Editors’ Selections from Relevant Scientific Publications

Leading modifiable cancer risk factors (icons from PowerPoint, world map from Wikimedia Commons)Nearly 40% of cancer cases worldwide are attributable to modifiable risk factors. Using GLOBOCAN data across 36 cancer sites in 185 countries, Fink and colleagues estimated the 2022 global cancer burden attributable to 30 modifiable risk factors, applying a 10-year exposure−cancer latency by using prevalence data from around 2012. They found that these risk factors accounted for approximately 37.8% of new cancer cases - representing 7.1 million diagnoses - including 29.7% of cases in women and 45.4% in men. Smoking, infections, and alcohol consumption emerged as the leading contributors, and lung, stomach, and cervical cancers together comprised about half of all preventable cases. The analysis also revealed substantial regional variations: among men, the highest preventable burden occurred in East Asia (57.2%), while among women, it was in sub-Saharan Africa (38.2%). Behavioral risk factors were more prevalent in high-income regions, whereas infections like HPV and H. pylori disproportionately affected populations in low-income regions. These findings highlight the urgent need for targeted, gender-sensitive, and context-specific prevention policies tailored to local risk factor patterns.Fink H, … Soerjomataram I. Nat Med. 2026 Feb 3.Prostate cancer screening (by Darryl Leja from National Human Genome Research Institute via Wikimedia Commons)Genomics-informed risk prediction tools may benefit precision cancer screening. Using data from 585,418 participants in the Million Veteran Program, Vassy and colleagues developed P-CARE, a prostate cancer risk prediction model that integrates polygenic score, family history and genetic ancestry. This model was externally validated in four cohorts from the PRACTICAL Consortium, and accurately predicted the risk of overall, metastatic, and fatal prostate cancer across diverse racial and ethnic groups. The authors also developed and validated a cost-efficient Blended Genome-Exome assay capable of detecting rare prostate cancer-associated variants, enabling the implementation of the P-CARE model in a randomized clinical trial of precision prostate cancer screening (the ProGRESS study). Together, these advances demonstrate the potential of a genomics-enabled learning health system to improve personalized, risk-adapted cancer screening and prevention.Vassy JL, … Seibert TM. Nat Cancer. Feb;7(2):352–367.Plasma biomarkers for early detection of PDAC (created using Gemini)A four-biomarker plasma panel may aid in the early detection of pancreatic ductal adenocarcinoma (PDAC). Using two mass spectrometry methods and plasma pools from University of Pennsylvania and the Mayo Clinic, Krusen and colleagues discovered two biomarkers, aminopeptidase N (ANPEP) and polymeric immunoglobulin receptor (PIGR), that were elevated in stage I/II PDAC compared with healthy or nonmalignant disease controls. In subsequent retrospective phase II studies using independent plasma samples, the researchers demonstrated that combining ANPEP and PIGR with thrombospondin 2 (THBS2) and carbohydrate antigen 19-9 (CA19-9), two previously identified markers, yielded an area under the receiver-operating characteristic curve of 0.87 for distinguishing stage I/II PDAC from nonmalignant controls. The four-marker panel (ANPEP, PIGR, THBS2 and CA19-9) achieved 87.5% sensitivity at 95% specificity for detecting stage I/II PDAC, suggesting potential utility for early pancreatic cancer detection pending validation in prediagnostic cohorts.Krusen BM, … Zaret KS. Clin Cancer Res. 2026 Feb 17;32(4):756–769.Imaging surveillance vs. mastectomy (created using Gemini)Survival outcomes were not compromised in women carrying pathogenic BRCA1 or BRCA2 variants (pvBRCA1/2) who chose surveillance instead of bilateral risk-reducing mastectomy (BRRM). In a prospective UK cohort study, Gandhi and colleagues determined overall survival (OS), breast cancer–specific death, and breast cancer incidence rates in pvBRCA1/2 carriers who elected to undergo either BRRM (n = 460, 38%) or imaging surveillance (n = 745, 62%). Over 4,627 woman-years of follow-up, the researchers found no significant differences in breast cancer–specific deaths or OS between the two groups, although they observed a 94% reduction in annual incidence rate of breast cancer in the BRRM group (0.15%) compared with the surveillance group (2.4%). Although longer follow-up is needed to confirm the survival findings, the results may reassure women who opt for surveillance due to concerns about surgery complications or effects on physical and sexual well-being, providing data to support individualized decision making.Gandhi A, … Evans DG. J Clin Oncol. 2026 Feb 4:JCO2500834.