Chemistry

Personalized medicine An Introduction
What is personalized medicine Any prescription of drugs today by doctors and physicians all over the world is based on the diagnosis of the patients condition and a few generalizations as to what those symptoms mean, which have been developed from experience. However, the few minutes the doctor spends with us, the information he receives based on the symptoms we communicate to him, and the ones he gleans from various tests performed give him only a decent idea of the kind of ailment we are suffering from. Personalized medicine is based on the fact that the diseases are heterogeneous in their effect on humans that is, they differ from person to person in their causes, rates of progression, and their response to drugs. Several disorders which were earlier thought to be alike are now considered distinct, such as rheumatism or apoplexy, based on more information which showed the differences. For example, recent discoveries in the molecular pathology of cancer have shown the differences in the gene expressions of a variety of tumors.1 (Waring, 2001).  As technology advances, the doctors receive more and more information, which aids them in understanding the disorders and to treat them more effectively. Personalized medicine only takes this a step further. It is anticipated that, from certain new genetic, genomic, imaging, etc. technologies, doctors may in fact be able to individualize treatment for patients depending upon their exact condition. This suggests necessary changes to their lifestyle to even prevent future onset of certain diseases. Treatments will now be targeted at the root cause of a disease and not just the symptoms. This is based on the hope that the quick progress of technologies will bring an unprecedented amount of information for the doctor, and the semi-empirical nature of the current medical practices will be transformed.

There are several advantages of personalized medicine, both for the pharmaceutical industry and for the patients and doctors. For the pharmaceutical industry, it will reduce the timelines and costs of clinical trials, and will lead to product differentiation in the market place. For the patients, it will increase the probability of curing a disease with a particular drug, prevent unwanted side-effects, and on the whole, lead to better health and healthcare.2 Not only will it benefit the patients since the treatment will be so much more effective, it will also help save the millions of dollars that go into the development and testing of just one new drug in the world today.

Pharmacogenomics and Personalized medicine  Overview and Advantages
Pharmacogenetics and pharmacogenomics are two overlapping disciplines, the latter being more broad-based than the former. Pharmacogenomic investigations use latest sequencing technologies to simultaneously study the structures and expressions of many genes or gene patterns. The growth of this field was stimulated by the discovery that gene expressions are variable, and that they can be modified due to a number of factors. A product of one gene might alter the expression of the others, or there might be environmental factors including sleep, emotions, diet, etc.3  Most of the common non-infectious diseases are multifactorial and have complex genetic causes hence, a similar looking disease in two humans may actually be due to different genes.4 The methods of pharmacogenomics can be used to study these disease-causing genes which control the cause, effect, and response to treatment of any organism to a disease. Thus, doctors may be able to target the therapy based on the particular gene that is causing the disease in a particular organism, that is, personalize the medicine that is offered. Also, pharmacogenomics could also be used to predict the onset of certain diseases. For example, cholesterol testing is used to understand a patients risk for heart disease and has been used for the development of cholesterol-reducing drugs called statins. Similarly, genomics could provide new genetic andor protein biomarkers in blood to identify a patients risk for developing a disease. An example is that patients are now being screened for a new CCR5 antagonist, maraviroc, to identify whether a strain of HIV uses this to access healthy cells.5 Figure 1 shows various stages of a disease where pharmacogenomics and personalized medicine could help an individual.

There are already several cases where pharmacogenomics is used to supplement the choice of the drug to be offered to a patient. For example, if the drug is known to be metabolized by a genetically variable enzyme, then unwanted side-effects of the drug in a patient can be avoided by pretesting this enzyme in a patient and by administering the drug only when the results are normal. The most commonly used drugs are generally safe thus, such pretesting is not usually performed. On the other hand, if the drug to be administered is known to be toxic, such pretesting can be extremely useful. For instance, the enzyme thiopurine-methyl-transferase (TPMT) metabolizes the immunosuppressive drug azathioprine which is used to cure Crohns disease.6 If the drug is not properly metabolized due to a lack of the enzyme, it is converted into a toxic thioguanine nucleotide. Thus, it is obviously useful for the patient to pretest the activity of TPMT before this particular drug is administered. This issue of drug safety makes a particularly important case for pharmacogenomics. For instance, the Rofecoxib drug, developed by Merck and approved by FDA, was later withdrawn by the company due to increasing evidence linking rofecoxib administration to the risk of suffering a heart attack.7  Such drugs could be tested for their effects on genes, and such incidents could be prevented.
In order to achieve personalized medicine, we first have to collect the data from each individual and consider the effects of single-gene mutations. Such a collection of data will require tremendous effort on a very large scale. However, it has also been found that since there is less genetic variation in certain groups as compared to others, analyzing the genes of one person belonging to a particular group may be useful in determining the appropriateness of a particular drug for his entire community. For instance, deficiency of the multidrug metabolizing enzyme CYP2D6 differs widely between human races.3 Another example is the new drug BiDil which is used to treat heart failure and has been found to be effective only for people of African origin.8 These examples demonstrate how pharmacogenomics based testing can be used to administer more accurate drugs and make the treatment more effective.

Challenges for Personalized medicine
There are two major reasons as to why the move towards personalized medicine has been slower than expected the first one is the limitations of genetic predictions itself. There are only very few pharmacogenomics based target therapies available today. Even the most common test, Trasutuzmab (Herceptin, Genentech) has been used only for a very narrow, monogenic type of disease. The limitation is that most diseases are multifactorial and the genotype-phenotype relations are much more complex and difficult to predict, which gives us a very limited number of diseases that can be understood individually.9 Also, disease genes have proved much more difficult to find and validate than previously expected after the understanding of the human genome sequence. Several studies claim discovery of disease genes which are responsible for most of the diseases, and variations in the gene which modify the individual responses to that particular disease and to its treatment.

The shift towards personalized medicine has also been slower mostly due to a lack of sufficient economic incentives. Most of the studies which are usually small-scale tests   unfortunately fail when a larger population is considered. Also, even if a biomarker for a particular disease is identified, only the first step has been accomplished. The biomarker also has to provide sufficient and good quality information for it to be clinically useful. It is evident that large scale studies have to be performed before we validate that genes and disease onset and medication-response are correlated, and if so, by how much. The basic research that is essential to validate a particular biomarker as an accurate predictor is also very high. Such a large scale collection and analysis of data is bound to be expensive. It is clear that the government will have to provide economic incentives for the companies and research institutions to develop pharmacogenomics based tests for various diseases and to study the correlation between the genes and the diseases in large sections of the population. There is also the factor of time. Given the complex nature of the relationships between the biomarkers, the disease, and the multifactorial nature of most of the diseases, the development and testing of a new biomarker will take several years to understand and several more years before the required certifications can be met and before it comes out as a product in the market. There is also the big challenge of convincing both the physicians and the patients to shift to objective, marker-based health care. Table 1 summarizes the various challenges faced by pharmacogenomics and personalized medicine.

    It is clear that pharmacogenomics-based personalized medicine has to face several challenges, both scientific and economic, before it manifests as the norm of medical practice. We need to still generate a large amount of knowledge about accurate biomarkers and the correlations between gene and disorder before successful applications of the same can be developed. However, although the promise of personalized medicine seems far off, there is clear evidence that the favor is turning towards precise marker-based treatments rather than the traditional trial and error practice of medicine. Diagnosis is the foundation of medicine, and new diagnostics will provide the foundation for personalized medicine.10 With a concerted effort from the research institutions, the companies and the government, we might see this happening on a regular basis sometime in the near future.