Combining semiconductor-based sequencing with amplicon-based cancer gene panels: A rapid next-gen approach to clinical cancer genotyping.

agreementkittensSemiconductor

Nov 2, 2013 (3 years and 1 month ago)

185 views

Combining semiconductor
-
based sequencing with
amplicon
-
based
cancer

gene panels: A
rapid next
-
gen approach to clinical cancer genotyping.

Christopher L. Corless, Tanaya Neff, Michael C. Heinrich, Carol
Beadling

Knight Cancer Institute, Oregon Health & Science University, and Portland Veteran's Affairs Medical Center, Portland, Oregon

0
500
1000
1500
2000
2500
3000
1
11
21
31
41
51
61
71
81
91
101
111
121
131
141
151
161
171
181
191
201
211
221
231
241
251
261
271
281
291
301
311
321
331
341
351
361
371
Average number of reads

Amplicon

0%
10%
20%
30%
40%
50%
60%
70%
80%
90%
100%
0
10
20
30
40
50
60
70
0
100
200
300
400
500
600
700
800
900
1000
1100
1200
1300
1400
1500
1600
1700
1800
Number of target regions with read depth

Read Depth

Number of targets within read depth
% of targets ≥ read depth

369/380 = 97% >0.2x mean

Average 695 reads

Background:

Bringing next
-
gen sequencing into clinical (CLIA
-
licensed) laboratories is an important step in the advancement of
personalized cancer care. We have validated a new
semiconductor

based sequencing approach (Ion Torrent PGM)
in combination with
amplicon
-
based library preparations. Using
FFPE
-
derived samples of tumor and normal DNA , the
performance of two panels has been examined in detail: 1) the
AmpliSeq
Cancer
Panel (available from Ion Torrent); 2) a custom
panel designed for genotyping GI
stromal

tumors (GISTs).






Designed to detect hotspot mutations across 46 genes.



Samples tested: 45 FFPE solid tumors previously genotyped on
Sequenom

MassArray

system, including :

-

53 point mutations

-

19 in/
dels

(4


63
bp
)




Input DNA: 10
ng




190 primer pairs (1 tube)




13.3 kb of target sequence




Barcoded

samples




4 samples per 316 chip


Results



On
-
target reads: 95%



Average read length:76
bp



Average read depth: 2000 (range 8


4,948)



All 53 point mutations were identified.




All 19 in/
dels

were visible in the sequence, but most were not
accurately identified by the Torrent Suite software (version 2.0).


Using a cut
-
off of 8% mutant allele, the sensitivity for known
mutations was 100% and the specificity was 92%.


There were 27
new non
-
synonymous mutations with >8% mutant
allele frequency
detected
in
regions
not covered by the mass
spectrometry
-
based
panel, including
24
point mutations and
3
deletions of 2
bp
; all
27 were confirmed by Sanger sequence
analysis.

Conclusions:





Combining solid
-
state sequencing with a highly multiplexed
PCR method for library construction is a rapid (48 hr)
approach for next
-
gen sequencing of clinical cancer samples.



The protocol requires very little DNA (10
-
20
ng
). Indeed,
t
wo
DNA samples from laser
-
captured tumor yielded excellent
data with the AmpliSeq Cancer Panel.




The process is highly scalable and larger, cancer
-
specific
amplicon

panels are in development.



Automated identification of in/
dels

remains a challenge in
next
-
gen sequencing output.



Recent improvements in the library preparation protocol and
in the variant caller software are leading to further reductions
in turn
-
around time and the number of background variants.



Both the Cancer Gene Panel and the GIST Panel are now
available in our CLIA
-
licensed/CAP
-
accredited laboratory.


Acknowledgements:
We are grateful for the help of our
collaborators at Ion Torrent, including Katherine Rhodes,
Michael Thornton, John
Leamon

and Mark Andersen.


ABL1

FGFR3

NRAS

AKT1

FLT3

PDGFRA

ALK

GNAS

PIK3CA

APC

HNF1A

PTEN

ATM

HRAS

PTPN11

BRAF

IDH1

RB1

CDH1

JAK2

RET

CDKN2A

JAK3

SMAD4

CSF1R

KDR

SMARCB1

CTNNB1

KIT

SMO

EGFR

KRAS

SRC

ERBB2

MET

STK11

ERBB4

MLH1

TP53

FBXW7

MPL

VHL

FGFR1

NOTCH1

FGFR2

NPM1

AKT1

KRAS

SDHA

AKT2

MAP2K1

SDHAF1

AKT3

NF1

SDHAF2

ATM

NRAS

SDHB

BRAF

PDGFRA

SDHC

CDKN2A

PIK3CA

SDHD

HRAS

PTEN

TP53

KIT

PTPN11

AmpliSeq Cancer Gene Panel (Ion Torrent)

Reads per
amplicon

(Mean + standard dev.)

Reads per
amplicon

(Mean + standard dev.)

Comparison of allele ratios:

MassArray

vs

AmpliSeq

53 Point Mutations

R=0.92, P<0.0001

Custom GI
Stromal

Tumor (GIST) Panel

AmpliSeq Cancer Gene Panel (Ion Torrent)




Input DNA: 20
ng




477 primer pairs (2 tubes)




Amplicons

tiled for


whole gene coverage




125 to 175
bp

amplicons




49.1 kb of target sequence




Barcoded

samples



Primers were designed with the assistance of collaborators at
Ion Torrent using the Ion AmpliSeq designer software v1.0.



Samples tested: 19 FFPE GISTs that were previously shown to
be negative for KIT and PDGFRA mutations.



8
barcoded

samples run on a single 318 chip, yielding an
average of 337,000 reads per sample



11
barcoded

samples run on 316 chips (@ 4 per chip), yielding
an average of 392,000 reads per sample


Results



On
-
target reads: 88%



Average read length: 86
bp



Average read depth: 695 (range 11


1,848)



An NRAS Q61L mutation was confirmed in one case; this
mutation is novel in GIST

Knight Diagnostic Laboratories