Emma Tinkler-Hundal

Cross-laboratory validation of the OncoScan® FFPE Assay, a multiplex tool for whole genome tumour profiling

BackgroundAdoption of new technology in both basic research and clinical settings requires rigorous validation of analytical performance. The OncoScan® FFPE Assay is a multiplexing tool that offers genome-wide copy number and loss of heterozygosity detection, as well as identification of frequently tested somatic mutations.MethodsIn this study, 162 formalin fixed paraffin embedded samples, representing six different tumour types, were profiled in triplicate across three independent laboratories. OncoScan® formalin fixed paraffin embedded assay data was then analysed for reproducibility of genome-wide copy number, loss of heterozygosity and somatic mutations. Where available, somatic mutation data was compared to data from orthogonal technologies (pyro/sanger sequencing).ResultsCross site comparisons of genome-wide copy number and loss of heterozygosity profiles showed greater than 95% average agreement between sites. Somatic mutations pre-validated by orthogonal technologies showed greater than 90% agreement with OncoScan® somatic mutation calls and somatic mutation concordance between sites averaged 97%.ConclusionsReproducibility of whole-genome copy number, loss of heterozygosity and somatic mutation data using the OncoScan® assay has been demonstrated with comparatively low DNA inputs from a range of highly degraded formalin fixed paraffin embedded samples. In addition, our data shows examples of clinically-relevant aberrations that demonstrate the potential utility of the OncoScan® assay as a robust clinical tool for guiding tumour therapy.

Intratumoral stromal morphometry predicts disease recurrence but not response to 5-fluorouracil-results from the QUASAR trial of colorectal cancer.

The biological importance of tumour‐associated stroma is becoming increasingly apparent, but its clinical utility remains ill‐defined. For stage II/Dukes B colorectal cancer (CRC), clinical biomarkers are urgently required to direct therapeutic options. We report here prognostic/predictive analyses, and molecular associations, of stromal morphometric quantification in the Quick and Simple and Reliable (QUASAR) trial of CRC.

Preoperative chemoradiation with capecitabine, irinotecan and cetuximab in rectal cancer: significance of pre-treatment and post-resection RAS mutations

Background:The influence of EGFR pathway mutations on cetuximab-containing rectal cancer preoperative chemoradiation (CRT) is uncertain.Methods:In a prospective phase II trial (EXCITE), patients with magnetic resonance imaging (MRI)-defined non-metastatic rectal adenocarinoma threatening/involving the surgical resection plane received pelvic radiotherapy with concurrent capecitabine, irinotecan and cetuximab. Resection was recommended 8 weeks later. The primary endpoint was histopathologically clear (R0) resection margin. Pre-planned retrospective DNA pyrosequencing (PS) and next generation sequencing (NGS) of KRAS, NRAS, PIK3CA and BRAF was performed on the pre-treatment biopsy and resected specimen.Results:Eighty-two patients were recruited and 76 underwent surgery, with R0 resection in 67 (82%, 90%CI: 73–88%) (four patients with clinical complete response declined surgery). Twenty–four patients (30%) had an excellent clinical or pathological response (ECPR). Using NGS 24 (46%) of 52 matched biopsies/resections were discrepant: ten patients (19%) gained 13 new resection mutations compared to biopsy (12 KRAS, one PIK3CA) and 18 (35%) lost 22 mutations (15 KRAS, 7 PIK3CA). Tumours only ever testing RAS wild-type had significantly greater ECPR than tumours with either biopsy or resection RAS mutations (14/29 [48%] vs 10/51 [20%], P=0.008), with a trend towards increased overall survival (HR 0.23, 95% CI 0.05–1.03, P=0.055).Conclusions:This regimen was feasible and the primary study endpoint was met. For the first time using pre-operative rectal CRT, emergence of clinically important new resection mutations is described, likely reflecting intratumoural heterogeneity manifesting either as treatment-driven selective clonal expansion or a geographical biopsy sampling miss.

The Effect of a Multidisciplinary Regional Educational Programme on the Quality of Colon Cancer Resection

Mesocolic plane surgery with central vascular ligation produces an oncologically superior specimen following colon cancer resection and appears to be related to optimal outcomes. We aimed to assess whether a regional educational programme in optimal mesocolic surgery led to an improvement in the quality of specimens.

Comparing mutation calls in fixed tumour samples between the affymetrix OncoScan® array and PCR based next-generation sequencing

BackgroundThe importance of accurate and affordable mutation calling in fixed pathology samples is becoming increasingly important as we move into the era of personalised medicine. The Affymetrix OncoScan® Array platform is designed to produce actionable mutation calls in archival material.MethodsWe compared calls made using the OncoScan platform with calls made using a custom designed PCR panel followed by next-generation sequencing (NGS), in order to benchmark the sensitivity and specificity of the OncoScan calls in a large cohort of fixed tumour samples. 392 fixed, clinical samples were sequenced, encompassing 641 PCR regions, 403 putative positive calls and 1528 putative negative calls.ResultsA small number of mutations could not be validated, either due to large indels or pseudogenes impairing parts of the NGS pipeline. For the remainder, if calls were filtered according to simple quality metrics, both sensitivity and specificity for the OncoScan platform were over 98%. This applied even to samples with poorer sample quality and lower variant allele frequency (5–10%) than product claims indicated.ConclusionsThis benchmarking study will be useful to users and potential users of this platform, who wish to compare technologies or interpret their own results.

Abstract 626: Cross-site reproducibility and orthogonal validation of copy number and somatic mutation calls of OncoScan® FFPE Assay Kit in solid tumors

Proceedings: AACR 106th Annual Meeting 2015; April 18-22, 2015; Philadelphia, PA Objectives Copy number (CN) and somatic mutation (SM) analysis in tumors is rapidly gaining importance in cancer management as a tool for differential diagnosis, determination of prognosis, and selection of therapeutic. Genome-wide copy number and LOH detection as well as a panel of frequently tested somatic mutations can be detected with OncoScan® FFPE Assay Kit. We report on a validation study of OncoScan FFPE Assay Kit CN and SM data by orthogonal technologies (FISH and NGS, respectively) to estimate the sensitivity and specificity parameters of the platform. In addition, we assessed the reproducibility of the platform across three different sites in the UK: Leeds Institute of Cancer and Pathology (LICP), West Midlands Regional Genetics Laboratory (WMRGL), and Almac Diagnostics (Almac). Methods Validation of CN data was performed on a cohort of cancer samples identified as Her2 positive/ambiguous by FISH. The panel of SMs available was validated by a custom targeted amplicon NGS panel on a diverse collection of samples derived from multiple cohorts across several tumor types sourced from both LICP and WMRGL. For the reproducibility study, 162 samples encompassing six different tumor types (breast, colorectal, lung, melanoma, prostate, and ovarian) were collected from LICP and WMRGL. DNA was extracted and plated in triplicate and distributed to the three testing sites: LICP, WMRGL, and Almac. Data from all sites was analyzed for reproducibility of genome-wide CN/LOH calls and SM calls. Results Cross-site comparisons of genome-wide CN and LOH profiles on 162 FFPE solid tumor samples showed greater than 95% average agreement between three sites (LICP, WMGRL, and Almac), while SM classification concordance between the sites averaged 97%. Initial orthogonal validation of Her2 amplification by FISH showed greater than 90% concordance, as did initial test samples used for validating OncoScan SM calls by a targeted amplicon NGS panel. Conclusion In this study we validated both CN and SM calls using OncoScan FFPE Assay Kit and demonstrated a high degree of agreement with orthogonal methods in all aspects. Reproducibility of whole-genome CN, LOH, and SM data using OncoScan FFPE Assay Kit has also been demonstrated for a range of FFPE samples, including highly degraded samples. This study is a step forward in evaluating the potential clinical utility of a platform combining genome-wide copy number and somatic mutation calls within the national health service of the UK. Citation Format: Joseph M. Foster, Assa Oumie, Fiona S. Togneri, Morag Taylor, Sofia Alyas, Paula Wojtowicz, Henry Wood, Emma Tinkler-Hundal, Katie Southward, Dominic McMullan, Phil Quirke, Katherine E. Keating, Mike Griffiths, Karen G. Spink, Fiona Brew, Eric Fung, Jeanette Schmidt. Cross-site reproducibility and orthogonal validation of copy number and somatic mutation calls of OncoScan® FFPE Assay Kit in solid tumors. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 626. doi:10.1158/1538-7445.AM2015-626