Genomic Testing: The Perils and the Pitfalls

Failure to Recommend Genetic Testing

In Downs v. Trias, 49 A. 3rd 180 (2012), a Connecticut woman won a $4 million verdict against her physician for his failure to warn her that her family's strong history of breast cancer also predisposed her to ovarian cancer. The plaintiff's mother, maternal grandmother and two maternal aunts had all died of breast cancer. In 1981, the plaintiff chose to undergo a bilateral mastectomy at the age of 22 in order to reduce her cancer risk. In 2005, she underwent a partial hysterectomy for painful fibroids not associated with cancer. She was not told she was at risk for ovarian cancer, and her ovaries were healthy at that time.

In its decision, the Connecticut Supreme Court wrote, “At the preoperative consultation, the defendant explained to the plaintiff that, although she had a significant family history of breast cancer, that history, unless supplemented by genetic testing, which the plaintiff had not undergone, did not point to an increased risk of ovarian cancer. The defendant further indicated that the plaintiff's ovaries were healthy, that there was no reason to remove them and that removal would result in unpleasant side effects[,] including early menopause and interference with sexual intercourse.”

Within a year of the cited discussion, the patient was diagnosed with advanced ovarian cancer, although the court found that she had not had it at the time of the partial hysterectomy. As the court said, “A physician has a professional duty to possess or obtain certain medical knowledge as well as an additional 'lay' duty to communicate a subset of that information to the patient. A physician who fails to apprise a patient of a certain fact may therefore, in appropriate circumstances be held liable for failing to know the fact in the first place (medical negligence) and for failing to convey the fact to the patient for his or her consideration in making medical treatment decisions (lack of informed consent).”

Another lawsuit in this vein was brought by a woman who claimed that her surgeon should have recommended genetic testing for BRCA ½ mutations before she had prophylactic surgery. Downey v. Dunnington, 895 N.E. 2d 271, (2008). The patient's mother and grandmother had both had breast cancer, although the surgeon was told incorrectly that the mother had had bilateral cancer. Nonetheless, genetic testing was discussed with the patient, who was on Medicaid. The doctor recommended the testing, but informed the patient that Medicaid would not pay for it. (There was some question as to whether other payment options were discussed with the patient.) The patient elected to have a double mastectomy instead of pursuing additional testing. When the surgical biopsies came back benign, she sued the doctor, claiming lack of informed consent. Eventually, the physician won, but the takeaway from a case such as this is that there is a need not only for a provider to recommend genetic testing, if appropriate, but also to discuss how it will be paid for and whether there are alternative sources from which a patient could seek payment if their own insurance will not cover the costs.

In Scholz V. Kaiser Foundation Hospital, RG12614636, Alameda Sup. Ct. (filed Jan. 30, 2012), a woman of Asian descent developed Stevens Johnson Syndrome, a rare and serious reaction to certain medications, after having been prescribed carbamazepine without the prescriber first recommending genetic testing, as the FDA-approved label indicated for patients of her ancestry. That case is still pending.

Incorrect Results of Genetic Testing

Even if testing is recommended, and the testing is done, it doesn't mean that the result will be the right one. The consequences of such mistakes can be even more grievous for providers than if testing wasn't recommended in the first place.

A particularly poignant example of this is the Washington State case, Wurth v. Laboratory of America, 359 P. 3d 841. (2015). There, prospective father Brock Rhea had a family history of birth defects, specifically chromosomal “translocation.” Rhea's cousin Jackie had “profound disabilities,” and had tested positive for chromosomal translocation. It was recommended that other family members be tested, including Rhea, who had “asymptomatic, balanced translocations.” Rhea's son was too young to be tested, but because Rhea and his wife wanted more children, they consulted with a genetic counselor. The counselor advised them that there was a 50% chance that any child of theirs would have a range of disabilities due to the defect. They were also counseled, however, that many of the pregnancies with the defect miscarried. The couple carefully followed the recommendations, conceived again, and brought Rhea's genetic report to each pregnancy appointment.

Rhea's wife suffered six miscarriages in the next five years. In November 2007, the couple conceived again, and the pregnancy progressed through the first trimester without incident. The obstetrician scheduled a chorionic villus sampling (CVS) procedure to obtain fetal tissue for testing.

Then a series of unfortunate events unfolded. The testing needed to be done within a short window of time, and the only opportunity available was on New Year's Eve, a day on which no genetic counselor was working at the hospital. The couple decided to have the perinatologist/obstetrician who was available go ahead with the procedure. Normally, the genetic counselor would have handled the lab forms, but in the absence of any such person, a medical assistant did it. The physician testified that he gave Rhea's genetic report to the assistant to send with the form, but it is clear that this did not happen. A cytogenetic technician-in-training was assigned to perform the testing, but did not perform the sensitive testing or determine that more testing was warranted. The policies of the laboratory, LabCorp, require supervision of technicians-in-training, but no one else looked at the case. Because of all (or a combination of some sub-set) of these circumstances, the needed test was not conducted.

The genetics counselor back at the hospital in Washington reported to the couple that the test results were “normal.” She did not report the results back to the obstetrician who did the procedure. Therefore, the obstetrician reported back to the primary care physician that the “fetal chromosome results were normal, with no evidence of … a translocation.”

Failing to get the warning that might have altered their plans, the couple carried on with the pregnancy and the baby was born extremely disabled. Genetic testing confirmed that he had inherited the catastrophic condition that Rhea's cousin suffered from. The Washington Supreme Court upheld the $50 million verdict against LabCorp and the obstetrician.

In another case, Levy v. Legacy Health System, 12 Or. Lit. Arb. Repts 78, 2012 WL 1646980, 34-year-old Deborah Levy had CVS testing because of concerns about possible genetic disorders. The plaintiffs claimed that the doctor removed maternal tissue, not fetal tissue, and no one realized that this had happened. When the test returned negative results, the couple evidently believed them, as opposed to two ultrasounds that showed potential indications for Down's syndrome. The mother claimed that she would have terminated the pregnancy had she known the diagnosis. The jury returned a plaintiff's verdict of $3 million.

Testing That Does Not Consider All Variations

Even if genomic testing is ordered, there may still be problems with the findings.

In Williams v. Quest Diagnostics, C.A. No. 3:16-cv-00972-MBS (filed 2016), pending in the U.S. District Court in South Carolina, a motion to dismiss was recently argued. In that case, a child was diagnosed at one year of age with a SCN1A mutation that causes an epileptic syndrome called Dravet. The plaintiff claims that the laboratory's misclassification of the child's mutation as Variants of Uncertain Significance (VUS) was the cause of his death by seizure several months after the test results were reported, because the misclassification led his physicians down the wrong diagnostic path. The gist of the claim is that the child's health care providers continued to treat the toddler as if he had a mitochondrial disorder, giving him increasing doses of drugs that were contraindicated for the child's mutation. The mother obtained the report, but not until 2014. In 2015, the laboratory reclassified the mutation as disease-creating.

The Williams motion to dismiss, argued in January, is currently focused on whether the laboratory is a “health care provider” within the meaning of the medical malpractice statute. If it is, the lawsuit is barred by the three-year medical malpractice statute of limitations. The plaintiff argues that the South Carolina Supreme Court already declared, in Swanigan v. American National Red Cross, 438 S.E.2nd 251 (1993), that such entities are governed by the negligence statute of limitations, not that of medical malpractice. While this case may be determined on statute-of-limitations grounds, the caution for testing facilities remains. If laboratories are not denominated as health care providers, their liability and responsibilities may be greater if the jurisdiction's statute of limitations for bringing general negligence claims is different from that for medical malpractice claims. In any case, testing that does not take into consideration all possibilities puts the provider, and potentially the laboratory, at risk for liability.

Reporting Incidental Findings in Genome Sequencing

The American College of Medical Genetics and Genomics recommends that whenever clinical sequencing is performed, the laboratory should conduct a deliberate search for medically significant mutations in a minimum list of 56 genes and report them to the patients' clinician, who would be responsible for counseling the patient. American College of Medical Genetics and Genomics: Incidental Findings in Clinical Genomics: A Clarification. Genet Med. 2013;15:664-666. These recommendations have generated discussion on how incidental findings should be reported and how best to manage potential liability risks for both providers and laboratories. Clayton EW, Haga S, Kuszler P, et al: Managing Incidental Genomic Findings: Legal Obligations of Clinicians. Genet Med. 2013;15:624–629. A Presidential Commission for the Study of Bioethical Issues has also issued its own recommendations, in parts at odds with the American College recommendations. See Anticipate and communicate: ethical management of incidental and secondary findings in the clinical, research, and direct-to-consumer contexts. Presidential Commission for the Study of Bioethical Issues. 2013.

To date, we have not found any cases regarding a requirement that incidental findings in genetic testing be reported. But, as some commentators suggest, it is only a matter of time. Powlege, Tabitha, What Should be Done with Unsettling 'Incidental Findings' in Genetic Testing? Genetic Literacy Project, March 2015. As genomic testing becomes more sophisticated, and the price decreases, there will be more access to such testing and, with that, greater expectations.

Conclusion

As we have seen, genomic testing has the potential to be of value to patients and their care providers in terms of identifying disease and managing it, but it also creates new perils for both providers and the laboratories that analyze the tests. We can expect more litigation regarding negative consequences for patients to be on the horizon.

*****
Linda Crawford, a member of the Medical Malpractice Law & Strategy Board of Editors, teaches trial advocacy at Harvard Law School.

The era of personalized medicine, which relies in part on genomic testing, is growing rapidly. Some hospitals, including Massachusetts General Hospital, are already using such testing in providing care. Ray, Turna: Study Finds Docs Could Face Greater Malpractice Risk in Personalized RX Era. July 29, 2011, http://www.genomeweb.com. It has the potential to be of value to patients in terms of diagnosing conditions, as well as deciding on proper treatments.

However, there are pitfalls for providers and companies that offer genomic testing. They might include not offering genetic testing if it is indicated, not offering the proper testing, or reporting inaccurate or misleading results. Each of these has the potential to generate a lawsuit, and indeed each of them already has. As Gary Marchant and Rachel Lindor indicated in an article published a few years back, “History demonstrates that new medical technologies usually spur increases in medical malpractice litigation. In part, this is a reflection of the reality that the more a provider is capable of doing, the more he or she becomes expected to do, and thus the more something can go wrong and lead to a lawsuit.” Merchant, Gary and Lindor, Rachel: Personalized Medicine and Genetic Malpractice. Genet Med. 2013 Dec. 15(12):921-922.

Let's look at some of the issues that may arise when a patient is (or is not) offered genomic testing.

Failure to Recommend Genetic Testing

In Downs v. Trias, 49 A. 3rd 180 (2012), a Connecticut woman won a $4 million verdict against her physician for his failure to warn her that her family's strong history of breast cancer also predisposed her to ovarian cancer. The plaintiff's mother, maternal grandmother and two maternal aunts had all died of breast cancer. In 1981, the plaintiff chose to undergo a bilateral mastectomy at the age of 22 in order to reduce her cancer risk. In 2005, she underwent a partial hysterectomy for painful fibroids not associated with cancer. She was not told she was at risk for ovarian cancer, and her ovaries were healthy at that time.

In its decision, the Connecticut Supreme Court wrote, “At the preoperative consultation, the defendant explained to the plaintiff that, although she had a significant family history of breast cancer, that history, unless supplemented by genetic testing, which the plaintiff had not undergone, did not point to an increased risk of ovarian cancer. The defendant further indicated that the plaintiff's ovaries were healthy, that there was no reason to remove them and that removal would result in unpleasant side effects[,] including early menopause and interference with sexual intercourse.”

Within a year of the cited discussion, the patient was diagnosed with advanced ovarian cancer, although the court found that she had not had it at the time of the partial hysterectomy. As the court said, “A physician has a professional duty to possess or obtain certain medical knowledge as well as an additional 'lay' duty to communicate a subset of that information to the patient. A physician who fails to apprise a patient of a certain fact may therefore, in appropriate circumstances be held liable for failing to know the fact in the first place (medical negligence) and for failing to convey the fact to the patient for his or her consideration in making medical treatment decisions (lack of informed consent).”

Another lawsuit in this vein was brought by a woman who claimed that her surgeon should have recommended genetic testing for BRCA ½ mutations before she had prophylactic surgery. Downey v. Dunnington , 895 N.E. 2d 271, (2008). The patient's mother and grandmother had both had breast cancer, although the surgeon was told incorrectly that the mother had had bilateral cancer. Nonetheless, genetic testing was discussed with the patient, who was on Medicaid. The doctor recommended the testing, but informed the patient that Medicaid would not pay for it. (There was some question as to whether other payment options were discussed with the patient.) The patient elected to have a double mastectomy instead of pursuing additional testing. When the surgical biopsies came back benign, she sued the doctor, claiming lack of informed consent. Eventually, the physician won, but the takeaway from a case such as this is that there is a need not only for a provider to recommend genetic testing, if appropriate, but also to discuss how it will be paid for and whether there are alternative sources from which a patient could seek payment if their own insurance will not cover the costs.

In Scholz V. Kaiser Foundation Hospital, RG12614636, Alameda Sup. Ct. (filed Jan. 30, 2012), a woman of Asian descent developed Stevens Johnson Syndrome, a rare and serious reaction to certain medications, after having been prescribed carbamazepine without the prescriber first recommending genetic testing, as the FDA-approved label indicated for patients of her ancestry. That case is still pending.

Incorrect Results of Genetic Testing

Even if testing is recommended, and the testing is done, it doesn't mean that the result will be the right one. The consequences of such mistakes can be even more grievous for providers than if testing wasn't recommended in the first place.

A particularly poignant example of this is the Washington State case, Wurth v. Laboratory of America , 359 P. 3d 841. (2015). There, prospective father Brock Rhea had a family history of birth defects, specifically chromosomal “translocation.” Rhea's cousin Jackie had “profound disabilities,” and had tested positive for chromosomal translocation. It was recommended that other family members be tested, including Rhea, who had “asymptomatic, balanced translocations.” Rhea's son was too young to be tested, but because Rhea and his wife wanted more children, they consulted with a genetic counselor. The counselor advised them that there was a 50% chance that any child of theirs would have a range of disabilities due to the defect. They were also counseled, however, that many of the pregnancies with the defect miscarried. The couple carefully followed the recommendations, conceived again, and brought Rhea's genetic report to each pregnancy appointment.

Rhea's wife suffered six miscarriages in the next five years. In November 2007, the couple conceived again, and the pregnancy progressed through the first trimester without incident. The obstetrician scheduled a chorionic villus sampling (CVS) procedure to obtain fetal tissue for testing.

Then a series of unfortunate events unfolded. The testing needed to be done within a short window of time, and the only opportunity available was on New Year's Eve, a day on which no genetic counselor was working at the hospital. The couple decided to have the perinatologist/obstetrician who was available go ahead with the procedure. Normally, the genetic counselor would have handled the lab forms, but in the absence of any such person, a medical assistant did it. The physician testified that he gave Rhea's genetic report to the assistant to send with the form, but it is clear that this did not happen. A cytogenetic technician-in-training was assigned to perform the testing, but did not perform the sensitive testing or determine that more testing was warranted. The policies of the laboratory, LabCorp, require supervision of technicians-in-training, but no one else looked at the case. Because of all (or a combination of some sub-set) of these circumstances, the needed test was not conducted.

The genetics counselor back at the hospital in Washington reported to the couple that the test results were “normal.” She did not report the results back to the obstetrician who did the procedure. Therefore, the obstetrician reported back to the primary care physician that the “fetal chromosome results were normal, with no evidence of … a translocation.”

Failing to get the warning that might have altered their plans, the couple carried on with the pregnancy and the baby was born extremely disabled. Genetic testing confirmed that he had inherited the catastrophic condition that Rhea's cousin suffered from. The Washington Supreme Court upheld the $50 million verdict against LabCorp and the obstetrician.

In another case, Levy v. Legacy Health System, 12 Or. Lit. Arb. Repts 78, 2012 WL 1646980, 34-year-old Deborah Levy had CVS testing because of concerns about possible genetic disorders. The plaintiffs claimed that the doctor removed maternal tissue, not fetal tissue, and no one realized that this had happened. When the test returned negative results, the couple evidently believed them, as opposed to two ultrasounds that showed potential indications for Down's syndrome. The mother claimed that she would have terminated the pregnancy had she known the diagnosis. The jury returned a plaintiff's verdict of $3 million.

Testing That Does Not Consider All Variations

Even if genomic testing is ordered, there may still be problems with the findings.

In Williams v. Quest Diagnostics, C.A. No. 3:16-cv-00972-MBS (filed 2016), pending in the U.S. District Court in South Carolina, a motion to dismiss was recently argued. In that case, a child was diagnosed at one year of age with a SCN1A mutation that causes an epileptic syndrome called Dravet. The plaintiff claims that the laboratory's misclassification of the child's mutation as Variants of Uncertain Significance (VUS) was the cause of his death by seizure several months after the test results were reported, because the misclassification led his physicians down the wrong diagnostic path. The gist of the claim is that the child's health care providers continued to treat the toddler as if he had a mitochondrial disorder, giving him increasing doses of drugs that were contraindicated for the child's mutation. The mother obtained the report, but not until 2014. In 2015, the laboratory reclassified the mutation as disease-creating.

The Williams motion to dismiss, argued in January, is currently focused on whether the laboratory is a “health care provider” within the meaning of the medical malpractice statute. If it is, the lawsuit is barred by the three-year medical malpractice statute of limitations. The plaintiff argues that the South Carolina Supreme Court already declared, in Swanigan v. American National Red Cross , 438 S.E.2nd 251 (1993), that such entities are governed by the negligence statute of limitations, not that of medical malpractice. While this case may be determined on statute-of-limitations grounds, the caution for testing facilities remains. If laboratories are not denominated as health care providers, their liability and responsibilities may be greater if the jurisdiction's statute of limitations for bringing general negligence claims is different from that for medical malpractice claims. In any case, testing that does not take into consideration all possibilities puts the provider, and potentially the laboratory, at risk for liability.

Reporting Incidental Findings in Genome Sequencing

The American College of Medical Genetics and Genomics recommends that whenever clinical sequencing is performed, the laboratory should conduct a deliberate search for medically significant mutations in a minimum list of 56 genes and report them to the patients' clinician, who would be responsible for counseling the patient. American College of Medical Genetics and Genomics: Incidental Findings in Clinical Genomics: A Clarification. Genet Med. 2013;15:664-666. These recommendations have generated discussion on how incidental findings should be reported and how best to manage potential liability risks for both providers and laboratories. Clayton EW, Haga S, Kuszler P, et al: Managing Incidental Genomic Findings: Legal Obligations of Clinicians. Genet Med. 2013;15:624–629. A Presidential Commission for the Study of Bioethical Issues has also issued its own recommendations, in parts at odds with the American College recommendations. See Anticipate and communicate: ethical management of incidental and secondary findings in the clinical, research, and direct-to-consumer contexts. Presidential Commission for the Study of Bioethical Issues. 2013.

To date, we have not found any cases regarding a requirement that incidental findings in genetic testing be reported. But, as some commentators suggest, it is only a matter of time. Powlege, Tabitha, What Should be Done with Unsettling 'Incidental Findings' in Genetic Testing? Genetic Literacy Project, March 2015. As genomic testing becomes more sophisticated, and the price decreases, there will be more access to such testing and, with that, greater expectations.

Conclusion

As we have seen, genomic testing has the potential to be of value to patients and their care providers in terms of identifying disease and managing it, but it also creates new perils for both providers and the laboratories that analyze the tests. We can expect more litigation regarding negative consequences for patients to be on the horizon.

*****
Linda Crawford, a member of the Medical Malpractice Law & Strategy Board of Editors, teaches trial advocacy at Harvard Law School.