From Blockbuster Medicine to Personalized Medicine

Jan Trøst Jørgensen

Personalized Medicine. 2008;5(1):55-63. 

Abstract and Introduction

Abstract

One of the biggest challenges for the biotechnology and pharmaceutical companies in the 21st century will be to develop and deliver drugs that fit the individual patient's biology and pathophysiology. This change from blockbuster medicine to personalized medicine will, to a large extent, influence the way that drugs are going to be developed, marketed and prescribed in the future. These changes could mean an end to the blockbuster philosophy in 'big pharma' and thereby impose major changes in company structures. The implementation of personalized medicine will be a stepwise process, where the division of patients into biological subgroups will be the first important step. Today, this is already the situation for several cancer diseases, for example, breast cancer. In the years to come, we will see more and more drugs being prescribed based on the results from pharmacodiagnostic testing. Within cancer medicine, which has been at the forefront of this field, it is expected that in 10-15 years time very few drugs will be prescribed without such a test.

Introduction

One of the biggest challenges for the pharmaceutical industry in the 21st century will be personalized medicine. Will 'big pharma' be able to develop and deliver drugs that fit the individual patient's biology and pathophysiology? It has been said that personalized medicine could be the deathblow for the blockbusters, thereby reducing the profitability of pharma companies. Most, if not all, of the big pharmaceutical companies are dependent on the profits generated by their blockbusters, and niche products do not seem to be sufficiently attractive to the pharmaceutical companies in today's market.^[1] In the future, it may be a pharmacodiagnostic test that decides which drug to choose for the individual patient and, to a much lesser extent, the marketing efforts of the pharmaceutical companies, opinion leaders and average results from large controlled clinical trials. When personalized medicine really takes off, the changes that will follow could have dramatic consequences for pharmaceutical companies and the healthcare system in general.

Whether the introduction of personalized medicine will produce the consequences described above will depend on how quickly research within the area evolves and how adaptive the big pharmaceutical companies prove to be to the new situation. For the patients, personalized medicine could mean more effective drugs with fewer side effects. The healthcare system will be able to provide a more targeted drug therapy to the patients with less medication errors and less costs for treatment of drug-related side effects. Altogether, this results in a more cost-effective drug therapy for the payers. It is also hoped that personalized medicine will provide mechanistic insight into the pathogenesis of different diseases and their response to drug therapy thereby adding a new scientific dimension to medicine.^[2] Personalized medicine will also have a downside, because not all drugs will be available to all patients owing to differences in patients' biology and pathophysiology, and this may be viewed as an ethical dilemma.

This article will briefly describe the background for the introduction of personalized medicine, who drives the process and their motives. Furthermore, a few examples of personalized medicine mainly from the treatment of cancer diseases will be given. Within oncology, the first step towards a more individualized drug treatment for patients was already taken decades ago.

Why Personalized Medicine?

All healthcare professionals who, one way or another, are involved in the medical treatment of patients know that the same drug does not work in the same way in each patient. Some patients do not respond positively to treatment at all and may only suffer the side effects. A relatively large number of patients treated for cancer, infectious diseases, psychiatric illnesses, respiratory diseases and cardiovascular conditions are not responding to the drugs they are given.^[3,4] The philosophy behind personalized medicine is that every patient has a unique biology and that this should be reflected in the choice of medical treatment, resulting in improved efficacy and reduction of side effects. In many ways personalized medicine can be regarded as the 21st century's answer for the rational use of drugs - the right drug to the right patient. With the new molecular diagnostic methods, physicians will be provided with an objective tool that will possibly improve medical treatment within a number of disease areas.

There are many definitions of personalized medicine, but the use of molecular diagnostic methods is within most. At a meeting in November 2006 in Boston organized by Harvard Medical School and Harvard Business School, the following definition was suggested: 'The management of a patient's disease or disposition by using molecular knowledge to achieve the best possible medicinal outcome for that individual'.^[24] The molecular methods that are frequently discussed in relation to personalized medicine are the 'omics', such as genomics, proteomics and metabolomics, which are factors affected by genetic variation. With respect to these variations, both germline (e.g., hereditary mutations) and somatic (e.g., nonhereditary mutations, such as most cancers) mutations must be taken into consideration.

Personalized Medicine and Cancer

From a historical perspective, personalized medicine should be of no surprise to the companies that work in cancer drug development. With the discovery of the estrogen receptor in the 1960s and the introduction of the anti-estrogen tamoxifen in the 1970s the road was prepared for a more individualized medical treatment approach to cancer. Results from clinical studies demonstrated that a high proportion of the patients with estrogen receptor-positive breast cancer responded to treatment with tamoxifen.^[5,6] In the 1990s, another targeted drug was introduced that was aimed at a selected group of cancer patients: women whose breast cancer tumors overexpressed the human epidermal growth factor receptor (HER)-2. The drug was the monoclonal antibody trastuzumab (Herceptin^®, Genentech, CA, USA), which was specifically targeted toward the HER2 protein of tumor cells.^[7,8] The effect of trastuzumab is dependent on overexpression of HER2, or overamplification of the HER2 gene, which is tested for by a pharmacodiagnostic test. These protein or gene changes occur in 20-25% of all women with breast cancer^[9] and the use of trastuzumab has been shown to be especially effective as adjuvant treatment in combination with chemotherapy. Results from large Phase III studies in HER2-positive women with breast cancer have shown that treatment with trastuzumab and chemotherapy can reduce the risk of relapse by approximately 50% compared to conventional treatment.^[10,11] During the last 10 years, a number of new anticancer drugs especially developed for a selected group of patients have been introduced. The requirement for use of some of these drugs is preselection with a pharmacodiagnostic test. A number of these targeted anticancer drugs, their targets, the indication and the type of pharmacodiagnostic used for selection of patients for treatment are listed in .

For most drugs, such as tamoxifen and trastuzumab, the pharmacodiagnostic tests are used to select the patients who are most likely to benefit from treatment; but such tests could also be used to predict toxicity. Many anticancer drugs produce severe side effects and some of them are due to variation in drug metabolism. That is the case for irinotecan (Campto^®, Pfizer, NY, USA), a topoisomerase-1 inhibitor used for treatment of colorectal cancer, which is metabolized by the UDP-glucuronosyltransferase 1A1 (UGT1A1) enzyme. Genetic polymorphism of the UGT1A1 gene is related to severe toxicity caused by the drug, such as leukopenia and diarrhea.^[12] In order to identify the group of patients with aberration of the UGT1A1 gene who will need a reduced dose of irinotecan, a pharmacodiagnostic test has been developed (Invader^®UGT1A1 Molecular Assay, Third Wave Technologies, WI, USA).^[25] A similar genetic test to predict toxicity of 5-fluorouracil or capecitabine (Xeloda^®, Roche, Basel, Switzerland) has recently been introduced (TheraGuide 5-FU™, Myriad Genetic Laboratories, Inc., UT, USA). 5-fluorouracil or capecitabine-based chemotherapy is frequently used for the treatment of breast cancer, colorectal cancer and head and neck cancer. The test detects variations in the genes for dihydropyrimidine dehydrogenase (DPYD) and thymidylate synthetase (TYMS).^[26]

It is not only for new anticancer drugs that pharmacodiagnostic tests are developed. As our knowledge about tumor biology increases and the mechanism of action of the drugs is explained, it has been possible to develop pharmacodiagnostic tests for anticancer drugs that have been used in the clinics for years. One example of this is the predictive fluorescence in situ hybridization (FISH) test for anthracyclines in the treatment of primary breast cancer (TOP2A FISH, pharmDx™, Dako, Glostrup, Denmark).^[4,13] Today, there are already a number of drugs within oncology that require a pharmacodiagnostic test in order to select the patients who are most likely to respond to treatment, and it is expected that more tests will be developed in the years to come.^[14]

Cancer has been at the forefront of personalized medicine compared with other disease areas and there are probably several reasons for this. The diagnosis of cancer is almost always based on a biopsy and subsequent examination of cells or tumor tissue. The different slide-based technologies, such as immunohistochemistry, FISH and chromogenic in situ hybridization, used in the pathology have paved the way for pharmacodiagnostic testing.^[4] Furthermore, cancer is a potentially life-threatening disease that develops from a series of genetic changes and our understanding of the genes and related proteins involved in the disease process has increased considerably over the last 10-20 years. In fact, most of the recently developed anticancer drugs listed in reflect this knowledge.

Personalized Medicine and Other Diseases

The fact that a group of patients with a certain genotype respond differently to a medical treatment compared with the average population is known from disease areas other than cancer. For instance, it has been shown that individuals with mutations of the melanocortin-1 receptor gene (MC1R) have a higher anesthetic requirement with desflurane (Suprane^®, Baxter, IL, USA) than persons with a normal genotype.^[15] The same observation has been found concerning topical and subcutaneous administration of lidocain.^[16] In relation to these observations, it has been suggested that the melanocortin-1 receptor is involved in the modulation of stimulus that might be perceived as painful. Interestingly, red hair nearly always results from mutations in MC1R, so it seems that the anecdotal impression that redheads are more sensitive to pain may have found a biological explanation.^[16]

There are several other examples where genetic factors can influence the outcome of drug treatment, mainly through the effect on metabolism. Among the enzymes involved in drug metabolism, the cytochrome P450 (CYP) family plays a central role. Several factors can influence CYP enzyme activity, such as diet, environmental chemicals, age, gender and, not least, genetic polymorphisms. The activity of the CYP enzymes has been reported to vary up to 50-fold between individuals for some metabolic reactions.^[17] Many drugs such as anti-epileptics, antidepressants, β -blockers, proton-pump inhibitors and anticancer drugs are metabolized through the CYP2D6 and CYP2C19 enzymes. Aberrations in the genes for these enzymes can contribute substantially to the variability in response to drug therapy, which may result in toxic blood levels or ultra-rapid clearance.^[17,27] A number of drugs that are metabolized by the CYP2D6 and CYP2C19 enzymes are listed in . A microarray-based pharmacodiagnostic test (AmpliChip CYP450 Test, Roche, Basel, Switzerland) has recently been introduced and such a test will hopefully be able to identify patients with polymorphisms in the CYP2D6 and CYP2C19 genes, thereby reducing the number of drug-induced side effects and drug failures.

One of the latest examples of a drug combined with diagnostic testing is related to the HIV drug maraviroc (Selzentry™, Pfizer, NY, USA), a CCR5 coreceptor antagonist. In the labeling for this drug it is stated that a tropism testing to identify patients infected with R5 virus should guide the therapeutic use of maraviroc.^[28] At the same time as maraviroc was approved by the US FDA, a tropism assay (Trofile™, Monogram Biosciences, CA, USA) was launched. This assay was also used for patient selection in Phase III clinical trials that Pfizer performed with maraviroc.^[29]

Another new important drug-related diagnostic test that has recently obtained FDA approval is the Nanosphere's Verigene Warfarin Metabolism Nucleic Acid Test (Nanosphere, IL, USA). This test detects variants of the CYP2C9 and VKORC1 genes that are related to metabolism of the anticoagulant warfarin (Coumadin^®, Bristol-Myers Squibb, NY, USA). The approval of this test has resulted in changes in the labeling for Coumadin and the generics of the drug. Warfarin is a difficult drug to manage and one-third of patients receiving this drug metabolize it quite differently than expected and experience a higher risk of bleeding. This new test may help the physician to assess whether a patient is especially sensitive to warfarin and whether dosage adjustment is needed.^[30]

Who Drives the Process?

Will all future drug treatments be individualized? Probably not, as there needs to be a clear rationale within a given therapeutic area that must be explained by an unmet medical need of today. The use of simple analgesics, such as acetylsalicylic acid, will probably not be subject to pharmacodiagnostic testing in the near future, but these will be the drugs that are likely to be used within oncology, cardiovascular and infection medicine. Based on the number of pharmacogenomics publications in PubMed at the end of 2006,^[24] it also appears that these disease areas are among the most research intensive (Figure 1). This is not surprising since within these areas of medicine there are still a great deal of unmet medical needs.

So far, the larger pharmaceutical companies have not been in the drivers seat with respect to the implementation of personalized medicine. Personalized medicine goes against the blockbuster philosophy where a drug should be used broadly by as many patients as possible and for different indications. With our increasing knowledge about drug therapy, we now know that 'one-size-fits-all' is very rarely the situation. An implementation of personalized medicine will lead to a segmentation of the market, thereby reducing the potential for a given drug.^[1]

The larger pharmaceutical companies have a problem with the efficiency of development as described by the FDA in relation to the Critical Path Initiative.^[31] Over the last 10-15 years, the number of new drug applications submitted to the regulatory agencies has declined while the research and development spendings have increased, especially the costs of clinical development, which have escalated rapidly.^[31,18] The use of predictive biomarker tests in clinical development may be one of the tools that could improve the efficiency. Pharmacodiagnostics could help to decrease the variability of treatment response through an identification of responsive subgroups. This would lead to a decrease in the size of clinical trials by reducing the number of patients who fail to respond to therapy.^[2] Within cancer it has been demonstrated that the number of patients needed can be reduced substantially by using a pharmacodiagnostic test.^[19] A drug like trastuzumab would probably not have been developed if a pharmacodiagnostic test did not exist to identify the patient population likely to respond.^[20]

The players that have been most active in promoting the idea about personalized medicine are the academic groups, patient advocacy groups, authorities and, to some extent, the health insurance companies. The reason for this interest by the authorities and insurance companies may partly be explained by the high drug costs of new targeted therapies. The use of a pharmacodiagnostic test will ensure that these expensive drugs are only given to the patients who have a reasonable chance of responding to treatment. In the UK, the annual costs, including administration and medical care, for several of these new anticancer drugs exceeds £50,000 (∼US$100,000; ) and the high costs of these type of drugs are likely to create a significant financial pressure on the healthcare budgets.^[21]

The FDA is a very active player in the implementation of personalized medicine and they see their Critical Path Initiative as a step in this direction.^[31] They regard personalized medicine as a way to improve the safety, efficacy and cost-effectiveness of drug therapy.^[18] Furthermore, in 2005, the FDA issued a concept paper on how they see the future drug-diagnostic codevelopment for new targeted drugs.^[32] It is expected that this concept paper will be translated into to a real guideline in the near future. With respect to the development of personalized medicine, the USA may also be regarded as the leading country today, which should mainly be attributed to the strong commitment seen among the important players such as the FDA. Relatively slower approaches have been taken in other parts of the world, but certainly activities are ongoing in Japan, China, Australia, New Zealand and Europe.^[33] Denmark seems to have taken the lead in Europe.^[34] A number of biotechnology, pharmaceutical and diagnostic companies in the major biotech cluster, known as 'Medicon Valley', embrace the concept of personalized medicine.^[35,36]

The interest in personalized medicine is increasing worldwide and in 2004 a nongovernmental organization was formed in the USA. The name of the organization is the Personalized Medicine Coalition (PMC), whos goal is to advance the understanding and adoption of personalized medicine.^[37] The members of the PMC are a diverse group comprised of participants from industry, universities, academic medical centers, relevant trade associations, patient advocacy groups, government officials, healthcare providers, health insurance companies, information technology companies, venture companies and strategic partners. If we take a closer look at the industry group, this mainly included the diagnostic companies whereas the large biotechnology and pharmaceutical companies are more sparsely represented. The membership composition may reflect the interest that the large biotechnology and pharmaceutical companies still have in personalized medicine.

Future Perspective

We as humans regard ourselves as unique and individual persons and when we as patients interact with the healthcare system we bring this perception with us. From a biological point of view this may not only be perception. The use of pharmacodiagnostic tests and other molecular methods has given us an insight into the biological variation that underlies the observed differences in the response to drug therapy. We are now starting to understand that some diseases, for example, cancer, can be divided into different biological subgroups that respond differently to a specific medical treatment and that drug therapy needs to be individualized to a much greater extent than it is today. One of the explanations for this interpatient variation probably lies in the genes that express the targets for drugs, such as receptor proteins, enzymes or signal peptides. The introduction of personalized medicine or stratified medicine, which may be a more precise name for this type of medicine, will be a stepwise process where the division of patients into biological subgroups will be the first important step.^[22] As our knowledge about the pathophysiology at the molecular level increases, the different biological subgroups will become smaller and smaller. Fully individualized drug therapy is not something that will be introduced in the near future; we are probably talking about several decades before it becomes a reality.

If personalized medicine is to have a real breakthrough there need to be incentives for those who are going to do the research and development work - the pharmaceutical and diagnostic companies. The use of predictive efficacy and safety biomarker tests will most likely reduce both development time and costs considerably, owing to a smaller clinical trial size and, most importantly, by reducing the failure rate of drug development programs.^[22] This could, of course, be one of the incentives, but pricing and reimbursement should also be attractive to the companies. Certainly, there needs to be a flexible pricing and reimbursement system in place for personalized medicine that is able to reward the higher ratio of benefit to risk for this type of treatment.^[23]

As yet, larger pharmaceutical companies, with a few exceptions, have not been very active in the development of personalized medicine. The fear of losing business could be one of the explanations, despite the fact that 'personalized drugs' like trastuzumab and imatinib (Glivec^®, Novartis, Basel, Switzerland) have become blockbusters, but the organizational structure of these companies may also be a likely cause. The transformation from 'blockbuster medicine' to 'personalized medicine' may be a very difficult process for the larger pharmaceutical companies.^[1] This would mean many changes in the whole chain from preclinical and clinical development to the way these drugs are going to be marketed. No doubt the development of drugs to smaller groups of patients could be attractive businesswise,^[22] if the right organization is in place. In the future, we may see that some of the larger pharmaceutical companies are split into smaller units or divisions in order to adapt to the new situation, thus mimicking the structure of the smaller biotech companies. We may even see research and development being split from the sales and marketing organization in order to put more creativity into the development process. The new situation with more individualized medical treatment will undoubtedly make room for new players, especially among biotech and diagnostic companies.

One important part of personalized medicine is the pharmacodiagnostic tests that will be developed in the future in parallel with the new drugs.^[32] These tests must possess high sensitivity and specificity with regards to their predictive performance and here there is still room for improvement. The performance characteristics of the pharmacodiagnostic tests will steadily be improved along with the increase in our knowledge about disease biology and the mechanisms of the drugs. In the years to come, we will see more and more drugs prescribed based on the use of a pharmacodiagnostic test. Within cancer medicine, which has been at the forefront with respect to personalized medicine, it is expected that very few drugs, if any, will be prescribed without such a test in 10-15 years time. With such development in mind, other important challenges will arise, such as education of healthcare professionals and a healthcare infrastructure that is able to cope with personalized medicine.

It is often the perception that personalized medicine is something that will arrive in the future. Personalized medicine has arrived, and within certain disease areas it has already been implemented into medical practice, although still to a limited extent. In the years to come we will see an increased use of personalized medicine, and when this concept really takes off, it will have huge consequences for the way that drugs are being developed, marketed and prescribed. Personalized medicine will impose changes in both pharmaceutical companies and the healthcare system, but these changes will not prove to be wasted; they will, without doubt, improve future drug therapy to the benefit of the individual patient and society in general.

Reprint Address

Dako Denmark A/S, Clinical Affairs, Produktionsvej 42, DK-2600 Glostrup, Denmark. E-mail: jan.trost@dako.com

 

 

 

 

 

Personalized Medicine. 2008;5(1):55-63. © 2008  Future Medicine Ltd.

No writing assistance was utilized in the production of this manuscript.

Cite this: From Blockbuster Medicine to Personalized Medicine - Medscape - May 28, 2008.