Virtual Library
Start Your Search
S. Rabizadeh
Author of
-
+
OA 18 - Lung Cancer Pathology and Genetics (ID 687)
- Event: WCLC 2017
- Type: Oral
- Track: Biology/Pathology
- Presentations: 1
- Moderators:George R. Simon, Yoon-La Choi
- Coordinates: 10/18/2017, 14:30 - 16:15, Room 316
-
+
OA 18.04 - Whole Genome Tumor-Normal Sequencing Reveals Potential False Positives Versus Standard CGP Sequencing in Patients with NSCLC (ID 10453)
14:30 - 16:15 | Author(s): S. Rabizadeh
- Abstract
- Presentation
Background:
Matched tumor-normal sequencing analyses are essential for precise identification and interpretation of somatic and germline alterations. A 2015 study of 815 paired tumor and normal genomes showed that both genomes are required for precise identification and interpretation of both somatic and germline variation. In this study, we further demonstrate the critical importance of both tumor and germline sequencing using a selected panel of genes that are relevant to cancer prognosis and treatment.
Method:
Tumor and normal (germline) genomes were sequenced using the Illumina HiSeqX platform. Data was aligned to GRCh37 using methods described previously. Tumor versus matched-normal variant analysis was performed using the NantOmics Contraster analysis pipeline to determine somatic and germline genomic variants. RNA-Seq libraries were prepared from tumor samples using KAPA stranded RNA-Seq with RiboErase kit and sequencing on the Illumina platform. RNA sequencing reads were aligned and variants were identified using methods that have been previously described.
Result:
44 patients with NSCLC with tumor-normal sequencing were analyzed. 29 patients also had RNA-Seq whole transcriptome data available for analysis. Focusing on somatic SNVs in ALK, BRAF, CDKN2A, CEBPA, DNMT3A, EGFR, ERBB2, EZH2, FLT3, IDH1, IDH2, JAK2, KIT, KRAS, KMT2A (MLL), NPM1, NRAS, MET, NOTCH1, PDGFRA, PDGFRB, PGR, PIK3CA, PTEN, RET; 427 germline plus somatic variants were detected in these 25 genes in 44 patients; 386 out of 427 (90%) variants detected in somatic tissue were also present in germline (true positive germline variant, false positive somatic variant). 5 out of 29 patients (17%) did not have detectable expression in at least one somatic variant. Focusing on germline SNVs in APC, MYH, MLH1, MSH2, MSH6, PMS2, EPCAM, POLE, POLD1, BMPR1A, PTEN, STK11; 350 germline plus somatic variants were detected in these 12 genes in 44 patients, 10 out of 44 patients (23%) had at least one variant detected in their tumor DNA in these 12 genes that was not detected in the patient’s germline DNA (false positive germline variant, true positive somatic variant).
Conclusion:
Somatic-only sequencing may lead to false positive variant calls which has important clinical implications for highly actionable targets and for the veracity of mutational load algorithms. Calling germline variants from somatic DNA has a false positive risk---called variants may truly be somatic mutations. This is further complicated by the presence of circulating tumor DNA which may also lead to a false positive “germline” result in the absence of a true normal comparator.
Only 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, select "Add to Cart" and proceed to checkout. If you would like to become a member of IASLC, please click here.
-
+
P2.01 - Advanced NSCLC (ID 618)
- Event: WCLC 2017
- Type: Poster Session with Presenters Present
- Track: Advanced NSCLC
- Presentations: 1
- Moderators:
- Coordinates: 10/17/2017, 09:00 - 16:00, Exhibit Hall (Hall B + C)
-
+
P2.01-056 - Use of Cell-Free Circulating RNA (cfRNA) Expression of PD-L1 and ERCC1 in Plasma to Monitor Response to Therapy in NSCLC (ID 9038)
09:00 - 16:00 | Author(s): S. Rabizadeh
- Abstract
Background:
There is an unmet need to evaluate tumor response by other means than radiology tests. Cell-free circulating tumor RNA (cfRNA) can be extracted from plasma of cancer patients (pts); measuring dynamic changes in gene expression and levels of b-actin; cfRNA (per ml of plasma) as a proxy for total cfRNA in metastatic patients has shown great potential for evaluating disease status and predicting outcome to anti-tumoral therapy in advance of imaging. We have previously shown that high levels of PD-L1 cfRNA expression correlates well with positive response to immunotherapies including nivolumab in pts with NSCLC.
Method:
Blood was drawn from pts at approximately 6-week intervals under various therapies, with CT scans at 3-month intervals. Total cfRNA was extracted from patient plasma and reverse transcribed to cDNA. Levels of b-actin, ERCC1 and PD-L1 were quantitated across multiple blood draws by RT-qPCR and correlated with pt response (PR/SD/PD), as determined by CT scans.
Result:
A total of 24 NSCLC patients were enrolled in a 1-year clinical study. Non-SCC comprised 87% (21/24). 19 pts completed the first two cycles of therapy. 1 pt with PR had decreasing levels of cfRNA, 10 pts achieved SD with decreasing or no change while 6/8 pts with PD had increasing levels of cfRNA. CfRNA levels were predictive of disease status about 4 weeks in advance of imaging in 6/19 pts and matched with disease status in 8/19 pts (74% ). Dynamic changes in PD-L1 expression correlated with response to nivolumab in 3/4 pts. In 2/4 pts with SD, PD-L1 remained undetected after therapy, whereas 1 patient continued to have PD despite loss of PD-L1. PD-L1 was undetectable in a pt initially with PD on nivolumab who achieved SD after one cycle of nivolumab plus radiation. Changing ERCC1 expression correlated with platinum-based therapy outcome in 8/8 patients. 4/4 patients with PD on pemetrexed/carboplatin had an increase in ERCC1. 4/4 patients with lower or decreasing levels of ERCC1 achieved PR or SD. In the only patient achieving PR, ERCC1 became undetectable during treatment.
Conclusion:
We found significant concordance between clinical response and changes in plasma cfRNA levels in NSCLC pts (74%). Levels of PD-L1 expression correlated with response in 3/4 pts treated with nivolumab . ERCC1 levels were predictive of outcome to platinum based therapy for 8/8 patients. ERCC1 and PD-L1 expression in cfRNA can be used to monitor response to platinum-based and immuno-therapy.
