LogP determination as part of biocompatibility screening

The octanol-water partition coefficient (logP) is an important predictor in toxicology for assessing bioaccumulation risk in the human body and the environment. Therefore, we experimentally determined the logP of composite monomers in vitro with the shake-flask method and compared them with in-silico computed logP values.

Monomers BisEMA3, BisEMA6, BisEMA10, BisGMA, HEMA, TEGDMA, and UDMA were added to a pre-saturated 1-octanol/water system. The mixture was equilibrated for 24, 48, and 72 h. Next, the samples were analyzed using UHPLC-MS/MS, and the logP was calculated. In addition, three in-silico models (KOWWIN, ACD/LogP, XLogP3) were used to predict the logP. The relationship between experimental and computed values was explored using linear regression and mean absolute error (MAE).

Since there was no statistical difference between the logP values at different time points (two-way ANOVA, p > 0.05), the final equilibrium was considered to be reached after 24 h. At the 72 h time point, the mean experimental logP values were ranked from highest (4.05 ± 0.09) to lowest (0.09 ± 0.03): BisEMA10 >BisEMA3 >BisEMA6 >BisGMA>UDMA>TEGDMA>HEMA. The in-silico predicted values consistently showed overprediction bias. This trend was more noticeable for more hydrophobic monomers.

Determination of logP is important for biocompatibility assessment. High logP values (≥5) warrant further investigation. It is advisable to determine logP values in vitro, as current in-silico models are not yet reliable enough. The in-vitro values were lower than the calculated values, which was reassuring in terms of bioaccumulation potential. In-silico models should be improved using large datasets of experimentally determined logP values.