Genetic testing is becoming more frequent and the results more complex. Not infrequently, genetic testing conducted for one purpose reveals information about other features of the genome that may be of clinical significance. These unintended findings have been referred to as "incidental" or "secondary" findings. In 2013, the American College of Medical Genetics ("ACMG") recommended that clinical laboratories inform people if their genetic analyses indicate that they have certain secondary mutations. These mutations were selected because they probably cause a serious disease, which is treatable, and may go undetected. The ACMG's recommendations galvanized critical responses by the genetics and ethics community. One of the most important open questions concerns the scope of negligence liability for clinical laboratories if they failed to provide any of these SFs to patients who never requested them. To answer this question, this article argues that while there might be an ethical or professional obligation to share knowledge about these specific genetic mutations, laboratories should not be subject to tort liability for failure to share secondary findings directly with patients.
INTRODUCTION II. MUCH DEPENDS ON HOW THE COMPLAINT IS FRAMED III. THE AMERICAN COLLEGE OF MEDICAL GENETICS REPORT IV. APPLYING THE INELEGANT BUT FLEXIBLE LAW OF NEGLIGENCE a. The History and Expansion of A Duty To Warn in Negligence b. Crucially, There is no "Special Relationship" Between the Lab and Patient V. THE DEVELOPMENT OF THE ROWLAND FACTORS VI. APPLYING THE ROWLAND FACTORS TO THE PROBLEM AT HAND a. The First Factor: The Foreseeability of Harm i. What is Penetrance and How Does it Impact the Foreseeability Analysis? ii. What is Expression and How Does it Impact the Foreseeability Analysis? iii. What is Analytic Validity and How will it Impact the Foreseeability Analysis? iv. Summary of the Foreseeability Factor Analysis b. The Second Factor: The Plaintiff Must Have Suffered A Certain, Concrete Harm c. The Third Factor: There Must be A Close Connection Between Defendant's Conduct and the Injury Suffered d. The Fourth Factor: The Moral Blame Attached to Defendant's Conduct e. The Fifth Factor: The Policy of Preventing Future Harm f. The Sixth Factor: The Potential Resource Burden on Labs g. The Seventh Factor: The Lack of Insurance to Spread the Cost of Imposing a Duty VII. TO WHOM WOULD THE DUTY BE OWED? VIII. PATIENT CONFIDENTIALITY SHOULD NOT BE BREACHED WHEN THERE IS NO RISK OF IMMINENT PHYSICAL VIOLENCE IX. CONCLUSION I. INTRODUCTION
Gone are the days of treating patients based upon their symptoms alone. Instead, in a growing number of contexts, physicians prescribe treatments based on an individual's unique genetic information, ("genotype."). (1) This is the essence of precision medicine: a bold new frontier of innovation in health care where treatments are no longer "one size fits all." It is already standard-of-care for oncologists to order a genetic test to determine which chemotherapy to use to treat small-cell lung cancer. (2) If the tumor possesses epidermal growth factor receptor ("EGFR") genetic mutations the cancer is more likely to respond to a chemotherapy drug called Tarceva[R]. (3) It is also common to perform a genetic test for human leukocyte antigen before prescribing abacavir, an HIV drug, as certain mutations predict adverse reactions including death. (4) The potential to improve treatments using genetic information is enormous. Looking for the next genetic discovery to more precisely treat cancers and other life-threatening diseases has resulted in a great deal of clinical research. (5)
Up until recently, genetic tests were mostly used to confirm the diagnosis of disease rather than to guide personalized treatments. Traditional genetic tests looked for the presence or absence of mutational "hotspots." (6) Often doctors needed to know what they were looking for, and roughly where it was in the genome, and the test would tell you whether the disease-causing mutation was present. This is how genetic tests for Sickle Cell, Cystic Fibrosis, and many other common genetic diseases have worked for decades. (7)
This all changed with the advent of next-generation genetic sequencing. One no longer needs to have a needle in mind. Tests can now search the entire haystack to see if there is anything of interest anywhere. While much is lost in resolution and validity, much is gained in scope. So-called "next generation whole genome sequencing," or simply "WGS" can decode three billion base pairs or the complete DNA sequence of an organism's genome with one tissue or blood sample. (8) Sequencing a genome used to be done manually, but the process has become more automated and therefore much faster, thus being labeled the "next generation." (9) In humans, this means that the chromosomal, and mitochondrial DNA, may be rapidly sequenced. (10) When WGS first hit the scene, the process was prohibitively expensive so it was only used in well-funded laboratories as demonstration projects. But that is no longer the case. The price has dropped to about $1500 per sample, so WGS is more commonly used in clinical research and treatment. (11) But it is not yet standard-of-care because insurance usually does not cover it. (12) WGS takes longer than other methods and yields a complex range of tricky-to-interpret results. Such results vary from those that predict risk for simple Mendelian diseases--caused by single genes--to those that implicate common, risk alleles with small effect sizes for multi-gene traits (caused by multiple genes). (13)
Rather than employing expensive WGS, more clinicians use whole-exome sequencing as a cheaper and faster alternative. (14) Whole exome sequencing ("WES") sequences only exons: the portions of a person's genome that code for proteins or peptides. (15) The exons represent only two percent of the genome but account for roughly eighty-five percent of the mutations that increase the risk of disease. (16) Clinical use of WES is rising in two contexts: cancer research (17) and assisting patients on diagnostic odysseys. (18) If the laboratory finds something interesting through WGS or WES, laboratories can, and should, confirm whether the result is a false positive or false negative through a more validated, targeted genetic test. (19)
These next-generation genetic analyses present novel questions for ordering physicians and laboratories. (20) Because physicians do not need to know what to look for before using WGS or WES, physicians can cast a wide net. Before, if a child presented symptoms of cystic fibrosis (CF), the physician ordered a specific genetic test to confirm whether the CF mutation was present. The laboratory took a blood sample, ran the cystic-fibrosis test, and confirmed the results. (21) But now--with one blood sample--the clinical laboratory can scan a genome or exome looking for thousands of other mutations. (22) Maybe the symptoms are caused by something the physician never considered? Maybe there is an interesting mutation, but its presence is too rare to say it's disease-causing?
WGS or WES invite a huge data dump. This data dump can come in various forms, depending on the way the lab reports its data. It can include preliminary information on many diseases for which the patient shows no symptoms and which were not related to the reason the physician ordered the test. (23) Instead of hearing that a child does or does not have cystic fibrosis, a parent might also learn that the child is at risk for developing early-onset Alzheimer's or colon cancer. These additional mutations are called "secondary findings," because they are discovered incidentally to the primary target of the test. (24) While secondary findings are possible with all types of genetic
testing, they are significantly more likely with WES and WGS due to the volume of data analyzed. (25)
This article addresses whether the tort law of negligence imposes any obligation on the laboratory to disclose genetic-risk information that is secondary to the primary reason for ordering the test, a "secondary finding" or (SF). (26) Whether there is a legal duty to warn is a threshold question for the judge to apply to similar future cases. Thus, before answering case-specific questions about whether there is a breach of the standard of care or whether the breach actually and proximately caused the patient's injury, there must be a duty. This article will proceed in two parts. First, it introduces the topic and the professional obligations imposed by the American College of Medical Genetics ("ACMG"). Second, it discusses the common-law factors courts routinely use when deciding whether to impose an affirmative duty to warn. The article argues against imposing this kind of duty on clinical laboratories.
MUCH DEPENDS ON HOW THE COMPLAINT IS FRAMED
Whether a judge will find that the laboratory has a particular duty will depend a great deal on how the patient frames his or her complaint. What sort of duty is being alleged? Is the plaintiff arguing that the defendant acted carelessly, or that the defendant did not act at all? These distinctions matter a great deal and make it impossible to answer the question of liability in the abstract. The outcome also depends on whether the plaintiff has brought a wrongful-death claim or a "lost chance" claim, for the loss of a better clinical outcome, because a medical diagnosis was delayed. Courts treat each type of negligence differently depending on the alleged duty and injury. Because this article seeks to provide practical, concrete guidance to judges, it will address a particular question: whether a laboratory should have disclosed a mutation directly to the patient. While some fact-specific questions arise in this context, a proper analysis of duty should proceed more in-abstractly, without wading too deep in the particulars of any case.
This article envisions one a very...