Sticky Cancer Genes

 

In the previous blog I talked about Breath Biopsy — a new method that aims to detect cancers from breath samples. I noted that it could end up complementing liquid biopsies — samples of tumour cell DNA pulled out of a teaspoon of blood — both being, as near as makes no difference, non-invasive tests. Just to show that there’s no limit to the ingenuity of scientists, yet another approach to the detection problem has just been published — this from Matt Trau and his wonderful team at The University of Queensland.

This new method, like the liquid biopsy, detects DNA but, rather than the sequence of bases, it identifies an epigenetic profile — that is, the pattern of chemical tags (methyl groups) attached to bases. As we noted in Cancer GPS? cancer cells often have abnormal DNA methylation patterns — excess methylation (hypermethylation) in some regions, reduced methylation in others. Methylation acts as a kind of ‘fine tuner’, regulating whether genes are switched on or off. In the methylation landscape of cancer cells there is an overall loss of methylation but there’s an increase in regions that regulate the expression of critical genes. This shows up as clusters of methylated cytosine bases.

Rather helpfully, a little while ago in Desperately SEEKing … we talked about epigenetics and included a scheme showing how differences in methylation clusters can identify normal cells and a variety of cancers and how these were analysed in the computer program CancerLocator.

The Trau paper has an even better scheme showing how the patterns of DNA decoration differ between normal and cancer cells and how this ‘methylscape’ (methylation landscape) affects the physical behaviour of DNA.

How epigenetic changes affect DNA. Scheme shows methylation (left: addition of a methyl group to a cytosine base in DNA) in the process of epigenetic reprogramming in cancer cells. This change in the methylation landscape affects the solubility of DNA and its adsorption by gold (from Sina et al. 2018).

Critically, normal and cancer epigenomes differ in stickiness — affinity — for metal surfaces, in particular for gold. In a clever ploy this work incorporated a colour change as indicator. We don’t need to bother with the details — and the result is easy to describe. DNA, extracted from a small blood sample, is added to water containing tiny gold nanoparticles. The colour indicator makes the water pink. If the DNA is from cancer cells the water retains its original colour. If it’s normal DNA from healthy cells the different binding properties turns the water blue.

By this test the Brisbane group have been able to identify DNA from breast, prostate and colorectal cancers as well as from lymphomas.

So effective is this method that it can detect circulating free DNA from tumour cells within 10 minutes of taking a blood sample.

The aim of the game — and the reason why so much effort is being expended — is to detect cancers much earlier than current methods (mammography, etc.) can manage. The idea is that if we can do this not weeks or months but perhaps years earlier, at that stage cancers may be much more susceptible to drug treatments. Trau’s paper notes that their test is sensitive enough to detect very low levels of cancer DNA — not the same thing as early detection but suggestive none the less.

So there are now at least three non-invasive tests for cancer: liquid biopsy, Breath Biopsy and the Queensland group’s Methylscape, and in the context of epigenetics we should also bear in mind the CancerLocator analysis programme.

Matt Trau acknowledges, speaking of Methylscape, that “We certainly don’t know yet whether it’s the holy grail for all cancer diagnostics, but it looks really interesting as an incredibly simple universal marker for cancer …” We know already that liquid biopsies can give useful information about patient response to treatment but it will be a while before we can determine how far back any of these methods can push the detection frontier. In the meantime it would be surprising if these tests were not being applied to age-grouped sets of normal individuals — because one would expect the frequency of cancer indication to be lower in younger people.

From a scientific point of view it would be exciting if a significant proportion of ‘positives’ was detected in, say, 20 to 30 year olds. Such a result would, however, raise some very tricky questions in terms of what, at the moment, should be done with those findings.

Reference

Abu Ali Ibn Sina, Laura G. Carrascosa, Ziyu Liang, Yadveer S. Grewal, Andri Wardiana, Muhammad J. A. Shiddiky, Robert A. Gardiner, Hemamali Samaratunga, Maher K. Gandhi, Rodney J. Scott, Darren Korbie & Matt Trau (2018). Epigenetically reprogrammed methylation landscape drives the DNA self-assembly and serves as a universal cancer biomarker. Nature Communications 9, Article number: 4915.

Advertisements

Food Fix For Pharma Failure

 

If you held a global quiz, Question: “Which biological molecules can you name?” I guess, setting aside ‘DNA‘, the top two would be insulin and glucose. Why might that be? Well, the World Health Organization reckons diabetes is the seventh leading cause of death in the world. The number of people with diabetes has quadrupled in the last 30 years to over 420 million and, together with high levels of blood glucose (sugar), it kills nearly four million a year.

There are two forms of diabetes: in both the level of glucose in the blood is too high. That’s normally regulated by the hormone insulin, made in the pancreas. In Type 1 diabetes insulin isn’t made at all. In Type 2 insulin is made but doesn’t work properly.

When insulin is released into the bloodstream it can ‘talk’ to cells by binding to protein receptors that span cell membranes. Insulin sticks to the outside, the receptor changes shape and that switches on signalling pathways inside the cell. One of these causes transporter molecules to move into the cell membrane so that they can carry glucose from the blood into the cell. When insulin doesn’t work it is this circuit that’s disrupted.

Insulin signalling. Insulin binds to its receptor which transmits a signal across the cell membrane, leading to the activation of the enzyme PIK3. This leads indirectly to the movement of glucose transporter proteins to the cell membrane and influx of glucose.

So the key thing is that, under normal conditions, when the level of blood glucose rises (after eating) insulin is released from the pancreas. Its action (via insulin receptors on target tissues e.g., liver, muscle and fat) promotes glucose uptake and restores normal blood glucose levels. In diabetes, one way or another, this control is compromised.

Global expansion

Across most of the world the incidence of diabetes, obesity and cancer are rising in parallel. In the developed world most people are aware of the link between diabetes and weight: about 90% of adults with diabetes are overweight or obese. Over 2 billion adults (about one third of the world population) are overweight and nearly one third of these (31%) are obese — more than the number who are underweight. The cause and effect here is that obesity promotes long-term inflammation and insulin resistance — leading to Type 2 diabetes.

Including cancer

The first person who seems to have spotted a possible connection between diabetes and cancer was the 19th-century French surgeon Theodore Tuffier. He was a pioneer of lung and heart surgery and of spinal anaesthesia and he’s also a footnote in the history of art by virtue of having once owned A Young Girl Reading, one of the more famous oil paintings produced by the prolific 18th-century artist Jean-Honoré Fragonard (if you want to see it head for the National Gallery of Art in Washington DC). Tuffier noticed that having type 2 diabetes increased the chances of patients getting some forms of cancer and pondered whether there was a relationship between diabetes and cancer.

It was a good question then but it’s an even better one now when this duo have become dominant causes of morbidity and mortality worldwide.

We now know that being overweight increases the risk of a wide range of cancers including two of the most common types — breast and bowel cancers. Unsurprisingly, the evidence is also clear that diabetes (primarily type 2) is associated with increased risk for some cancers (liver, pancreas, endometrium, colon and rectum, breast, bladder).

With all this inter-connecting it’s perhaps not surprising that the pathway by which insulin regulates glucose also talks to signalling cascades involved in cell survival, growth and proliferation — in other words, potential cancer initiators. The central player in all this is a protein called PIK3 (it’s an enzyme that adds phosphate groups (so it’s a ‘kinase’) to a lipid called phosphatidylinositol bisphosphate, an oily, water-soluble component of the plasma membrane). It’s turned out that PIK3 is one of the most commonly mutated genes in human cancers — e.g., PIK3 mutations occur in 25–40% of all human breast cancers.

Signalling pathways switched on by mutant PIK3. A critical upshot is the activation of cell survival and growth that leads to cancer.

Accordingly, much effort has gone into producing drugs to block the action of PIK3 (or other steps in this signal pathway). The problem is that these have worked as cancer treatments either very variably or not at all.

The difficulty arises from the inter-connectivity of signalling that we’ve just described: a drug blocking insulin signalling causes the liver to release glucose and prevents muscle and fats cells from taking up glucose. Result: blood sugar levels rise (hyperglycaemia). This effect is usually transient as the pancreas makes more insulin that restores normal glucose levels.

Blockade of mutant PIK3 by an inhibitor. This blocks the route to cancer but glucose levels rise in the circulation (hyperglycaemia) promoting the release of insulin (top). Insulin can now signal through the normal pathway (bottom), overcoming the effect of the anti-cancer drug. Note that the cell has two copies of the PIK3 gene/protein, one of which is mutated, the other remaining normal.

Is our journey really necessary?

By now you might be wondering whether there is anything that makes grappling with insulin signaling worth the bother. Well, there is — and here it is. It’s a recent piece of work by Benjamin Hopkins, Lewis Cantley and colleagues at Weill Cornell Medicine, New York who looked at ways of getting round the insulin feedback response so that the effect of PIK3 inhibitors could be boosted.

Sketch showing the effect of diet on the potency of an anti-cancer drug in mice. The red line represents normal tumour growth. The black line shows the effect of PIK3 blockade when the mice are on a ketogenic diet: tumour growth is suppressed. On a normal diet the drug alone has only a slight effect on tumour growth. Similar results were obtained in a variety of model tumours (Hopkins et al., 2018).

They first showed that, in a range of model tumours in mice, insulin feedback caused by blockade of PIK3 was sufficient to switch on signalling even in the continued presence of anti-PIK3 drugs. The really brilliant result was that changing the diet of the mice could offset this effect. Switching the mice to a high-fat, adequate-protein, low-carbohydrate (sugar) diet essentially stopped the growth of tumours driven by mutant PIK3 treated with PIK3 blockers. This is a ketogenic (or keto) diet, the idea being to deplete the store of glucose in the liver and hence limit the rise in blood glucose following PIK3 blockade.

Giving the mice insulin after the drug drastically reduces the effect of the PIK3 inhibitor, supporting the idea that that a keto diet improves responses to PIK3 inhibitors by reducing blood insulin and hence its capacity to switch on signalling in tumour cells.

A few weeks prior to the publication of the PIK3 results another piece of work showed that adding the amino acid histidine to the diet of mice can increase the effectiveness of the drug methotrexate against leukemia. Methotrexate was one of the first anti-cancer agents to be made and has been in use for 70 years.

These are really remarkable results — as far as I know the first time diet has been shown to influence the efficacy of anti-cancer drugs. It doesn’t mean that all tumours with mutations in PIK3 have suddenly become curable or that the long-serving methotrexate is going to turn out to be a panacea after all — but it does suggest a way of improving the treatment of many types of tumour.

References

Hopkins, B.D. et al. (2018). Suppression of insulin feedback enhances the efficacy of PI3K inhibitors. Nature 560, 499-503.

Kanarek, N. et al. (2018). Histidine catabolism is a major determinant of methotrexate sensitivity. Nature 559, 632–636.

Taking Aim at Cancer’s Heart

 

Cancer is a unique paradox. At one level it’s as easy as can be to describe: damage to DNA (aka mutations) drives cells to behave abnormally — to make more of themselves when they shouldn’t.

But we all know that cancer’s fiendishly complicated — at least at the level of fine detail. Over the last decade or so the avalanche of sequenced DNA has revealed that every cell in a tumour is different: compare one cell to its neighbour and you’ll find variations in the individual units (the bases A, C, G & T) that make up the chains of DNA.

It’s a nightmare: every cancer is different so we need an infinite number of treatments to control or cure each one. Time to give up and retire to the pub.

Drivers and passengers

Not quite. DNA sequencing has also revealed that, amongst all the genetic mayhem, some mutations are more important than others. The movers and shakers have been dubbed ‘drivers’: those that come along for the ride are ‘passengers’. The hangers-on are heavily in the majority but, even so, several hundred drivers (i.e. mutated genes that give rise to abnormal proteins) have been identified. As it needs a group of drivers to work together to make a cancer we still have the problem that the number of critical combinations that can arise is essentially infinite.

One way of reducing the scale of the problem has been to look at what ‘driver’ proteins do in cells and to target those acting at key points to push cell proliferation beyond the normal.

Playing games

Just recently Giulio CaravagnaAndrea Sottoriva and colleagues at the Institute of Cancer Research, London and the University of Edinburgh have come up with a different approach. The idea goes back to the 1950s when a clever chap from Kansas by the name of Arthur Samuel came up with a program for IBM’s first commercial computer so that it could play draughts (checkers as our American friends call it) in its spare time. The program defined the patterns that could be formed by the pieces on the chequerboard so that, given enough of these, IBM 701 could indicate the optimal moves. Samuel called this machine learning, a precursor of the idea of artificial intelligence.

Perhaps the most famous moment in this saga came in 1997 when a later IBM computer, Deep Blue, beat the then world chess champion Garry Kasparov. Unsurprisingly, Kasparov was a bit miffed and accused IBM of cheating — to wit, getting a human to tell the machine what to do. Let’s hope that in the end he came to terms with the fact that Deep Blue could crank through 200 million positions per second and, however many games Grandmasters have in their heads, they can’t compete with that.

The cancer team realized that the mutations driving the evolution of cancer cells emerge as patterns in the sequence of DNA as a cell moves towards becoming independent of normal controls. Think of each cancer as a family tree of mutations, the key question being which branch leads to the most potent combination.

To pick out these patterns they applied a machine-learning approach known as transfer learning to the DNA sequences from a large number of cancers. They called this ‘repeated evolution in cancer’ — REVOLVER — aimed at picking out mutation patterns at the heart of cancer that foreshadow future genetic changes and can be used to predict how they will evolve.

Identifying patterns of mutation common to different tumours.

Samples are taken from different regions of a patient’s tumour (represented by the coloured dots). Their DNA sequences will have multiple variations that can mask underlying patterns of driver mutations present in some subgroups. The five trees show mutations picked up in those patients. REVOLVER uses transfer learning to screen the sequence data from many patients and pull out evolutionary trajectories shared by subgroups. The dotted red lines highlight common patterns that are represented in the lower strip. From Caravagna et al. 2018.

REVOLVER was applied to sequences from lung, breast, kidney and bowel cancers but there’s no reason it shouldn’t work with other tumours. The big attraction is that if these mini-sequence mutation patterns can be associated generally with how a given tumour develops they should help to inform treatment options and predict survival.

We have in the past referred to the ways cancers evolve as ‘genetic roulette’ — so perhaps it’s appropriate if game-playing computer programs turn out to be useful in teasing out behavioural clues.

Reference

Caravagna, G. et al. (2018). Detecting repeated cancer evolution from multi-region tumor sequencing data. Nature Methods 15, 707–714.

Keeping Up With Cancer

 

Cancer enthusiasts will know that there are zillions of web sites giving info on cancer stats — incidence, mortality, etc. — around the world. Notable is the World Health Organization’s Globocan, an amazing compilation of data on all cancers from every country. The Global Burden of Disease Cancer Collaboration audits diagnosis rates and deaths for 29 types of cancer around the world each year. Needless to say, this too is a vast undertaking involving hundreds of scientists around the world. The organizing genius is Dr. Christina Fitzmaurice of the Institute for Health Metrics and Evaluation in the University of Washington, Seattle and, under her guidance, their update for 2016 has just come out.

What’s new?

In 2016 there were 17.2 million people diagnosed with cancer. 8.9 million died from cancers. By 2030 the number of new cancer cases per year is expected to reach 24 million. Well, you knew the numbers were going to be big — almost incomprehensibly so. But here’s the real shaker: the 17.2 M is 28% up on the 2006 figure — yes, that’s a rise of more than one quarter.

Yearly global cancer deaths from 1990 to 2016.

The green line is total deaths per 100,000 people.

Red line: Cancer death rates taking account of the increase in world population.

Blue line: Age-standardized death rates: these are corrected for population size and age structure. Age-standardization therefore gives a better indication of the prevalence and incidence of underlying cancer risk factors between countries and with time without the influence of demographic and population structure changes.

The numbers on the vertical axis are deaths per 100,000 people. From Our World in Data.

Any real surprises?

No. An increase of more than one quarter in cancer cases does indeed make you think but the grim numbers are only what you would predict from looking at the trends over the last 40 years. The graph shows death rates that, of course, reflect incidence. The total figure (top) shows starkly how the rise in the population of the world and our increasing life-span is steadily pushing up the overall cancer burden.

So it’s a mega-problem but the trends are smooth and gradual. There’s been no drastic upheaval.

Global trends

Yearly global cancer deaths from 1990 to 2016 for the four major cancer types.

The numbers on the vertical axis are age-standardized death rates per 100,000 people.

Because age-standardization assumes a constant population age & structure it permits comparisons between countries and over time without the effects of a changing age distribution within a population. From Our World in Data.

The global trends in deaths from the four major cancers look mildly encouraging (above). However, these should not cheer us up too much. In the developed world there are some positives. In the USA, for example, over the last 17 years deaths from prostate are down from 31.6 to 18.9, for breast from 26.6 to 20.3 and for lung from 55.4 to 40.6 per 100,000 people. For bowel cancer there’s been a slight increase (4.1 to 4.8).

In the wider world, however, the really dispiriting thing shown by the latest figures is that the increases in incidence and deaths are greatest in low- and middle-income countries.

What can we do?

Lung cancer (includes cancers of the trachea and bronchi) remains the world’s biggest cancer killer, accounting for 20% of all deaths in 2016. Over 90% of these were caused by tobacco. In the UK and the USA lung cancer deaths in men have markedly declined as a result of widespread smoking bans, as the graph below shows, and the female figures have started to show a sight decline.

Lung cancer deaths per 100,000 by sex from 1950 to 2002 for the UK and the USA. From Our World in Data.

Asia contributes over half the global burden of cancer but the incidence in Asia is about half that in North America. However, the ratio of cancer deaths to the number of new cancer cases in Asia is double that in North America. Although the leading cause of death world-wide is heart disease, in China it is cancer. Every year more than four million Chinese are diagnosed with the disease and nearly three million die from it. Overall, tobacco smoking is responsible for about one-quarter of all cancer deaths in China. Nevertheless, Chinese smoking rates continue to rise and air pollution in the major cities is fuelling the problem.

The under-developed world, however, continues to be targeted by the tobacco industry and the successful promotion of their products means that there is no end in sight to one of mankind’s more bizarre and revolting forms of self-destruction.

There are, of course, other things within our control that contribute significantly to the global cancer burden. If only we could give everyone clean water to drink, restrict our red meat and processed food consumption and control our exposure to uv in sunlight we would cut cancer by at least one half.

If only …

Reference

The Global Burden of Disease Cancer Collaboration (2018). Global, Regional, and National Cancer Incidence, Mortality, Years of Life Lost, Years Lived With Disability, and Disability-Adjusted Life-Years for 29 Cancer Groups, 1990 to 2016: A Systematic Analysis for the Global Burden of Disease Study. JAMA Oncol. Published online June 2, 2018. doi:10.1001/jamaoncol.2018.2706.

Fantastic Stuff

 

It certainly is for Judy Perkins, a lady from Florida, who is the subject of a research paper published last week in the journal Nature Medicine by Nikolaos Zacharakis, Steven Rosenberg and their colleagues at the National Cancer Institute in Bethesda, Maryland. Having reached a point where she was enduring pain and facing death from metastatic breast cancer, the paper notes that she has undergone “complete durable regression … now ongoing for over 22 months.”  Wow! Hard to even begin to imagine how she must feel — or, for that matter, the team that engineered this outcome.

How was it done?

Well, it’s a very good example of what I do tend to go on about in these pages — namely that science is almost never about ‘ground-breaking breakthroughs’ or ‘Eureka’ moments. It creeps along in tiny steps, sideways, backwards and sometimes even forwards.

You may recall that in Self Help – Part 2, talking about ‘personalized medicine’, we described how in one version of cancer immunotherapy a sample of a patient’s white blood cells (T lymphocytes) is grown in the lab. This is a way of either getting more immune cells that can target the patient’s tumour or of being able to modify the cells by genetic engineering. One approach is to engineer cells to make receptors on their surface that target them to the tumour cell surface. Put these cells back into the patient and, with luck, you get better tumour cell killing.

An extra step (Gosh! Wonderful GOSH) enabled novel genes to be engineered into the white cells.

The Shape of Things to Come? took a further small step when DNA sequencing was used to identify mutations that gave rise to new proteins in tumour cells (called tumour-associated antigens or ‘neoantigens’ — molecular flags on the cell surface that can provoke an immune response – i.e., the host makes antibody proteins that react with (stick to) the antigens). Charlie Swanton and his colleagues from University College London and Cancer Research UK used this method for two samples of lung cancer, growing them in the lab to expand the population and testing how good these tumour-infiltrating cells were at recognizing the abnormal proteins (neo-antigens) on cancer cells.

Now Zacharakis & Friends followed this lead: they sequenced DNA from the tumour tissue to pinpoint the main mutations and screened the immune cells they’d grown in the lab to find which sub-populations were best at attacking the tumour cells. Expand those cells, infuse into the patient and keep your fingers crossed.

Adoptive cell transfer. This is the scheme from Self Help – Part 2 with the extra step (A) of sequencing the breast tumour. Four mutant proteins were found and tumour-infiltrating lymphocytes reactive against these mutant versions were identified, expanded in culture and infused into the patient.

 

What’s next?

The last step with the fingers was important because there’s almost always an element of luck in these things. For example, a patient may not make enough T lymphocytes to obtain an effective inoculum. But, regardless of the limitations, it’s what scientists call ‘proof-of-principle’. If it works once it’ll work again. It’s just a matter of slogging away at the fine details.

For Judy Perkins, of course, it’s about getting on with a life she’d prepared to leave — and perhaps, once in while, glancing in awe at a Nature Medicine paper that does not mention her by name but secures her own little niche in the history of cancer therapy.

References

McGranahan et al. (2016). Clonal neoantigens elicit T cell immunoreactivity and sensitivity to immune checkpoint blockade. Science 10.1126/science.aaf490 (2016).

Zacharakis, N. et al. (2018). Immune recognition of somatic mutations leading to complete durable regression in metastatic breast cancer. Nature Medicine 04 June 2018.

No It Isn’t!

 

It’s great that newspapers carry the number of science items they do but, as regular readers will know, there’s nothing like the typical cancer headline to get me squawking ‘No it isn’t!” Step forward The Independent with the latest: “Major breakthrough in cancer care … groundbreaking international collaboration …”

Let’s be clear: the subject usually is interesting. In this case it certainly is and it deserves better headlines.

So what has happened?

A big flurry of research papers has just emerged from a joint project of the National Cancer Institute and the National Human Genome Research Institute to make something called The Cancer Genome Atlas (TCGA). This massive initiative is, of course, an offspring of the Human Genome Project, the first full sequencing of the 3,000 million base-pairs of human DNA, completed in 2003. The intervening 15 years have seen a technical revolution, perhaps unparalled in the history of science, such that now genomes can be sequenced in an hour or two for a few hundred dollars. TCGA began in 2006 with the aim of providing a genetic data-base for three cancer types: lung, ovarian, and glioblastoma. Such was its success that it soon expanded to a vast, comprehensive dataset of more than 11,000 cases across 33 tumor types, describing the variety of molecular changes that drive the cancers. The upshot is now being called the Pan-Cancer Atlas — PanCan Atlas, for short.

What do we need to know?

Fortunately not much of the humungous amounts of detail but the scheme below gives an inkling of the scale of this wonderful endeavour — it’s from a short, very readable summary by Carolyn Hutter and Jean Claude Zenklusen.

TCGA by numbers. The scale of the effort and output from The Cancer Genome Atlas. From Hutter and Zenklusen, 2018.

The first point is obvious: sequencing 11,000 paired tumour and normal tissue samples produced mind-boggling masses of data. 2.5 petabytes, in fact. If you have to think twice about your gigas and teras, 1 PB = 1,000,000,000,000,000 B, i.e. 1015 B or 1000 terabytes. A PB is sometimes called, apparently, a quadrillion — and, as the scheme helpfully notes, you’d need over 200,000 DVDs to store it.

The 33 different tumour types included all the common cancers (breast, bowel, lung, prostate, etc.) and 10 rare types.

The figure of seven data types refers to the variety of information accumulated in these studies (e.g., mutations that affect genes, epigenetic changes (DNA methylation), RNA and protein expression, duplication or deletion of stretches of DNA (copy number variation), etc.

After which it’s worth pausing for a moment to contemplate the effort and organization involved in collecting 11,000 paired samples, sequencing them and analyzing the output. It’s true that sequencing itself is now fairly routine, but that’s still an awful lot of experiments. But think for even longer about what’s gone into making some kind of sense of the monstrous amount of data generated.

And it’s important because?

The findings confirm a trend that has begun to emerge over the last few years, namely that the classification of cancers is being redefined. Traditionally they have been grouped on the basis of the tissue of origin (breast, bowel, etc.) but this will gradually be replaced by genetic grouping, reflecting the fact that seemingly unrelated cancers can be driven by common pathways.

The most encouraging thing to come out of the genetic changes driving these tumours is that for about half of them potential treatments are already available. That’s quite a surprise but it doesn’t mean that hitting those targets will actually work as anti-cancer strategies. Nevertheless, it’s a cheering point that the output of this phenomenal project may, as one of the papers noted, serve as a launching pad for real benefit in the not too distant future.

What should science journalists do to stop upsetting me?

Read the papers they comment on rather than simply relying on press releases, never use the words ‘breakthrough’ or ‘groundbreaking’ and grasp the point that science proceeds in very small steps, not always forward, governed by available methods. This work is quite staggering for it is on a scale that is close to unimaginable and, in the end, it will lead to treatments that will affect the lives of almost everyone — but it is just another example of science doing what science does.

References

Hutter, C. and Zenklusen, J.C. (2018). The Cancer Genome Atlas: Creating Lasting Value beyond Its Data. Cell 173, 283–285.

Hoadley, K.A. et al. (2018). Cell-of-Origin Patterns Dominate the Molecular Classification of 10,000 Tumors from 33 Types of Cancer. Cell 173, 291–304.

Hoadley, K.A. et al. (2014). Multiplatform Analysis of 12 Cancer Types Reveals Molecular Classification within and across Tissues of Origin. Cell 158, 929–944.

Please … Not Another Helping

 

You may have seen the headlines of the: “Processed food, sugary cereals and sliced bread may contribute to cancer risk” ilk, as this recently published study (February 2018) was extensively covered in the media — the Times of London had a front page spread no less.

So I feel obliged to follow suit — albeit with a heavy heart: it’s one of those depressing exercises in which you’re sure you know the answer before you start.

Who dunnit?

It’s a mainly French study (well, it is about food) led by Thibault Fiolet, Mathilde Touvier and colleagues from the Sorbonne in Paris. It’s what’s called a prospective cohort study, meaning that a group of individuals, who in this case differed in what they ate, were followed over time to see if diet affected their risk of getting cancers and in particular whether it had any impact on breast, prostate or colorectal cancer. They started acquiring participants about 20 years ago and their report in the British Medical Journal summarized how nearly 105 thousand French adults got on consuming 3,300 (!) different food items between them, based on each person keeping 24 hour dietary records designed to record their usual consumption.

Foods were grouped according to degree of processing. The stuff under the spotlight is ‘ultra-processed’ — meaning that it has been chemically tinkered with to get rid of bugs, give it a long shelf-life, make it convenient to use, look good and taste palatable.

What makes a food ‘ultra-processed’ is worked out by something called the NOVA classification. I’ve included their categories at the end.

Relative contribution of each food group to ultra-processed food consumption in diet (from Fiolet et al. 2018).

And the result?

The first thing to be said is that this study is a massive labour of love. You need the huge number of over 100,000 cases even to begin to squeeze out statistically significant effects — so the team has put in a terrific amount of work.

After all the squeezing there emerged a marginal increase in risk of getting cancer in the ultra-processed food eaters and a similar slight increase specifically for breast cancer (the hazard ratios were 1.12 and 1.11 respectively). There was no significant link to prostate and colorectal cancers.

Which may mean something. But it’s hard to get excited, not merely because the effects described are small but more so because such studies are desperately fraught and the upshot familiar.

One problem is that they rely on individuals keeping accurate records. Another problem here is that the classification of ‘ultra-processed’ is somewhat arbitrary — and it’s also very broad — leaving one asking what the underlying cause might be: ‘is it sugar, fat or what?’ Furthermore, although the authors tried manfully to allow for factors like smoking and obesity, it’s impossible to do this with complete certainty. The authors themselves noted that, for example, they couldn’t allow for the effects of oral contraception.

The authors are quite right to point out that it is important to disentangle the facets of food processing that bear on our long-term health and that further studies are needed.

I would only add ‘rather you than me.’

Perforce in these pages we have gone on about diets good and bad so there is no need to regurgitate. Suffice to say that my advice on what to eat is the same as that of any other sane person and summarized in Dennis’s Pet Menace — and it’s not been remotely affected by this new research which, in effect, says ‘junk food is probably bad for you in the long run.’ But let’s leave the last word to Tom Sanders of King’s College London: “What people eat is an expression of their life-style in general, and may not be causatively linked to the risk of cancer.” 

Reference

Fiolet, T. et al. (2018). Consumption of ultra-processed foods and cancer risk: results from NutriNet-Santé prospective cohort. BMJ 2018;360:k322 http://dx.doi.org/10.1136/bmj.k322

NOVA classification:

The ultra-processed food group is defined by opposition to the other NOVA groups: “unprocessed or minimally processed foods” (fresh, dried, ground, chilled, frozen, pasteurised, or fermented staple foods such as fruits, vegetables, pulses, rice, pasta, eggs, meat, fish, or milk), “processed culinary ingredients” (salt, vegetable oils, butter, sugar, and other substances extracted from foods and used in kitchens to transform unprocessed or minimally processed foods into culinary preparations), and “processed foods” (canned vegetables with added salt, sugar coated dried fruits, meat products preserved only by salting, cheeses, freshly made unpackaged breads, and other products manufactured with the addition of salt, sugar, or other substances of the “processed culinary ingredients” group).

Desperately SEEKing …

These days few can be unaware that cancers kill one in three of us. That proportion has crept up over time as life expectancy has gone up — cancers are (mainly) diseases of old age. Even so, they plagued the ancients as Egyptian scrolls dating from 1600 BC record and as their mummified bodies bear witness. Understandably, progress in getting to grips with the problem was slow. It took until the nineteenth century before two great French physicians, Laënnec and Récamier, first noted that tumours could spread from their initial site to other locations where they could grow as ‘secondary tumours’. Munich-born Karl Thiersch showed that ‘metastasis’ occurs when cells leave the primary site and spread through the body. That was in 1865 and it gradually led to the realisation that metastasis was a key problem: many tumours could be dealt with by surgery, if carried out before secondary tumours had formed, but once metastasis had taken hold … With this in mind the gifted American surgeon William Halsted applied ever more radical surgery to breast cancers, removing tissues to which these tumors often spread, with the aim of preventing secondary tumour formation.

Early warning systems

Photos of Halsted’s handiwork are too grim to show here but his logic could not be faulted for metastasis remains the cause of over 90% of cancer deaths. Mercifully, rather than removing more and more tissue targets, the emphasis today has shifted to tumour detection. How can they be picked up before they have spread?

To this end several methods have become familiar — X-rays, PET (positron emission tomography, etc) — but, useful though these are in clinical practice, they suffer from being unable to ‘see’ small tumours (less that 1 cm diameter). For early detection something completely different was needed.

The New World

The first full sequence of human DNA (the genome), completed in 2003, opened a new era and, arguably, the burgeoning science of genomics has already made a greater impact on biology than any previous advance.

Tumour detection is a brilliant example for it is now possible to pull tumour cell DNA out of the gemisch that is circulating blood. All you need is a teaspoonful (of blood) and the right bit of kit (silicon chip technology and short bits of artificial DNA as bait) to get your hands on the DNA which can then be sequenced. We described how this ‘liquid biopsy’ can be used to track responses to cancer treatment in a quick and non–invasive way in Seeing the Invisible: A Cancer Early Warning System?

If it’s brilliant why the question mark?

Two problems really: (1) Some cancers have proved difficult to pick up in liquid biopsies and (2) the method didn’t tell you where the tumour was (i.e. in which tissue).

The next step, in 2017, added epigenetics to DNA sequencing. That is, a programme called CancerLocator profiled the chemical tags (methyl groups) attached to DNA in a set of lung, liver and breast tumours. In Cancer GPS? we described this as a big step forward, not least because it detected 80% of early stage cancers.

There’s still a pesky question mark?

Rather than shrugging their shoulders and saying “that’s science for you” Joshua Cohen and colleagues at Johns Hopkins University School of Medicine in Baltimore and a host of others rolled their sleeves up and made another step forward in the shape of CancerSEEK, described in the January 18 (2018) issue of Science.

This added two new tweaks: (1) for DNA sequencing they selected a panel of 16 known ‘cancer genes’ and screened just those for specific mutations and (2) they included proteins in their analysis by measuring the circulating levels of 10 established biomarkers. Of these perhaps the most familiar is cancer antigen 125 (CA-125) which has been used as an indicator of ovarian cancer.

Sensitivity of CancerSEEK by tumour type. Error bars represent 95% confidence intervals (from Cohen et al., 2018).

The figure shows a detection rate of about 70% for eight cancer types in 1005 patients whose tumours had not spread. CancerSEEK performed best for five types (ovary, liver, stomach, pancreas and esophagus) that are difficult to detect early.

Is there still a question mark?

Of course there is! It’s biology — and cancer biology at that. The sensitivity is quite low for some of the cancers and it remains to be seen how high the false positive rate goes in larger populations than 1005 of this preliminary study.

So let’s leave the last cautious word to my colleague Paul Pharoah: “I do not think that this new test has really moved the field of early detection very far forward … It remains a promising, but yet to be proven technology.”

Reference

D. Cohen et al. (2018). Detection and localization of surgically resectable cancers with a multi-analyte blood test. Science 10.1126/science.aar3247.

Boldly Going

When you come across a very successful, middle-aged scientist jumping up and down shouting “This is going to be just amazing” you can only conclude that either the pressures of the life scientific have finally got to him and he’s flipped or there is something really remarkable going on. Thus my feeling this week when I noted the behaviour of Greg Hannon who now works at the Cancer Research Institute in Cambridge.

Probing further, it emerged that Hannon, who is collaborating with Xiaowei Zhuang at Harvard University in the ‘other’ Cambridge, has just been awarded a five-year grant of £20 million by the London-based charity Cancer Research UK as part of its Grand Challenge initiative – more than enough to get your jumping genes going.

But it’s the aim of the project rather than its monetary size that is truly astonishing and has almost a feel of science fiction about it. The plan is nothing less than to come up with an interactive virtual-reality map of breast cancers. That is, to reconstruct every cell that makes up a tumour, showing the different types of cell and what they are up to at any given time – meaning that the model will display the expression level of thousands of genes in each cell and the different proteins being made. Staggering.

What’s the point?

The project is driven by the fact that we have gradually come to realize that tumours are a complex mixture of cells (what’s been called the tumour microenvironment) and the signals that these cells send out and receive determine the extent of tumour growth and whether it can spread to other sites in the body (i.e. metastasize).

Where have we got to?

One approach to mapping what’s going on was laid out a couple of years ago by the converging studies of Rahul Satija and colleagues of the Broad Institute of MIT and Harvard and Kaia Achim et al. from labs in Heidelberg, Cambridge and Oxford using zebrafish embryos and worm brains, respectively.

The method has two parts:

  1. The tissue is dissociated into single cells and the power of sequencing is applied to obtain RNA sequences from each cell (revealing which genes are ‘switched on’ in that cell).
  2. The second step visualizes specific RNAs using tagged probes (fluorescently labeled RNAs that enter cells and bind to target RNAs molecules).

In essence a reference map is made by overlaying the intact tissue with a grid and matching a cell to a grid area by comparing expression of a number of ‘landmark’ genes with the fluorescence marker signal.

To do all this they devised a computational package that, using fewer than 100 landmark genes, maps hundreds of sequenced cells to their location in the tissue. In that arty way that scientists have, they named their package after Georges-Pierre Seurat, the French chappie who came up with the idea of painting in small dots of colour (though his weren’t fluorescent).

Cellular pointillism has just taken another step forward with Keren Bahar Halpern, Ido Amit and Shalev Itzkovitz at the Weizmann Institute of Science, Rehovot, Israel producing a cell-by-cell map of mouse liver, complete with RNA sequences from each cell. To be precise they mapped the hexagon-shaped units called lobules that are repeated to make up mammalian liver.

The shapes of things to come

So the next step for Hannon and his colleagues is an interactive map of a human tumour and, if you can’t wait, CLICK HERE to see their mock-up to give you some idea of what’s in store. In this synthetic video tumour cells are green, macrophages are blue and blood vessels are red.

Overwhelming?

So it’s warp factor 9 for Captain Hannon and his crew. It may be that the 3D images of tumours will look a bit the virtual graphics that the astrophysicists fob off on us whilst pretending they have some idea what a star’s doing umpti-zillion light years away. But in fact, rather than boldly going where no man has gone before“, this cellular journey is better summed up by Marcel Proust The real voyage of discovery consists not in seeking new landscapes, but in having new eyes” – the new eyes being the stunning combination of methods that permits 3D interrogation of individual cells.

Will this phase of the Grand Challenge produce overwhelming amounts of data? Undoubtedly. But, if we are to understand how living things work we have to front up to the complexity of nature. We then have to hope we are smart enough to resolve the crucial from the detail.

References

Satija, R. et al. (2015). Spatial reconstruction of single-cell gene expression data. Nature Biotechnology 33, 495–502.

Achim, K. et al. (2015). High-throughput spatial mapping of single-cell RNA-seq data to tissue of origin. Nature Biotechnology 33, 503–509.

Halpern, K. B. et al. (2017). Nature 542, 352–356.

In the beginning … 

You may have noticed that the American actress Angelina Jolie, who is now employed as a  Special Envoy  for the  United Nations High Commissioner for Refugees, has re-surfaced in the pages of the science media. She first hit the nerdy headlines by announcing in The New York Times that she had had a preventive double mastectomy (in 2013) and a preventive oophorectomy (in 2015).

We described the molecular biology that prompted her actions in A Taxing Inheritance. The essential facts were that she had a family history of breast and ovarian cancer: genetic testing revealed that she carried a mutation in the BRCA1 gene giving her a 87% risk of breast cancer and a 50% chance of getting ovarian cancer.

A star returns

BRCA1 and breast cancer are back in the news as a result of a paper by Jane Visvader, Geoffrey Lindeman and colleagues in Melbourne that asked a very simple question: which type of cell is driven to proliferate abnormally and give rise to a tumour by mutant BRCA1 protein? That is, pre-cancerous breast tissue contains a mixture of cell types: does cancer develop from one in particular –  and, if you blocked proliferation of that type of cell, could you prevent tumours forming?

Simple question but their paper summarises about 10 years of work to come up with a clear answer.

And the villain is …

The mature mammary gland is made up of lots of small sacs (alveoli) lined with cells that produce milk – called luminal cells. Groups of alveoli are known as lobules, linked by ducts that carry milk to the nipple. Most breast cancers start in the lobular or duct cells.

Breast fig copy

Left: Normal breast lobule showing alveoli lined with milk-producing luminal cells connected to duct leading to the nipple. Right: Normal milk sac, non-invasive cancer, invasive cancer.

Things are complicated by there being more than one type of progenitor cell but the Melbourne group were able to show that, in mice carrying mutated BRCA1, one subtype stood out in terms of its cancerous potential. These cells carried a protein on their surface called RANK (which is member of the tumour necrosis factor family). They had gross defects in their DNA repair systems (so they can’t fix genetic damage) and they’re highly proliferative. Luminal progenitors that don’t express RANK behave normally.

Slide1 copy

Scheme representing normal and abnormal cell development. The basic idea is that different types of cells evolve from a common ancestor. The Australian work identified one type of luminal progenitor cell that carries a protein called RANK on its surface (pink cell) as being a prime source of tumours. RANK+ cells have defective DNA repair systems so they accumulate mutations (red cells) more rapidly than normal cells, a feature of tumour cells.

In mice with mutant BRCA1 a monoclonal antibody (denosumab) that blocks RANK signalling markedly slowed tumour development. In a small pilot study blockade of RANK inhibited cell proliferation in breast tissue from human BRCA1-mutation carriers.

Next?

How effective blocking the activation of RANK signalling will be in preventing breast cancer is anyone’s guess but the idea behind the work of the Australian group cannot be faulted. Being able to prevent the ‘starter’ cells from launching themselves on the pathway to cancer driven by mutation in BRCA1 would mean that women in Angelina Jolie’s position would not have to contemplate the drastic course of surgery. The question is: will the preliminary mouse results lead to something that works in humans and, moreover, does so with high efficiency. As ever in cancer, watch this space – but don’t hold your breath!

References

Nolan, E. et al. (2016). RANK ligand as a potential target for breast cancer prevention in BRCA1-mutation carriers.