Embed Size (px)
Citation preview
Lessons from 6000 clinical exomes
Han G. Brunner
The first whole genomes James Watson Craig Venter
Nature 2008
Nature 2008
PloS Biology 2007
Nature 2008
Marjolein Kriek
Ozzy Osbourne genome sequencing project
Jim Lupski
TEDMED 2010
Ozzy Osbourne
Nijmegen Clinical Exome Sequencing 2011-‐ 2015
Clinical Exome Requests 2013-‐2014 increase
500 Exomes/ month in 2015
All individuals must agree with the entire procedure
All individuals must understand the possibility of incidental findings
and agree to be informed.
Data will be stored and may be shared with other researchers
Informed consent
Full exome or targeted analysis?
Most families with severe or progressive disease choose full exome e.g. severe ID in child Some paTents with milder or stable condiTons choose targeted analysis
e.g. deafness in adult
Three reasons why DiagnosTc Next GeneraTon Sequencing of all genes (exome) is becoming part of mainstream medicine: • 1500 Euros is not prohibiTve, and we are cuZng other costs • We are not as good at clinical diagnosis as we would like to think • Process not more difficult to handle than current diagnosTc strategies
Misdiagnosis Missed diagnosis Confusion Harm
Number of paHents with dedicated gene tests: 78 of 100 Total dedicated gene tests: 233 Number of different genes: 56 Range per paHent: 0-‐13 genes Also: 100 SNP microarrays
79 metabolic screens 34 brain MRIs 4 cerebral CT scans
0
100
200
300
400
500
600
700
1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4
2013 2014 2015
# of te
sts Sanger vs Exomes
Sanger
Exomes
Next generaTon sequencing is a tool for diagnosis and differenTal diagnosis Accross clinical specialiTes
The hit rate of diagnosTc exome sequencing (including CNVs) Clearly Exceeds The current diagnosTc standard of clinician driven invesTgaTon
An inconvenient truth
Targeted analysis across 5 diagnosTc groups
Compare with clinical diagnosTc pracTce 2011
Neveling et al. Human Muta:on 2013
Severe intellectual disability, auTsm
Lessons from doing exomes
-‐ Midface hypoplasia -‐ Broad nasal bridge One abnormal ear
Unilateral deafness Unilateral facial palsy
CHD7 de novo mutaHon > CHARGE syndrome!
Severe intellectual disability, auTsm -‐ Midface hypoplasia -‐ Broad nasal bridge One abnormal ear
Unilateral deafness Unilateral facial palsy
Doctors are great; exomes are beVer
Lessons
Doctors are good; Exomes are beeer In diagnosTc exomes / genomes:
Diagnosis = Discovery InterpreTng noncoding informaTon is sTll hard For rare diseases matchmaking is key!
Genomes make us rethink nosology
Lessons we learned
A new geneHc cause of intellectual disability
Exome sequencing 20.000 genes of Ender and his parents
Shows that Ender has just 1 de novo mutaTon. This mutaTon is in the PACS1 gene
Exome sequencing 20.000 genes of Siebe and his parents
Shows that Siebe has just 2 de novo mutaTons. One mutaTon is in the PACS1 gene
New genes
Exome sequencing 20.000 genes of Ender and his parents
Shows that Ender has just 1 de novo mutaTon. This mutaTon is in the PACS1 gene
Exome sequencing 20.000 genes of Siebe and his parents
Shows that Siebe has just 2 de novo mutaTons. One mutaTon is in the PACS1 gene
And they have the same PACS1 mutaTon
Schuurs-‐Hoeijmakers et al. Am J Hum Genet 2012
More paHents with PACS1 Arg203Trp mutaHon
More matchmaking
In less than 1 year: 14 new paTents worldwide All were found by exome
Got connected because: 6: The doctor found the arTcle 8: The mother found the Facebook group!
Schuurs-‐Hoeijmakers et al. Am J Med Genet 2015 in press
Challenge
We need beeer matchmaking for rare diseases A system where: paTents with a mutaTon in a gene can be compared and linked with other paTents with the same disease and gene
Databases
Matchmaker exchange
IRDiRC, ClinGen and GA4GH.
Rehm et al. Hum MutaTon 2015 in press
Gilissen et al. Nature 2014 De Ligt et al. New England Journal of Medicine 2012
De Vries et al. American Journal Human Gene:cs 2006
New GeneHcTechnologies Elucidate Intellectual Disability
-‐ Reanalysis of data 2012-‐2014 Beeer mapping algorithms Candidate genes confirmed Beeer variant interpretaTon
* 27% Diagnosis by 2014 -‐ Whole Genome sequencing Gilissen et al. Nature 2014 * 40% Diagnosis mostly coding and structural variaTon
Why do things get beVer over Hme?
De Ligt et al NEJM 2012 -‐ Trio exome sequencing in 100 paTents with unexplained ID * 16% Diagnosis
-‐ Reanalysis of data 2012-‐2014 Beeer mapping algorithms Candidate genes confirmed Beeer variant interpretaTon
* 27% Diagnosis by 2014 -‐ Whole Genome sequencing Gilissen et al. Nature 2014 * 40% Diagnosis mostly coding and structural variaTon
Nothing non-‐coding that we understand!
Why do things get beVer over Hme?
De Ligt et al NEJM 2012 -‐ Trio exome sequencing in 100 paTents with unexplained ID * 16% Diagnosis
Majority of new candidate ID genes already confirmed
YY1
DYNC1H1
DEAF1
CIC
ZMYM6
GRIA1
PHIP
WAC
MIB1
PPP2R5D
KIF5C
COL4A3BP
EEF1A2
MYTL1
CAMIIKG
ASH1L
PSMA7 RAPGEF1
PROX2
LRP1
TANC2
TNPO2
PHACTR
MTF1
GATAD2B
CTNNB1
AddiTonal mutaTons No addiTonal mutaTons
DDX3X
820 ID trios support 8 new genes for ID
Lelieveld et al under review
TLK2 is a new ID gene TLK2 mutaTons previously observed in Schizophrenia
DYRK1A SCN2A GRIN2B SYNGAP1 CDH8 PTEN TBL1XR1 TBR1 CTTNB1 ARID1B GPR98 KIRREL TCF4 NRXN1 SHANK2 SHANK3 MECP2 ANK3
ARID1B SYNGAP1 CTNNB1 TCF4 STXBP1 TLK2 DEAF1 MECP2 DYRK1A SCN2A ANK3 STXBP1
SYNGAP1 SCN2A SCN1A ARX CDKL5 PCDH19 MAGI2 DLG1 DLG2
Severe Intellectual Disability 60%?
AuTsm 5-‐20%
Severe Epilepsy 20%
70%
30%
100% 50%
??
??
De novo genes
An inconvenient truth: Genes do not respect diagnosTc boundaries
Schizophrenia 10%? ANK3 DLG1 TLK2 MAGI2 DLG2
??
>60% of Intellectual disability is by de novo gene mutaHons
Gilissen et al. Nature, 17 July 2014
No diagnosis
De novo SNVs
De novo SVs
Inherited 2%
De novo 60%
~1000 / 20.000 genes (5%) are crucial for the brain
Another inconvenient truth
We cannot prevent de novo mutaHons
So should we offer prenatal tesTng to everyone?
An inconvenient truth
We cannot prevent de novo mutaHons
Non invasive prenatal tesTng NIPT
So should we offer prenatal tesTng to everyone?
An inconvenient truth
We cannot prevent de novo mutaHons
Non invasive prenatal tesTng NIPT
Why offer this for Down syndrome (risk 1/1000) and not for other forms of ID (risk 5/1000)?
Another inconvenient truth
There is one known risk factor for de novo mutaHons
Kong et al. Nature 2011
Gilissen et al. Nature, 17 July 2014
A quesHon to ponder • Should we offer prenatal diagnosis to everyone?
• Should boys freeze their sperm at 17?
Doctors are good; Exomes are beeer In diagnosTc exomes / genomes:
Diagnosis = Discovery InterpreTng noncoding informaTon is hard For rare diseases matchmaking is key!
Genomes make us rethink nosology
Lessons we learned
Acknowledgements Willy Nillesen Dorien Lugtenberg Helger Ijntema Rolph Pfundt Marjolein Willemsen Bregje van Bon Tjitske Kleefstra Bert de Vries Lisenka Vissers Charloee Ockeloen Tjitske Kleefstra Anneke Vulto – van Silqout
Human GeneWcs ChrisTan Gilissen Jayne Hehir-‐Kwa Joep de Ligt Alex Hoischen Djie Tjwan Thung Maartje van de Vorst Petra de Vries Michael Kwint Irene Janssen Marloes Steehouwer Tan Bo Joris Veltman Complete Genomics Rick Tearle Robert Klein Richard Leach
Washington University, Sea\le Heather Mefford, Gemma Carvill Evan Eichler, Bradley Coe
Many clinical collaborators
Lessons from 6000 clinical exomes
Han G. Brunner
