The study procedure was illustrated in Figure 1 . In short, after the subject recruiting for two groups by meeting the inclusion criteria, informed consent was obtained from all participants. Subjects were then measured their skin aging parameters, skin physical properties and skin-aging related SNPs testing. After the data collection, data was analyzed to compare in two groups. A total of 278 healthy East Asian females were participated in the study. The participants were divided into two groups. The first group consisted of 141 Asian females (age range 28 to 76, mean age 47.1, SD 7.81) who were longitudinal FTE skincare product users with a written usage record (named Long-term FTE group, hereafter), having started regular daily use of FTE, twice a day in the morning and evening after cleaning their face, in their 20s or early 30s and continued this until the time of the present study. The second group consisted of 137 control females (age range 26 to 77, mean age 47.3, SD 9.51) who were in the same age range and had the same skincare habits and practices but no history of FTE product use (named Control group, hereafter), ensuring a consistent age distribution between the two groups. This research was performed in Tokyo, Nagoya, and Osaka, Japan, in May 2013. The examination room was maintained at a constant temperature and humidity (room temperature 20˚C ± 2˚C, relative humidity 50% ± 5%).

The average duration of FTE use in the Long-term FTE group was 13.1 years (SD 7.81) and the average age of starting FTE use was 28.9 (SD 7.49). The age of the subjects and the number of years of longitudinal FTE use are presented in Figure 2 and Table 1 . The number with history of longitudinal FTE use of more than 10 years was 88, more than 20 years was 22, and more than 30 years was 12 ( Table 2 ). The earliest age of starting FTE use was 6 years old; this subject had continued using FTE for 32 years since then. The longest duration of FTE use was 34 years.

None of the subjects underwent any type of esthetic treatment such as laser cosmetic procedures during the study period. This study was conducted in accordance with the tenets of the Declaration of Helsinki and approved by P&G Ethics Committee. Data acquisition and analysis were performed in compliance with protocols approved by the Ethical Committee of Global Product Stewardship in P&G Innovation Godo Kaisya (ethics approval number CT13-006). Written informed consent was obtained from all subjects prior to enrollment in the study.

The subjects washed their faces using the prescribed cleansing foam and then spent 20 min becoming accustomed to the environment of the measurement room at a constant temperature and humidity. Each subject’s face was photographed using an image capture system, which we named “Magic Ring” ( Figure 3 ), consisting of a portable image-capturing module that can be positioned on either the left-side or the right-side of the subject’s face [13] . Illumination is provided by a number of 5600-K light-emitting diodes (LEDs) mounted in the imaging module. A high-resolution complementary-symmetry metal-oxide-semiconductor (CMOS) digital camera, capable of generating 2592 (vertical) × 1944 (horizontal) effective picture elements (pixels), is also mounted in the imaging module, and the images that it collects are digitally transferred to a computer. The casing of the module encloses the area of skin under study and the mount containing the illumination source and the camera, so that the intensity of illumination and the distance between the camera head and skin surface are controlled to ensure the reproducible collection of images. A rectangular capturing guide appears on the monitor of the digital camera during image acquisition. A neutral 8.0 gray color board (GretagMacbeth GmbH, Munich, Germany) was used for white balancing of the camera. The region of interest (ROI) of the images was from the outer edge of the eyes to the cheek, and the following characteristic objects were extracted by measuring the contrast in the shape and pixels using an image analysis algorithm [1] . Wrinkles were defined as ≥ 5 mm in length, perimeter/length ratio ≤ 2.5, and circularity (perimeter²/area) ≥ 34. Total wrinkle area fraction was quantified as follows: total wrinkle area (pixels)/ROI (pixels). Hyperpigmented spots were defined as ≥ 5 mm² in area, color contrast DeltaE ≥ 3 compared with the surrounding skin region, and circularity (perimeter²/area) ≥ 20. Total spots area fraction was also quantified as follows: total hyperpigmented area (pixels)/ROI (pixels). As an index of skin surface roughness, total texture area fraction [total texture area (pixels)/ROI (pixels)] was quantified as ≤ 3 mm² in area, aspect ratio ranging from 0.5 to 2, and color contrast DeltaE ≥ 1.5, while pores were defined as ≤ 4 mm² in area, color contrast DeltaE ≥ 2, and circularity (perimeter²/area) ≥ 20, which differ from hyperpigmented spots in terms of size and circularity. Total pore area fraction was quantified as follows: a histogram-equalized image was filtered by fast Fourier transformation to remove one or more skin features (e.g., pigmented spots and moles). Then, skin feature segmentation was performed to extract images of follicles with one or more of the following features: threshold follicle area major diameter (200 - 1000 μm), width/length aspect ratio (0.3 - 1), and redness (a-value) contrast of more than 1.5 units compared with the surrounding skin. Facial skin color (lightness, redness, yellowness) was also measured in the region of interest. The mean values of the resulting data obtained by these evaluations were analyzed.

Skin hydration (water content), TEWL, and skin elasticity of the cheek were measured using Corneometer1 (Courage + Khazaka Electronic GmbH, Cologne, Germany), Vapometer^TM (Delfine Technologies), and Cutometer (Courage + Khazaka), respectively.

Using oral cavity swab samples ( Figure 4 ) and routine PCR sequencing, genotyping was performed on SNPs related to skin aging in Glutathione Peroxidase 1 (GPX1, SNP ID: rs1050450; CATCGAAGCCCTGCTGTCTCAAGGGC[C/T]CAGCTGTGCCTAGGGCGCCCCTCCT; Seq. ID Nos. 5 and 6), Superoxide Dismutase 2 (SOD2, SNP ID: rs4880; CAGCACCAGCAGGCAGCTGGCTCCGG[C/T]TTTGGGG-TATCTGGGCTCCAGGCAG (Seq. ID Nos. 3 and 4), and Matrix Metalloproteinase 1 (MMP1, SNP ID: rs1799750; GAATTGTAGTTAAATAATTAGAAAG[-G] ATATGACTTATCTCAAATCAATCCA (Seq. ID Nos. 1 and 2). These were SNPs reported to potentially confer a risk of skin aging, such as skin hyperpigmentation, anti-oxidation, or wrinkling. In brief, the genotyping was performed as follows: DNA extraction was performed using a Gentra Puregen Buccal Cell Kit following the manufacturer’s instructions (Qiagen, Valencia, CA, USA). The genotypes of the three SNPs (GPX1 rs1050450, SOD2 rs4880, MMP1 rs1799750) were determined by PCR-based TaqMan assays in the presence of two differently fluorescently labeled probes, which allow the detection of both alleles in a single reaction (Applied Biosystems Inc., Foster City, CA, USA). The specificity of the TaqMan assays was confirmed by Topo-cloning and capillary sequencing. Under optimized PCR conditions [volume 25 μl; using the TaqMan Universal PCR Master Mix as indicated by the manufacturer (Applied Biosystems Inc.), PCR: 95˚C for 10 min and then 50 cycles of 60˚C for 60 s and 92˚C for 15 s, performed on RotorGene 6000 (Qiagen)], reliable accurate results were obtained with at least 100 pg of chromosomal DNA. The presence or absence of SNPs in the particular genes was reflected by the following weighting: wild-type genotype = 1, heterozygous genotype = 2, and homozygous genotype = 3. The genetic skin score was determined, using the weightings corresponding to each polymorphism’s presence or absence in the biological samples of the subjects (e.g., MMP = 1, SOD2 = 2, GPX = 1).

SPSS 22 (IBM® SPSS® Statistics) for Windows 2010 was used for statistical analyses. Data were analyzed by nonparametric, paired, and unpaired tests including, if appropriate, analysis of (co-)variance and Welch’s t-test. Differences were considered to be significant at p < 0.05. In accordance with the above-described procedures, the parameters of texture, pores, wrinkles, skin tone, and spots were reverse-fitted on a scale from 0 to 1 (a higher score representing less texture, fewer pores and wrinkles, and a brighter tone), and the fitted scores of texture, pores, wrinkles, and skin tone were summed and defined as Total Aging Score. A higher Total Aging Score indicates more youthful-looking skin. In addition to these image analysis parameters, hydration, TEWL, and elasticity in the Long-term FTE group were compared to those in the Control group. Moreover, correlation coefficients (r) of the relationships of the skin imaging parameters of Total Aging Score, texture, pores, wrinkles, tone, and spots with the genotypes of the GPX1, MMP1, and SOD2 SNPs were determined.

We measured various facial skin optical and physical parameters in both the Long-term FTE group of 141 female volunteers and the Control group of 137 female volunteers. The facial skin hydration in the Long-term FTE group was significantly higher while TEWL was correspondingly lower than those in the Control group ( Table 3 ). In parallel with this, the values of six fit-scaled skin aging-related parameters (a higher score representing more youthful-looking skin), namely, texture, pores, wrinkles, tone (lightness), spots, and the total aging score, were also significantly higher than those in the Control group ( Table 4 ). In addition, facial skin in the Long-term FTE group had significantly lower levels of redness and yellowness than that in the Control group ( Table 3 ). Moreover, elasticity in the Long-term FTE group was significantly higher than that in the Control group ( Table 3 ). This trend of youthful-looking skin in the Long-term FTE group was observed in all age groups, including those in their 20 s/30 s, 40 s, 50 s, and 60 s/70 s ( Table 3 , Table 4 ).

†: Significant difference from the corresponding Control group (p<0.05).

*: Significant difference from those in their 20s/30s (p < 0.05). ⁑: Significant difference from those in their 40s (p < 0.05). ⁑*: Significant difference from those in their 50s (p < 0.05). ⁑⁑: Significant difference from those in their 60s (p < 0.05). †: Significant difference from the corresponding Control group (p < 0.05).

All skin aging parameters of texture, pores, wrinkles, skin tone (lightness), spots, and Total Aging Score in those in their 20s/30s, 40s, and 50s/60s/70s were significantly larger and better in the Long-term FTE group than in the Control group. Notably, based on the Total Aging Score, no aggravation of skin aging was observed in the long-term FTE group after reaching their 40s, and there were no significant differences in Total Aging Score in those in their 50s, 60s, and 70s ( Figure 5 ).

The genotypes of three representative SNPs in GPX1, MMP1, and SOD2, believed to be related to skin aging, were determined in the two groups. First, the associations of these three genotypes with skin aging-related parameters were determined in the Control group ( Table 5 ).

All three tested genotypes of GPX1, MMP1, and SOD2 showed significant associations with skin aging parameters and Total Aging Score in the Control Group ( Figure 6 ). Meanwhile, the GPX1 genotype was associated with the individual variables of spots and pores, and the MMP1 genotype was significantly associated with wrinkles, texture, and tone. Furthermore, the SOD2 genotype was associated with pores. These findings suggest that these three genotypes affect skin aging chronically. In contrast, no significant associations of these genotypes with skin aging-related parameters were observed in the Long-term FTE group. Notably, there were no significant differences in genotype frequencies of the SNPs in GPX1, MMP1, and SOD2 between the Long-term FTE group and the Control group ( Figure 7 ). The above findings suggested that most of the skin aging-related parameters measured in this study were significantly better in the Long-term FTE group than in the Control group; however, no differences were observed in the frequency of skin aging-related DNA genotypes between the two groups. Representative facial photos of the Long-term FTE group and the Control group in age 30s, 40s, 50s and 60s were shown in Figures 8-11 .