Personalized Medicine in Oncology: Ethical Implications for the Delivery of Healthcare

Nathalie Egalite; Iris Jaitovich Groisman; Beatrice Godard


Personalized Medicine. 2014;11(7):659-668. 

In This Article

Key Ethical Issues at Stake in Developing Personalized Medicine in Oncology

Scientists and healthcare providers have extensively discussed the benefits and risks associated with the introduction of NGS techniques in clinical practice. As stated in a position paper by the European Society of Medical Oncology (ESMO), "A new era of personalized cancer medicine will touch every aspect of cancer care – from patient counseling, to cancer diagnosis, tumor classification, treatment and outcome ― that demands a new level of in-depth education and collaboration between researchers, cancer specialists, patients and other stakeholders".[18] This stated goal hides much complexity and there are considerable challenges, including ethical ones, that need to be addressed. Although many ethical issues have been identified and analyzed, whether in oncology or in genomics, some are still problematic because they are faced by healthcare providers and exacerbated by the use of newer sequencing techniques (Table 2).

Regarding consent, informed consent remains a significant challenge when integrating personalized medicine into oncology. A tumor's heterogeneity, its molecular evolution and its relationship with drug response and resistance pose a significant challenge to personalized cancer medicine.[18] Many patients with cancer have difficulties in understanding the purposes and complexities of pharmacogenomics testing. Given the vast amount of information that needs to be conveyed to patients, the development of innovative methods of communication in order to explain the benefits and risks and to assess patient understanding has been recommended.[19,20] As for healthcare providers, many describe the consent process as a challenge in practice, either due to its duration or its contents. The time needed to have a detailed discussion about each gene being tested may be prohibitive. In addition, communicating the same amount of detail for each gene would likely lead to information overload, in which there is too much information to absorb in a short time, potentially impeding patient understanding and decision-making.[19] Particularly challenging are the explanations of the meaning of variants of unknown significance and of incidental findings[19,21] and the potential psychosocial implications of germline and somatic genetic testing.[22]

The concept of consent is related to the right to privacy. Privacy is respected if an individual has an opportunity to exercise control over his or her personal information by consenting to or withholding consent for the collection, use and/or disclosure of information. Patients' genetic privacy and discrimination concerns have been documented in case studies, but ambiguity remains in the law and in professional guidelines regarding how to manage patients concerns. According to providers of personalized medicine, these issues are likely to become increasingly challenging for clinicians as genomic testing is further integrated into cancer care.[22] Disclosure of results from genomic testing that are clinically and analytically valid can help patients to take control of their lives. As a matter of fact, there is wide public interest in being informed of such results.

The return of results to the patient remains an essential element of clinical genomic testing, and the decision on which results to return to the patient has become a focus in the implementation of NGS. However, before returning results, one of the greatest obstacles in bringing NGS to the clinical setting is, as reported by Machini et al., the difficulty in analyzing and interpreting huge amounts of data.[21] Such an amount of data requires having access to highly identifiable, integrated datasets of genomic and health information, which brings into question patients' concerns about their genetic privacy. The ESMO advocates for broad patient consent, allowing for additional studies to be undertaken without the need for ongoing reconsent,[18] but support for this type of consent is far from unanimous. The European Commission also recommends large-scale data disclosure from all players in the personalized cancer medicine field (including pharmaceutical companies) in order to create an 'open-source' molecular medicine global knowledge network and drive further innovation in this field. In spite of the promise of what open-source data may bring, every precaution should be taken in order to minimize the risks to patients. As for interpreting results, appropriately trained professionals who are skilled at interpreting and translating the data to patients are required, and many healthcare providers express difficulties in interpreting results.[21,22] In 2013, the American College of Medical Genetics (ACMG) published a practice statement listing the results that should always be reported to the ordering physician because of their significant impact on the health of patients and relatives.[15] As noted by several authors,[19,21] these recommendations have the potential to magnify the ethical and practical dilemmas associated with genomic medicine.[21] For instance, Ciardiello et al. have reported that there is not enough research on cancer patients' preferences and expectations concerning genomic testing.[18] Further research and continued discussions in order to develop ethical frameworks in addition to the enactment of counseling recommendations on disclosing information from genomic testing are required.

Finally, we cannot consider personalized medicine without addressing the costs associated with it. While some stakeholders see personalized medicine as a solution to the 'one-size-fits-all' approach that creates inefficiencies in the healthcare system caused by pouring scarce resources into nonresponders and putting them at risk for toxic therapy, others believe that population-wide access to cancer genomic testing and therapeutics is not presently attainable.[22,23] Access to healthcare is highly differentiated by patient socioeconomic status, insurance providers and site of cancer care, all of which raise ethical concerns regarding fair and just distribution of healthcare resources. Respondents to a study conducted by McGowan et al. described distribution challenges in terms of both patients' access to genomic tests and targeted therapeutics and clinicians' access to genomic resources.[22] Even when clinicians and patients have access to genomic resources, it may be cost prohibitive if patients' genotypes indicate expensive targeted therapeutics that fall outside the standard of care for their type of cancer or if they are not covered by insurance. These situations contribute to the emergence of new health disparities based on ability to pay (rich people paying for treatment out of pocket and poor people being unable to afford it) or whereby some cancers eventually become orphan diseases that affect an underserved patient population.[22] Some institutions offer NGS for research as an alternative solution when there are insurance coverage issues. As noted by Machini et al., such use of research as a clinical service might create confusion for patients due to the differences between clinical and research testing.[21] In addition, it is conceivable that such a service is more frequently offered to vulnerable individuals, making them susceptible to further risks resulting from sharing their genomic information.

These specific concerns are not unique to oncology, or even genomics. However, as mentioned by other authors,[20–22,24] the rapid emergence of personalized medicine approaches in clinical oncology provides a crucial lens for identifying and managing the potential pitfalls that emerge as healthcare paradigms shift in these directions. It is worth noting that the ACMG issued a statement in 2012 that, while listing criteria that should be satisfied when offering genomic testing to patients, provided no specific guidelines with regards to the implementation process.[25]