Cancer GPS?

The thing that pretty well everyone knows about cancers is that most are furtive little blighters. They kill one in three of us but usually we don’t they’re there until they are big enough to make something go wrong in the body or to show up in our seriously inadequate screening methods. In that sense they resemble heart problems of one sort or another, where often the first indication of trouble is unexpectedly finding yourself lying on the floor.

Meanwhile, out on the highways and byways you are about 75 times less likely to be killed in an accident than you are to succumb to either cancers or circulation failure. Which is a way of saying that in the UK about 2000 of us perish on the roads each year. That it’s ‘only’ 2000 is presumably because here your assailant is anything but furtive. All you’ve got to do is side-step the juggernaut and you’ll probably live to be – well, old enough to get cancer.

Did you know, by the way, that ‘juggernaut’ is said to come from the chariots of the Jagannath Temple in Puri on the east coast of India. These are vast contraptions used to carry representations of Hindu gods on annual festival days that look as though walking pace would be too much for them. So, replace the monsters on our roads with real juggernauts! Problem largely solved!!

Flagging cancer

But to get back to cancer or, more precisely, the difficulty of seeing it. After centuries of failing to make any inroads, recent dramatic advances give hope that all is about to change. These rely on the fact that tissues shed cells – and with them DNA – into the circulation. Tumours do this too – so in effect they are scattering clues to their existence into blood. By using short stretches of artificial DNA as bait, it’s possible to fish out tumour cell DNA from a few drops of blood. That’s a pretty neat trick in itself, given we’re talking about fewer than 100 tumour cells in a sea of several billion other cells in every cubic millimeter of blood.

There are two big attractions in this ‘microfluidics’ approach. First it’s almost ‘non-invasive’ in needing only a small blood sample and, second, it is possible that indicators may be picked up long before a tumour would otherwise show up. In effect it’s taking a biochemical magnifying glass to our body to ask if there’s anything there that wouldn’t normally be present. Detect a marker and you know there’s a tumour somewhere in the body, and if the marker changes in concentration in response to a treatment, you have a monitor for how well that treatment is doing. So far, so good.

And the problem?

These ‘liquid biopsy’ methods that use just a teaspoonful of blood have been under development for several years but there has been one big cloud hanging over them. They appear to be exquisitely sensitive in detecting the presence of a cancer – by sequencing the DNA picked up – but they have not been able to pinpoint the tissue of origin. Until now.

Step forward epigenetics

Shuli Kang and colleagues at the University of California at Los Angeles and the University of Southern California have broken this impasse by turning to epigenetics. We noted in Twenty More Winks that an epigenetic modification is any change in DNA, other than in the sequence of bases (i.e. mutation), that affects how an organism develops or functions. They’re brought about by tacking small chemical groups (commonly methyl (CH3) groups) either on to some of the bases in DNA itself or on to the proteins (histones) that act like cotton reels around which DNA wraps itself. The upshot is small changes in the structure of DNA that affect gene expression. You can think of DNA methylation as a series of flags dotted along the DNA strand, decorating it in a seemingly random pattern. It isn’t random, of course, and the target for methylation is a cytosine nucleotide (C) followed by a guanine (G) in the linear DNA sequence – called a CpG site because G and C are separated by one phosphate (p). Phosphate links nucleosides together in the backbone of DNA.

Cancer cells often display abnormal DNA methylation patterns – excess methylation (hypermethylation) in some regions, reduced methylation in others – that contributes to their peculiar behavior. It’s possible to determine the methylation profile of a DNA sample (by a method called bisulfite sequencing).

Kang & Co. developed a computer program to analyse methylation profiles from solid tumours and healthy samples in public databases and compare them to patient DNA of unknown tissue origin.

The peaks represent CpG clusters that characterize normal cells (top) and a variety of cancers. The key point is that the different patterns identify the tissue of origin (from Kang, S. et al., 2017).

The program’s called CancerLocator and in this initial study it was used to test samples from patients with lung, liver or breast cancer. In the modest words of the authors, CancerLocator ‘vastly outperforms’ previous methods – mind you, they struggle to even to distinguish most cancer samples from non-cancer samples. Nevertheless, CancerLocator’s a big step forward, not least because it can detect early stage cancers with 80% accuracy.

It’s also reasonable to expect major improvements as methylation sequencing becomes more extensive and higher resolution reveals more subtle signatures. What’s more, in principle, it should be able to detect all types of cancers – meaning that, after all so many centuries we may at last have a way of side-stepping the juggernaut.


Kang, S. et al. (2017). CancerLocator: non-invasive cancer diagnosis and tissue-of-origin prediction using methylation profiles of cell-free DNA. Genome Biology DOI 10.1186/s13059-017-1191-5.

A Taxing Inheritance

The centenary of the beginning of the First World War prompted me, as perhaps many others, to reflect on how successive generations have done since then in terms of what they’ve bequeathed to their offspring. I didn’t need to think for too long though, to find myself muttering ‘Thank heavens for science’—because most of the rest is a pretty dismal chronicle. I know, not all technological advances in the past one hundred years have been a cause of unrestrained joy but many of them transformed life in the most wonderful ways. Would that we could point to such success in other fields.

Our best defence may be to aver: “Man cannot control the current of events. He can only float with them and steer”, a saying attributed to Otto von Bismarck. If the ‘Iron Chancellor’ actually did utter those words it seems to me he was being coy beyond belief. He is, after all, generally credited with unifying Germany, seeing off the last French monarch (Napoleon III) and establishing the peaceful domination of Europe by the German Empire that lasted until long after his death—and setting up the first welfare state along the way. “The main thing is to make history, not to write it” sounds much more like Bismarck in full and frank mode.

Nature and Nurture

One form of history that we do write but indeed we cannot control comes in the form of the genetic material that we pass to the next generation. We’re all familiar with some of this legacy because we literally see it in physical resemblances and other attributes between parents and children (“He’s got his Mum’s eyes”) or shared by siblings (“Jack and Jill are wonderful musicians”). They’re shared because large chunks of the genetic code (i.e. DNA) are identical between the individuals concerned. But if conserved DNA makes for similarities, what of the differences—the fact that our parents and brothers look different to all the seven thousand million other people on the planet? Our unique features come from variations in the genetic code—odd changes in the units (bases) of DNA scattered through our genome. Called SNPs (pronounced ‘snips’ for single nucleotide polymorphisms), they’re what make the differences between us. In other words, a SNP is a difference in a single nucleotide—A, T, C or G—within a stretch of DNA sequence that is otherwise identical between two individuals. For example, you have AAGCCTA whereas I have AAGCTTA. These genetic variations that make individuals different are the basis of DNA fingerprinting.

There’s about three million SNPs scattered throughout the human genome (so, on average, you’d come across one in every 1,000 bases if you scanned your DNA from beginning to end) and they’re what makes each of us unique. Within ethnic groups common patterns of such variants confer characteristics (dark skin/light skin, tall/short, etc) and, with that in mind, you might guess that there will also be variants that make such groups more (or less) susceptible to diseases.

Of course, there’s an endless debate about the border between our genetic inheritance and how the world we experience makes us what we are—how much of Jack and Jill’s precocious talent is because Mum and Dad made them practice twelve hours a day from age five? Fortunately we can ignore nurture here and stick to genes because we’re trying to pin down the good and the bad of our genetic legacy.

What’s all this got to do with cancer?

A good bit is that we’re distinct from everyone else but still share family features. However, our genetic baggage may also contain some unwanted freebies—the most potent of which can give a helping hand to a variety of diseases, including cancers. Cancers are caused by damage to DNA—a build-up of changes, i.e. mutations, that affect the activity of proteins critically involved in controlling cell growth. For most cancers (90%) these mutations accumulate over the lifetime of the individual—they’re called “somatic mutations”—so you can’t blame anyone but yourself and Lady Luck. But about 10% get a kind of head start when someone is born with a key mutation. That is, the mutated gene came from either egg or sperm (so it’s a germline mutation). This effect gives rise to cancers that “run in families”: a critical mutation is passed from generation to generation so that children who inherit it have a greatly increased risk of developing cancer. Two of the most common cancers that can come in hereditary form are those of the breast and bowel.


A mutational steeplechase leads to cancer. Of the tens of thousands of mutations that accumulate over time in a cancer cell, a small number of distinct “drivers” make the cancer develop (four are shown as Xs). Almost all mutations arise after birth, but about one in every ten cancers start because a person is unfortunate enough to be born with a mutation: they are already one jump ahead and are much more likely to get cancer than those born with a normal set of genes. The rate at which mutations arise is increased by exposure to carcinogens, e.g., in tobacco smoke.

Breast cancer is about twice as common in first-degree relatives of women with the disease as it is in the general population (you’re a first degree relative if you’re someone’s parent, offspring, or sibling). About 5% of all female breast cancers (men get the disease too but very rarely—about 1% of all breast cancers) arise from inherited mutations. In the 1990s two genes were identified that can carry such mutations. These are BRCA1 and BRCA2 and their abnormal versions can increase the lifetime risk of the disease to over 50%, compared with an average of about 10%. Since then heritable mutations in some other genes have also been shown to increase the risk.

Angelina Jolie

Angelina Jolie

A star turn

Breast cancer genetics came under the spotlight with the much-publicised saga of Angelina Jolie, the American film actress. Jolie’s mother and maternal grandmother had died of ovarian cancer and her maternal aunt from breast cancer—a family history that persuaded Jolie to opt for genetic testing that indeed revealed she was carrying a mutation in BRCA1 (BRCA1 and BRCA2 mutations account for about 10% of breast cancers and 15% of ovarian cancers). For Jolie the associated lifetime risk of breast cancer was estimated as 87%, prompting her to have a preventative double mastectomy, thereby reducing her risk to less than 5%. The months after she revealed her story saw the “Angelina effect”, a doubling in the number of women being referred for genetic testing for breast cancer mutations.

What’s all this got to do with SNPs?

The story so far is of the one in ten cancers that get kicked off by a powerful, inherited mutation that changes the action of the affected protein—the BRCAs being the best-known examples. However, the BRCAs and other known mutated genes account for only about 25% of familial breast cancers, meaning that for three quarters of cases the genetic cause remains unknown. And yet we know there is an inherited (genetic) cause simply because of the generational thread. Which brings us back to those other, more subtle tweaks to DNA that we mentioned—SNPs—alterations that don’t directly affect proteins, so they’re often called variants to distinguish them from mutations.

It seems very likely that the missing culprits are indeed SNPs—lots of them. These DNA variants each make a contribution so small that on its own would have no detectable effect on the chances that the carrier will get cancer. Their impact comes from a cumulative effect. They’re like pieces of straw, individually easily bent or broken but put a dozen of them together and you have a rope. Thus combinations of individually insignificant SNPs can raise the risk of cancer by, say, 10%—not a massive increase but not negligible either. Twins who are genetically identical have similar risks of developing breast cancer, consistent with the idea that many variants, each having a very small effect, can combine to give a substantial increase in risk. Very slowly, by sequencing lots of genomes, these rare variants are being identified. Given that clusters of appropriate variants confer risk, people with the “other” variant have, in effect, a degree of protection against cancer.

And in our more distant relatives?

All this comes from the huge effort that has gone into finding genetic variants linked to one of the most common cancers but, unsurprisingly, almost all the attention has focused on European women. Not before time, someone has got round to looking for breast cancer variants in East Asians who, after all, make up over one fifth of all the people in the world. Cai Qiuyin and his colleagues at the Vanderbilt University School of Medicine compared the genomes of over 20,000 cancer cases from China, Japan and South Korea with a similar number of disease-free controls. After much selecting and comparing of sequences, three particular DNA variants consistently associated with significant cancer risk. The variants were much less common in European women, suggesting that as the DNA keyboard has been strummed by evolution, distinct patterns associated with breast cancer have emerged in diverse populations.

Just two problems then. First it’s a huge task to assemble the lists of runners (and as the Asian results show, they will differ between ethnic groups). But the real challenge is yet to come. Almost all of these variants (99.9%) don’t change the sequence of proteins (i.e. how the proteins work). What they do is exert subtle effects on, for example, how much RNA or protein is made from a DNA gene at any time. At the moment we have little understanding of how this works, yet alone ideas on how to intervene to change the outcome.

Although identifying the BRCA genes that help to drive breast and ovarian cancers was a giant breakthrough, we still have no effective therapy for countering their malign influences. The intervening twenty-five years of effort have brought us to a new era of revealing the more subtle effects of variants. But the price we pay for unveiling the complete picture is perceiving just how tough is the therapeutic challenge.


Qiuyin Cai, et al. (2014). Genome-wide association analysis in East Asians identifies breast cancer susceptibility loci at 1q32.1, 5q14.3 and 15q26.1. Nature Genetics 46, 886–890. doi:10.1038/ng.3041.

Risk Assessment

For UK readers a title that instantly raises the spectre of the ’Elf & Safety police and the annoyance, irritation and amusement generated by the seemingly ubiquitous injunctions of their minions. Even my department is not spared, the harbinger of warm weather invariably being an email reminding us that this is no reason for abandoning the rule that at all times we should wear a lab coat – though, to be fair, our local enforcer usually includes the cheeky inference that we retain the option of going naked underneath. Ah, The Joy of Science! ’E & S’s reputation comes, of course, from periodically making the headlines by banning a centuries-old tradition in some rustic backwater involving such fun activities as rolling cheeses down a hill.

Stuart Kettell and sprout

Stuart Kettell and sprout

Mind you, they’ve slipped up recently by allowing Stuart Kettell to push a Brussels sprout up Mount Snowdon with his nose. As that’s 3,560ft (vertically) he probably did lasting damage to his knees, to say nothing of his hooter, as well as inflicting grievous bodily harm on 22 sprouts (they wear out on the basalt, obviously). By his own admission, he’s probably mad – but he did at least raise some money for Macmillan Cancer Support.


But why are we bothered about assessing risk?

Setting the above entertainment to one side, estimating risk can be a really serious business and never more so than when it comes to cancer. It’s an especially contentious, long-running issue for breast cancer and both in Betrayed by Nature and more recently in Behind the Screen I tried to crystallize some clear guidelines from the vast amount of available info. In short these were: ignore commercial plugs for thermography – the only test to go for is mammography – i.e. X-ray imaging to find breast cancer before a lump can be felt. And the simple message you were relieved to read in BbN was that, whilst the matter is controversial, if you are offered screening, accept – but be aware that the method is not perfect. There’s a small risk that a cancer may be missed and a bigger chance that something abnormal but harmless will be picked up – a signal for intervention (by surgery and drugs) and that, in those cases, would be unnecessary.

And we’re revisiting this question?

Because there have been some recent contributions to the debate that might well have increased confusion and concern in equal measure for women who are desperately trying to make sense of it all. The most controversial of these comes from a panel of experts (The Swiss Medical Board) who reviewed the history of mammography screening – and recommended that the current programmes in Switzerland should be phased out and not replaced.

Needless to say, their report caused a furore, not only in Switzerland, with experts damning its conclusions as ‘unethical’ – mainly because they ran counter to the consensus that screening has to be a good thing.

So what did the Swiss Big Cheeses point out to get into such hot water? Their view after considering the cumulative evidence was that systematic mammography might prevent about one breast cancer death for every 1,000 women screened. However, two other things struck them. First, it was not clear that this result outweighed the disadvantages of screening – what are inelegantly referred to as the ‘harms’ – the detection and treatment of something ‘abnormal but harmless’ mentioned earlier. Second that, on the basis of a survey by American group, women had a grossly optimistic idea of the benefits of mammography.

Good versus bad

Two of the weightiest bits of evidence that led them to conclude that screening does more harm than good were studies that had combined several independent investigations – what’s called a meta-analysis – which is a way of increasing your sample size and hence getting a more meaningful answer. One of these (The Independent United Kingdom Panel on Breast Cancer Screening) pulled together 11 trials from which it emerged that women invited to screening had a reduction of about 20% in their risk of dying from breast cancer compared with controls who were not offered screening. So far so good. However, inevitably there were differences in methods between the trials, which made the UK Panel very cagey about drawing more specific conclusions but their best estimate was that, for every 10,000 UK women aged 50 years invited to screening for the next 20 years, 43 deaths from breast cancer would be prevented and 129 cases would be over-diagnosed. Over-diagnosis means detection of cancers that would never have been emerged during the lifetime of the individuals and these healthy women will be needlessly subjected to some combination of surgical interventions, radiotherapy and chemotherapy.

The second combined study is from The Cochrane Collaboration, the trials involving more than 600,000 women. Their review also emphasized the variation in quality between different studies and noted that the most reliable showed that screening did not reduce breast cancer mortality. However, less rigorous methods introduced bias towards showing that screening did reduced breast cancer mortality. In this sort of trial “less rigorous” relates particularly to the problem of ensuring that the two groups of subjects are truly randomized – i.e. that nothing influences whether a woman is assigned to receive screening mammograms or not. This is much harder than it sounds, mainly because human beings do the assigning so there is always a chance of either a genuine mistake or a flaw in the design of a particular study. One simple example of how the best laid plans … The consent form for a study specifically states that women are assigned, at random, to either the mammography or no mammography group. Women are then examined by a specially trained nurse. However, if these two steps are reversed, assignment may be biased by the findings of the examination. Precisely such a failure to adhere to a protocol has been revealed in at least one study.

Making the liberal assumption that screening reduces mortality by 15% and that over-diagnosis occurs at a rate of 30%, they estimated that for every 2000 women invited for screening over 10 years, one will avoid dying of breast cancer and 10 will be treated unnecessarily. In addition, false alarms will subject 200 women to prolonged distress and anxiety.

All of which explains why, taking everything into consideration, the Big Cheeses recommended that the Swiss abandon mammography screening.

MammogramWhat does the NHS say?

Actions speak louder than words and in the UK women aged 50 to 70 are invited for mammography screening every three years. By way of explanation, the NHS document (NHS breast screening: Helping you decide) says that for every 200 screened about one life is saved from breast cancer. The American Cancer Society recommends screening annually from age 40 – so it’s clear that Britain and the USA are firmly in favour.

You will have noted that the NHS figure of one saved for every 200 screened is seriously at odds with the findings summarized above and they don’t say where it comes from. However, they are clear about the critical point in saying “for every 1 woman who has her life saved from breast cancer, about three women are diagnosed with a cancer that would never have become life-threatening.”

Misplaced optimism

It will be obvious by now that attaching precise numbers to the effects of screening is next to impossible but the overall message is clear. At best screening yields a small reduction in breast cancer deaths but this comes with a substantially greater number of women who are treated unnecessarily – hence the Swiss position that it is ethically difficult to justify a public health program that does more harm than good.

It’s a bit difficult to assess just how knowledgeable women are about the benefits of mammography screening but one study that tried came up with some positively alarming pointers. A telephone survey of more than 4000 randomly chosen females over 15 years of age in the USA, the UK, Italy and Switzerland revealed that a substantial majority believed that (i) screening prevents or reduces the risk of getting breast cancer, (ii) screening at least halves breast cancer mortality, and (iii) 10 years of regular screening prevents 10 or more breast cancer deaths per 1000 women.

A clear conclusion?

Rates of breast cancer mortality are declining. Hooray! And the five-year survival rate in developed countries is now about 90%. Hooray again! It seems probable that this trend is more though improved treatments and greater awareness – leading to early detection – than because of screening. Nevertheless, all that doesn’t alter the fact that where women are offered the choice they need to be as well informed as possible. The weaknesses of the telephone survey are obvious but the implication that misconceptions are widespread indicates that we need to do much better at explaining the facts of mammography screening.


Biller-Andorno N. and Jüni P. (2014). Abolishing mammography screening programs? A view from the Swiss Medical Board. New England Journal of Medicine 370:1965-7.

Independent UK Panel on Breast Cancer Screening. (2012). The benefits and harms of breast cancer screening: an independent review. Lancet 380:1778-86.

Gøtzsche, P.C. and Jørgensen, K.J. (2013). Screening for breast cancer with mammography. Cochrane Database Syst Rev; 6:CD001877.

Domenighetti G, D’Avanzo B, Egger M, et al. (2003). Women’s perception of the benefits of mammography screening: population-based survey in four countries. Int J Epidemiol., 32:816-21.

Keeping Cancer Catatonic

Over a century ago there lived in London an astute physician by the name of Stephen Paget. He was one of those who may or may not be envied in being part of a super-talented family. His Dad, Sir James Paget, was pals with Charles Darwin and, together with Rudolph Virchow, laid the foundations of modern pathology, though today medical students usually encounter his infinitesimal immortality through several diseases that bear his name. These include a rare condition, Paget’s disease of the breast, in which malignant cells form in the skin of the nipple creating an itchy rash, usually treatable by surgery. His Uncle George had been Regius Professor of Physic at Cambridge and he had several brothers, two of whom became bishops. Fortunately Stephen continued the medical thread of the family and Paget’s passion became breast cancer.

A Key Question

Paget had that invaluable scientific gift of being able to pinpoint a key question – in his case ‘What is it that allows tumour cells to spread around the body?’ – and it was such a good question that to this day we don’t have a complete answer. That it happens had been known long before the appearance of Paget Junior. René-Théophile-Hyacinthe Laënnec, French of course, in the early years of the 19th century described how skin cancer could spread to the lungs before he went on to invent the stethoscope in 1816. The mother of this invention was a young lady whom he described as having a ‘great degree of fatness’ that made her heartbeat inaudible by the then conventional method of placing ear to chest. Using a piece of paper rolled into a tube as a bridge, Laënnec was somewhat taken aback that the beat was more distinct than he’d ever heard before. Needless to say, medicine being a somewhat reactionary profession, not all its practitioners had ears tuned to receive this advance with glee but in the end, of course, it caught on and we can therefore award Laënnec first prize in reducing human cumulative embarrassment. It was another French surgeon, Joseph Récamier, who subsequently coined the term metastasis, (to be precise ‘métastase’) to describe the formation of secondary growths derived from a primary tumour.

Early Ideas about Metastasis

The notion that primary tumours could give rise to a diaspora gradually took root but it was not until 1840 that the Munich-born surgeon Karl Thiersch showed that it was actually cells – malignant cells – that wandered off and found new homes. Rudolf Virchow had come up with the idea that spreading was via a ‘juice’ released by primaries that somehow converted normal cells at other sites into tumours. As Virchow was jolly famous, having not only made the study of disease into a science but also discovered leukemia, it took a while for Thiersch to triumph, notwithstanding the evidence of Laënnec and others. Funnily enough, and as quite often happens in scientific arguments, it now looks as though both were right if for ‘juice’ you substitute ‘messengers’ – that is, chemicals dispatched by tumour cells – as we shall see.

Paget’s attention had been drawn to this subject through his observations on breast cancer, and he’d taking as a starting point the most obvious question: ‘How do tumour cells know where to stick?’ Or, as he elegantly phrased it in a landmark paper of 1889: ‘What is it that decides what organs shall suffer in a case of disseminated cancer?’ The simplest answer would be that it just depends on anatomy: when cells leave a tumour and get into the circulation they stick to the first tissue they meet. But in looking at over 700 cases he’d found this just didn’t happen and that secondary growths often appeared in the lungs, kidneys, spleen and bone. Paget acknowledged the uncommonly prescient suggestion a few years earlier by Ernst Fuchs that certain organs may be ‘predisposed’ for secondary cancer and concluded that ‘the distribution of secondary growths was not a matter of chance.’ This led him to a botanical analogy for tumour metastasis: ‘When a plant goes to seed, its seeds are carried in all directions; but they can only live and grow if they fall on congenial soil.’ From this, then, emerged the ‘seed and soil’ theory of metastasis, its great strength being the image of interplay between tumour cells and normal cells, their actions collectively determining the outcome. Rather charmingly, Paget concluded his paper with: ‘The best work in the pathology of cancer is now done by those who are studying the nature of the seed. They are like scientific botanists; and he who turns over the records of cases of cancer is only a ploughman, but his observation of the properties of the soil may also be useful.’


Bookmarking cancer: Primary tumours mark sites around the body to which they will spread (metastasize) by sending out chemical signals that create sticky ‘landing sites’ (red protein A) on target cells. Cells released from the bone marrow carry proteins B and C. B attaches to A and tumour cells ‘land’ on C. Cells may remain quiescent in a new site for years or decades, their growth suppressed by signals (e.g., TSP-1) released from nearby blood vessels. Only when appropriate activating signals dominate (e.g., TGFbeta) is secondary tumour growth switched on.

Finding a Landing Strip

For well over a century Paget’s aphorism of  ‘seed and soil’ pretty well summed up our knowledge of metastasis. It’s obvious that before any rational therapy can be designed we need to unravel the molecular detail but we’ve had to wait until the twenty-first century for any further significant insight into the process. As so often in science, the hold-up has been largely due to waiting for the appropriate combination of methods to be developed – in this case fluorescently tagged antibodies to detect specific proteins in cells and tissues and genetically modified mice.

In the forefront of this pursuit has been David Lyden and his colleagues at Weill Cornell Medical College and other centers and their most extraordinary finding is that cells in the primary tumour release proteins into the circulation and these, in effect, tag what will become landing points for wandering cells. Extraordinary because it means that these sites are determined before any tumour cells actually set foot outside the confines of the primary tumour. These are chemical messengers rather equivalent to Virchow’s ‘juice’: they don’t change normal cells into tumour cells but they do direct operations. However, it’s a bit more complicated because, in addition to sending out a target marker, tumours also release proteins that signal to the bone marrow. This is the place where the cells that circulate in our bodies (red cells, white cells, etc.) are made from stem cells. The arrival of signals from the tumour causes some cells to be released into the circulation; these carry two protein markers on their surface: one sticks to the pre-marked landing site, the other to tumour cells once they appear in the circulation. It’s a double-tagging process: the first messenger makes a sticky patch for bone marrow cells that appear courtesy of another messenger, and they become the tumour cell target. It’s molecular Velcro: David Lyden calls it ‘cellular bookmarking.’

Controlling Metastatic Takeoff

Tumour cells that find a new home in this way, after they’ve burrowed out of the circulation, could in principle then take off, growing and expanding as a ‘secondary.’ However, and perhaps surprisingly, generally they do the exact opposite: they go into a state of hibernation, remaining dormant for months or years until some trigger finally sets them off. The same group has now modeled this ‘pre-metastatic niche’ for human breast cancer cells, showing that the switch between dormancy and take-off is controlled by proteins released by nearby blood vessels. The critical protein that locks tumour cells into hibernation appears to be TSP-1 (thrombospondin-1). As long as TSP-1 is made by the blood vessel cells metastatic growth is suppressed. This effect is overridden by stimuli that turn on new vessel growth and in so doing switch secretion from TSP-1 to TGFB (transforming growth factor beta). Now proliferation of the disseminated tumour cells is activated and the micro-metastasis becomes fully malignant. It should be said that this is a model system and may possibly bear little relation to what goes on in real tumours. However, the fact that specific proteins that are, moreover, highly plausible candidates, can control such a switch strongly suggests its relevance and also highlights potential targets for therapeutic manipulation.

Stranger Than Fiction

The system for directing tumour cells to a target seems extraordinarily elaborate. Given that tumour cells cannot evolve in the sense of getting better at being metastatic – they just have to go with what they’ve got – how on earth might it have come about? We don’t know, but the most likely explanation is that they are taking advantage of natural defense mechanisms. Although tumours start from normal cells, the first reaction of the body is to see them as ‘foreign’ – much as it does bugs that get into a cut – and the response is to switch on inflammation and an immune response to eliminate the ‘invader.’

Perhaps what is happening in these mouse models is that the proteins released by the tumour cells are just a by-product of the genetic disruption in cancer cells. Nevertheless, they may signal ‘damage somewhere in the body’. That at least would explain why the bone marrow decides to release cells that are, in effect, a response to the tumour. The second question is trickier: Why should tumours release proteins that mark specific sites? We’ve known since Dr. P’s studies that cells from different tumours do indeed head for different places and it may just be that the messengers arising in the genetic mayhem happen to reflect the tissue of origin. The mouse models, encouragingly, show that the target changes with tumour type (e.g., swap from breast to skin and the cells go somewhere else). In other words, tumours send out their own protein messengers that set up sticky landing strips in different places around the circulation.

As for take-off, it may be that newly arrived tumour cells simply adapt to the style of their neighborhood. By and large, the blood vessels are pretty static structures: they don’t go in for cell proliferation unless told to do so by specific signals, as happens when you get injured and need to repair the damage. TSP-1 appears to be a ‘quiescence’ signal, telling cells to sit tight. The switch to proliferation comes when that signal is overcome by TGFB, activating both blood vessels and tumour. All of which would delight Paget: not only is our expanding picture consistent with ‘seed and soil’ but the control by local signals over what happens next makes his rider that ‘observation of the properties of the soil may also be useful’ spot on.


Kaplan, R.N., Riba, R.D., Zacharoulis, S., Bramley, A.H., Vincent, L., Costa, C., MacDonald, D.D., Jin, D.K., Shido, K., Kerns, S.A., Zhu, Z., Hicklin, D., Wu, Y., Port, J.L., Altork, N., Port, E.R., Ruggero, D., Shmelkov, S.V., Jensen, K.K., Rafii, S. and Lyden, D. (2005). VEGFR1-positive haematopoietic bone marrow progenitors initiate the pre-metastatic niche. Nature 438, 820-827.

Ghajar, C.M. et al. (2013). The perivascular niche regulates breast tumour dormancy. Nature Cell Biology 15, 807–817.



Signs of Resistance

In Beware of Greeks … we noted that in one sort of leukemia at least, tumour cells have come up with an extraordinary way of escaping from the bone marrow where they start life into the circulation where they cause trouble – by releasing pieces of their own DNA that then break down the retaining barrier.

Keeping track of tumors

Curious behaviour though it may be, there’s nothing new about the idea of cells shedding bits of their genetic code – that was first shown to happen over 60 years ago. What is novel is the evidence that not only does this happen in a variety of cancer cells but that modern methods enable those fragments to be isolated from just a teaspoonful of blood: the sequence of the DNA can then be determined – which gives the mutational signature of the original tumour. A remarkable development has now shown that repeating these steps over a period of time can reveal the response of secondary tumours (metastases) to drug treatment (chemotherapy).


One great advantage of this blood sampling method is that it is as near as makes no difference ‘non-invasive’. That is, it uses only a (small) blood sample and there’s no need for painful excavations to dig out tumour samples. The study, largely funded by Cancer Research UK, looked at three major cancers (breast, ovarian and lung) and identified specific mutations caused by drugs over a period of one to two years. For good measure they also took tumour samples to show that the mutation patterns found in circulating DNA did indeed represent what had gone on in the tumour itself. In other words, they had established what scientists like to call ‘proof of principle’ – i.e. we can do it!

There’s another more subtle advantage of this approach in that it gets round a problem we described in Molecular Mosaics: tumours are a mixture and the mutational signature differs depending on which bit you sample and sequence. The cell-free DNA fragments collected from blood are a gemisch – an averaged signature if you like – that may therefore give a better picture of the target for drug cocktails at any given time during tumour evolution.

Why is this so important?

There are two main reasons why it’s difficult to exaggerate the potential important of this step. The first is that metastasis accounts for over 90% of cancer deaths, the second that the fiendish ingenuity with which tumours negate chemotherapy, i.e. develop drug resistance, is one of the biggest challenges to successful treatment. So, the sooner changes that enable tumours to become insensitive to drugs can be detected the better in terms of adjusting the treatment regime. Even more exciting, however, is that notion that the DNA shed by cancers into the circulation may permit detection years or even decades earlier than is possible with any of the current methods (e.g., mammography) – with screening being carried put routinely from blood samples. Being even more optimistic, very early stage tumours may be particularly susceptible to appropriate drug combos, so that we might look forward to the day when chemotherapy replaces surgery as the first line of treatment for most cancers.


Murtaza, M. et al., (2013). Non-invasive analysis of acquired resistance to cancer therapy by sequencing of plasma DNA. Nature 497, 108–112.

Wake up at the back

Living with someone of the opposite sex, or getting married as it used to be known, is an interesting experience. One of the things you rapidly discover that your Mum never warned you about is that women are a distinct species.  You missed that revelation in your biology classes? Serves you right for snoozing on the back row but, as a recap of the evidence, consider the following. Species often show major differences in sensory perception – thus our cat is much better than I am at seeing in the dark, though he misses out a bit in daylight as cats don’t have colour vision. When it comes to hearing it’s a bit the other way round: most of the time you can shout at him til you’re hoarse with absolutely no effect – but one faint clink of a food bowl at the back door and, yet again, he’ll set a new Feline Fifty metres Steeplechase record from the front garden. And dogs, as is well known, hear frequencies way beyond what we can pick up.

Not in my lectures!!

The gentle sex has similarly evolved beyond what mere man can manage. Take colour, for example, at which men are, as we’ve noted, quite good – compared to cats. But, as you discover the first time you are taken ‘clothes shopping’ by your wife, other half, inamorata, partner, mistress or whatever, women have evolved far beyond merely spotting that blue is different from red and being able to recite Richard Of York (to remind themselves of the rainbow sequence). They see ‘combinations’ – so you are curtly informed that what has taken your fancy ‘just doesn’t go together’ in the sort of voice that adds ‘any nitwit can see that’ without the need to expend breath on the last seven syllables.

They’re at a similarly lofty level of evolution when it comes to sound. My lady wife avers that I snore – all the time (when asleep, that is) and very loudly. So much so that she tends to use a bed at the opposite end of the house for sleeping and only ventures within sonar range for other purposes. I’d always explained this behaviour as a manifestation of the amazing imagination possessed of the female that us boys are, of course, completely lacking. However, I’ve now come to appreciate that, like Fido (who sleeps in the kitchen), she simply has exquisitely sensitive aural apparatus. So maybe I do snore – but only very quietly or at ultra high frequency, so that I would be undetectable at rest to my own species and only my beloved and the dog would know what was going on (oh, and the cat because he can see the heaving chest).

Which is very reassuring since some fellows at the Universities of Wisconsin and Barcelona have got together to discover that snoring makes you nearly five times more likely to develop cancer. Strictly the problem is sleep disordered breathing (SDB) – which happens when there’s some kind of blockage of the upper airway and, apart from disrupting sleep, it can make you snore. Of course, there’s evidence that sleep disruption can contribute to all sorts of problems from heart disease to car crashes but this is the first study making a link to cancer.

No problem for me (discounting the wife’s super sonar) but should real, habitual snorers panic? Please don’t for most of the usual reservations to this type of study apply – relatively small numbers (1522) for example. The volunteers came from an alluringly named body of men and women called the Wisconsin Sleep Cohort, set up in 1988 for prospective studies of sleep disorders. In fact the interesting ones here are what we might call the Winsomniacs – the 365 of the Cohort who can’t do it rather than the majority of Badger State dreamers. Split in this case into sub-groups of SDB severity – the strongest association being with the most severe SDB. Although the authors did their best to allow for other factors (obesity – a common cause of SDB – diabetes, smoking, etc.) it’s almost impossible in this type of study to eliminate everything bar the one factor you’re focussing on.

The most frequent linked cancer was of the lung, followed by bowel, ovary, endometrial, brain, breast, bladder, and liver. And the cancer risk was up to four-fold greater for the worst afflicted.

Do the boffins have any helpful suggestions? Not really. Those unlucky enough to be severely affected can try a gadget called a continuous positive airway pressure device but, for the rest, console yourselves that the risk is small and the data so far are very preliminary. Put another way, you have more important things to think about – like finding a partner (preferably with sub-standard sonar detection capability) who loves you so much they’re willing to poke you in the ribs whenever you become aurally intrusive.


Javier Nieto, F.J. et al. (2012). Sleep-disordered Breathing and Cancer Mortality: Results from the Wisconsin Sleep Cohort Study. American Journal of Respiratory and Critical Care Medicine 186, Iss. 2, pp 190–194.

And Now For Something …

It’s not often that science and rugby collide (though to be fair they do a couple of times in BbN) but when it happens we should enjoy the spectacle. This item therefore comprises the cover page of my referees’ society newsletter (which goes by the name of Contact) for November. It is the work of the editor, Mr. Michael Dimambro, who has given immeasurable service to the game but is quite cheeky and has an inimitable style. Read on:

A Radiant Visitor

In an historic first, Cancer For All welcomes a guest, Stacey McGowan, who is a physicist just starting a Ph.D. on something called Proton Therapy. She is a member of the Department of Oncology in Cambridge and you can find out more about her in her blog but today she is going to take us into her world with a simple guide to radiotherapy in the treatment of cancer.

As undergraduate there was a lot of pressure to know what you wanted to do after graduation. I knew I wanted to stay in physics as it was what I loved; I also knew I wanted a job that meant something to me. I did not want to work in finance or for a defence company. At the time I also didn’t think I wanted to go into research! This seemed to have left me with two options, to work in the energy industry, or in medicine.

A lot of people, including my undergrad self, are unaware of medical physicists and their role in the hospital and in treating patients. After an inspiring talk at a careers event from a medical physicist working in the NHS I knew that this was what I wanted to do after graduation: I wanted to be a medical physicist.

There are three main methods for treating cancer; surgery, chemotherapy and radiotherapy. A patient will usually receive one or more of these methods as part of their treatment. Of the cures achieved about 49% of them involve surgery, 11% involve chemotherapy and 40% involve radiotherapy. However of the NHS’s cancer budget surgery costs around 22%, chemotherapy 18% and radiotherapy just 5%. This makes radiotherapy both a successful treatment option, sometimes on its own but usually in combination with surgery or chemotherapy, and it is extremely cost effective. Despite this many people don’t really know what radiotherapy is and the prospect of it as a treatment often makes patients apprehensive. As much as radiation sounds scary, we are exposed to it all the time in nature from the sun and soil and nowadays in our homes from electrical devices including Wi-Fi and mobile phones. In addition, we use it in many diagnostic applications including X-rays, CT scanners and nuclear medicine.

The difference between the radiation used for cancer treatment and that received from other sources is in the amount of radiation, or dose, delivered. When I talk about dose, think of it in the same way you would any other type of medicine. An oncology doctor will prescribe a course of radiotherapy with a specific dose to be delivered to the patient every weekday for between 4 and 6 weeks. The radiation is delivered in the form of X-rays – highly energetic particles of light – delivered at higher energies and doses than those used to image a broken bone (Editor’s enlightenment: physicists tend to use the word ‘light’ to mean electromagnetic radiation of any wavelength – not just what the eye sees). To create such highly energetic light we need a powerful machine that can also precisely deliver the X-rays to the part of the patient where the cancer lies. This machine is known to the medical community as a linac, and to the scientific community as a linear accelerator!

The linacs used in the hospital differ from those used in physics research as medical linacs have a very different role and it is the medical physicists’ job to ensure they work as intended. The X-rays delivered to the patient will harm cells in their body, both cancerous and healthy, by damaging their DNA. It is extremely important that the cancer cells receive the dose necessary to kill them so that they cannot continue to grow, resulting in a cure. It is also a priority that healthy tissue receives the smallest possible radiation dose to ensure a low chance of long term side effects. To accomplish these goals linacs are designed to rotate about the patient so that the tumour can be targeted from more than one direction. Treatment is usually delivered in daily doses (known as fractions) over a period of a few weeks because healthy cells are better at repairing damage to their DNA than cancer cells, so they can recover from each dose, whereas damage will accumulate in the tumour cells. Cumulative DNA damage leads to cell death, stopping the cancer in its tracks.

Linacs can also shape the beam so that it will match the shape of the tumour, shielding the adjacent healthy tissue from the highest radiation doses. To produce such patient-specific and intricate treatments powerful computer programs are used to design the treatment based on images of the patient (usually CT scans). Oncologists and physicists will work together, distinguishing cancer tissue from healthy, choosing beam directions and designing beam shapes to ensure that the patient receives the optimal treatment.

Many types of cancers respond to radiotherapy including those of the lung, breast, prostate, brain and spine and the method can be used to treat both adults and children. The short term side effects from radiotherapy vary depending on the region being treated. For example, radiation of the abdominal area may cause digestive and bowel discomfort or if the head and neck is the target, the patient may experience difficulty swallowing and develop a dry mouth. Generally radiotherapy can lead to tiredness, nausea and skin irritation in the targeted areas. Long term side effects can include secondary cancer, more probably in young patients, and growth problems in children.

The future of radiotherapy in the NHS is to use of protons and not X-rays to deliver radiation for specific types of cancer. The nature of protons makes the aim of cure without complication more achievable and is the topic of my PhD research.

Unlike X-rays, protons have a finite range (we can choose where they stop) which reduces the amount of radiation exposure to the patient, making this form of therapy especially beneficial for spine and brain tumours in adults and for most cancers in children. Proton therapy is particularly attractive for treating childhood cancers because it is less likely than conventional radiotherapy to cause growth defects and other health complications, including the development of cancers in later life.

Despite the UK lacking the facilities necessary to treat cancer using proton radiotherapy, a limited number of NHS patients are currently offered this option abroad as part of the NHS Proton Overseas Programme. The Government also announced in April 2012 that two proton centres will be established in England, in Manchester and in London. It is hoped that these will start to treat patients by early 2017.

Stacey McGowan

Department of Oncology, University of Cambridge

Behind the Screen

Anyone who’s romped through to the closing pages of Betrayed by Nature will (we hope) be grateful to have come upon a succinct summary of the pros and cons of screening, particularly for breast cancer. In the UK screening (i.e. mammography: see Breast Cancer – Seeing Red, Jan 2012 for explanation) is offered to women aged 50-70 every three years – so clearly ‘we’ think it’s a Good Thing. However, the waters are less than crystal clear because several studies have concluded that as many as one in three cancers identified by mammography would not cause any symptoms during the lifetime of the patient  – and suggested that countries should spend the money on other things. The simple message you were relieved to read in BbN was that, whilst the matter is controversial, if you are offered screening, accept – whilst being aware that a ‘positive’ is not always a signal for intervention (by surgery and drugs) and that in deciding on a course of action you should be guided by the best advice your clinicians can give.

Photo: Alamy

Trying to bring resolution to this complicated and important matter, yet another ‘official review’ has just appeared in The Lancet – predictably accompanied by some absurdly inflammatory press headlines. So, ignoring them and your groans – and because it is important – can we recap the key points and reassess the clear BbN message? Of course we can – that’s what we do in “Cancer for All.”

Any type of screen for signs of cancer has two problems for those on the receiving end. First it will miss some and second it will sometimes pick up things that, although abnormal, will never become life-threatening. This latest report estimates that screening reduces the relative risk by 20%, i.e. prevents one breast cancer death for every 235 women invited for screening, equivalent to 43 preventions per 10,000 women aged 50 who are screened over the next 20 years. The downside is that about 130 women in every 10,000 are what is called ‘overdiagnosed’: they receive treatment for something that could simply be left alone. Of course that’s highly undesirable as well as stressful and unpleasant for the patient. However, it is rarely fatal – and at least carries an element of reassurance that they won’t develop breast cancer.

So, in the light of the newest info, is BbN’s take in need of modification? No, it’s just fine. What a relief!! But just bear a couple of other points in mind: the sensitivity of screening is gradually improving and dramatic improvements in analysing tumours at the molecular level mean that ‘overdiagnosing’ will decline. As we’ve pointed out, the system isn’t perfect – then neither are we or we wouldn’t get cancers – but it’s heading in the right direction.


Independent UK Panel on Breast Cancer Screening. The benefits and harms of breast cancer screening: an independent review. The Lancet October 2012.

Not another ‘Great Cancer Breakthrough’!!!

Since I started writing Betrayed by Nature and this accompanying blog, my take on science reporting in the ‘media’ has undergone considerable change. I guess most of it used to wash over me: now I feel obliged to read it, with a view to making sense of it from the point of view of non-scientists. The dramatic headlines generally fall into two groups – one telling us what not to do/eat, the other revealing how wonderful scientists are.

I admitted recently (Whose side are you on?) that as far as the eating, drinking and exercising injunctions go, I’m beginning to side with those who just wish ‘they’d’ keep quite and let us get on doing whatever we want to do. The other group is trickier because there’s almost always some interesting stuff beneath the press rhetoric. The latest Scientists hail revolutionary breast cancer breakthrough is a case in point. The media coverage refers to a paper just published in Nature that has applied the formidable power of nucleic acid sequencing methods to a large number of breast tumours. The sheer amount of information generated is almost stupefying and the efforts of folk – called ‘bioinformaticists’ – who make sense of the raw data are remarkable.

But the overall message is relatively simple. Like every other tumour, each breast cancer is different at the level of the molecular changes it carries. However, the DNA sequences of genes and the extent to which they are ‘switched on’ to make RNA and protein (‘gene expression’) permit these tumours to be sub-divided into 10 major categories.

So is this a ‘Great Cancer Breakthrough’. Not really. It’s a terrific piece of science but it’s just one more small step towards better designed therapies that’s come from using the wonderful methods that have become available over the last ten years or so.

Did the guys who did the work use the hyped-up language of Mr. Connor in The Independent? Not exactly. This paper is a stunning technical tour-de-force – but the authors merely sign off with the comment that their work ‘reveals novel subgroups that should be the target of future investigation’.


Curtis, C., Shah, S.P., Chin, S.-F., Turashvili, G., Rueda, O.M. et al. (2012). The genomic and transcriptomic architecture of 2,000 breast tumours reveals novel subgroups. Nature (2012) doi:10.1038/nature10983