Kelly Li, Devin Do, Patricia Hegerich, Nivedita Ma jumdar, David Keys, Stephen Jackson, Junko Stevens and Caifu Chen Life Science Solutions Division, Life Technologies , A part of Thermo Fisher Scientific, 180 Oyster Point Blvd, South San Francisco, CA 94080

RESULTS

Figure 1. QuantStudio TM 3D Digital PCR System

Fig. 3 demonstrates high sensitivity of 5% or even lower of the dPCR system in differentially detecting CN in heterogeneous samples. Two samples from Coriell (NA 17258 and NA 17132) have known 2 or 3 copy number for CCL3L1 gene. The 2 samples with CN of 2 and 3 are mixed at different proportion to generate the predicted CN at 0.1 incremental changes between CN 2 and 3 ( like 2.0. 2.1, 2.2…3.0). CN for the mixed samples is calculated from absolute quantitation reported by the QS3D Analysis Suite. Measured CN is plotted against predicted CN and the legends are the same as Fig. 2.

Figure 4. Testing TaqMan Copy Number (CNV) Assay fo r HER2 Gene by qPCR

HER2 gene amplification was analyzed for 43 breast cancer FFPE samples using QuantStudioTM 3D dPCRsystem. HER2 copy number was determined by ratio of absolute quantitation of FAM-labeled HER2 target and VIC-labeled RNaseP reference exported from QS3D Analysis Suite software. Each sample was also analyzed by SISH. For SISH, HER2 gene amplification was defined as positive(P) or negative (N) according to ASCO/CAP 2007 guideline (HER2/C17 ratio >2.2 as positive or <1.8 as negative, in between as equivocal). The high concordance between the two completely different technologies (dPCR and SISH) is shown. The red line is drawn roughly CN=4.4 corresponding to the cutoff of HER2/C17 ratio (2.2) for calling positive in SISH.

Figure 6. HER2 Copy Number Analysis of Breast Canc er FFPE Samples by dPCR and SISH

ABSTRACT

Human epidermal growth factor receptor 2 (HER2) gene amplification (increasing copy number), not only is a strong indicator for tumor progression/poor prognosis but also presents as an effective drug target. Accurate measurement of HER2 copy number (CN) is so critical that it determines patient qualification for HER2-targeted therapy with Trastuzumab (Herceptin) for early or advanced breast cancer. Currently, in situ hybridization-based and immunohistochemical analysis are commonly used, but they can be subjective, lengthy, labor intensive and less quantitative, especially for samples whose HER2/CEP17 ratio fall into the equivocal range (1.8-2.2). Life Technologies offers TaqMan® Copy Number Assays for copy number analysis on real-time qPCR platform, but qPCR also becomes less quantitative for CN calls reach higher than 4, due to its relative quantitation (RQ) methodology. Recently launched QuantStudioTM 3D (QS3D) Digital PCR System provides a new platform with absolute quantification capability to address these challenges. We have developed a copy number application on the dPCR system and we can calculate copy number based on absolute quantitation exported from QS3D Analysis Suite software. We demonstrate that the system not only can quantify gene copy numbers from zero to eight with high accuracy and precision but also is able to differentially detect heterogeneous samples with 5% resolution. We also present our technical assessment for quantifying HER2 copy number using breast cancer FFPE samples and our data show high concordance with results from SISH method. As a robust platform, the dPCR system not only leverages our existing TaqMan® CNV assays but also enables customers to measure higher CN with absolute accuracy, simple workflow and fast turnaround time.

Accurate Quantification of HER2 Gene Amplification Using QuantStudio TM 3D Digital PCR System

Sealed Consumables

3. Amplify2. Load 4. Read1. Mix

• Sealed consumables• Limited hands-on• Minimal sample loss

• Interrogated volume similar to real-time PCR

• Hexagonal packing enables 20,000 wells / 10×10 mm2 chip

• Each reaction well is isolated from its neighbors

B. QuantStudio™ 3D Digital PCR 20K Chip

A. Workflow

60 µµµµm

AACR 2014Poster #1521

Figure 3. High Sensitivity of dPCR in Quantifying C opy Number in Heterogeneous Samples

0.5

1

2

4

8

16

32

0.8 1.1 1.1 1.4 1.4 1.5 1.6 1.6 1.7 1.7 1.8 1.8 1.9 2.0 2.0 2.1 2.1 2.2 2.2 2.3 2.4 2.5 2.7 2.8 3.0 3.2 3.5 4.7 5.0 5.2 5.6 6.8 6.9 7.3 7.4 8.2 9.7 9.8 10.521.022.424.631.1

N N N N N N N N N N N N N N N N N N N N N N N N N N P P P P P P P P P P P P P P P P P

Co

py

Nu

mb

er

by

dP

CR

(lo

g2)*

*

SISH

CN by dPCR

Fig.2A The limitation of qPCR for CNV detection is that it measurement lose accuracy when copy number is above 4 due to its relative quantitation method. We demonstrate here that the dPCR can accurately measure copy number from 0 to 8 for CCL3L1 gene using nine gDNA samples with known copy number for the target. For each sample, 6-8 replicates were run and their standard deviation is shown in the error bars. The CV for each sample is within the range of 0.1-0.25%.

Fig.2B shows high precision of copy number (CN) measurement from individual replicate chips. Measured CN is plotted against expected CN for each of a given sample (shown in Fig.2A). The blue dots are CN from individual chips and red cross is mean of all the measurement for the given sample. The grey shaded rectangular bars are their standard deviations, and the dotted line are 99% prediction interval. Copy number is calculated from absolute quantitation reported by QS3D Analysis Suite .

Figure 5 HER2 Copy Number Determination Using FFPE Samples and Two Distinct Copy Number Reference Assays with dPCR

INTRODUCTIONGene amplification is a common genetic abnormality in many types of cancers and has been implicated in playing important roles in cancer development. HER2 gene amplification occurring in about 10-30% of breast cancer cases, is strongly associated with aggressive tumor progression and poor prognosis, and also an effective therapeutic target. Breast cancer patients with HER2+ are eligible for Herceptin treatment. Therefore, accurately determining HER2 gene amplification is essential yet challenging for cancer researchers and clinicians. Currently, IHC (immunohistochemical analysis), FISH (fluorescence in situ hybridization) or SISH (silver in situ hybridization) are routinely used. However, the challenge is that these methods can be laborious and less quantitative, especially when CN falls into equivocal range. In the study, we aim to 1). demonstrate the utility of QuantStudio™ 3D Digital PCR for quantifying DNA copy number with high precision, accuracy, resolution, 2). show its application in measuring HER2 gene amplification in breast cancer FFPE samples, and its high concordance with current method (SISH ).

MATERIALS AND METHODSThe workflow for QuantStudio™ 3D Digital PCR System and its 20K chip is shown in Figure 1. For copy number analysis, 14.5 ul volume of dPCR reaction mix, including QS3D dPCR master mix, a FAM-labeled TaqMan® CNV assay for target of interest and VIC-labeled RNaseP reference assay (P/N 4400294) and gDNA, is loaded onto a chip. The amount of gDNA used usually targets range of 200-2,000 copies/ul on the chip and adjustment for input DNA may be required depending gene copy number to be detected. PCR thermalcycling condition and other details can be found in QS3D User Guide. After PCR, chips are read on QS3D dPCR Instrument and based on absolute quantitation exported from QS3D Analysis Suite software, copy number is calculated by 2 times the ratio of FAM-labeled target and VIC-labeled RNaseP reference.

Regular gDNA samples are purchased from Coriell Repository. TaqMan® CNV assays for CCL3L1 (Hs03198166_cn) and for HER2 (Hs00817646_cn) are from Life Technologies. For qPCR, FAM-labeled TaqMan CNV assay for target of interest was duplexed with VIC-labeled RNaseP CN reference assay, and copy number is determined by a relative quantitation (RQ) based-∆∆Ct method, as described in TaqMan® CNV assay protocol.

CONCLUSIONS

• The QuantStudio™ 3D Digital PCR System provides a sensitive, accurate, robust technology with a simple workflow for copy number analysis. Copy number can be calculated based the absolute quantitation exported from QS3D Analysis Suite software. It is a reliable platform for measuring gene amplification and copy number variation (CNV).

• We can accurately measure copy number from 0 to 8 with high precision (with CV of 1.0%-2.5%) using CCL3L1 as a model gene and the samples with known copy number for the gene.

• We also demonstrate the ability of QS3D to differentially detect copy number differences in heterogeneous samples with high sensitivity of 5% or even lower.

• We have assessed HER2 gene amplification in 43 breast cancer FFPE samples and high concordance is achieved compared to SISH results from the same set of samples.

ACKNOWLEDGEMENTS

We thank Bruno Ping, Molecular Diagnostics Department at Royal Surrey County Hospital, Guildford, United Kingdom for sharing the breast cancer FFPE samples and their SISH data.

TRADEMARKS/LICESING© 2014 Thermo Fisher Scientific Inc. All rights reserved. All trademarks are the property of Thermo Fisher Scientific and its subsidiaries unless otherwise specified. TaqMan is a registered trademark of Roche Molecular Systems, Inc., used under permission and license.

Thermo Fisher Scientific • 5791 Van Allen Way • Carlsbad, CA 92008 • www.lifetechnologies.com

0

1

2

3

4

5

6

7

8

9

Co

py

Nu

mb

er

Samples

0

1

2

3

4

NA

1085

9_1

NA

1085

9_2

NA

1085

9_3

NA

1085

9_4

NA

1447

4N

A14

476

NA

1710

2N

A17

103

NA

1710

4N

A17

105

NA

1710

6N

A17

107

NA

1710

8N

A17

109

NA

1711

0N

A17

111

NA

1711

2N

A17

113

NA

1711

4N

A17

115

NA

1711

6N

A17

117

NA

1711

8N

A17

119

NA

1712

0N

A17

121

NA

1712

2N

A17

123

NA

1712

4N

A17

125

NA

1712

6N

A17

127

NA

1712

8N

A17

129

NA

1713

0N

A17

131

NA

1713

2N

A17

134

NA

1713

6N

A17

137

NA

1713

9N

A17

140

NA

1714

4N

A17

147

NA

1714

8N

A17

149

NA

1715

5N

A17

188

NA

1719

4N

A17

201

NA

1720

2N

A17

203

NA

1720

4N

A17

205

NA

1720

6N

A17

207

NA

1720

8N

A17

209

NA

1721

0N

A17

211

NA

1721

2N

A17

213

NA

1721

4N

A17

215

NA

1721

6N

A17

217

NA

1722

0N

A17

221

NA

1722

3N

A17

225

NA

1722

6N

A17

227

NA

1722

8N

A17

230

NA

1723

1N

A17

232

NA

1723

5N

A17

237

NA

1723

9N

A17

240

NA

1724

1N

A17

242

NA

1724

5N

A17

247

NA

1725

1N

A17

253

NA

1725

4N

A17

255

NA

1725

8N

A17

259

NA

1726

0N

A17

261

NA

1726

2N

A17

263

Co

py N

um

ber

gDNA

0

1

2

3

4

5

6

30 11 6 14 10 7 3-2 3-1 19-1 19-2

CN_using RNaseP Ref

CN_using TERT Ref

To ensure high performance of TaqMan CNV assay for HER2, we first tested it with qPCR using a panel of 94 gDNAs from Coriell. Four replicates were run for each sample. HER2 copy number was determined by a relative quantitation (RQ) based-∆∆Ct method. All the samples are expected to have 2 copies of the gene and the HER2 CNV assay shows the robust performance cross the gDNA panel.

It is a concern of using a single copy number reference assay with cancer samples due to cancer genome instability. We tested HER2 copy number by dPCR with randomly selected 8 breast cancer FFPE samples using both RNaseP and TERT CN reference assays. Two of the samples (#3 and #19) were run in duplicate chips. **CN is calculated based on absolute quantitation reported by QS3D Analysis Suite.

Figure 2. QuantStudio TM 3D Digital PCR Can Accurately Quantify Higher Copy Number

Fig. 1A illustrates the workflow of QuantStudioTM 3D (QS3D) dPCR system. The steps are simple and straightforward with minimal hands-on requirement. The system consists of an autoloader, a modified thermocycler 9700 and a reader (QS3D dPCR Instrument). Copy number is calculated from absolute quantitation reported by the cloud-based QS3D Analysis Suite software .Fig. 1B shows a 20K chip used in the dPCR system. PCR reaction mix (volume of 14.5 ul) is loaded onto a chip (roughly 0.8nl/well).

A

B

Cop

y N

umbe

r**

Samples