PURPOSE: To predict the crystallization time of amorphous solid dispersions by controlling the combined effect of temperature and moisture content. METHODS: The authors exposed amorphous samples of spray-dried API and Hydroxypropylmethylcellulose Phtalate to various temperature and humidity conditions below and above the glass transition temperature (Tg) until crystallization of the API was observed. The crystallization of API was detected by XRPD, while the T ( g ) and the water absorption by the amorphous dispersion are quantified by mDSC and water sorption analysis. RESULTS: Extrapolation of the data obtained at a temperature above T ( g ) to conditions below T ( g ) gives only a qualitative trend. By contrast, in conditions below T ( g ) the logarithm of onset of crystallization time was shown to vary linearly with the T ( g ) /T ratio. A statistical analysis shows that the data obtained in the highest temperature/humidity conditions, for which the onset of crystallization is below 3 months, can be extrapolated over 15 months. CONCLUSIONS: The proposed methodology can be used as a stress program to predict long-term stability from a relatively short observation period and to design appropriate temperature and humidity conditions for long-term storage to prevent crystallization.
PURPOSE: To predict the crystallization time of amorphous solid dispersions by controlling the combined effect of temperature and moisture content. METHODS: The authors exposed amorphous samples of spray-dried API and Hydroxypropylmethylcellulose Phtalate to various temperature and humidity conditions below and above the glass transition temperature (Tg) until crystallization of the API was observed. The crystallization of API was detected by XRPD, while the T ( g ) and the water absorption by the amorphous dispersion are quantified by mDSC and water sorption analysis. RESULTS: Extrapolation of the data obtained at a temperature above T ( g ) to conditions below T ( g ) gives only a qualitative trend. By contrast, in conditions below T ( g ) the logarithm of onset of crystallization time was shown to vary linearly with the T ( g ) /T ratio. A statistical analysis shows that the data obtained in the highest temperature/humidity conditions, for which the onset of crystallization is below 3 months, can be extrapolated over 15 months. CONCLUSIONS: The proposed methodology can be used as a stress program to predict long-term stability from a relatively short observation period and to design appropriate temperature and humidity conditions for long-term storage to prevent crystallization.