Predicting Disease--Personalized Medicine in the Post-Genome Era

As the post-genome era approaches, what do you know about genes, and especially about DNA and the genetic code it contains?

DNA is a double-helix structure formed from four types of bases, and organized into 46 human chromosomes. The human genome contains approximately 3 billion base pairs, but most of them are "junk DNA." It is estimated that only about 35,000 genes have a real function. Scientists are studying these 35,000 genes that make up humanity's life code intensively, but their knowledge remains very limited, and they are still far from decoding most of them. But in the vast world of genes, even a single clue can bring about a change in medical treatments.

There are ethnic and genetic differences among people. Now that genetics is all the rage, society must carefully consider how to confront the problem of unequal treatment, discrimination, and segregation based on ethnicity or genetic makeup.

From natural to artificial selection

Genetic screening of embryos has allowed modern medical science to reduce congenital defects, and uncertainty about them, to a minimum. But some people have misgivings that such artificial selection, which smacks of eugenics, goes against nature and the will of God.

It is estimated that there are some 7000 different diseases caused by an abnormality in a single gene. Most of these are rare diseases and in most cases scientists have yet to identify the exact location of the culprit genes.

At present, more than 20 single-gene disorders can be identified by prenatal diagnosis in Taiwan, including thalassemia, Huntington's disease, dwarfism, glycogen storage disease, spinal muscular atrophy (SMA), Duchene muscular dystrophy (DMD), hemophilia, favism, congenital color blindness, and congenital deafness. The list goes on. Some of these diseases are not serious, such as glucose-6-phosphate dehydrogenase (G6PD) deficiency, which can cause hemolytic favism. All that people afflicted with this condition need to do to lead a normal life is to avoid certain drugs and foods. For them, prenatal genetic testing makes little sense.

Ko Tsang-ming, obstetrician-gynecologist at NTUH, says that in Taiwan prenatal genetic diagnosis is in principle carried out to identify serious, incurable diseases that would place a very significant physical and psychological burden on mother and child. When a serious disease is found, the pregnancy can be terminated quickly, and a tragedy can be avoided. In principle, no screening is carried out for non-serious hereditary diseases that don't make life intolerable.

In Taiwan 90% of prenatal genetic diagnoses are currently carried out to test for thalassemia. It is estimated that 300 million people around the world suffer from this disorder, including 1.4 million carriers in Taiwan alone. Ko Tsang-ming explains that there are two types of thalassemia: alpha and beta. If both parents are carriers of the same type of thalassemia, there is a 25% chance that their child will suffer from a severe form of anemia. Thalassemia leads to impaired hemoglobin synthesis and hemoglobin deficiency, and adversely affects the oxygen-carrying capacity of the hemoglobin in the blood, necessitating periodic blood transfusions or bone marrow transplants.

The thalassemia gene was discovered in 1978, and in 1991 the Department of Health promoted thalassemia screening for pregnant women. It is estimated that more than 1000 screening tests are conducted in Taiwan every year. Ko Tsang-ming says that because Taiwanese people are working ever more closely with their doctors, there has been a sharp drop in the annual number of children born with severe thalassemic anemia. Twenty years ago, almost 40 people with severe thalassemia were born every year; in recent years the number has dropped to less than four.

Newborn babies undergo genetic screening tests at birth. In 1985, genetic screening for five metabolic disorders was introduced. This is the largest genetic screening program currently underway in Taiwan. (photo by Diago Chiu)

A heavy burden

Without a doubt, prenatal diagnosis effectively helps many women who carry a hereditary disease to lower the risk of passing it on to their children.

There was the case of a man with dwarfism who married a foreign woman. Because dwarfism is an autosomal dominant condition, there is a 50% chance that the offspring of an affected person will be born with the condition. This particular couple's first child was unfortunately born with dwarfism. During the woman's second and third pregnancies, prenatal diagnoses indicated the presence of the dwarfism gene in the fetus, so the couple decided to terminate the pregnancies. The fourth time the woman got pregnant, the test showed a normal fetus. To the great delight of the couple, she went on to give birth to a healthy child.

However, the severity of some hereditary disorders is difficult to predict. This is particularly true of dominant hereditary genetic mutations, because it is extremely difficult to determine whether they will be transmitted, even if a screening test is carried out. Neurofibromatosis is a good example.

Ko Tsang-ming explains that neurofibromatosis is a common autosomal dominant inherited single-gene disorder. In approximately 60% of patients the symptoms are slight, and in another 20% the tumors can be surgically removed. But in the remaining 20% of patients the symptoms are very serious and severely affect their quality of life. The severity of neurofibromatosis depends on the type of a second mutation of the pathogenic gene, and genetic mutations are extremely complex, so that gene sequencing allows doctors to determine the location of the mutation in only 50% of cases. This makes it very difficult to predict how serious the condition will be.

Generally speaking, when prenatal diagnosis confirms the presence of an inherited disease, in 99% of cases the parents choose to abort. But there are a few exceptions in which screening tests show an abnormality but parents refuse to give up their child and turn to spirits and divination. Ko Tsang-ming laments that if the medium tells them that the child is normal, "they prefer to believe him rather than science, and there's nothing we can do about it."

DNA plays a key role in screening for inherited disorders, and in recent years it has also played an important role in paternity testing. The father, mother, and younger brother of a third grade junior high school student had attached earlobes, which is considered a recessive genetic trait, while he was the only one in the family with detached earlobes, a dominant trait. Because biology textbooks state that parents who share a recessive trait cannot produce offspring with the corresponding dominant trait, the boy suspected that he had been given to the wrong parents in the maternity ward. In the past, such doubts had to be repressed, because there was no scientific way of determining the truth. Today DNA testing is a technique with proven results. When the four members of the family underwent a paternity test at Dr. Ko's clinic, the test showed that the boy had not been switched at birth, and that they were all members of the same family.

Dr. Ko says, "The textbooks are wrong." How are attached earlobes passed from one generation to the next? Is it a dominant trait? And how is it manifested? Given that scientists have not reached a final conclusion on this question, textbooks should not be categorical about it, to avoid causing further confusion.

"Like father, like son" is an old saying, but it conveys quite accurately how genes pass on hereditary traits.

Neonatal screening

In addition to prenatal genetic diagnosis, newborn babies also face genetic screening tests at birth.

In July 1985, Taiwan introduced a comprehensive neonatal screening program targeting phenylketonuria, homocystinuria, galactosemia, G6PD deficiency (commonly associated with favism), and congenital hypothyroidism. Major medical centers in Taiwan such as NTUH and Veterans General Hospital use tandem mass spectrometry to screen some 200,000 newborns a year for metabolic disorders. This is the largest genetic screening program currently underway in Taiwan.

There are screening tests for more than 20 other metabolic disorders. But because the tests are very costly, and the disorders very rare and currently untreatable, the tests are not really worth the trouble. The DOH therefore has not included these tests in the subsidized screening program. If private individuals insist on them, they can pay for them out of their own pocket.

Chen Yuan-tsung, director of the Institute of Biomedical Sciences, Academia Sinica, says that what makes neonatal screening different from embryonic screening is that it aims to detect rare metabolic disorders that can be treated or prevented if they are discovered at an early stage. Infants are therefore given the chance to grow up healthy, or to suffer from fewer sequelae. In Taiwan, one in 34,000 infants are born with phenylketonuria, but provided they are given an appropriate diet during infancy, their prognosis is good and they are likely to attain a normal IQ.

The pathogenic gene must be inherited from both parents for the disease to occur.

Waiting to get sick?

Genetic screening after the embryonic stage and infancy is very controversial, because groundbreaking treatments still remain to be discovered for a large number of hereditary diseases. Even when screening tests indicate the presence of a disorder before any symptoms appear, no radical cures are available. This can cause considerable unnecessary distress.

Huntington's disease is the best known example. Its pathogenic gene was discovered in 1984, but medical science still has no treatment for this disease. All patients can do after the disease is identified in a screening test is to wait for it to manifest itself. The same is true of insulin-dependent diabetes mellitus: it can be tested for, but not prevented.

ARCAP, the first mutant cancer gene to have been patented in Taiwan, also falls into this category. Lu Ming-fong, president and CEO of Targetgen Biotechnology Co., points out that his company is currently not ready to advocate screening for mutant genes that cause liver cancer, because there is no treatment for this disease. Lu says, "Once you identify the gene, what are you supposed to do? Allow life to become hopeless?" According to Lu, as long as there is no drug treatment for a particular cancer, all screening does is to alarm patients. It does them no good at all.

Chang Tai-jay, associate principal investigator at the Genome Research Laboratory of the Department of Medical Research and Education, Veterans General Hospital Taipei, points out that ARCAP cannot be used to predict whether a person will acquire a congenital gene that causes liver cancer, but it does facilitate the early detection of postnatal mutations in liver cancer genes. Early detection gives patients the opportunity to undergo treatment and to change their lifestyle, including their diet, and may prevent activated cancer genes from producing cancer symptoms. In some cases this may save the patient's life.

A little over a year ago, NTUH began to conduct breast cancer screening for women with a high-risk family history of breast cancer. The hospital is the first medical center in Taiwan to put genetic screening for breast cancer into clinical practice.

Su I-ning, professor of medicine at NTUH's Department of Medical Genetics, notes that in the majority of cases, breast cancer develops randomly and cannot be predicted. But breast cancer is hereditary in 5-10% of cases, of which 30-50% are caused by a BRCA1 or BRCA2 gene mutation.

Dr. Su points out that BRCA1 and BRCA2 are autosomal dominant genes, which means that offspring of individuals with one of these gene mutations have a 50% chance of inheriting the gene mutation. Even without a family history of the gene, one person in 200-300 is a carrier of the gene. If a screening test reveals that an individual carries one of these gene mutations, there is an 80% that she will develop breast cancer.

Su therefore recommends that women in high-risk groups consider undergoing genetic screening. High-risk groups are those who have a family member who has had breast cancer and ovarian cancer, or two or more family members who have contracted breast cancer before age 50, or a male family member who has contracted male breast cancer, or someone who has had early-onset cancer (before age 40).

Su explains that NTUH currently offers breast cancer screening for whole families. The fee is NT$20,000 per family, and the degree of accuracy is 95-98%.

Screening easy, treatment hard

Although the intentions behind genetic screening are good, it is an unfortunate fact that screening is easy but treating and curing a disease much more difficult. Once a test reveals a disorder, what do you do next? It's a tough question.

Su I-ning points out that given that there is no effective treatment for breast and ovarian cancer, all she can recommend at present is that affected women undergo a mastectomy, ovariectomy, or chemotherapy, or take medication to prevent the onset of the cancer. Otherwise, the patient ought to be urged to undergo regular tests.

BRCA1 and BRCA2 gene mutations are associated with breast cancer as well as ovarian cancer. Dr. Su explains that because ovarian cancer has no symptoms, it is often not detected until it has reached an advanced stage. A preventive ovariectomy not only prevents ovarian cancer but can also affect hormone levels and reduce the incidence of breast cancer.

Nevertheless, given the destructiveness and substantial side effects of preventive interventions, as well as the fact that BRCA1 and BRCA2 gene mutations do not necessarily mean that the patient will develop breast cancer, women often hesitate to subject themselves to surgery.

"It's a bit of a gamble," says Dr. Su. In the past year or so, she has seen 40 to 50 cases of breast or ovarian cancer, but only a tiny minority of women has agreed to have a mastectomy or an ovariectomy. Given that the effectiveness of preventive chemotherapy has not yet been demonstrated, even fewer women are interested in taking that risk.

"It's not just a case of better safe than sorry," says Dr. Su. Given that effective treatments have yet to be discovered, doctors certainly don't always urge people to undergo genetic screening. It depends on the patient's attitude and individual maturity.

A brave new world?

With mapping of the human genome nearly complete, scientists around the world have set themselves the goal of finding the genes responsible for a wide variety of human diseases.

The trouble is that although numerous single-gene disorders have been identified, the majority of chronic diseases and cancers are caused by the complex interactions of a multitude of genes. Moreover, a single gene can have a variety of functions, each one of which can affect the entire human organism.

For example, the gene that causes sickle-cell anemia, which can cause severe pain in the bones, joints, and abdomen and is found predominantly in black people, also has an antimalarial effect. Scientists have therefore conjectured that in Africa, where malaria is rife, the sickle-cell gene has survived in the population as an evolutionary protection mechanism. In other words, repairing the sickle-cell gene may actually increase the incidence of malaria. Moreover, the manipulation of even a single gene may have an overall impact that will not be apparent over the short term.

Dr. Ko Tsang-ming says, "Nature works in exceedingly complex ways. Unlocking its secrets may be easy at first, but understanding how genes are formed and how they operate is only the beginning of our challenge." Although the completion of the Human Genome Project will constitute a great achievement in the history of medical science, there is a long road ahead in the study of genetic disease, and gene therapy is still very limited in its scope. Consequently, it is imperative not to raise people's hopes too high that cures will be found for the great majority of disorders whose genetic structure and characteristics are extremely complex.

Although important breakthroughs are difficult to achieve in a short space of time, the global medical community is still sparing no effort to investigate the human genome. The Academia Sinica's Taiwan genetic database project plans to collect genes from some 500,000 people and to conduct follow up studies over 20 to 30 years to investigate the relationship between genes and the environment. Chen Yuan-tsung, director of the Institute of Biomedical Sciences, says that it is difficult to predict what results this project will yield, but he is confident in saying that it must be undertaken.

In addition, Academia Sinica's "National Clinical Core for Genomic Medicine" has begun a research project to find the genetic basis of five disorders commonly found in Taiwan: diabetes, hypertension, bipolar disorder, asthma, and osteoarthritis. Hypertension has been investigated over the past three years, and initial results will be published shortly.

If the mother has one defective X chromosome and one normal X chromosome, and the father's X and Y chromosomes are both normal, a son has a one-in-two chance of manifesting the disorder, and a daughter has a one-in-two chance of being a carrier.

A holistic approach

Unlike genetic screening to identify disease genes, which is beset by practical and ethical dilemmas, genetic screening to identify drug reactions is relatively uncontroversial, and can be immediately applied to clinical practice.

Most people would be surprised to learn that medical science does not know why and for whom most drugs we take today are effective. In fact, drug companies take a buckshot approach to drug manufacture, and then try them out on various diseases. In other words, people who take medication are essentially guinea pigs.

Lu Ming-fong, president and CEO of Targetgen Biotechnology Co., says that clinical trials have shown that on average a given drug is only effective in 70% of people. Of the remaining 30%, 20% do not respond to the drugs they take. What's worse, drugs are not only ineffective in the remaining 10% of patients, but actually harm them.

The different ways people react to drugs are due to what in traditional Chinese medicine is known as differences in constitution.

Lu Ming-fong says, "If we combine genetic research with traditional Chinese medicine we will make the road ahead much shorter." Traditional Chinese medicine divides the human organism into the five different phases, or constitutions, of metal, wood, water, fire, and earth. Genes also show that different people have different constitutions. For example, it appears that people with cardiovascular diseases rarely get cancer, because heart disease is what Chinese medicine calls a "hot" condition, whereas cancer is a "cold" one. Chinese medicine merely describes various bodily constitutions, but not the underlying causes.

Given that constitutions differ, the efficacy and dosage of medicines also varies from person to person. Ku Chih-chieh, vice-president of Vita Genomics Inc., points out that in the past Taiwanese doctors never called into question new drugs developed in advanced countries such as the US, Europe, or Japan. They assumed that as long as a drug underwent clinical trials and was approved abroad, it would naturally have the same effect on Chinese people in Taiwan. As a matter of fact, this is far from being the case, as the example of anesthesia demonstrates. Ku notes that Chinese people have a relatively strong response to anesthetics. If Western dosages were the standard applied in Taiwanese hospitals, 1-2% of patients would die on the operating table.

A pathogenic gene inherited from either parent will cause the disease.

We won't be guinea pigs

Thanks to pharmacogenomics, which integrates gene research and drug research, Western medicine is moving away from "treating the disease," and toward "treating the patient." Thus Western medicine is moving in a new direction: that of personalized medicine.

Two prominent success stories are screening tests to determine the effectiveness of interferon and tests to identify allergic reactions to antiepileptic agents.

Ku Chih-chieh points out that interferon, which is used to treat hepatitis C, is only effective in 50% of patients, and has side effects in 100%. In some patients, interferon causes headaches, aches from head to foot, long-term cold symptoms and diarrhea, and in severe cases even obesity or suicidal tendencies. Moreover, interferon is very expensive: one four-to-six-month course of treatment costs US$9,000-20,000.

Interferon treatment has painful side effects, takes six months to a year to show results, and is very expensive. If the treatment fails, it will still have cost a considerable amount of money and suffering.

But if hepatitis C is not treated, it has a 10-15% chance of turning into liver cancer. It costs 30 times more to treat liver cancer than hepatitis. Considering that in mainland China a third of medical resources are spent on caring for patients in the final stages of liver cancer, the situation is clearly very serious.

How do we know in advance whether a patient will respond to interferon? An effective method currently available is a DNA test used in pharmacogenomics. Ku Chih-chieh says that more than ten genes that have a correlation with this drug's efficacy have been identified to a level of accuracy of approximately 85%. Vita Genomics is discussing cooperation and commercialization plans with medical schools in Taiwan, Hong Kong, and the United States.

Drug effectiveness is an important question, but the harm caused by drug allergies-for instance, drug-induced skin reactions such as Stevens-Johnson syndrome (SJS)-is an even more serious issue. SJS causes mucosal ulcerations and blisters all over the body, and carbamazepine, an antiepileptic medication, is the most common cause. Chen Yuan-tsung points out that up 15% of people who take carbamazepine die of an allergic reaction to it. In the remaining 85%, the drug can have side effects or sequelae such as long-term inflammation of the cornea or an esophageal disorder.

Academia Sinica and Chang Gung Memorial Hospital have jointly discovered the HLA-B1502 genetic marker, which can be used to predict carbamazepine-induced skin reactions with 100% reliability. Thanks to genetic screening, deaths caused by allergic reactions to carbamazepine should be a thing of the past.

The rapid advance of genetic research is giving substance to the medical world's high hopes for the future. As the function and interactions of our 35,000 genes are analyzed and explained, genetic screening will facilitate effective disease prevention and drug treatment at every step of human development, from the embryonic stage to old age. The age of personalized medicine seems sure to arrive, with all the benefits it will bring.

Genes contain an individual's life code. Genetic screening makes it possible to trace a person's ancestry and to detect hereditary diseases.