NEW YORK (GenomeWeb) – At a hearing this week in the US District Court for the District of South Carolina, Judge Margaret Seymour seemed to be mulling if she should ask the state supreme court to consider whether a diagnostics lab qualifies as a healthcare provider under state laws.
The answer will decide whether Amy Williams' wrongful death lawsuit against Quest Diagnostics and subsidiary Athena Diagnostics is dismissed or moves toward a jury trial. Courts in other states have ruled that labs that provide genetic testing are protected by statutes of limitations and repose afforded to healthcare services, and it's been a decade since Athena issued the genetic test report at issue in the case.
Williams sued Quest and Athena last year for negligence in classifying an SCN1A mutation in her son Christian as a variant of unknown significance (VUS) in 2007. Several months after that report, on Jan. 5, 2008, two-year-old Christian had a fatal seizure and died.
SCN1A is a highly variable gene, and mutations are well known to cause an epileptic syndrome called Dravet, which occurs in one in 21,000 infants. Around 80 percent of Dravet patients have an SCN1A mutation, though some SCN1A variations are benign. Some Dravet patients have mutations in other genes or no known genetic cause.
Williams alleges Athena's misclassification of Christian's SCN1A mutation as a VUS is the proximal cause of his death, because it led his doctors down the wrong diagnostic path. According to the complaint, based on the VUS result, Christian's doctors continued to treat him as if he had a mitochondrial disorder, giving him increasing doses of sodium channel-blocking drugs. Although such drugs are standard treatments for epileptic seizures, they worsen seizures in Dravet patients and weren't helping Christian.
Quest has asked Judge Seymour to dismiss the lawsuit, arguing that Williams is alleging medical malpractice and not negligence, and, as a result, is time-barred. In South Carolina, plaintiffs have three years to bring a medical malpractice suit against a licensed healthcare provider from the time they discover they were harmed, but they lose the right to sue entirely after six years.
In matters of negligence, which is how Williams is describing her complaints against Quest, there is a three-year limitation but also the "discovery rule," which would give her three years from the time she realized there was cause for a lawsuit. Williams claims she did not know about her son's 2007 report until 2014, when she requested and received it. By then, the variant had been deemed disease-causing and Quest issued a revised report with the new classification in 2015.
Quest is arguing that the clock for filing a medical malpractice lawsuit should start once the 2007 report was issued to Christian's clinical geneticist John Shoffner at Horizon Molecular Medicine, a practice that dissolved in December 2007, while Christian was a patient there. Williams is arguing that the timeline should start from when she learned about Athena's "negligent misclassification" and saw the report in 2014.
How the judge decides this matter hinges on whether Quest — a reference lab that provides clinical lab testing services to half the physicians and hospitals in US — is a healthcare provider under South Carolina law. At the hearing yesterday, Judge Seymour asked both the defense and plaintiff's lawyers whether this question should be put before the South Carolina Supreme Court.
According to Williams' lawyer Brad Cranshaw, the Supreme Court has already taken up the issue in Swanigan v Red Cross. In that 1993 case, Pumpy Swanigan received a blood transfusion collected and processed by the Red Cross, but a year later learned that the blood he received was HIV positive.
Swanigan's wife sued the Red Cross for negligence after he died, but the Red Cross argued that it was a healthcare provider, and that she was making a medical malpractice claim barred by the three-year statute of limitations. The Supreme Court held that the Red Cross was not a healthcare provider, "because it plays no role in the care of patients" and merely employing healthcare professionals wasn't enough.
Tally Casey, an attorney representing Quest and Athena from the firm Wyche, said that while the Red Cross provided a defective product — HIV-positive blood — it did not provide Swanigan a diagnosis. In contrast, Casey noted that the results of the genetic test performed on Christian had a diagnostic purpose and was important for treatment decisions.
In its motion to dismiss, Quest has highlighted that courts in other states have determined that similarly situated genetic testing services were protected from lawsuits by statutes of limitations and repose as healthcare providers. Based on other case law, Casey further asserted that the lawsuit should be medical malpractice since Williams relies on affidavits from Robert Cook-Deegan and Max Wiznitzer, both of whom are doctors of medicine.
Cook-Deegan, now at Arizona State University, held an appointment in internal medicine at Duke University at the time the lawsuit was filed. Casey highlighted in court the portion of the affidavit signed by Cook-Deegan, alleging that Athena "breached the standard of care of a CLIA-certified diagnostic laboratory performing high-complexity genetic testing by its negligent failure to correctly diagnose the DNA missense mutation in the decedent’s SCN1A gene" — placing special emphasis on the words "diagnose" and "standard of care."
Similarly, Casey noted that the affidavit of Wiznitzer, a pediatric neurologist at Rainbow Babies and Children's Hospital, attests to "a reasonable degree of medical certainty, that if [Christian's] … condition had been properly diagnosed and had he received appropriate care … he would not have suffered the fatal seizure on Jan. 5, 2008."
It's an error that cost my client his life.
Weighing the evidence
In addition to trying to get this case dismissed on statute of limitations and repose grounds, a key part of Quest's defense is that the 2007 report contained "repeated and emphatic" warnings that Christian's parents needed to be tested to determine whether he had inherited the missense variant or it arose de novo, meaning it was unique to the child.
More than 90 percent of missense mutations known to cause Dravet are de novo. The 2007 report states: "Testing of the biological parents is strongly recommended to resolve the uncertainty of these test results."
Meanwhile, Williams' allegation that Quest erred in classifying Christian's mutation as a VUS in 2007 focuses on the fact that at the time there were two publications that reported the same SCN1A mutation Christian had in another patient with Dravet. An author of one of the papers was Sat Dev Batish, who then was and still is Athena's chief director of genetics. Moreover, Athena outlines in the 2007 report its classification criteria where the only requirement for deeming a variant to be disease causing is whether it was reported in the literature to be associated with the disease.
Based on this, Williams asserts that Athena should have classified the mutation as disease causing and not a VUS. "The 2007 report is as close to a smoking gun as I've come across," Cranshaw told the judge. "It's an error that cost my client his life."
According to Cranshaw, Christian's doctors had a right to rely on the report that the detected variant's links to Dravet were "unknown" and continue to treat the child as if he had a mitochondrial disorder. He further claimed — counter to the opinion of experts in the field and possibly common practice even back in 2007 — that because missense mutations tend to show up de novo in Dravet, that parental testing wasn't necessary for a pathogenic determination.
Heidi Rehm, director of the Laboratory for Molecular Medicine at Partners Healthcare Personalized Medicine, who wasn't at the hearing but is familiar with the case, is of the view that Christian's clinical geneticist, Shoffner, really should have ordered parental testing to determine de novo status, since this was and is a fairly common basis for classifying mutations in rare diseases like Dravet. The fact that Athena did not cite the two publications where another child with Dravet had the same mutation is a failing, Rehm acknowledged, particularly since an author of one of those papers was the lab's own chief director of genetics.
"My belief is at that time, had they found the other papers that documented the mutation as a de novo occurrence in another case, lab standards at the time would have been to interpret that variant as likely pathogenic," said Rehm, who is an expert in variant classification and the lead author of the American College of Medical Genetics and Genomics 2015 guidelines on the topic. She added, though, that even without the papers, de novo status would have upgraded a VUS to likely pathogenic status.
Robert Lawther, professor emeritus in the biology department at the University of South Carolina, who was at the hearing out of interest in the case, is more troubled by the allegation that Athena missed a key publication on Christian’s variant that one of its directors contributed to. "How can you write this up and not acknowledge this information?" he said. "It's difficult to understand why they didn't know that."
But he also wondered about the responsibilities of Christian's doctors. "There is this huge unknown about Shoffner," he said. "I think there isn't any place in all of this, in terms of the care of Christian, that wasn't fouled up. Everybody played their part, including the technology."
A physician’s responsibility
At the hearing, Judge Seymour wondered how Williams could have gotten tested if she did not know about the 2007 report until 2014, as she claims. To that Casey said providing the report to Christian's doctor sufficed as constructive notice to her through the so-called "learned intermediary doctrine".
This rule has been used in jurisdictions, including South Carolina, mostly to exempt pharmaceutical companies from having to directly warn patients about drug risks, if they have warned doctors. But some courts have found there are special circumstances where, for example, pharmacists have a duty to break with this general doctrine and warn the patient about dangerous side effects or dosing errors.
Judge Seymour asked Casey whether Williams could have a claim against Shoffner for not informing her about the report. Casey acknowledged she could but wouldn’t comment further about when the clock would start for filing suit against the doctor. To date, there hasn't been much discussion about the role of Christian's doctors in court filings, but if the case reaches discovery, this would be of critical interest to Quest and Athena. Of course, a medical malpractice suit against Christian's doctors would also contend with statute of limitation and repose hurdles.
Casey quoted from different parts of the 2007 report to assert that it was unreasonable for the doctors to rely only on four words — "variant of unknown significance" — to decide Christian's diagnosis and treatment course, when throughout the report there is language indicating that the SCN1A gene is associated with Dravet and that such a diagnosis would impact treatment decisions.
Rehm noted that in the case of rare variants, labs often don't have details about a patient's phenotype to make more definitive judgements, but that the doctor has this information. "Two thirds of the variants we report out as clinically significant, including VUS, are unique to the proband," she said. "They've never been reported before. In those cases, you get one shot to figure it out, and the onus is on the physician to do something." Particularly a medical geneticist, she said, should consider the variant in the context of the phenotype and the likelihood that it could be causative, and pursue relevant tests that could clarify the significance of the variant.
It's a mistake in general to see a lab report on a genetic test as the last word.
The ACMG's 2015 guidelines state that a VUS "should not be used in clinical decision making." However, the ACMG adds that efforts to arrive at a pathogenic or benign classification should be made, and while reclassification efforts are ongoing "additional monitoring of the patient for the disorder in question may be prudent."
"It's a mistake in general to see a lab report on a genetic test as the last word," said Bryce Mendelsohn, a pediatrician and medical geneticist at the University of California, San Francisco Medical Center. "Clinical judgement is critical in these situations. Doctors aren't robots who respond to just whatever the lab report says."
Mendelsohn wasn't at the hearing and didn't want to comment on the case, but offered some insights on how he deals with uncertain genetic test results. He recently saw a patient who was suffering from seizures and who received testing for several genes associated with the condition. The patient received a VUS result for one of these genes, "but we didn't stop there," he said. Mendelsohn performed additional testing and investigated the patient's metabolic signature to conclude that the VUS was pathogenic for pyridoxine dependent epilepsy, even though there was no evidence in the literature that the variant was pathogenic.
But even before the diagnosis was made, Mendelsohn initiated the patient on vitamin B6 which is the treatment for pyridoxine dependent epilepsy, and a generally safe therapeutic option. "It's not so much that if you have an uncertain variant you can't act on it," he said. "It's a clinical judgment about the benefits and risks of how you act on that variant. If the treatment for that condition is brain surgery, you'd want to be very conservative, but if the benefits of acting on a variant outweigh the risks, that is a matter of clinical judgement that is a fundamental part of practicing medicine."
While variant classification requires facility with a lot of technical information — how often a variant occurs in the population, has it been seen before, and how it impacts protein function — Mendelsohn said doctors should still try to investigate these aspects, especially in the case of a VUS report. They should also familiarize themselves the latest guidelines on variant classification and what pieces of evidence pushes a variant toward a benign, uncertain, or pathogenic classification.
Despite the availability of guidelines, labs employ different processes, which sometimes result in discrepancies in how they classify a particular variant. In NIH's public database of genotype/phenotype relationships, called ClinVar, 17 percent of variants submitted by multiple labs had conflicting interpretations.
"Labs don't always find the critical paper for that variant and don't always look in the database that happens to have that variant," Mendelsohn said. "Many medical tests have some degree of uncertainty, and the reason why people get paid to be doctors is to take that uncertainty and make a decision."
In a Medscape article last year, Mendelsohn and Jeanette McCarthy, a professor at Duke University Medical Center, used Williams' case to discuss "the complexities and subjective nature" of variant classification. They have even used Christian's variant as a teaching tool in educational workshops, where participants used the latest guidelines to interpret the variant, with some concluding it was likely pathogenic and others classifying it a VUS.
Any place where there was any doubt about what the lawyers were saying, she asked very direct, pointed questions.
Piecing together the facts
However, current variant classification standards aren't relevant in a court case concerned with the accuracy of a 2007 test report. Casey argued that the genetic testing field is much changed from when Christian got tested and so it would be nearly impossible to recreate what happened and how decisions were made.
Still, during the hearing, Judge Seymour seemed interested in piecing together what happened a decade ago with regard to Christian's report. She asked Quest's lawyers which doctors oversaw the 2007 and 2015 reports.
They said William Seltzer, who is currently at Oncocyte Corporation but who previously held various director roles at Athena, was responsible for the 2007 report, and Joseph Higgins, Athena's current CLIA director, was responsible for the revised 2015 report. When Judge Seymour pressed Casey as to who made the decision at Athena to update the report, she stated she was not prepared to discuss Athena's policies for reclassifying variants and didn't want to provide inaccurate information.
However, both the 2007 report Williams received from Quest in 2014 and the 2015 revision bear Higgins' name. The 2015 revised report also lists the names of two other directors at Athena, Narasimhan Nagan and Hui Zhu, who had left the lab several years before. This further convolutes the facts of the case and suggests that Williams still doesn't have access to the original 2007 report that Christian's doctor received.
GenomeWeb consulted several experts in laboratory procedures, and they opined that industry best practice is to provide the patient with the original report, which a board-certified clinical molecular geneticist has signed off on. There is no signature on the 2007 or 2015 reports Williams received, but the names of the directors are printed. Some experts acknowledged the possibility that when the 2007 report was issued to Williams in 2014, Athena's electronic system automatically generated the name of the current CLIA license holder.
But Williams and her lawyers point to these discrepancies as CLIA violations and signaling something more nefarious — that Quest is trying to cover up Athena's mistake in classifying Christian's variant in 2007. The judge pressed Cranshaw regarding the allegation that Quest and Athena conspired to cover up a variant classification error once Williams began asking questions in 2014. But by then Quest had acquired Athena, and the defendants point out in their motion to dismiss that "South Carolina Supreme Court has repeatedly ruled that a corporation cannot conspire with itself."
Cranshaw replied that if the case reaches the discovery stage then it might reveal certain dealings between the companies when they were separate entities negotiating the acquisition that could provide support for the conspiracy allegations. "The plaintiff's council is making things up," Casey responded. "There is no good faith basis for these allegations."
Cranshaw meanwhile maintained that Quest and Athena were trying to obfuscate the facts of the case in order to deflect fault. "All I've seen is blame for the mother," he told the judge.
USC's Lawther wasn't impressed with the ability of the lawyers to argue the facts of the case, but he felt Judge Seymour was giving careful consideration to the complex issues involved. "Any place where there was any doubt about what the lawyers were saying, she asked very direct, pointed questions," he said.