The Daily Telegraph reports on a "remarkable breast cancer drug that could save the lives of thousands of women".
This solid and believable headline, along with similar ones from The Times and The Guardian, was based on large-scale, high-quality research looking at whether the drug anastrozole could reduce the risk of cancer in postmenopausal women who were at high risk of breast cancer.
Researchers gave these women either anastrozole (an aromatase inhibitor) or a placebo dummy pill. They found that women taking anastrozole reduced their risk of getting breast cancer from 4% to 2% over a five-year period compared with women taking a placebo. This equates to a halving of relative risk, which is welcome given that these women were at high risk of cancer.
Promisingly, anastrozole didn't appear to have many – or any serious – side effects. However, we can't tell whether the drug will work as well as the other existing drugs in use, because this study only used a placebo.
However, all postmenopausal women in this study had a higher than average risk of developing the disease because of a family history of breast cancer and other specific medical criteria. The results don't apply to other groups.
Anastrozole has a drug licence to treat breast cancer in postmenopausal women, but it can't currently be used to prevent breast cancer. If it gets a licence for this use, the NHS watchdog the National Institute for Health and Care Excellence (NICE) may have to reconsider its current recommendations on drugs to reduce the risk of breast cancer, factoring in the new evidence.
Where did the story come from?
The study was carried out by a large international collaboration of researchers led by the University of London. It was funded by Cancer Research UK, the National Health and Medical Research Council, Australia, and drug manufacturers Sanofi-Aventis and AstraZeneca.
Anastrozole was originally developed by British company Zeneca Pharmaceuticals, now AstraZeneca, and goes under the brand name Arimidex. Because of the drug company's involvement in this research, there is a clear potential conflict of interest. However, the publication states that, "The sponsors of the study had no role in study design, data collection, data analysis, data interpretation, or writing of the report."
The media reporting generally reflected the underlying results of the study accurately, with most focusing on the 50% reduction figure, its relative effectiveness with tamoxifen, and the observation that there were few side effects.
What kind of research was this?
This was a double-blind, randomised, placebo-controlled trial assessing the effectiveness of the drug anastrozole to prevent the development of breast cancer in postmenopausal women with a higher than average risk of developing breast cancer.
A double-blind randomised control trial is the most robust and appropriate study design to assess the health effects of this drug.
Anastrozole is an "aromatase inhibitor", a type of hormone treatment currently used to treat breast cancer in postmenopausal women with oestrogen receptor-positive breast cancers (oestrogen stimulates the breast cancer cells to grow).
Postmenopausal women no longer produce oestrogen from their ovaries, but they do produce a small amount of oestrogen in their body.
Anastrozole is approved by the medicines regulator to treat breast cancer in postmenopausal women fulfilling specific criteria, but it is not yet approved for the prevention of breast cancer in the way it was used in this trial.
Until it is licensed to be used in this way, the National Institute for Health and Care Excellence (NICE), which is the agency that recommends the drugs to be used in the NHS, is unlikely to be able to recommend its use.
In June 2013 NICE published a guideline on breast cancer recommending that the drugs tamoxifen (another hormone treatment more commonly used in premenopausal women with breast cancer) or raloxifene (used for the treatment and prevention of osteoporosis in postmenopausal women) are considered for women at higher than average risk of breast cancer who fulfil specific criteria.
However, this guideline was produced before the current study results were published, so they were not taken into consideration. The new evidence will be considered when the guideline is next updated, but it is not clear when this will be.
What did the research involve?
Between 2003 and 2012 the researchers recruited postmenopausal women between the ages of 40 and 70 from 18 countries into an international double-blind randomised placebo-controlled trial.
To be eligible, women had to be at "higher risk" of breast cancer than average, based on specific criteria related to their medical and family history of disease. These are listed in full below as definitions of "higher risk" can vary from study to study.
For women aged 45 to 70:
- first-degree relative who developed breast cancer at age 50 or less
- first-degree relative who developed bilateral cancer
- two or more first- or second-degree relatives who developed breast or ovarian cancer
- no previous births (nulliparous) or age 30 or above at first birth
- no previous births (nulliparous) or age 30 or above at first birth, and first-degree relative who developed breast cancer
- benign biopsy with proliferative disease and first-degree relative who developed breast cancer
- mammographic opacity covering at least 50% of the breast
- first-degree relative with breast cancer at any age
- age at menopause of 55 years or more
For women aged 40 to 44:
- two or more first- or second-degree relatives who developed breast cancer or ovarian cancer at age 50 or less
- first-degree relative with bilateral breast cancer who developed first breast cancer at age 50 or less
- no previous births (nulliparous) or age 30 or above at first birth, and first-degree relative who developed breast cancer at age 40 or less
- benign biopsy with proliferative disease and first-degree relative who developed breast cancer at age 40 or less
For women in all age groups:
- lobular carcinoma in situ
- atypical ductal or lobular hyperplasia in a benign lesion
- ductal carcinoma in situ (oestrogen receptor-positive) diagnosed within last six months with completed adequate local treatment
- women with a clearly apparent family history indicating appropriate increased risk
Eligible women were randomly assigned by central computer allocation. Half received 1mg oral anastrozole and half received a placebo every day for five years. Except for the trial's statistician, none of the trial personnel, participants and clinicians knew which women had been allocated which treatment.
The researchers' main outcome of interest was breast cancer confirmed by biopsy (invasive cancers or non-invasive ductal carcinoma in situ, a very early stage of breast cancer that may or may not develop into invasive).
The researchers analysed their results using the "intention to treat" method, the preferred and more conservative way of measuring a drug effect in clinical trials.
What were the basic results?
A total of 1,920 women were randomly assigned to receive anastrozole and 1,944 to placebo.
After an average (median) follow-up of five years (interquartile range 3 to 7.1 years):
- Forty women in the anastrozole group (2%) and 85 in the placebo group (4%) had developed breast cancer (hazard ratio [HR] 0.47, 95% confidence interval [CI] 0.32-0.68). This means there was a 53% reduction in relative risk of developing breast cancer in women using the drug compared with placebo. It shows there was a 2% reduction in the absolute risk of developing breast cancer in women using the drug.
- The predicted cumulative incidence of all breast cancers after seven years was 5.6% in the placebo group and 2.8% in the anastrozole group.
- Eighteen deaths were reported in the anastrozole group and 17 in the placebo group (this was not significantly different) and no specific causes were more common in one group than the other.
- 51% of women in the anastrozole group and 50% in the placebo group had completed five years of treatment.
- The main reasons for stopping treatment were side effects and patient refusal. Side effects were given as the reason for stopping in 20% of the anastrozole group and 15% in the placebo group. Patient refusal was 5% in the anastrozole group and 5% in the placebo group.
- Many side effects were reported in both treatment groups. No significant differences were seen between treatment groups for fracture-related side effects, but musculoskeletal and vasomotor side effects were increased using anastrozole. Hypertension was also reported more in the anastrozole group.
How did the researchers interpret the results?
The researchers interpreted their results simply, saying: "Anastrozole effectively reduces incidence of breast cancer in high-risk postmenopausal women.
"This finding, along with the fact that most of the side effects associated with oestrogen deprivation were not attributable to treatment, provides support for the use of anastrozole in postmenopausal women at high risk of breast cancer."
They also point out that the reduction in breast cancer seen in their trial was larger than reported for the main alternative drug, tamoxifen.
Overall, the study provides solid and believable evidence that anastrozole can reduce the number of new cases of breast cancer in postmenopausal women at higher than average risk of developing breast cancer.
The study has many strengths, including its large sample size and robust study design. However, the study also has limitations to note.
The results only apply to a specific group of postmenopausal women at a higher than average risk of developing breast cancer. "Higher risk" was defined using a number of very specific criteria. This means the results are not applicable to other groups of postmenopausal women.
Around half of women in both the placebo and anastrozole groups had dropped out after five years, in part because of side effects. This highlights that although the side effects may not have been drug related, longer term treatment compliance may be an issue.
Anastrozole was not tested against existing drugs already used to prevent cancer in higher risk women, only against a placebo treatment.
This tells us that anastrozole is better than giving no drug, but doesn't really tell us if it is better or worse than the other drugs that are currently available. No studies have looked at this directly, but it is possible to make some indirect comparisons, although these are prone to error.
Anastrozole is approved by the medicines regulator to treat specific types of breast cancer in postmenopausal women, but it is not yet approved for the prevention of breast cancer in the way it was used in the trial discussed here.
If this drug obtains a licence for the prevention of breast cancer, it will then be up to NICE to recommend whether the drug is a sensible use of NHS resources, and whether or not to recommend anastrozole ahead of tamoxifen or raloxifene based on all the evidence available.
The study authors mention that the effect of tamoxifen has been shown to persist for at least 10 years, so further follow-up is needed to establish whether anastrozole has such a sustained effect. It was only tested for five years in the current study.
The bottom line is that the drug appears much more effective than a placebo pill, but it is less clear whether it is better than other drugs available from this research alone.