Loading...
Publication
Genotyping tumor DNA for precision oncology
| Name: | Description: | Size: | Format: | |
|---|---|---|---|---|
| 6.81 MB | Adobe PDF |
Authors
Abstract(s)
The genomic revolution marked the beginning of the 21st century in biology and medicine. Completion of the Human Genome Project in 2003 has brought a flood of technological innovations that enabled decoding the entire genetic information in health and disease, leading to the foundation of the precision medicine movement. Genomics has revolutionized cancer research and transformed our understanding of how cancer arises. In recent years cancers have been re-classified based on their mutations, and a multitude of new drugs were developed that target specific molecular features of the tumor. Accurate and sensitive assays for cancer genotyping are crucial to enable precision oncology. Moreover, in addition to genotyping tissue samples from the primary tumor and metastasis, methodologies for cancer genotyping in circulating blood and other biological fluids are attracting much attention. This project was initiated with the goal of implementing and validating cutting-edge methodological approaches for profiling tumor DNA in the clinic. First, we used Sanger sequencing to determine the frequency of rare mutations in the gene that encodes epidermal growth factor receptor (EGFR) that are not detected by the widely used PCR-based Idylla™ EGFR Mutation Assay. Mutations in the EGFR gene are biomarkers that predict how non-small cell lung cancer (NSCLC) patients respond to EGFR-targeted therapies collectively known as tyrosine kinase inhibitors (TKIs). Thus, EGFR genotyping provides crucial information for treatment decision. Both Sanger sequencing and real-time PCR methodologies are used for EGFR genotyping. However, methods based on real-time PCR have limitations, as they may not detect rare or novel mutations. We sought to determine the prevalence of rare mutations in the tyrosine kinase domain (exons 18 to 21) of the EGFR gene not targeted by the most frequently used real-time PCR approaches, i.e., the cobas® EGFR Mutation Test, and the Idylla™ EGFR Mutation Assay. A total of 1228 NSCLC patients were screened for mutations in exons 18 to 21 of the EGFR gene using Sanger sequencing. We observed that 252 patients (~20%) had at least one mutation in the EGFR gene, and 38 (~3%) carried uncommon genetic alterations that could not be identified by the cobas® or the Idylla™ tests. We further found six new single mutations and seven previously unreported compound mutations. Clinical information and patient outcome are presented for these cases. In conclusion, this study highlights the value of sequencing based approaches to identify rare mutations. Our results add to the inventory of known EGFR mutations, thus contributing to improved lung cancer precision treatment. After the detection of an EGFR-TKI-sensitive mutation, at initial diagnosis, and after treatment with 1st or 2 nd generation inhibitors, about half of tumors develop a resistance mutation. To identify this mutation, so that the treatment can be readjusted, it is mandatory to carefully choose the sample and the method to be used. Therefore, we implemented and assessed the performance of a droplet digital PCR (ddPCR) assay for detecting the EGFR T790M mutation in liquid biopsies. Liquid biopsy allows the identification of targetable cancer mutations in a minimally invasive manner. In patients with NSCLC, ddPCR is increasingly used to genotype the EGFR gene in circulating cell-free DNA (cfDNA). However, the sensitivity of this method is still under debate. We optimized a ddPCR assay and used it to detect the EGFR T790M mutation in plasma samples from 77 patients with NSCLC in progression. Our ddPCR assay enabled the detection and quantification of the EGFR T790M mutation at cfDNA allele frequency as low as 0.5%. The mutation was detected in 40 plasma samples, corresponding to a positivity rate of 52%. The number of mutant molecules per mL of plasma ranged from 1 to 6,000. A re-biopsy was analyzed for 12 patients that had a negative plasma test and the mutation was detected in 2 cases. A second liquid biopsy was performed for 6 patients and the mutation was detected in 3 cases. In conclusion, this study highlights the value of ddPCR to detect and quantify the EGFR T790M mutation in liquid biopsies in a real-world clinical setting. Our results suggest that repeated ddPCR tests in cfDNA may obviate tissue re-biopsy in patients unable to provide a tumor tissue sample suitable for molecular analysis. Despite the advantages above described, the major disadvantage of ddPCR is the difficulty in testing multiple targets simultaneously. As such, we explored massively parallel sequencing methodologies for mutation discovery in cfDNA extracted from circulating blood. We analysed a patient with metastatic breast cancer who was selected for enrolment in an open-label clinical phase IIIb trial with ribociclib combined with letrozole. We genotyped cfDNA isolated from blood samples collected before (Pre_cfDNA) and after (Post_cfDNA) treatment with ribociclib and letrozole. Analysis of Pre_cfDNA using the Memorial Sloan Kettering-Integrated Mutation Profiling of Actionable Cancer Targets (MSK-IMPACT) revealed two alterations: a missense hotspot mutation in the PIK3CA gene (3:178952085, A>G, H1047R) and a copy number amplification of the CCND1 gene. The presence of both molecular alterations was confirmed in the primary tumor. After treatment, the patient presented a significant clinical improvement, and the two alterations were no longer detected in cfDNA (Post_cfDNA). This work underlines that the minimally invasive nature of liquid biopsy allows monitoring of disease progression, demonstrating a practical alternative to tissue biopsy. The use of massively parallel sequencing enables to interrogate of several genes simultaneously, the search for different types of mutation, and identify variants with low allele frequency. Based on the previously mentioned results, we decided to implement a targeted ddPCR assay for detecting PIK3CA mutations in cfDNA. We used this assay to screen urine samples from the patient with a PIK3CA mutation identified in the plasma, but we failed to detect any positive result in the urine. In conclusion, we implemented at GenoMed the main methods that are currently recommended for molecular diagnosis in Precision Oncology, from massively parallel sequencing to ddPCR. We characterized the main advantages and limitations of each technology applied to a variety of biological samples including tumor tissue, blood, and other fluids. Taken together, the results of our work represent an important contribution towards a more achievable and cost-effective Precision Medicine in the context of real world clinical practice.
Description
Keywords
Diagnóstico molecular biópsia líquida ddPCR sequenciação massiva em paralelo cancro do pulmão cancro da mama Molecular diagnosis liquid biopsy massively parallel sequencing lung cancer breast cancer