Ethical, Legal, and Social Issues Associated with Pharmacogenomics
On completion of the Human Genome Project in 2003, the field of pharmacogenetics and/or pharmacogenomics has rapidly developed with the intent of clinical application on individualizing therapy to maximize efficacy and/or minimize toxicity. There are numerous challenges and barriers that impact the implementation of pharmacogenomics into clinical practice. These include ethical, legal, and social issues that highlight risk, benefit, and public and professional acceptance.1 Attempts to establish a framework of accelerating integration of appropriate pharmacogenomic and human genome discoveries into clinical practice have been reported.2,3 Haddow and Palomaki4 first reported the ACCE model for evaluating data on genetic tests. ACCE that takes its name from the four main criteria for evaluating a genetic test—analytic validity, clinical validity, clinical utility, and associated ethical, legal, and social implications—is a model process that includes collecting, evaluating, interpreting, and reporting data about DNA (and related) testing for disorders with a genetic component in a format that allows policy makers to have access to up-to-date and reliable information for decision making. One of the major components for evaluation includes the associated ethical, legal, and social implications. The authors identified challenges and barriers to genetic testing including genetic discrimination, stigmatization, privacy/confidentiality, and personal/family social issues. Additional challenges highlighted legal issues regarding consent, ownership of data and/or samples, patents, licensing, disclosure obligations, and reporting requirements. Although the ethical, legal, and social issues revolve around establishing safeguards in the context of genetic testing, these components are applicable to pharmacogenomic testing.5–7 Several examples in this chapter will highlight these issues in relation to genetic testing to aid in the understanding of specific concepts and application. The purpose of this chapter is to provide a framework of the ethical, legal, and social issues surrounding the rapidly evolving field of pharmacogenomics.
Individuals who obtain pharmacogenomic and/or genetic testing may be exposed to genetic discrimination. Genetic discrimination refers to when an individual, with a known genetic disorder or genetic polymorphism, is treated differently by his or her employer or insurance company.8 Knowledge of genetic and/or pharmacogenomic information may be used to deny, limit, or cancel health insurance. There is suggestion that health insurers should have limited access to such information. Insurers would be able to use pharmacogenomic information for drug formulary management, but should be prohibited from using the same information in determining copayments or premiums, or negotiating contracts.9 Additional concerns regarding genetic discrimination include employers using such information to only employ or retain individuals who do not have the genetic disorder or genetic polymorphism, limiting access to social services, and in the delivery of health care.10 Consequently, genetic discrimination potentially generates social, health, and economic burdens for society due to decreasing opportunities of genetically predisposed individuals in a range of scenarios.
Several reports of genetic discrimination have been documented.10–12 In 2001, the Equal Employment Opportunity Commission (EEOC) filed suit against the Burlington Northern State Santa Fe (BNSF) Railroad.13 Without the knowledge and consent of its employees, BNSF tested 36 employees for a genetic condition associated with carpal tunnel syndrome as part of a comprehensive diagnostic exam. BNSF employees were examined by company-paid physicians and were not informed that genetic testing was being performed. Those employees who refused testing were also threatened with possible job termination.14 BNSF defended such testing as a means of determining if repetitive stress injuries were work-related. Under the Americans with Disabilities Act, the EEOC argued that such testing as a basis for employment was unlawful and a cause for illegal discrimination.13 The suit was eventually settled, with BNSF admitting that testing employees for a genetic condition was performed and that such testing was no longer being conducted.13 In a case history study , genetic discrimination involved insurance companies in areas of applications or coverage changes for health, life, disability, mortgage, and auto insurance. Seven cases involved employment in areas relating to hiring, termination, promotion, and transfer.10 Due to the fear of discrimination, the authors also noted that respondents either withheld mentioning such information to health care providers, insurers, and employers or provided incomplete or dishonest information on insurance application forms.10
Stigmatization is defined as “a social process that begins with distinguishing and labelling some feature of a person such as occupation, disease, or skin color.”15 An individual may experience stigmatization from family, friends, and coworkers on knowledge that a specific disease will not respond to therapy or if one is identified as a “poor metabolizer” of a specific medication. This may lead the individual to feeling lonely, isolated, hopeless, and depressed.1 A classic example was during the 1980s, with human immunodeficiency virus/acquired immunodeficiency syndrome (HIV/AIDS) described as a sexually transmitted infection diagnosed in patients identified as homosexuals or intravenous drug users. Patients, as well as parents, children, and caregivers for HIV-positive patients, were susceptible to stigmatization. Negative connotations by the public included misperceptions that people associated with such behaviors were immoral, predatory, and dangerous.15,16 Consequently, pharmacogenomic testing may also reveal similar negative connotations and public perceptions, as evidenced with traditional genetic testing.17
Privacy and Confidentiality
Patients, health care professionals, and policy makers are concerned about security and loss of privacy. The capacity for such information to be adequately stored and rapidly disseminated and the relative ease of accessibility all pose a risk for a loss of privacy.18,19 Examples of methods to ensure privacy include use of de-identified subject data, use of password-protected and/or encrypted files, and limiting access. Additionally, federal legislation such as the Health Insurance Portability and Accountability Act (HIPAA) and private sector self-regulation exists. The HIPAA Privacy Rule provides federal protections for personal health information held by covered entities (e.g., health care providers, hospitals, pharmacies, insurance companies) and provides patients rights regarding personal health information. However, uncertainty remains in the capacity and ability of current systems to ensure privacy, security, and confidentiality. For example, health care practitioners who use telemedicine technologies to disseminate pharmacogenomic information may be faced with contrasting federal and state law privacy standards. In one scenario, if a provider orders a genetic test from one state and the results of the test are made available electronically to another out-of-state provider, clarification is warranted as to which state privacy laws are followed.20,21 Additionally, HIPAA only applies to employer-based and commercially based issued health insurance and does not apply to individuals who seek private health insurance in the individual market.14
Balancing Harms and Benefits
Foreseeable harms should not outweigh anticipated benefits for participants in pharmacogenomic research. Minimizing harm implies a duty to ensure that research subjects are not subjected to unnecessary risks of harm and that their participation in research is essential to achieving the scientific outcomes necessary for future treatment for the subjects themselves, other individuals, or society as a whole.22 Pharmacogenomic research includes the potential for physical risk and negative psychological impact. The physical risks are related to using a drug in which the drug itself has the potential to cause harms such as intolerable side effects. The psychological impact may include being excluded from enrolling in a clinical study if the individual does not have the genetic polymorphism or genotype, which may lead to despair if other treatments have failed or are not available. The implementation of a pharmacogenomic test to determine if an individual may respond appropriately to a given drug before beginning treatment may also result in a dilemma for health care providers and patients. For example, what is appropriate therapy for an individual who is not a candidate to receive a specific therapy based on a pharmacogenomic test result and for whom there are no other known treatments? Is it morally permissible to offer the treatment anyway with the knowledge of there being little chance for response or possibly even inducing harm? This is very unlikely to occur as most health care providers would find this option to be in conflict with their inherent duty to do no harm. However, this could theoretically create fear among consumers, which may result in patients unwilling to take the associated test for fear of being abandoned in their treatment.
With genetic testing, the DNA collected not just will include the known genetic polymorphisms obtained from participating in a research study but may also include information about the individual’s entire genome. As a result, these DNA samples could provide invaluable information about not only the participant but also possibly his or her extended family, which could be entirely unrelated to the original study. Typically, such additional information would not be revealed to the original participants but rather be used in an anonymous fashion. However, this possibility should be discussed up front as part of obtaining informed consent. Although future use of such data may not be known at the time of consent, specific language should be described in the informed consent form. Consideration of using narrow (specific limited use) or broad (current and future use) terminology may be appropriate to help participants understand how their information may be used.1,23
In pharmacogenomic research one of the fundamental requirements to enter a study is to test individuals for a particular genotype of interest. Study participation in such research will reveal personal health information, which would not be known in other clinical studies. This particular knowledge does not necessarily compromise privacy and confidentiality but does require it be addressed in the consent process as well as dissemination of results that may or may not be revealed to the participants.23
Several issues related to the informed consent process for pharmacogenomic testing warrant discussion. First, there appears to be no standard at the national or international level concerning how to consent individuals for research that involves genetic testing. Although some have attempted to address these issues,23 there currently exist two informed consent standards. The professional standard evaluates “what a reasonably prudent physician with the same background, training, and experience, practicing in the same community, would have disclosed to a patient in the same or similar situation.”1 In contrast, the patient standard attempts to address “what a reasonable patient in the same or similar situation would need or want to know to make an informed decision regarding a medical intervention or treatment.”1 Second, there is an inconsistency as to the placement of a consent form for genetic testing that was completed by the patient as part of a clinical study. Some advocate placement of the consent form in the patient’s medical record, while others suggest that such information be kept separate from the patient’s medical record. Our own experience has been that some institutional review boards (IRBs) do not recommend consent forms (for studies where genetic testing is to be done) to be placed in the medical record. Rather, such information should be kept in a file by both the investigator and the IRB. This attempts to ensure privacy by avoiding inadvertent disclosure of such information to anyone reviewing the patient’s medical chart.
Ownership of Data and/or Samples
Pharmacogenomic testing requires collection of DNA. A biological specimen (e.g., blood, saliva, or tissue) is obtained from the patient. Due to genomic scale-up (e.g., genomic sequencing of patients), such specimens are stored in repositories known as biobanks. At times, biological specimens are aliquoted whereby material is used for pharmacogenomic testing and any remaining specimen is archived. The archiving of biological specimens in a biobank facilitates use of the same sample for future exploratory research with no defined time limit for use. These biological samples and their respective data are invaluable, not only to academic or medical geneticists but also to pharmaceutical companies and the biotechnology industry.
Claims of ownership of the biological specimen involve the subject-donor, the institution whereby research was conducted, and the individual investigator. Due to a lack of clearly defined regulations and multiple stakeholder involvement, the issue of ownership of biological specimens, such as those reserved for pharmacogenomic testing, remains contentious. From a subject-donor perspective, his or her biological specimen is considered a property right and denying such a right over biological material is unfair. The contrasting perspective “defends an absolute non-patrimonial view, denying the possibility of the existence of a property right.”24
In the Moore v Regents of the University of California case,25 the plaintiff sued the defendant claiming deprivation of property interest, lack of adequate informed consent, and breach of fiduciary duty. Mr Moore’s biological specimens were collected by the defendant’s physician for a research project. Unknown to the patient, the results of the research led to a patent application by the defendant. The state court ruled that Mr Moore did not retain the right of ownership of his biological specimen that was used to develop a new therapy/product. In addition, the court also noted that disclosure to the patient of additional research or economic interests was necessary.25,26
Research is underway to examine the issues related to the ownership of data and/or samples.27,28 Legal writers have suggested theories related to trusteeship, benefit sharing, commodity, and “waste” models.27 Current research is focused on understanding the different perspectives from the multiple stakeholders and to develop policies. Input should be acquired from all interested stakeholders including the general public, researchers, as well as the private sector. It also seems reasonable given the global nature of research and development that these regulations be formulated at an international level.29
Approximately 20% of all human genes are under US patents.30 In 2005, this equated to 4,382 of the 23,688 genes described in the National Center for Biotechnology Information database.30 US patent laws are intended to stimulate innovation and to protect the discoveries of inventors. For a period of 20 years, a patent provides the inventor the exclusive right to manufacture, use, and sell the discovery. To qualify for a patent, the invention needs to be useful, novel, and not obvious.31,32 In the context of genomic medicine and pharmacogenomics, gene patents remain a contentious issue. The most common argument against gene patents is that since genes are naturally occurring biological processes, they exist to be discovered and not invented.32 In contrast, such an argument is considered unpersuasive since isolating DNA does not occur naturally and that the “patent system has long recognized useful applications of discoveries as inventions.”32 In Diamond v Chakrabarty, the US Supreme Court ruled that genetically engineered bacteria were subject to patent protection as they did not occur in nature.33 Although the genes themselves are not patentable, patent protection is allowable under the auspices of DNA that is modified, purified, or isolated resulting in a form that is nonexistent in nature.31
In 2009, a lawsuit was filed by the American Civil Liberties Union (ACLU) and others against Myriad Genetics for their US patent on the human genes BRCA1 and BRCA2. BRCA1 and BRCA2 mutations are associated with an increased risk of breast and ovarian cancers. Myriad Genetics currently owns at least seven patents directed at BRCA1 and BRCA2.21,34