Virtual Library
Start Your Search
M. Doherty
Author of
-
+
P2.02 - Biology/Pathology (ID 616)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Biology/Pathology
- Presentations: 1
- Moderators:
- Coordinates: 10/17/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
-
+
P2.02-052 - A Clinically-Validated Universal Companion Diagnostic Platform for Cancer Patient Care (ID 8212)
09:30 - 09:30 | Author(s): M. Doherty
- Abstract
Background:
The increase in targeted therapies and associated companion diagnostics (CDx) has led to the need for efficient determination of therapeutic eligibility from a single assay. Comprehensive genomic profiling (CGP) provides a solution, but due to the complexity and number of assays available today, standardization of validation has become critically important. We present here the first NGS-based universal CDx platform developed and performed in compliance with FDA 21 CFR part 820. The assay interrogates 324 genes, and is anticipated initially to have eight CDx indications (Table 1). The versatile assay design will facilitate streamlined development of future CDx indications.
Method:
DNA extracted from FFPE tumor tissue underwent whole-genome shotgun library construction and hybridization-based capture, followed by sequencing using Illumina HiSeq 4000. Sequence data were processed using a proprietary analysis pipeline designed to detect base substitutions, indels, copy number alterations, genomic rearrangements, microsatellite instability (MSI), and tumor mutational burden (TMB).
Result:
Concordance with FDA-approved CDx are shown in Table 1. Clinical validity was established such that the concordance between CGP and approved CDx were statistically non-inferior to that of two runs of approved CDx. For analytical validity, limit of detection (LoD) was at allele frequency 4% for known substitutions and indels. LoD was 16% tumor content for copy number amplifications, 30% for homozygous deletions, 11% for genomic rearrangements, 12% for MSI, and estimated 20% for TMB. Positive percent agreement (PPA) with an orthogonal NGS platform was 95.8% in substitutions and indels. PPA with FoundationOne was 98.3% across all variant types. Within-assay reproducibility was measured with PPA 99.4%.Companion diagnostic Indicated use PPA Comparator CDx assay EGFR exon 19 deletions and L858R erlotinib, afatinib or gefitinib in NSCLC 98.1% (106/108) cobas® EGFR Mutation Test v2 EGFR T790M osimertinib in NSCLC 98.9% (87/88) cobas® EGFR Mutation Test v2 ALK rearrangements crizotinib in NSCLC 92.9% (78/84) Ventana ALK (D5F3) CDx Assay Vysis ALK Break-Apart FISH KRAS cetuximab or panitumumab in CRC 100% (173/173) therascreen® KRAS PCR ERBB2 (HER2) Amplifications trastuzumab, pertuzumab and ado-trastuzumab-emtansine in breast and gastric cancer 89.4% (101/113) Dako HER2 FISH PharmDx® BRAF V600E/K vemurafenib, dabrafenib, trametinib in melanoma 99.4% (166/167) cobas® BRAF V600 PCR BRCA1/2 rucaparib in ovarian cancer N/A N/A: Novel CDx that was previously validated in PMA P160018
Conclusion:
Rapid expansion of targeted therapies and CDx has necessitated a new approach and urgency to defining performance standards. We developed a universal CDx assay and established a robust approach for demonstrating clinical and analytical validity to support and accelerate the use of CGP for routine clinical care.
-
+
P3.02 - Biology/Pathology (ID 620)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Biology/Pathology
- Presentations: 2
- Moderators:
- Coordinates: 10/18/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
-
+
P3.02-061 - An ALK Follow-On Companion Diagnostic Using CGP for Clinical Care of Patients with NSCLC (ID 10396)
09:30 - 09:30 | Author(s): M. Doherty
- Abstract
Background:
Advanced NSCLC patients may benefit from treatment with ALK inhibitors if they harbor ALK rearrangements such as an EML4-ALK fusion. While the FDA has approved companion diagnostics (CDx) using IHC and FISH-based assays for ALK, molecular diagnostic testing in NSCLC is rapidly evolving towards comprehensive genomic profiling (CGP) to test for a growing number of established predictive biomarkers. Clinical validity of ALK testing using CGP however, has not yet been demonstrated in an FDA approved manner. We present here the first follow-on CDx ALK test using CGP as a part of our universal CDx platform.
Method:
Clinical validity was established against the FDA-approved IHC and FISH CDx, using 291 samples from the ALEX clinical trial (NCT02075840), including ALK-negative screen-failed patients. Clinical validity was established such that the CGP concordance versus an FDA-approved test was statistically non-inferior to the performance comparison between IHC and FISH. Concordance reported in Table 1 was calculated on samples that were in agreement between IHC and FISH.
Result:
Concordance in patients in consensus between IHC and FISH is shown in Table 1. Positive percent agreement of 92.9% and negative percent agreement of 100% was achieved. Statistical non-inferiority demonstrates a non-inferiority margin of 5.4%. Limit of detection studies demonstrated the ability to detect ALK rearrangements down to 1.8% tumor content. Table 1.Companion diagnostic ALK rearrangements Indicated use ALK inhibitors in NSCLC Approved CDx comparator Ventana ALK (D5F3) CDx Assay Vysis ALK Break-Apart FISH Probe Kit Positive percent agreement 92.9% (78/84) Negative percent agreement 100% (75/75) Median unique sequence coverage 807X Precision 100% Limit of Detection 1.8% tumor content
Conclusion:
We developed and validated a follow-on CDx for ALK as part of a universal CDx platform. The platform provides a single assay with well-defined performance characteristics, identifies patients with cancer who may be eligible to receive FDA-approved targeted therapeutics, conserves tissue by avoiding serial testing and can serve as clinical trial assay for studies requiring a molecular biomarker for eligibility. The data demonstrating clinical and analytical validity of ALK may accelerate the use of CGP for routine clinical care.
-
+
P3.02-062 - An EGFR Follow-On Companion Diagnostic for Clinical Care of Patients with NSCLC (ID 10414)
09:30 - 09:30 | Author(s): M. Doherty
- Abstract
Background:
Late-stage NSCLC patients may benefit from treatment with EGFR tyrosine kinase inhibitors if they harbor certain activating mutations in EGFR. While the FDA has approved companion diagnostics (CDx) using PCR to detect EGFR mutations, molecular diagnostic testing is evolving towards comprehensive genomic profiling (CGP). However, clinical validity of EGFR testing using CGP has not yet been demonstrated in an FDA-approved manner. We present here the first follow-on CDx EGFR test using CGP as part of our universal CDx platform.
Method:
Clinical validity was established against FDA-approved PCR-based cobas® EGFR Mutation Test v2, using 406 procured samples containing EGFR L858R mutations and exon 19 deletions, and 354 samples from the AURA clinical trials for T790M mutations. For each sample, two tests of cobas were run, and clinical validity was established such the concordance between CGP and cobas were statistically non-inferior to the performance between two runs of cobas. Concordance (PPA and NPA) reported on Table 1 was calculated on samples that were in agreement between the two cobas runs.
Result:
Concordance is shown in Table 1. For L858R and exon 19 deletions, statistical non-inferiority demonstrated a non-inferiority margin of 1.9%. For T790M, our assay has higher sensitivity: of the 61 samples that were below 10% mutant allele fraction (MAF) as detected by NGS, 30 were missed by cobas. Table 1.Companion diagnostic EGFR exon 19 deletions and L858R EGFR T790M Indicated use erlotinib, afatinib, or gefitinib in NSCLC osimertinib in NSCLC Approved CDx comparator cobas® EGFR Mutation Test v2 cobas® EGFR Mutation Test v2 Positive Percent Agreement (PPA) 98.1% (106/108) 98.9% (87/88) Negative Percent Agreement (NPA) 99.4% (153/154) 86.1% (93/108) Median unique sequence coverage L858R: 1165X Exon 19: 1028X 1126X Precision 100% 100% Limit of Detection L858R: 1.9% MAF Exon 19 deletions: 3.4% MAF 1.8% MAF
Conclusion:
We present a follow-on CDx for EGFR as part of the universal CDx platform. The platform provides a single assay with well-defined performance characteristics, identifies patients with cancer who may be eligible to receive FDA-approved targeted therapeutics, conserves tissue by avoiding serial testing and can serve as clinical trial assay for studies requiring a molecular biomarker for eligibility. The data demonstrating clinical and analytical validity of EGFR may accelerate the use of CGP for routine clinical care.
-
+
P3.03 - Chemotherapy/Targeted Therapy (ID 719)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Chemotherapy/Targeted Therapy
- Presentations: 1
- Moderators:
- Coordinates: 10/18/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
-
+
P3.03-016 - Morphometric Genotyping Identifies Lung Cancer Cells Harboring Target Mutations; Cell-CT® Platform Detects Gene Abnormalities (ID 9491)
09:30 - 09:30 | Author(s): M. Doherty
- Abstract
Background:
The advent of genotype-directed therapy in personalized medicine requires the identification of driver-mutations that are often under-diagnosed due to limitations in tissue biopsy and high false negative rates associated with genomic tests. Studies have demonstrated that the mutation status of cancer cells correlates with changes in cellular morphology. The automated Cell-CT[®] platform produces isometric, high-resolution 3D images of cells in liquid biopsies, such as sputum, where published studies have demonstrated 92% sensitivity to biopsy confirmed lung cancer with 95% specificity. This study reports the development of cell classifiers for lung cancer cell lines that harbor known mutations, helping pave the way to driver-mutation targeted therapy.
Method:
Non-invasive sputum specimens from patients without lung cancer (“normal cells”) and the following cell lines were analyzed using the Cell-CT[®] platform: Small Cell Lung Cancer cell line NCI-H69 Adenocarcinoma cell lines A549 (EGFR wild-type, -c.1_471del471, KRAS- p.G12S) NCI-H1650 (EGFR- p.E746_A750del, CDKN2A- c.1_471del471, TP53- c.673-2A>G) NCI-H1975 (EGFR-T790M, CDKN2A- p.E69*, PIK3CA- p.G118D, TP53- p.R273H) NCI-H2228 (EML4-ALK+, CDKN2A- c.1_471del471, RB1- p.E204fs*10, TP53- p.Q331* high PD-L1).
Result:
15,000 normal cells from sputum and 500 malignant cells from each of the five cancer cell lines were analyzed using Cell-CT[®] platform, measuring 704 structural biomarkers to sub-classify the cancer cells by mutation status. Cell classifiers were operated to drive the highest specificity (avoidance of false positives) while maintaining sensitivity above 50%. The area under ROC (aROC), sensitivity and specificity for each classifier were:Cell Classifiers aROC Sensitivity % Specificity % Small cell lung cancer (NCI-H69) 0.991 74.8 99.98 Adenocarcinoma, EGFR wild-type (A549) 0.950 58.8 99.94 Adenocarcinoma, EGFR – pE746_A750del (NCI-H1650) 0.993 59.6 99.99 Adenocarcinoma, EGFR -T790M (NCI-H1975) 0.972 66.0 99.99 Adenocarcinoma, ALK+ (NCI-H2228) 0.992 76.8 99.97
Conclusion:
This study demonstrates the feasibility of processing non-invasive sputum specimens by the Cell-CT[®] platform to accurately identify driver mutations in cancer cells to promote mutation-directed targeted therapy for the treatment of lung cancer.
