NEW HAVEN — Death is not final in the Yale New Haven Hospital Pathology Department.
By going beyond the standard autopsy to examine a deceased patient’s DNA, the team led by Dr. Jon Morrow can find out the cause of an unexplained sudden death, including newborns, and inform families of inherited diseases that could help them make decisions about their health care.
“As our knowledge of the genome has expanded … there has increasingly been identified a number of genetic changes … that have turned out to be causal in many of these cases of sudden death,” said Morrow, chief of pathology for the hospital.
Yale New Haven is one of the few hospitals to do the intensive postmortem examination, and Morrow said they are interested in performing molecular autopsies on anyone whose family is willing to have one done or who has requested it before death, That’s because every exam adds to the body of knowledge about how genes contribute to disease.
“It builds a database that helps us understand disorders at a deeper level than we’ve been able to do from normal autopsy,” Morrow said.
Morrow’s team collaborates with the Yale School of Medicine’s Center for Genomic Health, which launched the Generations Project, which began in September with a goal of sequencing the DNA of 100,000 patients.
In a molecular autopsy, the pathology team looks at the exome, the 3 percent or less of the human genome that includes all of the genes involved in producing proteins. “That 3 percent is responsible for everything you see in the body,” Morrow said, because our organs are all made up of proteins. “It’s a very efficient way to look at almost everything that’s important,” he said.
The rest of the genome is involved in cellular structure, regulatory processes, how and when genes are switched on and off.
By examining the exome, pathologists can compare the individual to a reference database at the National Institutes of Health. The individual differences, called polymorphisms, are what can tell the doctors whether a patient who died from cancer did so simply because he smoked or because he had a genetic predisposition.
“You always run what you find against this reference database,” Morrow said. “You can expect up to 70,000 variations in an individual. … Most of those variations are innocuous. They don’t have any real or deep pathologic significance.”
By sequencing the genome, variations in alleles — the pairs of genes that control characteristics — will be identified and, “if you find a variation in a patient that is extremely rare, lightbulbs go off,” Morrow said. “If that was a variation in a gene that’s rare and a gene is known to control cardiac function, you would be very suspicious that that had to do with a cardiac malfunction.”
In addition to finding the cause of death when a regular autopsy does not, Morrow said molecular autopsies have a larger goal of finding answers on the genetic level. He and his staff hope many people will consent to the procedure. Now, 12 to 14 percent of deceased patients undergo autopsies.
“By looking at this exome, we get a pretty good view of what should be right. … There’s wide variations in individuals,” Morrow said. “Some are pathologic, disease-causing. By moving beyond sudden death and extending these studies to all individuals now, we have a hope of explaining causes of all these conditions.”
When word got out about Yale New Haven’s program, “We had almost universal enthusiasm from a number of major medical centers [asking], ‘Can we partner with you?’” Morrow said. “I’m hopeful we’d be able to expand this to multi-institutional practices.”
Some diseases, such as many cancers, start with genetic mutations that are caused by external factors, such as smoking. Those are called somatic mutations. What the molecular autopsy finds are inherited mutations that may give a predisposition to the cancer.
“We’re not sequencing tumors, we’re sequencing normal tissue,” Morrow said. “When we find a damaging lesion, that gene was present from birth and that gene was inherited.”
Another example is alpha-1 antitrypsin deficiency, an inherited disorder that is responsible for both lung and liver disease. “It’s not that uncommon in the population and people get emphysema and liver disease later in life because of it,” Morrow said. Finding out that a deceased relative had the disorder can help family members lessen their risk, by not smoking, for example.
Finding a genetic cause of death is especially valuable in fetal deaths “that have no anatomic cause of death,” he said. “For a lot of these mothers, just reading their notes, they have a lot of guilt. So if we can identify through whole-exome sequencing, it would at least give them something.” As many as 30 percent of stillbirths and infant deaths may be genetically caused, Morrow said.
“They want to know sometimes, am I at risk?” said Amanda Masters, a pathologists’ assistant on Morrow’s team. “If we do find a genetic link, if it was inherited, it could help them and future generations.”
Some families do not want the autopsy done, and a few consent to it but don’t want to know the results, Morrow said. But almost all do want to understand how their relative’s DNA may affect their health.
“The hospital has agreed to provide one genetic counseling session free of charge to those families,” Morrow said, adding that the results of the autopsy are sent to the genomics lab. “Decisions are discussed. Do you need to be tested? … You have to be very careful because you never want to send up false flags. We’re very conservative in the way we approach this.”
That is important because even genetic testing isn’t infallible. “When the genomic revolution started … there was overpromising of what it could tell you,” Morrow said. Besides wasting money on genetic testing, “it also caused angst.”
Often, even after a standard autopsy, “we still don’t know the reason why they died, even with slides and looking at tissue at a very, very close level,” said Dr. Keith Churchwell, executive vice president and chief operating officer of the Yale New Haven Health System. The lack of information leaves a “real void,” he said.
Conducting a genetic examination of a deceased patient will reveal information “for the next generations, for that family and for their children,” Churchwell said.
“There’s three aspects of doing this that I think are good,” Morrow said. “One is it helps us move the autopsy from what we’ve been doing … into a new realm that helps get deeper answers into disease causation and relevance into that diseased individual. … It offers a pathway for more rational decision-making for individuals related to the [deceased patient].
“The third thing, which goes back to the Generations Project, [is] whether we find anything relevant or not to the disease. It provides a growing database that we can correlate the variations into,” Morrow said. For example, doctors might discover a “liver disease of a certain type that we might not know before.”
“It is a wonderful platform for future discovery on … genetic changes,” he said. “It’s a huge task to figure out which variations are meaningful.”
Morrow said people who are concerned about privacy, worried that their life insurance rates will rise or that they’ll lose their job, shouldn’t be. Since the autopsy is done on a relative, and the medical record is protected by federal privacy laws known as HIPAA, the information wouldn’t appear on a relative’s record anyway. “In a way it’s the perfect vehicle for getting some information … at almost no risk,” he said.