Background: Machine learning models learn about general behavior from data by finding the relationships between features. Our purpose was to develop a predictive model to identify and predict which subset of colorectal cancer patients are more likely to experience chemotherapy-induced toxicity and to determine the specific attributes that influence the presence of treatment-related side effects. Methods: The predictor was general toxicity, and for the construction of our data training, we selected 95 characteristics that represent the health state of 74 patients prior to their first round of chemotherapy. After the data were processed, Random Forest models were trained to offer an optimal balance between accuracy and interpretability. Results: We constructed a machine learning predictor with an emphasis on assessing the importance of numerical and categorical variables in relation to toxicity. Conclusions: The incorporation of artificial intelligence in personalizing colorectal cancer management by anticipating and overseeing toxicities more effectively illustrates a pivotal shift towards more personalized and precise medical care.

We performed a meta-analysis of chemo-brain diagnostic, pooling sensitivities, and specificities in order to assess the accuracy of a machine-learning (ML) algorithm in breast cancer survivors previously treated with chemotherapy. We searched PubMed, Web of Science, and Scopus for eligible articles before 30 September 2022. We identified three eligible studies from which we extracted seven ML algorithms. For our data, the χ² tests demonstrated the homogeneity of the sensitivity’s models (χ² = 7.6987, df = 6, p-value = 0.261) and the specificities of the ML models (χ² = 3.0151, df = 6, p-value = 0.807). The pooled area under the curve (AUC) for the overall ML models in this study was 0.914 (95%CI: 0.891–0.939) and partial AUC (restricted to observed false positive rates and normalized) was 0.844 (95%CI: 0.80–0.889). Additionally, the pooled sensitivity and pooled specificity values were 0.81 (95% CI: 0.75–0.86) and 0.82 (95% CI: 0.76–0.86), respectively. From all included ML models, support vector machine demonstrated the best test performance. ML models represent a promising, reliable modality for chemo-brain prediction in breast cancer survivors previously treated with chemotherapy, demonstrating high accuracy.

Breast cancer remains a leading cause of cancer-related mortality worldwide, with treatment resistance and tumor heterogeneity posing major clinical challenges. MicroRNAs (miRNAs), small non-coding RNAs regulating gene expression, have emerged as key players in breast cancer biology, influencing tumor initiation, progression, and therapy resistance. This narrative review synthesizes recent evidence on the involvement of miRNAs in breast cancer subtypes and their impact on treatment response. Notably, miR-155, miR-503, and miR-21 have shown potential as non-invasive biomarkers and modulators of pathways such as PI3K-Akt, MAPK, and TNF signaling. Additionally, exosomal miRNAs may reflect chemoresistance profiles and predict pathological response to neoadjuvant therapy. Emerging data also support the use of specific miRNAs to sensitize tumors to radiotherapy or modulate immune checkpoints like PD-L1 in triple-negative breast cancer. However, challenges persist regarding standardization, sample types, and study heterogeneity. Further translational research is needed to validate miRNA signatures and their utility in guiding personalized treatment. By highlighting mechanistic insights and potential clinical applications, this review aims to contribute to the ongoing efforts of integrating miRNAs into precision oncology for breast cancer.

Background: Advanced-stage ovarian cancer presents a significant therapeutic challenge, with primary cytoreductive surgery (PCS) followed by chemotherapy and neoadjuvant chemotherapy (NACT) with interval debulking surgery (IDS) as the two main treatment modalities. This study aims to compare the clinical outcomes, surgical complexity, and survival rates between these approaches and to assess the impact of molecular markers such as BRCA and HRD status. Methods: This retrospective, single-center observational study included 100 patients diagnosed with stage III-IV high-grade serous ovarian cancer. The patients were divided into two cohorts based on their treatment strategy: PCS followed by adjuvant chemotherapy or NACT followed by IDS. Clinical outcomes, recurrence-free survival (RFS), progression-free survival (PFS), and overall survival (OS) were analyzed, along with the impact of genetic biomarkers. Results: No statistically significant differences were observed in OS and PFS between the two treatment approaches. Patients who underwent NACT followed by IDS had lower surgical complexity scores and reduced perioperative morbidity. The HRD-positive patients exhibited improved responses to PARP inhibitors, reinforcing the significance of molecular profiling in therapeutic decision-making. The KELIM scores demonstrated prognostic relevance, particularly in the patients receiving neoadjuvant chemotherapy. Conclusion: Both PCS and NACT-IDS are viable treatment options for advanced ovarian cancer, with similar survival outcomes. The choice between strategies should be tailored based on patient-specific factors, including tumor burden, performance status, and molecular profile. The integration of biomarkers such as BRCA mutations and HRD status into clinical practice can further refine treatment selection and improve personalized management strategies.

Neoadjuvant chemotherapy (NAC) is the standard of care for locally advanced breast cancer, yet the molecular predictors of pathological response remain incompletely defined, particularly regarding microRNA (miRNA) dynamics. We investigated paired pre- and post-NAC miRNA expression profiles in relation to molecular subtype, residual cancer burden (RCB), and clinical timing parameters. Seven patients with invasive breast cancer (Luminal A n = 3, Luminal B n = 1, TNBC n = 2, HER2+ n = 1) who received NAC (AC-T or TCHP) were included in this pilot study. Small-RNA sequencing (NovaSeq X Plus, CeGaT GmbH, project S17293) was performed on 14 FFPE specimens (7 pre-NAC core needle biopsies, 7 post-NAC surgical specimens). Differential expression analysis used the paired Wilcoxon signed-rank test with Benjamini–Hochberg correction. Spearman correlations assessed associations between miRNA expression, RCB score, and clinical timing intervals. Candidate miRNAs were subsequently annotated using experimentally validated miRNA–target interactions. After filtering (≥ three counts in ≥ three samples), 759 miRNAs were analysed. No miRNA reached strict significance (padj < 0.05, |log₂FC| > 1.0) after multiple testing correction, consistent with the limited statistical power (n = 7). Under exploratory criteria (p < 0.10, |log₂FC| > 0.5), 156 candidate miRNAs were identified: 70 upregulated and 86 downregulated post-NAC. Leading candidates included hsa-miR-139-3p (+2.60), hsa-miR-139-5p (+2.36), and hsa-miR-1323 (+1.55) as upregulated, and hsa-miR-429 (−2.53), hsa-miR-141-3p (−1.79), and hsa-miR-1277-5p (−1.25) as downregulated post-NAC. The single patient achieving the lowest residual disease burden (P3, Luminal B, RCB-I, score 1.32) displayed a distinct pre-treatment miRNA profile, separating from all other pre-NAC specimens on principal component analysis and characterised by higher baseline hsa-miR-139-3p/-5p and lower baseline hsa-miR-429 and hsa-miR-141-3p expression, suggesting that baseline miRNA expression patterns may contribute to differential chemotherapy response. RCB score showed a non-significant positive trend with post-NAC Ki-67 (ρ = +0.71, p = 0.07). This pilot study identifies NAC-modulated candidate miRNAs in breast cancer and establishes a paired FFPE-based small-RNA-sequencing workflow applicable in routine clinical settings. The distinct pre-treatment profile of the single best responder generates the testable hypothesis that baseline expression of tumour suppressor miRNAs of the miR-139 family, together with low miR-200-family expression, may track chemosensitivity. As no candidate reached statistical significance after multiple testing correction and none has been validated in an independent cohort or by an orthogonal method, all findings are exploratory and hypothesis-generating. The results support larger prospective validation studies examining miRNA signatures as predictive biomarkers of NAC response across breast cancer molecular subtypes.