Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 https://doi.org/10.1186/s13148-022-01302-x RESEARCH Open Access Noninvasive early detection of colorectal cancer by hypermethylation of the LINC00473 promoter in plasma cell‑free DNA Juan Ruiz‑Bañobre1,2†, Aitor Rodriguez‑Casanova3,4,5†, Nicolas Costa‑Fraga3,5, Aida Bao‑Caamano3,5, Ana Alvarez‑Castro6, Martín Carreras‑Presas6, Elena Brozos‑Vazquez1, Yolanda Vidal‑Insua1, Francisca Vazquez‑Rivera1, Sonia Candamio‑Folgar1,2, Manuel Mosquera‑Presedo3,5, Ramón M. Lago‑Lestón7, Laura Muinelo‑Romay2,7, José Ángel Vázquez‑Bueno8, Rebeca Sanz‑Pamplona9,10,11, Víctor Moreno9,10,11,12, Ajay Goel13,14,15, Lourdes Castillo16, Ana C. Martin16, Rocio Arroyo16, Manel Esteller2,17,18,19, Ana B. Crujeiras20,21, Rafael López‑López1,2,4* and Angel Díaz‑Lagares2,3* Abstract Background: Current noninvasive assays have limitations in the early detection of colorectal cancer. We evaluated the clinical utility of promoter methylation of the long noncoding RNA LINC00473 as a noninvasive biomarker to detect colorectal cancer and associated precancerous lesions. Methods: We evaluated the epigenetic regulation of LINC00473 through promoter hypermethylation in colorec‑ tal cancer cell lines using bisulfite genomic sequencing and expression analyses. DNA methylation of LINC00473 was analyzed in primary colorectal tumors using 450K arrays and RNA-seq from The Cancer Genome Atlas (TCGA). Tissue-based findings were validated in several independent cohorts of colorectal cancer and advanced colorectal polyp patients by pyrosequencing. We explored the clinical utility of LINC00473 methylation for the early detection of colorectal cancer in plasma cell-free DNA by quantitative methylation-specific PCR and droplet digital PCR. Results: LINC00473 showed transcriptionally silencing due to promoter hypermethylation in colorectal cancer cell lines and primary tumors. Methylation of the LINC00473 promoter accurately detected primary colorectal tumors in two independent clinical cohorts, with areas under the receiver operating characteristic curves (AUCs) of 0.94 and 0.89. This biomarker also identified advanced colorectal polyps from two other tissue-based clinical cohorts with high diagnostic accuracy (AUCs of 0.99 and 0.78). Finally, methylation analysis of the LINC00473 promoter in plasma † Juan Ruiz-Bañobre and Aitor Rodriguez-Casanova share first authorship *Correspondence: rafael.lopez.lopez@sergas.es; angel.diaz.lagares@sergas.es 1 Translational Medical Oncology Group (ONCOMET), Health Research Institute of Santiago de Compostela (IDIS), University Clinical Hospital of Santiago (CHUS/SERGAS), 15706 Santiago de Compostela, Spain 3 Cancer Epigenomics Laboratory, Epigenomics Unit, Translational Medical Oncology Group (ONCOMET), Health Research Institute of Santiago de Compostela (IDIS), University Clinical Hospital of Santiago (CHUS/SERGAS), 15706 Santiago de Compostela, Spain Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 2 of 13 cell-free DNA accurately identified patients with colorectal cancer and advanced colorectal polyps (AUCs of 0.88 and 0.84, respectively), which was confirmed in an independent cohort of patients. Conclusions: Hypermethylation of the LINC00473 promoter is a new promising biomarker for noninvasive early detection of colorectal cancer and related precancerous lesions. Keywords: DNA methylation, LINC00473, Cell-free DNA, Early detection, Colorectal cancer Background in their promoters [13, 14]. DNA methylation is a well- Colorectal cancer (CRC) is the third most frequently known epigenetic mechanism based on the incorporation detected cancer in both sexes worldwide and is expected of a methyl group ­(CH3) into the 5′ carbon of cytosines to increase by 60% to more than 2.2 million new cases in cytosine-phosphate-guanine (CpG) dinucleotides that by 2030 [1]. CRC is usually diagnosed at an advanced generates 5-methylcytosine (5mC). In carcinogenesis, stage of disease [2] and represents a leading cause of hypermethylation of tumor suppressor genes (TSGs) cancer mortality worldwide [3]. For CRC patients, the represents an early event usually associated with their 5‐year survival rate ranges from 90 to 14% depending on transcriptional silencing, which leads to tumor initiation whether they are diagnosed at early or advanced stages, and disease progression. This epigenetic mechanism can respectively [4]. The growing incidence and high mor- be detected in tumors but also in cell-free DNA (cfDNA) tality rates of CRC highlight the clinical need for novel released into circulation by tumor cells, showing great strategies to improve early CRC detection and patient potential as a dynamic and noninvasive tool for CRC management [5]. diagnosis [5]. Colorectal carcinogenesis is a multistep process involv- LINC00473 is a lncRNA downregulated in CRC that ing genetic and epigenetic alterations [6]. The majority of exerts tumor suppressor functions in this disease by pro- CRCs (70%) originate from a common colorectal precur- moting apoptotic protease-activating factor 1 (APAF1) sor lesion, adenomatous polyp or conventional adeno- IRES activity through competitively sponging miR574- mas, which can potentially become malignant through 5p and miR15b-5p in tumor initiation and pathogenesis the “traditional” carcinogenesis pathway (adenoma–car- [15]. Of note, Diaz-Lagares et al. recently identified that cinoma sequence). In addition, another type of colorectal the CpGI in the promoter of LINC00473 is hypermethyl- lesion, serrated polyps, has been recently recognized as a ated in CRC [14]. Therefore, in this study, we evaluated precursor of 30% of all CRCs through the “serrated” car- the epigenetic regulation of LINC00473 by DNA meth- cinogenesis pathway. Among these precancerous lesions, ylation and its clinical impact for noninvasive early detec- some are defined as advanced colorectal polyps (ACPs) tion of CRC. and present a higher risk of cancer transformation [7]. There are several screening assays used to detect CRC Results at an early stage, including the fecal immunochemical Epigenetic regulation of LINC00473 by promoter test (FIT), which is a widely used noninvasive and cost- methylation in colorectal cancer effective assay for detecting the presence of fecal hemo- CpGI hypermethylation in the promoter of LINC00473 globin [8]. However, this type of stool-based assay has has been recently described in the colorectal cancer cell shown some limitations, including the inability to reliably line HCT-116 using the Infinium HumanMethylation450 detect colorectal precancerous lesions, such as ACPs [9]. (450K) microarray (Illumina) [14]. To confirm this epige- Although colonoscopy is considered the gold standard netic feature, we first analyzed the methylation status of CRC diagnosis technique, it is an invasive procedure that, LINC00473 in HCT-116 cells in comparison with normal in addition to requiring tedious and time-consuming colon mucosa by bisulfite genomic sequencing of multiple preparation, can potentially cause serious complications clones (Fig. 1A). Importantly, the CpGI promoter region [10] and has low patient adherence [11]. Therefore, novel of LINC00473 that we analyzed contains CpGs included non‐invasive tests for early detection of CRC are urgently in the 450K array. As expected, we observed hyper- needed. methylation of all the CpGs analyzed in HCT-116 cells Long noncoding RNAs (lncRNAs) are an emerging compared to unmethylated normal colon mucosa. The group of heterogeneous noncoding transcripts longer effect of the CpGI hypermethylation on gene expression than 200 nt involved in the regulation of many biologi- (Fig. 1B) was analyzed by qRT-PCR, showing the down- cal processes in normal cells [12]. Of note, the expression regulation of LINC00473 in HCT-116 cells with respect of lncRNAs can be disrupted in cancer by several mecha- to unmethylated normal colon mucosa. Importantly, nisms, such as hypermethylation of CpG islands (CpGIs) Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 3 of 13 Fig. 1 Epigenetic silencing of the LINC00473 in colorectal cancer cells. A Bisulfite genomic sequencing analysis of LINC00473 promoter CpG island in the colorectal cancer cell line HCT-116 and normal tissue. Locations of CpG dinucleotides (vertical lines) and the TSS (long black arrow) are shown. Ten single clones are represented for each sample. The presence of unmethylated and methylated CpGs is indicated by white and black squares, respectively. B DNA methylation-associated transcriptional silencing of LINC00473 in the colorectal cancer cell line HCT-116. Expression levels of LINC00473 were determined by qRT-PCR in the methylated cancer cell line HCT-116 and in colorectal normal tissues (N = 3). C Restored LINC00473 expression in the methylated cancer cell line HCT-116 after AZA treatment analyzed by qRT-PCR. Values were determined from triplicates and are expressed as the mean ± SEM. D Methylation and expression analysis of LINC00473 in common colorectal cancer cell lines and normal colon mucosa using 450K array and RNA-seq public datasets. Expression levels obtained by RNA-seq were expressed as transcripts per million (TPMs). TSS, transcription start site; NC; normal colon mucosa the use of the demethylating agent 5-aza-2’-deoxycyti- this analysis revealed a significantly higher methylation dine (AZA) in HCT-116 cells restored the expression of level of LINC00473 in colorectal tumors than in controls LINC00473 (Fig. 1B, C). Thus, we extended the methyla- (Fig. 2A), which was consistent across all TNM tumor tion and expression analysis to other common colorectal stages (Fig. 2B). In addition, LINC00473 promoter hyper- cancer cell lines (Fig. 1D) using 450K array and RNA- methylation was associated with a significant down- seq datasets obtained from Gene Expression Omnibus regulation of its expression levels in primary colorectal (GEO): GSE49143 and GSE138734, respectively. These tumors (Additional file 1: Fig. S1). Consistent with this, analyses confirmed CGI promoter hypermethylation promoter methylation of LINC00473 showed a signifi- and downregulation of LINC00473 in all CRC cell lines cant inverse correlation (r = − 0.19; p = 0.0009) with its analyzed (COLO-205, HCC-2998, HCT-116, HCT-15, expression levels in CRC patients from the TCGA dataset HT-29, KM12, and SW-620) in comparison with normal (Cohort 1). colon mucosa. Next, we generated receiver operating character- istic curves (ROCs) to evaluate the robustness of the Methylation status of the LINC00473 promoter in colorectal methylation status of the LINC00473 promoter for tumors CRC diagnosis across all tumor stages. This analysis After confirming the epigenetic regulation of LINC00473 revealed a significantly high CRC detection accuracy by promoter methylation in CRC cell lines, we decided to with an area under the receiver operating characteristic evaluate whether this epigenetic alteration was a general curve (AUC) of 0.941 (95% CI 0.915–0.966, p < 0.0001) event in primary colorectal tumors. Thus, we evaluated (Fig. 2C), a sensitivity of 91% (CI 95% 87–94%), a speci- the methylation status of the LINC00473 promoter and ficity of 100% (CI 95% 91–100%), a positive predictive the expression level of the corresponding gene by using value (PPV) of 100% and a negative predictive value a 450K array and RNA-seq, respectively, from a TCGA (NPV) of 58%. Moreover, ROC curve analyses yielded dataset (Cohort 1) of primary colorectal tumors (from a very high diagnostic accuracy across all tumor stages TNM stage I to IV) and matched normal tissues (con- separately (Additional file 2: Fig. S2). Subsequently, trols). As expected from our results with CRC cell lines, these results were confirmed in an independent cohort Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 4 of 13 A P<0.0001 B P<0.0001 C 1.0 1.0 Ctrl vs CRC LINC00473 methylation (b-values) LINC00473 methylation (b-values) 100 0.8 0.8 80 AUC=0.941 Sensitivity (%) 0.6 0.6 60 P<0.0001 0.4 0.4 40 0.2 0.2 20 0.0 0.0 Ctrl CRC 0 Ctrl I II III IV 0 20 40 60 80 100 (N=38) (N=293) (N=38) (N=44) (N=113) (N=85) (N=41) 100 - Specificity (%) D P<0.0001 E P<0.0001 F 100 100 Ctrl vs CRC 100 LINC00473 methylation (%) LINC00473 methylation (%) 80 80 80 AUC=0.893 Sensitivity (%) 60 60 60 P<0.0001 40 40 40 20 20 20 0 0 0 Ctrl CRC Ctrl I II III IV 0 20 40 60 80 100 (N=82) (N=180) (N=82) (N=10) (N=84) (N=56) (N=28) 100 - Specificity (%) Fig. 2 Methylation of LINC00473 in colorectal cancer tissues. Methylation analysis of LINC00473 promoter in tissues from primary CRC and matched normal colorectal mucosa (controls) of two independent cohorts analyzed by 450K array (Cohort 1) and pyrosequencing (Cohort 2). A, B Methylation status of LINC00473 in (A) all CRC patients (N = 293) and (B) in those CRC patients (N = 283) with available clinical information on their TNM tumor stage (Cohort 1). C ROC curve analysis evaluating the methylation of LINC00473 for the detection of CRC in tissue samples in Cohort 1. D, E Validation of the methylation status of LINC00473 in (A) all CRC patients (N = 180) and (B) in those CRC patients (N = 178) with available clinical information on their TNM tumor stage in Cohort 2. F ROC curve analysis to validate the methylation of LINC00473 for the detection of colorectal tumors in Cohort 2. Horizontal lines represent mean methylation levels of LINC00473. P, p value analyzed by Mann–Whitney U test or ROC curve; AUC, area under the ROC curve; Ctrl, controls; CRC, colorectal cancer (Cohort 2) of primary colorectal tumors (from TNM 71–84%), a specificity of 98% (CI 95% 92–100%), a PPV stage I to IV) and matched non-tumor controls using of 99% and an NPV of 67%. The detection accuracy was bisulfite pyrosequencing (Fig. 2D–F). This analysis constant across all CRC stages separately, as shown in also showed a significantly higher methylation level Additional file 3: Fig. S3. of the LINC00473 promoter in primary colorectal tumors than in controls (Fig. 2D), which was constant Methylation status of the LINC00473 promoter in tissue across all TNM tumor stages (Fig. 2E). As expected, from precancerous colorectal lesions the LINC00473 methylation level identified primary DNA methylation is an epigenetic mechanism that can colorectal tumors with an AUC of 0.893 (95% CI 0.851– be deregulated in precancerous colorectal lesions [16]. 0.935, < 0.0001) (Fig. 2F), a sensitivity of 78% (CI 95% To confirm this feature, we analyzed the methylation (See figure on next page.) Fig. 3 Methylation levels of LINC00473 in tissue precancerous colorectal lesions. A Methylation levels of LINC00473 promoter in tissues from premalignant colorectal polyps, CRC and matched normal colorectal mucosa (controls) by pyrosequencing (Cohort 3). B, C Methylation levels of LINC00473 promoter in tissues from N-ACP, ACP and CRC (Cohort 3) analyzed by pyrosequencing. D ROC curve analysis evaluating the methylation of LINC00473 promoter for the detection of ACPs with respect to the combination of controls and N-ACPs (Cohort 3). E ROC curve to evaluate the methylation of LINC00473 promoter for the detection of AN with respect to the combination of controls and N-ACPs (Cohort 3). Horizontal lines represent mean methylation levels of LINC00473. P, p value analyzed by Mann–Whitney U test or ROC curve; AUC, area under the ROC curve; Ctrl, controls; P, polyps; CRC, colorectal cancer; N-ACP, non-advanced colorectal polyps; ACP, advanced colorectal polyps; AN, advanced neoplasia Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 5 of 13 A P=0.0002 B P=0.0040 80 80 P=0.3144 LINC00473 methylation (%) LINC00473 methylation (%) 60 60 40 40 20 20 0 0 Ctrl P CRC N-ACP ACP (N=12) (N=12) (N=12) (N=4) (N=8) C P=0.0132 D Ctrl & N-ACP vs ACP 60 100 LINC00473 methylation (%) 80 40 AUC=0.992 Sensitivity (%) 60 P=0.0001 20 40 20 0 0 N-ACP CRC 0 20 40 60 80 100 (N=4) (N=12) 100 - Specificity (%) E Ctrl & N-ACP vs AN 100 80 AUC=0.947 Sensitivity (%) 60 P<0.0001 40 20 0 0 20 40 60 80 100 100 - Specificity (%) Fig. 3 (See legend on previous page.) Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 6 of 13 status of the LINC00473 promoter by bisulfite pyrose- high accuracy of the methylation status of LINC00473 quencing in tissues from premalignant colorectal polyps, to detect polyps (AUC = 0.831, CI 95% 0.738–0923, CRC and matched normal colorectal mucosa (controls) p = 0.0009; sensitivity = 70%, CI 95% 57–81%; specific- (Cohort 3) (Fig. 3). As expected, the methylation of ity = 100%, CI 95% 69–100%, PPV = 100%; NPV = 36%) LINC00473 was significantly higher in polyps and in (Additional file 5: Fig. S5D) and, even more importantly, CRC than in healthy controls, while no differences were to detect ACPs (AUC = 0.776, CI 95% 0.665–0.888, found between polyps and CRC (Fig. 3A). Importantly, p < 0.0001; sensitivity = 71%, CI 95% 54–85%; specific- from a clinical viewpoint, we also found significant dif- ity = 79%, CI 95% 62–91%, PPV = 78%; NPV = 73%) ferences between LINC00473 methylation levels in (Additional file 5: Fig. S5E). non-ACP (N-ACP) compared to ACP (Fig. 3B) and in Additionally, we also assayed for a possible effect of N-ACP compared to CRC (Fig. 3C) but not in controls age on LINC00473 methylation levels in tissue samples; compared to N-ACP or in ACP compared to CRC (Addi- however, no significant effect was found (p > 0.05). Fur- tional file 4: Fig. S4A-B). Next, we generated ROC curves thermore, we also did not find any significant difference to evaluate the robustness of this methylation biomarker (p > 0.05) in methylation levels according to the sex of the to detect CRC or premalignant colorectal polyps, and we individuals (data not shown). observed a high capacity of LINC00473 to differentiate controls from polyps (AUC = 0.840, CI 95% 0.657–1.00, Diagnostic potential of methylation of the LINC00473 p = 0.0047; sensitivity = 83%, CI 95% 52–98%; specific- promoter to detect colorectal cancer and precancerous ity = 92%, CI 95% 62–100%; PPV = 91%; NPV = 85%) and lesions in plasma cell‑free DNA controls from CRC (AUC = 0.917, CI 95% 0.794–1.00, Beyond tissue samples, promoter hypermethylation of p = 0.0005; sensitivity = 83%, CI 95% 52–98%; specific- several genes in plasma cfDNA of patients with ACPs or ity = 100%, CI 95% 74–100%; PPV = 100%; NPV = 86%) CRC has also been described [5]. Based on this fact, we (Additional file 4: Fig. S4C-D). More importantly, the analyzed the methylation of the LINC00473 promoter in methylation status of LINC00473 was able to accurately plasma cfDNA of a cohort of self-declared healthy con- detect ACP (AUC = 0.992, CI 95% 0.967–1, p = 0.0001; trols and CRC patients by qMSP (Cohort 5). The result sensitivity = 100%, CI 95% 63–100%; specificity = 94%, of this analysis showed significantly higher methylation CI 95% 70–100%, PPV = 89%; NPV = 100%) (Fig. 3D) and levels in CRC than in controls (Fig. 4A). In addition, the AN (AUC = 0.947, CI 95% 0.873–1.00, p < 0.0001; sensi- methylation levels of LINC00473 in cfDNA exhibited tivity = 85%, CI 95% 62–97%; specificity = 100%, CI 95% a very high diagnostic accuracy to detect CRC patients 79–100%, PPV = 100%; NPV = 84%) (Fig. 3E). with an AUC of 0.881 (CI 95% 0.776–0983, p < 0.0001), To better appreciate the translational potential of our a sensitivity of 81% (95% CI 61–93%), a specificity of previous findings, we asked whether this methylation 100% (95% CI 88–100%), a PPV of 100% and an NPV of biomarker could be successfully validated using pyrose- 85% (Fig. 4B). Furthermore, we also analyzed the plasma quencing in an additional cohort (cohort 4) composed of cfDNA of a cohort of self-declared healthy controls and colorectal polyps and normal mucosa from non-cancer patients with ACPs presenting at least one polyp > 10 mm patients (controls). In addition to confirming methyla- previously confirmed by colonoscopy (Cohort 6). Of tion differences between controls and polyps (Additional note, this assay revealed significantly higher methylation file 5: Fig. S5A), with a larger number of N-ACPs, in this levels of LINC00473 in cfDNA of ACPs than in controls cohort, we were able to detect significantly higher meth- (Fig. 4C), showing a high diagnostic accuracy to detect ylation levels in controls than in N-ACPs (Additional ACPs (AUC = 0.836; 95% CI 0.722–0.949, p < 0.0001) file 5: Fig. S5B). Importantly from a clinical viewpoint, with a sensitivity of 79% (95% CI 58–93%), a specificity of we confirmed significantly higher methylation levels of 88% (95% CI 73–97%), a PPV of 83% and an NPV of 86% LINC00473 in ACPs than in N-ACPs (Additional file 5: (Fig. 4D). Fig. S5C). Furthermore, ROC curves demonstrated the (See figure on next page.) Fig. 4 Methylation of LINC00473 in cfDNA from patients with advanced colorectal polyps and colorectal cancer. A Methylation status of LINC00473 promoter in plasma cfDNA of CRC patients analyzed by qMSP (Cohort 5). B ROC curve analysis evaluating the methylation of LINC00473 for the detection of CRC in plasma cfDNA (Cohort 5). C, Methylation status of LINC00473 promoter in plasma cfDNA of patients with ACP analyzed by qMSP (Cohort 6). D ROC curve analysis evaluating the methylation of LINC00473 for the detection of ACP in plasma cfDNA (Cohort 6). E Validation of the methylation status of LINC00473 promoter in plasma cfDNA of patients with ACP by ddPCR analysis (Cohort 7). F, ROC curve analysis to validate the methylation of LINC00473 for the detection of ACP and CRC in plasma cfDNA (Cohort 7). Horizontal lines represent mean methylation levels of LINC00473. P, p value analyzed by Mann–Whitney U test or ROC curve; AUC, area under the ROC curve; Ctrl, controls. ACP, advanced colorectal polyps; CRC, colorectal cancer. Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 7 of 13 A P<0.0001 B 100 Ctrl vs CRC LINC00473 methylation (%) 100 80 80 AUC=0.881 Sensitivity (%) 60 60 P<0.0001 40 40 20 20 0 0 Ctrl CRC 0 20 40 60 80 100 (N=28) (N=26) 100 - Specificity (%) C P<0.0001 D 1.0 Ctrl vs ACP 100 LINC00473 methylation (%) 0.8 0.4 80 AUC=0.836 Sensitivity (%) 0.3 60 P<0.0001 0.2 40 0.1 20 0.0 0 Ctrl ACP 0 20 40 60 80 100 (N=34) (N=24) 100 - Specificity (%) P=0.0003 E P=0.0001 F CRC ACP AN 50 P=0.1216 100 40 LINC00473 methylation (%) 80 AUCCRC=0.833 (%) Sensitivity (%) 30 60 P=0.0003 Sensitivity 20 AUCAN=0.783 P=0.017 40 10 P=0.0005 20 AUCACP=0.721 5 P=0.0235 0 0 0 20 40 60 80 100 Ctrl ACP CRC AN (N=19) (N=17) (N=21) (N=38) 100 - Specificity (%) Fig. 4 (See legend on previous page.) Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 8 of 13 To confirm the feasibility of using methylation of the increased with disease progression. Notably, in some LINC00473 promoter for the noninvasive early detec- cases (e.g., cases #1, #4, and #5), LINC00473 methyla- tion of CRC, we retrospectively analyzed an independ- tion preceded CEA in detecting the presence of CRC. ent cohort of plasma cfDNA samples (Cohort 7) obtained Together, these six cases represent proof of concept of either prior to a scheduled colonoscopy as part of the utility of LINC00473 methylation as a potential bio- standard CRC screening or prior to colonic surgery for marker to detect the presence of CRC during the follow- primary tumors (Fig. 4E, F). Due to the need for very sen- up of metastatic patients. sitive methodologies for the early detection of cancer in liquid biopsy [5], we used ultrasensitive droplet digital Discussion PCR (ddPCR) for the methylation analysis of LINC00473. Colorectal cancer (CRC) is one of the most common Consistent with our previous results in tissue and malignancies and is a major cause of cancer-related cfDNA, the methylation levels of LINC00473 were sig- deaths worldwide [17]. The high incidence and mortality nificantly higher in ACP and CRC than in confirmed of CRC highlight the clinical need for novel strategies to healthy controls (Fig. 4E). In addition, no significant dif- improve early detection and personalize the management ferences were found between the methylation status of of patients with this type of tumor. Deregulation of epi- LINC00473 of ACP and CRC (Fig. 4E). Of note, methyla- genetic mechanisms, such as the promoter hypermeth- tion of the LINC00473 promoter showed high accuracy ylation of lncRNAs, has relevant implications for CRC for the detection of ACP, with an AUC of 0.721 (95% CI development and progression [14, 18]. Importantly, DNA 0.553–0.890, p = 0.0235), a sensitivity of 76% (95% CI methylation can be detected both in tumor cells and in 50–93%), a specificity of 63% (95% CI 38–84%), a PPV of various components of liquid biopsy, such as cfDNA, and 65% and an NPV of 75% (Fig. 4F). Similarly, methylation has shown clinical utility as a cancer biomarker [5]. Thus, of LINC00473 was able to identify CRC patients with a in this work, we studied epigenetic regulation through high AUC of 0.833 (95% CI 0.709–0.958, p = 0.0003), a DNA methylation of lncRNA LINC00473 and evaluated sensitivity of 90% (95% CI 70–99%), a specificity of 63% the clinical utility of hypermethylation of its promoter (95% CI 38–84%), a PPV of 73 and an NPV of 86. In addi- for the detection of CRC and its precancerous lesions in tion, LINC00473 showed the ability to detect AN with an both tissue and noninvasively collected patient samples. AUC of 0.783 (95% CI 0.662–0.904, p = 0.0005), a sensi- LINC00473 is a lncRNA with pro-apoptotic tumor sup- tivity of 84% (95% CI 69–94%), a specificity of 63% (95% pressor properties in CRC whose expression is downreg- CI 38–84%), a PPV of 82% and an NPV of 67% (Fig. 4F). ulated in this tumor type. This lncRNA is able to sponge Similar to our previous analysis in tissue samples, endogenous miR574-5p or miR15b-5p, inhibit cell pro- we also found no significant effect of age (p > 0.05) on liferation and colony formation capacity, and induce LINC00473 methylation levels in cfDNA samples. In cell apoptosis by activating the APAF1-CASP9-CASP3 addition, no significant difference (p > 0.05) in meth- pathway [15]. In addition, a genome-wide analysis of ylation levels according to the sex of the individuals was the CRC cell line HCT-116 recently identified promoter found (data not shown). hypermethylation of LINC00473 [14]. However, informa- tion is lacking on the methylation status of LINC00473 Clinical utility of LINC00473 promoter methylation in CRC and the regulation of this lncRNA by this epige- for the noninvasive detection of colorectal cancer netic mechanism. Thus, in this work, we confirmed that during the follow‑up of metastatic patients LINC00473 is hypermethylated in CRC and that this epi- After confirming its utility in the detection of precancer- genetic modification is associated with the transcriptional ous lesions and CRC, we decided to evaluate whether silencing of this lncRNA. In line with this, previous works the methylation status of the LINC00473 promoter in have described the aberrant hypermethylation of various plasma cfDNA could be useful to diagnose the presence lncRNAs in several tumor types, including CRC [14, 19]. of CRC in the palliative setting. For this purpose, we The deregulation of DNA methylation is an epigenetic analyzed the methylation of the LINC00473 promoter alteration that takes place during colorectal carcino- in plasma cfDNA by ddPCR at various clinically rele- genesis and can be detected from precancerous lesions, vant time points from a cohort of six randomly selected such as ACP, to advanced stages of CRC [5]. In this con- mCRC patients (Cohort 8), whose disease evolution was text, our study revealed that the promoter of LINC00473 evaluated according to a standard clinical practice (serial is hypermethylated in primary colorectal tumors and serum carcinoembryonic antigen (CEA) determinations that this epigenetic deregulation is present from early to and computed tomography scans). As shown in Addi- advanced stages of CRC. Importantly, LINC00473 was tional file 6: Fig. S6, the plasma cfDNA methylation lev- also hypermethylated in ACP tissues, indicating that els of LINC00473 decreased with effective therapy and the hypermethylation of LINC00473 is an early event Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 9 of 13 in colorectal carcinogenesis that is present throughout screening context. While high PPV is desired when all stages of tumor development. In addition, we found the costs or risks of further testing are significant, NPV in primary colorectal tumors that the hypermethyla- gains importance when the disease is serious and curable tion of LINC00473 was associated with a decrease in the in its preclinical phase, as is the case of ACP and CRC expression of this lncRNA. These results are in line with [26]. Another advantage of analyzing the methylation of a previous work, which identified the downregulation of LINC00473 in plasma would be the higher adherence that expression and tumor suppressor properties of LINC00473 blood-based assays provide in comparison with widely in association with the initiation and pathogenesis of CRC used fecal tests, such as FIT [27]. [15]. Similarly, other authors have found epigenetic dereg- While we only performed a small proof-of-concept ulation of important TSGs, such as p16/CDKN2A and study, the methylation levels of LINC00473 in cfDNA were hMLH1, throughout the colorectal carcinogenesis process able to detect the presence of the disease during the fol- [20, 21]. In addition to biological effects on gene expression, low-up of metastatic CRC patients, anticipating, in some epigenetic alterations have shown clinical utility as tumor cases, the response and progression compared with a biomarkers at different stages of CRC disease [5]. In this standard protocol based on serial CEA determinations and regard, we found that the methylation status of LINC00473 CT scans [28]. These results show the promising potential in tissue samples was able to accurately identify ACPs and of LINC00473 methylation in cfDNA to facilitate disease the different stages of CRC from early to advanced dis- detection in the palliative setting of CRC patients. ease, indicating the feasibility of using the methylation of Altogether, the results of this work indicate that the LINC00473 as a biomarker for the early detection of CRC. methylation status of LINC00473 in cfDNA is a prom- There is a pressing need in the clinic to detect CRC at ising biomarker for the noninvasive early detection of the beginning of the disease, since the majority of patients CRC and its precancerous lesions that could be used in can be successfully treated if detected early [4]. Although the clinic. Although methylation analysis of cfDNA was colonoscopy is an effective diagnostic tool for the screen- performed in a limited number of patients, our results ing and early detection of CRC, it is an invasive method support further validation to confirm the clinical appli- that limits its application [10]. Currently, FIT is a nonin- cations of this novel epigenetic biomarker in large-scale vasive stool-based assay widely used for the early detec- prospective studies with asymptomatic screening partici- tion of CRC in screening programs; however, it has a low pants and colorectal cancer patients. Because the combi- sensitivity for the early detection of precancerous lesions nation of different types of biomarkers may increase the [9]. To date, several noninvasive biomarkers have been accuracy of diagnostic assays [22], future studies should proposed for the early detection of CRC, including cir- also evaluate whether the combination of LINC00473 culating epigenetic biomarkers [9]. Alterations in DNA with other circulating biomarkers may improve its high methylation are promising candidates for early diagnosis, diagnostic accuracy for the early detection of CRC. since they are covalent and stable marks that can be found early in carcinogenesis [22]. In line with this, the detection Conclusions of SEPT9 methylation in plasma cfDNA (Epi proColon), CRC is a leading cause of cancer mortality worldwide which was approved by the US Food and Drug Adminis- that is usually diagnosed at an advanced stage, highlight- tration (FDA) in 2016 and has been proposed as a nonin- ing the need for new early detection strategies. We evalu- vasive test for the early detection of CRC [23]. However, ated the clinical utility of the promoter methylation of the a large prospective study revealed that the detection of long noncoding RNA LINC00473 in plasma cfDNA as SEPT9 methylation has low diagnostic accuracy for CRC a noninvasive biomarker for the detection of CRC. The (sensitivity, 48.2%) and advanced precancerous lesions methylation of LINC00473 showed high diagnostic accu- (sensitivity, 11.2%) [24], highlighting the need for new racy to detect CRC and associated precancerous lesions accurate, noninvasive tests for the early detection and both in tissues and in cfDNA, indicating that the meth- management of CRC. In this work, we found that, similar ylation of LINC00473 has a huge potential as a biomarker to tissue samples, the promoter of LINC00473 was hyper- for the noninvasive early detection of CRC and related methylated in plasma cfDNA of CRC and ACP patients. precancerous lesions. Of note, the methylation of LINC00473 in cfDNA allowed the identification of CRC and ACP with very good AUCs, Methods indicating that it is an assay with high diagnostic accu- Cancer cell lines and treatments racy for the very early detection of CRC [25]. In addition The human CRC cell line HCT-116 (American Type Cul- to showing interesting specificity and PPV, methylation ture Collection [ATCC]) was cultured in DMEM with of the LINC00473 promoter in cfDNA also showed high GlutaMAX (Gibco) supplemented with 10% fetal bovine sensitivity and NPV in a CRC and precancerous lesions serum (Sigma-Aldrich) and 1% penicillin/streptomycin Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 10 of 13 (Gibco) at 37 °C and 5% C­ O2. To promote DNA dem- to sodium bisulfite treatment using the EZ DNA Meth- ethylation, HCT-116 cells were treated with 5-aza-2′- ylation-Gold kit (Zymo Research). A 490-bp fragment deoxycytidine (AZA) (Sigma-Aldrich) at 5 μM for 72 h. of the LINC00473 promoter was amplified using 2 μL of Treatment was performed in triplicate, and data were bisulfite-converted DNA with ImmolaseTaq polymerase uct was gel-purified (2% agarose) with NucleoSpin® Gel compared with the corresponding non-treated cell line. (Bioline) at 60 °C for 40 cycles. The resulting PCR prod- Study participants and PCR Clean-up (Macherey–Nagel) and then cloned In this retrospective study, methylation levels of into the pGEMT Easy Vector System (Promega) follow- LINC00473 were analyzed in eight independent clini- ing the manufacturers’ protocols. For all samples, 10 using NucleoSpin® 96 Plasmid (Macherey–Nagel) and cal cohorts of patients, including four cohorts of colo- colonies were randomly chosen, and DNA was purified rectal tissue samples (Cohorts 1 to 4) and four cohorts of plasma samples (Cohorts 5 to 8) (Additional file 7: sequenced with a 3730 DNA Analyzer (Applied Bio- Table S1). Polyps were categorized into advanced colo- systems). Results were transformed into percentages of rectal polyps (ACPs) and non-ACPs (N-ACPs). ACP was CpGs showing methylation. defined as adenomas of at least 10 mm in size, contain- ing high-grade dysplasia or with tubulovillous or villous Gene expression analysis was reverse-transcribed using the ThermoScript™ RT- histology, or a serrated polyp of at least 10 mm in size or For quantitative RT-PCR (qRT-PCR), 2 μg of total RNA containing any grade of dysplasia. Advanced neoplasia (AN) was defined as CRC or ACP [29]. The main clini- PCR System (Invitrogen) according to the manufacturer’s cal characteristics of the respective cohorts are described recommendations. Reactions for qRT-PCR were per- in Additional file 8: Table S2 and Additional file 9: formed in triplicate on an Applied Biosystems 7,900HT Table S3. Inclusion and exclusion criteria are detailed in Fast Real-Time PCR using 25–50 ng cDNA and TaqMan Additional file 10: Methods. gene expression assays (LINC00473: Hs03677577_m1; GAPDH: Hs02758991_g1) as previously described [14]. Blood sample collection and plasma isolation GAPDH was used as an endogenous control, and water Blood samples were collected by phlebotomy into collec- as a negative control. RNA-seq data of LINC00473 from tion tubes with EDTA as an anticoagulant. Plasma was CRC cell lines and normal colon mucosa were obtained isolated within 2 h after collection by an initial centrifu- from the public database Gene Expression Omnibus gation at 1,500–1,600 g for 10 min at 4 °C, followed by (GEO) (GSE138734). RNA-seq data of LINC00473 in a second centrifugation at 15,000–16,000 g for 10 min at CRC primary tumors and matched normal tissues were 4 °C. Isolated plasma was stored at −80 °C until analysis. obtained from TCGA. Isolation of nucleic acids from cell lines, tissues and plasma DNA methylation analysis of 450K array data samples To analyze the DNA methylation levels of the LINC00473 Genomic DNA (gDNA) and total RNA were isolated promoter in CRC cell lines, we obtained 450K array data from cell lines using TRIzol (Invitrogen) according to (β-values) from the public database GEO (GSE49143). the manufacturer’s protocol. gDNA was also isolated For the methylation analysis of LINC00473 promoter in from formalin-fixed paraffin-embedded (FFPE) colorec- primary colorectal tumors and matched normal tissues, tal tissues with the AllPrep DNA/RNA FFPE Kit (Qia- we obtained the β-values of the 450K array from TCGA. the QIAamp® Circulating Nucleic Acid Kit (Qiagen) and gen). cfDNA was isolated from 2–4 mL of plasma using Methylation analysis of the LINC00473 promoter the vacuum system QIAvac 24 Plus (Qiagen) following in colorectal tissues by pyrosequencing the manufacturer’s recommendations. The quality and The methylation status of the LINC00473 promoter was quantity of gDNA and RNA were evaluated with a Nan- analyzed in primary colorectal tumors and matched nor- the QuantiFluor® ONE dsDNA kit with the Quantus™ oDrop (Thermo Fisher), and cfDNA was quantified by mal tissues by pyrosequencing. Primers for PCR amplifica- tion and sequencing were designed using PyroMark Assay Fluorometer (Promega). DNA and RNA were stored Design 2.0 software (Qiagen). 500 ng of DNA was bisulfite- at − 80 °C until analysis. converted with the EZ-96 DNA Methylation kit (Zymo Research) and used as a template for subsequent PCR. Bisulfite genomic sequencing Before pyrosequencing, PCR products were observed on Methyl Primer Express v1.0 software (Applied Biosys- 2% agarose gels. Pyrosequencing and methylation quanti- tems) and Primer3 (v.0.4.0) were used to design primers fication were performed in a PyroMark Q96 System (Qia- for the methylation analysis. DNA (1 µg) was subjected gen) according to the manufacturer’s instructions. CpG Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 11 of 13 site methylation was quantified using Pyro Q-CpG 1.0.9 of 4 °C. The temperature ramp increment was 2.5 °C/s the QX200™ Droplet Reader (Bio-Rad), and QuantaSoft (Qiagen). Water was used as negative control. for all steps. Droplets were counted and analyzed using Methylation analysis of the LINC00473 promoter in cfDNA analysis (Bio-Rad) was performed to acquire data. Water by qMSP was included as a no-template control, and HCT-116 and The methylation levels of the LINC00473 promoter in NL as positive controls for methylation and unmethyla- plasma cfDNA were determined by quantitative meth- tion, respectively. Reactions were performed in triplicate. ylation-specific PCR (qMSP) in a StepOne Plus system DNA methylation was expressed according to the follow- (Applied Biosystems). cfDNA (15–50 ng) was bisulfite- ing formula: Methylation (%) = [M/(U + M)] × 100, where converted with the EZ DNA Methylation-Lightning Kit M represents the copies/μl of methylated cfDNA, and U (Zymo Research) according to the manufacturer’s recom- the copies/μl of unmethylated cfDNA. converted cfDNA as a template, 10 µl Power SYBR™ mendations. Each reaction contained 2 µL of bisulfite- Statistical analysis Green PCR Master Mix (Thermo Fisher) and 150 nM The Kolmogorov–Smirnov test was used to evaluate the each forward and reverse primers in a total volume of normality of the distribution of the data. Subsequently, 20 µl. Thermocycling conditions in the StepOne Plus the nonparametric Mann–Whitney U test was used for system were as follows: 10 min at 95 °C, followed by 50 the comparison of methylation data. Nonparametric cycles of 94 °C for 15 s and 60 °C for 30 s. Water was regression with generalized additive models (GAMs) [31] included as a no-template control. HCT-116 and normal was used to evaluate the effect of age on methylation lev- leukocytes (NLs) were used as positive controls for meth- els. To assess the diagnostic accuracy, a receiver operat- ylation and unmethylation, respectively. All the samples ing characteristic (ROC) curve was generated. To obtain and controls were analyzed in triplicate. The DNA meth- the greatest combination of sensitivity and specificity, ylation level was expressed as a percentage of methylation the Youden index (J) was used: J = sensitivity + specific- (%) according to the following formula [30]: Methylation ity − 1 [32]. The positive predictive value (PPV) and neg- (%) = 100/[1 + 2^(CTCG–CTTG)], where ­CTCG and ­CTTG ative predictive value (NPV) were calculated: PPV = true represent, respectively, the threshold cycle (CT) of the positive/(true positive + false positive); NPV = true nega- methylation and unmethylation status of the LINC00473 tive/(true negative + false negative). Correlation analysis promoter. between DNA methylation and expression levels was evaluated using Pearson’s coefficient. GraphPad Prism 7.0 Methylation analysis of the LINC00473 promoter in cfDNA software was used for statistical analysis and graphic rep- by ddPCR resentation. All expressed p values were calculated with Methylation of the LINC00473 promoter was analyzed two-tailed tests and were considered significant when the by droplet digital PCR (ddPCR) in a QX200 system (Bio- p value < 0.05. Rad). cfDNA (30–50 ng) was bisulfite-converted with the EZ DNA Methylation-Lightning Kit (Zymo Research) Abbreviations according to the manufacturer’s recommendations. A ACP: Advanced colorectal polyp; AN: Advanced neoplasia; AUC​: Area under custom Bio-Rad assay to detect LINC00473 methyla- the curve; AZA: 5-Aza-2′-deoxycytidine; CpGI: CpG island; CRC​: Colorectal cancer; LINC00473: Long intergenic non-protein coding RNA 473; N-ACP: Non- tion (LINC00473-M) or unmethylation (LINC00473-U) advanced colorectal polyp; ROC: Receiver operating characteristic. was designed. First, a multiplex preamplification reac- DNA, 25 μl SsoAdvanced™ PreAmp Supermix (Bio- tion was performed using ~ 2 ng of bisulfite-converted Supplementary Information The online version contains supplementary material available at https://​doi.​ Rad), 0.5 μl of LINC00473-M and 0.5 μl of LINC00473- org/​10.​1186/​s13148-​022-​01302-x. U in a total volume of 50 μl. PCR conditions were as follows: 3 min at 95 °C, 10 cycles of 95 °C for 15 s and Additional file 1: Figure S1. Expression levels of LINC00473 in colorectal cancer tissues. Expression levels of LINC00473 were determined in tissues 56.2 °C for 4 min, and a final hold step of 4 °C. Next, a from primary colorectal cancer and matched normal colorectal mucosa multiplex reaction mix was prepared by combining 2 µL (controls) by RNA-seq data obtained from The Cancer Genome Atlas of the preamplification product, 11 μl ddPCR Supermix (TCGA). P, p value analyzed by Mann–Whitney U test. Ctrl, controls; CRC, colorectal cancer. for Probes (No dUTP) (Bio-Rad), 2.2 μl of LINC00473- Additional file 2: Figure S2. Evaluation of the diagnostic accuracy of the The QX200™ Droplet Generator (Bio-Rad) was used to M and 2.2 μl of LINC00473-U in a total volume of 22 μl. methylation of LINC00473 to detect colorectal cancer stages. ROC curve analysis evaluating the methylation of LINC00473 for the detection of CRC generate droplets. Thermocycling conditions were as at stage I (A), II (B), III (C) and IV (D), in tissue samples from primary colorec‑ tal cancer and matched normal colorectal mucosa (controls) by 450K array follows: 10 min at 95 °C, 40 cycles of 95 °C for 15 s and 56.2 °C for 30 s; 98 °C for 10 min and a final hold step Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 12 of 13 Funding (Cohort 1). P, p-value analyzed by ROC curve; AUC, area under the ROC This research was co-funded by the ISCIII (PI18/00307) and the European curve; Ctrl, controls; CRC, colorectal cancer; Se, sensitivity; Sp, specificity. Regional Development Fund (FEDER), PRIS3 grant from ACIS and Xunta de Additional file 3: Figure S3. Validation of the diagnostic accuracy of the Galicia, 2015 Merck Serono research grant, all donors who participated in methylation of LINC00473 to detect colorectal cancer stages. ROC curve the Liquid Biopsy Crowdfunding campaign in 2017, and Advanced Marker analysis evaluating the methylation of LINC00473 for the detection of Discovery S.L. (Amadix). J.R.-B. was supported by a Río Hortega fellowship from CRC at stage I (A), II (B), III (C) and IV (D), in tissue samples from primary ISCIII (CM19/00087) and is supported by a Juan Rodés contract (JR21/00019) colorectal cancer and matched normal colorectal mucosa (controls) by and a 2020 TTD Research Grant from the Spanish Cooperative Group for the pyrosequencing (Cohort 2). P, p-value analyzed by ROC curve; AUC, area Treatment of Digestive Tumors. A.R.-C. is supported by the Roche-Chus Joint under the ROC curve; Ctrl, controls; CRC, colorectal cancer; Se, sensitivity; Unit (IN853B 2018/03) funded by GAIN, “Consellería de Economía, Emprego Sp, specificity. e Industria.” N.C.-F. is funded by a predoctoral contract from “Xunta de Galicia” (IN606A-2020/004). A.B.-C. is funded by a predoctoral contract PFIS from Additional file 4: Figure S4. Methylation levels of LINC00473 in tissue ISCIII (FI19/00240) co-funded by “Fondo Social Europeo” (FSE). L.M.-R. is sup‑ colorectal polyps and colorectal cancer. A-B Comparison of methyla‑ ported by the AECC. ABC is a Miguel Servet researcher (ISCIII; CP17/0008). tion of LINC00473 between tissues from colorectal polyps, N-ACP, ACP A.G. is funded by CA181572, CA184792, CA202797 and CA214254 grants and CRC and matched normal colorectal mucosa (controls) analyzed by from the National Cancer Institute, National Institutes of Health. A.D.-L. is pyrosequencing (Cohort 3). C-D ROC curve analysis evaluating the meth‑ funded by a contract Juan Rodés from ISCIII (JR17/00016). V.M. is funded by ylation of LINC00473 for the detection of colorectal polyps and colorectal the Agency for Management of University and Research Grants (AGAUR) of cancer with respect to controls (Cohort 3). Horizontal lines represent the Catalan Government grant 2017SGR723, the Instituto de Salud Carlos III, mean methylation levels of LINC00473. P, p-value analyzed by Mann–Whit‑ co-funded by FEDER funds—a way to build Europe—grants PI17-0009, and ney U test or ROC curve; AUC, area under the ROC curve; Ctrl, controls; P, the Spanish Association Against Cancer (AECC) Scientific Foundation grant polyps; CRC, colorectal cancer; N-ACP, non-advanced colorectal polyps; GCTRA18022MORE. ACP, advanced colorectal polyps. Additional file 5: Figure S5. Validation of the methylation levels of Availability of data and materials LINC00473 to detect tissue precancerous lesions. A-C Methylation levels of The methylation and expression data obtained from the 450K array and RNA- LINC00473 promoter in tissues from premalignant colorectal polyps and seq used in this study are publicly available in TCGA and GEO (GSE49143 and normal colorectal mucosa (controls) by pyrosequencing (Cohort 4). D-E GSE138734). ROC curve analysis evaluating the methylation of LINC00473 promoter for the detection of premalignant colorectal polyps (Cohort 4). Horizontal lines represent mean methylation levels of LINC00473. P, p-value analyzed Declarations by Mann–Whitney U test or ROC curve; AUC, area under the ROC curve; Ctrl, controls; P, polyps; N-ACP, non-advanced colorectal polyps; ACP, Ethics approval and consent to participate advanced colorectal polyps. This study was approved by the Galician Ethical Committee (2017/538) and conducted in accordance with the guidelines for Good Clinical Practice Additional file 6: Figure S6. Clinical utility of LINC00473 methylation and the Declaration of Helsinki. Patients signed the informed consent to for the noninvasive detection of colorectal cancer during the follow-up. participate. Methylation levels of LINC00473 promoter (blue color) were evaluated in serial plasma cfDNA samples at clinically relevant time points from 6 Consent for publication randomly selected metastatic CRC patients by ddPCR. CEA (red color) Not applicable. was also analyzed in the same patients. ddPCR, droplet digital PCR; CEA, carcinoembryonic antigen; CRC, colorectal cancer; 1L, first-line therapy ini‑ Competing interests tiation; PD, progressive disease; PR partial response; Pre-PTR, pre-primary J.R.-B. has received honoraria for educational activities from Roche; honoraria tumor resection; Pre-MR, pre-metastases resection; SD, stable disease; F-U, for consultancies from Boehringer Ingelheim; institutional research funding follow-up from Roche; and travel, accommodations, expenses from Bristol-Myers Squibb, Additional file 7. Table S1. General description of the patient cohorts Merck Sharp & Dohme, Ipsen, PharmaMar, Merck, Pfizer, and Roche. E.B.-V. has included in the study. received honoraria for consultancies from Leo, and Rovi; and travel, accommo‑ dations, expenses from Servier, Merck, Sanofi, Roche, Pierre Fabre, and Amgen. Additional file 8. Table S2. Demographic and clinical characteristics of Y.V.-I. has received honoraria for consultancies from PharmaMar. F.V.-R. has the patient cohorts included in the study. received honoraria for consultancies from Roche, Eisai, and Servier; and travel, Additional file 9. Table S3. Demographic and clinical characteristics of accommodations, expenses from Lilly, Merck, and Pierre Fabre. S.C.-F. has metastatic colorectal cancer patients from cohort 8. received honoraria for educational activities from Servier, and Pierre Fabre; and travel, accommodations, expenses from Lilly, and Merck. R.L.-L. has received Additional file 10. Methods. Inclusion and exclusion criteria. honoraria for participation in Advisory Boards from Roche, AstraZeneca, Merck, Merck Sharp & Dohme, Bayer, Bristol-Myers Squibb, Novartis, Janssen, Lilly, Pfizer, and Leo; travel, accommodations, and expenses from PharmaMar, Acknowledgements Roche, Bristol-Myers Squibb, and Pierre Fabre; research funding from Roche We would like to thank all the patients and healthy volunteers included in this and Merck; and is co-founder and shareholder in Nasasbiotech, S.L., Mtrap Inc. study for their kind collaboration, all donors who participated in the Liquid A.R.-C., A.B.C., R.L.-L. and A.D.-L. are inventors in one patent application over Biopsy Crowdfunding campaign in 2017, and all personnel who provided these results licensed to Advanced Marker Discovery, S.L. (Amadix). L.C., A.C.M. technical support with sample collection. and R.A. are employees of Advanced Marker Discovery, S.L. (Amadix). The rest of the authors declare no potential conflicts of interest. Author contributions J.R.-B., A.R.-C., R.L.-L. and A.D.-L. contributed to conceptualization; J.R.-B., Author details A.A.-C., M.C-P., L.M.-R., J.A.V.-B., E.B.-V., Y.V.-I., F.V.-R., S.C.-F., R.S.-P., V.M., A.G., L.C., 1 Translational Medical Oncology Group (ONCOMET), Health Research Institute A.C.M., R.A. and M.E. provided resources; A.R.-C., A.A.-C., J.R.-B., E.B.-V. and R.S.-P. of Santiago de Compostela (IDIS), University Clinical Hospital of Santiago curated the data; A.R.-C., J.R.-B., N.C.F, A.B.C., M.M.-P., R.M.L.-L. and A.D.L. con‑ (CHUS/SERGAS), 15706 Santiago de Compostela, Spain. 2 Centro de Inves‑ tributed to methodology; J.R.-B., A.R.-C., N.C.-F., A.B.-C. and A.D.-L. performed tigación Biomédica en Red Cáncer (CIBERONC), ISCIII, 28029 Madrid, Spain. formal analysis; J.R.-B., A.R.-C. and A.D.-L. helped in writing—original draft 3 Cancer Epigenomics Laboratory, Epigenomics Unit, Translational Medical preparation; J.R.-B., A.R.-C., A.C.M., A.B.-C, R.L.-L. and A.D.-L. helped in writing— Oncology Group (ONCOMET), Health Research Institute of Santiago de review and editing; A.D.-L supervised the study. All authors have read and Compostela (IDIS), University Clinical Hospital of Santiago (CHUS/SERGAS), agreed to the published version of the manuscript. 15706 Santiago de Compostela, Spain. 4 Roche‑Chus Joint Unit, Translational Ruiz‑Bañobre et al. Clinical Epigenetics (2022) 14:86 Page 13 of 13 Medical Oncology Group (ONCOMET), Health Research Institute of Santiago 13. Léveillé N, Melo CA, Rooijers K, Díaz-Lagares A, Melo SA, Korkmaz G, et al. (IDIS), 15706 Santiago de Compostela, Spain. 5 Universidade de Santiago Genome-wide profiling of p53-regulated enhancer RNAs uncovers a de Compostela (USC), 15782 Santiago de Compostela, Spain. 6 Department subset of enhancers controlled by a lncRNA. Nat Commun. 2015;6:6520. of Gastroenterology and Hepatology, University Clinical Hospital of Santiago 14. Diaz-Lagares A, Crujeiras AB, Lopez-Serra P, Soler M, Setien F, Goyal A, et al. (CHUS/SERGAS), 15706 Santiago de Compostela, Spain. 7 Liquid Biopsy Analy‑ Epigenetic inactivation of the p53-induced long noncoding RNA TP53 sis Unit, Translational Medical Oncology Group (ONCOMET), Health Research target 1 in human cancer. Proc Natl Acad Sci USA. 2016;113:E7535–44. Institute of Santiago de Compostela (IDIS), 15706 Santiago de Compostela, 15. Wu H, Hu X, Li Y, Chen Q, Sun T, Qiao Y, et al. LNC473 regulating APAF1 Spain. 8 Department of Pathology, Complejo Hospitalario Universitario de IRES-dependent translation via competitive sponging miR574 and Ferrol (SERGAS), 15405 Ferrol, Spain. 9 Unit of Biomarkers and Susceptibil‑ miR15b: implications in colorectal cancer. Mol Ther Nucleic Acids. ity, Oncology Data Analytics Program, Catalan Institute of Oncology (ICO), 2020;21:764–79. 08907 Barcelona, Spain. 10 Colorectal Cancer Group, Bellvitge Biomedical 16. Petko Z, Ghiassi M, Shuber A, Gorham J, Smalley W, Washington MK, et al. Research Institute (IDIBELL), 08907 Barcelona, Spain. 11 Biomedical Research Aberrantly methylated CDKN2A, MGMT, and MLH1 in colon polyps and in Centre Network for Epidemiology and Public Health (CIBERESP), 28029 Madrid, fecal DNA from patients with colorectal polyps. Clin Cancer Res Off J Am Spain. 12 Department of Clinical Sciences, Faculty of Medicine and Health Assoc Cancer Res US. 2005;11:1203–9. Sciences, University of Barcelona, 08907 Barcelona, Spain. 13 Center for Gastro‑ 17. Bray F, Ferlay J, Soerjomataram I, Siegel RL, Torre LA, Jemal A. Global intestinal Research, Center for Translational Genomics and Oncology, Baylor cancer statistics 2018: GLOBOCAN estimates of incidence and mortality Scott and White Research Institute, Charles A Sammons Cancer Center, Baylor worldwide for 36 cancers in 185 countries. CA Cancer J Clin Am Cancer University Medical Center, Dallas, TX, USA. 14 Department of Molecular Diag‑ Soc. 2018;68:394–424. nostics and Experimental Therapeutics, Beckman Research Institute of City 18. Du Z, Yu T, Sun M, Chu Y, Liu G. The long non-coding RNA TSLC8 inhibits of Hope, Monrovia, CA, USA. 15 City of Hope Comprehensive Cancer Center, colorectal cancer by stabilizing puma. Cell Cycle. 2020;19:3317–28. Duarte, CA, USA. 16 Advanced Marker Discovery (AMADIX), 47004 Valladolid, 19. Yang Z, Xu F, Wang H, Teschendorff AE, Xie F, He Y. Pan-cancer charac‑ Spain. 17 Josep Carreras Leukaemia Research Institute (IJC), Barcelona, Spain. terization of long non-coding RNA and DNA methylation mediated 18 Institucio Catalana de Recerca I Estudis Avançats (ICREA), Barcelona, Spain. transcriptional dysregulation. EBioMedicine. 2021;68: 103399. 19 Physiological Sciences Department, School of Medicine and Health Sciences, 20. Gonzalez-Zulueta M, Bender CM, Yang AS, Nguyen T, Beart RW, Van Tor‑ University of Barcelona (UB), Barcelona, Spain. 20 Epigenomics in Endocrinology nout JM, et al. Methylation of the 5’ CpG island of the p16/CDKN2 tumor and Nutrition Group, Epigenomics Unit, Health Research Institute of San‑ suppressor gene in normal and transformed human tissues correlates tiago de Compostela (IDIS), University Clinical Hospital of Santiago (CHUS/ with gene silencing. Cancer Res (United States). 1995;55:4531–5. SERGAS), 15706 Santiago de Compostela, Spain. 21 Centro de Investigación 21. Kane MF, Loda M, Gaida GM, Lipman J, Mishra R, Goldman H, et al. Biomédica en Red Fisiopatología de La Obesidad y Nutrición (CIBERobn), ISCIII, Methylation of the hMLH1 promoter correlates with lack of expression of 28029 Madrid, Spain. hMLH1 in sporadic colon tumors and mismatch repair-defective human tumor cell lines. Cancer Res (United States). 1997;57:808–11. Received: 22 April 2022 Accepted: 17 June 2022 22. Diaz-Lagares A, Mendez-Gonzalez J, Hervas D, Saigi M, Pajares MJ, Garcia D, et al. A novel epigenetic signature for early diagnosis in lung cancer. Clin Cancer Res. 2016;22:3361–71. 23. Pickhardt PJ. Emerging stool-based and blood-based non-invasive DNA tests for colorectal cancer screening: the importance of cancer References prevention in addition to cancer detection. Abdom Radiol (New York). 1. Arnold M, Sierra MS, Laversanne M, Soerjomataram I, Jemal A, Bray F. 2016;41:1441–4. Global patterns and trends in colorectal cancer incidence and mortality. 24. Church TR, Wandell M, Lofton-Day C, Mongin SJ, Burger M, Payne SR, et al. Gut (England). 2017;66:683–91. Prospective evaluation of methylated SEPT9 in plasma for detection of 2. John SKP, George S, Primrose JN, Fozard JBJ. Symptoms and signs in asymptomatic colorectal cancer. Gut. 2014;63:317–25. patients with colorectal cancer. Colorectal Dis Off J Assoc Coloproctology 25. Šimundić A-M. Measures of diagnostic accuracy: basic definitions. EJIFCC. (Gt Britain Irel England). 2011;13:17–25. 2009;19:203–11. 3. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, et al. 26. Trevethan R. Sensitivity, specificity, and predictive values: foundations, Global cancer statistics 2020: GLOBOCAN estimates of incidence and pliabilities, and pitfalls in research and practice. Front Public Health. mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin Am 2017;5:307. Cancer Soc. 2021;71:209–49. 27. Parikh RB, Prasad V. Blood-based screening for colon cancer: a dis‑ 4. Siegel RL, Miller KD, Goding Sauer A, Fedewa SA, Butterly LF, Anderson JC, ruptive innovation or simply a disruption? JAMA (United States). et al. Colorectal cancer statistics, 2020. CA Cancer J Clin. 2020;70:145–64. 2016;315:2519–20. 5. Rodriguez-Casanova A, Costa-Fraga N, Bao-Caamano A, López-López R, 28. National Comprehensive Cancer Network. Colon Cancer (Version 4.2020). Muinelo-Romay L, Diaz-Lagares A. Epigenetic landscape of liquid biopsy https://​www.​nccn.​org/​profe​ssion​als/​physi​cian_​gls/​pdf/​colon.​pdf. in colorectal cancer. Front Cell Dev Biol. 2021;9:622459. Accessed 6 Dec 2020. 6. Nguyen LH, Goel A, Chung DC. Pathways of colorectal carcinogenesis. 29. Regula J, Rupinski M, Kraszewska E, Polkowski M, Pachlewski J, Orlowska Gastroenterology. 2020;158:291–302. J, et al. Colonoscopy in colorectal-cancer screening for detection of 7. Rex DK, Boland CR, Dominitz JA, Giardiello FM, Johnson DA, Kaltenbach advanced neoplasia. N Engl J Med (United States). 2006;355:1863–72. T, et al. Colorectal cancer screening: recommendations for physicians 30. Cottrell S, Jung K, Kristiansen G, Eltze E, Semjonow A, Ittmann M, et al. and patients from the U.S. multi-society task force on colorectal cancer. Discovery and validation of 3 novel DNA methylation markers of prostate Gastroenterology. 2017;153:307–23. cancer prognosis. J Urol (United States). 2007;177:1753–8. 8. Song L-L, Li Y-M. Current noninvasive tests for colorectal cancer screen‑ 31. Benedetti A, Abrahamowicz M. Using generalized additive models to ing: an overview of colorectal cancer screening tests. World J Gastrointest reduce residual confounding. Stat Med (England). 2004;23:3781–801. Oncol. 2016;8:793–800. 32. Fluss R, Faraggi D, Reiser B. Estimation of the Youden Index and its associ‑ 9. Inadomi JM. Screening for colorectal neoplasia. N Engl J Med (United ated cutoff point. Biom J (Germany). 2005;47:458–72. States). 2017;376:149–56. 10. Triantafillidis JK, Vagianos C, Malgarinos G. Colonoscopy in colorectal cancer screening: current aspects. Indian J Surg Oncol. 2015;6:237–50. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ 11. Adler A, Geiger S, Keil A, Bias H, Schatz P, deVos T, et al. Improving compli‑ lished maps and institutional affiliations. ance to colorectal cancer screening using blood and stool based tests in patients refusing screening colonoscopy in Germany. BMC Gastroenterol. 2014;14:183. 12. Esteller M. Non-coding RNAs in human disease. Nat Rev Genet. 2011;12:861–74.