Search

ESCC prediction accuracy using various number of DMRs identified from ESCC vs. healthy plasma sample analysis

PDF file, 56K, Methylation in multiple metachronous tumors.

Supplementary tables

Supplementary Table-S4

Hierarchical clustering of HCC and healthy plasma samples

Supplementary Table Legends

Supplementary Figure 1: Recurrence prediction comparison between CMS and MRC in the A) TCGA discovery and B) Validation cohorts.

Supplementary Table 2: Data of miRNAs which are analyzed in the TCGA discovery cohort with associated statistical parameters.

Supplementary Figure 4: MRC significantly predicted benefit of adjuvant chemotherapy in microsatellite stable stage III colorectal cancers KM plots depicted in panel A, B, C and D belong to MRC high risk category, while the KM plots in panel E, F, G and H belong to MRC low risk category. KM plots in panel A and E represents the analysis performed on combined stage II and III CRC patients, while panel B and F represents stage II analysis and panel C and G represents stage III patient's analysis separately. KM plots depicted in panel D and H represents the analysis performed on microsatellite stable stage III CRC patients.

Hierarchical clustering of GC and healthy plasma samples

Supplementary data

Purpose:Emerging evidence indicates that gut microbiome plays a crucial role in the cancer pathogenesis. Although Fusobacterium nucleatum (F. nucleatum) is associated with poor prognosis in multiple cancers, its clinical significance in predicting response to chemotherapy in patients with esophageal squamous cell carcinoma (ESCC) remains unclear.Experimental Design:The F. nucleatum levels were quantified by qPCR assays in tumor tissues from 551 patients with ESCC from two independent cohorts, including 101 patients who received neoadjuvant chemotherapy prior to curative resection. Associations between F. nucleatum burden and recurrence-free survival (RFS), as well with chemotherapeutic response were evaluated using response evaluation criteria in solid tumors (RECISTs), primary tumor metabolic response defined by maximum standardized uptake value (SUVmax) changes in positron emission tomography-CT (PET/CT), and pathologic tumor regression grade (TRG).Results:High burden of F. nucleatum in patients with ESCC associated with poor RFS in both training [log-rank P = 0.02; HR = 1.61; P = 0.03] and validation cohorts (log-rank P = 0.003; HR = 1.96; P = 0.004). Importantly, patients with ESCC with high levels of F. nucleatum displayed poor chemotherapeutic response for all three evaluation methods: RECIST (P = 0.04), SUVmax change in PET/CT (P = 0.0004), and TRG (P = 0.003).Conclusions:We conclude that high levels of intratumoral F. nucleatum have a prognostic significance for predicting poor RFS in patients with ESCC. More importantly, our data indicates that higher F. nucleatum burden correlates with poor response to neoadjuvant chemotherapy, suggesting the possibility that an antibiotic intervention against this bacterium may significantly improve therapeutic response in patients with ESCC.

PDF file, 43K, Probability Value plot of the 3-plex methylation assay.

Purpose:Pancreatic ductal adenocarcinoma (PDAC) is a lethal disease with dismal survival rates. Tumor microenvironment (TME), comprising of immune cells and cancer-associated fibroblasts, plays a key role in driving poor prognosis and resistance to chemotherapy. Herein, we aimed to identify a TME-associated, risk-stratification gene biomarker signature in PDAC.Experimental Design:The initial biomarker discovery was performed in The Cancer Genome Atlas (TCGA, n = 163) transcriptomic data. This was followed by independent validation of the gene signature in the International Cancer Genome Consortium (ICGC, n = 95), E-MTAB-6134 (n = 288), and GSE71729 (n = 123) datasets for predicting overall survival (OS), and for its ability to detect poor molecular subtypes. Clinical validation and nomogram establishment was undertaken by performing multivariate Cox regression analysis.Results:Our biomarker discovery effort identified a 15-gene immune, stromal, and proliferation (ISP) gene signature that significantly associated with poor OS [HR, 3.90; 95% confidence interval (CI), 2.36–6.41; P < 0.0001]. This signature also robustly predicted survival in three independent validation cohorts ICGC [HR, 2.63 (1.56–4.41); P < 0.0001], E-MTAB-6134 [HR, 1.53 (1.14–2.04); P = 0.004], and GSE71729 [HR, 2.33 (1.49–3.63); P < 0.0001]. Interestingly, the ISP signature also permitted identification of poor molecular PDAC subtypes with excellent accuracy in all four cohorts; TCGA (AUC = 0.94), ICGC (AUC = 0.91), E-MTAB-6134 (AUC = 0.80), and GSE71729 (AUC = 0.83). The ISP-derived high-risk patients exhibited significantly poor OS in a clinical validation cohort [n = 119; HR, 2.62 (1.50–4.56); P = 0.0004]. A nomogram was established which included the ISP, CA19-9, and T- and N-stage for eventual clinical translation.Conclusions:We report a novel gene signature for risk-stratification and robust identification of patients with PDAC with poor molecular subtypes.

Supplementary figures

PDF file, 89K.

Supplementary Figure 2: CONSORT diagram for patients and specimens from both fresh-frozen and FFPE clinical validation cohorts. A) Fresh-frozen cohort (cohort-1), B) FFPE cohort (cohorts 2 and 3).

Methylation ratio distribution of the gitBS performed on normal plasma samples (a) and GI cancer plasma samples (b).

Hierarchical clustering of EAC and healthy plasma samples

Hierarchical clustering of ESCC and healthy plasma samples

PDF file, 27K, Consistency of the multiplex methylation assay.

PDF file, 15K, ROC curve of methylation assay (dotted line) and methylation + FGFR3 assay (thick line) for the validation set.

Hierarchical clustering of PDAC and healthy plasma samples

Hierarchical clustering of CRC and healthy plasma samples

Supplementary Table 3: Spearman's rho correlation coefficients of MRC vs clinicopathological risk factors for all the clinical cohorts.

PDF file, 19K, Sensitivity of the methylation markers for detection of different stage and grade recurrences.

Purpose: The current tumor–node–metastasis (TNM) staging system is inadequate at identifying patients with high-risk colorectal cancer. Using a systematic and comprehensive biomarker discovery and validation approach, we aimed to identify an miRNA recurrence classifier (MRC) that can improve upon the current TNM staging as well as is superior to currently offered molecular assays.Experimental Design: Three independent genome-wide miRNA expression profiling datasets were used for biomarker discovery (N = 158) and in silico validation (N = 109 and N = 40) to identify an miRNA signature for predicting tumor recurrence in patients with colorectal cancer. Subsequently, this signature was analytically trained and validated in retrospectively collected independent patient cohorts of fresh-frozen (N = 127, cohort 1) and formalin-fixed paraffin-embedded (FFPE; N = 165, cohort 2 and N = 139, cohort 3) specimens.Results: We identified an 8-miRNA signature that significantly predicted recurrence-free interval (RFI) in the discovery (P = 0.002) and two independent publicly available datasets (P = 0.00006 and P = 0.002). The RT-PCR–based validation in independent clinical cohorts revealed that MRC-derived high-risk patients succumb to significantly poor RFI in patients with stage II and III colorectal cancer [cohort 1: hazard ratio (HR), 3.44 (1.56–7.45), P = 0.001; cohort 2: HR, 6.15 (3.33–11.35), P = 0.001; and cohort 3: HR, 4.23 (2.26–7.92), P = 0.0003]. In multivariate analyses, MRC emerged as an independent predictor of tumor recurrence and achieved superior predictive accuracy over the currently available molecular assays. The RT-PCR–based MRC risk score = (−0.1218 × miR-744) + (−3.7142 × miR-429) + (−2.2051 × miR-362) + (3.0564 × miR-200b) + (2.4997 × miR-191) + (−0.0065 × miR-30c2) + (2.2224 × miR-30b) + (−1.1162 × miR-33a).Conclusions: This novel MRC is superior to currently used clinicopathologic features, as well as National Comprehensive Cancer Network (NCCN) criteria, and works regardless of adjuvant chemotherapy status in identifying patients with high-risk stage II and III colorectal cancer. This can be readily deployed in clinical practice with FFPE specimens for decision-making pending further model testing and validation. Clin Cancer Res; 24(16); 3867–77. ©2018 AACR.See related commentary by Rodriguez et al., p. 3787

Workflow of model training and test for GI cancer prediction. (a) Initial markers discovery was conducted with 1653 GI cancer and 287 normal tissues. Targeted bisulfite sequencing that covers the initial selected CpG sites was performed on 300 plasma samples, including 254 GI cancer and 46 normal samples. GI plasma samples set was split into training set (70%) and test set (30%). The training set was used for calling differential methylated regions, feature selection and training random forest prediction models. The test set was used for model performance evaluation. (b) Annotations of panGI markers selected from GI cancer tissues.

PDF file, 113K, 3-plex methylation assay.

Purpose:DNA methylation alterations have emerged as front-runners in cell-free DNA (cfDNA) biomarker development. However, much effort to date has focused on single cancers. In this context, gastrointestinal (GI) cancers constitute the second leading cause of cancer-related deaths worldwide; yet there is no blood-based assay for the early detection and population screening of GI cancers.Experimental Design:Herein, we performed a genome-wide DNA methylation analysis of multiple GI cancers to develop a pan-GI diagnostic assay. By analyzing DNA methylation data from 1,781 tumor and adjacent normal tissues, we first identified differentially methylated regions (DMR) between individual GI cancers and adjacent normal, as well as across GI cancers. We next prioritized a list of 67,832 tissue DMRs by incorporating all significant DMRs across various GI cancers to design a custom, targeted bisulfite sequencing platform. We subsequently validated these tissue-specific DMRs in 300 cfDNA specimens and applied machine learning algorithms to develop three distinct categories of DMR panelsResults:We identified three distinct DMR panels: (i) cancer-specific biomarker panels with AUC values of 0.98 (colorectal cancer), 0.98 (hepatocellular carcinoma), 0.94 (esophageal squamous cell carcinoma), 0.90 (gastric cancer), 0.90 (esophageal adenocarcinoma), and 0.85 (pancreatic ductal adenocarcinoma); (ii) a pan-GI panel that detected all GI cancers with an AUC of 0.88; and (iii) a multi-cancer (tissue of origin) prediction panel, EpiPanGI Dx, with a prediction accuracy of 0.85–0.95 for most GI cancers.Conclusions:Using a novel biomarker discovery approach, we provide the first evidence for a cfDNA methylation assay that offers robust diagnostic accuracy for GI cancers.

PDF file, 58K, Study design.

PDF file, 21K, Primers and probes used for the BS-SNaPshot assay and FGFR3 assay.

BACKGROUND/AIM N6-Methyladenosine (m6A), the most abundant internal modification of RNA, plays a critical role in cancer development. However, the clinical implications of m6A in hepatocellular carcinoma (HCC) remain unclear. MATERIALS AND METHODS We analyzed 177 HCC and paired noncancerous liver tissues from patients who underwent hepatectomy according to global m6A quantification and expression of m6A demethylases fat mass and obesity-associated protein (FTO) and alpha-ketoglutarate-dependent dioxygenase alkB homolog 5 (ALKBH5). RESULTS The global m6A quantification revealed no significant difference between HCC and non-cancerous tissue. The expression of m6A demethylases FTO and ALKBH5, was significantly lower in HCC than in non-cancerous tissues (both p

BACKGROUND: While emerging evidence indicates that N(6)-methyladenosine (m(6)A) regulators play crucial roles in cancer progression, their clinical significance in gastric cancer (GC) has thus far not been elucidated. METHODS: We investigated the expression of the m(6)A regulator genes and their prognostic potential in a large clinical cohort of 173 GC patients using qRT-PCR assays. In addition, we undertook a series of in-vitro and in-vivo functional studies to investigate the oncogenic role of FTO. RESULTS: GC patients with low expression of METTL3, METTL14, ALKBH5, WTAP and YTHDF1 demonstrated significantly poor OS, while patients with high FTO expression exhibited markedly worse OS. Furthermore, the cumulative risk-score derived from these gene panel also significantly associated with poor OS, with a corresponding hazard ratio of 5.47 (95% CI: 3.18–9.41, p

PURPOSE The development and use of predictive biomarkers to guide treatment decisions are paramount not only for improving survival in patients with metastatic colorectal cancer (mCRC), but also for sparing them from unnecessary toxicity and reducing the economic burden of expensive treatments. We conducted a systematic review of published studies and evaluated the predictive biomarker landscape in the mCRC setting from a molecular and clinical viewpoint. METHODS Studies analyzing predictive biomarkers for approved therapies in patients with mCRC were identified systematically using electronic databases. Preclinical studies and those providing no relevant information were excluded. RESULTS A total of 173 studies comprising 148 biomarkers were selected for final analysis. Of all the biomarkers analyzed, 1.4% (two of 148) were explored in a prospective manner, whereas 98.6% (146 of 148) were evaluated in retrospective studies. Of the latter group, 78.8% (115 of 146) were not tested in subsequent phases, 9.6% (14 of 146) were tested in other retrospective cohorts, 8.9% (13 of 146) were retrospectively tested in at least one or more randomized cohorts, and only 2.7% (four of 146) were prospectively tested in a clinical trial. Finally, only 1.4% (two of 148) were validated sufficiently and are recognized as biomarkers for guiding treatment decision making in patients with mCRC. These markers were RAS mutational status for anti-EGFR antibodies and microsatellite instability status for anti–programmed cell death-1 drugs. CONCLUSION Despite notable efforts to identify predictive biomarkers for various therapies used in the mCRC setting, because of a lack of data beyond retrospective studies and successful biomarker-driven approaches, only two molecular biomarkers have thus far found their translation into the clinic, highlighting the imperative need for implementing novel strategies and additional research in this clinically important field.

OBJECTIVE: This study aimed to do a genomewide transcriptomic profiling to develop a miRNA-based signature for the identification of peritoneal metastasis (PM) in patients with gastric cancer (GC). SUMMARY BACKGROUND DATA: Even though PM in patients with GC has long been recognized to associate with poor survival, currently there is lack of availability of molecular biomarkers for its robust diagnosis. METHODS: We performed a systematic biomarker discovery by analyzing miRNA expression profiles in primary tumors from GC patients with and without PM, and subsequently validated the expression of candidate miRNA biomarkers in three independent clinical cohorts of 354 patients with advanced GC. RESULTS: Five miRNAs (miR-30a-5p, -134-5p, -337-3p, -659-3p, and -3917) were identified during the initial discovery phase; three of which (miR-30a-5p, -659-3p, and -3917) were significantly overexpressed in the primary tumors from PM-positive patients in the testing cohort (p=0.002, 0.04 and 0.007 respectively), and distinguished patients with vs. without peritoneal metastasis with the value of area under the curve (AUC) of 0.82. Furthermore, high expression of these miRNAs also associated with poor prognosis (HR=2.18, p=0.04). The efficacy of the combination miRNA-signature was subsequently validated in an independent validation cohort (AUC=0.74). Finally, our miRNA signature when combined together with the macroscopic Borrmann’s type score offered a much superior diagnostic in all three cohorts (AUC=0.87, 0.76, 0.79, respectively), and led us to establish a risk-prediction nomogram for the diagnosis of PM in GC patients. CONCLUSIONS: We have established a miRNA-based signature that have a potential to identify peritoneal metastasis in GC patients.

Purpose: Emerging evidence indicates that gut microbiome plays a crucial role in the cancer pathogenesis. Although Fusobacterium nucleatum (F. nucleatum) is associated with poor prognosis in multiple cancers, its clinical significance in predicting response to chemotherapy in patients with esophageal squamous cell carcinoma (ESCC) remains unclear. Experimental Design: The F. nucleatum levels were quantified by qPCR assays in tumor tissues from 551 patients with ESCC from two independent cohorts, including 101 patients who received neoadjuvant chemotherapy prior to curative resection. Associations between F. nucleatum burden and recurrence-free survival (RFS), as well with chemotherapeutic response were evaluated using response evaluation criteria in solid tumors (RECISTs), primary tumor metabolic response defined by maximum standardized uptake value (SUVmax) changes in positron emission tomography-CT (PET/CT), and pathologic tumor regression grade (TRG). Results: High burden of F. nucleatum in patients with ESCC associated with poor RFS in both training [log-rank P = 0.02; HR = 1.61; P = 0.03] and validation cohorts (log-rank P = 0.003; HR = 1.96; P = 0.004). Importantly, patients with ESCC with high levels of F. nucleatum displayed poor chemotherapeutic response for all three evaluation methods: RECIST (P = 0.04), SUVmax change in PET/CT (P = 0.0004), and TRG (P = 0.003). Conclusions: We conclude that high levels of intratumoral F. nucleatum have a prognostic significance for predicting poor RFS in patients with ESCC. More importantly, our data indicates that higher F. nucleatum burden correlates with poor response to neoadjuvant chemotherapy, suggesting the possibility that an antibiotic intervention against this bacterium may significantly improve therapeutic response in patients with ESCC.

Colorectal cancer (CRC) is one of the leading causes of cancer-related deaths worldwide. Similar to many other malignancies, CRC is a heterogeneous disease, making it a clinical challenge for optimization of treatment modalities in reducing the morbidity and mortality associated with this disease. A more precise understanding of the biological properties that distinguish patients with colorectal tumors, especially in terms of their clinical features, is a key requirement towards a more robust, targeted-drug design, and implementation of individualized therapies. In the recent decades, extensive studies have reported distinct CRC subtypes, with a mutation-centered view of tumor heterogeneity. However, more recently, the paradigm has shifted towards transcriptome-based classifications, represented by six independent CRC taxonomies. In 2015, the colorectal cancer subtyping consortium reported the identification of four consensus molecular subtypes (CMSs), providing thus far the most robust classification system for CRC. In this review, we summarize the historical timeline of CRC classification approaches; discuss their salient features and potential limitations that may require further refinement in near future. In other words, in spite of the recent encouraging progress, several major challenges prevent translation of molecular knowledge gleaned from CMSs into the clinic. Herein, we summarize some of these potential challenges and discuss exciting new opportunities currently emerging in related fields. We believe, close collaborations between basic researchers, bioinformaticians and clinicians are imperative for addressing these challenges, and eventually paving the path for CRC subtyping into routine clinical practice as we usher into the era of personalized medicine.

Purpose: DNA methylation alterations have emerged as front-runners in cell-free DNA (cfDNA) biomarker development. However, much effort to date has focused on single cancers. In this context, gastrointestinal (GI) cancers constitute the second leading cause of cancer-related deaths worldwide; yet there is no blood-based assay for the early detection and population screening of GI cancers. Experimental Design: Herein, we performed a genome-wide DNA methylation analysis of multiple GI cancers to develop a pan-GI diagnostic assay. By analyzing DNA methylation data from 1,781 tumor and adjacent normal tissues, we first identified differentially methylated regions (DMR) between individual GI cancers and adjacent normal, as well as across GI cancers. We next prioritized a list of 67,832 tissue DMRs by incorporating all significant DMRs across various GI cancers to design a custom, targeted bisulfite sequencing platform. We subsequently validated these tissue-specific DMRs in 300 cfDNA specimens and applied machine learning algorithms to develop three distinct categories of DMR panels Results: We identified three distinct DMR panels: (i) cancer-specific biomarker panels with AUC values of 0.98 (colorectal cancer), 0.98 (hepatocellular carcinoma), 0.94 (esophageal squamous cell carcinoma), 0.90 (gastric cancer), 0.90 (esophageal adenocarcinoma), and 0.85 (pancreatic ductal adenocarcinoma); (ii) a pan-GI panel that detected all GI cancers with an AUC of 0.88; and (iii) a multi-cancer (tissue of origin) prediction panel, EpiPanGI Dx, with a prediction accuracy of 0.85–0.95 for most GI cancers. Conclusions: Using a novel biomarker discovery approach, we provide the first evidence for a cfDNA methylation assay that offers robust diagnostic accuracy for GI cancers.

Simple Summary Colon cancer is one of the leading causes of cancer-related death worldwide. Therefore, the development of new therapeutic strategies is of the utmost importance. Previously, we identified a subset of colon cancers that are characterised by DNA methylation and have a poor prognosis. In this study, we therefore treated ten colon cancer patients with a demethylating agent, decitabine, to investigate if reversal of methylation is feasible and can be used as a novel therapy. Unfortunately, this study revealed that while decitabine treatment is effective in vitro, it only marginally decreased global methylation in patients and had no effect on the specific regions of DNA methylation in the tumours. Future studies should therefore focus on optimisation of treatment schedules in patients with highly methylated tumours. Abstract DNA hypermethylation is common in colon cancer. Previously, we have shown that methylation of WNT target genes predicts poor prognosis in stage II colon cancer. The primary objective of this study was to assess whether pre-operative treatment with decitabine can decrease methylation and increase the expression of WNT target genes APCDD1, AXIN2 and DKK1 in colon cancer patients. A clinical study was conducted, investigating these potential effects of decitabine in colon cancer patients (DECO). Patients were treated two times with 25 mg/m2 decitabine before surgery. Methylation and expression of LINE1 and WNT target genes (primary outcome) and expression of endogenous retroviral genes (secondary outcome) were analysed in pre- and post-treatment tumour samples using pyrosequencing and rt-PCR. Ten patients were treated with decitabine and eighteen patients were used as controls. Decitabine treatment only marginally decreased LINE1 methylation. More importantly, no differences in methylation or expression of WNT target or endogenous retroviral genes were observed. Due to the lack of an effect on primary and secondary outcomes, the study was prematurely closed. In conclusion, pre-operative treatment with decitabine is safe, but with the current dosing, the primary objective, increased WNT target gene expression, cannot be achieved.

Purpose:Due to the lack of effective screening approaches and early detection biomarkers, ovarian cancer has the highest mortality rates among gynecologic cancers. Herein, we undertook a systematic biomarker discovery and validation approach to identify microRNA (miRNA) biomarkers for the early detection of ovarian cancer.Experimental Design:During the discovery phase, we performed small RNA sequencing in stage I high-grade serous ovarian cancer (n = 31), which was subsequently validated in multiple, independent data sets (TCGA, n = 543; GSE65819, n = 87). Subsequently, we performed multivariate logistic regression-based training in a serum data set (GSE106817, n = 640), followed by its independent validation in three retrospective data sets (GSE31568, n = 85; GSE113486, n = 140; Czech Republic cohort, n = 192) and one prospective serum cohort (n = 95). In addition, we evaluated the specificity of OCaMIR, by comparing its performance in several other cancers (GSE31568 cohort, n = 369).Results:The OCaMIR demonstrated a robust diagnostic accuracy in the stage I high-grade serous ovarian cancer patients in the discovery cohort (AUC = 0.99), which was consistently reproducible in both stage I (AUC = 0.96) and all stage patients (AUC = 0.89) in the TCGA cohort. Logistic regression-based training and validation of OCaMIR achieved AUC values of 0.89 (GSE106817), 0.85 (GSE31568), 0.86 (GSE113486), and 0.82 (Czech Republic cohort) in the retrospective serum validation cohorts, as well as prospective validation cohort (AUC = 0.92). More importantly, OCaMIR demonstrated a significantly superior diagnostic performance compared with CA125 levels, even in stage I patients, and was more cost-effective, highlighting its potential role for screening and early detection of ovarian cancer.Conclusions:Small RNA sequencing identified a robust noninvasive miRNA signature for early-stage serous ovarian cancer detection.

Aim To develop a decision model for the population-level evaluation of strategies to improve the selection of stage II colon cancer (CC) patients who benefit from adjuvant chemotherapy. Methods A Markov cohort model with a one-month cycle length and a lifelong time horizon was developed. Five health states were included; diagnosis, 90-day mortality, death other causes, recurrence and CC death. Data from the Netherlands Cancer Registry were used to parameterize the model. Transition probabilities were estimated using parametric survival models including relevant clinical and pathological covariates. Subsequently, biomarker status was implemented using external data. Treatment effect was incorporated using pooled trial data. Model development, data sources used, parameter estimation, and internal and external validation are described in detail. To illustrate the use of the model, three example strategies were evaluated in which allocation of treatment was based on (A) 100% adherence to the Dutch guidelines, (B) observed adherence to guideline recommendations and (C) a biomarker-driven strategy. Results Overall, the model showed good internal and external validity. Age, tumor growth, tumor sidedness, evaluated lymph nodes, and biomarker status were included as covariates. For the example strategies, the model predicted 83, 87 and 77 CC deaths after 5 years in a cohort of 1000 patients for strategies A, B and C, respectively. Conclusion This model can be used to evaluate strategies for the allocation of adjuvant chemotherapy in stage II CC patients. In future studies, the model will be used to estimate population-level long-term health gain and cost-effectiveness of biomarker-based selection strategies.

Risk stratification in Stage II and III colorectal cancer (CRC) patients is critical, as it allows patient selection for adjuvant chemotherapy. In view of the inadequacy of current clinicopathological features for risk-stratification, we undertook a systematic and comprehensive biomarker discovery effort to develop a risk-assessment signature in CRC patients. The biomarker discovery phase examined 853 CRC patients, and identified a gene signature for predicting recurrence-free survival (RFS). This signature was validated in a meta-analysis of 1212 patients from nine independent datasets, and its performance was compared against established prognostic signatures and consensus molecular subtypes (CMS). In addition, a risk-prediction model was trained (n = 142), and subsequently validated in an independent clinical cohort (n = 286). As a result, this mesenchymal-associated transcriptomic signature (MATS) identified high-risk CRC patients with poor RFS in the discovery (hazard ratio [HR]: 1.79), and nine validation cohorts (HR: 1.86). In multivariate analysis, MATS was the most significant predictor of RFS compared to established prognostic signatures and CMS subtypes. Intriguingly, MATS robustly identified CMS4-subtype in multiple CRC cohorts (AUC = 0.92-0.99). In the two clinical cohorts, MATS stratified low and high-risk groups with a 5-year RFS in the training (HR: 4.11) and validation cohorts (HR: 2.55), as well as predicted response to adjuvant therapy in Stage II and III CRC patients. We report a novel prognostic and predictive biomarker signature in CRC, which is superior to currently used approaches and have the potential for clinical translation in near future.

OBJECTIVE The aim of the study was to perform mRNA-miRNA regulatory network analyses to identify a miRNA panel for molecular subtype identification and stratification of high-risk patients with pancreatic ductal adenocarcinoma (PDAC). BACKGROUND Recent transcriptional profiling effort in PDAC has led to the identification of molecular subtypes that associate with poor survival; however, their clinical significance for risk stratification in patients with PDAC has been challenging. METHODS By performing a systematic analysis in The Cancer Genome Atlas and International Cancer Genome Consortium cohorts, we discovered a panel of miRNAs that associated with squamous and other poor molecular subtypes in PDAC. Subsequently, we used logistic regression analysis to develop models for risk stratification and Cox proportional hazard analysis to determine survival prediction probability of this signature in multiple cohorts of 433 patients with PDAC, including a tissue cohort (n = 199) and a preoperative serum cohort (n = 51). RESULTS We identified a panel of 9 miRNAs that were significantly upregulated (miR-205-5p and -934) or downregulated (miR-192-5p, 194-5p, 194-3p, 215-5p, 375-3p, 552-3p, and 1251-5p) in PDAC molecular subtypes with poor survival [squamous, area under the receiver operating characteristic curve (AUC) = 0.90; basal, AUC = 0.89; and quasimesenchymal, AUC = 0.83]. The validation of this miRNA panel in a tissue clinical cohort was a significant predictor of overall survival (hazard ratio = 2.48, P < 0.0001), and this predictive accuracy improved further in a risk nomogram which included key clinicopathological factors. Finally, we were able to successfully translate this miRNA predictive signature into a liquid biopsy-based assay in preoperative serum specimens from PDAC patients (hazard ratio: 2.85, P = 0.02). CONCLUSION We report a novel miRNA risk-stratification signature that can be used as a noninvasive assay for the identification of high-risk patients and potential disease monitoring in patients with PDAC.

Treatment modalities in esophageal squamous cell carcinoma (ESCC) depend largely on lymph node metastasis (LNM) status. With suboptimal detection sensitivity of existing imaging techniques, we propose a methylation signature which identifies patients with LNM with greater accuracy. This would allow precise stratification of high-risk patients requiring more aggressive treatment from low-risk ESCC patients who can forego radical surgery. An unbiased genome-wide methylation signature for LNM detection was established from an initial in silico discovery phase. The signature was tested in independent clinical cohorts comprising of 249 ESCC patients. The prognostic potential of the methylation signature was compared to clinical variables including LNM status. A 10-probe LNM associated signature (LNAS) was developed using stringent bioinformatics analyses. The area under the curve values for LNAS risk scores were 0.81 and 0.88 in the training and validation cohorts respectively, in association with lymphatic vessel invasion and tumor stage. High LNAS risk-score was also associated with worse overall survival [HR (95% CI) 3 (1.8-4.8), p < 0.0001 training and 3.9 (1.5-10.2), p = 0.001 validation cohort]. In conclusion, our novel methylation signature is a powerful biomarker that identifies LNM status robustly and is also associated with worse prognosis in ESCC patients.

Colorectal cancer (CRC) is a highly heterogeneous disease both from a molecular and clinical perspective. Several distinct molecular entities, such as microsatellite instability (MSI), have been defined that make up biologically distinct subgroups with their own clinical course. Recent data indicated that CRC can be best segregated into four groups called consensus molecular subtypes (CMS1-4), each of which has a unique biology and gene expression pattern. In order to develop improved, subtype-specific therapies and to gain insight into the molecular wiring and origin of these subtypes, reliable models are needed. This study was designed to determine the heterogeneity and identify the presence of CMSs in a large panel of CRC cell lines, primary cultures and patient-derived xenografts (PDX). We provide a repository encompassing this heterogeneity and moreover describe that a large part of the models can be robustly assigned to one of the four CMSs, independent of the stromal contribution. We subsequently validate our CMS stratification by functional analysis which for instance shows mesenchymal enrichment in CMS4 and metabolic dysregulation in CMS3. Finally, we observe a clear difference in sensitivity to chemotherapy-induced apoptosis, specifically between CMS2 and CMS4. This relates to the in vivo efficacy of chemotherapy, which delays outgrowth of CMS2, but not CMS4 xenografts. Combined our data indicate that molecular subtypes are faithfully modelled in CRC cell cultures and PDXs, representing tumour cell intrinsic and stable features. This repository provides researchers with a platform to study CRC using the existing heterogeneity.

Purpose Liquid biopsy has become increasingly important in cancer diagnosis, personalized medicine, and disease progression monitoring. Conventional liquid biopsy relies on targeted cancer gene panels which often contain fewer than 500 genes. Despite the revolutionary impact of liquid biopsy on cancer research and patient care, targeted gene panels may miss key novel mutations involved in cancer development and drug response, and other novel genomic alternations underlining cancer development, such as whole genome structural or epigenetic changes. Additionally, current commonly used methodologies require relatively large amounts of cell free DNA (cfDNA) input material, thus limiting its application when only small amounts of cfDNA input is available. Experimental Design cfDNA was isolated from whole blood collected from commercially available prostate cancer patients or age-matched healthy donors and evaluated using Predicine’s platform for whole exome sequencing, plasma WES, and compared to the targeted 500 gene PredicineATLAS panel (cfDNA input 10 or 30 ng). Additionally, low-input cfDNA (cfDNA input 10 or 20 ng) was also evaluated for whole genome methylation sequencing using PredicineECM and compared with standard whole genome bisulfite sequencing (WGBS). Results Plasma WES profiling of 30 or 10 ng cfDNA detected variants with mutation allele frequency (MAF) of 1% and above in the entire exome region, which enabled more accurate genome wide copy number variation prediction. All mutations detected using the PredicineATLAS panel that were ≥1% allele frequency were also detected by plasma WES, although low input required greater depth (10 ng, 3,500x) compared with high input (30 ng, 2500x depth). Furthermore, plasma WES identified 1.5 to 15.3x additional mutations compared with the PredicineATLAS panel. PredicineECM methylation analysis was superior to WGBS in reducing DNA damage and GC bias, resulting in increased NGS read mapping rate and quality score. Profiling of 2.5, 5 and 10 ng cfDNA using PredicineECM showed higher mapping rate and quality compared with WGBS profiling of 10 ng cfDNA. Additionally, clustering analysis of CpG methylation using PredicineECM separated tumor samples from normal samples and may serve as additional markers for minimal residue disease (MRD) tracing or early cancer detection. Conclusion In conclusion, we have successfully developed a multi-omics analysis platform for low-input cfDNA that integrates whole exome mutation and whole genome methylation profiling to obtain a comprehensive picture of cancer genomics to enable precision medicine in prostate cancer. This technology can also be applied to other cancers. Citation Format: Raju Kandimalla, Binggang Xiang, Chao Dai, Amy Wang, Pan Du, Shujun Luo, Debbie Liao. Whole exome and whole genome methylation sequencing of low-input cfDNA to implement precision medicine in metastatic castration resistant prostate cancer [abstract]. In: Proceedings of the AACR-NCI-EORTC Virtual International Conference on Molecular Targets and Cancer Therapeutics; 2021 Oct 7-10. Philadelphia (PA): AACR; Mol Cancer Ther 2021;20(12 Suppl):Abstract nr CC07-01.