A breakthrough cure for acute leukemia?


This week a group of scientists from Memorial Sloan-Kettering Cancer Center published what may be a genuine breakthrough in the search for a cure for cancer. The simple word "cancer" disguises what is really hundreds of diseases, most caused by genetic mutations that make a cell start replicating out of control.  The new study looks at a deadly form of leukemia, called B cell acute lymphoblastic leukemia, or B-ALL, and describes a radically new type of cancer therapy that uses genetic modification of a patient's own cells.

First a bit of background. Leukemias are cancers of the blood, which means they don't form solid tumors at all. Instead, the cancer cells circulate in the blood, going virtually everywhere in the body.  The blood has many cell types within it, including white blood cells or leukocytes, which is the origin of the word leukemia.  Lymphocytes are a type of white blood cell, and the two most common subtypes are called B cells and T cells.  B-ALL, then, is a cancer where a B lymphocyte has suddenly turned cancerous.

Scientists have developed chemotherapies to treat B-ALL, but if the first line of treatments fail and the cancer returns, the patient faces a very grim prognosis.  We really don't have any effective treatments at this point, and most patients die.

The new leukemia therapy is a technological tour de force.  The scientific team at Memorial Sloan-Kettering, led by Renier Brentjens and Michel Sadelain, genetically modified T cells from each patient so that these T-cells would target cancerous B cells.  The modified T cells use something called a "chimeric antigen receptor," or CAR, which they designed so that it would attach itself to a specific protein, called CD19, that sticks out of the surface of most cancerous B cells.  (This technique was invented by another group at Memorial Sloan-Kettering, led by Isabelle Rivière.)

What was amazing about this study is that in all five patients, their cancers virtually disappeared in just a few weeks.  As described in a The New York Times report, one of the sickest patients, 58-year-old David Aponte, saw his leukemia disappear in just eight days.  But we cannot know for certain yet if the therapy alone is a cure.  This is because the patients also had stem cell transplants within a few months after therapy, which may have helped eliminate any remaining cancer cells.  The therapy itself was temporarily very toxic, requiring close monitoring and steroid treatments to lessen the toxic side effects.  This may have been the reason that one patient died: the steroids may have hampered the therapy by destroying the genetically modified cells.

If this therapy were used as a first-line treatment of B-ALL, rather than in patients who have tried conventional therapy and relapsed, it might be even more effective (as the authors themselves suggest).  In particular, it is likely to be less toxic in patients who are at earlier stages of the disease and have fewer tumor cells in their blood.  Thus despite its promise, CAR T-cell therapy needs to be refined and tested further before we can declare it a success.

I'd like to make a final observation: this work, like almost all biomedical research in the U.S., was supported by the NIH, whose budget was just cut severely by Congress with its ill-conceived "sequester."  If we keep cutting biomedical research, we won't see many more breakthroughs like this one.  Even this dramatic result, promising though it is, needs more research to improve it and to test it on other leukemias, including chronic lymphycytic leukemia and non-Hodkins lymphoma, where it has already shown very promising results.

This leukemia treatment didn't just appear out of thin air.  It uses a technology that was invented four years ago, and that technology in turn is based on other discoveries that now make it possible to re-engineer a patient's own cells and turn them into a treatment.  Who knows where the next ten years will take us?   Now is the time to be increasing our investment in biomedical research.

Maryland legislature scores one for science

Just one month ago, I wrote a post entitled "Naturopathic shenanigans in the Maryland State legislature," about the efforts of naturopaths to get a new law passed in the state of Maryland that would license them to practice medicine.  This effort to get a set of pseudoscientific practices certified as "medicine" has been pursued by naturopaths across the U.S., and Maryland is just one of their latest targets.  I was dismayed when I learned that laws had been introduced in both houses of the Maryland legislature to give naturopaths what they wanted.

Well, the votes just came in, and the law was unanimously defeated in committee in both the House and Senate.  Without knowing precisely what happened (or if any of them read my article, either here or on the Forbes magazine site) I cannot explain why the legislators voted as they did, but after my previous article I want to give them a rousing cheer of approval.  Hurrah!  Sometimes lawmakers get it right, and when they do, they deserve our applause.

Everything in this journal is wrong

Moxibustion at work.

In real science, we agonize over every detail.  When we publish a paper, we strive to get everything just right.  We qualify our findings, always allowing for the possibility that we might have missed something.

Oh, to be freed of the constraints of reality.  But fiction, alas, doesn't work in the real world.  Fantasy medicine is, well, a fantasy.  Or is it?  Let's enter the world of Complementary and Alternative Medicine.

Imagine a scientific journal in which every single article was wrong.  Not just an occasional mistake, as happens with many journals.  No, I'm talking about a publication where every single article has at least one fundamental flaw.

No, it's a real journal, published by BMJ, the publishers of the well-known journal by the same name (the British Medical Journal).  "Helping doctors make better decisions since 1840", they proclaim.  Then why, among their 40+ journals, do they publish Acupuncture in Medicine?

It appears - I can't prove this, but it seems to be so - that every single article in this journal is wrong.  Not just a little bit wrong, either.  Let's look at the "Editor's choice" article from the latest issue, which must be one of their best.  It's called "Using moxibustion in primary healthcare to correct non-vertex presentation: a multicentre randomised controlled trial".

In this study, a group of Spanish doctors looked at the use of moxibustion to correct a common problem in pregnancy known as a breech position - that's a baby that is turned the wrong way round in the mother, head up instead of down.  What, you may ask, is moxibustion?  Well, let me use the authors' explanation:
"The application of heat from the combustion of Artemisia vulgaris (moxibustion) for therapeutic purposes has long been used in China. Among other effects, it is believed to contribute to correcting the non-vertex presentation of the fetus when applied at a specific acupuncture point (BL67 Zhiyin) located at the outer corner of the little toenail."
That's right: you burn an herb (Artemisia) next to the little toenail of the pregnant mother, right next to an acupuncture point, and that's supposed to make baby flip right around so its head comes out first.

You can't make this stuff up.

These guys are serious.  They ran a study where they divided pregnant women into 3 groups, with about 135 women in each.  One group got moxibustion, another got "fake" moxibustion, where they burned the herb next to a different toe (really!), and the last group got standard care.

Amazingly, the doctors found that it works!  More of the women with moxibustion had babies born the right way round, head down, than in either of the other two groups.  To be precise, 79 babies in the moxibustion group came out the right way, versus 60 and 59 in the other two.  By the authors' own analysis, this difference was statistically significant.

Wow.  Maybe this stuff really does work.  And moxibustion is cheap and easy to administer.  Maybe this is the solution to our rising health care costs.

How to explain this result?  I see several possibilities:

  1. The small effect was just a random variation, not due to the treatment.  Previous studies of exactly the same treatment (such as this one) showed that moxibustion did not work.  So either those studies are wrong, or this one is.
  2. There was bias in how the women were assigned to treatment, and the effect can be explained by that.
  3. The authors manipulated the data to make the numbers come out better.  (Quelle surprise! Not in a BMJ journal!)  Of course we can't prove this without much more investigation.
  4. Burning an herb next to the little toe at just the right place - and nowhere else - stimulates a mystical "qi" pathway, and zaps that little baby back around the way he's supposed to be.
Hmmm, which seems the most likely?

OK, I admit it: I haven't read every single article in BMJ's acupuncture journal.  So I don't really know that they are all wrong.  But who has time to read all this bad science?  The latest issue has a disturbing number of articles about acupuncture for cancer, which I find particularly upsetting because that practice takes advantage of highly vulnerable patients.  It also includes an article on ear acupuncture, a practice invented out of whole cloth by an Army doctor in the U.S., who has been inflicting it on injured service members, as I've written about before.

Actually there is one article that seems plausible.  It reports a case of arterial hemorrhage caused by an acupuncture needle.  But we don't need to dwell on that.

Acupuncture in Medicine has all the trappings of a real journal, including an editorial board whose members work at respectable medical schools.  I know that BMJ wants us to believe it; after all, they make money on this stuff.  As for the journal itself: I can't bear to read any more.  It is just too painful.  Perhaps acupuncture would help.

Whatever happened to swine flu?

What happened to the flu pandemic?  In 2009, a new flu strain swept across the world. The new strain, called H1N1, emerged from pigs and jumped over to humans sometime in late 2008, and then swept through the human population starting in the spring of 2009.  Panic ensued.  Egypt responded by slaughtering all of its pigs, about 300,000.

Was the panic justified?  If so, where are all the victims?

I first wrote about this soon after the outbreak began, and we now know that hundreds of millions of people were infected, somewhere in the range 11% to 21% of the population.  That's an awful lot of sick people.  However, H1N1 turned out to be a very mild flu: many people experienced little more than a few days of sniffles, much like a common cold.  This surprising mildness of swine flu led to great confusion.  Conspiracy theorists claimed that the threat had been overblown, hyped by vaccine manufacturers and their government co-conspirators.  A wacky German lawmaker, Wolfgang Wodarg, even claimed that the swine flu vaccine caused cancer, a claim that was picked up and amplified by famed internet snake oil salesman, Joseph Mercola.

The swine flu now seems routine, just another human flu circulating among the population. As I wrote back in 2010, the seasonal flu vaccine now includes the H1N1 pandemic strain, so if you get your flu shot, you're protected.  But as this figure from the CDC shows, the current season has been dominated by H3N2. 
See the little tiny brown bits at the top of each bar?  Those are swine flu cases.  The swine flu has nearly vanished.

This is a big surprise, because in all three of the previous pandemics: 1918 ("Spanish" flu), 1957, and 1968, the new pandemic strain completely replaced the older strain.  That hasn't happened this time, and it looks like the old strain, H3N2, is winning.  That's rather unfortunate, because H3N2 is a much nastier flu than the swine flu.  And this year we had a big spike in deaths due to flu, all because of H3N2.

So no, the panic back in 2009 wasn't justified, but the warnings beforehand, about the possibilities of a pandemic, were legitimate.  All we knew in early 2009 was that a new pandemic strain had jumped from pigs to humans, and we didn't know for several months how bad (or mild) it would be.  The human species got lucky this time.

Can anyone say when the next pandemic will arrive?  Well, no.  Look at the past century: 4 pandemics, separated by 39 years, 11 years, and 41 years.  From that record it seems we should be safe for a while.  But until 2009, the pandemics had always pushed out the previous flu.  We're still living with the 1968 flu strain, and no one knows when a new flu will truly replace it.

Meanwhile, get your flu shot, because the flu mutates so fast that we need a new vaccine every year to keep ahead of it.  Work continues to try to develop a permanent flu vaccine - one that we will only have to take once in a lifetime.  If you like that idea, then keep supporting NIH, which is the biggest source of funding for flu research.

Oh right: we just cut NIH across the board because Congress couldn't get its act together.  I guess we may have to wait a bit longer for a better flu vaccine.