Abstract

The inevitable switch from standard molecular methods to next-generation sequencing for the molecular profiling of tumors is challenging for most diagnostic laboratories. However, fixed validation criteria for diagnostic accreditation are not in place because of the great variability in methods and aims. Here, we describe the validation of a custom panel of hotspots in 24 genes for the detection of somatic mutations in non-small cell lung carcinoma, colorectal carcinoma and malignant melanoma starting from FFPE sections, using 14, 36 and 5 cases, respectively. The targeted hotspots were selected for their present or future clinical relevance in solid tumor types. The target regions were enriched with the TruSeq approach starting from limited amounts of DNA. Cost effective sequencing of 12 pooled libraries was done using a micro flow cell on the MiSeq and subsequent data analysis with MiSeqReporter and VariantStudio. The entire workflow was diagnostically validated showing a robust performance with maximal sensitivity and specificity using as thresholds a variant allele frequency >5% and a minimal amplicon coverage of 300. We implemented this method through the analysis of 150 routine diagnostic samples and identified clinically relevant mutations in 16 genes including KRAS (32%), TP53 (32%), BRAF (12%), APC (11%), EGFR (8%) and NRAS (5%). Importantly, the highest success rate was obtained when using also the low quality DNA samples. In conclusion, we provide a workflow for the validation of targeted NGS by a custom-designed pan-solid tumor panel in a molecular diagnostic lab and demonstrate its robustness in a clinical setting.

Highlights

  • Targeted therapies for solid tumors have shown great promise and based on ongoing clinical studies, the potential to enlarge the current set of approved anti-cancer drugs is huge [1,2]

  • We performed the validation of the targeted next-generation sequencing (NGS) workflow on 55 Formalin-fixed paraffin-embedded (FFPE) samples (14 non-small cell lung carcinoma (NSCLC), 36 CRC, 5 MELA) as well as the Multiplex reference sample

  • Key factors determining our gene selection were 1) the present or likely future clinical significance in terms of therapeutic, prognostic or diagnostic value, both from clinical research and pre-clinical data; 2) the frequency of known hotspot mutations within the types of tumors predominantly screened in the hospital; and 3) the cost per sample, taking into account the number of samples weekly screened and the desired depth of coverage to allow for sufficient sensitivity

Read more

Summary

Introduction

Targeted therapies for solid tumors have shown great promise and based on ongoing clinical studies, the potential to enlarge the current set of approved anti-cancer drugs is huge [1,2]. Most diagnostic labs are applying standard molecular technologies (qPCR, melt curve analysis, Sanger and pyrosequencing, etc.) to screen solid tumor samples for the presence of a selected number of actionable hotspot mutations, mainly in BRAF, EGFR, KRAS, NRAS, KIT and PIK3CA. These methods are well installed and highly reliable, they require a significant amount of DNA since the analysis of each gene needs a specific assay. The standard assays are currently being replaced by multiplex parallel analysis i.e. next-generation sequencing (NGS)

Objectives
Methods
Results
Discussion
Conclusion
Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call