Seeing a New World

May I wish readers a Happy New Year – and indeed extend my felicitations to non-readers with the hope that they too will become followers! What a good idea! Not least because I suspect many are viewing the new year with a mixture of anxiety and despair. But I can promise there’s nothing like the sanity of science to restore you after a few minutes contemplating how we’re doing on the economic and political fronts.

Your starter for 2017

By happy chance a few weeks ago I tried to explain how it’s now possible to ‘re-write the manual of life’ – that is, to engineer our DNA, to fix broken genes if you like. This means that, in theory, it’s possible to correct errors in our genetic code that cause genetic diseases. As there are over 6,000 of these and they include Down syndrome, cystic fibrosis and Alzheimer’s disease, there’s no need to say it’s important. There are several ways of going about this but the one I described is called CRISPR and it’s had a lot of media coverage.

Right on cue

Well done then Keiichiro Suzuki, Juan Carlos Belmonte and friends from the Salk Institute in California together with colleagues from other centres in Spain, Saudi Arabia and China for their December paper describing a new CRISPR twist. They used a rat model of retinitis pigmentosa, a genetic disease that is a major cause of inherited blindness, afflicting about one and a half million people worldwide (one in 4,000 in the UK).

The CRISPR-Cas9 system is great but it works best in dividing cells (e.g., in skin and gut that are renewing all the time) and it’s particularly useful for knocking out genes rather than inserting new DNA. The latest modification allows a new gene to be inserted into a specific site in the DNA of cells that are not dividing (e.g., those of the eye or brain).

The bits of CRISPR-Cas9, which insert DNA at very precise locations within the genome, are delivered to target cells as part of an inert virus. However, the package also includes DNA that encourages the cells to use a repair process that can be turned on even in non-dividing cells. So CRISPR-Cas9 cuts the cell’s DNA at an exact sequence and the cell then repairs the double-strand breaks (by a process called non-homologous end joining (NHEJ) that glues the broken ends directly together). Give the cell a new bit of DNA (e.g., your favorite gene) and that will get patched in – bear in mind that the cell doesn’t ‘know’ what it’s doing: it just tries to fix damaged DNA with whatever’s at hand.

And the target?

Retinitis pigmentosa occurs when a chunk of a gene called Mertk is lost. After quite a lot of experiments to show that their method worked, Suzuki, Belmonte & Co made a viral carrier that included a normal Mertk gene and injected it under the retina of rats with the disease. After about 5 weeks the rats were making Mertk RNA as a result of the gene being correctly ‘knocked-in’ to eye cells. The light-detecting region of the eye, greatly reduced by the disease, was significantly restored, with associated appearance of MERTK protein.

      Diseased    Normal     Treated                         Diseased         Normal         Treated

pic

Left trio: Sections of the light-detecting layers of the eye in diseased (left), normal (centre) and diseased post-treatment rats (right). Right trio: corresponding fluorescence images showing MERTK expression (red: highlighted by white arrows); Cells labeled blue. (Suzuki et al. Nature 1–6 (2016) doi:10.1038/nature20565)

How did the rats see it?

Well, after treatment they were able to detect light and had significantly recovered their visual functions, albeit not to completely normal levels.

The usual caveats apply: the method isn’t hyper-efficient and a human treatment is still a long way off. Nevertheless, it’s a significant step.

The same group has also shown, using a way of re-programming the expression of just four genes, that it’s possible to arrest the signs of ageing. In other words, in mice this time, tinkering with these genes can increase lifespan – and yes, we have versions of these genes and in us they also control cell renewal.

So the New Year message is clear to see. If we can avoid turning the planet into a desert or blowing ourselves to smithereens the future is really rosy – and maybe even infinite!

References

Suzuki, K. et al. (2016). In vivo genome editing via CRISPR/Cas9 mediated homology-independent targeted integration. Nature 540, 144-149.

Ocampo, A. et al. (2016). In Vivo Amelioration of Age-Associated Hallmarks by Partial Reprogramming. Cell 167, 1719–1733.

Advertisements

Bigger is Better

“Nonsense!” most males would cry, quite logically, given that we spend much of our time trying to persuade the opposite sex that size doesn’t matter. But we want to have it both ways: in the macho world of rugby one of the oldest adages is that ‘a good big ’un will always beat a good little ’un’.  Beethoven doubtless had a view about size – albeit unrecorded by history – but after he’d written his Eroica symphony, perhaps the greatest revolutionary musical composition of all, his next offering in the genre was the magical Fourth – scored for the smallest orchestra used in any of his symphonies. And on the theme of small can be good, the British Medical Journal, no less, has just told us that if we cut the size of food portions and put ’em on smaller plates we’ll eat less and not get fat!

Is bigger better?

Is bigger better?

All of which suggests that whether bigger is better depends on what you have in mind. Needless to say, in these pages what we have in mind is ‘Does it apply to cancer?’ – that is, because cancers arise from the accumulation in cells of DNA damage (mutations), it would seem obvious that the bigger an animal (i.e. the more cells it has) and the longer it lives the more likely it will be to get cancer.

Obvious but, this being cancer, also wrong.

Peto’s Paradox

The first person to put his finger on this point was Sir Richard Peto, most famous for his work with Sir Richard Doll on cancer epidemiology. It was Doll, together with Austin Bradford Hill, who produced statistical proof (in the British Doctors’ Study published in 1956) that tobacco smoking increased the risk of lung cancer. Peto joined forces with Doll in 1971 and they went on to show that tobacco, infections and diet between them cause three quarters of all cancers.

Whenever this topic comes up I’m tempted to give a plug to the unfortunate Fritz Lickint – long forgotten German physician – who was actually the first to publish evidence that linked smoking and lung cancer and who coined the term ‘passive smoking’ – all some 30 years before the Doll study. Lickint’s findings were avidly taken up by the Nazi party as they promoted Draconian anti-smoking measures – presumably driven by the fact that their leader, Gröfaz (to use the derogatory acronym by which he became known in Germany as the war progressed – from Größter Feldherr aller ZeitenGreatest Field Commander of all Time) was a confirmed non-smoker. Despite his usefulness, Lickint’s political views didn’t fit the ideology of the times. He lost his job, was conscripted, survived the war as a medical orderly and only then was able to resume his life as a doctor – albeit never receiving the credit he deserved.

Returning to Richard Peto, it was he who in 1975 pointed out that across different species the incidence of cancer doesn’t appear to be linked to the number of cells in animal – i.e. its size.   He based his notion on the comparison of mice with men – we have about 1000 times the number of cells in a mouse and typically live 30 times as long. So we should be about a million times more likely to get cancer – but in fact cancer incidence is another of those things where we’re pretty similar to our little furry friends. That’s Peto’s Paradox.

It doesn’t seem to apply within members of the same species, a number of surveys having shown that cancer incidence increases with height both for men and women. The Women’s Health Initiative found that a four inch increase in height raised overall cancer risk by 13% although for some forms (kidney, rectum, thyroid and blood) the risk went up by about 25%. A later study found a similar association for ovarian cancer: women who are 5ft 6in tall have a 23% greater risk than those who only make it to 5 feet. A similar risk links ovarian cancer to obesity (i.e. a rise in body mass index from 20 (slim) to 30 (slightly overweight) puts the risk up by 23%). Statistically sound though these results appear to be, it’s worth nothing that, as my colleague Paul Pharoah has pointed out, these risk changes are small. For example, the ovarian cancer finding translates to a lifetime risk of about 16-in-a-1000 for shorter women going up to 20-in-a-1000 as they rise by 6 inches.

It’s true that there may be a contribution from larger animals having bigger cells (whale red blood cells are about twice as big as those of the mouse) that divide more slowly but at most that effect seems small and doesn’t fully account for the fact that across species the association of size and age with cancer breaks down: Peto’s Paradox rules – humans are much more likely to get cancer than whales.

What did we know?

Well, since Peto picked up the problem, almost nothing about underlying causes. The ‘almost’ has been confined to the very small end of the scale and we’ve already met the star of the show – the naked mole rat – a rather shy chap with a very long lifespan (up to 30 years) but who never seems to get cancer. In that piece we described the glimmerings of an explanation but, thanks to Xiao Tian and colleagues of the University of Rochester, New York we now know that these bald burrowers make an extraordinarily large version of a polysaccharide (a polymer of sugars). These long strings of glucose-like molecules (called hyaluronan) form part of the extracellular matrix and regulate cell proliferation and migration. They’re enormous molecules with tens of thousands of sugars linked together but the naked mole rat makes versions about four times larger than those of mice or humans – and it seems that these extra-large sugar strings restrict cell behaviour and block the development of tumours.

Going up!

Our ignorance has just been further lifted with two heavyweight studies, one from Lisa Abegglen, Joshua Schiffman and chums from the University of Utah School of Medicine who went to the zoo (San Diego Zoo, in fact) and looked at 36 different mammalian species, ranging in size from the striped grass mouse (weighing in at 50 grams) to the elephant – at 4,800 kilogram nearly 100,000 times larger. They found no relationship between body size and cancer incidence, a result that conforms to Peto’s paradox. Comparing cancer mortality rates it transpires that the figure for elephants is less than 5% compared with the human range of 11% to 25%.

107 final pic

Cancer incidence across species by body size and lifespan. A selection of 20 of the 36 species studied is shown. Sizes range from the striped grass mouse to the elephant. As the risk of cancer depends on both the number of cells in the body and the number of years over which those cells can accumulate mutations, cancer incidence is plotted as a function of size (i.e. mass in grams × life span, years: y axis: log scale). Each species is represented by at least 10 animals (from Abegglen et al., 2015).

It can be seen at a glance that cancer incidence is not associated with mass and life span.

The Tasmanian devil stands out as a remarkable example of susceptibility to cancer through its transmission by biting and licking.

How does Jumbo do it?

In a different approach to Peto’s Paradox, Michael Sulak, Vincent Lynch and colleagues at the University of Chicago looked mainly at elephants – more specifically they used DNA sequencing to get at how the largest extant land mammal manages to be super-resistant to cancer. In particular they focused on the tumor suppressor gene P53 (aka TP53) because its expression is exquisitely sensitive to DNA damage and when it’s switched on the actions of the P53 protein buy time for the cell to repair the damage or, failing that, bring about the death of the cell. That’s as good an anti-cancer defence as you can imagine – hence P53’s appellation as the ‘guardian of the genome’. It turned out that elephants have no fewer than 20 copies of P53 in their genome, whereas humans and other mammals have only one (i.e. one copy per set of (23) chromosomes). DNA from frozen mammoths had 14 copies of P53 but manatees and the small furry hyraxes, the elephant’s closest living relatives, like humans have only one.

The Utah group confirmed that elephants have, in addition to one normal P53 gene, 19 extra P53 genes (they’re actually retrogenes – one type of the pseudogenes that we met in the preceding post) that have been acquired as the animals have expanded in size during evolution. Several of these extra versions of P53 were shown to be switched on (transcribed) and translated into proteins.

Consistent with their extra P53 fire-power, elephant cells committed P53-dependent suicide (programmed cell death, aka apoptosis) more frequently than human cells when exposed to DNA-damaging radiation. This suggests that elephant cells are rather better than human cells when it comes to killing themselves to avoid the risk of uncontrolled growth arising from defective DNA.

More genes anyone?

Those keen on jumping on technological bandwagons may wish to sign up for an extra P53 gene or two, courtesy of genetic engineering, so that bingo! – they’ll be free of cancers. Aside from the elephant, they may be encouraged by ‘super P53’ mice that were genetically altered to express one extra version of P53 that indeed significantly protected from cancer when compared with normal mice – and did so without any evident ill-effects.

We do not wish to dampen your enthusiasm but would be in dereliction of our duty is we did not add a serious health warning. We now know a lot about P53 – for example, that the P53 gene encodes at least 15 different proteins (isoforms), some of which do indeed protect against cancer – but there are some that appear to act as tumour promoters. In other words we know enough about P53 to realize that we simply haven’t a clue. So we really would be playing with fire if we started tinkering with our P53 gene complement – and to emphasise practicalities, as Mel Greaves has put it, we just don’t know how well the elephants’ defences would stack up if they smoked.

Nevertheless, on the bright side, light is at long last beginning to be shed on Peto’s Paradox and who knows where that will eventually lead us. Meanwhile Richard Peto’s activities have evolved in a different direction and he now helps to run a Thai restaurant in Oxford, a cuisine known for small things that pack a prodigious punch. Bit like Beethoven’s Fourth you could say.

a-gem-of-a-find-in-oxford

References

Peto, R. et al. (1975). Cancer and ageing in mice and men. British Journal of Cancer 32, 411-426.

Doll, R. and Peto, R. (1976). Mortality in relation to smoking: 20 years’ observations on male British doctors. Br Med J. 2(6051):1525–36.

Maciak, S. and Michalak, P. (2015). “Cell size and cancer: A new solution to Peto’s paradox?”. Evolutionary Applications 8: 2.

Doll, R. and Hill, A.B. (1954). “The mortality of doctors in relation to their smoking habits”. BMJ 328 (7455): 1529.

Doll, R. and Hill, A.B. (November 1956). “Lung cancer and other causes of death in relation to smoking; a second report on the mortality of British doctors”. British Medical Journal 2 (5001): 1071–1081.

Tian, X. et al. (2013). High-molecular-mass hyaluronan mediates the cancer resistance of the naked mole rat. Nature 499, 346-349.

Abegglen, L.M., Schiffman, J.D. et al. (2015). Potential Mechanisms for Cancer Resistance in Elephants and Comparative Cellular Response to DNA Damage in Humans. JAMA. doi:10.1001/jama.2015.13134.

Sulak, M., Lindsey Fong, Katelyn Mika, Sravanthi Chigurupati, Lisa Yon, Nigel P. Mongan, Richard D. Emes, Vincent J. Lynch, V.J. (2015). TP53 copy number expansion correlates with the evolution of increased body size and an enhanced DNA damage response in elephants. doi: http://dx.doi.org/10.1101/028522.

García-Cao, I. et al. (2002). ‘Super p53’ mice exhibit enhanced DNA damage response, are tumor resistant and age normally. EMBO Journal 21, 6225–6235.

Blowing Up Cancer

To adapt the saying of the sometime British Prime Minister Harold Wilson, a month is a long time in cancer research. {I know, you’ve forgotten – as well you might. He was PM from 1964 to 1970 and again from 1974 to 1976. His actual words were “A week is a long time in politics”}. When I started to write the foregoing Self Helps (Parts 1 & 2) I had absolutely no intention of mentioning the subject of today’s sermon – viral immunotherapy. But how times change and a recent report has hit the headlines – so here goes.

The reason for my reticence is that this is not a new field – far from it. Folk have been trying to target tumour cells with active viruses for twenty years but efforts have foundered to the extent that the new report is the first time in the western world that a phase III trial (when a drug or treatment is first tested on large groups of people) of cancer “virotherapy” has definitively shown benefit for patients with cancer, although a virus (H101) made by the Shanghai Sunway Biotech Co. was licensed in China in 2005 for the treatment of a range of cancers.

Hard bit already done

I appreciate that getting the hang of immunotherapy in the two Self Helps wasn’t a total doddle – but it was worth it, wasn’t it, bearing in mind we’re dealing with life and death here. My friend and correspondent Rachel Bown had to resort to her GCSE biology notes (since she met me I think she keeps them on the coffee table) but is now up to speed.

Fortunately this bit is pretty easy to follow – it’s just an extension of the viral jiggery-pokery we met in Self Help Part 2. There we saw that using ‘disabled’ viruses is a neat way of getting new genetic material into cells. The viruses have key bits of their genome (genetic material) knocked out – so they don’t have any nasty effects and don’t replicate (make more of themselves) once inside cells. Inserting new bits of DNA carrying a therapeutic gene turns them into a molecular delivery service.

Going viral

In virotherapy there’s one extra wrinkle: the viruses, though ‘disabled’, still retain the capacity to replicate – and this has two effects. First, more and more virus particles (virions) are made in an infected cell until eventually it can hold no more and it bursts. The cell is done for – but a secondary effect is that the newly-made virions spill out and drift off to infect other cells. This amplifies the effect of the initial injection of virus and, in principle, will continue as long as there are cells to infect.

A new tool

The virus used is herpes simplex (HSV-1) of the relatively harmless type that causes cold sores and, increasingly frequently, genital herpes. The reason for this choice is that sometimes, not very often, science gets lucky and Mother Nature comes up with a helping hand. For HSV-1 it was the completely unexpected discovery that when you knock out one of its genes the virus becomes much more effective at replicating in tumour cells than in normal cells. That’s a megagalactic plus because, in effect, it means the virus targets tumour cells, thereby overcoming one of the great barriers to cancer therapy. In this study another viral gene was also deleted, which increases the immune response against infected tumour cells.

All this cutting and pasting (aka genetic engineering) is explained in entertaining detail in Betrayed by Nature but for now all that matters is that you end up with a virus that:

  1. Gets into tumour cells much more efficiently than into normal cells,
  2. Makes the protein encoded by the therapeutic gene, and
  3. Replicates in the cells that take it up until eventually they are so full of new viruses they go pop.

The finished product, if you can get your tongue round it, goes by the name of talimogene laherparepvec, mercifully shortened by the authors to T-VEC (made by Amgen). So T-VEC mounts a two-pronged attack – what the military would call a pincer movement. Infected tumour cells are killed (they’re ‘lysed’ by viral overload) and the inserted gene makes a protein that soups up the immune response – called GM-CSF (granulocyte macrophage colony-stimulating factor). The name doesn’t matter: what’s important is that it’s a human signaling molecule that stimulates the immune system, the overall result being production of tumour-specific T cells.

Fig. 1 Viral Therapy

Virotherapy. Model of a virus (top). The knobs represent proteins that enable the virus to stick to cells. Below: sequence of injecting viruses that are taken up by tumour cells that eventually burst to release new virions that diffuse to infect other tumour cells.

And the results?

The phase III trial, led by Robert Andtbacka, Howard Kaufman and colleagues from Rutgers Cancer Institute of New Jersey, involved 64 research centres worldwide and 436 patients with aggressive, inoperable malignant melanoma who received either an injection of T-VEC or a control immunotherapy. Just over 16% of the T-VEC group showed a durable response of more than six months, compared with 2% given the control treatment. About 10% of the patients treated had “complete remission”, with no detectable cancer remaining – considered a cure if the patient is still cancer-free five years after diagnosis.

Maybe this time?

We started with Harold Wilson and it was in between his two spells in Number 10 that President Nixon declared his celebrated ‘War on Cancer’, aimed at bringing the major forms of the disease under control within a decade or two. It didn’t happen, as we might have guessed. Back in 1957 in The Black Cloud the astrophysicist Sir Fred Hoyle has the line ‘I cannot understand what makes scientists tick. They are always wrong and they always go on.’ To be fair, it was a science fiction novel and the statement clearly is only partly true. But it’s not far off and in cancer there’s been rather few of the media’s beloved ‘breakthroughs’ and a great deal of random shuffling together with, overall, some progress in specific areas. Along the bumpy highway there have, of course, been moments of high excitement when some development or other has briefly looked like the answer to a maiden’s prayer. But with time all of these have fallen, if not by the wayside, at least into their due place as yet another small step for man. The nearest to a “giant leap for mankind” has probably been coming up with the means to sequence DNA on an industrial scale that is now having a massive impact on the cancer game.

When Liza Minnelli (as Sally Bowles in Cabaret) sings Maybe this time your heart goes out to the poor thing, though your head knows it’ll all end in tears. But this time, maybe, just maybe, the advent of cancer immunotherapy in its various forms will turn out to be a new era. Let us fervently hope so but, even if it does, the results of this Phase III trial show that a long struggle lies ahead before treatments arrive that have most patients responding.

We began Self Help – Part 1 with the wonderful William Coley and there’s no better way to pause in this story than with his words – reminding us of a bygone age when the scientist’s hand could brandish an artistic pen and space-saving editors hadn’t been invented:

“While the results have not been as satisfactory as one who is seeking perfection could wish, … when it comes to the consideration of a new method of treatment for malignant tumours, we must not wonder that a profession with memories overburdened with a thousand and one much-vaunted remedies that have been tried and failed takes little interest in any new method and shows less inclination to examine into its merits. Cold indifference is all it can expect, and rightly too, until it has something beside novelty to offer in its favour.”

References

Mohr, I. and Gluzman, Y. (1996). A herpesvirus genetic element which affects translation in the absence of the viral GADD34 function. The EMBO Journal 15, 4759–66.

Andtbacka, R.H.I. et al. (2015). Talimogene Laherparepvec Improves Durable Response Rate in Patients With Advanced Melanoma. 10.1200/JCO.2014.58.3377

Self Help – Part 2

In the second type of cancer immunotherapy a sample of a patient’s T lymphocytes is grown in the lab. This permits either expansion of the number of cells that recognize the tumour or genetic engineering to modify the cells so they express receptors on their surface that target them to the tumour cell surface. Infusion of these manipulated cells into the patient enhances tumour cell killing. We’re now in the realms of ‘personalized medicine’.

A little more of a good thing

The first of these methods picks up a weakness in the patient’s immune system whereby it makes lymphocytes that kill tumour cells but can’t make enough – their protective effect is overwhelmed by the growing cancer. By taking small pieces of surgically removed tumours and growing them in the lab, it’s possible to select those T cells that have killing capacity. These are expanded over a few weeks to make enough cells to keep on growing when they’re infused back into the patient. The upshot is a hefty boost for the natural anti-tumour defence system. The pioneer of this method, called adoptive cell therapy, is Steven Rosenberg (National Cancer Institute, Bethesda) and it has been particularly effective for melanomas. Responses are substantially improved by treatment with drugs that reduce the white cell count before samples are taken for T cell selection – probably because the system responds by making growth factors to restore the balance and these drive the expansion of the infused cells.

A wonderful benefit of this method is its efficacy against metastases – i.e. tumour growths that have spread from the primary site – perhaps not surprising as it’s what Rosenberg calls a “living” treatment, in other words it just gives a helping hand to what nature is already trying to do.

93. Fig. 1Selecting naturally occurring T cells with anti-tumour activity

Tumour fragments are grown in the laboratory: lymphocytes that kill tumour cells are selected and expanded in culture.  About 6 weeks growth yields enough cells to infuse into the patient.

Gene therapy

A more sophisticated approach to boosting innate immunity is to introduce new genes into the genetic material (the genome) of T cells to target them to tumour cells with greater efficiency. An ordinary blood sample suffices as a starting point from which T cells are isolated. One way of getting them to take up novel genes uses viruses – essentially just genetic material wrapped in an envelope. The virus is ‘disabled’ so that it has none of its original disease-causing capacity but retains infectivity – it sticks to cells. ‘Disabling’ means taking just enough of the original genome to make the virus – a viral skeleton – and then inserting your favourite gene, so the engineered form is just a handy vehicle for carrying genes. No need to panic, therefore, if you see a press headline of the “HIV cures cancer” variety: it just means that the human immunodeficiency virus – well and truly disabled – has been used as the gene carrier.

93. Fig. 2

Genetic modification of blood lymphocytes

T cells are isolated from a blood sample and novel genes inserted into their DNA. The engineered T cells are expanded and then infused into the patient.

 This method of re-directing T cells to a desired target was pioneered by Gideon Gross and colleagues at The Weizmann Institute of Science in Israel in the late 1980s and it has led to sensational recent results in treating chronic lymphocytic leukemia (CLL), albeit in just a few patients so far. To the fore have been Renier Brentjens and his group from the Memorial Sloan-Kettering Cancer Center, New York. The genetic modification they used made the patient’s T cells express an artificial receptor on their surface (called a chimeric antigen receptor). This T cell receptor was designed to stick specifically to a protein known to be displayed on the surface of CLL cells. The result was that the T cells, originally unable to ‘see’ the leukemic cells, now homed in on them with high efficiency. Astonishingly, and wonderfully, the modified cells divide in the patient so that, in effect, their immune system has been permanently super-charged.

A critical part of the strategy is that CLL cells carry a known molecular target but the absence of such defined markers for most cancers is currently a severe limitation. On the bright side, however, this type of gene therapy has now been attempted in at least three different centres and, despite inevitable minor differences in method, it clearly works.

One of the leading figures in gene therapy is Carl June of the University of Pennsylvania. Some of his colleagues have made a brilliant video explaining how it works whilst June himself has described in wonderfully humble fashion what it means to work in this field.

References

Rosenberg, S.A. and Restifo, N.P. (2015). Adoptive cell transfer as personalized immunotherapy for human cancer. Science 348, 62-68.

Gross, G., et al. (1989). Expression of immunoglobulin-T-cell receptor chimeric molecules as functional receptorswith antibody-type specificity. Proc. Natl. Acad. Sci. U.S.A. 86, 10024–10028.

Brentjens, R.J., et al. (2013). CD19-Targeted T Cells Rapidly Induce Molecular Remissions in Adults with Chemotherapy-Refractory Acute Lymphoblastic Leukemia. Sci Transl Med., 5, 177ra38. DOI:10.1126/scitranslmed.3005930.

Kalos, M., et al. (2011). T cells with chimeric antigen receptors have potent antitumor effects and can establish memory in patients with advanced leukemia. Sci. Transl. Med. 3, 95ra73.

Kochenderfer, J.N., et al. (2012). B-cell depletion and remissions of malignancy along with cytokine-associated toxicity in a clinical trial of anti-CD19 chimeric-antigen-receptor–transduced T cells. Blood 119, 2709–2720.