases, if the rate and heat of reaction are sufficiently large, microcalorimetry data can be obtained in a short period of time, at a number of temperatures, and be used to determine if, and at what temperatures, non-Arrhenius kinetics take place.

In the example outlined below emphasis will be given on practical and rapid microcalorimetry screening techniques versus exact thermodynamic and kinetic studies.  For the later the reader is directed to references 2,3, and 4.


Example Data

Solids, suspensions, or solutions can be studied using microcalorimetry without the need for any method development.  In fact, the large sample sizes that can be used with microcalorimeters (versus traditional aluminum DSC pans) permits adding a small vial of a saturated salt solution into the measurement chamber.  Figure 1 shows such an arrangement, which makes it possible to study chemical and physical processes taking place in a solid sample over a range of temperatures and humidity conditions.



Figure 1.  Using a saturated salt solution to accelerate degradation.

In a typical stability screening experiment a solution, suspension, or solid is placed in the microcalorimeter and the thermal activity (heat flow) is monitored.  The basic assumption is that the rate of heat production, at any given temperature, is proportional to the rate of chemical and/or physical processes taking place in the sample.

 A simple example would be conversion of A to B:

A -------> B

Thermal activity = dq/dt = DH * dn/dt


Where q is the heat (e.g., calories), t is time in seconds, DH is the heat of reaction, and n is the number of moles of B formed.  Therefore, the output of the calorimeter is directly proportional to the rate of formation of B at any given time t.

It was mentioned above that using higher temperatures makes it possible to use smaller sample sizes during a microcalorimetry measurement.  The example in figure 2 shows data collected using scanning microcalorimetry from 25 up to 90 °C.  The solutions were relatively dilute, 10mg per ml, suggesting the same samples used for the HPLC stability testing could be used for the microcalorimetry measurements. 

Notice that at low temperatures, there was no sign of thermal activity associated with any chemical processes for either the pH 6 or pH 10 solutions.  However, as the temperature was raised very slowly (0.2 °C/min) significant non-zero thermal activity was observed.  As the temperature continued to rise, the thermal activity continued to increase, as expected for chemical reactions. 

Observe that under basic conditions the reaction was endothermic.  In acidic media the reaction was exothermic.  This is useful information when attempting to understand the reaction pathway (enthalpy versus entropy driven).



Figure 2.  Reactivity as a function of temperature and pH (exo up)


In traditional HPLC stability screening studies the common practice is to assume the UV response for all degradation products formed are equal.  While this certainly is not true, and can be in error by orders of magnitude, there generally is not enough time available to run prep level HPLC to isolate each product so individual extinction coefficients can be determined. 

Using a similar assumption, that the heats of reaction in figure 2 were the same for both solutions, and remembering that the thermal activity is proportional to rate, one can conclude the apparent rate of reaction is significantly faster at pH 10 than at pH 6 at any given temperature.  As all common heats of reaction are within one order of magnitude of each other, the microcalorimetry method is in fact potentially more accurate than HPLC for measurement of relative kinetics of degradation.

It can be shown that graphing the logarithm of the thermal activity versus the reciprocal temperature results in an Arrhenius plot of the microcalorimetry data6.  Figure 3 shows this type of plot for the solutions from figure 2.  Notice that at pH 6 there appears to be a single activation energy, while at pH 10 there appears to be at least two reaction pathways with different activation energies.  As mentioned earlier, this information can be very useful in the design of traditional HPLC stability screening studies since at pH 10 the analyst would quickly realize the need to qualify results obtained at low versus higher stressing temperatures.



Figure 3.  Arrhenius plots of pH 6 and pH 10 microcalorimetry results.

Having generated scanning thermal activity versus temperature data as shown in figure 2, it would be possible to estimate the thermal output expected at any given temperature.  With his information it would be possible to define good operating parameters (sample size and temperature) to carry out more exact isothermal kinetic studies using microcalorimetry.


Microcalorimetry was shown to be a practical tool to complement traditional stability screening of pharmaceutical compounds.  The ability to test, for example, numerous pH conditions within less than a day can yield useful information to guide the design of more traditional HPLC based assays.  In many cases fundamental information can be derived from the same data, including whether degradation is enthalpy or entropy driven, as well as if HPLC data generated at elevated temperatures can reliably be used to predict relative reactivity at lower temperatures. 

It was pointed out that the microcalorimetry method is very simple, and basically always the same.   This  makes it possible to collect significant amounts of stability information, in a very short time, early in the development phase of API characterization.  It should also be pointed out that in late phase development the microcalorimetry technique can be used to rapidly determine that if any proposed processing conditions are altered (for example salt form isolation from basic versus acidic media), what effect the change can be expected to have on the stability of the drug candidate.

For more information contact:


Quoted References

  1. Karen M.Alsante, Linda Martin, and Steven W.Baertschi, A Stress Testing Benchmark Study, Pharmaceutical Technology (February 2003) 60 – 72.
  2. Selzer, T. / Radau, M. / Kreuter, J., Use of isothermal heat conduction microcalorimetry to evaluate stability and excipient compatibility of a solid drug.  International Journal of Pharmaceutics, vol. 17, Issue 2, (1998) 227-241.
  3. Skaria, C.V. / Gaisford, S. / O'Neill, M.A.A. / Buckton, G. / Beezer, A.E., Stability assessment of pharmaceuticals by isothermal calorimetry: two component systems.  International Journal of Pharmaceutics, vol. 292, Issues 1-2.  (2005) 127-135.
  4. Phipps, M. A. et alApplication of Isothermal Microcalorimetry in Solid State Drug Development.  Pharmaceutical Sciences & Technology Today vol. 3 no. 1,  (2000) 9 – 17.
  5. Angberg, M. (1995) Lactose and Thermal Analysis with Special Emphasis on Microcalorimetry.  Thermochimica Acta 248,  (1995) 161 – 176.
  6. Hansen, L. et al. Kinetics of Drug Decomposition by Heat Conduction Calorimetry.  Pharmaceutical Research, Vol. 6, No 1 (1989) 20 – 27.