Wednesday April 29 2009
Before (top) and six months after using the product (bottom)
Scientists say that, “a skin cream sold on the high street has been clinically proven to reduce fine wrinkles and improve the appearance of weathered skin,” The Times reported. It said that a trial of 60 volunteers with signs of sun-damaged skin, found that the cream from the chemist Boots, helped to reduce signs of ageing. The newspaper reported that 70% of people who used the cream for a year had significantly fewer wrinkles compared with volunteers using a placebo. This popular story was covered by most UK national newspapers.
This new study was a double-blind randomised controlled trail and has been welcomed for "raising the bar" (The Guardian) on the kinds of tests cosmetic companies should do before making claims for their products. Although the study has a few limitations in how it was reported, and there was no significant difference in benefit at six months between the test group and a group given a placebo, there is a clear trend towards benefit. If the cream were to be regulated as a medicine, the next step would be larger trials that confirmed the affect.
Where did the story come from?
The research was carried out by Dr Rachel Watson and colleagues from the Dermatological Sciences Research Group at the University of Manchester. One of the authors is employed by Alliance Boots Ltd, the manufacturer of the cream. Sources of funding are not reported. The study was published in the peer-reviewed British Journal of Dermatology.
What kind of scientific study was this?
The researchers say that very few over-the-counter cosmetic ‘anti-aging’ products have been subjected to rigorous testing. This was an investigation of one such product involving a double-blind randomised controlled trial and an ‘in vivo’ patch testing study. Both parts of the study tested the same product: No.7 Protect and Perfect Intense Beauty Serum manufactured by Alliance Boots.
This product consists of a water in silicone emulsion. It contains glycerine and other emollients, plus a number of ‘anti-ageing’ ingredients. These ‘anti-ageing’ ingredients include extracts from plants (Panax ginseng, Morus alba, Lupinus alba, Medicago sativa), as well as small pieces of protein called peptides (palmitoyl oligopeptide, palmitoyl tetrapeptide-7) as well as vitamin derivatives (sodium ascorbyl phosphate, tocopherol and retinyl palmitate). Retinyl palmitate (a retinoid) is thought to be the active ingredient and is a synthetic ester derived from vitamin A. More potent, prescription-only, retinoids are used to treat mild to moderate acne. The researchers used one of these, all-trans retinoic acids (ATRA) as a comparison (a positive control), in the ‘in vivo’ patch test study. ATRA is known to cause the skin to produce fibrillin-1, which itself predicts the formation of collagen. More collagen should result in skin rejuvenation.
The researchers recruited 10 people (four men and six women, aged 61 to 76 years) whose skin that had been photoaged (sun damaged). Each person underwent a patch test then had small amounts (20 microlitres) of the No.7 product or a control ‘vehicle’ cream applied to separate small areas (6mm in diameter) of clean, photoaged skin on their forearms. The areas were then covered over with dressings. Similarly sized areas of skin had no cream applied but were covered over to act as untreated controls. The control and the test product were applied to skin on the first, fourth, and eight days of the test. In addition, an all-trans retinoic acid (ATRA) cream (Retin-A cream produced by Janssen-Cilag Ltd. containing 0.025% retinoic acid) was applied to an area of untreated skin on the eighth day and left on the skin for four days. The vehicle cream represented a negative control, while the ATRA cream represented a positive control.
Twelve days after the test began, skin samples (biopsies) were taken from the four forearm areas (the No.7 test product area, control cream treated area untreated area, and the ATRA positive control area). The researchers used a standard technique using antibodies to look at how much of the fibrillin protein was present in thin slices from these skin samples, scoring the levels from zero (lowest) to four (highest).
The randomised-controlled trial involved 60 recruits with photo-aged skin (11 men and 49 women, aged 45 to 80 years). In it, the researchers compared the No.7 Protect and Perfect Intense Beauty Serum and a control (vehicle) cream without the ‘anti-ageing’ ingredients.
Participants were randomly assigned to receive the test or control cream, each in identical packaging so that neither the volunteers nor the researchers knew which they were receiving. They applied the cream to their face and the backs of their hands, wrists and forearms every evening for six months.
The participant’s faces and the backs of their hands were examined by researchers at the start of the study and at one, three and six months. The researchers looked for fine lines and wrinkles, areas of abnormal colouration, overall level of signs of ageing of the skin by the sun and roughness of the skin to the touch. All four of these aspects of the skin were rated from zero to eight. Zero indicated the least evidence of photoageing or fine lines and wrinkles, no areas of abnormal pigmentation or totally smooth skin. Eight indicated the most severe photoageing or fine lines or wrinkles, severe abnormal pigmentation or very roughened skin. Scores for these four aspects of skin appearance were compared in the test and control groups.
Twenty-eight of the volunteers (13 from the control and 15 from the test group) agreed to provide skin biopsies from their wrist at the start and end of the study. The researchers looked for the levels of fibrillin in these tissue samples. All volunteers were monitored for serious side effects of the treatment during the treatment period and for 28 days afterwards.
After the six-month initial trial period, all volunteers were offered and used the No.7 product for a further six months. The researchers compared 12-month results from the treated volunteers with what would be expected in the control group by using statistical modelling based on the vehicle response at six months to predict the expected outcome at 12 months.
What were the results of the study?
The patch testing showed that the skin treated with the test cream contained more fibrillin than the untreated control skin. Skin treated with the all-trans retinoic acid cream (positive control) had similarly increased levels of fibrillin but treating the skin with vehicle cream alone did not significantly affect the level of this protein.
On a scale of zero (lowest) to four (highest), the average level of fibrillin in the skin was rated as 1.3 for the untreated control skin, 1.7 for the vehicle cream, 2.5 for the all-trans retinoic acid cream, and 2.6 for the test product.
After six months, 43% of people in the randomised-controlled trial using the test product, showed an improvement in facial wrinkles compared to the start of the study, this was a statistically significant increase. After six months, 22% of those using the control cream had improvements in facial wrinkles compared to the start of the study, but this was not statistically significant. The difference in the proportion of people with improved appearance of wrinkles between the tested and the control cream (21% or one in five) was also not statistically significant.
When the researchers gave all participants the test cream, they found that at 12 months 70% of the people showed an improvement in facial wrinkles compared to the start of the study. Statistical modelling predicted that 33% of people would have shown improvement in their wrinkles if they continued to use the control (vehicle) cream for 12 months. The proportion of participants who showed improvement in this model (37%) was statistically significant.
The improvement in the test group was described as “clinically significant”. Skin samples showed that wrist skin treated with the test product for six months contained more fibrillin than control (vehicle) treated wrist skin.
The researchers did not observe any other benefits of the test product over the control. For example, there were no improvements in the amount of abnormal colouration (mottled dyspigmentation) of the skin and both the test and control products produced similar improvements in skin roughness from the start of the study.
What interpretations did the researchers draw from these results?
The researchers say that they have shown “for the first time that a commercially available over-the-counter anti-ageing product improves the appearance of facial wrinkles when used in the long-term”. They say that the improvement is associated with “restoration of fibrillin-1” and that this further supports the use of fibrillin-1 as a marker for assessing the efficacy of similar products.
What does the NHS Knowledge Service make of this study?
This randomised-controlled trial used double blinding to reduce the risk of bias for the six-month outcomes and a validated scale to provide as objective a measure of the actual skin appearance as possible. There are aspects of the reporting and the statistical analysis in this study of note when:
- The characteristics of the treated and control groups before the study began are not described. Even in randomised trials, there can be important differences between groups that can sometimes partly explain the results.
- The results and significance between the cream and control groups at six months are not fully reported. If these had been provided, it would be possible to judge how close to a significant result the six-month trend came.
- The assumptions underlying the extrapolation of the control group results to 12-months are not reported.
- The dropout rates are not reported, nor is any measure of compliance with the cream between the groups.
- The study was small, therefore limiting its ability to detect a true difference in outcomes.
- Any minor adverse effects, such as skin redness or irritation, common with the more potent topical retinoids, are not reported.
As all participants knew that they were receiving the test cream in the ‘open’ part of this trial from six months to 12 months, this has effectively compromised the benefits of blinding. Although there are understandable reasons why this was done, it remains to be seen what benefits there may be if a larger trial was conducted for the full 12 months.