Search

Background: The impacts of long-term storage and varying preanalytical factors on the quality and quantity of DNA and miRNA from archived serum have not been fully assessed. Preanalytical and analytical variations and degradation may introduce bias in representation of DNA and miRNA and may result in loss or corruption of quantitative data.Methods: We have evaluated DNA and miRNA quantity, quality, and variability in samples stored up to 40 years using one of the oldest prospective serum collections in the world, the Janus Serumbank, a biorepository dedicated to cancer research.Results: miRNAs are present and stable in archived serum samples frozen at −25°C for at least 40 years. Long-time storage did not reduce miRNA yields; however, varying preanalytical conditions had a significant effect and should be taken into consideration during project design. Of note, 500 μL serum yielded sufficient miRNA for qPCR and small RNA sequencing and on average 650 unique miRNAs were detected in samples from presumably healthy donors. Of note, 500 μL serum yielded sufficient DNA for whole-genome sequencing and subsequent SNP calling, giving a uniform representation of the genomes.Conclusions: DNA and miRNA are stable during long-term storage, making large prospectively collected serum repositories an invaluable source for miRNA and DNA biomarker discovery.Impact: Large-scale biomarker studies with long follow-up time are possible utilizing biorepositories with archived serum and state-of-the-art technology. Cancer Epidemiol Biomarkers Prev; 24(9); 1381–7. ©2015 AACR.

Supplementary Tables 1-7 Supplementary Table 1. PCR primers and conditions Supplementary Table 2. List of included diagnoses in each histological subgroup Supplementary Table 3. Associations between published genetic risk variants and risk of disease, calculated by conditional logistic regression; Glioma cases and matched controls Supplementary Table 4. Associations between published genetic risk variants and risk of disease, calculated by conditional logistic regression; GBM cases and matched controls Supplementary Table 5. Associations between published genetic risk variants and risk of disease, calculated by conditional logistic regression; Oligodendroglioma cases and matched controls Supplementary Table 6. Associations between published genetic risk variants and risk of disease, calculated by conditional logistic regression; Astrocytoma cases and matched controls Supplementary Table 7. Associations between published genetic risk variants and risk of disease, calculated by conditional logistic regression; Ependymoma cases and matched controls

Background: Although glioma etiology is poorly understood in general, growing evidence indicates a genetic component. Four large genome-wide association studies (GWAS) have linked common genetic variants with an increased glioma risk. However, to date, these studies are based largely on a case–control design, where cases have been recruited at the time of or after diagnosis. They may therefore suffer from a degree of survival bias, introduced when rapidly fatal cases are not included.Methods: To confirm glioma risk variants in a prospective setting, we have analyzed 11 previously identified risk variants in a set of prediagnostic serum samples with 598 cases and 595 matched controls. Serum samples were acquired from The Janus Serum Bank, a Norwegian population-based biobank reserved for cancer research.Results: We confirmed the association with glioma risk for variants within five genomic regions: 8q24.21 (CCDC26), 9p21.3 (CDKN2B-AS1), 11q23.3 (PHLDB1), 17p13.1 (TP53), and 20q13.33 (RTEL1). However, previously identified risk variants within the 7p11.2 (EGFR) region were not confirmed by this study.Conclusions: Our results indicate that the risk variants that were confirmed by this study are truly associated with glioma risk and may, consequently, affect gliomagenesis. Though the lack of positive confirmation of EGFR risk variants may be attributable to relatively limited statistical power, it nevertheless raises the question whether they truly are risk variants or markers for glioma prognosis.Impact: Our findings indicate the need for further studies to clarify the role of glioma risk loci with respect to prolonged survival versus etiology. Cancer Epidemiol Biomarkers Prev; 24(5); 810–6. ©2015 AACR.

Figure s1: Bioanalyzer trace, long RNA vs miRNA Figure s2: miRNA sequencing saturation curves Figure s3: Small RNA detected in Janus samples Table s1: Adjusted p-values for mean DNA yield between treatments. Significant values in bold. Figure s4: DNA yield by serum treatments Figure s5: Example of DNA Bioanalyzer trace Figure s6: DNA sequencing results Figure s7: Example of sequence distribution

Background: Although glioma etiology is poorly understood in general, growing evidence indicates a genetic component. Four large genome-wide association studies (GWAS) have linked common genetic variants with an increased glioma risk. However, to date, these studies are based largely on a case–control design, where cases have been recruited at the time of or after diagnosis. They may therefore suffer from a degree of survival bias, introduced when rapidly fatal cases are not included. Methods: To confirm glioma risk variants in a prospective setting, we have analyzed 11 previously identified risk variants in a set of prediagnostic serum samples with 598 cases and 595 matched controls. Serum samples were acquired from The Janus Serum Bank, a Norwegian population-based biobank reserved for cancer research. Results: We confirmed the association with glioma risk for variants within five genomic regions: 8q24.21 (CCDC26), 9p21.3 (CDKN2B-AS1), 11q23.3 (PHLDB1), 17p13.1 (TP53), and 20q13.33 (RTEL1). However, previously identified risk variants within the 7p11.2 (EGFR) region were not confirmed by this study. Conclusions: Our results indicate that the risk variants that were confirmed by this study are truly associated with glioma risk and may, consequently, affect gliomagenesis. Though the lack of positive confirmation of EGFR risk variants may be attributable to relatively limited statistical power, it nevertheless raises the question whether they truly are risk variants or markers for glioma prognosis. Impact: Our findings indicate the need for further studies to clarify the role of glioma risk loci with respect to prolonged survival versus etiology. Cancer Epidemiol Biomarkers Prev; 24(5); 810–6. ©2015 AACR.

The purpose of the study was to identify and describe cervical cancer specific changes in humoral immune response by using a novel technology platform Mimotope Variation Analysis (MVA) developed by Protobios. The Janus Serum Bank is one of the world's oldest and largest population-based research biobanks established in 1973. The cohort is annually linked to the Cancer Registry of Norway using personal identification numbers. More than 1000 women have developed cervical cancer after donating sera. We identified sets of successive sera samples from ten patients with invasive cervical cancer, ten patients with pre-invasive cervical neoplasia and twenty cancer-free individuals (matched for age, gender, length of sample storage +3 months). For each subject we identified 4-10 samples donated to Janus over a period of up to 18 years, in total 213 samples. We applied MVA which combines phage display technology and high-throughput sequencing analysis to generate quantitative serologic profiles of millions of 12-mer peptide antigens called mimotopes from 2 μl of blood serum per analysis. Hierarchical clustering analysis of the top 5000 mimotopes from each sample resulted in individual-specific immunoprofiles. Multiple samples from the same person clustered together. Moreover, clustering reflected the time the sera was drawn suggesting that part of the individual immunoprofile is stabile over time. In conclusion, preliminary results suggests that Mimotope Variance Analysis generates individual-specific immunological profiles. This profile most likely reflect prior exposure environmental pathogens. Further, we aim to decode the differences in the immunological profiles between cancer patients and cancer-free subjects. Citation Format: Espen Enerly, Mari Nygård, Madleen Orumaa, Arno Pihlak, Susan Pihelgas, Hilde Langseth, Toomas Neuman, Kaia Palm. Mimotope variance analysis: A novel immunoprofiling method to monitor progression of cervical cancer. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4039.

Background: Lung cancer in never smokers is a significant contributor of cancer mortality. As the incidence of lung cancer in never smokers is increasing, there is a need to investigate risk factors for lung cancer in this population. There is data to support that polyomaviruses are potentially carcinogenic in the human lung. We explored the role of polyomaviruses in lung cancer development in never smokers using a multiplex assay to detect serum antibodies to viral capsid proteins. Methods: We conducted a nested case-control study of never-smoking cases of lung cancer identified from four established prospective cohorts- NYU Women's Health Study (NYU-WHS), Janus Serum Bank, Shanghai Women's Health Study (SWHS) and Singapore Chinese Health Study (SCHS). Controls were matched to cases on gender, never-smoking status, age and calendar period of entry. Serological analysis was performed using a 100 μL pre-diagnostic serum sample at the German Cancer Research Center (Deutsches Krebsforschungszentrum, DKFZ) in Heidelberg, Germany using fluorescent bead-based multiplex serology and included antibodies to viral capsid protein-1 and T-antigens of 9 human polyomaviruses: JC virus, BK virus, KI virus, WU virus, Trichodysplasia Spinulosa-associated Polyoma Virus, Merkel Cell Polyoma Virus, Human Polyoma Virus 6, Human Polyoma Virus 7 and Human Polyoma Virus 10. Results: The final analysis included 511 cases and 508 controls. Mean age of participants was 56.1±11.0 years. SWHS and SCHS included only Asian participants. NYU-WHS and SWHS had only female participants. Nearly 85% of the participants in our pooled analysis were females and 74% were Asians. Seropositivity for each polyomavirus showed significant heterogeneity by study but overall there were no statistical significant differences between cases and controls for any of the polyomaviruses. 72.0% of the cases and 71.5% of the controls were seropositive for JC virus antibody. Seropositivity for BK virus was higher at 89.0% in cases and 89.8% in controls. We did not find any difference in seropositivity between cases and controls in our stratified analysis based on gender, histology or time interval from sample collection to cancer diagnosis. We also performed sensitivity analyses and found the seroprevalence data to be very robust to alterations in the MFI cutpoints. Conclusions: Our study is the largest epidemiological study in never smokers to investigate the role of polyomaviruses in lung cancer development. We did not find a significant difference in serological measurements of antibodies against each of the polyomaviruses between the cases and controls. Future research to evaluate the relationship between polyomaviruses and lung carcinogenesis should focus on evaluating viral replication in tumor in combination with serological markers of infection especially as antibody reactivities can vary considerably across different populations and geographical areas as demonstrated by our study. Citation Format: Jyoti Malhotra, Tim Waterboer, Michael Pawlita, Angelika Michel, Qiuyin Cai, Wei Zheng, Qing Lan, Nathaniel Rothman, Hilde Langseth, Tom K. Grimsrud, Jian-Min Yuan, Woon-Puay Koh, Alan A. Arslan, Anne Zeleniuch-Jacquotte, Paolo Boffetta. Serum biomarkers of polyomavirus infection and risk of lung cancer in never smokers. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 4290.

Introduction: Human papillomavirus type 16 (HPV16) infection is associated with an increasing fraction of oropharyngeal cancer. The aim of our study was to determine the association of HPV16 antibodies (Abs) and oropharyngeal cancer (OPC) risk in sera obtained prior to clinical diagnosis. Methods: We identified 92 participants with incident oropharyngeal cancers and 460 age-, gender-, region- and year- matched controls from the Janus Serumbank in Norway. Serum samples were collected from cases, on average, 9.2 years before diagnosis (range, -0.6-14.9 years). 11 cases had serum samples from multiple time points. IgG Abs to the HPV16 antigens E1, E4, E5, E6, E7, L1, L2, and the N-terminal and C-terminal fragments of E2 (NE2, CE2) were quantified in blinded fashion using a custom RAPID ELISA assay. Full-length genes were expressed as GST fusion proteins with a mammalian protein expression system, incubated with sera, and bound IgG was measured. The ratio of relative luminescence units (RLU) values for each antigen to control GST protein was determined. Cutoff values were pre-established as 3 SD>median of 247 healthy controls. A logistic regression classifier of a 7-Ab biomarker panel (E1, NE2, CE2, E4, E5, E6, E7) for the predictive diagnosis of OPC was trained on an independent sample set, and applied to these samples. P-values were calculated by Fisher's exact test. Odds ratios (ORs) of cancer and 95% CIs were calculated. Results: HPV16 early antigen seropositivity was present in 13% of patients with OPC and 1% of controls (OR, 17.0; 95% CI, 5.0-74.0, p10 years, 5.4). Abs to CE2 were strongly associated with cases 0-2 years pre-diagnosis (OR, 112.0; 95% CI, 15.0-971.2, p Conclusion: HPV16 Abs to multiple early viral antigens are detectable years prior to diagnosis of oropharyngeal cancer, and the probability of seropositivity increases closer to diagnosis. A model for screening would help identify individuals at risk for whom interventions and surveillance may be evaluated. Citation Format: Karen S. Anderson, Hilde Langseth, Garrick Wallstrom, Marshall Posner, Julia N. Cheng, Rizwan Alam, Diego Chowell, Jon Mork. Association of a serum HPV16 IgG signature and risk of oropharyngeal cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr 1566. doi:10.1158/1538-7445.AM2015-1566

Introduction miRNAs are a distinct class of small non-coding RNA molecules that regulate the expression of protein coding genes. They are present and stable in the cell-free body fluids such as plasma, serum and urine. The Janus Serum Bank contains 317 000 serum samples and more than 61 000 incident cancer cases diagnosed during follow-up. It is well-suited for studies of miRNAs as early cancer biomarkers since samples are collected pre-diagnostically. Aim To establish a high-throughput sequencing approach for obtaining high quality miRNA expression profiles from archived Janus serum samples stored for up to 41 years. Further aims were to assess the quantity and quality of small RNAs in the samples, and establish guidelines for the optimal sequencing depth. Methods Total RNA was extracted from eight pilot samples using Trizol and miRNeasy (Qiagen) columns. Sequencing libraries were made using ScriptMiner (EpiCentre) and the samples were sequenced on one HiSeq (Illumina) lane. After adaptor removal the sequences were aligned unguided to the genome. Reads mapping to mirBASE and Ensembl smallRNA features were counted and sequencing saturation curves were calculated. DESeq was used for differential expression analyses. Results High-throughput sequencing results from the Janus samples, with storage times ranging from 13 to 39 years, show that miRNAs are preserved in the samples. High quality miRNA expression profiles, including on average 641 miRNAs per sample, were obtained from our pilot samples. Most of the uniquely mapped reads, on average 84%, are miRNA. We can also identify protein coding RNAs and other groups of small RNA, on average respectively 9% and 6% of the uniquely mapped reads. The preliminary results from sequencing saturation curves, tracking miRNA discovery in each sample as a measure of input reads, have identified a guideline for optimal sequencing depth. Conclusion and future perspectives Here we have shown that the stability, quantity and quantity of long-term stored sera from the Janus Serum Bank are sufficient for obtaining high-quality miRNA expression profiles. High-throughput sequencing approach will be used in the search for miRNA expression profiles as early cancer biomarkers for early diagnosis, improved therapy and surveillance of cancer. Citation Format: Trine B. Rounge, Robert Lyle, Marianne Lauritzen, Kristina Kjaerheim, Tom Grotmol, Tom K. Grimsrud, Randi E. Gislefoss, Steinar Tretli, Giske Ursin, Hilde Langseth. Optimizing small RNAseq of archived serum samples from the Janus Serum Bank - a search for cancer biomarkers. [abstract]. In: Proceedings of the 105th Annual Meeting of the American Association for Cancer Research; 2014 Apr 5-9; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2014;74(19 Suppl):Abstract nr 1501. doi:10.1158/1538-7445.AM2014-1501

Background: In recent years there has been an increased interest in utilizing biorepositories in cancer research. More specifically to detect novel cancer biomarkers for early detection, treatment and prognosis, and to investigate etiological factors. This demands large series of cases including high quality biospecimens and their associated data. The present study communicate the potentials in combining information from a high quality cancer registry with analytical information from the population-based Janus biobank of Norway. It was established in 1973 with the intention to collect premorbid serum for future use in cancer research. The philosophy is to make valuable biological samples and data available to researchers worldwide. Studies of cancer etiology and early cancer detection typically measure biochemical and immunological changes several years before cancer diagnosis. Most human cancer studies involve data from patients after the tumor has been diagnosed. The opportunity to detect important factors in the pathogenesis or etiology of cancer may then be lost. Use of the Janus biobank facilitates evaluation of the preclinical stage, the latent period before the cancer has been diagnosed. For a small sub-group of the Janus donors both pre-diagnostic and diagnostic samples from the same individuals are available. Cancer incidence and scientific results: Today the Janus biobank is one of the world's largest and oldest population-based biobanks. The repository consist of prediagnostic serum samples from 317 000 individuals of whom 52 464 have developed cancer at a later point in life. In male donors prostate cancer is the most common cancer (n=7468), followed by cancer of the lung (n=3222), colon (n=2205), bladder (n=2040) and malignant melanoma (n=1663). Breast cancer is the most common cancer in female donors (n=6774) followed by genital organs (cervix uteri, corpus uteri and ovaries combined n=4176), colon (n=1707), lung (n=1650) and malignant melanoma (n=1423). A considerable number of rare cancer cases have developed among Janus donors, for instance 179 cases of liver cancer and 660 multiple myeloma cases, giving opportunity to set up valid studies of cancers with poor prognosis, and where the knowledge of etiology is limited. Predictions of the future cancer burden indicates that the aging Janus cohort will generate a considerable number of new cancer cases in the years to come. The biobank has been used in a large number of studies investigating biomarkers of susceptibility, exposure and effect. Important contributing results for cancer prevention were early observations of the association between PSA, CA125, HPV, Helicobacter pylori and risk of prostate, ovarian, cervical and stomach cancer respectively. Furthermore, important quality and stability studies have shown that the sample quality in the Janus repository is good for a number of components after long-term storage at −25°. Future potentials: The Janus biobank comprises samples stored over a long period of time and present the opportunity to investigate accumulated, prospectively occurred cancer cases. The samples accessibility to a large number of healthy controls make it very timely to set up large studies with adequate statistical power to demonstrate contrasts in exposure between cancer cases and healthy controls, and to address the long-term effect of past exposure profiles, individual susceptibility and time-lag effects. An important future perspective is to utilize modern technology to enable the use of the Janus biobank in gene-environment-interaction studies, microRNA profiling, and omics-based studies. Statement of conclusion: The unique arrangement of the integration between the Janus biobank and the Cancer Registry of Norway, and the comprehension of this type of resources in the scientific milieus, makes the use of the biobank in cancer prevention and care very timely. Citation Information: Cancer Prev Res 2011;4(10 Suppl):A7.