Sunday, May 22, 2016

Sickle Cell Anemia: An underprioritized disease?

The Sunday Boston Globe today had a front page piece by STAT's Sharon Begley that asks some challenging questions about prioritization of disease research.  Poking around the STAT site, I found that the original article was even longer and better, but between the important issues it raises, some interesting peripheral stuff and at least one gaping hole, there's plenty to discuss.

As the article points out, we've known the biochemical basis for sickle cell anemia for nearly 70 years.  Linus Pauling and colleagues linked altered hemoglobin to the sickling phenonomenon in 1949. with Vernon Ingram (working with Francis Crick) later showing that a single amino acid change was responsible.  In that time, only a single treatment for sickle cell, which can cause debilitating pain and ultimately shortens lives, has been approved, hydroxyurea, which was apparently stumbled upon.  The article asks why?

Sickle cell and the related beta thalassemias are most prevalent in persons of African or Mediterranean descent.  According to Begley's piece, only about 10 labs in the U.S. have focused efforts on sickle cell anemia.  Several figures quoted in the article say that donors are reluctant to fund sickle cell programs, preferring to fund instead pediatric cancer based on photogenic patients.  The ugly spectre of racism is certainly suggested by this pattern.

To me, this is an illustration of one key reason why demographic diversity is an important goal in bioscience.  Many researchers are passionate about their work because it has deep personal meaning for them, as a relative, friend or even the researcher themselves is a patient of a particular disease.  I know that I personally keep a mental list of those I've known (or should have known) who have been devastated by what should be treatable diseases, starting with my paternal grandmother, who died of leukemia far before my parents met.  Ensuring that scientific personnel are sampled from a wide variety of cultures, backgrounds, geographies and ethnic groups is one way to minimize the awful possibility that important diseases are "out of sight, out of mind".

The article spends a bit of time discussing work by Stuart Orkin.  It has long been known that humans generate a different beta globin, fetal hemoglobin, during development in utero.  Hydroyurea works, at least in part, by activating fetal hemoglobin production, and apparently just 15 percent fetal hemoglobin can stop sickling.  Orkin had identified Saudi Arabian individuals who retained fetal hemoglobin production into adulthood. Orkin had pursued finding transcription factors key to the production of the different globins, with much frustration.  As related in the article, a student named Vijay Sankaran wanted to take on the challenge -- and Orkin made sure Sankaran was willing to risk failure.  Indeed, it took more than three years, but in 2009 Sankaran and colleagues as well as a second group showed that polymorphisms in the hematopoetic transcription factor BCL11A were linked to adult expression of fetal hemoglobin (Begley omitted mention of the second group).  Further work culminated in Orkin, Sankaran and others showing that inactivating BCL11A in a mouse model led to therapeutic levels of fetal hemoglobin production.

The Globe piece describes further efforts by Orkin to develop a gene therapy for sickle cell anemia by silencing BCL11A.  Unfortunately cut out of that version, but in the original STAT article, are descriptions of several other ongoing gene therapy efforts.  Novartis and Intellia are working on an ex vivo treatment, in which CRISPR-based engineering would be used to generate cells producing fetal hemoglobin, which would then be infused.    Immunogenicity of a wild-type protein the patient's immune system has never seen?  Sangamo Bioscience is also trying a gene therapy route, but by targeting an enhancer element which is key to BCL11A expression.  

Strangely missing from the STAT article is any mention of Bluebird Bio's ex vivo gene therapy treatment for sickle cell.  (Disclosure: Bluebird & my employer were both funded by Third Rock Ventures, though I've never had any involvement in Bluebird).  Bluebird is attempting to replace the mutant beta chain gene with a wildtype form.  Bluebird has had mixed success with this approach in clinical trials, which is a painful reminder of just how difficult clinical biology can be -- what could be more of a slam dunk than fixing a disease-causing mutation?  Well, nothing is a slam dunk in this business.

Also perhaps not simple is what the effects of interfering with BCL11A expression will be.  The BCL prefix expands out to B-Cell Lymphoma -- BCL11A over-expression appears to be a driver in some lymphomas.   Now, at least this is the other direction; sickle cell therapy attempts are trying to reduce expression of BCL11A.  Still, this is a reminder that we are tinkering with fundamental control mechanisms in blood-forming cells, and rude surprises could easily be in the future.

Begley's article points out one serious issue with even successful gene therapy treatments: these are expensive treatments, requiring cells to be removed from the patient, cultured and edited, and then reintroduced.  Sickle cell and the thalassemias are serious scourges of developing countries, whose health care systems would be unable to deliver such treatments to the bulk of their populations.

So, ideally a good small molecule would be found.  Hydroxyurea shows that this is possible, but has problematic side effects that limit its utility.  However, BCL11A (as noted in the longer version of the article) is a transcription factor, a class that has defied small molecule attacks. Orkin and others believe that a much larger push would get past this problem, but undruggability is often a reflection of very real biophysical challenges.  The American Society of Hematology plans to call for a "moonshot" on sickle cell, bringing in much more funding.

I could rant extensively on the poor analogy of a moonshot -- when Kennedy called for Apollo there were few serious scientific questions around the feasibility of landing a man on the moon.  But more worrisome is the declaration of this moonshot or that war on a particular disease.  Behind the scenes of every advance detailed in the STAT article -- discovery of the biochemistry of sickling, identifying BCL11A as a regulator of fetal hemoglobin or generating possible gene therapies -- was a mountain of basic research that wasn't directed towards any disease.  Just to give one example, our ability to study and manipulate DNA (from mapping genes to producing unlimited quantities of BCL11A protein to study to constructing lentiviral vectors for gene therapy) relies on work in the 1960s to understand the basics of DNA replication and why some viruses can infect certain bacteria but not others.  CRISPR fell out of bacterial genome sequencing -- but it took a lot of curiosity about strange repeats in bacterial genomes to move it along a non-obvious, twisting path to the genome editing powerhouse Cas9 is today.  Cas9 may even be superseded as a genome editing tool by a very different protein from an archean so weird it had only 6 prior PubMed citations on the organism.

Unfortunately, nobody calls for a basic research moonshot.  Hunting for biological weirdness or simply studying what is interesting is far more likely to get congressional calumny than to be funded in the first place.  One can hate the trend, but the public likes well-defined, targeted efforts -- until they don't (hello 40+ year War on Cancer!).  Those who call for moonshots also forget (or ignore) that once minimal success was achieved, Apollo was first ignored and then abandoned.  Apollo 13 didn't get live TV coverage when it all was "nominal"; only the threat to the astronauts lives brought network coverage back.  Apollo was intended to lead to long stays on the moon, but the later missions were cancelled.  I've thrilled to stand beneath the Saturn V at the Kennedy Space Center, but that thrill is tempered with the knowledge that it is not a mock-up above me, but a fully constructed moon rocket that was put on the shelf.  It is also worth remembering that Kennedy's 1970 deadline got us to the moon, but at the price of committing to a single technology, three astronauts dying horrible deaths in the Apollo I fire, the moral stain of relying on a bunch of  horribly tainted ex-Nazis and a program that was in many ways a technological dead end (as opposed to the Earth Orbit Rendevous plan, which would have evolved into permanent space stations).

If we must have "moonshots", perhaps we could at least have a small fraction of each program paid back into a fund to support basic research.  I know that is a very, very difficult sell -- advocates for any disease will be understandably loathe to let go of any funds they have rounded up.  But we must restock the larder of technologies, and there is always a possibility that the cure for sickle cell will come from a direction not even imaginable today. Targeting BCL11A expression appears to be a very well-validated approach to generating fetal hemoglobin expression, but perhaps there are others.  A decade or so ago there was a debate as to how much cancer funding should pour into cancer genomics.  I remain convinced that cancer genomics has yielded many important insights and led to new treatment approaches -- but that enthusiasm is tempered by the fact that cancer immunotherapy is right now looking like the approach that will have the highest yield of durable cures.   

Clinical medicine is tough, and setting funding priorities perhaps tougher.  Begley's article surfaces a lot of important points; let's hope that lots of powerful people absorb these and contemplate them.  There will probably never be a shortage of tragic unmet medical needs, but we cannot let some go underfunded due to our personal blinders.

No comments: