Correlation of in vitro SPF values from two laboratories

Abstract

Keywords: Correlation, R – Squared, Sunscreen Products, SPF

Objectives: To compare and correlate in vitro SPF values obtained from two different laboratories

Methods: Six sunscreen products in markets label with SPF in a range of 7 - 30 (7, 15, 20, 25, 30 and 30 on label), in container closer systems proposed for marketing were evaluated for their SPF values. The in vitro SPF values of each sample were determined in two different laboratories; i.e. Chulalongkorn University Drug and Health Products Innovation Promotion Center (CU-D-HIP) and a company laboratory. Both laboratories utilized the same equipment, the Optometric SPF 290 S Analyzer, and the same substrate, Transpore^® tape. The only different step in these two laboratory procedures was product weight per square centimeter that applied on the substrate; i.e. around 2  mg/cm² vs. 13 mg/cm² of sample on the substrate according to CU-D-HIP vs the company laboratory procedure, respectively. The obtained SPF values were correlated using the linear regression. Two COLIPA reference sunscreen formulations were prepared and employed in this study

Results: The SPF values from the company laboratory were about 1.5 times higher than that from CU-D-HIP. The SPF values from the same products obtained from two laboratories, showed linear correlation with high deviation (R² » 0.60 – 0.65). Sunscreen products containing Diethylamino Hydroxy Benzolyhexyl Benzoate (DHHB) caused high deviation from linear regression since these products were either non-homogeneous or viscous. Product with non-homogeneity or with high viscosity directly affects surface smoothness and weight uniformity of sample when applied on the substrate. Therefore, three sunscreen products were withdrawn from this correlation study. Two COLIPA reference sunscreen formulations were evaluated by the company laboratory under a hypothesis that COLIPA standard sunscreen formulation should give the SPF values equivalent to their specified values. The remaining three sunscreen products and two COLIPA reference sunscreen formulation showed good linear correlation with R² around 0.95 - 0.96.

Conclusion: Small differences in in vitro SPF evaluation procedures led to obvious different in results. However, results from different procedures can be correlated using simple linear regression technique.