---
title: "Autoantibodies to DNA repair proteins found in cancer patients and in healthy children before diag"
id: "frontiers-in-immunology-5-pathogenic-autoantibodies-targeting-dna-repair-proteins-are-shared-by-cancer"
canonical_url: "https://medichelpline.com/clinical-feed/frontiers-in-immunology-5-pathogenic-autoantibodies-targeting-dna-repair-proteins-are-shared-by-cancer"
content_type: "clinical_feed_article"
specialty: "Oncology"
source_name: "Frontiers in Immunology"
source_url: "https://www.frontiersin.org/articles/10.3389/fimmu.2026.1911932"
published_at: "2026-10-14T00:00:00.000Z"
evidence_level: "Journal Feed"
license: "CC-BY-NC-4.0 / Informational Use"
---
# Autoantibodies to DNA repair proteins found in cancer patients and in healthy children before diag
## Provenance & Clinical Metadata
- **Canonical URL:** https://medichelpline.com/clinical-feed/frontiers-in-immunology-5-pathogenic-autoantibodies-targeting-dna-repair-proteins-are-shared-by-cancer
- **Specialty:** [Oncology](https://medichelpline.com/clinical-feed/oncology.md)
- **Primary Source:** Frontiers in Immunology
- **Source URL:** [Original Journal Publication](https://www.frontiersin.org/articles/10.3389/fimmu.2026.1911932)
- **Published At:** 2026-10-14T00:00:00.000Z
- **Evidence Rating:** Journal Feed
## Executive GIST (TL;DR)
- Investigators profiled sera from 375 individuals using high-density native **proteome arrays** to map circulating IgG autoantibodies across multiple cancer types and controls. - The study cohort combined adult and pediatric patients from Norwegian biobanks and pre-diagnostic samples from UK birth cohorts (ALSPAC, BiB); array data deposited as E-MTAB-13995. - Analysis identified an enrichment of autoantibodies directed at **DNA repair proteins** in pediatric patients across several tumor types; the repertoire showed age specificity. - Some of the most enriched autoantibodies in pediatric patients were also present in sera from otherwise healthy children who later developed cancer years afterward, based on pre-diagnostic cohort analysis. - Investigators tested the capacity of serum IgG to penetrate live cultured cells; they confirmed IgG internalization and described a specific intracellular localization pattern (TISa) more prevalent in pediatric patients. - Proteome microarrays were processed with standardized blocking, 1:1000 serum dilution, and detection using anti-GST and anti-human IgG; positive calls used z-score >2.51, rsd 3.5. - For pre-diagnostic samples, antigen-specific associations were examined with logistic regression followed by mixed-effects models to account for repeated samples; FDR correction was applied for six core antigens (ADRM1, DDX58, DHX58, DMC1, MGMT, UNG). - A pooled analysis created a composite binary variable indicating presence of at least one core antigen; individuals with multiple samples were classified positive if any sample was positive. - Confocal microscopy used a 1:40 serum dilution to evaluate IgG internalization (2 h incubation at 37 °C), followed by anti-human IgG staining and protein-specific antibodies for subcellular localization. - The authors note technical limitations inherent to protein arrays (batch effects) and removed affected proteins from analyses; discovery cohort lacked childhood controls but later stages included pediatric controls to adjust for false positives.
## Clinical Analysis & Structured Key Points
Frontiers | Pathogenic autoantibodies targeting DNA repair proteins are shared by cancer patients and healthy children later diagnosed with cancer ORIGINAL RESEARCH article Front. Immunol. , 14 October 2026 Sec. Cancer Immunity and Immunotherapy Volume 17 - 2026 | https://doi.org/10.3389/fimmu.2026.1911932 Published in Frontiers in Immunology Cancer Immunity and Immunotherapy 7 impact factor 11.3 citescore Part of a Research Topic Autoimmunity and Chronic Inflammation in Cancer: Mechanisms, Biomarkers, and Therapeutic Implications Submission open 4404 views 4 articles Editor & Reviewers Edited by Z C Zorica Cvetkovic Reviewed by E S Evgeny Suspitsin S T Samira Tabaei Outline Figures and Tables Figure 1 View in article Figure 2 View in article Figure 3 View in article Figure 4 View in article Figure 5 View in article Table 1 Heat map showing the frequencies (%) of the most enriched autoantigens in cancer patients and healthy controls. View in article Table 2 Data on the positive samples from British cohort. View in article ORIGINAL RESEARCH article Front. Immunol. , 14 October 2026 Sec. Cancer Immunity and Immunotherapy Volume 17 - 2026 | https://doi.org/10.3389/fimmu.2026.1911932 Pathogenic autoantibodies targeting DNA repair proteins are shared by cancer patients and healthy children later diagnosed with cancer A S Anastassia Serguienko 1 * J S Jeetesh Sharma 1 R S Ronas Shakya 2 M C Monica Cheng Munthe-Kaas 3 K B Kjetil Boye 4 O M Ola Myklebost 1,5,6 * 1. Department of Clinical Science, University of Bergen, Bergen, Norway 2. Department of Global Public Health and Primary Care, University of Bergen, Bergen, Norway 3. The Norwegian Childhood Cancer Biobank, The Department of Pediatric Oncology and Hematology, Oslo University Hospital, Oslo, Norway 4. Department of Oncology, Oslo University Hospital, Oslo, Norway 5. Department of Tumor Biology, Institute for Cancer Research, Oslo University Hospital, Oslo, Norway 6. Western Norway Familial Cancer Center, Haukeland University Hospital, Bergen, Norway See more Article metrics View details Abstract The etiology of pediatric cancer remains poorly understood and does not fit the classical view as an age-associated genetic disease, suggesting yet unknown mechanisms might be involved. Using high-density native proteome arrays, we analyzed sera from 375 healthy individuals and patients. We found a significant enrichment of autoantibodies against DNA repair proteins shared by cancer patients and healthy children diagnosed with cancer years later. These autoantibodies were detected in multiple cancer types and exhibited remarkable age-specificity. To explore their potential pathogenic significance, we assessed the ability of serum IgG to penetrate live cells. We confirmed that IgG can indeed enter live intact cells and identified a specific pattern of subcellular localization prevalent in pediatric patients. These findings offer a novel direction for the investigation of etiology of pediatric cancer and cancer in general. Introduction Autoantibodies - antibodies recognizing the body’s own proteins - are common in healthy people but are also involved in autoimmune disorders ( 1 , 2 ). They can be present at birth or can appear shortly after and increase toward older ages ( 3 – 6 ). Antinuclear autoantibodies at low concentrations have been reported in up to 40% of healthy people ( 7 , 8 ). Although the repertoires are highly heterogeneous, many autoantibodies are shared by a large fraction of healthy individuals ( 1 ). The pathogenic effects of autoantibodies are not limited to classical autoimmune diseases. In cancer patients, mutated proteins, so-called neoantigens, can trigger the generation of autoantibodies that cross-react with wild-type (wt) proteins, and wt proteins aberrantly expressed or post-translationally modified in the tumor can also induce autoantibodies ( 9 ). Such tumor-associated autoimmunity may give rise to paraneoplastic syndromes in distant tissues ( 10 , 11 ). Both positive and negative associations between autoimmune diseases and cancer have been reported, including an interesting population-based study showing a positive association between all tested autoimmune conditions and gastric cancer ( 12 ). While the mechanistic aspects of tumor-induced autoimmunity are quite well characterized, the higher cancer incidence among patients with some autoimmune diseases has not been well explained, with a major role being attributed to inflammation ( 12 , 13 ). Pediatric cancers arise at ages when their tissues of origin undergo critical developmental processes and waves of proliferation ( 14 ). Despite the increased risk of mutations due to the large number of cell divisions in growing tissues, pediatric cancers are rare and characterized by a paucity of somatic mutations ( 15 – 17 ). Although genetic predisposition is thought to play an important role in early cancers, only 6-10% of patients have so far been shown to have likely pathogenic gene variants ( 18 ). Interestingly, novel mutation signatures caused by unknown mechanisms have recently been identified ( 15 ). Therefore, the understanding of the etiology of pediatric cancer remains unsatisfactory. Using high-density human proteome arrays, we performed autoantibody profiling of serum samples collected at diagnosis from 175 patients with nine different cancer types and 141 healthy controls of corresponding ages. We found that the sera of pediatric patients with different cancer types were enriched in autoantibodies against DNA repair proteins. Strikingly, we found some of the most enriched autoantibodies also in the serum of healthy children who got cancer years later. Our data suggest that these autoantibodies might be involved in cancer etiology, which is further supported by our finding of specific intracellular localization of patients’ serum IgG. Therefore, pathogenic penetrating autoantibodies may represent a novel mechanism of cancer predisposition. Materials and methods Patient samples All patient and control groups are listed in Supplementary Table S1 . The main cohort, which included 91 adult sarcoma and ALL patients and 84 pediatric patients, was obtained from the NoSarC Norwegian Sarcoma Study, a collection of tumor and blood samples from >700 sarcoma patients from all Norwegian health regions collected from 2014-2017, from the Leukemia Center at UiB, and from the Norwegian Childhood Cancer Biobank, Oslo. Additional pediatric sarcoma samples were provided by the Oslo University Hospital Cancer Research Biobank. Anonymous control samples were obtained from blood donors, the Bergen Growth Study 2 ( 19 ), and anonymous surplus serum samples were obtained from small children visiting the Helse Bergen Trauma Clinic and their Pediatric Cancer Unit. Serum from healthy school children who later got cancer and matched controls were obtained from the UK ALSPAC (14901 donors) and BiB cohorts ( 20 , 21 ). Study data were collected and managed using REDCap electronic data capture tools hosted at the University of Bristol, a secure, web-based software platform designed to support data capture for research studies ( 22 ). At age 18, study children were sent ‘fair processing’ materials describing ALSPAC’s intended use of their health and administrative records and were given clear means to consent or object via a written form. For these cases, we have incomplete information on histological diagnoses. All patients provided informed consent to participate according to ethical approval 17866 from the Regional Committee for Medical Ethics, which also approved amendments for the inclusion of anonymous samples from healthy controls and additional samples from pediatric patients from the General Cancer Biobank and Norwegian Childhood Cancer Biobank ( 23 ), both at Oslo University Hospital, as part of the same approval number. Proteome microarray analysis HuProt v4.0 Human Proteome Microarrays (CDI Labs) were purchased from Cambridge Protein Arrays Ltd. Probing of the arrays and statistical data analyses were performed according to the manufacturer’s protocol. Specifically, the arrays were blocked with 2% BSA at 4 °C overnight. Serum samples were diluted 1:1000 in blocking buffer and incubated with the arrays with gentle shaking for 2h at RT. The arrays were washed 3 times in rapid succession and 2 times for 5 min. The arrays were then incubated with an anti-GST antibody (Cat. no ab117497, Abcam) and anti-human IgG secondary antibodies (Cat. no A-21089, Invitrogen) for 2h at RT, followed by washing as described previously. Detection was performed using the GenePix scanner, with the same settings maintained throughout the entire study. Preprocessing of raw data and statistical analysis were performed according to the standard protocol provided by Cambridge Protein Arrays Ltd. For neighborhood background correction, the net intensities were calculated by subtracting the median local background intensities from the median raw intensities. To obtain the corrected net intensities, a negative control array (no serum) was produced for each batch, and the net intensities were recalculated by subtracting the corrected background intensities from the net intensities for each corresponding spot on the serum-probed array from the same batch. The relative standard deviation (rsd) was calculated for each duplicate spot. For z-score-positive proteins, the cutoff was set at >2.51, corresponding to a p-value of 0.02; for rsd, the cutoff was set at 3.5, as recommended. The significant proteins were ranked using z-scores. An intrinsic technical drawback of protein array technology is the so-called “batch effect”. We observed that some batches of arrays failed to detect antigens that were previously efficiently detected by other batches. Proteins that were observed to be affected by batch effects were removed from the analysis and disregarded in this study. Statistical analysis For pre-diagnostic cohort, we examined antigen-specific immune responses in cases and controls using generalized linear mixed models. Six positive antigens were analyzed individually (ADRM1, DDX58, DHX58, DMC1, MGMT, and UNG) using a two-stage analytical strategy. First, for each antigen, we fitted a standard logistic regression model to obtain an initial estimate of the association between disease status and antigen reactivity, expressed as odds ratios (ORs). Second, because multiple blood samples were available from some individuals, we fitted mixed-effects logistic regression models to account for within-individual correlation. These models included a random intercept, allowing individual-specific baseline probabilities of antigen reactivity and thereby reducing bias arising from repeated measurements. To control for multiple testing across the six antigen-specific analyses, p-values were adjusted using the Benjamini–Hochberg false discovery rate (FDR) procedure. To further assess differences between cases and controls at the group level, we conducted a pooled analysis of the six core antigens. A composite binary variable was created to indicate the presence of at least one core antigen. Individuals with multiple blood samples were classified as positive if any core antigen was detected in any sample and negative otherwise. The association between pooled core antigen positivity and disease status was then evaluated using binary logistic regression, with odds ratios and 95% confidence intervals reported. The discovery cohort lacked childhood controls, which would allow false positives among pediatric patients for autoantibodies common in healthy children, but this would not be the case in the full study ( Table 1 ) where we had obtained a large number of samples from healthy children. In fact, in subsequent stages of our study, the analysis of controls from 0.6 to 19 years of age and of age-matched patients allowed the adjustment of the pediatric cohort for false positives and fully confirmed the same patterns we observed at the discovery stage. Table 1 Pediatric patients Adult patients Healthy controls total 84 total 91 total 141 N of samples 5 10 10 11 11 16 21 13 12 38 12 16 60 81 Tumor type Enbryonic NBL Wilm's Glioma ALL RMS OS ALL RMS OS UPS LPS Young Adult ALKBH7 20 20 18 9 8 2 1 APBB1 10 30 27 9 13 5 8 3 5 2 DMC1 20 18 19 10 2 1 MGMT 20 9 5 38 42 8 5 5 NUMA1 40 20 10 18 9 6 8 3 8 2 RAD52 20 40 30 36 36 31 14 23 8 3 UNG 9 25 5 62 58 10 1 DDX58 20 20 30 18 55 19 10 8 3 17 5 2 DHX58 20 30 18 64 13 5 3 Heat map showing the frequencies (%) of the most enriched autoantigens in cancer patients and healthy controls. Colored cells are shaded by prevalence. This core list includes autoantigens that were positive in at least 10% of patients in at least two different cancer groups.NBL, neuroblastoma; ALL, Acute lymphoblastic leukemia; RMS, rhabdomyosarcoma; OS, osteosarcoma; UPS, Undifferentiated pleomorphic sarcoma; LPS, Liposarcoma. Cell lines The panel of osteosarcoma cell lines was either established in-house from patient-derived xenografts (PDXes) from patient tumors (NRH-OS2 and NRH-OS3) or older cell lines (U2OS – RRID: CVCL_0042, SJSA-1 - RRID: CVCL_1697, IOR/OS18 - CVCL_9993, IOR/OS15 -CVCL_9991) available from the STR-validated EuroBoNeT collection ( 24 ), BJ (RRID: CVCL_3653) and MRC5 (RRID: CVCL_0440) primary human fibroblasts were obtained from ATCC (RRID: SCR_001672); primary MSC were produced in-house, and NHOst primary human osteoblasts (cat. no CC-2538) were obtained from Lonza, Walkersville, MD, US. Primary antibodies The following antibodies were used for confocal microscopy: ADRM1 (RRID: AB_10794044), APBB1 (RRID: AB_2882385), CGGBP1 (RRID: AB_2079283, RRID: AB_2675433), CRAT (RRID: AB_1847212), DHX58 (RRID: AB_1847665), EEPD1 (RRID: AB_2682223), FBL (RRID: AB_2105788), Golgin-97 (CDF4) (RRID: AB_221447, RRID: AB_2115315), GST (RRID: AB_10900368), HNRNPC (RRID: AB_2117500), Hsc7 (RRID: AB_2120153), anti-Human IgG (Invitrogen cat. no A-21089, RRID: AB_2535745) ING1 (RRID: AB_2681876), INO80E (RRID: AB_10794923), Nucleolin (ZN004) (RRID: AB_2533426), NuMA (RRID: AB_2154616), PLCD4 (RRID: AB_2675855), RSL1D1 (RRID: AB_2678504), RAD52 (Atlas Antibodies, cat. no HPA028957, RRID: AB_2672842), SON (RRID: AB_2674029), SRP19 (Proteintech cat. no 16033-1, RRID: AB_11041714), ZFP36L1 (Proteintech cat. no 12306-1, RRID: AB_2918780), ZFP36L1/2 (RRID: AB_2737443), and for ZO-1 (RRID: AB_2533456). Confocal microscopy The cells were seeded on coverslips in 12-well plates in regular culture medium. The next day, the medium was replaced with fresh culture medium containing donor serum at a 1:40 dilution. The cells were then incubated at 37 °C for 2h. After incubation, the cells were washed twice with PBS, fixed with 4% formaldehyde solution for 15 min, washed twice with PBS, permeabilized with 0.02% Tween-100 solution for 10 min, washed twice with PBS, and blocked for 30 min with 2% BSA solution. The cells were then incubated with anti-human IgG antibody overnight at 4 °C. The next day, the cells were either incubated for 2h at room temperature with primary antibodies against the protein of interest, followed by secondary antibodies, or incubated for 30 min with phalloidin. Finally, the coverslips were mounted on slides with Vectashield mounting medium containing DAPI. The samples were analyzed on Leica SP8-STED confocal microscope at 100X magnification using the basic setting, lightning mode, and z-stack mode for 3D reconstruction. The serum dilution 1:40 is a standard used in other studies on penetrating autoantibodies. However, we tested a range of dilutions below and above 1:40 to establish the best experimental protocol and concluded that 1:40 indeed was the optimal one. This concentration gives clear signal from internalized IgG with low background and saves scarce serum samples. Flow cytometry Cell Cycle analysis strategy: after exclusion of debris, single cells were gated for the IgG + and IgG - cell population, which was analyzed separately across the samples via the in-built cell cycle analysis tool in FlowJo. Due to biological variability between control serum-treated and patient serum-treated samples, the proportion of IgG + cells differed across conditions, with control serum samples consistently showing a lower percentage of IgG + cells. Each subset (IgG + /IgG - ) in both groups was redefined as 100% to allow calculation of the relative distribution of cells across G0/G1, S, and G2/M phases within that specific fraction. Importantly, this normalization reflects the internal cell cycle composition of each IgG-defined subpopulation and does not represent the proportion of the total sample. TISa pattern quantification The .vsi files from the Olympus VS120 S6 slide scanner acquired with the 10x UPlanSApo NA 0.4 objective were analyzed using QuPath software for quantitative image analysis (version 0.5.1) ( 25 ). First, DAPI-stained nuclei were detected with “Cell Detection”, and a radius of the investigated area (ROI) of 10 µm was allowed around the nuclei. Thereafter, “subcellular spot detection” was performed to identify spots with anti-human IgG staining within the ROI. The spot size was defined as an expected size of 50 µm 2 , a minimum spot size of 30 µm 2 and a maximum spot size of 250 µm 2 . The range was optimized to minimize the number of false-positive and false-negative results. The results of the analysis were confirmed by visual inspection. IgG isolation IgG were purified from the blood serum using the Nab Protein G Spin Kit according to the manufacturer’s protocol (Thermo Fisher Scientific Inc.). Availability of data and materials Requests for materials will be considered. The complete dataset was deposited in the ArrayExpress repository (RRID: SCR_002964) with the accession number E-MTAB-13995. ALSPAC data are available through a fully searchable data dictionary and variable search tool http://www.bristol.ac.uk/alspac/researchers/our-data/ . Data Access Statement: Research data supporting this publication are available from the ArrayExpress repository with the accession number E-MTAB-13995 (Public only after publication, use Link for anonymous access). Results Sera from pediatric cancer patients are enriched in IgG against DNA repair proteins Osteosarcoma is more common in children and young people, with an incidence that peaks at approximately 19 years of life and then flattens towards older ages ( Figure 1A ). Our discovery cohort consisted of 60 osteosarcoma patients aged between 2 and 76 years at diagnosis and 55 healthy controls aged between 22 and 69 years. Serum samples were screened using high-density proteome arrays to determine the repertoire of circulating IgG autoantibodies. Figure 1 Age distribution of pediatric cancer patients and autoantibodies. (A) Age incidence of the cancers included in this study in Norway from 1980 to 2019. The data are plotted as 5-year moving averages for better visibility. The colored arrows indicate the peaks of incidence in children. The bar indicates the pediatric (0-19), intermediate (20-29), and adult (30+) age groups. ALL – acute lymphoblastic leukemia; GLM – glioma; NBL – neuroblastoma; RMS – rhabdomyosarcoma; OS – osteosarcoma; WT – Wilms tumor; UPS – undifferentiated pleomorphic sarcoma; LPS – liposarcoma. Data courtesy of the Norwegian Cancer Registry. (B) The three most enriched (highest k/K ratio) molecular pathways for the autoantigen sets (z score > 2.5) from patient groups by GSEA. Full GSEA data are shown in Supplementary Data Set 1 . (C) Graphical presentation of the analyzed cohorts, colored by cancer types as in (D) , with healthy controls in green. Adult fractions (above 20 years) in solid color, children and young fractions in dotted shading. (D) Number and distribution of autoantigens according to age. Each bar shows the total number of autoantigens per patient from the three
## Related Clinical Research

- [PHF19 in AML: role in leukemic initiation, prognosis, and survival stratification](https://medichelpline.com/clinical-feed/pubmed-42913019.md) (DOI: 10.1126/sciadv.aed6757)
- [A CAR T cell–amplified suppressive B cell circuit in AML — preprint report](https://medichelpline.com/clinical-feed/biorxiv-22-a-car-t-cell-amplified-suppressive-b-cell-circuit-in-aml.md)
- [Cancer incidence trends in the Ecuadorian Amazon, 1990–2019: a registry-based analysis](https://medichelpline.com/clinical-feed/plos-one-19-cancer-incidence-in-the-ecuadorian-amazon-a-retrospective-study-1990-2019.md)
- [SEC31A drives gastric cancer growth via PI3K/AKT/mTOR and CCL5-mediated M2 macrophage recruitment](https://medichelpline.com/clinical-feed/pubmed-42885291.md) (DOI: 10.3892/or.2026.9211)
- [Pope Leo XIV Diagnosed with Lung Adenocarcinoma: Key Insights](https://medichelpline.com/clinical-feed/medical-news-today-0-pope-leo-lung-cancer-diagnosis-what-to-know-about-lung-adenocarcinoma.md)

## Navigation
- [← Back to Oncology Feed](https://medichelpline.com/clinical-feed/oncology.md)
- [← All Clinical Specialties](https://medichelpline.com/clinical-feed.md)
## Medical & Regulatory Disclaimer

> [!CAUTION]
> MedicHelpline content is structured for research, educational, and professional discovery purposes. It does not constitute individual medical advice, clinical diagnosis, or treatment recommendations.
> Always verify dosing, contraindications, and regulatory alerts against official product labeling and primary regulatory sources before clinical decision-making.