Nature Biotechnology, 16/98, Charles R. Cantor, page 212
As our catalog of human DNA variability becomes more complete, the first practical consequences will be improved abilities to predict disease risk. Sometimes these predictions will be extremely accurate. In other cases, there will be only a loose association between the DNA difference detected and the probability of developing illness.
By itself, DNA sequence information can rarely provide definitive information about the function of particular genes. Studies of gene function include where and when a given gene is expressed and what the gene actually does. Function is complicated since it inevitably involves the interplay of many gene products at once. Methods that might reveal important insights into the functions of many genes at once do not appear to be close at hand. In the past, a lifetime of scientific research has often been required to understand the function of a single gene. The Human Genome Project has now provided close to 100,000 genes for study. Anyone who seriously contemplates the currently available data is overwhelmed by it. We will need to train a new generation of scientists with quite different perspectives to fully exploit the wealth of information already available.
Many genome scientists argue that studies in model organisms will rapidly accelerate our understanding of gene function. This will certainly be true for many genes, but it will surely not be true for all. Many functions important to the human species have no detectable counterparts in conveniently studied laboratory workhorses like yeast, worms, flies, or even mice. Yeast do not have brains, hearts, or any of the other major organs that can show pathophysiological effects in disease. Models for genetic diseases in mice have been created by laboratory procedures that place the precise mutation that leads to disease in the human into the corresponding gene in the mouse. Often, the resulting diseases are remarkably different. It is also self-evident that the brains of mice and humans are not equivalent. We cannot tell if mice have headaches, if they get depressed, or if they are schizophrenic.
Early advocates of the Human Genome Project often painted it as a medical panacea. This oversell was important in attracting attention and funding to the project, but now we must face reality. Many human diseases result from defects in genes. Many other diseases show variable symptoms that reflect the enormous genetic diversity of the human population. Genetic information does have the potential to revolutionize medicine, but realizing this potential is not necessarily straightforward. Knowledge of the molecular defect that leads to a disease is useful, but it does not usually provide a full understanding of disease pathophysiology or immediately useful methods of prevention or therapy. For example, we have known the precise molecular defect responsible for sickle cell disease for more than 20 years. Only recently has this knowledge provided a useful framework for improved disease management.
Improved diagnostics
The immediate consequence of the data provided by the Human Genome Project is a reference sequence for a presumably normal human genome. Differences from this norm are being continually discovered, and some of these differences are responsible for particular disease phenotypes or other abnormalities. Other differences correlate with these phenotypes but may not be causative.
As our catalog of human DNA variability becomes more complete, the first practical consequences will be improved abilities to predict disease risk. Sometimes these predictions will be extremely accurate. In other cases, there will be only a loose association between the DNA difference detected and the probability of developing an illness.
Most diseases arise from defects in function that are not determined by a single gene. DNA differences in any of the genes involved may result in a disease phenotype. Differences in some genes may even compensate for what would otherwise be a disease causing allele in another gene.
The situation is complicated by several additional factors: ( 1 ) Pathophysiology is complex-it can arise from many functions and their interactions; and (2), the human species is genetically very heterogeneous unsuspected or undetected variations can disguise the simple patterns expected from the rather clean genetics observed in inbred laboratory animals or plants. The phenotype reflects the genotype as it functions in an environment that is extremely heterogeneous in fundamental and important ways, such as the food we eat, the air we breath, the disease causing agents we are exposed to, and the medical care we get.
Despite these complications, the Human Genome Project is going to provide revolutionary improvements in predicting disease risk. However, a major problem is that this improved diagnostic power will not be, initially, matched by any significant improvement in disease management. For example, we can now diagnose, with near certainty, whether someone will contract Huntington's disease, a fatal inherited neurological disorder, but the only current benefit of performing this test is psychological. Those who test negative will know they are risk free.
Some heath-care insurers or potential employers may seek access to the results of such genetic tests to screen out individuals who could place a disproportionate cost burden on their economics. This trend should be vigorously resisted. There is essentially unanimous agreement among those who have seriously considered the results of the Human Genome Project that the genetic variations that characterize individuals should be their private information, like other medical records.
DNA sequence information is certain to lead to improved medical care. The first impact is likely to be improved choices among the therapies that already exist to manage particular illnesses. Medical practice currently stereotypes most individuals as 170-pound males and sets drug doses accordingly, even though it is well known that individual abilities to metabolize drugs vary widely. Many diseases are diagnosed by therapy. A patient's response to a certain treatment is taken as evidence of the presence of a disease known to respond to that treatment. Unfortunately, many therapies have risks, and, often, unpleasant procedures or side effects have to be endured before the correct therapy is found and a disease diagnosis assigned by it. However, by measuring human DNA variability and correlating it with therapeutic responsiveness, new diagnostics will be developed that can precede any therapy.
Another fairly certain early benefit from the Human Genome Project will be improved assessment of disease risk in cases where prevention is possible by changes in diet or behavior. For example, someone genetically predisposed toward a high risk of skin cancer will learn this and may make relatively simple lifestyle control measures to minimize risk. In other cases, diet control can reduce the risk of individuals genetically predisposed to common, complex diseases like cancer or hypertension.
Improved clinical trials
Knowledge of specific disease gene alleles will inevitably lead to methods for treating or curing these diseases. In most cases, this process will take one to two decades. For example, imagine a disease that involves improper communication between cells. The cause could be faulty signal reception, faulty signal transmission, or a faulty signal itself. Intervention at the level of many different gene products may be possible, but at present we are almost powerless to predict which ones will actually be clinically effective.
Usually, different alternatives must be tried, first in experimental animals and then in clinical trials. DNA sequence information is likely to have a major impact on reducing the costs of these trials. Individuals in trials often respond differently for reasons that are not immediately discernible. Genetic heterogeneity is frequently responsible. For example, if only 10% of the individuals in a trial respond well, the drug may be a poor choice and the trial should be ended.
Alternatively, for 10% of the population, those with particular genotypes, the drug may be excellent. DNA sequence analysis of variations in genes involved in drug responsiveness will help decide whether the drug is good for a subset of the population. In these cases an informative DNA diagnosis will then be available before the drug is used. This is an attractive scenario, since improved clinical trials will result in more drugs, more effective drugs, and cheaper drugs.
Gene Therapy and genetic variability
Pandora’s box for the Human Genome Project is, of course, the issue of gene therapy. Once we know all the genes and the variations that can lead to disease, we can contemplate replacing deleterious alleles with normal ones. Such gene therapy can assume two rather different forms. Somatic gene therapy uses a gene just like a drug. The goal is to target a particular cell type responsible for disease in order to replace or compensate for the defect causing that disease. Preliminary clinical studies of some such therapies look promising. This approach is only a small step beyond using the gene product, a protein, as a therapeutic. For some diseases like diabetes, gene products-proteins like insulin-have been used safely and successfully for decades. Issues of efficacy, side effects, and cost will decide whether it is better to use the gene or the gene product. No substantial ethical or legal issues are involved.
Germ line gene therapy, on the other hand, presents quite a different situation. Its goal is to change the genetic content of all of the cells of an individual in an inheritable way. Germ line gene therapy is already used in such procedures as sex selection by selective abortion. Disease alleles can be eliminated in precisely the same manner. This approach is relatively safe and cost effective. Whether it should encouraged or condoned must be decided by issues of quality of life and morality.
More precise forms of germ line therapy by manipulation of individual genes is a well developed technique in mice, and it will not be long before similar procedures are seriously considered in humans. In mice, we can add or destroy almost any gene at will, and there seems to be little interest in restricting this kind of manipulation. In contrast, it seems prudent that initial attempts at human germ line therapy should be restricted to rare cases in which a couple is genetically incapable of giving birth to a normal child without such drastic intervention. However, in cases in which selective abortion will eventually produce a normal child, it will be hard to justify other, more costly, and indeed more risky, procedures.
Soon humanity will be faced with a much more difficult decision. We have gained the power to alter the genetic constitution of our species, to improve it by whatever standards we choose to use. Whether this should be done is best judged from an evolutionary perspective. Evolution operates at the species level to ensure that the most fit species survive and prosper. Humans are already the dominant species on earth. We control the environment in a way that no species before us has even remotely approached. Thus we already control most aspects of our future evolution, but it is foolish to think that we totally control the environment.
Future disease susceptibility is unpredictable, and past examples are known where genetic variations have protected individuals against otherwise fatal diseases. The text book examples an sickle cell disease and thalassemia, two kinds of blood disorders that can be fatal if both gene copies are disease alleles. However, those individuals with one single disease allele and one normal allele fare much better, and as a side benefit are much less susceptible to fatal malaria. Thus people who live in or come from zones where malaria is endemic have a high frequency of these alleles. The current alarming increase in drug-resistant malaria may mean that such individuals are once again genetically advantaged.
It seems clear that genetic variability is a valuable resource for a species, even though some of its manifestations may cause consternation for individuals. Variability must be preserved, and the pitfall of genetic homogenization is easily avoided. However, a really significant moral dilemma posed by germ line gene therapy remains. Humankind is already master of its evolution by environmental control. Should it not take a much more active role and guide that evolution by genetic control? Perhaps the natural next step in evolution is when a species finally gains this ability. This is a terribly thorny question, relieved only by the realization that the author, and perhaps even the readers of this article, will probably not live long enough to have to answer it.
Moral dilemmas predate the Human Genome Project
It is unfair to hold the Human Genome Project responsible for moral dilemmas like population control and germ line gene therapy. These problems predate the project; they have been apparent ever since it was recognized that genetics applies to people as well as peas. The Human Genome Project raised no new issues, it just enlarged their scope.
Consider current concerns about human cloning. The procedure now seems technically feasible. Who wouldn't want to have a bank of spare parts, a source of perfect transplants, whenever they were needed? Fortunately, it seems easy to draw a line here. Identical twins are already clones. It would be difficult to justify banning a procedure that increased twinning, provided that the twins were treated as separate but equal individuals. One cannot be a source of spare parts for the other. The future prospect of banking tissues that can be turned into transplants seems harmless too.
Fortunately, the human brain is a big monkey wrench in any plans to bank whole human clones for future use. We are what we are because of the combined effects of genetics and the environment. This is especially true for the brain, which changes and evolves continuously in response to the environment. A brain transplant from a clone that has been kept in a dark freezer for forty years is a brain transplant no one will ever want.