Pomeranian Medical University (PMU) was established in 1948 in Szczecin, Poland. It is referred to as Pomorski Uniwersytet Medyczny in Polish.
BACKGROUND:Machine-learning (ML) driven molecular diagnostics based on omics data has a potential to revolutionize personalized medicine. However, implementation of ML into diagnostic protocols is hindered by methodological challenges which often lead to inflated performance assessment of models during development followed by poor performance of these models in implementation phase. Here, we aimed to develop and validate a pan-cancer classification framework based on DNA methylation data, that addresses methodological challenges of omics data powered ML. METHODS:We curated a primary dataset of DNA methylation profiles for 10756 samples, that included 54 healthy and cancer tissue types and validation dataset comprising data for 2306 samples from 28 independent studies. The classification framework was build using custom biomarkers selection strategy based on effect size metric that considers variance and class imbalance. The ML models were trained, tuned and evaluated using nested cross-validation approach. Local outlier factor algorithm was built into the inference pipelines to identify and filter samples displaying technical or biological anomalies. Additionally, for methodological validation of our framework we used methylation profiles for 3905 central nervous system (CNS) tumors. RESULTS:We found that relatively simple ML models outperformed complex algorithms such as deep neural network. A logistic regression classifier achieved a balanced accuracy (BACC) of 0.90 to classify 54 cancer and healthy tissue types using methylation levels at 1208 CpG sites. Similarly, our CNS tumor classifier also based on logistic regression algorithm reached a BACC of 0.94 across 59 CNS tumor subtypes. The anomaly filtering improved performance across all categories of samples tested. CONCLUSIONS:Our study demonstrates that DNA methylation profiling, when combined with carefully controlled ML practices allows for development of robust solutions that might substantially increase the efficacy of oncological diagnosis. Finally, we deployed our inference pipelines for public access via secure web platform - https://opp.pum.edu.pl/ .
Breast cancer is the most prevalent malignancy among women worldwide. A significant portion of patients possess homologous recombination deficiency (HRD), often caused by BRCA1/2 mutations, which may sensitize tumors to PARP inhibitors and platinum-based chemotherapy through synthetic lethality. Since mutations in BRCA genes have been previously suggested in association with impaired biology of telomeres, in the present study we investigated leukocyte telomere length (LTL) to evaluate its potential utility as a biomarker for BRCA1 mutations and HRD. LTL was measured using multiplex monochrome real-time qPCR in four groups: breast cancer patients with pathogenic hereditary BRCA1 mutations (n = 99), age-matched non-cancerous controls carrying the same BRCA1 mutations (n = 99), breast cancer patients with wild-type BRCA1 (n = 105), and age-matched non-cancerous controls with wild-type BRCA1 (n = 107). BRCA1 mutations were tested by the DNA sequencing approach. A significant negative correlation between age and LTL was observed across all studied groups, except in breast cancer patients carrying pathogenic hereditary BRCA1 mutations. Interestingly, after adjusting for age, BRCA1 mutation carriers had shorter LTL compared to non-carriers, regardless of the presence of cancer (P = 0.024). LTL shortening is associated with BRCA1 mutations, regardless of cancer status. Further validation studies are needed.
BACKGROUND:The pes anserinus (PA) is a tendinous complex located on the anteromedial surface of the proximal tibia, formed by the conjoined insertions of the sartorius, gracilis, and semitendinosus muscles. It contributes to internal rotation and flexion of the tibia, providing stability to the medial knee compartment. CASE REPORT:During the routine dissection of a 75-year-old female cadaver, an unusual PA variant was found. On the right limb, four accessory tendons were identified: one accessory sartorius tendon (aSART), one accessory gracilis tendon (aGRT), and two accessory semitendinosus tendons (aSTT1 and aSTT2). Distally, the aSTT1 fused with the aGRT and subsequently joined the aSART, forming a single fascial band that inserted slightly inferior to the typical PA footprint. No corresponding variations were present on the contralateral side. The coexistence of accessory slips from all three PA muscles appears exceedingly rare. CONCLUSIONS:This case report presents a unique variant of PA, emphasizing the extent of its anatomic variability. Awareness and recognizing such variants of PA is crucial for surgeons, as this knowledge may help prevent the misidentification of tendons during medial knee procedures and tendon graft harvesting.
Germline genetic testing practices for hereditary cancer vary across the European Reference Network on Genetic Tumour Risk Syndromes (ERN GENTURIS) member countries. We surveyed experts in genetic testing from 20 EU member countries and Norway to assess multi-gene panel usage, availability of genome-wide sequencing, first-tier testing approaches, implementation of polygenic risk scores, and the roles of non-genetic healthcare professionals. National experts and members of the ERN GENTURIS completed a structured questionnaire covering founder germline pathogenic variants (gPV) testing, panel testing for common genetic tumour risk syndromes, use of whole-exome sequencing (WES) and whole-genome sequencing, polygenic risk score implementation, use of formalin-fixed paraffin-embedded tumour samples, laboratory accreditation, and the clinical roles of physicians, genetic counselors and nurses. Significant inter-country heterogeneity was observed. Most countries rely on next-generation sequencing (NGS) multi-gene panels. Founder gPV testing is first-line in a few high-prevalence populations (e.g., BRCA1/2 founders). All 21 countries offer NGS panel tests for hereditary breast and ovarian cancer, and ≥19 countries do so for colorectal and prostate cancers. However, NGS panel size and gene composition exhibit substantial variability. WES is available in 12 countries on a routine basis. Most countries implemented genetic testing on stored tumour tissue from deceased patients. In all countries, clinical geneticists can order germline genetic tests, and in 9 countries, any physician can do so. These findings show differences in accessibility to germline genetic testing of hereditary cancer in Europe. We propose EU-wide guidance via pathways, standards of care, and sharing of best practices to further optimize access to hereditary cancer genetic molecular diagnostics.
BACKGROUND:Virological failure (VF) with long-acting injectable cabotegravir and rilpivirine (CAB/RPV-LA) is uncommon but often associated with selection of resistance, potentially limiting future treatment options. Registration trials associated VF risk with baseline RPV resistance, A1/A6 subtypes, and body mass index (BMI) >30 kg/m2, but these factors have rarely been analyzed in other clinical settings. We summarize the first ∼100 reported VF cases, focusing on subtypes, drug levels, and resistance patterns. METHODS:Published data on CAB/RPV-LA through July 2025 were analyzed for risk factors. Resistance mutations were interpreted using the Stanford HIVdb database. RESULTS:After excluding duplicates, 94 VF cases were analyzed. Only 4.4% met the high-risk threshold of ≥2 risk factors. Subtype A lineages were reported in 26.4%, preexisting RPV mutations in 14.7%, and BMI >30 kg/m2 in 36.9%. At failure, low CAB or RPV levels were observed in 29% but did not differ from treatment successes. Predicted reduced susceptibility to CAB or RPV was observed in 87.2% (56% for both), with CAB resistance mutation N155H more frequently observed among subtype A lineages. Predicted susceptibility to dolutegravir/bictegravir (44.3%), doravirine (39.7%), or etravirine (35.9%) was common, but high-level resistance was rare. CONCLUSIONS:Emergent resistance in VF cases often resulted in cross-resistance to other nonnucleoside reverse transcriptase inhibitors and integrase strand transfer inhibitors. Although most cases did not meet the high-risk profile as defined by registration trials, subtype A lineages were overrepresented. Low drug levels were not elevated versus treatment successes. These data suggest that subtype-specific factors beyond A6 may influence VF risk and merit further study.
