Send to

Choose Destination
Int J Cosmet Sci. 2016 Dec;38(6):576-580. doi: 10.1111/ics.12322. Epub 2016 May 3.

SPF and UVA-PF sunscreen evaluation: are there good correlations among results obtained in vivo, in vitro and in a theoretical Sunscreen Simulator? A real-life exercise.

Author information

GlaxoSmithKline (GSK), Rua Hungria, 1.240 - 4° andar, São Paulo, SP, Brazil.
Allergisa - Pesquisa Dermato-Cosmética Pele Ltda, Av. Romeu Tórtima n° 452, Campinas, SP, Brazil.
Medcin Instituto da Pele, Rua Atílio Delanina n° 178, Osasco, SP, Brazil.



Strategies to optimize the development of sunscreens include the use of theoretical sunscreen simulators to predict sun protection factor (SPF) and UVA protection factor (UVA-PF) and in vitro measurements of UVA-PF. The aims of this study were to assess the correlations between (1) SPF and UVA-PF results obtained in a theoretical sunscreen simulator with those observed in vivo (SPF and UVA-PF) and in vitro (UVA-PF) and (2) the results of UVA-PF observed in vitro and in vivo for products in different galenic forms containing or not pigments.


BASF Sunscreen Simulator software was used to evaluate the theoretical performance of formulations regarding SPF and UVA protection. In vitroUVA-PF and in vivoSPF were determined for all formulations. UVA-PFin vivo measurements were carried out only on products for which the galenic forms (compact foundations and lip balms) or the presence of dye or pigments could make the results of UVA-PFin vitro less reliable (due to a possible uneven film formation).


The results of the SPF calculated by the BASF Sunscreen Simulator presented a very good correlation with SPF observed in vivo in the absence of pigments (r = 0.91; P < 0.05) and a good correlation in the presence of pigments (r = 0.70; P < 0.05). The UVA-PF calculated by the BASF Sunscreen Simulator also exhibited a very good correlation with UVA-PF measured in vitro (r = 0.88; P < 0.05) for the formulations not containing pigment and a good correlation (r = 0.75; P < 0.05) for the formulations containing pigment. The correlation of same UVA-PF calculated by BASF Sunscreen Simulator with UVA-PF measured in vivo for the formulations containing pigment was r = 0.74 (P < 0.05), which is considered good. In addition, the measurements of UVA-PFin vivo presented a good correlation with the values obtained in vitro (r = 0.74; P < 0.05).


In the present study, the use of BASF Sunscreen Simulator and in vitroUVA tests showed good correlations with in vivo results and could be considered as valuable resources in the development of sunscreens.


claim substantiation in vivo/in vitro; colour cosmetics; formulation skin care/stability/texture/emulsions; suncare/UV protection

[Indexed for MEDLINE]

Supplemental Content

Full text links

Icon for Wiley
Loading ...
Support Center