This work reports the generation and validation of a new genome-wide genetic data resource for the Australian 45 and Up Study, one of the largest population-based cohorts in the Southern Hemisphere. The cohort, recruited 2005–2009, contains extensive longitudinal and linked health data. The principal aims were to produce genome-wide data for a large sub-sample of participants, validate the genomic resource, characterise participant and ancestry composition, and demonstrate integration with existing linked health records through polygenic risk score analyses for common cancers.
Between 2022 and 2023, a total of 30,541 participants were invited to provide a DNA sample. Invitations targeted a random sub-cohort and all cohort members with a recorded history of the four most common invasive cancers in the cohort: breast, prostate, melanoma and colorectal cancer. The study team performed in-depth mapping of characteristics for participants who consented to provide a DNA sample versus those who did not, to document potential selection differences. The source reports these mapping efforts but does not provide detailed tabulations in the abstract.
The genetic resource was generated using low-coverage whole-genome sequencing with coverage ranging from 0.4× to 6.3× across samples, followed by imputation. After these steps, roughly 79 million variants were called across the sequenced individuals. Initial variant calls covered 7,408 individuals prior to quality control filtering.
Following stringent quality control procedures, data for 6,827 individuals remained for analysis. Validation included concordance checks: genotype concordance was high between duplicate samples (n=85) and between the low-coverage sequence-derived genotypes and dense genotyping arrays (n=188). These validation comparisons support the reliability of the low-coverage sequencing plus imputation approach used to produce the genome-wide dataset.
Ancestry inference indicated that most unrelated participants with high-quality data have inferred European genetic ancestry: 6,631 individuals (98%). A smaller subset, 141 individuals (2%), were inferred to have non-European or admixed ancestry. The source provides these counts but does not elaborate further on ancestral subgroups or the methods used for ancestry inference in the abstract.
To demonstrate proof of principle for integration of genomic data with the cohort's extensive linked health records, the authors calculated polygenic risk scores (PGS) for the four most common cancers represented in the invited sample (breast, prostate, melanoma, colorectal). These analyses leveraged the newly generated genomic data together with the established linked clinical and registry data available for the 45 and Up Study.
The PGS analyses produced predictive performance broadly consistent with prior studies. Reported example results include an association between higher prostate cancer PGS and earlier age at prostate cancer diagnosis. The authors also report no meaningful differences in cancer stage at diagnosis by PGS. The abstract does not provide detailed effect sizes, confidence intervals, or model specifications; the statement indicates concordance with established PGS performance but specific numeric results were not reported in the abstract.
Study data are available for approved research projects through the established data custodians: the Sax Institute for the 45 and Up Study and genomic data, and the NSW Ministry of Health for NSW Cancer Registry and NSW Registry of Births, Deaths and Marriages. Direct provision of participant-level data by the authors is not permitted by custodianship rules because of confidentiality and privacy obligations. Data access enquiries can be made to the Sax Institute by email (45andUp.research@saxinstitute.org.au) or via the Sax Institute website for guidance on using the 45 and Up Study. The authors state that, upon publication, all newly generated genomic data will also be deposited to the European Genome-phenome Archive (EGA), subject to appropriate access controls.
Ethical approvals for the work were obtained from the Human Research Ethics Committee of the University of New South Wales (HC210602) and the New South Wales Population and Health Services Research Ethics Committee (2019/ETH01746; HREC/14/CIPHS/54). The authors declared no competing interests and confirmed that participant consent procedures and relevant reporting guidelines were followed. Funding sources declared include the Medical Research Future Fund (Genomic Health Futures Mission), the National Health and Medical Research Council (NHMRC), Cancer Institute NSW, and Cancer Council NSW.
The addition of high-quality genome-wide data for 6,827 participants substantially enhances the research value of the 45 and Up Study by enabling genetic epidemiology, disease aetiology research, risk stratification and precision health investigations within an Australian context. The resource's validation steps and linkage to deep phenotypic and registry data make it a potentially valuable asset for national and international research, subject to custodial access procedures. The abstract does not provide exhaustive methodological detail or full analytic results; readers should consult the full preprint and supplementary material for comprehensive methods, tables and effect estimates.