In March 2017, in a room at the outpatient clinic on the fifth floor of the Hospital das Clínicas (HC) of the University of São Paulo School of Medicine (FMUSP), endocrinologist Alexander Jorge called in a couple to present the probable cause of their 4-year-old daughter’s microcephaly and short stature. The culprit was not the mother being infected with syphilis during pregnancy, as had been initially suggested, but a mutation in a DNA-repairing gene called BRCA1. Jorge explained that this mutation, identified to be present in homozygosis (two copies of the gene) in the daughter and in heterozygosis (one copy) in each parent, favors the development of breast and ovarian cancer. Then, the 32-year-old woman said: “Doctor, I have a lump in my armpit.” It was a sign that she could have an undiagnosed–and already growing–breast tumor. The endocrinologist referred the mother of his patient to the oncology team in the neighboring building, for a diagnosis.
Jorge has seen this type of situation frequently since 2013, when he started using a technique that completely sequences a group of DNA segments known as exons, which make up 1% to 2% of our genetic material. The exome–this full group of exons–contains all 19,000 human genes, whose mutations can cause diseases. Complete sequencing of genetic material (the genome) is rarely performed for medical purposes, due to its higher cost and the output of information on DNA segments called introns, which are rarely associated with diseases.
In recent years, complete exome sequencing superseded the previous approach, which read few exons at a time. Exome sequencing has been shown to be effective in identifying the mutations responsible for 8,500 genetic diseases caused by a single gene, in up to 40% of people examined. “The success rate can reach 80% when candidate genes have already been pre-selected to explain clinical symptoms,” says Jorge.
The cost of exome sequencing is still high–in Brazil, it can reach R$10,000. An additional difficulty is that sequencing produces a monumental quantity of information, as any given person will carry approximately 50,000 mutations, most of them harmless. The analyses require specialized teams of biologists, bioinformaticians and physicians to determine the 20 or 30 most relevant alterations that debilitate the original function of genes and could explain the symptoms of a disease.
Despite its limitations, the technique has helped resolve uncertain diagnoses and steer treatments in the right direction. Having identified the BRCA1 gene mutation, which could explain the girl’s microcephaly and short stature, and with the help of similar reports from other countries, Jorge concluded that he should not apply the traditional treatment for shortness, which is based on growth hormone, as this could increase her risk of cancer.
For researchers and clinicians, exome sequencing is introducing a new, more open and collaborative way of working. In September 2016, Jorge had no way of confirming whether a mutation in gene BCL11B was behind the intellectual disability, dental arch defects and short stature of a teenager. He visited the GeneMatcher website, in which information about genetic mutations and their possible consequences are shared with physicians, researchers and patients. His report got the attention of researchers in Germany and the United States who had seen similar cases associated with that mutation. It also allowed him to confirm the diagnosis and to gather information on a new disease, possibly facilitating the work of other doctors.
“The more detailed the descriptions of clinical symptoms of patients and families, with consistent diagnostic hypotheses, the higher the chance of interpreting the exome correctly,” says biologist Maria Rita Passos Bueno, a coordinator of the Human Genome and Stem Cell Research Center (CEGH-CEL) at the USP Biosciences Institute. Physicians, research center and diagnostics laboratories generally work in an isolated manner, but team integration, information exchange and a harmonization of technical and ethical procedures could benefit patients, researchers and medical teams, according to a study by a research group at the Netherlands Cancer Institute, published in May 2017 in the journal Annals of Oncology.
One gene, many diseases
Exome analysis has been showing that diseases once seen as distinct could actually have the same genetic origin. The team at Mendelics, a genetic analysis company started in 2013 in the city of São Paulo, determined that the TRPV4 gene causes both muscle and bone diseases. While pursuing her doctorate at the São Paulo State University (Unesp), advised by biologist Silvia Rogatto, Maísa Pinheiro, also a biologist, identified an alteration in a DNA-repairing gene as a possible cause of breast and thyroid cancers. Her findings will be presented in a paper currently in press. Pinheiro identified the mutation in patients who had one or both types of cancer and were treated at the A.C.Camargo Cancer Center in São Paulo, and who also had family members afflicted by these tumors. “Breast and thyroid cancer may be part of a syndrome that involves hereditary predisposition,” suggests Rogatto, currently at the University of Southern Denmark, in the city of Odense.
Exome sequencing has been increasingly used by research groups in São Paulo, Belo Horizonte, Porto Alegre, Recife, Brasília and other cities to discover the causes of tumors, as well as skeletal, muscle, neurological and other genetic diseases. A research group at the Baylor College of Medicine, in the United States, sequenced the exomes of 2,000 patients, most of them children, with an unconfirmed suspicion of genetic diseases associated with delays in neurological development. All of them had undergone biochemical blood tests that failed to pinpoint the cause of the diseases. As reported in a paper published in 2014 in the Journal of the American Medical Association, the mutations revealed by exome sequencing elucidated the causes of diseases afflicting 504 patients, enabling doctors to plan their treatment and family genetic counseling.
CEGH-CEL has been serving families with genetic diseases since 1968, exome sequencing since 2012, and has identified new mutations that cause autism, muscle and skeletal diseases, intellectual disability, obesity and deafness. In September 2017, the genetic testing laboratory at CEGH-CEL is expected to start offering mini-exome sequencing, at a cost of approximately R$4,000, to search for mutations in up to 6,709 genes that have already been associated with genetic diseases, instead of examining all 19,000 genes. The price reduction could increase the use of this technique within the Brazilian healthcare system.
Physician and bioinformatician Guilherme Yamamoto, who works at the Hospital das Clínicas Children’s Institute and at CEGH-CEL, assessed the effectiveness of mini-exome sequencing in detecting mutations in 500 genes responsible for a group of 26 genetic diseases of newborns, with a focus on inborn errors of metabolism. Using this focused gene sequencing strategy, he re-ran genetic tests for 90 people: 45 children who were patients of the Hospital das Clínicas in the city of Porto Alegre and 45 control patients. With no knowledge of previous diagnoses, Yamamoto found one false-positive and one true-positive among the group of 45 controls that did not have a genetic disease. In the other group, he properly identified the genetic diseases of 22 of the 45 children. The test presented a total sensitivity of 50%, increasing to 73% for inborn errors of metabolism, although it failed to detect cases of muscular dystrophy, epilepsy and immunodeficiency.
“Screening via genetic testing has high specificity, with only one false-positive being found,” Yamamoto says. He argues that this approach could be complementary to the neonatal screening known as the heel prick test, which detects genetic, metabolic and infectious diseases. “The heel prick test has low specificity, with a high rate of false-positives for some of the tested diseases,” he says. According to Yamamoto, increased accuracy in genetic disease detection in newborns could get treatment started earlier.
“Exome sequencing effectively identifies inborn errors of metabolism,” agrees neurologist Fernando Kok, professor at FMUSP and medical director at Mendelics. By the end of 2017, his company wants to start offering a test to identify mutations in 260 genes responsible for 150 genetic diseases in newborns.
The biologists, geneticists and medical doctors at Mendelics receive the results after they are processed by a computer program called Abracadabra, developed by a group headed by neurologist David Schlesinger, who is also the president of the company and was featured in 2014 by MIT Technology Review among the top 10 most innovative researchers under age 35 in Brazil (see Pesquisa FAPESP Issue No. 220). Abracadabra filters mutations by groups of diseases, pointing out harmful ones and eliminating incidental findings from genes of lesser concern. It also automatically searches for similar reports in online databases.
Another program, developed by the Mendelics team under bioinformatician João Paulo Kitajima, started initial tests in August to identify genes that escape from conventional exome analyses. These genes have a peculiar characteristic: they do not tolerate the loss of function of one of their copies, in what are known as constraints. With this strategy, they hope to elucidate unresolved cases. In two days of work, physician geneticist Luiza Ramos found a genetic alteration that could explain one of the first 52 unresolved cases examined using the new software. Sitting at another computer in the same room, physician geneticist Fabíola Monteiro said that exome analysis is expanding existing knowledge about genetics by revealing multiple forms and causes for the same diseases. “What is written in the genetics books,” she says, “is very limited, compared to what we are seeing”.
Understanding the results
At the University of Campinas (Unicamp), a research group coordinated by physician geneticist Iscia Lopes Cendes is developing computer programs to facilitate the interpretation of exome sequencing results, which may be helpful in finding the causes of diseases set off by a single gene. These programs also integrate that information with the data available in databases like the Brazilian Initiative on Precision Medicine (BIPMed), which contains specific genetic information on the Brazilian population. The first such programs are expected to be released for public use by late 2017, says Cendes, who is a professor at the Unicamp School of Medical Sciences and researcher at the Brazilian Institute of Neuroscience and Neurotechnology (BRAINN), one of the Research, Innovation and Dissemination Centers (CEPID) funded by FAPESP.
In addition to seeking new strategies for exome analysis, Cendes’ group is also investigating the cost-benefit aspects of incorporating this type of test into the public healthcare system. In a study currently being wrapped up with 150 patients, physician geneticist Joana Trotta determined that exome sequences could indicate the probable causes of 80% of adult-onset neurodegenerative diseases, such as ataxia and dementia, which are not always detected even after a series of imaging exams. The proportion of positive diagnoses for intellectual disability was 20%, 10 times more than that obtained via chromosome (karyotype) analysis, the technique used by the public health system. “By increasing the proportion of accurate diagnoses on the origin of diseases, expenditures on other tests and clinical examinations will drop, and treatment, if possible, can start quickly,” says Cendes.
1. New approaches and methodologies in molecular-genetic studies of growth and pubertal development disorders (No. 13/03236-5); Grant Mechanism Thematic Project; Principal Investigator Alexander Augusto de Lima Jorge (USP); Investment R$2,948,891.06.
2. CEGH-CEL – Human Genome and Stem Cell Research Center (No. 13/08028-1); Grant Mechanism Research, Innovation and Dissemination Center (CEPID); Principal Investigator Mayana Zatz (USP); Investment R$27,221,413.39 (for entire project).
VIS, D. J. et al. Towards a global cancer knowledge network: Dissecting the current international cancer genomic sequencing landscape. Annals of Oncology. V. 28, No. 5, pp. 1145-51. 2017.
YANG, Y. et al. Molecular findings among patients referred for clinical whole-exome sequencing. JAMA. V. 312, No. 18, pp. 1870-79. 2014.