Editors' Selections from Relevant Scientific Publications
Bibliographic record
Abstract
Cell-free DNA test (by Köhnke et al via Cancer Discovery)Bruhm et al. conducted a genome-wide somatic mutation analysis of cancer tissues to identify tumor-specific somatic mutations in cell-free DNA fragments for early cancer detection. The authors analyzed genome sequencing data from 2,511 individuals across 25 cancers in the Pan-Cancer Analysis of Whole Genomes study and developed the GEnome-wide Mutational Incidence for Non-Invasive detection of Cancer (GEMINI) approach. GEMINI was then applied to 489 individuals from four prospective patient cohorts. When combined with DELFI fragmentation analyses and low-dose CT imaging, GEMINI detected >90% of patients with lung cancer, including those with stage I and II disease. The GEMINI scores in patients with liver cancer were also higher than those in patients with cirrhosis. The results suggest that cancer can be detected non-invasively through single-molecule mutation profiles to facilitate cancer screening and monitoring.Bruhm DC… Velculescu VE. Nat Genet. 2023 Aug;55(8):1301–1310.Skin cancer (by Mittapalli et al via Cancer Research)People living in the United Kingdom accumulate four times as many mutations—including many associated with keratinocyte cancer—in their facial skin as do people in Singapore. Singapore receives up to three times as much UV radiation as the UK but has only one-seventeenth the rate of keratinocyte carcinomas, suggesting that some factors protect Singaporean skin from harmful radiation. To study this, King et al. assayed eyelid epidermis from five people in Singapore and six in the UK who were in their 60s, on average, and did not have skin cancer. The UK samples had far more mutations, including copy number variations, mutations caused by UV radiation, and p53 mutations associated with cancer. In contrast, samples from Singapore had more NOTCH mutations. These results suggest that aging skin in a high-incidence country displays cancer-associated features not observed in a low-risk country, possibly reflecting variations in germline UV-protective genes.King C, … Jones PH. Nat Genet. 2023 Aug 3. doi: 10.1038/s41588-023-01468-x. Epub ahead of print. Erratum in: Nat Genet. 2023 Sep;55(9):1440–1447.Breast cancer (by Lynch and Jaffe from West Coast Surgical Oncology)Variants in genes including BRCA1 and BRCA2 can serve as red flags regarding the danger of developing breast cancer, but the genes identified to date account for less than half of the hereditary risk for such tumors. Wilcox et al. set out to fill in the blanks by using patient data (26,368 cases and 217,673 controls) from three major whole-exome sequencing efforts to identify protein-truncating and missense variants that additionally contribute to breast cancer risk. Their findings largely underlined the outsized impact of known genetic factors, but also spotlighted modest but statistically significant associations with mutations in additional tumor suppressor genes including MAP3K1, CDKN2A, and SAMHD1. These newly-identified variants may meaningfully heighten breast cancer risk, but the authors conclude that most of the unidentified hereditary factors likely reside within noncoding genomic regions.Wilcox N, … Simard J. Nat Genet. 2023 Sep;55(9):1435–1439.Blood cancer (by Ernesto del Aguila III, NHGRI via Flickr)The identification of robust risk signatures for the various myeloid neoplasms could enable early prevention for a category of blood cancers that currently prove fatal for most patients. To this end, Gu et al. analyzed data from 454,340 participants in the UK Biobank study, and identified subsets of genomic mutations associated with risk of developing acute myeloid leukemia, myelodysplastic syndrome, or myeloproliferative neoplasm. By combining these with blood testing-derived parameters such as hemoglobin concentration, the authors derived risk profiles that could predict the risk of developing these three subtypes of myeloid neoplasm within the next year, achieving an area under curve of roughly 90%. After validating their models in two additional cohorts, the researchers developed a web-based tool called MN-predict that allows clinicians and researchers to further test and implement this predictive strategy.Gu M, … Vassiliou GS. Nat Genet. 2023 Sep;55(9):1523–1530.Breast cancer screening via mammogram (by Navy Medicine via Flickr)Breast cancer screening for women between the ages of 40 and 49 raises their survival rate without increasing the rate of diagnosis. Using registry data from more than 21,000 women diagnosed with breast cancer in Canada, Wilkinson et al compared the survival rates of women in Canadian provinces that include 40–49 year-olds in routine screening programs to those of provinces that begin screening at age 50. The ten-year survival rate showed a 1.9 percentage point increase in women screened in their 40s, and a 2.6 percent point increase among those aged 45–49. The diagnosis rate, however, did not differ depending on the screening program, suggesting that overdiagnosis is not a problem in this population. The higher survival rate should inform screening guidelines, potentially allowing women in their 40s to begin treatment sooner.Wilkinson AN, …Seely JM. J Clin Oncol. 2023 Oct 10;41(29):4669–4677.Medical exam reminder (by Jernej Furman via Flickr)A study by Atlas et al. investigated the impact of primary care intervention and patient outreach with patient navigation to improve follow-up of abnormal cancer screening test results. The researchers developed and evaluated a multilevel intervention care model that included a comprehensive health informatics platform and a team including primary care clinicians and specialists. This study involved 11,980 patients in 44 primary care practices who required follow-up for abnormal breast, cervical, colorectal, or lung cancer screening test results within 120 days of enrollment. The rate completion of follow-up was higher among patients exposed to electronic health record (EHR) reminders, outreach, and navigation (31.4%); or EHR reminders and outreach (31.0%), than those exposed to EHR reminders only (22.7%) or usual care (22.9%). This study indicates that systems-based outreach in primary care settings can improve timely follow-up of abnormal cancer screening results.Atlas SJ, … Haas JS. JAMA. 2023 Oct 10;330(14):1348–1358.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame distilled prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. Learned from the 10,348 direct Codex labels and 10,348 direct Gemma labels. Candidate is the union of thresholded teacher heads; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels or direct frontier model labels.
Codex and Gemma teacher scores by category
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.001 | 0.000 |
| Meta-epidemiology (narrow) | 0.000 | 0.000 |
| Meta-epidemiology (broad) | 0.000 | 0.000 |
| Bibliometrics | 0.000 | 0.001 |
| Science and technology studies | 0.001 | 0.000 |
| Scholarly communication | 0.000 | 0.000 |
| Open science | 0.000 | 0.000 |
| Research integrity | 0.000 | 0.000 |
| Insufficient payload (model declined to judge) | 0.000 | 0.000 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one teacher head, not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".