A stark, anonymous office overlooks the Surrey countryside. Inside, Professor Anthony Swerdlow is doing what feels to me like the most worthwhile job in the world. He is making the discoveries that could save my four daughters from going through the nightmare that has dominated my life this year: breast cancer.
I discovered I had the disease at the end of March, and having a mother with it automatically increases the chances of my girls - Rosie is 22, Elinor 20, Miranda 15 and Catriona 12 - one day getting the same heart-stopping diagnosis. Aside from doing all I can to ensure I survive it, my single biggest hope is that my daughters will never have to sit in a clinic, as I have done, bandying words like "mastectomy", "lumpectomy", "reconstruction" and "wide local excision" across a doctor's desk. They're ugly words for an ugly condition, and when you have it you're presented with an unpalatable and unpleasant set of choices.
And the person who could help ensure my girls never need to use these terms is Swerdlow, along with the study he masterminded and which next week celebrates its 10th anniversary. It's one of the most ambitious breast cancer studies on the planet; it's tracking the lives of more than 113,000 British women across half a century, and it promises answers to an epidemic that worries away in the back of every female mind, and brings the world crashing down at some point for one in every eight of us.
Swerdlow is an epidemiologist. He works out of the Institute of Cancer Research's futuristic and ever-expanding science campus in Sutton, but in Oxford, where he lives and where he used to work, he had occasional contact with the most distinguished epidemiologist in history. Sir Richard Doll, who died in 2005 aged 92, carried out a long-term study of 40,000 British doctors that began in the 1950s and which finally proved what seems obvious now, but was not at all obvious then, which is that smoking causes lung cancer (until Doll's study it was being linked with car exhaust fumes and road tar).
As lung cancer was to the 1950s and 60s, so breast cancer has been to the 90s and 00s. Since the late 1970s its incidence has gone up by more than 70%, and the question why is one of the great medical conundrums of our age. So in the spirit of Doll, who realised that it was only with a huge, forward-facing study that he could properly uncover the cause of lung cancer, Swerdlow thought big, and he thought long-term.
In effect, he says, there are two sorts of studies of cancer causation: those that look backwards, asking retrospective questions about the lifestyles of people who have been diagnosed with the disease; and those that recruit healthy people, and wait and watch to see who gets cancer. "The first sort of study is easier and quicker to do, but you may get biased answers from people who already have cancer about the way they behaved in the past," says Swerdlow. "The second type, a cohort study, gives much purer, less biased information because it asks people about their lifestyle before they developed cancer, so their answers aren't coloured by knowing their diagnosis."
So the Breakthrough Generations study, funded by Breakthrough Breast Cancer, was born, first as a 1,000-strong pilot study and then, after appeals in the national press and on radio, with the recruitment of 100,000 more women. That, at least, was the target: but on the day in August 2004 when the phone lines opened for volunteers, Swerdlow and his team were inundated and the study eventually recruited way over its target. In more than 30 years of signing people up to studies, Swerdlow has never experienced anything like it. "We were completely overwhelmed," he says. "It was astonishing."
Volunteers ranged from 16-year-old schoolgirls to a woman of 102. In some families several generations signed up, providing researchers with important information about the role genetics has to play.
And it's been participatory enthusiasm that has marked this study out; it's about women recognising their own role in assisting medical science to combat breast cancer, the same sort of spirit seen each year at the annual MoonWalk, when thousands of women march across London through the night to raise funds for research.
"Studies like ours only work if we can recruit sufficiently dedicated women to take part: women who are prepared to share with us information and provide us with the blood samples we need, and to answer questions about their changing lifestyle, health and illnesses in the future, as their lives go on," says Swerdlow. In effect, he says, his job is a bit like running a detective agency that is keeping tabs on an extraordinarily large number of people. "We don't want to lose any of them, because what happens to each is important. And of course sometimes the women who are most difficult to find are the very ones who have the most useful information for us, so we have to try very hard to track them down - and we do," he says.
Around 1,800 participants have been diagnosed with breast cancer since joining the study, and their doctors are then asked to share their mammograms and other clinical information, so Swerdlow's team can look for patterns, risk factors and markers. Because this is the ultimate aim of the study. "We want to dissect exactly what it is that raises women's risk of breast cancer, and convey this in a way that can be fed back to women in terms of what actions they can take to reduce their risk of ever developing breast cancer in the first place and of developing a second cancer if they have already had cancer once," he says.
But, though Doll's study and methodology are Swerdlow's gold standard, his research is more complicated than Doll's. Cancer is never a straightforward disease, and in every case lifestyle interacts with genetics and the exposure to specific dangers. But with lung cancer the link between its development and smoking was relatively simple. With breast cancer, in contrast, there is no single "bad" lifestyle choice that raises the risk; instead a crucial factor at its centre is one of the most elusive and intricate of all our body's mechanisms, its hormone system.
Women's reproductive lives are governed by a delicate interplay between various sex hormones; and just as the hypothesis at the root of Doll's work was that smoking had a role to play in lung cancer, so Swerdlow's work is founded on a well-documented understanding that these hormones are important to the development of the disease in the majority of breast cancer sufferers. Precisely how these hormones can help pave the way for cancer, plus the way lifestyle issues, like exercise levels, weight and alcohol consumption can impact on the hormones and their behaviour, is the $64m question at the heart of his work. Swerdlow's study will help cancer specialists piece the puzzle together: but even a decade in, these are still early days. "We have some answers, some pointers, some information that is helpful," he says. "But over the next 30 and 40 years these women's experiences are going to give us many more." What he has done to date, he says, is to plant a vineyard: the best wine will take a while to come through.
But when it comes, the study's results will be extremely significant: not only for families like mine, but for women everywhere, and they're eagerly awaited by doctors such as Peter Barry, consultant breast surgeon at the Royal Marsden hospital, who has to deal every day with the sharp end of the epidemic - women in their 50s and 60s struck down by cancer. One of the most important outcomes of the study, he feels, could be a much more tailored approach to screening. At the moment it operates on a "one size fits all" policy, with mammograms offered at three-yearly intervals to all women aged 50-70. In the future, using information gleaned from the study, women could be traced from their 20s with blood tests looking for specific markers; those found to have these markers could be offered enhanced tests or advised on lifestyle changes.
The study could focus attention too, he says, on when and whether it's a good idea to remove healthy breasts. Specialists report that some women, on hearing they have the disease, have a knee-jerk reaction that double mastectomy is the "best" way forward. "It's a response we see all the time, and it's driven by fear," says Barry. In fact, he says, there is no research to show that removing healthy breasts reduces the risk of dying of the disease. "The current trend, and it's being driven from the US, is towards bilateral mastectomy even in women who are not at particularly high risk," he says. "But it's got to be better for us to be able to say there are lifestyle changes you can make, because mastectomy and reconstruction is no walk in the park - it can involve many operations and it's life-changing."
And there are other implications too: encouraging women to make specific, tailored changes to their lives to avoid cancer, where there is a clearly demonstrable advantage to doing so, will have overall health advantages. The study will have cost pounds 20m to run by the end of 2019, but its findings could eventually save the NHS many times that amount, not to mention families like mine a colossal amount of heartache. The sponsoring charity, Breakthrough, estimates that by 2050 what they have discovered could prevent 30% of breast-cancer cases. My third daughter will be 51 by then, my age now: her sisters will be 48, 56 and 58. More than anything, I hope they are in that 30%.
The Breakthrough Generations Study is sponsored by Breakthrough Breast Cancer. To support its work, go to breakthrough.org.uk/donate or text 'PINK' to 70300 to give pounds 3
¦ Data from the Breakthrough Generations Study has contributed to identifying more than 50 new genetic markers associated with an increased breast cancer risk.
¦ Being heavier, drinking alcohol and eating red meat all delay the menopause, and that in turn is linked with a higher risk of breast cancer.
¦ Breast density - ie the opacity of breast tissue seen on mammograms - has emerged from this and other studies as a clear risk factor for breast cancer. In the future, women may have the density of their breasts taken into account when deciding on measures to reduce risk.
¦ The study has identified several factors that make breast cancer more likely in later life. Having a low birth weight, not being breastfed, not exercising much and starting periods early have been linked with increased risk.
¦ The study has produced some surprising risk factors, including that women who live in areas with high levels of light (ie street lights) tend to be more overweight. After the menopause, being overweight is a risk factor for breast cancer. One study found women who work at night have an increased risk of breast cancer; it's likely to be linked to levels of melatonin, a hormone that regulates the sleep cycle.
Joanna Moorhead with her youngest daughter Catriona, 12, on Wandsworth Common in London.
Photograph by Sophia Evans for the Observer
¦ Around 50,000 women and 400 men are diagnosed with breast cancer each year. It is the most common cancer in the UK.
¦ Around eight in 10 breast cancers are diagnosed in women aged 50 and older.
¦ Forty years ago, roughly half those diagnosed survived beyond five years. Now it is more than eight in 10.
¦ Women with a mother, sister or daughter diagnosed with breast cancer have almost double the risk of being diagnosed with it themselves.
¦ Around 11,000 women die each year from breast cancer in the UK.