Liquid biopsy has emerged as a promising non-invasive way to diagnose tumor and monitor its progression. Different types of liquid biopsies have different advantages and limitations. In the present research, we compared the use of two types of liquid biopsy, extracellular vesicle-derived DNA (EV-DNA) and cell-free DNA (cfDNA) for identifying tumor mutations in patients with colon carcinoma. Method: DNA was extracted from the tumor tissue of 33 patients diagnosed with colon carcinoma. Targeted NGS panel, based on the hotspots panel, was used to identify tumor mutations. Pre-surgery serum and plasma were taken from the patients in which mutation was found in the tumor tissue. Extracellular vesicles were isolated from the serum followed by the extraction of EV-DNA. CfDNA was extracted from the plasma. The mutations found in the tumor were used to detect the circulating tumor DNA using ultra-deep sequencing. We compared the sensitivity of mutation detection and allele frequency obtained in EV-DNA and cfDNA. Results: The sensitivity of mutation detection in EV-DNA and cfDNA was 61.90% and 66.67%, respectively. We obtained almost identical sensitivity of mutation detection in EV-DNA and cfDNA in each of the four stages of colon carcinoma. The total DNA concentration and number mutant copies were higher in cfDNA vs. EV-DNA (p value = 0.002 and 0.003, respectively). Conclusion: Both cfDNA and EV-DNA can serve as tumor biomarkers. The use of EV-DNA did not lead to improved sensitivity or better detection of tumor DNA in the circulation.

Circulating tumor DNA (ctDNA) has been suggested as a surrogate biomarker for early detection of cancer recurrence. We aimed to explore the utility of ctDNA as a noninvasive prognostic biomarker in newly diagnosed head and neck squamous cell carcinoma (HNSCC) patients. Seventy HNSCC specimens were analysed for the detection of TP53 genetic alterations utilizing next-generation sequencing (NGS). TP53 mutations were revealed in 55 (79%). Upon detection of a significant TP53 mutation, circulating cell-free DNA was scrutinized for the presence of the tumor-specific mutation. ctDNA was identified at a minimal allele frequency of 0.08% in 21 out of 30 processed plasma samples. Detectable ctDNA correlated with regional spread (N stage ≥ 1, p = 0.011) and poorer 5-year progression-free survival (20%, 95% CI 10.9 to 28.9, p = 0.034). The high-risk worst pattern of invasion (WPOI grade 4–5) and deep invasion were frequently found in patients whose ctDNA was detected (p = 0.087 and p = 0.072, respectively). Detecting mutated TP53 ctDNA was associated with poor progression-free survival and regional metastases, indicating its potential role as a prognostic biomarker. However, ctDNA detectability in early-stage disease and the mechanisms modulating its release into the bloodstream must be further elucidated.

Introduction: Accurate assessment of mitotic activity is an important component of melanoma prognostication but remains challenging due to interobserver variability and difficulties in identifying representative mitotic hotspots. We developed and evaluated a deep learning–based mitosis detection algorithm for whole‐slide images (WSIs) of melanoma and investigated its utility for hotspot identification and spatial analysis of mitotic distribution. Methods: The model was trained on manually annotated histopathology images and applied to a cohort of 114 melanoma cases comprising 378 WSIs. Performance was assessed against expert annotation using sensitivity, precision, and F1‐score. Algorithm‐identified hotspots were reviewed by a pathologist and compared with mitotic counts reported in routine clinical practice. Spatial organization of mitotic figures was evaluated using nearest‐neighbour distance analyses. Results: The algorithm achieved a sensitivity of 88%, precision of 75%, and an F1 score of 0.81, demonstrating strong agreement with expert assessment. Across the cohort, 30,547 mitotic figures were detected, enabling comprehensive whole‐slide analysis of proliferative activity. AI‐assisted hotspot identification yielded significantly higher mitotic counts than those reported in routine pathology practice (5.35 ± 7.9 versus 2.96 ± 3.73 mitoses/mm ² , P = 0.004), suggesting improved identification of regions with maximal proliferative activity. Whole‐slide analysis further enabled characterization of mitotic spatial organization. Distinct patterns ranging from clustered to relatively uniform distributions were observed across tumours. Moreover, melanomas arising in chronically sun‐damaged (CSD) sites demonstrated significantly greater nearest‐neighbour distances than melanomas from non‐CSD sites (423.3 μm versus 285.4 μm, P = 0.023), indicating differences in the spatial organization of proliferating tumour cells. Conclusion: Our findings demonstrate that AI‐assisted mitosis detection can accurately identify mitotic figures while improving hotspot detection compared with routine assessment. Beyond mitotic quantification, large‐scale whole‐slide analysis enables novel spatial characterization of mitotic organization, providing additional insights into melanoma biology and highlighting new opportunities for computational pathology–based biomarker discovery.