Tuesday March 16 2010
Does hair loss reduce the risk of prostate cancer?
“Hair loss 'almost halves the risk of prostate cancer',” the Daily Mail reported. The news is based on a study looking at hair loss in about 2,000 men aged between 35 and 74, about half of whom had prostate cancer.
This study does appear to suggest a possible association between male pattern baldness that begins before the age of 30, and a reduced risk of prostate cancer. No link was found between baldness in later life and prostate cancer risk. However, other studies have had contradictory findings, with some linking hair loss to an increased risk of prostate cancer. This study also has some limitations, including having to rely on men’s recollections of whether they began to lose their hair by age 30 – up to 44 years in the past for some.
These inconsistent results and the limitations to the current study make it difficult to draw firm conclusions about the relationship between male pattern baldness and prostate cancer. Further well-designed prospective cohort studies looking at this question are needed.
Where did the story come from?
Dr Jonathan L Wright and colleagues from the University of Washington and Fred Hutchinson Cancer Research Center carried out this research. The study was funded by the National Institutes of Health and the Fred Hutchinson Cancer Research Center. The study was published in the peer-reviewed medical journal Cancer Epidemiology.
The Daily Mail and The Daily Telegraph report the research and both papers make the important point that other studies have had different findings. However, both papers concentrate on the figure of a 45% reduction in the risk of prostate cancer – the largest risk reduction calculated in the study. This figure only applies to men aged 60 or over who showed hair loss at the top of the head and forehead at age 30, a fact not reported by the newspapers. In addition, the small numbers of men in this analysis are likely to make it less reliable than the overall analyses, which showed smaller reductions in risk.
What kind of research was this?
This case-control study looked at the relationship between early onset male pattern baldness and prostate cancer. The researchers thought that similar factors might be involved in the development of the two conditions as they are both affected by how much testosterone men’s bodies produce. They report that previous studies have had mixed conclusions – some finding male pattern baldness to be associated with an increased risk of prostate cancer, with others finding no such association. However, some of these studies were quite small and assessed male pattern baldness in different ways.
This study design compares past exposures/events in people with a disease and those without the disease. If an exposure/event is more common in people with the disease than those without the disease, then it could potentially be related to the cause of the disease. This type of study has some limitations. Firstly, cases and controls need to come from the same population, so that any differences between them are more likely to be related to the disease. Secondly, this type of study is usually retrospective – looking back at exposures in the past. It can be difficult for people to remember what happened in the past, which can reduce the reliability of results. If a link is found in a case-control study, ideally this link should be confirmed in a prospective cohort study that follows people over time and looks at whether their hair loss is related to their risk of prostate cancer later in life.
What did the research involve?
The researchers enrolled 999 men aged 35 to 74 who had been diagnosed with prostate cancer between 2002 and 2005 in one county in Washington (cases), and 942 age-matched men without prostate cancer (controls). They then assessed the men’s hair loss at age 30, and in the year before diagnosis for cases, or a similar date (called the reference date) for matched controls. Whether or not the extent of hair loss at these two ages differed between cases and controls was then assessed.
Participants gave information about their lifestyles, medical and family history, whether they had been screened for prostate cancer in the past five years, and whether they used drugs that could affect their testosterone metabolism, such as finasteride (a drug used to treat male pattern baldness and some prostate problems). Hair loss was assessed by showing participants drawings of heads with different extents of hair loss and asking which one best represented their hair loss at the two ages of interest. The drawings showed men with little or no hair loss, loss at the forehead only, or loss at the top of the head and the forehead.
Comparisons were then made between cases and controls in the odds of having hair loss at either time point.
The researchers also looked at whether the pattern of hair loss – at forehead only, or at the top of the head and forehead – had an effect, and whether effects were different in men aged 60 or over. Analyses took into account the men’s ages, race, prostate screening history, family history of prostate cancer, body mass index and use of finasteride.
What were the basic results?
The researchers found that about 20% of men who had prostate cancer had shown hair loss at age 30, compared to about 25% of men who did not have prostate cancer. This represented a 29% reduction in the odds of prostate cancer in men who had hair loss at age 30 (odds ratio [OR] 0.71, 95% confidence interval 0.56 to 0.91). If they looked only at men who were aged 60 and over at cancer diagnosis, the reduction in risk was 37% for men who showed hair loss at age 30.
There was no difference in the proportion of men in the case and control groups who only started losing hair after the age of 30, suggesting that hair loss after 30 was not related to prostate cancer risk.
Different patterns of baldness were associated with different reductions in risk of prostate cancer. Loss of hair at the forehead only and loss of hair at both the forehead and top of the head at age 30 were associated with a reduction of 25% to 31% in the odds of prostate cancer. This reduction was only statistically significant for loss of hair at the forehead only – possibly because this type of hair loss is more common. When looking at men aged 60 and over, men showing loss of hair at the top of the head and forehead at age 30 had a 45% reduction on the odds of prostate cancer compared to men with little or no hair loss.
How did the researchers interpret the results?
The researchers concluded, “early-onset [male pattern baldness] was associated with a reduced relative risk of [prostate cancer] in this population-based study”. They suggest that further research into why this may be the case is needed.
Though this study suggests there is an an association between early onset male pattern baldness and risk of prostate cancer, the findings are not conclusive. Its strengths include its relatively large size and that it assessed men’s baldness at a standard point in time (at age 30). However, there are a number of points to consider:
- Hair loss was assessed retrospectively, and this may reduce the reliability of these reports, particularly for hair loss at age 30, which may become increasingly difficult to recall the further in the past this date was.
- Some of the analyses of specific subgroups of men, particularly those examining patterns of baldness in men aged 60 and over, only included small numbers of men, and therefore these results should be interpreted cautiously.
- Although the researchers took into account some factors that could be affecting the results in their analyses, other unknown or unmeasured factors could be having an effect.
- The researchers report that some previous studies have produced conflicting results, whereby hair loss had been associated with an increased risk of prostate cancer. The researchers suggest that this is because of differences in when hair loss was assessed. A more detailed systematic review and analysis of all studies on this issue would give a clearer idea of whether male pattern baldness is associated with prostate cancer in any way.
The conflict between these findings and those from previous studies – and the limitations of the current study – mean that it is difficult to draw firm conclusions about the relationship between male pattern baldness and prostate cancer. Further well-designed prospective cohort studies looking at this question are needed.