questions in genomic medicine…

In Genes, Genomic Medicine on Wednesday, 31 October 2012 at 15:12

Whole Genomes, Small Children, Big Questions

Jeantine E Lunshof

Personalized Medicine. 2012;9(7):667-669. © 2012  Future Medicine Ltd.


Whole-genome sequencing (WGS) is entering clinical practice. Expectations are high: a better understanding of disease, more accurate diagnosis and targeted therapies are hoped for; however, while some answers are being found, many new questions are arising. At the current stage, the dimensions of the -omics data sets in particular are overwhelming and physician–researchers are just starting to explore the ways in which they can be used. Apart from the bioinformatics challenges of genome data management there are the questions of meaningful interpretation of the emerging knowledge for health information users, both patients and professionals. At the individual and the societal level, questions are being asked about the extent and the usefulness of the newly available knowledge, for example: do whole-genome or -exome sequences tell us more than we want to know about ourselves, and doctors more than they need to know to treat their patients? In addition, do whole genomes tell parents more than they should know about their children? Since the early days of cytogenetic diagnosis, similar questions have been raised about the extent of the information yielded by testing, the communication of findings, and the ethical implications of testing.[1] Today, the availability of WGS and whole-exome sequencing (WES) for clinical diagnostics puts these old questions in a new context. Most recently, the feasibility of WGS was described for the application of noninvasive prenatal diagnosis.[2]

Diagnostics of Last Resort

WES and WGS are not routine diagnostic tools yet. However, while the introduction into the clinic of methods that are promising in the research setting is notoriously slow and usually accompanied by lengthy deliberations, medical emergencies can sometimes accelerate the process: a therapy of last resort that saves the life of an individual patient, may lead to a breakthrough in general clinical application. One striking example is the application of WES, coupled with meticulous bioinformatics analysis, that in a grand multidisciplinary effort saved the life of a little boy.[3]

The child was first seen at the age of 15 months presenting with an unusual, severe and progressive Crohn’s disease-like illness. A definitive diagnosis could not be made, restricting effective clinical management. The boy underwent more than 100 surgical procedures over 3 years. A hematopoietic progenitor cell transplant, that might be a therapy of last resort, was considered too risky without a clear diagnosis of the underlying disorder. Established genetic tests for known congenital immune disorders presenting with inflammatory bowel disease-like illness were not informative. Ultimately, with institutional review board approval, research-stage, nonvalidated, WES was performed to make a clinical diagnosis and guide the treatment decision. The identification of the causative mutation enabled treatment of the underlying immune disorder and of the child’s bowel disease as well.[3]

In a commentary article, alongside with the report describing the clinical and methodological details of the case, the Wisconsin group presents their ethical deliberations.[4] The authors raise some crucial questions – notwithstanding the impressive outcome and the indisputable huge benefit to the young patient – and suggest further discussion. This editorial aims to follow up on their suggestions.

The first issue raised is about the elusive boundary between research and clinical care. As this case shows, therapies of last resort can be considered a type of n = 1 studies and this is relevant from a regulatory point of view. n = 1 studies connect the bedside and the bench in a very specific and direct way.[5,6]

The other major issue, identified as an intrinsic concern by the Wisconsin group, is concerned with the data flood that results from WGS and WES and that greatly exceeds actual clinical needs. Inevitably, much of this information is ‘off target’ and may predict risks for late-onset disorders, which is considered particularly problematic when the genome analysis is performed in children.

Research & Clinical Care

Clinical application of research-stage procedures can save lives, as the exemplary case of the 15-month-old boy shows. In this case, the institutional review board-approved the use of nonvalidated experimental methods precisely because the primary purpose was to obtain a diagnosis for a patient; had the aim been gaining generalizable knowledge this would have turned it formally into research.[4] This reasoning, however, is based on a questionable and probably outdated distinction between research and clinical care that takes systematic recording of outcomes as the decisive criterion for research.[5,6] Moreover, can there be any instance of a diagnostic or therapeutic procedure – experimental or routine – that does not record results or yield generalizable knowledge? Also, clinical care and n = 1 studies are essentially connected. One could say that in ‘personalized’ medicine – and good medicine is always personalized – every medical intervention in an individual is a type of n = 1 study. The results from regular clinical care, however, will often hide in the patient’s medical record and potential new knowledge of general interest will remain with the attending physician unless so striking that it is shared through publication.

Revising the distinction between research and clinical care is urgent, now that WGS begins to enter the clinic.[7–9] For the development of personalized genomic medicine, a modified n = 1 study design is needed that enables resolving the old dichotomy and that also takes the normative (i.e., ethical and legal) aspects into account, as well as the translation in terms of health technology assessment.[5,6,10]

Small Children, Whole Genomes: Too Much Information?

The other key issue concerns the amount and extent of the data generated through WGS or WES: do we get too much information? And, what makes information too much?

One very simple answer to the latter question is that how much we need to know is determined by the problem that the sequencing was intended to solve. That problem seems to be clear at least in a clinical context, as the question there is: what is causing this disease in this patient, what is the actual diagnosis? This was actually the question the medical team in Wisconsin asked, desperate to know the cause of ther patient’s symptoms and to be able to make a diagnosis. Indeed, for their acute clinical purpose, the information about the XIAP deficiency was sufficient to balance potential harms and benefits of a risky intervention and to make a rational treatment decision. At this point, any further information from the patient’s exome could be regarded as redundant – ‘off target’ in the traditional clinical context.

In the context of systems biology-based medicine it is much harder to declare data as redundant or ‘off target’: the disease state of an individual results from particular network perturbations, and many known and unknown interactions between network components can play a role.[11] The interpretation is the problem, rather than the amount of data. Thus, information that may seem redundant today may later be found to be essential for a correct understanding of a disorder. For example, WGS data may contain clues about modifier genes (or non-coding RNAs)that seem of minor importance, but whose products do affect the disease phenotype.

Yet, the big question of how to handle the comprehensive genome information is still in the room, in the clinical context as well as in the research setting: what to communicate to patients, or to research participants? When children are concerned, the communication includes the parents and, depending on their age, the children themselves. From the ethical, legal and psychological point of view, WGS/WES in children is particularly interesting because areas of long-standing hot debate converge: children as research participants, genetic testing in children for clinical purposes, returning results to patients/study participants, and the major issue of generating predictive information.[12] The common denominator in these debates is the issue of data access and sharing information – in particular the sharing with the true data owners, namely, the participants, patients and with parents as caretakers. However, parents’ entitlement to access all of their child’s data is not undisputed, as it could be seen as a breach of the child’s privacy.[12] The inherent lack of informed consent – depending on age and maturity – for clinical interventions and research participation presents a major and unavoidable dilemma for which no generally valid solution exists.

On the Horizon: New Methods, Old Questions

In their commentary, the group from Wisconsin present the outline of a plan for returning results to parents that shall allow the parents to decide what type of ‘off-target’ information they want to receive.[4] Other researchers recently found in a focus group study among parents of children with serious conditions that these parents wished to receive all types of research results, not only those that would be actionable.[13] The early application of WGS/WES for solving clinical puzzles has shown great therapeutic benefit for a number of children with serious conditions.[7] Similar to the case of the young child described above, the stories about these children were not only published in scholarly journals but also fully disclosed in the general media.[101,102]

Beyond the clinical applications, WGS greatly advances more fundamental research, as demonstrated by the Pediatric Cancer Genome Project.[14]

New applications of WGS are on the horizon, even if they are not around the just corner yet. Noninvasive prenatal diagnosis through WGS is technically feasible.[2] The genome of a fetus was successfully predicted, based on inferences made from the genome of the mother, the father and the estimated fetal proportion of the cell-free DNA in the mother’s circulation. However, this was no more than a proof-of-principle at the moment, and the major technical difficulty in the future will be the data interpretation, as the authors conclude.

On the nontechnical side, however, the difficulties could be even bigger, as a decades-old debate on some deep and unsolvable moral issues in prenatal diagnosis will be rejuvenated.

Retrieved from: http://www.medscape.com/viewarticle/771374?src=nl_topic


  1. Hsu LY, Durbin EC, Kerenyi T, Hirschhorn K. Results and pitfalls in prenatal genetic diagnosis. J. Med. Genet.10,112–119 (1973).
  2. Kitzman JO, Snyder MW, Ventura M et al. Noninvasive whole-genome sequencing of a human fetus. Sci. Transl Med.4(137),137ra76 (2012).
  3. Worthey EA, Mayer AN, Syverson GD et al. Making a definitive diagnosis: successful clinical application of whole exome sequencing in a child with intractable inflammatory bowel disease. Genet. Med.13(3),255–262 (2011).
  4. Mayer AN, Dimmock DP, Arca MJ et al. A timely arrival for genomic medicine. Genet. Med.13(3),195–196 (2011).
  5. Tsapas A, Matthews DR. Using n-of-1 trials in evidence-based clinical practice. JAMA301(10),1022–1023 (2009).
  6. Chalmers I. Regulation of therapeutic research is compromising the interests of patients. Int. J. Pharm. Med.21(6),395–404 (2007).
  7. Bainbridge MN, Wiszniewski W, Murdock DR et al. Whole-genome sequencing for optimized patient management. Sci. Transl Med.3(87),87re3 (2011).
  8. Dixon-Salazar TJ, Silhavy JL, Udpa N et al. Exome sequencing can improve diagnosis and alter patient management. Sci. Transl Med.4(138),138ra78 (2012).
  9. Check Hayden E. Sequencing set to alter clinical landscape. Nature482,288 (2012).
  10. Becla L, Lunshof JE, Gurwitz D et al. Health technology assessment in the era of personalized health care. Int. J. Technol. Assess. Health Care27(2),118–126 (2011).
  11. Vidal M, Cusick ME, Barabási AL. Interactome networks and human disease. Cell144,986–998 (2011).
  12. Mand C, Gillam L, Delatycki MB, Duncan RE. Predictive genetic testing in monors for late onset conditions: a chronological and analytical review of the ethical arguments. J. Med. Ethics doi:10.1136/medethics-2011-100055 (2012) (Epub ahead of print).
  13. Harris ED, Ziniel SI, Amatruda JG et al. The beliefs, motivations, and expectations of parents who have enrolled their children in a genetic biorepository. Genet. Med.14(3),330–337 (2012).
  14. Downing JR, Wilson RK, Zhang J et al. The pediatric cancer genome project. Nat. Genet.44(6),619–622 (2012).

    101. One In A Billion: A boy’s life, a medical mystery. http://www.jsonline.com/features/health/111224104.html
    102. Genome Maps Solve Medical Mystery For Calif. Twins. http://www.npr.org/blogs/health/2011/06/18/137204964/genome-maps-solve-medical-mystery-for-calif-twins


Thanks for your comments!

Please log in using one of these methods to post your comment:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: