In my recent
blog post, I discussed the p53 gene, and experiments testing sunscreens effect
on p53 in mice. In this blog post, I will discuss a study testing sunscreen’s
relation to p53 protein production done on humans. From 1992-1996, the NambourSkin Cancer Prevention Trial conducted a study to test the association of time
spent outdoors and the expression of p53, as well as its relation to the use of
sunscreen. Researchers predicted that sunscreen counteracts p53 production, and
carried out their research on over 160 random participants for 6 months. There
were four different treatment groups: 1. Application of SPF 16 sunscreen daily
+ beta-carotene supplementation, 2. Application of SPF 16 sunscreen daily +
placebo tablets, 3. Beta-carotene only, 4. Placebo only. Beta-carotene is a
type of pigment called a carotenoid, and produces Vitamin A, which helps
prevent cell damage. It’s thought to be possibly effective for people who are
sensitive to sun exposure in order to help prevent sunburn, but the exact
efficiency is not known.
Figure 1. Proportion of p53-positive cells in the
whole epidermis (n=139) of participants’ dorsal hand.
Based on the above
graph, it seems like there’s a high frequency of individuals with <5% of
their cells being p53 positive, and that the frequency for p53 positive cells
becomes relatively low after 25%. What this graph leaves unclear is what the
frequency represents—is it measured by number of individuals who had that
amount of p53 positive cells? Does it carry forward? For example, if a person
is 80% positive, will they be counted into the frequency for 0-5, 5-10, 10-15,
etc. in addition to 75-80? If so, this could be why the data looks so skewed
toward a higher frequency for lower levels of p53 positive cells. The study
says that 61% of participants had over 5% of p53 cells and 22% had over 20%,
but 17% had <5%, and so I would think that the frequencies would correlate
with these numbers but it doesn’t seem like it does. Not to mention, does it
make a difference that they took biopsies from the hand and not other areas of
the body? Different parts of the body will most likely have different
susceptibilities to sunburn or sensitivity to UV exposure due to their cell
makeup, so does using the hand also skew the data
Table 1 shows
that there is a significant difference in having p53 positive cells when
comparing males to females (p<0.01), as well as females smokers to male
smokers (p<0.05). One thing the researchers could control if they were to
repeat the study would be to not accept past smokers, whose history could affect
the results of the trial. It is indicated that the subsequent analysis were
adjusted for sex and smoking status, which could also affect the data by making
p53 percentages higher even though they might not be attributed to UV exposure.
According to the table, it seems like other variables such as age, skin color,
skin type, and cancer history did not significantly influence the proportion of
p53-positive cells.
Table 3 (above) shows
the effect of sunscreen on proportion of p53-positive cells. The study found
that when sunscreen is the only variable used to compare participants, there is
no significant difference in proportion of p53 positive cells. But they also
conducted a multivariable model to take into account time spend outdoors, sex,
smoking status, skin type, and skin color.
Overall, the
researchers concluded that sunscreen is independently associated with lowering
p53 protein production, and I somewhat agree with their argument. What makes
the data difficult to interpret is the amount of external factors that could
contribute. Also, the numbers in Table 3 suggest that there is a significant
difference between the groups, but between never wearing sunscreen and wearing
sunscreen 7 days per week, there appears to be the same percentage of p53
positive cells, but when comparing sunscreen use from 1-6 days, it appears that
there is a lower proportion in the 5-6 range, which is confusing. Another downside
to the experiment is that it was done in Australia, which is known as the “skin
cancer capital of the world,”(3) and so another variable could be location of the
homes of these individuals. The paper also did not differentiate results for
the four different groups, and did not go into depth about why they also used
beta-carotene. Wouldn’t that add another variable other than sunscreen when
analyzing data? As a result, I don’t fully trust the conclusion of the paper
because there are so many variables that could contribute to the experiment,
and there are also parts of the experiment that aren’t fully explained, which
makes it difficult to interpret the data and propose conclusions.
Sources:
1. Jolieke
C. van der Pols, Chunxia Xu, Glen M. Boyle, Peter G. Parsons, David C.
Whiteman, and Adele C. Green. Expression of p53 Tumor Suppressor Protein in
Sun-exposed Skin and Associations with Sunscreen Use and Time Spent Outdoors: A
Community-based Study Am. J. Epidemiol. (1 June 2006) 163 (11): 982-988
first published online April 19, 2006 doi:10.1093/aje/kwj137
2.
"Beta-carotene." U.S
National Library of Medicine. U.S. National Library of Medicine, 19 July
2011. Web. 26 May 2014. <http://www.nlm.nih.gov/medlineplus/druginfo/natural/999.html>.
3.
"SunSmart." Cancer
Council Australia. Cancer Council Australia, 19 Nov. 2013. Web. 25 May
2014.
<http://www.cancer.org.au/policy-and-advocacy/position-statements/sun-smart/>.
