Genetic Cause and Developmental Mechanisms Underlying Congenital Heart Disease

April 30, 2021

ID6549

To CiteDCA Citation Guide

00:00Good afternoon everyone.
00:01My name is Nicole Lake and I'm a postdoctoral
00:05associate in the Department of Genetics.
00:08It's my great pleasure to now introduce
00:10Doctor Martina Brueckner Doctor Bruckner
00:12received her undergraduate and medical
00:14degrees from the University of Virginia,
00:17then trained in Pediatrics at the
00:19University of Pittsburgh and completed her
00:21pediatric cardiology fellowship at Yale.
00:23She has been on the
00:25faculty at Yale since 1991,
00:27where she provided clinical,
00:29pediatric cardiology care and founded
00:31the Yale Pediatric Cardi Attic.
00:33Cardiac genetics clinic.
00:34Her laboratory now focuses on
00:37the molecular and genetic causes
00:39of congenital heart disease,
00:41with a special focus on the role of
00:44cilia in heart development and disease.
00:47She has LED Yale's participation
00:48in the pediatric Cardiac Genomics
00:50Consortium since its inception in 2009.
00:53The zoom floor is now yours, Doctor Bruckner.
00:57Thank you, let me share screen.
01:10Wait? There we go.
01:13Is that visible for everybody?
01:17I believe so yes. OK, so I'm going
01:20to be talking about the genetic
01:23causes of congenital heart disease,
01:25which is and I don't have
01:27any conflicts of interest.
01:29So congenital heart disease
01:30is quite a common problem.
01:32It affects one in 100 life born infants.
01:3690% of our patients survived to adulthood,
01:38which is definitely a change over the
01:41time that I've been in this field,
01:44but many suffer comorbidities,
01:46including neurodevelopmental
01:47and respiratory problems.
01:48What we've come to believe is
01:50that about 90% of patients with
01:53congenital heart disease have a
01:55significant genetic contribution.
01:57And unlike in the past,
01:59this is no longer an itch kind of disease.
02:02If you look at these statistics from 2010,
02:05there are two point 4,000,000
02:06people in the United States living
02:08with congenital heart disease.
02:101,000,000 are under 18 years,
02:121.4 million are over 18 years,
02:14and since 210,
02:15the proportion of those that
02:16are in the over 18 years of age
02:20range has grown significantly.
02:21Also,
02:22looking at the population disease
02:24statistics in other countries,
02:25there are in Canada 257 thousand people
02:28living with congenital heart disease
02:30as compared to for instance 71,000
02:33with HIV or 4000 with cystic fibrosis.
02:36So this is a common problem that touches
02:39quite many aspects of the medical system.
02:42The outcome in congenital heart
02:44disease is incredibly variable,
02:46so a very common type of congenital
02:49heart disease tetralogy of fellow can be
02:52surgically repaired very affectively.
02:54Post operatively him,
02:55the children will quite compromised.
02:57However many of them grow up and are
03:00quite fine as seen here by Shaun
03:02White winning an Olympic gold medal.
03:05He is status post.
03:06I believe three tetralogy surgeries however.
03:09If you look at the serious adverse
03:12events in this population as a whole,
03:15they are quite significant,
03:16and by the age of 30,
03:19about 910% will have suffered a serious
03:22adverse event and somewhere between
03:2415 and 20% a significant adverse event.
03:28So the question we have is whether
03:30genetics can lead to personalized
03:31treatment of congenital heart disease
03:33and the associated comorbidities,
03:35and so a national consortium was
03:37formed in 2009 called the Pediatric
03:39Cardiac Genomics Consortium,
03:41and Yale was one of the founders of
03:43that group when it was first developed.
03:46Here are the current set of
03:48principle investigators.
03:49The main thing that this group has done.
03:52And we wanted to discover all the genes
03:55that cause congenital heart disease,
03:57identify the mutations responsible
03:59for congenital heart disease,
04:00and then link those findings to the outcome.
04:04What we've done so far as we for Crew did
04:0714,000 patients and over 14,000 relatives.
04:11This is the largest congenital
04:13disease cohort anywhere.
04:14The cardiac phenotyping was done by Echo.
04:16There's more superficial
04:18extracardiac phenotyping.
04:18We developed a database called
04:20Heart Smart and Mind you,
04:22this database was developed in 2009 and 2010,
04:26so we're now realizing it is
04:28grossly outdated and we have work
04:30in progress to link the genomic.
04:33An electronic medical record data.
04:35Or yell came into this is that
04:37the El Center
04:38for Genome Analysis has been the sequencing
04:40Center for the program since its inception.
04:44A man they have sequenced
04:464075 program parent trios.
04:48So that's about 13,000 exomes
04:5017125 Singleton probands. Answer.
04:52Currently in the process of doing many more,
04:55there's also whole genome
04:57sequencing available.
04:58And as I'll talk about later why SGA
05:01helped develop some targeted sequencing,
05:04that allowed us to expand our
05:06cohort size were also banking,
05:08cardiac surgical tissue.
05:10So the main findings from the
05:131st 10 years of this effort is.
05:16There are really 22 major sort of
05:19biological mechanisms that might
05:20link to congenital heart disease.
05:23Recessive mutations in the cilia genes.
05:25These are patients that might also
05:28have some respiratory compromise.
05:29Dominant mutations in chromatin
05:31modifier genes that associate
05:33specific subtypes of congenital
05:34heart disease and most importantly,
05:36these are mutations that really
05:38provide a biomarker that a patient
05:40is at very high risk for developing
05:43neurodevelopmental compromise and
05:44might benefit from more aggressive
05:47neurodevelopmental follow-up.
05:50The congenital heart disease genetics,
05:52however, is really incredibly
05:54complicated and characterized
05:55by vast genetic heterogeneity,
05:57so this is the sequencing data.
05:59Over the course of the project,
06:01we started with 362 trios at a time when we
06:05were told that that was a vast overreach,
06:09and we would never be able to pay for it.
06:13I believe a single exon at that
06:16time was somewhere around $800.
06:19That we have now progressed,
06:21obviously to a much larger number,
06:23and you can see that the number of genes
06:26with significant IKA more than one damaging
06:30de Novo mutation has expanded from 2 to 95.
06:33However, when you take all the data together,
06:36we predict that at least 400 genes
06:39contribute to congenital heart
06:40disease by a denovo mechanism alone,
06:43and that's not counting all the genes and
06:46biological pathways that may be implicated
06:49as recessive or contributing genes.
06:51We think we need about 10,000 trios
06:53to identify 50% of the genes that
06:56cause congenital heart disease
06:57by de Novo mechanism.
06:59Probably 20 to 30,000 to begin to define
07:01the multigenic genetic mechanisms.
07:03So we're really now lost in that
07:06middle heavy digging part that IRA
07:08alluded to at the very beginning
07:10of this session of this symposium.
07:13And we probably need very large
07:15numbers to understand the genetics of
07:18specific congenital heart disease sub
07:20phenotypes because really congenital
07:22heart disease as a whole is just a
07:25mishmash of anything that went wrong
07:28with cardiac development prenatally.
07:30So what we really need to do is
07:32increase the number of patients that
07:35we have sequenced and this goes into
07:38the whole issue of genomic health
07:40and recruitment and sequencing
07:42of really large patient cohorts.
07:46I took a first step together with the
07:48Yale Center for Genome Analysis because
07:50as you can see on the table on the left,
07:53the cost of a whole exon.
07:55When we did this was 140 actually
07:57was at one point it was $200 and so
08:00the limitation was not the number of
08:02patients who were willing to participate,
08:05but just the number of dollars
08:06we had to do the sequencing.
08:09So Michael Serrant,
08:09who's a graduate student,
08:11said, OK,
08:12I'm going to develop a targeted
08:14sequencing approach that allows
08:15us to get all the best guests.
08:17Congenital heart disease genes.
08:19In a lot of patients,
08:21for a very low cost,
08:23$37 a sample.
08:24He did this with molecular inversion probe
08:27sequencing that's outlined on the right.
08:30This panel included 248 known and
08:33putative congenital heart disease genes,
08:35and he was able to do a statistically
08:39robust meta analysis combining
08:41our whole exon data and the
08:44targeted sequencing data so that
08:46now he has genomic data on 11,000.
08:49500 programs with congenital heart
08:51disease and this really lets us
08:54see what expanding cohort size
08:56and expanding the amount of data
08:58we have available in forms,
09:00so we were able to increase
09:02from a small number of genes.
09:04The genes that are underlined here
09:07that had significant statistical
09:09validation of a role in congenital
09:11heart disease to 61
09:12jeans, and this is by increasing from
09:152700 programs to 11,000 programs.
09:17Now remember this is only across 248 genes,
09:20so there's still.
09:22A lot of room to learn more.
09:25The other thing is this is across the
09:27broad range of congenital heart disease.
09:30However, having this very large cohort
09:32now allows us to separate the genes to
09:35those that provide risk for specific
09:37subtypes of congenital heart disease.
09:39For instance,
09:40you can see on the right that the
09:42jeans flip for Jaguar include H
09:45and TBX really put you at risk for
09:48developing a type of congenital heart
09:50disease called tetralogy of fellow.
09:52And when you look even more carefully
09:54you can now submit segregate.
09:56Those patients to identify the patients that,
09:59for instance,
09:59are at high risk for neurodevelopmental
10:02abnormalities and those that are not.
10:03You can see that these genes on
10:05the right are the ones that have
10:08very specific contributions.
10:09The ones on the left are genes
10:11that are more globally involved in
10:13heart development and cause a broad
10:15range of congenital heart disease.
10:17And we're now hypothesising,
10:18but some of these changes may
10:20actually affect more than cardiac
10:22structure may also affect,
10:23for instance,
10:24that progressive risk for myocardial
10:26dysfunction over patients lifetime.
10:28So what do these discoveries from
10:30genomic analysis do to inform clinical care?
10:34And here's an example from the
10:36Yale Health System,
10:37a 2 month old ex pre term infant
10:40was transferred for hepatic
10:42failure and liver transplant and
10:44the liver transplant team obtained
10:46rapid whole exome sequencing.
10:48Patient also had congenital heart disease,
10:51atrioventricular canal and
10:52repeating ductus arterio sis,
10:53although that particular type of congenital
10:56heart disease should not have resulted.
10:59In this degree of distress at this age,
11:01here is an outline of the heart disease.
11:05An extensive metabolic work up was
11:07done to show A cause for the hepatic
11:10failure yielded nothing exon sequencing,
11:12which why SGA turned over in five days
11:14revealed a loss of function mutation
11:17in a chromatin modifier called CHD 7.
11:19This mutation was diagnostic for
11:21something called Charge syndrome
11:23which nobody thought the patient had.
11:25This is the list of the particular
11:27features of Charge syndrome.
11:29A coloboma choanal atresia.
11:30These are usually quite obvious.
11:32Patient didn't have any of those.
11:34The only things he had was heart defect.
11:37He had somewhat unusual looking external
11:39ears and a C HD7 gene mutation.
11:44And then we did a literature search on
11:47liver function in charge syndrome and found
11:49not a single report of liver abnormalities
11:51or liver failure in that setting.
11:54So the thought began to percolate to the
11:56top that this patient actually was much
11:59more affected by the congenital heart
12:01disease than primary liver disease went to.
12:04The Cath lab,
12:05had the patent ductus arteriosus closed.
12:08And within about five days,
12:10hepatic function started returning
12:11to normal in the patient was
12:13discharged without a liver transplant.
12:15So this kind of highlights how
12:17knowing the genetics lets you predict
12:20what a patient is at risk for,
12:22but also lets you predict maybe what a
12:25patient is not at risk for and provides
12:28a more objective lens through which
12:30clinicians can view how patient is doing.
12:33Was this just an isolated patient or
12:35a more global finding an answer from
12:38the large sequencing effort from the
12:40PC GC where we did look at those 11,000
12:43patients is that this is actually
12:44much more common than we thought,
12:47so this list is all the patients
12:49that we found.
12:50A molecular charge diagnosis and the
12:52ones up top had a clinical diagnosis.
12:54The ones on the bottom do not.
12:56There is nothing about their mutations
12:59that predicts they will fall into
13:01the clinical charge diagnosis or
13:03the no clinical charge diagnosis.
13:05If you look at their phenotypic spectrum,
13:08the ones with the molecular diagnosis.
13:10Obviously since the entrance mechanism
13:12here was congenital heart disease,
13:14100% of them had congenital heart disease,
13:17but a much smaller number had the other
13:20expected findings of charge syndrome,
13:22and So what I think we're at now is that
13:25a molecular diagnosis is a prediction
13:27of risks and is somewhat different
13:30from the old clinical diagnosis
13:32that were based on very specific,
13:35well defined clinical findings.
13:37These molecular diagnosis,
13:38especially if they're returned
13:40rapidly enough,
13:40can really significantly inform
13:42patient care and give us some guidance
13:45as to what are the risks that that
13:48patient is going to face.
13:49What are some of the things we
13:52should look for or not look for
13:55and significantly enhance patient
13:56care going forward?
13:58So where do we go from here?
14:00The diagram on the left is what
14:02we know so far,
14:03and what we've really learned over
14:05the last 11 years of participating
14:08in this project.
14:0956% of congenital heart disease
14:11still has an unknown cause,
14:13and so we really need to go after
14:16this more challenging aspect.
14:19These could be due to rare common
14:22variant interactions,
14:22recessive contribution across the genome,
14:24Multigenic contribution and contribution of.
14:26Then we are also interested in
14:28looking at the contribution of
14:30the genetic data to outcome.
14:32But to do this,
14:34you really need to expand the
14:35genomic data and why SGA has
14:38recently very generously agreed
14:39to lower their sequencing rates.
14:42So now we can really get whole exomes on
14:45everybody and don't need to focus on it.
14:48Targeted 100 and 248 genes
14:49we're going to be doing
14:517000 additional probands
14:52with whole exome sequencing,
14:54so this is really going to be the largest
14:57sequence congenital disease cohorts.
14:59These patients are also
15:01going to get sniper a data.
15:03And we're hoping that as
15:05sequencing modalities get more
15:07efficient and more cost effective,
15:09and we're able to recruit more
15:11patients across the country,
15:13we're going to be able to expand
15:15this cohort to a large enough size
15:17to really start drawing genotype
15:20phenotype correlations and defining
15:21some of the more complex genetics that
15:24probably underlie the large part of
15:27the congenital heart disease cases.
15:29In addition,
15:29the pediatric Cardiac Genomics
15:31Consortium is involved in developing.
15:33Links between the genomic data and
15:36the electronic medical records,
15:38and this is being done collaboratively
15:41with Mike Murray and a group at
15:46Cincinnati Children's Hospital.
15:48So I want to thank the people
15:50that did all the work,
15:52in particular the patients and
15:55families who generously contributed
15:57their information and their genomic
15:59sample and data to this project
16:01and a special hands up for the
16:04Yale Center for Genome Analysis,
16:05CIAM and Shrikant,
16:06who have been incredibly involved with
16:09this project since its inception.
16:11It was initially driven by
16:13Rick Lifton's idea that, hey,
16:15this has to be a de Novo mechanism.
16:18Ann has evolved significantly since then.
16:21I'm happy to take questions
16:23by email or by chat box.