Start Your Search
OA 10 - Liquid Biopsy for Genomic Alterations (ID 678)
- Event: WCLC 2017
- Type: Oral
- Track: Advanced NSCLC
- Presentations: 1
- Moderators:Adrian G. Sacher, Pasi A Jänne
- Coordinates: 10/18/2017, 11:00 - 12:30, F201 + F202 (Annex Hall)
OA 10.06 - Longitudinal Mutation Monitoring in Plasma by Deep Sequencing as a Potential Predictor of Disease Progression in NSCLC (ID 9595)
11:00 - 12:30 | Author(s): J.F. Palma
Circulating tumor DNA (ctDNA) sequencing and analysis has the potential to transform clinical management of patients with advanced NSCLC. Non-invasive sampling of blood draws at different time points during treatment could potentially be used for routine monitoring of disease progression and detection of therapy resistant mutations by using next generation sequencing (NGS).
448 longitudinal plasma samples (mean 6.3 per subject) collected from 71 subjects with advanced NSCLC during 1[st] line treatment were analyzed by NGS. Of these 71 subjects, 47 also had matched baseline tumor tissue samples. The AVENIO ctDNA Surveillance kit and AVENIO FFPET Analysis kit (RUO, Roche, Pleasanton, CA, USA) were used for sequencing analysis. The Surveillance kit contains 17 cancer driver genes and additional 180 frequently mutated genes mainly selected for NSCLC and colorectal cancer. This kit is capable of detecting four mutation classes: SNVs, fusions, CNVs and InDels. CT images were reviewed centrally using RECIST v1.1.
Somatic, disease-associated mutations were detected with allele frequency (AF) of >5% in 94% of baseline tumor samples (44/47), and in 100% of plasma samples with AF in ctDNA ranging from ≥0.5% to ≤30%. The most commonly mutated genes in tumors were TP53 (22/47 subjects), KRAS (14/47), BRAF (7/47), STK11 (5/47), and ERBB2 (5/47). Tracking the AF’s of key tumor mutations by the Surveillance panel in the paired longitudinal plasma samples allowed the monitoring of treatment response at the molecular level. We identified a number of subjects in which the AF of cfDNA mutations increased three to four months before clinical evidence of progression of disease detected by CT scans that were centrally reviewed according to RECIST v1.1. Cases were also observed where the AF’s of key mutations decreased in 1[st] line chemotherapy to nearly zero which correlated with clinical partial response and stable disease. . Additionally, first post treatment plasma samples collected during first line treatment showed a difference of 96 days in median survival times of ctDNA- vs ctDNA+ groups (logrank p value =0.0371).
ctDNA testing with molecular bar coded duplex sequencing and digital background error suppression of a large 197 gene panel offers high sensitivity for tumor variant detection. The study demonstrated that the presence of tumor variants detected in blood at the beginning and end of 1[st] line treatment is a risk factor for early disease progression. Longitudinal mutation monitoring has the potential to predict disease progression earlier than regular CT imaging.
Only Active Members that have purchased this event or have registered via an access code will be able to view this content. To view this presentation, please login or select "Add to Cart" and proceed to checkout.
P1.15 - SCLC/Neuroendocrine Tumors (ID 701)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: SCLC/Neuroendocrine Tumors
- Presentations: 1
- Coordinates: 10/16/2017, 09:30 - 16:00, Exhibit Hall (Hall B + C)
P1.15-011 - Longitudinal Mutation Monitoring in Plasma Without Matching Tumor Tissue by Deep Sequencing in Small Cell Lung Cancer (SCLC) (ID 9622)
09:30 - 16:00 | Author(s): J.F. Palma
SCLC is a devastating cancer with poor overall survival. Mutation profiles and treatment regimens differ significantly from non-small cell lung cancer (NSCLC). Here we demonstrate feasibility of monitoring patients with SCLC by deep sequencing from only 2 ml of plasma, without prior knowledge of the tumor tissue mutations relative to CT imaging.
Cell free DNA (cfDNA) was isolated from 64 longitudinal plasma samples (2ml) from 23 subjects with advanced SCLC using the cobas® cfDNA Sample Preparation Kit. The AVENIO ctDNA Surveillance kit (RUO, Roche, Pleasanton, CA, USA) with 197 genes detects SNVs, fusions, CNVs and InDels, and was used for sequencing the cfDNAs. Library preparation with 10-50ng cfDNA yielded a mean pre-capture library yield of 1,728ng. Mean % reads on-target was 65% with a mean deduped coverage of 3,397 fold. CT scans were reviewed to assess disease burden.
47 longitudinal plasma samples from 8 subjects were successfully analyzed without access to matched tumor tissue. Sixteen subjects with only one baseline plasma sample were excluded from further analysis. Subjects had 3-10 longitudinal plasma samples. Somatic variants were detected with allele frequency (AF) of >0.5% to 30% and 60-90% if they were present in cfDNA from at least three different time points. Germline variants were identified and removed if AFs were 40-60%, and >90%. All variants with frequencies >1% in ExAC were removed. Somatic variants were identified in TP53 (5), APC (2), NPAP1 (2), MKRN3 (2), BRAF, NFE2L2, CDKN2A, TIAM1, LRRTM1, NYP2, FAM135B, FAM71B, PIK3CG, KEAP1, DCAF12L1, PCDH15, EGFLAM. Tracking somatic variants in the longitudinal plasma samples allowed monitoring of treatment response at the molecular level for 6 of 8 subjects. In one subject with 10 longitudinal plasma samples mutations in five different genes were tracked at 1-3% AF before rising to 5- 20% at month 8 and 12-55% at month 9. Molecular progression was detected 1 month earlier than clinical progression by CT. Another subject had a TP53 splice mutation over 10 time points and through 3 lines of treatment and AF correlated with clinical response as measured by imaging.
Subjects with advanced SCLC and mixed SCLC/NSCLC can be monitored at the mutation level using molecular barcoded duplex sequencing in longitudinal plasma samples. 2ml of plasma yielded sufficient cfDNA for testing with the Surveillance kit. Somatic mutation monitoring is possible without matching tumor tissue samples where longitudinal mutation profiles correlate with clinical response by CT imaging.