p53 (23 page)

In the great majority of cases, people with Li-Fraumeni syndrome, or LFS, are born with a mutant copy of p53 in all their cells. This is a so-called ‘germline mutation’, which means
that it occurs in a sperm or egg cell, and the faulty gene is then passed on from generation to generation by the affected parent, leaving their offspring especially vulnerable to cancer at almost
any age.

The Berkeleys were offered genetic counselling. The three who had suffered from cancer were tested and all proved positive for mutant p53. John’s brother died in a car accident in 2004
while in the throes of an epileptic fit. His father, after suffering several bouts of sarcoma, died of pancreatic cancer in 2007. In 2010 John was one of a small group of people to set up the
Li-Fraumeni Syndrome Association – a web-based organisation offering mutual support to people coping with the intensely lonesome experience of living with the constant threat of cancer. He
himself has since had another brush with the disease, being diagnosed with myofibroblastic sarcoma in 2012.

THE DISEASE DETECTIVES

Jo Fraumeni started his career as an epidemiologist with the National Cancer Institute working in a dingy little office
above a dress shop in downtown
Bethesda, Maryland. Epidemiology was a fledgling discipline at NIH in the mid-1960s and Fraumeni shared the space with just one other person, Bob Miller, also a recent recruit. Fraumeni had studied
medicine at Duke University, North Carolina. During his residency at Memorial Sloan Kettering in New York he had discovered that he had an eye for patterns of disease and a special interest in
looking beyond the patient seated before him in the clinic to the context in which he or she had become sick – the environment in which his patients lived and worked and the families they
came from – for clues to what ailed them. Reaching a decision point in his career as a physician, he might well have joined the armed forces and been sent to Vietnam, but he chose instead to
specialise in epidemiology at NIH.

Bob Miller was a paediatrician, but he had a degree in epidemiology as well. The two men struck up a firm friendship. ‘Bob was kind of an iconoclast,’ commented Fraumeni. ‘I
remember one of the first things he did was to show me a picture on his wall of a cover from the
New Yorker.
It was of seagulls standing in a row along the roof of a house. All except one
bird, standing a little apart from the others, were facing in the same direction. The ones looking forward, he told me, are considering the diagnosis and how to treat the patient. “We should
be like this one,” he said, pointing at the odd man out, “looking sideways to see what else is going on in the patient, in the community, in the environment, in the
family.”’

Fraumeni still has that picture and he pulled it from a sheaf of papers he had gathered to show me when I visited him at the NIH’s National Cancer Institute in the summer of 2012. He was
just about to retire, at the age of 79, as Director of the Department of Cancer Epidemiology and Genetics, which he had founded 50 years earlier and which had grown under his leadership from its
humble roots above the dress shop into a huge and prestigious unit housed
today in purpose-built laboratories and offices on an elegant modern campus outside Bethesda. After
a sweltering hot day the sky had gathered thunderclouds which broke just as I arrived. As we sat talking in his book-lined office with its myriad framed certificates and awards dotting the walls,
hailstones slashed at the window and leaves stripped from the trees danced in the wind behind the glass. After a pause for reflection, Fraumeni, a reserved man with a spare frame and large owlish
spectacles, who chooses his words carefully, said, ‘I was very fortunate to have met Bob. He was what I call a kindred spirit.’

Miller sparked his curiosity in childhood cancers, about which almost nothing was known beyond the fact that children with Down’s syndrome are especially prone to leukaemia. With this one
example of a link in mind, the two epidemiologists decided to look for other patterns in the occurrence of paediatric cancers for clues to what might be triggering the disease at unusually young
ages. Cancer of any kind is relatively rare in children, so in order to get enough cases for patterns to become apparent, they needed to scour the records of multiple hospitals. Their first
investigation was into Wilms’ tumour, a cancer of the kidneys for which their searches turned up 440 cases. These were associated with a ‘constellation of anomalies’ that included
genitourinary defects and mental and physical retardation, as well as aniridia, or absence of the iris, in the eyes of some affected children. This condition is so rare that the six cases of
aniridia they found among their sample were ‘off the wall statistically’ – showing a frequency more than 1,000 times what would be expected in the general population – and
therefore strongly suggestive of a link with the tumour.

‘We were excited by all this and we went after all different kinds of childhood tumours,’ Fraumeni told me. ‘We found not only associations between the tumours and the
anomalies, but between tumours and tumours. Kids with one cancer sometimes developed a second cancer, unrelated to
the first. We found that some of them were related to
treatment – radiation and chemotherapy – but a lot of them were not, so it led us to conceptualise the notion that genetic susceptibility can result either in multiple tumours in the
same individual, or multiple cancers scattered over the family tree.’

By this time, Fred Li, also a paediatric oncologist by training, had joined their tiny unit. ‘Fred was a wonderful guy,’ commented Fraumeni. ‘He was a prince. Easy to work
with, very bright, very modest – he added a tremendous amount to our group.’ (Fraumeni has a quaint habit of describing any man he admires – including Alfred Knudson, whose work
on retinoblastoma, you will recall, led to the discovery of tumour-suppressor genes – as ‘a prince’.) The story of the syndrome to which he and Li gave their names began with a
survey of children with adrenocortical tumours – a type of cancer that involves the outer layer of tissue of the adrenal glands. These tiny organs sit atop the kidneys and produce the
hormones that control many of the vital functions of living, such as heart rate, blood pressure, the fight-or-flight response to stress, growth and sexual characteristics. ‘Adrenocortical
tumours are very, very rare,’ said Fraumeni, ‘so we had to go to about 10 hospitals to get 21 cases. And two of them developed brain tumours . . . Unusual!’ he said, pausing for
emphasis. ‘Then when we finished that survey, I got a call to say that a third case had developed a brain tumour and the family was riddled with sarcomas. That suggested to me there was enemy
action . . . Three cases like that was unusual to say the least. So there were all these little clues coming up that something was going on.’

The eureka moment came when a family was referred to Fred Li in which two young cousins had rhabdomyosarcoma (the muscle tumour that afflicted John Berkeley, whose story opened this chapter);
the mother of one child, still in her twenties, had breast cancer; and the father of the other
had acute leukaemia. ‘It was explosive, you know, so much cancer at the
same time,’ commented Fraumeni. The two doctors homed in on the family, taking a detailed medical history across the generations that revealed an abundance of cancers, sometimes with multiple
tumours in the same individual. ‘That was it – we knew there was a syndrome.’

Thoroughly caught up in their detective work by now, the two returned to their multi-centre survey of childhood cancers and pulled out the cases of rhabdomyosarcoma for further investigation.
Among these they found another three families in which the children’s tumours were associated with cancer in their still-young parents. Convinced they had uncovered something important in a
field that still had little clue as to how and why children should be afflicted by a disease normally considered to be caused by the ravages of age and long-term exposure to harmful environments,
they rushed to prepare a paper for the
Annals of Internal Medicine
as quickly as they could. ‘The thing that took the longest time was deciding what to call our syndrome,’
laughed Fraumeni. ‘Fred was more conservative than I was and we ended up with a question mark after the words “a familial cancer”.’

The year was 1969, way before the genetic basis of cancer had been convincingly demonstrated, and most of the seagulls on the roof were looking to viruses. Li and Fraumeni’s paper with its
suggestion of a familial syndrome – by definition genetically based – was received with frank scepticism. A handful of familial, and therefore hereditary, cancers were already known,
but they tended to give rise to the same tumour type in related cases – either breast or colon or ovary, for instance – and to affect adults. ‘Our syndrome was unusual in that
those affected were all children and very young adults – and a
bizarre
array of tumours . . . Every conceivable cell type, from leukaemias to gliomas to sarcomas to adrenal tumours
to breast cancer. And there were some other strange tumours like choroid plexus tumours, which are in the lining of the brain where the cerebrospinal fluid
circulates. These
tumours have come up so often in LFS that they’re almost pathognomonic (i.e. a defining characteristic of the disease). It’s like adrenal cortical cancers in children – when you
see them, you think LFS.’

The sheer variety of tumour types was what puzzled the sceptics. Some suggested that a virus, transmitted from mother to child during pregnancy and birth much like HIV, was behind the family
pattern. Others said there was no real pattern and that what they were seeing was ‘just the play of chance’. But while, as good scientists, they had to keep open minds, Li and Fraumeni
were pretty convinced that a gene or genes were at the root of the clusters they had observed. They decided to keep a close eye on the four original families in their survey, and over a 12-year
period they saw a further 16 cases of cancer develop – all of them part of the same constellation of tumours that characterised their syndrome.

During this time two other groups had joined the field of familial cancer epidemiology: one led by Louise Strong at MD Anderson in Houston and the other by Jill Birch at Manchester University in
England. Together they provided much new evidence in support of Li and Fraumeni’s genetic hypothesis – and, incidentally, were responsible for the name Li-Fraumeni syndrome, or LFS,
creeping into the medical literature and common parlance in the 1980s. Almost another decade was to pass, however, before mutant p53 was identified as the gene bringing grief to the unfortunate
families.

HOMING IN ON THE RESPONSIBLE GENE

In the late 1980s Fred Li, then Head of Cancer Epidemiology and a practicing oncologist at the Dana-Farber Institute in Boston, was joined on the staff by a young Canadian
named David Malkin, trained in paediatrics at Toronto’s Sick Kids Hospital, who was looking to gain some experience
in children’s cancers to take back to the
clinic in Canada. But the time he spent with Li – whom he describes as ‘a gentleman and a scholar; not physically imposing but with a real presence, and extraordinarily
knowledgeable’ – gave Malkin an appetite for research, and he decided to pursue a career as a scientist alongside his work in the clinic seeing patients. He got a position as a postdoc
with Steve Friend, the scientist who had discovered the first tumour-suppressor gene, Rb, just three years earlier. Friend had recently left Weinberg’s lab to set up on his own at Harvard,
and over a casual meal in a restaurant he and Malkin brainstormed ideas for a research project to work on together. ‘Steve explained that he liked working with Fred (Li) and they were
interested in exploring the genetics of this weird syndrome. I knew relatively little about LFS at the time,’ commented Malkin, ‘but it sounded intriguing.’

With a vast arena to work in and no clues to guide them, the two decided to start their search for the faulty gene by looking at Rb – if only because, with an inherited condition, a
malfunctioning tumour suppressor that would leave carriers unprotected against cancer seemed like the most obvious candidate. Rb was familiar territory to Friend and there was little to excite
their interest in p53, which had yet to emerge from the doldrums of being miscast as an oncogene. Coincidentally, the two scientists drew a blank with Rb just as big things began to happen again
with p53. In 1989 Suzy Baker and Bert Vogelstein in Baltimore revealed that wild-type p53 was in fact a tumour suppressor, and later that same year scientists working with mouse models in Toronto
published a paper describing the multiple tumour types that developed in animals with mutant p53. The constellation of mouse tumours didn’t exactly match what oncologists were seeing in
families with LFS, said Malkin, but it was a dramatic enough demonstration of p53’s broad effect to make the researchers switch their focus to this newly revealed tumour suppressor.

Working with material provided by members of LFS families seen by Fred Li and Louise Strong in their cancer clinics, the two scientists set about isolating p53 to look
for possible mutations. In those days it was an almost infinitely laborious process of cloning and sequencing stretches of DNA one after another until they found their target gene. ‘A lot of
sequences didn’t work, and it was well over a year before we had our first hit that we knew was real,’ commented Malkin. ‘I remember I looked at the gel; I sequenced the gene; I
compared it against the normal p53 and there was the mutation. We were very lucky because the first one that popped up was at codon 248, which happens to be about the most common mutation we have
in p53. That was fortuitous, because if it had been in some slightly more obscure site we may not have paid so much attention to it.’

He and Friend knew there were months of hard work ahead if they were to confirm the finding and look for more mutations but, once they realised they were on the right track, the whole lab went
into overdrive. ‘That’s where you turn your alarm clock off and you’re just in the lab the whole time because you’re having a lot of fun!’ said Malkin with a grin.
‘That summer of 1990 was very exciting. We were in a brand new building, it had a beautiful view over Boston Harbour . . . There was no good reason not to enjoy yourself and the whole lab was
really on fire.’