Clinical relevance of circulating tumour (ct)DNA genotyping for first-line cetuximab-based treatment monitoring in metastatic colorectal cancer (mCRC): A prospective multicentric study

abstracts Legal entity responsible for the study: The authors. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest. 623P

Plasma clearance of RAS mutation under therapeutic pressure is a rare event in metastatic colorectal cancer

Background: In metastatic colorectal cancer (mCRC), circulating tumour DNA (ctDNA) monitoring can be used to genotype tumors and track clonal evolution. We investigated the clearance of RAS mutated clones under chemotherapy pressure by ctDNA analysis in patients with a RAS mutated mCRC. Methods: Patients with a RAS mutated tumour included in the prospective PLACOL study, were monitored for ctDNA. Analyzes were based on optimized targeted next generation sequencing and/or droplet-based digital PCR (ddPCR). For plasma samples without detectable mutations, we tested the methylation status of WIF1 and NPY genes using methylation-ddPCR (met-ddPCR) to validate the presence of ctDNA. Results: Among the 36 patients with positive plasma samples for RAS mutations at inclusion, 28 (77.8%) remained RAS positive at disease progression, and 8 (22.2%) became negative. Subsequent met-ddPCR for methylated markers showed that only 2 out of the 8 patients with RAS negative plasma had detectable ctDNA at progression. Therefore, only two samples among 36 were confirmed for clearance of RAS mutation in our series. Conclusions: This study suggests that the clearance of RAS mutations in patients treated by chemotherapy for a RAS mutated mCRC is a rare event. Monitoring tumor mutations in plasma samples should be combined with a strict control of the presence of ctDNA. The therapeutic impacts of RAS clearance need to be further explored. Legal entity responsible for the study: The authors. Funding: The Ministe`re de l’Enseignement Supe´rieur et de la Recherche, the Universite´ Paris-Descartes, the Centre National de la Recherche Scientifique (CNRS), the Institut National de la Sante´ et de la Recherche Me´dicale (INSERM), the Institut National du Cancer (INCA, n 2009-1-RT-03-US-1 and 2009-RT-03-UP5-1), the Association pour la recherche contre le cancer (ARC, no. SL220100601375), the Agence Nationale de la Recherche (ANR Nanobiotechnologies; no. ANR-10-NANO-0002-09), the SIRIC CARPEM, the ligue nationale contre le cancer (LNCC, Program “Equipe labelise´e LIGUE”; no. EL2016.LNCC/VaT) and Advanced Merieux Research Grant and canceropole funding (no. 2011-1-LABEL-UP5-2). Disclosure: H. Blons: Advisory / Consultancy: AstraZeneca; Advisory / Consultancy: Merck; Advisory / Consultancy: Amgen; Advisory / Consultancy: Boehringer Ingelheim; Advisory / Consultancy: Pfizer. V. Taly: Advisory / Consultancy: Servier; Advisory / Consultancy: Boehringer Ingelheim; Advisory / Consultancy: Raindance Technologies. W. Shu-Fang: Research grant / Funding (institution): Servier. S. Pernot: Advisory / Consultancy: Amgen; Advisory / Consultancy: Sanofi. J. Taieb: Advisory / Consultancy, Speaker Bureau / Expert testimony: Merck; Advisory / Consultancy, Speaker Bureau / Expert testimony: Roche; Advisory / Consultancy, Speaker Bureau / Expert testimony: Amgen; Advisory / Consultancy, Speaker Bureau / Expert testimony: Eli Lilly; Advisory / Consultancy, Speaker Bureau / Expert testimony: Sanofi; Advisory / Consultancy, Speaker Bureau / Expert testimony: Celgene; Advisory / Consultancy, Speaker Bureau / Expert testimony: Sirtex; Advisory / Consultancy, Speaker Bureau / Expert testimony: Shire; Advisory / Consultancy, Speaker Bureau / Expert testimony: MSD; Advisory / Consultancy, Speaker Bureau / Expert testimony: Servier; Advisory / Consultancy, Speaker Bureau / Expert testimony: Pierre Fabre. P. Laurent-Puig: Advisory / Consultancy: Amgen; Advisory / Consultancy: Merck-Serono; Advisory / Consultancy: Sanofi; Advisory / Consultancy: Boehringer Ingelheim; Advisory / Consultancy: Roche; Advisory / Consultancy: Lilly. A. Zaanan: Advisory / Consultancy: Amgen; Advisory / Consultancy: Baxter; Advisory / Consultancy: Lilly; Advisory / Consultancy: Merck-Serono; Advisory / Consultancy: MSD; Advisory / Consultancy: Sanofi; Advisory / Consultancy: Roche; Advisory / Consultancy: Servier. All other authors have declared no conflicts of interest.

624P

High circulating miR-1247 is a marker for poor prognosis in patients with metastatic colorectal cancer treated with chemotherapy and cetuximab

J.V. Schou1, Z. Sztupinszki2, J.S. Johansen1, D.L. Nielsen1, B. Glimelius3, C. Qvortrup4, H. Sorbye5, B. Vittrup Jensen6, P. Pfeiffer7 1 Department of Oncology, Herlev Hospital, Herlev, Denmark, 2Danish Cancer Society, Copenhagen, Denmark, 3Section of Experimental and Clinical Oncology, Department of Immunology, Genetics and Pathology, Uppsala, Sweden, 4Department of Oncology, Finsen Centre - Rigshospitalet, Copenhagen, Denmark, 5Oncology, Haukeland Universitetssykehus, Bergen, Norway, 6Department of Oncology, Herlev and Gentofte Hospital, Herlev, Denmark, 7Experimental Research in Medical Cancer Therapy, Odense University Hospital, Odense, Denmark Background: MicroRNAs (miRs) are small non-coding RNAs. Circulating miRs in plasma are deregulated in patients with metastatic colorectal cancer (mCRC). The aim

v236 | Gastrointestinal Tumours, Colorectal

was to find prognostic circulating miRs for overall survival (OS) in patients with mCRC treated with chemotherapy and cetuximab. Methods: A prospective biomarker study was conducted. A full plasma miR-panel was tested with TaqMan Human MicroRNA assay in 151 patients treated with 3rd line irinotecan and cetuximab (Discovery cohort, Biweekly study). Significant miRs were validated with Fluidigm arrays, in 95 patients, treated with 1st line chemotherapy and cetuximab (Validation cohort, NORDIC-7.5 study1). Cox regression analysis was conducted, and multivariate analyses performed with adjustment for sidedness, number of metastatic sites, RAS/BRAF mutation status, age and performance status. Consecutive samples after 3 months of treatment were available for 117 and 86 patients in the discovery and validation cohorts, respectively. Results: Baseline high expression of miR-1247 in the discovery and validation cohorts, was significantly associated with short OS (hazard ratio [HR]) ¼ 2.33, p ¼ 0.0005) based on the multivariate analysis in the validation cohort. High miR-1290 expression after 3 months was prognostic for short progression free-survival (HR ¼ 1.57, p ¼ 0.0465) in both cohorts. In both cohorts, we found a statistically significant decrease in, miR-1290 (p ¼ 0.0051), miR-200b (p ¼ 0.036), and miR-375 (p ¼ 0.00005), from baseline to 3 month-sample, in patients with partial/complete (n ¼ 78) response compared to patients with progressive disease (n ¼ 72). Conclusions: High expression of miR-1247 was prognostic for short OS in the multivariate analysis in a discovery and validation cohort of patients treated with chemotherapy and cetuximab in both first and third line. A decrease in miR-1247 and miR-1290 during treatment was associated with favorable response. 1Pfeiffer et al, Clin Colorectal Cancer 2015. Legal entity responsible for the study: Jakob Vasehus Scho. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest.

625P

Clinical relevance of circulating tumour (ct)DNA genotyping for firstline cetuximab-based treatment monitoring in metastatic colorectal cancer (mCRC): A prospective multicentric study

J. Vidal Barrull1, D. Casadevall2, M.C. Fernandez3, M. Castro-Henriques1, P. GarcıaAlfonso4, B. Martin-Cullell5, V. Alonso-Orduna6, R.M. Rodrıguez Alonso7, M. Benavides8, C. Santos Vivas9, E. Elez Fernandez10, R. Garcia-Carbonero11, R. Ferreiro12, J.L. Manzano13, F. Losa14, R. Salazar15, J. Tabernero16, E. Aranda Aguilar17, B. Bellosillo18, C. Montagut Viladot1 1 Medical Oncology, Hospital del Mar, Barcelona, Spain, 2Oncology, University Hospital del Mar, Barcelona, Spain, 3Pathology Department, Hospital del Mar, Barcelona, Spain, 4 Servicio de Oncologıa Me´dica, General Universitario Gregorio Mara~ n on, Madrid, Spain, 5 Medical Oncology, Hospital de la Santa Creu i Sant Pau, Barcelona, Spain, 6Medical 7 Oncology Service, Hospital Miguel Servet, Zaragoza, Spain, Medical Oncology, IMIBIC, Reina Sofıa Hospital, University of C ordoba, Cordoba, Spain, 8Oncology, Universitario Regional Virgen de la Victoria, Malaga, Spain, 9Oncologia Radioterapica, Catalan Institute of Oncology. Hospital Druan y Reynals, Barcelona, Spain, 10Department Servicio de Oncologıa Me´dica, Vall d’Hebron University Hospital, Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain, 11Medical Oncology, University Hospital 12 de on y Cajal, Octubre, Madrid, Spain, 12Medical Oncology, Hospital Universitario Ram Madrid, Spain, 13Oncology Medical, Hospital Universitario Germans Trias i Pujol, 14 Badalona, Spain, Medical Oncology Department, Hospital de Sant Joan Despı Moise`s a d’Oncologia Hospital Broggi, Barcelona, Spain, 15Medical Oncology, Institut Catal Duran i Reynals, Barcelona, Spain, 16Oncology, Vall d’Hebron University Hospital and Vall d’Hebron Institute of Oncology (VHIO), Barcelona, Spain, 17Oncology, University Hospital Reina Sofia, Cordoba, Spain, 18Pathology Department, Hospital del Mar, Barcelona, Spain Background: Genotyping of ctDNA characterizes the molecular profile and monitors tumor molecular dynamics, but the clinical applicability of next generation sequencing (NGS) DNA sequencing has not been assessed in a large prospective cohort of mCRC patients (pts). Methods: mCRC pts with RAS wt tumors treated in first line with chemotherapy (CT) þ cetuximab in 16 Spanish hospitals were included. Plasma samples were collected baseline (BL) and every 2 cycles until progression. ctDNA was isolated and sequenced with NGS platform Oncomine Colon cfDNA Assay which identifies hotspot mutations (mut) in AKT1, BRAF, CTNNB1, EGFR, ERBB2, FBXW7, GNAS, KRAS, MAP2K1, NRAS, PIK3CA, SMAD4, TP53 and APC. RAS ctDNA mut were also assessed by dPCR (BEAMing) at BL. Detection of any mut in plasma was considered as ctDNAþ. Mut in BRAF/RAS/MAP2K1/PIK3CA were considered as anti-EGFR resistance. Results: 101 pts were enrolled (65.3% male; median age 64.5y; 77.6% left colon). RAS mut were detected in 11 pts BL by BEAMing. After quality assessment, 84 samples were sequenced by NGS BL and at least one mut was detected in 65 samples (77.3%; median 1, range 0-5) and 10 RAS mut were detected. Plasma samples from 47 pts were also analyzed after 4 cycles of treatment (C4) and mut were detected in 30 pts (63.8%; median 1, range 0-3). Median mutant allele fraction for paired samples was 24.9 at BL and 0.56 at C4 (p ¼ 0.0045). Treatment clinical benefit (CB: PR, CR and SD  16weeks) was observed in 72 out of 85 pts. BL ctDNAþ or C4 ctDNA change were not associated with CB (p ¼ 0.722). Assessment of RAS BL mut was not associated with CB (p ¼ 0.243 by NGS and p ¼ 0.712 by BEAMing), while the absence of RAS mut at C4 predicted treatment CB (93.2% vs. 6.8% RAS mut p ¼ 0.02). Median PFS for all pts was 10.13 month. ctDNAþ at BL or C4 were not correlated with PFS (HR 0.8; p ¼ 0.73). No differences in

Volume 30 | Supplement 5 | October 2019

Downloaded from https://academic.oup.com/annonc/article-abstract/30/Supplement_5/mdz246.102/5577857 by guest on 26 October 2019

E. Moati1, H. Blons2, V. Taly3, F. Garlan3, W-R. Shu-Fang3, D. Pietrasz3, A. Didelot3, S. Garrigou3, A. Saint4, S. Pernot1, J. Taieb1, P. Laurent-Puig3, A. Zaanan1 1 Department of Gastroenterology and Digestive Oncology, European Georges Pompidou Hospital, Paris, France, 2Molecular Oncology, Hopital European George Pompidou, Paris, France, 3Sorbonne Universite´, USPC, Universite´ Paris Descartes, Universite´ Paris Diderot, Centre de Recherche des Cordeliers, INSERM, CNRS, Paris, France, 4Oncologie Digestive, Hopital European George Pompidou, Paris, France

Annals of Oncology

abstracts

Annals of Oncology PFS were found in ctDNA RAS mut vs RAS wt BL (BEAMingþ HR ¼ 1.2 p ¼ 0.79 and NGSþ HR ¼ 2.6 p ¼ 0.052). Among all anti-EGFR resistant mut, only ctDNA RAS mut were detected at C4 (N ¼ 4). Persistence or emergence of RAS mut at C4 were strongly associated with PFS (12.9m vs. 4.2. HR 8.8 p ¼ 0.0003). Conclusions: Early RAS ctDNA dynamics predicts benefit to first line CTþcetuximab in mCRC pts. Legal entity responsible for the study: Grupo de Tratamiento de los Tumores Digestivos (TTD), Spain. Funding: Merck-Serono. Disclosure: J. Vidal Barrull: Honoraria (self), Advisory / Consultancy: Merck-Serono; Advisory /

628P 626P

Clonal hematopoiesis mutations in plasma cfDNA RAS/BRAF genotyping of metastatic colorectal cancer 1

1

1

1

1

1

2

1

1

B. Wang , F. Huang , M. Shen , S. Wu , H. Wang , H. Jiang , Y. Yu , Q. Yu , Y. Yang , Y. Zhao1, Y. Zhou1, B. Pan1, T. Liu2, W. Guo1 1 Laboratory Medicine, Zhongshan Hospital, Fudan University, Shanghai, China, 2 Medical Oncology, Zhongshan Hospital, Fudan University, Shanghai, China Background: Clonal hematopoiesis (CH) leads to blood-derived somatic mutations in KRAS, NRAS and BRAF. Our aim is to identify the prevalence of CH-derived mutations in these three genes in metastatic colorectal cancer (mCRC) patients and reveal the practical clinical implication of these mutations on plasma cell-free DNA (cfDNA) genotyping. Methods: We analyzed KRAS, NRAS and BRAF genotypes in plasma and matched tumor tissues of 236 mCRC patients through next-generation sequencing (NGS). Suspected CH mutations were defined as those only present in plasma with variant allelic frequencies (AFs) <5% and were confirmed by paired peripheral blood cells (PBCs) using droplet digital PCR (ddPCR). The hemopoietic lineage harboring a CHderived mutation was analyzed through flow cytometry. Results: We identified suspected CH mutations in twenty patients (8.4%, 20/236). Three of these patients (1.27%, 3/236) had a CH-derived KRAS mutation. Two of them had a KRAS G12X and the third had a KRAS Q61H. We did not detect CH-derived NRAS or BRAF mutations. Patients harboring a CH-derived mutation previously received chemotherapy treatment. In one CH-derived KRAS G12X case, the mutation was enriched in lymphocytes and persisted in cfDNA over the course of 4 months of therapy. Conclusions: We confirmed the existence of CH-derived KRAS mutations in a small proportion of mCRC patients. This should be noted to prevent misclassification as tumor somatic mutations when performing cfDNA sequencing. Legal entity responsible for the study: The authors. Funding: The National Natural Science Foundation of China (81572064, 81772263, 81772511, 81602038), the Key Developing Disciplines of Shanghai Municipal Commission of Health and Family Planning (2015ZB0201), the Projects from the Shanghai Science and Technology Commission (16411952100), the Projects from Zhongshan Hospital, Fudan University (2018ZSLC05). Disclosure: All authors have declared no conflicts of interest.

627P

Immune status of patients with different stages of colorectal cancer with and without circulating tumour cells

A.O. Sitkovskaya1, E.Y. Zlatnik1, I.A. Novikova2, E.S. Bondarenko1, A.B. Sagakyants1, O.I. Kit2, L.Y. Vladimirova3 1 Laboratory of Immunophenotyping of Tumors, Rostov Research Institute of Oncology, Rostov-on-Don, Russian Federation, 2Administration, Rostov Research Institute of Oncology, Rostov-on-Don, Russian Federation, 3Medical Department, Rostov Research Institute of Oncology, Rostov-on-Don, Russian Federation

Volume 30 | Supplement 5 | October 2019

Detection of 5-hydroxymethylcytosine in circulating-free DNA for early diagnosis of colorectal cancer

T. Liu1, W. Chang1, W. Ye2, G. He1, L. Ren1, W. Tang1, J. Chen3, J. Xu3 General Surgery, Zhongshan Hospital, Fudan University, Shanghai, China, 2Colorectal Cancer Center; Department of General Surgery, Zhongshan Hospital, Fudan University, Shanghai, China, 3Liver Surgery, Zhongshan Hospital, Fudan University, Shanghai, China

1

Background: A convenient and accurate method to improve the effectiveness of colorectal-cancer screening is still under exploration. This may be achieved by a diagnostic model based on 5-hydroxymethylcytosine (5-hmC). Methods: We collected peripheral venous blood from 285 colorectal cancer patients and 285 healthy people at Zhongshan Hospital, Shanghai. The 5-hmC level of each gene locus in circulating-free DNA (cfDNA) was detected using a highly sensitive sequencing method. Of 285 subjects, 178 in each group were randomly selected as training set. The first 100 gene loci with the most significant difference of 5-hmC level were selected for establishing a diagnostic model (Logistic regression model) based on machine-learning method. The rest subjects in each group were used as validating set to validate this model. Further, we analysed the influence of several clinicopathological factors on 5-hmC level. Results: In training set, the comprison of the 5-hmC level suggested that the expression of 5-hmC in cfDNA was significantly different between the two groups. Of the 100 gene loci, 35 were included in the model. In internal validation, the sensitivity and specificity of the model were 88.8% and 89.9%, respectively. While the sensitivity and specificity 81.3% and 89.7%, respectively, in external validation using the validating set. The area under ROC curve of external validation was 0.946. For colorectal cancer stages I-IV, the diagnostic sensitivities were 78.7%, 81.9%, 88.2% and 95.6%, respectively. We also found that age, gender, Ras status or the site of primary tumor had no significant effect on the expression of 5-hmC. However, there was a statistically difference of 5-hmC level between stage I-III and stage IV. Conclusions: Our study identified the gene loci with significantly different 5-hmC level between colorectal cancer patients and healthy people, then we established a diagnostic model of colorectal cancer with high sensitivity and specificity. However, the utility of the model still need further verification in large-scale screening experiments. Clinical trial identification: NCT03599947 09/16/2018. Editorial acknowledgement: Shanghai Yibien Gene Technology Co., Ltd. Provides Testing Platform. Legal entity responsible for the study: The authors. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest.

629P

Detection of 5-hydroxymethylcytosine in circulating-free DNA for prediction of the efficacy of conversion therapy for colorectal cancer liver metastases

W. Chang1, T. Liu1, W. Ye2, L. Ren1, G. He1, J. Xu2

doi:10.1093/annonc/mdz246 | v237

Downloaded from https://academic.oup.com/annonc/article-abstract/30/Supplement_5/mdz246.102/5577857 by guest on 26 October 2019

Consultancy: Amgen; Honoraria (self): Sysmex-Inostics. E. Elez Fernandez: Honoraria (self), Advisory / Consultancy: Amgen; Honoraria (self), Advisory / Consultancy, Research grant / Funding (self): Merck-Serono; Honoraria (self), Advisory / Consultancy: Roche; Honoraria (self), Advisory / Consultancy: Sanofi. F. Losa: Honoraria (self): Amgen; Honoraria (self): Merck-Serono. R. Salazar: Honoraria (self), Advisory / Consultancy: Amgen; Honoraria (self), Advisory / Consultancy: MerckSerono; Advisory / Consultancy: Guardant-Helth; Honoraria (self), Advisory / Consultancy: Roche; Honoraria (self), Advisory / Consultancy: Ferrer; Advisory / Consultancy: Pfizer; Honoraria (self), Advisory / Consultancy: Novartis; Advisory / Consultancy: Lilly; Honoraria (self), Advisory / Consultancy: MSD. J. Tabernero: Honoraria (self), Advisory / Consultancy: Array Biopharma; Honoraria (self), Advisory / Consultancy: Merck-Serono; Honoraria (self), Advisory / Consultancy: Molecular Partners; Honoraria (self), Advisory / Consultancy: Sanofi; Honoraria (self), Advisory / Consultancy: Roche; Honoraria (self), Advisory / Consultancy: Novartis; Honoraria (institution), Advisory / Consultancy: MSD; Honoraria (self), Advisory / Consultancy: Foundation Medicine; Honoraria (self), Advisory / Consultancy: Halio DX SAS. E. Aranda Aguilar: Honoraria (self): Celgene; Honoraria (self): Merck-Serono; Honoraria (self): Roche; Honoraria (self): Sanofi. B. Bellosillo: Advisory / Consultancy: AstraZeneca; Advisory / Consultancy: Merck-Serono; Advisory / Consultancy: Roche; Honoraria (self), Advisory / Consultancy: Biocartis; Advisory / Consultancy: Novartis; Advisory / Consultancy: Thermofisher. C. Montagut Viladot: Honoraria (self), Advisory / Consultancy: Amgen; Honoraria (self), Advisory / Consultancy, Research grant / Funding (self): Merck-Serono; Advisory / Consultancy: Sysmex-Inostics; Advisory / Consultancy: Sanofi; Honoraria (self), Advisory / Consultancy: Roche; Advisory / Consultancy: Bayer; Honoraria (self), Advisory / Consultancy: Biocartis; Honoraria (self), Advisory / Consultancy: BMS; Honoraria (self), Advisory / Consultancy: Guardant-Helth; Advisory / Consultancy: Symphogen. All other authors have declared no conflicts of interest.

Background: Our purpose was to study parameters of the cellular immunity in patients with different stages of colorectal cancer with and without circulating tumor cells (CTCs). Methods: The presence of CTCs and parameters of the cellular immunity were studied in 60 colorectal cancer (CRC) patients: stage II – 20, stage III – 12, stage IV – 28 patients, adenocarcinoma in all cases. Blood levels of CTCs were determined before the treatment using the CellSearch System (Janssen Diagnostics, LLC) with iron microparticles coated with antibodies to epithelial cell adhesion markers EpCAM, CD45 and cytokeratins 8,18,19. Subpopulations of lymphocytes were studied using the FACSCantoII flow cytometer (BD): percentage of T-B-NR-cells, Tregs; activated CD4þ and CD8þ cells, naive T-lymphocytes and memory T cells; expression of CD335, perforin and granzyme B was evaluated in NR cells. Results: The results revealed a number of differences in immunological parameters in patients with and without CTCs. Stage II CRC with CTCs showed statistically significantly higher content of lymphocytes (19.162.7 vs 15.661.4%), CD335þ NK cells (44.967.6 vs 19.264.0%) and the respiratory burst index in granulocytes (97.561.6 vs 92.361.2%); p < 0.05 in all cases. Stage III CRC with CTCs was characterized by decreased lymphocyte and monocyte levels, while granulocyte content increased. These patients demonstrated elevated levels of CD4þ cells with early and late activation markers: CD38þ and HLA-DRþ by 2 times, CD25þ by 6 times, p < 0.05 in all cases. CTC-positive patients with stage IV CRC, compared to CTC-negative ones, showed lower levels of NK cells (16.662.0 vs 28.465.5%), CD4þCD69 þ (5.160.8 vs 8.261.0%), and CD8þCD25 þ (2.860.48 vs 5.561.1%; p < 0.05). Conclusions: CTCs in locally advanced CRC is associated with stimulation of functional parameters of innate immunity and activated Th, while in metastatic CRC - with inhibition of cytotoxic lymphocytes of innate and adaptive immunity, which, apparently, is associated with poor prognosis. Legal entity responsible for the study: The authors. Funding: Has not received any funding. Disclosure: All authors have declared no conflicts of interest.