Thursday July 8 2010
Oily fish can be a good source of nutrients
Fish oil may cut the risk of breast cancer by a third, said the Daily Mail.
The news is based on a large study that followed just over 35,000 postmenopausal women for up to seven years to investigate how their use of supplements, including fish oil, affected their risk of developing breast cancer. It found that women currently using fish oil supplements had a reduced risk of developing ductal carcinoma, the most common type of breast cancer.
While the size of this study was a strength, it has several important limitations, such as not measuring the dosage or frequency of fish oil. Also, the small size of some groups in the study, and the use of multiple statistical analyses, increase the likelihood that associations have been found by chance.
This large study warrants further research into possible association between fish oil supplements and the risk of breast cancer, but until these results are confirmed by further studies, it is too early to recommend fish oil supplements as a method of breast cancer prevention, as the researchers of this study themselves conclude.
Where did the story come from?
The study was carried out by researchers from the University of Washington and the University of California in the US. It was funded by the US National Cancer Institute and published in the peer-reviewed medical journal, Cancer Epidemiology, Biomarkers and Prevention.
The Daily Mail’s report of the study was accurate and did mention the study authors’ conclusions that further research is needed. It also featured a quote from an independent expert who implied that the results of a single study are not normally sufficient evidence to make any health recommendations. It did not mention the important limitations of this research.
What kind of research was this?
This was a prospective cohort study that aimed to investigate the possible association between the use of ‘speciality supplements’ and breast cancer risk. The authors define speciality supplements as non-vitamin, non-mineral supplements promoted used for various purposes, such as glucosamine, black cohosh (often taken for menopausal symptoms), St John’s wort, garlic pills, acidophilus, Coenzyme Q10 and fish oils.
They point out that use of these supplements has increased substantially over recent decades and that several have been claimed to have anti-inflammatory or anticancer properties. Despite their rise in popularity, there have been no prospective studies looking at their long-term use and breast cancer risk, the researchers say.
Cohort studies, in which large groups of people are followed for many years, are useful in helping to assess potential links between lifestyle factors (in this case, taking supplements) and health outcomes. However, a randomised controlled trial could more reliably demonstrate whether taking a particular supplement affected the risk of breast cancer over time.
To make it possible to detect small differences in the rates breast cancer developed, any randomised controlled trial performed would need to recruit a large number of women and follow them for a sufficient period of time, which may not be practical.
What did the research involve?
Between 2000 and 2002, the researchers recruited 40,337 postmenopausal women, aged between 50 and 76 years. These women were members of a larger cohort designed to specifically look at possible links between all types of supplements and cancer risk.
At entry into the study the women were asked to complete a 24-page questionnaire that included a detailed assessment of supplement use, both currently and during the 10 years prior to the study start. They were asked how often they took supplements and for how many years. Information was also gathered from the women on known and suspected risk factors for breast cancer including BMI, physical activity, medication use, family and medical history and diet. The researchers excluded any women reporting breast cancer or a history of cancer, leaving a total of 35,016 for inclusion in the study.
The women were then followed from 2000 to 2007, to see who developed breast cancer. This was determined using these results from a cancer registry. The researchers then used established statistical methods to analyse any association between supplement use and breast cancer risk. Their models were adjusted to take account of many other things that might influence risk, such as known risk factors of age, race, reproductive history, alcohol consumption, use of hormone replacement therapy and diet.
What were the basic results?
The researchers found that women who reported they were currently using fish oil had a 32% reduced risk of ductal breast cancer, the most common type (hazard ratio [HR)], 0.68; 95%, confidence interval [CI], 0.50-0.92), but not of a type called lobular cancer.
There was no significant reduction in risk for women who had used fish oil supplements in the past. However, more frequent use over the past 10 years demonstrated a non-significant trend towards decreasing risk of breast cancer:
- low use was associated with a non-significant 25% reduction in risk when compared to non-use
- high use was associated with a non-significant 18% reduction compared to non-use
None of the other supplements, including those often used for menopausal symptoms, such as black cohosh and dong quai, were associated with either higher or lower risk of breast cancer.
How did the researchers interpret the results?
The researchers say that fish oil may reduce the risk of ductal but not lobular breast cancer and that this warrants further investigation. This should focus on the timing of exposure and the dose, as well as the mechanism of action that might explain the different effects by cancer stage or type. They stress that, until these results are confirmed by further studies, fish oil supplements should not be promoted for preventing breast cancer.
This large study is likely to be one of the first to assess any association between specialty supplements and breast cancer risk. It has strengths in that it carried out a detailed assessment of supplement use in 35,016 women, and also adjusted for numerous known and suspected risk factors for breast cancer (possible confounders) when calculating cancer risk by supplement use.
However, the study provides no firm evidence that fish oil supplements reduce breast cancer risk and therefore they should not be recommended for this purpose. Further research is needed.
It is important to note that:
- The study carried out multiple statistical analyses examining the relationships between breast cancer and the use of numerous supplements, in addition to other medical and lifestyle factors. The multiple analyses increase the possibility of finding associations by chance.
- Of all the supplements examined, a link was only found with current use of fish oil supplements (at study start); however, only 47 women who were currently taking cod liver oil at study start went on to develop breast cancer meaning this small number again increases the risk of chance findings on statistical analysis.
- The questionnaire asked about ‘current use’ of fish oil a term that provides little information on the preparation, dose, frequency or duration of use. However further analysis of pattern of use over 10 years demonstrated no significant effect on breast cancer risk.
- Although researchers tried to exclude women with a cancer history and certain risk factors for breast cancer at the start of the study, it is possible that some of the women in the study had undiagnosed breast cancer at the time, which could have affected results.
- The study only followed women up until 2007, with an average follow up of six years. This is a relatively short period and many cases of breast cancer may go on to develop after this time. Diagnoses of breast cancer after the follow-up date could have affected the results.
- Although it tried to control for potential confounders, there is always a possibility that in this type of study, both measured and unmeasured confounders could be having an effect.
- The study relied on the women self-reporting their use of supplements and the factors that might have affected their risk of breast cancer. This may have introduced some inaccuracy.
- The study was specifically of postmenopausal women, and results may be different if supplement use were studied in premenopausal women.