Figuring out those breast cancer risk figures

This story was first published in The Straits Times on May 25, 2013

HOLLYWOOD star Angelina Jolie, 37, wrote recently in The New York Times about carrying "a 'faulty' gene, BRCA1" that she said gave her a "65 per cent risk of getting (breast cancer) on average".

The BRCA1 gene mutation, along with other factors, led her doctors to estimate her risk of breast cancer to be 87 per cent, so she chose to have a preventive double mastectomy which brought it down to 5 per cent, she said.

Her decision grabbed media headlines worldwide and sparked controversy. Some thought she or her doctors had over-reacted. Underlying those views are the figures: How were those statistics derived and what do they mean?

First, BRCA1 and BRCA2 are tumour suppressor genes, which make sure that DNA stays stable, so cell growth is orderly. But mutations impair their ability to repair cells with genetic defects or limit their proliferation. As a result, cell growth may become disordered and specific cancers may develop.

BRCA mutations affect few people and account for only between 5 and 10 per cent of breast cancer cases. They are acquired from one's father or mother or both.

A woman with BRCA mutations from one or both parents is likely to develop breast cancer younger - before menopause. She also has higher ovarian cancer risk. Hence Jolie's plans to have a preventive double oophorectomy (to remove the ovaries, fallopian tubes and womb).

What matters to a woman is the absolute risk she faces of getting a particular cancer over a lifetime. For US women, it is one in eight, or about 12 per cent. That is, about 120 out of every 1,000 women in the general US female population will get breast cancer in their lifetime.

How do we know this?

Basically, healthy women similar in ethnicity and age are followed up to see how many develop breast cancer over the next 10 years of their lives.

The number of actual cases for each age group divided by the total number of women in that age group gives the absolute likelihood of getting breast cancer.

Using such US data collected between 2005 and 2007, the absolute risks a woman there would develop breast cancer in the 10-year age band beginning from age 30, 40, 50, 60 and 70 were 0.43 per cent, 1.45 per cent, 2.38 per cent, 3.45 per cent and 3.74 per cent, in that order.

All these 10-year absolute risks over a woman's lifespan when totalled up give us 11.5 per cent or about 12 per cent, which is the cumulative or total absolute lifetime risk of breast cancer for a US woman, on average. When data from various years is used, this figure is consistently about 12 per cent.

When similar data is gathered for patients with BRCA1 mutations, we find their risks to be much higher, rising from 3.2 per cent at age 30 to reach 31 per cent at age 70, which total to a lifetime risk of 85 per cent. Likewise for BRCA2, the risks increase from 4.6 per cent at age 30 to 24 per cent at age 70, for a total of 85 percent also. Over a lifetime, a woman with BRCA mutations has about seven times the absolute risk of another woman without them.

This was probably how Jolie's doctors got to 87 per cent, using older data perhaps. Dr Tan Min Han, consultant medical oncologist of the Cancer Genetics Service at National Cancer Centre Singapore, feels that this figure was likely "based on estimates derived from high-risk families ascertained in the early years when BRCA was discovered".

There is, however, one problem with deriving such figures. One issue in genetic studies is the ascertainment bias. As BRCA mutations are rare, most BRCA carriers are identified in families attending breast/ovarian cancer clinics. By selectively looking at these cases, the relevant gene mutations, associated cancers and the most severely affected are more likely to be found.

By contrast, if the larger community were sampled, it is likely that more cases may be found where the mutations do not lead to cancers. Thus the present method of ascertaining a rare gene's impact is likely to exaggerate its significance.

So experts feel now that, if corrected for this bias, the impact of BRCA mutations is probably about 65 per cent, a figure Jolie cited for her BRCA1.

So was her double mastectomy a wise move? Though the underlying risk factors including the BRCA mutations remain, Associate Professor Lee Soo Chin, senior principal investigator at the Cancer Science Institute of Singapore, National University of Singapore, argues: "The procedure reduces cancer risk as the bulk of the at-risk tissues are removed."

Studies in the last decade comparing breast cancer incidence in high-risk women monitored closely but not operated on to those who had preventive mastectomies "estimated the procedure to reduce risk by 90 per cent".

But surgery does not guarantee that the woman will not have breast cancer in the future as cancer can develop in any breast tissue that remains. Prof Lee adds: "Her risk is not completely eliminated, and indeed, may remain higher than some women with the lowest risk (under 5 per cent, for instance) because she still has small amounts of breast tissue and her genetic predisposition remains."

Not all breast tissue can be removed surgically since it is found widely on the chest wall, even just under the skin, as far as the lower rib margins and in the armpit and just above the collarbone. The surgeon tries to remove breast tissue from just under the skin down to the chest wall and around the chest borders.

Jolie's doctors said her surgery reduced her heightened risk from 87 per cent to just 5 per cent, like some women with the lowest risk.

But in the end, whatever the figures show, experts concur that the patient must choose for herself and not be swayed by others - like Jolie's activism.

Should local women use these computer simulation models to assess their own risk?

A model's accuracy depends on the prevalence of risk factors in the general population. In Singapore, the average woman has a lifetime risk of breast cancer of 5 per cent. This is lower than the US figure, so the risk factors here may differ, including BRCA mutations.

In general, mutations tend to be population-specific, so the spectrum of mutations may differ in Chinese populations, say. A 2013 study of Hangzhou Chinese and a 2012 study of Hong Kong Chinese both found BRCA1 mutations to be less prevalent in southern Chinese than in Caucasian populations. And while both sets of mutations are rarer in these Chinese than Caucasians, BRCA2 mutations are more common than BRCA1 mutations in them, a reverse of the pattern found in Caucasians. So US simulation models must be tweaked for local use.

But all this should be academic for women here who have no strong family history of the cancer and no signs or symptoms. Your chance of living to age 70 is about 83 per cent, with just 5 per cent risk of breast cancer by age 70 and 1 per cent risk of dying from it, Dr Tan notes.

Just stick with the common sense approach: regular breast self-examination, healthy eating, regular exercise, alcohol abstention and seeing a doctor quickly if you have a nipple discharge or detect a breast lump. These suffice.

This story was first published in The Straits Times on May 25, 2013

To subscribe to The Straits Times, please go to