Too much optimism at NIH: Opinion

One researcher's reservations about Francis Collins's nomination as head of the agency

Jul 22, 2009
Neil Greenspan
I am not celebrating the nomination of Francis Collins to head the National Institutes of Health (NIH). My modest level of enthusiasm does not stem from concerns relating to his scientific accomplishments, administrative talent, political astuteness, or ability to communicate with average citizens, all of which are impressive. Nor do my reservations primarily follow from his religious commitments, although I disagree with a number of his assertions regarding the relationship between science and religion as well as some of his claims about the relative merits of evolution-related science versus non-scientific belief systems in explaining human behavior.My concern about Dr. Collins assuming the leadership of the NIH relates principally to the content of his claims for the medical promise of genomics, especially pertaining to therapy, and his assertions regarding the coming era of personalized medicine, which have included a promise of what he termed "medical nirvana" in responding to one radio interviewer. Where some see his enthusiasm for genomics as merely linkurl:"irrepressible optimism"; (1), I see insufficient concern with informing the public about biomedical realities -- specifically, this should be done in a more balanced manner, preserving hope while not generating unrealizable expectations. As Gardiner Harris put it in linkurl:The New York Times; of July 9 (presumably conveying the views of scientists not willing to be identified), "Dr. Collins cannot be blamed for the unexpected scientific hurdles facing genetic research, but he played an important role in raising expectations impossibly high" (2).So, before Dr. Collins takes the reins of the NIH, I would want to receive, and our representatives in Congress should demand, satisfactory explanations for many statements such as the following from an article he co-authored in the December 1999 Scientific American (3):"Within 20 years, novel drugs will be available that derive from a detailed molecular understanding of common illnesses such as diabetes and high blood pressure. The drugs will target molecules logically and therefore be potent without significant side effects" [emphasis added]. This statement is surprising, and was so at its publication, regardless of the effectiveness of genomics in identifying genes associated with diseases. The notion of drugs without side-effects (even a single drug!), is scientifically dubious if not preposterous. Molecules can never be absolutely specific for other molecules (4) and even if molecular interactions were absolutely precise, the inhibition or activation of one cellular or physiological process frequently affects functionally linked processes not intentionally targeted by the therapy. Even those with no special training in or knowledge of biology or medicine should be skeptical of such a prediction if only because every drug they have ever taken or heard about has been associated with significant side effects at some doses in some people.Consider the experience already in hand for drugs that target specific molecules "logically." While Vioxx worked as intended to inhibit a particular enzyme involved in inflammation, linkurl:cyclooxygenase-2,; it had a very long list of side effects including, most importantly for the lawsuits it engendered, increased risk for heart attack and stroke (5, 6). Epogen, an enormously profitable drug that effectively binds to a particular blood cell receptor, nevertheless has serious side effects, such as enhanced tumor growth in some cancer patients, some of which took years to fully appreciate (7).Quoting further from the same paragraph quoted above:"Drugs such as those for cancer will routinely be matched to a patient's likely response, as predicted by molecular fingerprinting. ... When people become sick, gene therapies and drug therapies will home in on individual genes, as they exist in individual people, making for precise, customized treatment."Collins's enthusiastic pitch for the notion that in the future drugs will be matched to the genotypes of individual patients is supported by some current examples in which particular treatments are avoided for patients possessing particular genes or are varied in dosage as a function of genotype at one or a few loci. However, he tends not to acknowledge the limitations of even these examples.An illustrative case is provided by the anti-retroviral drug, abacavir, used to treat AIDS. Abacavir causes a dangerous hypersensitivity reaction in patients with a particular gene involved in immune responses. However, based on recent studies, only about 55% of the individuals with the relevant gene are expected to exhibit the undesirable response on exposure to the drug (8, 9). Therefore, as many as 40-45% of those testing positive for the gene, and who could potentially use the drug safely, will receive an alternative agent. In other clinical situations, there may not be so many alternative agents of comparable effectiveness so that the inefficiency in such a genetic test, even if justified on balance, could incur its own costs in non-optimal treatment.The bolder implication of pharmacogenomics hinted at in the above quote, that everyone will receive genotype-based customization of treatment for whatever ails them, is problematic in at least two fundamental respects. Scientifically, it is doubtful that safe and effective genotype-specific drugs will be found to meet every need. Financially, it is difficult to imagine that pharmaceutical companies will be willing to invest the massive sums required to create such drugs for ever smaller genetically-constrained markets.Another disturbing aspect of Collins's genomic advocacy is his tendency, shared with some other proponents of genomics, to attribute to recent studies of human (or other) genomes insights that were in fact already arrived at through earlier genetics research. Consider the following quote from an article in The New Republic in 2001 (10): "Fortunately, ten years of intensive study of the human genome have provided ample evidence that these fears of genetic determinism are unwarranted." Anyone familiar with the term "genetic determinism" may have encountered one of the many articles or books (e.g., 11-13) that significantly predated the conclusion of the Human Genome Project and that were devoted to documenting the potentially complex relationships between genotypes and phenotypes. Collins and his co-authors admit as much in what follows, but the statement quoted above will conceivably cause unwary readers to credit our understanding of the subsequently described instances of "genes not being everything" to the power of genomics. For example, the authors of the article note that sickle cell disease, initially regarded as a so-called single-gene defect, can vary in severity even though the same disease-associated alleles at the hemoglobin beta-chain locus are present (in two copies) in the individuals with the clinically more and less severe forms of the condition. This interesting and important insight can be traced back to a study from 1978 (14). So, while Dr. Collins has many impressive credentials, talents, and skills relevant to directing the NIH, his tendency to make dubious claims for the future benefits of genomics is unsettling. The director of NIH should be a reliable and realistic source of medical information if the entire biomedical research enterprise is to remain credible. Therefore, in the future, Dr. Collins should harness his intellect to control his enthusiasm so that he is more realistic in his public pronouncements regarding improvements in medical care that will undoubtedly develop in part from new insights into human genetics and genomics.linkurl:Neil Greenspan; is an immunologist and professor of pathology at the Case Western Reserve University School of Medicine and director of the Histocompatibility and Immunogenetics Laboratory of University Hospitals Case Medical Center. The opinions expressed above are solely his own and do not reflect official views of the institutions with which he is affiliated. References: 1. J. Couzin-Frankel, "The promise of a cure: 20 years and counting," Science, 324:1504-07, 2009. 2. G. Harris, "Pick to lead health agency draws praise and some concern," The New York Times, July 9, 2009. 3. F.S. Collins and K.G. Jegalian, "Deciphering the code of life," Scientific American, 281:86-91, 1999. 4. B. Alberts et al., Molecular Biology of the Cell, Second Edition, Garland Publishing, Inc., New York & London, 1989, p. 94.5. R.S. Bresalier et al., "Adenomatous Polyp Prevention on Vioxx (APPROVe) Trial Investigators. Cardiovascular events associated with rofecoxib in a colorectal adenoma chemoprevention trial," N Engl J Med, 352:1092-102, 2005. Erratum in: N Engl J Med, 355:221, 2006.6. L.E. Levesque et al., "Time variations in the risk of myocardial infarction among elderly users of COX-2 inhibitors," CMAJ, 174:1563-69, 2006.7. M. Henke et al., "Erythropoietin to treat head and neck cancer patients with anaemia undergoing radiotherapy: randomised, double-blind, placebo-controlled trial," Lancet, 362:1255-60, 2003.8. S. Mallal et al., "HLA-B 5701 screening for abacavir hypersensitivity," N Engl J Med, 358:568-79, 2008.9. Personal communication, Simon Mallal, March 31, 2009.10. F.S. Collins, L. Weiss, and K. Hudson, "Have no fear. Genes aren't everything," The New Republic, June 25, 2001.11. R. Hubbard and E. Wald, Exploding the Gene Myth: How Genetic Information Is Produced and Manipulated by Scientists, Physicians, Employers, Insurance Companies, Educators, and Law Enforcers, Beacon Press, Boston, 1993.12. R. Lewontin, Biology as Ideology: The Doctrine of DNA, HarperPerennial, New York, 1991. 13. R. Lewontin, The Triple Helix: Gene, Organism, Environment, Harvard University Press, Cambridge, MA, 2000.14. R.P. Perrine et al., "Natural history of sickle cell anemia in Saudi Arabs. A study of 270 subjects," Ann Intern Med, 88:1-6, 1978.
**__Related stories:__***linkurl:Collins seen as boon to NIH;
[9th July 2009]*linkurl:The Human Genome Project +5;
[February 2006]