RESUMO
Fecal immunochemical test (FIT) followed by colonoscopy in positive cases is commonly used for population-based colorectal cancer (CRC) screening. However, specificity of FIT for CRC is not ideal, and has poor performance for advanced adenoma detection. Fecal Fusobacterium nucleatum (Fn) detection has been proposed as a potential non-invasive biomarker for CRC and advanced adenoma detection. We aimed to evaluate the diagnostic performance of Fn detection using droplet digital PCR (ddPCR) in FIT samples from individuals enrolled in a CRC screening program with colorectal adenoma or cancer. We evaluated Fn presence in DNA isolated from FIT leftover material of 300 participants in a CRC Screening Program using ddPCR. The Fn DNA amount was classified as Fn-low/negative and Fn-high, and the association with patients clinicopathological features and accuracy measurements was calculated. Fn high levels were more prevalent in FIT-positive (47.2%n=34 of72) than FIT-negative samples (28.9%, n=66 of 228) (p<0.04). Among FIT-positive samples, high Fn levels were significantly more frequent in cancer patients (CA, n=8) when compared to normal (NT, n=16) (p=0.02), non-advanced adenomas (NAA, n=36) (p=0.01), and advanced adenomas (AA, n=12) (p=0.01). Performance analysis of Fn in FIT-positive samples for colorectal cancer detection yielded an AUC of 0.8203 (CI: 0.6464-0.9942), with high sensitivity (100%) and specificity of 50%%. Concluding, we showed the feasibility of detecting Fn in FIT leftovers using the ultrasensitive ddPCR technique. Furthermore, we highlighted the potential use of Fn levels in fecal samples to ameliorate CRC detection.
RESUMO
Medulloblastoma (MB) is the most prevalent malignant brain tumor in children, known for its heterogeneity and treatment-associated toxicity, and there is a critical need for new therapeutic targets. We analyzed the somatic mutation profile of 15 driver genes in 69 Latin-Iberian molecularly characterized medulloblastomas using the Illumina TruSight Tumor 15 panel. We classified the variants based on their clinical impact and oncogenicity. Among the patients, 66.7% were MBSHH, 13.0% MBWNT, 7.3% MBGrp3, and 13.0% MBGrp4. Among the 63 variants found, 54% were classified as Tier I/II and 31.7% as oncogenic/likely oncogenic. We observed 33.3% of cases harboring at least one mutation. TP53 (23.2%, 16/69) was the most mutated gene, followed by PIK3CA (5.8%, 4/69), KIT (4.3%, 3/69), PDGFRA (2.9%, 2/69), EGFR (1.4%, 1/69), ERBB2 (1.4%, 1/69), and NRAS (1.4%, 1/69). Approximately 41% of MBSHH tumors exhibited mutations, TP53 (32.6%) being the most frequently mutated gene. Tier I/II and oncogenic/likely oncogenic TP53 variants were associated with relapse, progression, and lower survival rates. Potentially actionable variants in the PIK3CA and KIT genes were identified. Latin-Iberian medulloblastomas, particularly the MBSHH, exhibit higher mutation frequencies than other populations. We corroborate the TP53 mutation status as an important prognostic factor, while PIK3CA and KIT are potential therapeutic targets.