Search

3542 Background: Cancer screening in asymptomatic individuals who meet guideline criteria has yielded reductions in cancer death rates. However, adherence to screening guidelines remains below targets set forth by leading health-care organizations. A blood-based multi-cancer screening assay with clinically meaningful sensitivity and specificity, in cancer types where early detection and intervention can save lives, that is integrated with existing clinical pathways may increase access to and adherence with guideline recommendations, ensuring more individuals benefit from these proven interventions. We evaluated the performance of a blood-based multi-cancer screening assay that interrogates cell-free DNA (cfDNA) methylation signatures for cancer detection and tissue of origin prediction in a set of tumor types where cancer screening can save lives. Methods: Whole blood from 1,607 individuals with and 3,298 individuals without cancer was obtained from multiple unique cohorts. Plasma-derived cfDNA was profiled using a custom assay that enriches fragments with dense CpG methylation and further depletes uninformative background molecules containing unmethylated CpGs. We utilized a broad genomic panel (16 Mb) targeting regions with low rates of methylation in individuals without cancer. The panel captures tumor-associated molecules and allows for high sensitivity of detection at low sequencing costs. A cross-validated analysis was used to estimate the performance of the predictive model upon the sample set. Classification thresholds corresponding to 90%, 95%, and 98% specificities were established using samples from individuals without a cancer diagnosis. Results: At 90% specificity, overall sensitivity for lung cancer detection was 92.1% (95% CI: 80-100%; 90.2% in Stage I/II disease (N = 82) and 93.1% in Stage III/IV disease (N = 159)) and 93.1% (CI: 88-98%) for CRC detection (92% in Stage I/II disease (N = 743) and 94.5% in Stage III/IV disease (N = 623)). Tissue of origin prediction evaluated at 98% specificity yielded accurate identification in 99% of CRC and 98% of lung cancers. Lung cancer histology was known for approximately 74% of the cohort. Across Stage I – IV cancers, at 90% specificity, sensitivity was 97.3% in lung squamous cancer (N = 73) and 86.8% in lung adenocarcinoma (N = 106). At 95% and 98% specificity thresholds, overall sensitivity was 86.3% (CI: 75-98%) and 66.4% (CI: 56-77%) for lung cancer and 85.7% (CI:81-91%) and 71.6% (CI: 67-76%) for CRC, respectively. Conclusions: This blood-based multi-cancer screening assay yields clinically meaningful sensitivity and specificity for early-stage cancers. This assay is undergoing further development to expand detection capabilities to additional cancer types where screening can save lives. Clinical evaluation in registrational screening trials is ongoing (SHIELD; NCT05117840).

3627 Background: Blood-based colorectal cancer (CRC) screening tests can improve adherence to screening guidelines. Yet, current commercially available options have poor sensitivity and specificity inhibiting incorporation into routine clinical care. Here we report the validation of a blood-based test for the detection of colorectal cancer and advanced neoplasia. Methods: This blood-based test aims to detect colorectal neoplasia by identifying tumor-associated biomarkers including genomic or epigenomic (methylation and fragmentomics) signatures in cell-free DNA (cfDNA). cfDNA is partitioned based on methylation level, enriched for informative genomic regions, and sequenced. This novel workflow enables high-fidelity analysis of multi-modal information in majority of extracted cfDNA molecules. Results are integrated into a binary “detected” versus “not-detected” result using a proprietary bioinformatic pipeline (Guardant Health, USA). The assay was trained on samples obtained from >6,000 unique individuals (2,685 cancer-free and 1,698 with advanced colorectal neoplasia (ACN) for training, 1,072 cancer-free and 551 with ACN for threshold setting). The thresholds were frozen prior to validation targeting a specificity of > 91.5%. Each aspect of the validation study followed Nex-StoCT CLIA working group and CLSI guidelines. Results: Limit of detection (LoD) was established across six dilutions. Even for low cfDNA mass inputs of less than 4ng, the 95% LoD was determined to be less than 1 tumor-derived genomic equivalent (0.5), indicating over at least 10-fold increase in assay sensitivity compared to best-in-class assays for somatic mutation detection. Precision studies in 60 positive and negative replicates from clinical samples yielded >90% average positive and negative percent agreement both within and between batches. Endogenous interference studies yielded > 90% positive and negative percent agreement between reference control and common endogenous substances, including albumin, bilirubin, hemoglobin, triglycerides, and genomic DNA, in clinical positive and negative samples and minimally manipulated samples. The clinical validation of the test was conducted in > 300 cases (biobanked pre-operative cohort for CRC cases and screening cohort for advanced adenoma and negative cases). Conclusions: Here we present the validation of a multi-modal blood-based test for the detection of colorectal cancer. This test is currently being evaluated in a registrational study (ECLIPSE: NCT04136002).

3045 Background: Detection of minimal residual disease (MRD) by circulating tumor DNA (ctDNA) post-curative intent treatment is predictive of recurrence in many solid tumors. Due to biological challenges with low ctDNA shed in early-stage disease and potential to detect non-tumor derived alterations in plasma (e.g. CHIP), most ctDNA MRD assays are dependent on a priori knowledge of genomic alterations from tumor tissue to achieve high sensitivity and specificity. Tissue-dependent methods limit the clinical application of a MRD assay, especially in cancer types where tissue biopsy is challenging, neoadjuvant therapy (NAT) is standard of care, and/or rapid turnaround times are needed for clinical decision making. We previously validated an assay (Guardant Reveal) that combines somatic and epigenomic analysis to detect ctDNA from early-stage colorectal tumors without tumor tissue or peripheral blood cells. Here we describe the expansion of this assay to detect MRD across multiple tumor types. Methods: Cell-free DNA (cfDNA) fragments are extracted from patient plasma, partitioned based on methylation fraction, enriched using a panel to target informative genomic and epigenomic regions, barcoded, and pooled for sequencing. Methylation status is determined non-destructively and with minimal loss of molecules, allowing sensitive genomic and epigenomic analysis of the same cfDNA fragments. A single assay with a total panel size of 5.3 Mb was developed for MRD analysis in multiple cancer types, including CRC, Lung and Bladder. A “ctDNA detected” result is defined by the de novo identification of tumor-derived somatic variants and/or the observation of a tumor-specific methylation profile exceeding predefined thresholds. Results: The assay performance was tested using 163 pre-treatment clinical samples from patients with early-stage non small cell lung cancer (NSCLC) and bladder cancer and 133 self-declared healthy donors. Sensitivity for pre-treatment detection in NSCLC was 68.9% at 95% specificity (20/42 Stage I: 47.6%, 25/30 Stage II: 83.3%, 19/21 Stage III: 90.5%). Sensitivity for bladder cancer was 44.2% at 95% specificity (13/43 NMIBC: 30.2%, 18/27 MIBC: 66.7%). Additional development from larger cohorts and other tumor types is ongoing and data will be presented as available. Conclusions: Using cancer-specific genomic and epigenomic signals combined with learning-based classifiers, we developed a highly specific method for detecting the presence of ctDNA in early-stage NSCLC and bladder cancer patients from plasma without the need for tumor tissue but with comparable sensitivities to tissue-dependent approaches. A plasma-only MRD assay will allow for greater clinical impact by overcoming the challenges of tissue procurement, particularly following NAT, and enabling faster time to results.

3536 Background: To improve average risk CRC screening compliance, additional options are needed, especially options that address patient and provider reported barriers such as time and convenience. LUNAR-2 is a multimodal blood-based colorectal neoplasia detection assay incorporating ctDNA assessment of somatic mutations and tumor derived methylation and fragmentomic patterns, aimed to maximize sensitivity for early stage CRC detection. We evaluated this test in a large patient cohort with newly diagnosed CRC. Methods: Individuals diagnosed with CRC between 2013-2016 consented to provide blood samples prior to surgical resection. Those treated with neoadjuvant chemotherapy were excluded. Isolated plasma samples (median 3mL from EDTA) from 434 individuals were analyzed with LUNAR-2 (Guardant Health, USA) and included in the analysis. Median age at CRC diagnosis was 63 years (range 28 - 89) and 41% were female. Control samples were from 271 age-matched cancer free individuals. “ctDNA detected” and “ctDNA not detected” results were generated by a model trained on a separate sample set (N=614) from both cancer free individuals and those with CRC. Calling threshold was determined based on this held-out set to target 90% specificity. ctDNA results and clinical characteristics were correlated. Results: Overall CRC sensitivity was 91% (393/434), with high sensitivity across all stages; 88% Stage I/II, 93% Stage III (Table). Specificity was 94% (255/271). There was no difference in sensitivity when excluding those with early (84 years) onset CRC (90% sensitivity; 388/429; p=0.95; 88% Stage I/II, 93% Stage III). There were no differences in sensitivity for asymptomatic CRC (88%) compared to symptomatic CRC (91%; p=0.4; Table). However, higher cell-free DNA tumor fractions were observed in the symptomatic cohort. Sensitivity for detection of right-sided and left-sided CRC was similar (93% vs. 90%; p=0.5; Table). Conclusions: In this large early-stage CRC cohort, multimodal ctDNA assessment has high sensitivity for CRC detection with high specificity. Equivalent sensitivity in the asymptomatic cohort suggests this test will have clinically meaningful performance in an average risk screening population. A prospective registrational study is ongoing to evaluate the test in an average risk CRC screening cohort.[Table: see text]

3572 Background: Serial ctDNA can measure dynamic changes in disease burden over time, however utility of serial profiling to detect changes in actionable alterations remains unclear. Methods: We evaluated 501 patients with ≥3 serial Guardant360 assays performed between 09/2016 and 11/2020 and compared MSI, fusion, amplification and single nucleotide variant (SNV) detection over time. This comprised 2147 assays with a median of 4 assays per patient (min 3, max 18) occurring an average of 163 days apart (+/- SD of 147 days). Maximum detected variant allele frequency in samples (maxVAF) was assessed for relation to changes in detected alterations as a surrogate for tumor volume. Results: Among 406 patients with assays assessable for MSI-status, 17 (4.2%) had MSI detected. New MSI detection on a subsequent assay always occurred with a rising maxVAF (3/3) that was also ≥0.7%, while loss of detectable MSI between assays always associated with falling maxVAFs (7/7) with 6/7 occurring when maxVAF fell below 0.4%. Fusions were noted in 9/501 (2%) patients. Among 3 patients who lost a detectable fusion, maxVAF decreased in 1 patient and changed ≤0.2% between assays in 2, while 2/3 patients with new fusions had rising maxVAFs and 1 patient had a falling maxVAF. Amplifications were detected in 242/501 patients (48%). While most genes had highly variable amplification detection between assays (9% serially detected), ERBB2 amplifications were more consistent and serially detected in 39% of detected cases (P < 0.0001). New detection of amplifications occurred more commonly in cases with rising maxVAF (OR 11.70, 95% CI 7.61-18.00, P < 0.0001) and loss of detectable amplifications occurred more between samples with falling maxVAF (OR 12.37, 95% CI 8.35-18.66, P < 0.0001). Change in maxVAF correlated with change in number of detected amplifications (r = 0.62, P < 0.0001), but only partially explained changes seen (R2= 0.39). Between serial assays, SNVs changed a median of 0 variants (IQR -1 to 1), however some patients had significant changes (max gain 21/max loss 18). Among 1646 serial time points, 454 (28%) had no change in SNVs, 674 (41%) gained SNVs, and 518 (31%) lost SNVs on subsequent assays. Gains were more common in samples with rising maxVAF (OR 7.76, 95% CI 6.18-9.73, P < 0.0001) while losses were more common when maxVAF fell (OR 6.90, 95% CI 5.47-8.66, P < 0.0001). The correlation between maxVAF change and SNV change was significant (r = 0.29, P < 0.0001), but minimally explained SNV changes (R2= 0.086) and was a much weaker association than noted for amplification changes. Conclusions: We noted significant differences in detection of actionable alterations across serial ctDNA assays. Increased ctDNA volume (higher maxVAF) due to tumor progression may explain some variation over time, but variability also occurs outside these changes, likely reflecting clonal evolution following therapy.

e15549 Background: Liquid biopsy is a candidate for detection of minimal residual disease (MRD) in early colorectal cancer. Detection of circulating tumor (ct)DNA in early stage cancer is challenging given the low abundance of ctDNA molecules. We developed a tissue agnostic MRD Test to optimize sensitivity and specificity of ctDNA detection by integrating tumor-specific genomic mutation and DNA methylation signatures coupled to noise-reduction of non-tumor background cfDNA signals. Methods: This MRD Test was developed following CLIA, Nex-StoCT Working Group, and AMP/CAP guidance and validation principles. Using a single input plasma sample, data from somatic and epigenomic targeted capture panels are integrated to generate a final test result: ctDNA detected or ctDNA not detected. ctDNA detected is defined by the presence of mutation or epigenomic classifier inputs exceeding a defined threshold. Quality control (QC) measures were implemented in both process and final sequencing results. Analytical specificity, sensitivity, and accuracy were determined using 130 samples. Results: The MRD Test specificity was 95% (70/74) using colonoscopy screened negative samples.A retrospective review of the clinical histories of these false positive samples showed high risk clinical features such as smoking, family history of CRC, and sub-optimal bowel prep prior to the colonoscopy. Analytical sensitivity (LoD) was established by diluting late stage CRC samples at two clinically relevant cfDNA inputs at three dilutions (30ng or 10ng, ranging 0.1% to 0.75% tumor fraction). The 95% LoD was determined to be 0.3% for 10ng and < 0.1% for 30ng, the lowest level tested for each input. Accuracy was determined by diluting advanced CRC samples to 0.5-0.75% MAF tested at 30ng or 10ng cfDNA with a detection result of 100% (38/38). Conclusions: We present a CLIA validated assay with performance characteristics sufficient to detect residual disease in Stage II/III CRC post-curative intent therapy. This assay is currently being used in multiple interventional clinical trials.

3545 Background: FGFR2/ 3 rearrangements are promising therapeutic targets, especially in advanced urothelial cancer (aUC) with FDA-approved erdafitinib. Liquid biopsy is an attractive non-invasive method to identify these fusions, but detection in cfDNA is technically challenging due to low tumor shedding levels, short molecules, and wide variation in gene partners. To address this, we developed an assembly-based fusion detection algorithm to call rearrangements in a de novo fashion without reliance on a fixed partner set and applied it to > 15,000 clinical samples. Methods: A cohort of 15,218 patients with mixed cancer types (including 698 aUC patients, as well as breast, cholangiocarcinoma, colorectal, and gastric), plus 276 healthy control samples were previously tested with Guardant360(R), a clinical 74-gene cfDNA NGS-based assay. The median unique molecule coverage was approximately 3,000 molecules sequenced to 15,000x read depth. Samples were reanalyzed in silico using the novel algorithm: in brief, reads aligned to candidate fusion breakpoints were assembled into de Bruijn graphs. Resulting contigs were aligned to the reference and filters were applied to remove technical artifacts. Results: The majority of FGFR2 (86%) and FGFR3 fusion partners (73%) in the mixed cancer cohort were observed only once, consistent with previous reports (Helsten 2016). FGFR3-TACC3 was the most common fusion, occurring in 72% of FGFR3 fusion-positive patients. In 37% of FGFR2 fusion positive patients, the de novo caller detected partners not previously described. In the aUC cohort, FGFR3 fusions were detected in 3.3% of patients, with 8/10 (80%) partner genes/intergenic regions occurring only once, which is in line with previous reports (Nassar 2018). No fusions were identified in 276 healthy control samples. In the mixed cancer cohort, common mutations co-occurring with FGFR2 fusions were FGFR2 N549K, PIK3CA H1047R, and TP53 R175H (5.6% each); KRAS Q61H was observed in 28% of patients with FGFR3 fusions. Conclusions: FGFR2/ 3 fusion partners detected by a highly specific assembly-based de novo fusion caller were heterogeneous and individually rare, highlighting the importance of a de novo approach. We observed an FGFR3 fusion prevalence in cfDNA from aUC patients that is comparable to previous reports for tissue testing, demonstrating an ability to capture targetable genomic rearrangements with plasma-based NGS in this patient population.

415 Background: Testicular germ cell tumors (GCT) infrequently harbor somatic mutations. ctDNA assessment allows the noninvasive genomic profiling of malignancies and may assist with understanding molecular evolution of resistance. We report ctDNA profiling of patients (pts) with testicular GCTs. Methods: 40 patients (pts) with advanced testicular GCTs from multiple institutions in the USA that underwent ctDNA analysis using the Guardant (G)-360 platform were eligible and a total of 48 samples were collected. 36 pts had one sample, 3 pts had 2 samples, 1 pt had 6 samples. De-identified demographic data were collected in addition to data for ctDNA alterations. G360 employed a CLIA-certified ctDNA panel that assessed single nucleotide variant and copy number alterations in 68 to 73 genes for potentially actionable genomic alterations. Variants reported at least 3 times in the Catalogue of Somatic Mutations in Cancer (COSMIC) database or found in OncoKB were considered pathogenic. Results: Of 40 patients with testicular GCTs, 13pts (33%) were post systemic therapy. The median age was 36 years (range 20-61). 199 ctDNA alterations were detected in 35 patients (87.5%) across 41 genes. Among the 199 alterations, 102 were believed to be pathogenic and detectable in 26 samples from 25 pts (62.5) (%). The most common pathogenic somatic alterations were KRAS (n = 16/102, 16%), TP53 (n = 16/102, 16%), CCND2 (n = 9/102, 9%), CDK6 (n = 9/102, 9%), MET (n = 9/102, 9%), and RAF1 (n = 6/102, 6%). Conclusions: ctDNA alterations were frequently detected in resistant testicular GCTs and appear similar to alterations previously described in tumor tissue analyses of testicular GCTs. Given that ctDNA offers a non-invasive means of profiling tumor DNA, further development of this promising modality is warranted to study the evolution of resistance to cisplatin-based chemotherapy and new potentially actionable alterations.

6 Background: Penile cancer is a rare disease associated with HPV infection and harbors recurrent somatic genomic alterations in the ERBB (HER)-family, CDKN2A, TP53, NOTCH1 and PIK3CA. ctDNA assay allows the noninvasive genomic profiling of malignancies and may assist with understanding molecular evolution. To our knowledge, the genomic alterations observed in ctDNA for penile cancer have not been described before. We report ctDNA profiling of patients with advanced penile cancer. Methods: Sixteen pts with metastatic penile cancer from multiple institutions in the United States that underwent ctDNA analysis using the Guardant360 platform were eligible. Three patients had at least one serial ctDNA sample. De-identified demographic data were collected. Guardant 360 is CLIA-certified ctDNA panel that assesses single nucleotide variants and copy number alterations in 68 to 73 genes for potentially actionable genomic alterations. Variants seen at least 3 times in the Catalogue of Somatic Mutations in Cancer (COSMIC) database or reported in OncoKB were considered pathogenic. Results: The median age was 64 years (range 40-77). 4 pts (25%) were documented to be post platinum-based chemotherapy. Among the entire cohort, 51 ctDNA alterations were detected (median=2, range 0-6) in 15/16 patients (94%) across 21 genes (table). Of the 51 alterations, 24 (47%) were actionable and had approved targeted therapies in other cancers. Alterations were most frequently detected in TP53 (9/16, 56%), CDKN2A (5/16, 31%), and TERT promoter (5/16, 31%) (table). In 3 patients with serial samples, 9 novel pathogenic alterations were detected in the second sample including ATM, CDKN2A, ARID1A, CCND1, CDK6, EGFR, PDGFRA, PIK3CA, and SMAD4. Conclusions: ctDNA alterations in patients with advanced penile cancer were frequently detected and appeared similar to previously described tumor tissue analyses. New alterations found on serial ctDNA assays shed light on patterns of tumor evolution and may inform drug development for this challenging orphan malignancy.[Table: see text]

e16063 Background: Testicular germ cell tumors (GCT) infrequently harbor somatic mutations. ctDNA assay allows the noninvasive genomic profiling of malignancies and may assist with understanding molecular evolution of resistance. To our knowledge, ctDNA genomic alterations observed in in testicular GCTs have not been heretofore described. We report ctDNA profiling of patients (pts) with testicular GCTs. Methods: 31 Pts with testicular GCTs from multiple institutions in the United States that underwent ctDNA analysis using the Guardant360 platform were eligible. Two patients had one serial ctDNA sample. De-identified demographic data were collected in addition to data for ctDNA alterations. Guardant360 is a CLIA-certified ctDNA panel that assesses single nucleotide variant and copy number alterations in 68 to 73 genes for potentially actionable genomic alterations. Variants reported at least 3 times in the Catalogue of Somatic Mutations in Cancer (COSMIC) database or found in OncoKB were considered pathogenic. Results: Of 31 patients with testicular GCTs, 162 ctDNA alterations were detected in 26 patients (84%) across 41 genes (Table). A median number of 3 alterations (range 1-19) was detected per sample. Among the 162 alterations, 88 were believed to be pathogenic and detectable in 20 patients (65%). 12/31 pts (39%) were documented to be post systemic therapy. The median age was 38 years (range 20-61). The most common pathogenic somatic alterations were KRAS (n = 12/88, 14%), CCND2 (n = 8/88, 9%), MET (n = 8/88, 9%), CDK6 (n = 7/88, 8%), TP53 (n = 6/88, 7%), and RAF1 (n = 5/88, 6%). In 2 patients with serial samples, 5 novel pathogenic alterations were detected in the second sample including FGFR2, APC, CDK6, RAF1, and MAPK1. Conclusions: ctDNA alterations were frequently detected in resistant testicular GCTs and appear similar to alterations previously described in tumor tissue analyses of testicular GCTs. Given that ctDNA offers a non-invasive means of profiling tumor DNA, further development of this promising modality is warranted to determine it's relevance in clinical practice. [Table: see text]

3590 Background: While tissue-based assays have yields above 90% in solid tumors, there is less known about factors that influence the sensitivity of ctDNA for detecting mutations. Methods:We retrospectively evaluated mCRC patients (pts) who had plasma-derived NGS utilizing a highly-sensitive targeted 68-73-gene ctDNA assay. In a case-control design, pts with a known mutation on tissue and radiologic evidence of metastatic disease but no detectable ctDNA mutation were matched 1:3 with randomly selected pts with detectable mutations and compared according to clinical, laboratory, and radiologic characteristics. A prediction score for ctDNA detection was built using a binary logistic backward stepwise regression analysis and tested in two independent data sets from different institutions. Area under the curve (AUC) from receiver operating characteristics curves (ROC) were used for internal and external validation. Results: From 416 pts who met inclusion criteria, plasma-derived NGS did not find tumor mutations in 66 cases (15.9%); 198 pts with detectable alterations were selected as controls. After multivariate analysis, the detection of ctDNA was associated with increasing age (OR 1.05; 95%CI 1.02-1.09; p = .001), presence of liver (OR 5.82; 95%CI 2.55-12.49; p < .001) and lymph node metastases (OR 3.28; 95%CI 1.51-7.60; p = .004), archival TP53 mutations (OR = 2.88; 95%CI 1.37-6.17; p = .006). A key determinant was timing of collection relative to disease status: plasma collected in newly diagnosed metastatic disease or after evidence of progression was substantially more likely to have detectable alterations (OR 9.24; 95%CI 4.11-22.40; p < .001); The simplified prediction model performed well in internal (AUC = 0.88) and external validation (AUC = 0.95; 163 pts). Conclusions: Our validated prediction model provides clinicians and researchers with a tool to screen for patients in whom ctDNA testing can outperform tissue-based testing in detecting genomic alterations.

3602 Background: ctDNA identifies patients (pts) at high risk for disease recurrence post CRC resection (post-op). Current ctDNA residual disease detection approaches assess only genomic alterations (alts) and rely on tissue sequencing to identify tumor-derived alts. We evaluated a plasma-only ctDNA assay to identify high risk pts. Methods: 72 CRC pts (surgery only = 42; adjuvant therapy (adj) = 30) had post-op and/or post-adj plasma samples (3-4mL). Extracted cfDNA (median 27 ng) was analyzed using a single-sample NGS test validated in early stage CRC that integrates assessment of genomic alts with epigenomic cancer signature (Guardant Health, CA). A variant classifier was applied to differentiate tumor-derived from non-tumor derived alts in a tumor tissue-uninformed approach. Results: In the surgery cohort, samples were collected a median of 31 days (d) post-op. 7/8 pts with post-op ctDNA detected (ctDNA+) recurred (PPV 88%; median time to recurrence (mTTR) 248d). The recurrence-free pt has < 180d follow-up. 7/34 pts without ctDNA detected (ctDNA-) recurred (NPV 79%; mTTR 333d). 1/1 Stage 0-II ctDNA+ pt recurred (PPV 100%; TTR 440d) while 1/20 ctDNA- recurred (NPV 95%; TTR 440d). 27 pts in the adj cohort had samples collected a median of 37d post-adj. 6/6 ctDNA+ pts recurred (PPV 100%, mTTR 239d). 4/21 ctDNA- pts recurred (NPV 81%, mTTR 466d). 2/2 ctDNA+ and 0/11 ctDNA- Stage III pts recurred (PPV, NPV 100%, mTTR 420d). All 3 post-op ctDNA+/post-adj ctDNA+ (ctDNA persistence) pts recurred. 1/2 post-op ctDNA+/post-adj ctDNA- (ctDNA clearance) pts is recurrence free (306d). 2 post-op ctDNA-/post-adj ctDNA+ pts recurred. In the entire cohort, ctDNA+ after standard therapy completion had a recurrence PPV 93%, NPV 80%, HR 11.29 (p < 0.0001). Conclusions: In post-op CRC, ctDNA detection utilizing a tumor-uninformed integrated genomic and epigenomic assay has high recurrence PPV and NPV following standard therapy completion. ctDNA identifies pts who may benefit from post-op adj therapy or additional/modified post-adj therapy. These findings demonstrate that ctDNA detection from a single post-op/post-adj plasma sample stratifies high/low risk pts and informs therapy decision making.

3057 Background: ctDNA sequencing has been rapidly adopted for the identification of targetable somatic alterations (alts) in patients with advanced cancers. However, early stage disease detection has been hindered by low levels of ctDNA in circulation and the presence of confounding non-tumor-related somatic alts. We developed and validated a ctDNA assay that combines somatic and epigenomic signals to detect early stage tumors without tumor tissue or white blood cells (WBC). Methods: Using a single input sample, our assay integrates the sensitive detection of genomic alts with quantification of epigenomic signals associated with cancer. Non-tumor alts (e.g., clonal hematopoiesis of indeterminate potential; CHIP) are excluded using a newly developed bioinformatic classifier. To assess analytical sensitivity, specificity, and positive and negative reproducibility, we tested 337 clinical and contrived samples. Results: Clinical specificity was determined using 80 plasma samples from 50-75 year old presumptive cancer-free donors, and resulted in a single false positive (99% specificity). Analytical sensitivity (limit of detection) was established using a dilution series of 4 different late stage CRC pts tested in triplicate at a clinically relevant DNA input (30 ng) across multiple batches. 100% sensitivity was maintained even at the lowest tested level (estimated 0.1% tumor level). Positive/negative reproducibility was assessed by testing triplicates of diluted late stage samples and age-matched healthy donors, respectively, across different cfDNA inputs, and multiple reagent lots. Both positive and negative calls were 100% concordant across all replicates. Independent estimation of tumor levels from epigenomic or genomic signals produced highly concordant results (correlation r-value: 0.82, p-value: 3e-16). Conclusions: We designed and validated a highly specific ctDNA assay that integrates both genomic and epigenomic signals to allow for accurate and quantitative tumor level detection in early stages of the disease without requiring tumor tissue or WBC.

e14557 Background: Circulating tumor DNA (ctDNA) analysis has been successfully applied to therapy selection and treatment monitoring in advanced cancer patients. However, it is not yet established whether ctDNA can be used clinically for early cancer detection or predicting tumor recurrence in early stage lung cancer patients. Methods: We analyzed pre-operative plasma samples from 55 early stage NSCLC patients (stages I-IIIA) using next-generation sequencing to detect somatic mutations and differential epigenomics patterns, including methylation signatures. Results: Using somatic mutation analysis alone, ctDNA was detected in 42% (23/55) of patients, whereas combined mutational and epigenomic analysis detected ctDNA in 71%. ctDNA detection rate also varied markedly between lung squamous cell carcinoma (SCC) and adenocarcinoma (ADC);using combined analysis of somatic mutations and epigenomic patterns, ctDNA was detected in all SCC patients, while only 55% of ADC (12/22) were ctDNA-positive (p= 0.006). Within the ADC subgroup, ctDNA detection rates using the combined approach were dependent on disease stage: 47% (8/17) in stage I, 100% (2/2) in stage II, and 100% (2/2) in stage IIIA. Importantly, pre-operative ctDNA status was correlated with tumor recurrence post-resection; three of eight (38%) ctDNA-positive stage I ADC patients recurred within 2 years of resection, while only one of nine (11%) ctDNA-negative stage I ADC patients recurred (p= 0.29). Conclusions: Taken together, we show that the combination of somatic mutation detection and epigenomic analysis outperforms each individual biomarker in the detection of ctDNA in early stage lung cancer. Importantly, we also demonstrate that pre-operative ctDNA detection may identify a high-risk population of early stage lung cancer patients that may benefit from (neo)adjuvant therapy.

3511Background: Colorectal cancers (CRC) treated with EGFR inhibitor (EGFRi) evolve by a repetitive process of genetic diversification and clonal evolution. After discontinuation of EGFRi, KRAS/NRA...

3555Background: MET amp is a well described driver of acquired EGFR antibody (Ab) resistance. Blood-based genomic profiling of cell free (cf)DNA is a safe and efficient means to identify pts with a...

12044Background: ctDNA in post-op CRC pts correlates with molecular residual disease and may be useful to guide adjuvant therapy. However, initial studies employed clinically impractical assays ind...

641 Background: Circulating tumor DNA (ctDNA) represents an ideal platform to obtain the most current genomic profile of a patient’s tumor. We aimed to investigate how stable these profiles remain during serial ctDNA assays in metastatic colorectal cancer (mCRC). Methods: In 77 patients (pts) with mCRC and serial Guardant360 assays with a detectable mutation (mt), we compared mt stability by assessing whether variants were gained/lost between serial assays and changes in relative mutant allele frequency (rMAF). rMAF of a mt was defined as (mt allele frequency / mt present at the maximum allele frequency in that assay). rMAF results were normalized to detected ctDNA concentration changes between assays to ensure changes in rMAF were not due to changes in ctDNA concentration. MAPK pathway mutations were defined as RAS, BRAF, EGFR, KIT, or MET mutations. Results: Of 77 pts, 64 (83%) had 2 serial assays and 13 (16.9%) had 3 or 4 assays performed. Serial assays occurred an average of 138 days apart (+/- SD of 111 days). Only 13/77 (17%) pts had no change in the number of mts detected between assays. A new mt was detected in 42/77 (55%) pts, while 43/77 (56%) lost a previously detected mt. Of 52 mts detected in patients with > 2 assays, 16 (31%) were gained and subsequently lost. After controlling for ctDNA concentration, mts were equally likely to have an increasing (129/308 – 42%) or decreasing (150/308 – 49%) allele frequency. Potentially clinically relevant MAPK variants were gained/lost in 29% of patients; though MAPK mts developed in a large number of pts (16/77 – 21%), many pts also lost MAPK mts (9/77 – 12%), showing ongoing subclonal dynamics. Median time between assays did not differ between pts with gain/lost mts or stable mt profiles (P = 0.73), however mt rMAF shift of > 25% was more common if assays were > 90 days apart (OR 4.3, P < 0.0001). Conclusions: Serial ctDNA assays demonstrate ongoing mutational changes in mCRC, with emergence/disappearance of MAPK variants being more common than expansion of a pre-existing clone. Our results suggest repeated sampling may be important to optimize selection of targeted therapies at each regimen alteration.

634 Background: Utilization of ctDNA has been rapidly adopted as a predictive diagnostic in advanced NSCLC and indications in GI cancers may be emerging. While tissue-based assays have yields above 90%, there is less known about factors that influence the sensitivity of ctDNA for detecting mutations. Methods: We retrospectively evaluated mCRC patients (pts) who had plasma-derived NGS utilizing a highly-sensitive 68-73-gene assay. Tissue from prior resections or biopsies underwent concurrent 46-gene sequencing. In a case-control design, pts with a known mutation on tissue and radiologic evidence of metastatic disease but no detectable ctDNA mutation were matched 1:3 with randomly selected pts with detectable mutations and compared according to clinical, laboratory, and radiologic characteristics. A binary logistic regression was performed and Kaplan-Meier and Cox-proportional hazards models were used to compare overall (OS) and progression-free (PFS) survivals. Results: Of 427 mCRC pts who underwent ctDNA testing, 416 pts met inclusion criteria. Plasma-derived NGS did not find tumor mutations in 66 cases (15.9%); 198 pts with detectable alterations were selected as controls. After multivariate analysis, the lack of detection of ctDNA was associated with decreasing age (OR 0.94; 95%CI 0.91-0.98; p = 0.004), absence of liver (OR 0.19; 95%CI 0.08-0.45; p < 0.001) and lymph node metastases (OR 0.28; 95%CI 0.12-0.70; p = 0.006). A key determinant was timing of collection relative to disease status: plasma collected after evidence of progression was substantially more likely to have detectable alterations (OR 10.95; 95%CI 4.23-28.33; p < .001); in these pts, the modeled rate of detection was 98%, which increased to > 99% if the pts also had either liver or nodal disease. Pts with no detectable ctDNA had better OS (HR 0.38; 95%CI 0.21-0.69, p = 0.002) and PFS (HR 0.52; 95%CI 0.32-0.85; p = 0.009). Conclusions: When limited to ctDNA collected in newly diagnosed or recently progressing pts, the yield of ctDNA is 98%, which equals or exceeds the yield of tissue based testing. Our findings support the notion that ctDNA testing, when appropriately utilized, can replace tissue based testing and may provide prognostic information.

616 Background: For mCRC, the contribution of clinical and pathologic factors to concordance between formalin-fixed, paraffin-embedded (FFPE) tissue-based and ctDNA-based mutation profiling remains unclear. MSI is hypothesized to confer a higher rate of somatic alteration resulting in clonal evolution over time compared to microsatellite stable (MSS) patients, but this has not been previously confirmed. Methods: All mCRC patients were consented for a prospective genomic matching protocol (Assessment of Targeted Therapies Against Colorectal Cancer [ATTACC]) using CLIA-certified platforms. Archived tumor DNA from primary or metastatic tissue was sequenced on a 46- or 50-gene panel (Ion Torrent). Paired ctDNA samples were isolated from blood and sequenced with an ultra high-sensitivity assay (Guardant360). Mutation data were normalized by excluding genomic regions not covered on both platforms and were filtered to remove germline or synonymous variants. Results: An initial cohort of 139 patients was included in our analyses, with a median of two lines of intervening chemotherapy between FFPE and plasma DNA collection; 6 (4.3%) of the patients had MSI (MSI-H) mCRC. We detected 472 total mutations in either tissue or ctDNA, with a global concordance rate of 34.5%. Global concordance was not associated with tissue source, nor treatment with specific standard cytotoxic and/or biologic agents. Mutations detected in MSI-H CRC showed significantly greater discordance compared to MSS CRC (OR = 2.5, p = 0.025). This finding was validated using an independent cohort of 17 MSI-H mCRC patients (OR = 2.78, p = 0.00015). Among recurrently altered genes, we found that MSI-H cases were significantly more likely to have gained new TP53 mutations (OR = 8.17, p = 0.001) and to have lost PIK3CA mutations (OR = 8.04, p = 0.036) in ctDNA compared to MSS cases. Conclusions: MSI correlates with discordance between tissue DNA and ctDNA-based mutation profiling, which suggests that MSI-H CRC undergoes distinct patterns of clonal evolution including acquisition of new TP53 mutations. This may have implications for targeted and immunologic therapies in this unique population, and suggests a utility for repeated molecular testing.

4534 Background: Cell-free ctDNA may be potentially actionable, may have prognostic/predictive role and evolve after therapy. We updated our analysis of our retrospective study to shed light on UC/BC biology. Methods: Patients (pts) with UC/BC with ctDNA analysis for potentially actionable alterations using Guardant360 were identified. A 70-gene ctDNA next generation sequencing panel from a CLIA-licensed, CAP-accredited laboratory (Guardant Health, Inc.) offers complete exon sequencing for 29 cancer genes, critical exons in 39 genes and amplifications (16 genes), fusions (6 genes) and indels (3 genes) harvested from 10 mL of peripheral blood. Descriptive statistics were used. Results: There were 246 pts with 276 samples. At least 1 alteration was detected in 249 (90%) samples. Median age at time of ctDNA collection was 67 years (39-85), 78% men, median number of alterations per sample was 3.5 (1-35) most pts had MIBC. In MIBC pts, the most common alterations at the 1st ctDNA sample were in TP53 (52%), PI3KCA (18%) ARID1A (17%), FGFR2 (15%), MET & NF1 (14%), EGFR (13%), BRAF (12%), FGFR3 (11%), RAF1 (10%), BRCA1 & CCNE1 (9%). In MIBC pts, the most common genes with increased copy number were RAF1 & CCNE1 (8%), ERBB2 & PI3KCA (7%), EGFR, BRAF, FGFR1, MYC (each 5%), MET (4%), KRAS (3%). Most common altered pathways included TP53 signaling (56%), RAS/RAF/MEK/ERK (51%), RTK (48%), cell cycle (38%), FGFR family (34%), DNA damage response (25%), PI3KCA/AKT/mTOR (23%) and chromatin remodeling (17%). Interestingly, FGFR3 and RAS alterations were mutually exclusive in most cases, but each may co-occur with TP53alterations. 54 serial ctDNA samples from 24 pts (18 pts with 2 samples; 6 pts with 3 samples) revealed persistent, lost and new gene alterations. Conclusions: ctDNA was detected in 90% of pts and alterations were similar to those previously seen in UC tumor tissue. Tumor heterogeneity, interim therapy, genomic instability and clonal evolution can explain differences in serial samples. Correlation assessment with prior therapies and outcomes is being pursued to inform trial designs. Prospective validation, assessment of ctDNA concordance with tumor tissue DNA, and evaluation of clinical utility is warranted.

3522 Background: Preliminary data suggests that ctDNA can serve as a marker of minimal residual disease following colorectal cancer (CRC) tumor resection. Applicability of current ctDNA testing is limited by the requirement of sequencing known individual tumor mutations. We explored the applicability of a multi-gene panel ctDNA detection technology in CRC. Methods: Plasma was prospectively collected from CRC patients (pts) undergoing hepatic resections with curative intent between 1/2013 to 9/2016. In a blinded manner 5ml of preoperative (preop) and immediate post-operative (postop) plasma were tested using a novel 30kb ctDNA digital sequencing panel (Guardant Health) covering SNVs in 21 genes and indels in 9 genes based on the landscape of genomic alterations in ctDNA from over 10,000 advanced cancer pts with a high theoretical sensitivity (96%) for CRC. Median unique molecule coverage for this study is 9000 for cfDNA inputs ranging from 10 – 150 ng (media input preop = 27 ng, median input postop = 49 ng) with 120,000X sequencing depth on an IIlumina HiSeq2500. Results: A total of 54 pts underwent liver metastectomies with curative intent with a median follow-up of 33 months. Preop blood was a median of 49 days from last systemic chemotherapy and 3 days prior to surgery; postop blood was a median of 17 days after resection. Tumor mutations from standard of care hotspot multigene panel testing (at MDACC) were identified in 46 of 54 pts (85%). Preop ctDNA mutation detection rate was 80% (43/54) and 44% (24/54) in postop setting, with postop median allele frequency of 0.16% (range 0.01% to 20%). In pts with a minimum of 1 year follow up, sensitivity of postop ctDNA for residual disease was 58% (95%CI; 41%-74%), and specificity was 100% (66%-100%). In 43 patients who underwent successful resection of all visible disease, postop detection of ctDNA significantly correlated with RFS (P = 0.002, HR 3.1; 95% CI 1.7-9.1) with 2-year RFS of 0% vs. 47%. Recurrence was detected in ctDNA a median of 5.1 months prior to radiographic recurrence. Conclusions: The detection of postop ctDNA using an NGS panel-based approach is feasible and is associated with a very high rate of disease recurrence.

11593 Background: NGS of blood-derived ctDNA allows non-invasive tumor profiling. Liquid biopsy studies with clinical correlation have so far been mainly limited to small size cohorts. Methods: We performed comprehensive plasma genomic testing of ctDNA (NGS) in 670 patients (pts) (54-70 genes); Guardant Health, Inc.; (Clinical Laboratory Improvement Amendment certified and College of American Pathologists accredited). Results: The most represented cancers were gastrointestinal (31.8%), brain (22.7%), and lung (20.7%) (Table). Sixty-three percent of pts (N = 423) had ≥1 alteration. The most frequent alterations (characterized and variants of unknown significance (VUSs)) were in TP53 (32.5% of pts), followed by EGFR (13%), KRAS (12.5%), and PIK3CA (9.1%); for characterized alterations, the breakdown was 30.7% ( TP53), 7.6% ( EGFR), 12.2% ( KRAS), and 7.7% ( PIK3CA). Interestingly, 32% of brain tumors had ≥1 ctDNA alteration. Head and neck tumors were independently associated with a higher number of alterations (P=0.019). Forty-eight percent of pts (320/670) had potentially actionable alterations; in 241, (75% of 320), by an FDA-approved drug (mostly off label). Illustrative examples of clinical utility will be presented such as a patient with gastric cancer and EGFRamplification in ctDNA who received anti-EGFR treatment (60% regression), as well a patient with aggressive gynecologic malignancy who received immunotherapy based on a hypermutated ctDNA profile. Conclusions: Most pts, including a subset of those with brain tumors, demonstrated ctDNA alterations. Pts with head and neck tumors harbored higher numbers of alterations. Overall, three quarters of pts with alteration(s) had ≥1 aberration that could potentially be pharmacologically tractable, suggesting the need to further assess the utility of ctDNA in a therapeutic setting. [Table: see text]

e23068 Background: Analysis of cell-free circulating tumor DNA (ctDNA) by next-generation sequencing (NGS) is a promising method for detecting and monitoring earlier stage cancers. Liquid biopsies have a large potential to detect cancer earlier upon initial recurrence and to aid in adjuvant decision making through a non-invasive and highly accurate mechanism. Methods: We developed a 30kb ctDNA capture panel based on the landscape of genomic alterations in ctDNA of over 10,000 advanced cancer patients with high theoretical clinical sensitivity for colorectal (96%) and lung (87-93%) cancers. The panel was validated with high analytical PPV (86% at MAFs < 0.025%, 96% at MAFs > 0.025%). We applied the panel to a clinical study of 63 healthy donors and also 40 early stage (II/III) CRC and lung cancer patients with both pre- and post-tumor resection blood draws. Tumor samples were also collected at the time of the surgical resection. Results: Overall, the detection rate of ctDNA in pre-op blood draws was 72% (13/18) in CRC and 67% (12/18) in lung cancer patients. In the post-op blood draws ctDNA was detectable in 11% (2/18) of CRC and 11% (2/18) of lung cancer cases. For lung cancer patients, the median MAF in pre-op was 0.06% (0.01%-8.47%, n = 51) and in post-op was 0.06% (0.02%-4.14%, n = 38). For CRC patients, the median MAF in pre-op was 0.18% (0.01%-9.26%, n = 51) and in post-op was 0.36% (0.04%-9.5%, n = 18). Sixty-three healthy donor plasma samples were screened at similar cfDNA input amounts and nine (14%) had mutations; the majority of the mutations were also detected in matched leukocyte DNA. To assess a clinically relevant biological false positive background rate in lung cancer patients, screening of cancer-free smokers is ongoing. Conclusions: We have developed a highly sensitive and specific universal assay for the detection of ctDNA in early stage CRC and lung cancer patients without the requirement of a priori knowledge of tumor mutations. This technology allows for a promising non-invasive route for molecular monitoring of residual disease post-surgery and for early detection of relapse compared to traditional methodologies.

334 Background: Cell-free ctDNA may be potentially actionable, prognostic for outcomes and evolve after therapy. We conducted a retrospective study to evaluate these issues and shed light on UC biology. Methods: Patients (pts) with advanced UC who underwent ctDNA analysis for potentially actionable alterations using Guardant360 were identified. Data were requested for prognostic factors, current and prior therapies, TTF (time to treatment failure) and overall survival (OS). A 70-gene ctDNA next generation sequencing panel from a CLIA-licensed, CAP-accredited laboratory (Guardant Health, Inc.) offers complete exon sequencing for 29 cancer genes, critical exons in 39 genes and amplifications (16 genes), fusions (6 genes) and indels (3 genes) harvested from 10 mL of peripheral blood. Alterations were reported and association of non-synonymous potentially functional alterations with outcomes and prior therapy was examined. Results: There were 217 pts with 238 samples. ctDNA was detectable in 212 (89%) samples. Median age was 62 (range 39-91). The most common recurrent somatic mutations were in TP53 (n = 122, 57%) ARID1A (n = 47, 22%), NF1 (n = 35, 16%), FGFR2, FGFR3, and BRCA1 (n = 30, 14% each), MET (n = 27, 13%), ERBB2 (n = 25, 12%), PIK3CA (n = 24, 11%), and EGFR (n = 22, 10%). Most common genes with increased copy numbers were ERBB2 (n = 19, 9%) and RAF1 (n = 18, 8%). Clinical data were available for 64 pts, of whom 38 (59%) had prior chemotherapy. FGFR1 alterations noted in 3 of those pts (5%, 2 functional, 1 amplification) were associated with shorter OS (HR 2.95, p = 0.05). Pts with prior chemotherapy showed trend towards frequent alterations in DNA repair genes (45% vs. 23%, p = 0.11). Serial ctDNA profiling of 21 pts receiving therapy revealed the clonal evolution of mutations in BRCA2, NF1 and GATA3. Conclusions: ctDNA was very frequently detected in pts with advanced UC, and alterations were similar to those previously seen in muscle-invasive UC tumor tissue. FGFR1 alterations predicted for unfavorable outcome and DNA repair gene alterations evolved commonly after chemotherapy. Drugs exploiting these targets such as FGFR1 and PARP inhibitors may warrant exploration as salvage therapy in selected pts.

149 Background: Cell-free ctDNA may be prognostic and evolve following therapy. We report ctDNA profiling of patients (pts) with mCRPC and their association with clinical outcomes and evolution with therapy. Methods: Pts with mCRPC that underwent baseline ctDNA analysis for potentially actionable alterations using Guardant360 before new systemic therapy were identified. Data were requested for clinical factors, current and prior therapy, TTF (time to failure) and survival. A 70-gene cfDNA next generation sequencing panel from a CLIA-licensed, CAP-accredited laboratory (Guardant Health, Inc.) offers complete exon sequencing for 29 genes, critical exons in 39 genes and amplifications (16 genes), fusions (6 genes) and indels (3 genes) harvested from 10 mL of peripheral blood. Alterations were reported and association of alterations with outcomes and prior therapy was examined. Results: Of 514 with confirmed mCRPC, 482 (94%) had ≥ 1 ctDNA alteration. The median age was 70 years (range 39-91). The most common recurrent somatic mutations were in TP53 (36% of patients), AR (22%), APC (10%), NF1 (9%), EGFR, CTNNB1 and ARID1A (6% each) and BRCA1, BRCA2 and PIK3CA (5% each) The most common genes with increased copy numbers were AR (30%), MYC (20%) and BRAF (18%). Clinical outcomes were available for 163 pts, of whom 46 (28.8%) were untreated for mCRPC. A higher number of ctDNA alterations was associated with shorter TTF (HR: 1.05, p = 0.026). AR alterations had a trend for shorter TTF (HR: 1.42, p = 0.053) and survival (HR: 2.51, p = 0.09). Pts who received prior therapy had new alterations in AR (56% vs. 37%, p = 0.028) compared to untreated pts. Serial ctDNA profiling of 64 pts revealed the evolution of alterations in AR, BRCA1 and BRCA2 following therapy. Conclusions: ctDNA was frequently detected in patients with mCRPC, and alterations appear similar to tumor tissue alterations. A higher number of overall gene alterations and AR alterations appeared associated with poor clinical outcomes and new AR and BRCA alterations appeared following therapy. These data suggest that developing salvage agents targeting AR alterations and PARP inhibitors hold promise.

584 Background: ctDNA is shed into the bloodstream by tumor cells throughout the body, offering a non-invasive means of genomic testing, and a way to detect heterogeneous, subclonal genomic alterations present in distinct tumor lesions within an individual pt. However, a broad comparison of mutation prevalence in CRC ctDNA versus CRC tumor tissue has not yet been performed. Methods: ctDNA from 1397 CRC pts was analyzed using a CLIA-certified digital sequencing assay (Guardant360, Guardant Health) capable of detecting single nucleotide variants (SNV) in up to 70 genes, as well as selected insertions/deletions, amplifications, and fusions. Subclonal mutations were defined as mutations with mutant allele fractions (MAF) ≤ 50% of the greatest somatic MAF in the sample. Frequencies of mutations detected were compared to two large tissue-based sequencing databases (TCGA and NHS/HPS). Results: 1500/1772 (85%) tests had at least one genomic alteration (1397 unique pts). The most common SNV mutations included TP53 (62%), APC (47%), KRAS (39%), PIK3CA (17%), EGFR (11%), SMAD4 (11%), and BRAF (11%); these frequencies were comparable to rates in TCGA and NHS/HPS. In contrast, EGFR extracellular domain (ECD) mutations (42 pts) and JAK2 V617F mutations (16 pts) detected in ctDNA were not seen in tissue sequencing, reflecting acquired resistance to EGFR antibodies and clonal hematopoiesis of indeterminate clinical potential, respectively. 88% of pts with ECD mutations had at least one additional non-ECD resistance alteration detected in ctDNA (range 1-9, median 2.6), including KRAS, NRAS, BRAF, MAP2K1, MET and ERBB2. EGFR mutations were most likely to be detected as subclonal (86%), while mutations most likely to be clonal included KRAS (71%), TP53 (65%), BRAF (65%), and APC(63%). In 84 pts with serial monitoring, 87% had either gain (61%) or loss (63%) of clones over time. Conclusions: Blood-based genomic profiling can effectively detect common genomic alterations in CRC at comparable frequencies as observed in tissue and provide novel insights into tumor clonality and clonal dynamics. Clinical trials to target EGFR ECD mutations may be limited by the multiplicity of resistance mechanisms in each pt.

47 Background: Esophageal (EC), gastroesophageal junction (GEJ) and gastric cancer (GC), together GEC, have a poor prognosis with few targeted therapeutic options. As genomic profiling is becoming increasingly useful, we queried whether comprehensive genomic profiling of ctDNA would reveal relevant genomic alterations leading to targeted therapies. Methods: Patients (pts) with advanced GEC undergoing next-generation sequencing (NGS) of ctDNA in a CLIA certified lab were identified. ctDNA was analyzed using Guardant360, a digital NGS assay to identify single nucleotide variants, indels, amplifications and fusions in 54-70 genes. Results: 546 pts with GEC had ctDNA testing and 54 pts (10%) had more than one ctDNA test (range 2-6). Pts were similar across the 3 groups, with the exception of an increased male:female ratio (5:1) in the GEJ and EC cohorts. Mean age was 61.5 yrs (range 24-91). ctDNA alterations were detectable in 455 pts (83.3%). Recurrent alterations and therapeutic options for each cohort are shown in Table 1. Serial monitoring of ctDNA correlated with tumor markers, imaging and clinical response. Multiple antitumor responses to targeted therapy will be presented. Conclusions: ctDNA was detected in 83.3% of pts with advanced GEC with a frequency similar to TCGA. The diverse genomic landscape of GEC, obtained noninvasively, along with matched therapies have potential to improve outcomes for this aggressive disease. [Table: see text]

e23021Background: In non-small cell lung cancer (NSCLC), targeted treatment with first line tyrosine kinase inhibitors (TKIs) selects for the evolution of resistance at EGFR T790M in 50% of cases. ...

5035Background: Genomic profiling of metastatic PCa tissue is challenging and circulating tumor cells may not recapitulate entire tumor biology. Cell-free ctDNA is detectable in most patients (pts)...

LBA11501 Background: Next-generation sequencing (NGS) of circulating tumor DNA (ctDNA) enables non-invasive profiling of solid tumor cancers. Liquid biopsy studies to date have been limited to modest-size cohorts and case studies. Methods: Somatic genomic profiles of over 15,000 patients with advanced-stage clinical cancer were determined by a highly accurate, deep-coverage (15,000x) ctDNA NGS test targeting 70 genes (Guardant360). Frequencies of somatic ctDNA alterations per gene were compared to those previously described in tissue sequencing projects (e.g., TCGA). Accuracy of ctDNA sequencing (PPV) was assessed by comparing with matched tissue tests for 386 patients. Results: The cohort consisted of lung (37%), breast (14%), colorectal (10%) and other cancers (38%), with ctDNA clinical sensitivity of 86%, 83%, 85%, and 78%, respectively. Cancer-type-specific frequencies and mutual exclusivity patterns among major driver alterations largely recapitulated those seen in tissue sequencing studies. Mutation frequencies per codon correlated well between ctDNA and published tissue data, both for commonly altered tumor suppressors and for oncogenes (Pearson correlations: TP53, r = 0.94; KRAS, r = 0.99; PIK3CA, r = 0.99). The overall accuracy of ctDNA sequencing in comparison with matched tissue tests was 87% (336/386). The accuracy increased to 98% when blood and tumor were collected less than six months apart. Four distinct classes of clinical outcome benefits have been observed by liquid biopsy: 1) actionable mutations in cases with tissue QNS ( ALK fusion, or EGFR or BRAF activating mutations in lung; ERBB2 amp in gastric), 2) actionable resistance mutations at time of progression ( MET amp or EGFRT790M in lung), 3) evolution of sensitivity upon progression ( ERBB2-amplified metastatic breast cancer with triple negative primary), 4) under-genotyped tumors ( BRAFV600E or ERBB2indel in lung). Conclusions: Somatic alteration patterns in ctDNA samples largely agree with tissue alteration patterns, with the exception of resistance mutations. Clinical outcome benefits have been observed for patients treated based on ctDNA findings.

11576Background: A major advancement in patient diagnosis and monitoring is the use of ctDNA analysis using next-generation sequencing. Sequencing of patient samples using Guardant Health’s 70-gene...

e23097Background: Liquid biopsies were used to assess genomic alterations in plasma circulating tumor DNA (ctDNA) from patients with non-small cell lung adenocarcinoma (NSCLC), and therapeutic outc...

e23167Identification of novel EGFR ectodomain (ECD) mutations based on a database of clinical circulating cell-free DNA sequencing testsBackground: The Cancer Genome Atlas primarily sequenced early...

8553Background: Outcomes for curative-intent chemoradiotherapy of unresectable stage III non-small cell lung cancer (NSCLC) remains poor. Heterogeneity in tumors is correlated with therapeutic resi...

11541Background: Circulating tumor DNA (ctDNA) testing must include detection of single nucleotide variants (SNVs), indels, fusions, and copy number amplifications (CNAs) to identify all NCCN-recom...

11540Background: Analysis of cell-free circulating tumor DNA (ctDNA) by next-generation sequencing (NGS) is a promising method for early detection of cancer. Applications include screening high ris...

4528Background: Urothelial carcinoma of the bladder (UCB) exhibits one of the highest somatic mutation burdens. Circulating cell-free DNA (cfDNA) is obtained non-invasively from peripheral blood an...

11534Background: Analysis of cell-free circulating tumor DNA (ctDNA) by next-generation sequencing (NGS) is a promising method to identify druggable genomic alterations, especially when tumor speci...

358 Background: Urothelial carcinoma (UC) exhibits one of the highest somatic mutation burdens. Circulating cell-free DNA (cfDNA) is obtained non-invasively from peripheral blood and appears detectable in most patients (pts) with advanced UC. We report cfDNA profiling of patients with advanced UC using biopsy-free cfDNA sequencing. Methods: Twenty-nine patients with advanced UC that underwent cfDNA analysis using Guardant360 were identified. A 68-gene cfDNA next generation sequencing (NGS) panel from a CLIA-licensed, CAP-accredited laboratory (Guardant360) offers complete sequencing for all exons in 29 cancer associated genes, as well as the critical exons in 39 other genes, and copy number amplifications (16), fusions (4) and indels (1) in selected genes, harvested from 20 mL of peripheral blood. The mutant allele fraction (MAF) is defined as the number of mutated cfDNA molecules divided by the number wild type cfDNA molecules at a given nucleotide position. Results: Of 29 patients with advanced UC, cfDNA was detectable in 25 patients (86.2%). Twenty-nine patients were available with bladder primary in 27 pts (93.1%), and the median age was 75 years (range 52-91). The most common recurrent somatic mutations in the 25 patients with ≥ 1 alteration were in TP53 (n = 17), BRCA1/2 (n = 6), FGFR2 (n = 5), EGFR (n = 5), PIK3CA (n = 4), APC (n = 4), KRAS (n = 4), RAF1/BRAF (n = 4), and FGFR3, ALK, PDGFRA, NF1, CDKN2A and ARID1A were observed in 3 patients each. The most common genes with ≥ 2 copy numbers in the 25 evaluable patients were ERBB2 (n = 2) and PIK3CA, EGFR, RAF1, and FGFR1 in one patient each. Analysis of additional patients is ongoing in this expanding dataset. MAF data will be presented. Conclusions: cfDNA is frequently detected in patients with advanced UC, and alterations appear similar to those previously reported from UC tumor tissue. The correlation of cfDNA profiling with tumor tissue profiling within patients and ability to predict response requires study. Given that cfDNA offers a non-invasive means of profiling tumor DNA and MAF, further development of this promising modality is warranted to guide therapy with biologic agents and immunotherapy, and serial monitoring may provide insights regarding tumor biology.

11004 Background: Analysis of cell-free DNA using next generation sequencing (NGS) is a potentially powerful tool for the detection/monitoring of circulating tumor DNA (ctDNA). We report the findin...

11072 Background: Glioblastoma is the most aggressive type of primary brain tumor with a median survival of 15 months and limited therapy options. Trials of genomically targeted matched therapies f...

e12540 Background: The NEXT-2 trial is cfDNA genomics master protocol to enroll refractory cancer patients and to guide matched trial according to cfDNA genomics of 56 genes. Methods: Using Digital...

107 Background: Pancreatobiliary carcinomas (PC) carry a poor prognosis but have not yet benefitted from the revolution in precision oncology, in part because biopsy tissue is often inadequate for ...

3604 Background: Concordance of somatic mutations between primary and metastatic colorectal tumors is high in colorectal cancer (CRC), circulating cell-free DNA (cfDNA) accurately reflects the mutation status of metastatic disease. Intervening treatment and metastatic dissemination may result in clonal evolution from pre-existing intratumoral heterogeneity, with the resulting discordance in minor clones.Methods: Patients (pts) were prospectively consented for collection of cfDNA from plasma samples for sequencing on a 54-gene platform optimized for very low allele frequencies (Guardant360), with concurrent sequencing of clinically available and macrodissected historic FFPE tissue (50-gene Ion-Torrent panel), both performed in CLIA-compliant labs. Eligible cases had plasma and tissue sequencing where the modal mutation was present with ≥ 1% and ≥ 5% allele frequencies, respectively. Minor clones were defined as those with an allele frequency < 10% of the highest detected allele frequency in the sample.Resu...

3601 Background: Circulating cell-free DNA (cfDNA) isolated from the plasma of cancer patients has been shown to accurately reflect the genomic mutation profile of the tumor, and assays are now ava...

572 Background: MET inhibition is emerging as a potent therapeutic strategy and MET gene amplification has shown predictive significance. MET amplification rate in mCRC, as previously reported in literature, varies from 9% in primary to 18% in metastases but intermixes increased copy number from chromosomal level aberrations with focal gene amplification. Validation of MET amplification rate in mCRC is needed. Methods: We performed analyses of MET amplification in mCRC patients (pts) (n = 636) across multiple cohorts. Cohort 1 (n = 103) included tissue microarray from liver metastases analysed using fluorescence in situ hybridization (FISH) [cMET and CEP7 probes, MET/CEP7 ratio > 2]. Cohort 2 (n = 205) included pts referred for phase I trials who had MET amplification testing using FISH. Cohort 3 (n = 279) included cases sequenced with HiSeq (Illumina) with full exome coverage for 202 genes including MET (average depth 800) with focal gene amplification (≥ 4 copies) identified by an in-house algorithm. Cohort 4 (n = 49) included pts refractory to EGFR monoclonal antibodies enrolled in the ATTACC (a prospective molecular screening) program for mCRC, in whom plasma circulating-free DNA (cfDNA) was analyzed by Guardant sequencing technology. Results: In tissue based analyses, focal MET amplification rate was 1.7% and was higher in primary tumors compared to metastases [3.1% (9/291) vs. 0.4% (1/288), p = 0.02] [Table]. In cohorts 2 & 3 MET amplification was found in 4 [MET/CEP7: 2.1 – 7.7; primary (4/130), metastases (0/75)] and 6 [copy number: 4.0 – 6.7; primary (5/161), metastases (1/110)] cases, respectively. MET amplification rate in pts who had progressed on anti-EGFR therapy was 14.3% (Table). Conclusions: Contrary to prior reports, in this large cohort, MET amplification was a rare event in mCRC pts and the rate was not higher in metastatic sites. However, MET amplification occurred in a sizable subset of pts refractory to anti-EGFR therapy as identified by cfDNA analysis. MET amplification appears to play a minor role in de novo colorectal carcinogenesis but may play an important role in acquired resistance to anti-EGFR therapy. [Table: see text]

11093 Background: MBC is an incurable disease with complex molecular features including somatic mutations that evolve in relation to genomic instability and selective treatment pressure. Circulatin...

e22041 Background: Analysis of genomic alterations in advanced malignant disease is quickly becoming the standard of care in oncology. Biopsies are risky, often costly and may not capture genetic c...

11117 Background: cfDNA allows non-invasive assessment of gene mutations, but has not been reported with copy number alterations. Methodologies to integrate mutation and copy number would allow imp...