Background: MYH9-related disease (MYH9-RD) is the most common form of inherited thrombocytopenia (IT). It is caused by pathogenic variants in the MYH9 gene. It manifests as early-onset macrothrombocytopenia with variable later-onset extra-hematological features, including hearing loss, renal disease, and cataracts. In pediatric patients, early recognition is critical to avoid misdiagnosis as immune thrombocytopenia (ITP) and unnecessary immunosuppressive therapy. Methods: We conducted a retrospective unicentric study at the Pediatric Oncology and Hematology Department, Fundeni Clinical Institute, Bucharest, Romania, including patients aged 0–18 years with suspected IT, tested between 2017 and 2025 by next-generation sequencing (NGS). Clinical, laboratory, and genetic data were reviewed. Results: Among 66 patients who underwent genetic testing, 31 (48.5%) had IT-associated genetic variants; 8 (25.8%) carried MYH9 mutations. Four patients (50%) had disease onset before age 1 year, three with neonatal presentation; 3 (37.5%) reported a family history of thrombocytopenia. Six variants were previously reported, and two were novel variants. Five variants (62.5%) were pathogenic, while three (37.5%) were initially classified as variants of uncertain significance (VUS). Most mutations were missense in the coiled-coil tail domain, correlating with milder thrombocytopenia and absence of extra-hematological features. No life-threatening bleeding was recorded; hemorrhagic symptoms were limited to minor mucocutaneous bleeding. Conclusions: This is the first Romanian pediatric cohort and one of the few existing pediatric cohorts describing the genetic and clinical spectrum of MYH9-RD. Early genetic confirmation enables precise diagnosis, tailored management, and family screening, while preventing inappropriate therapies.

RUNX1 alterations contribute to pediatric myeloid malignancies through both germline predisposition syndromes and somatic leukemogenic events, but their clinical and biological significance in children remains incompletely defined. This retrospective single-center study evaluated six pediatric patients with myelodysplastic syndromes or acute leukemias harboring RUNX1 variants, integrating clinical, cytogenetic, and targeted next-generation sequencing data, with germline status confirmed using non-hematopoietic tissues. Three patients carried germline RUNX1 variants, characterized by antecedent cytopenias, dysplastic features, and increased treatment-related toxicity, including severe infections, persistent cytopenias, and transplant-related mortality. In contrast, somatic RUNX1 alterations were associated with overt high-risk disease, frequently accompanied by complex cytogenetics or monosomy 7, and demonstrated heterogeneous outcomes ranging from sustained remission to post-transplant relapse. Mixed-phenotype acute leukemia was observed in both groups. These findings support a model of RUNX1-driven leukemogenesis, in which germline and somatic alterations represent distinct yet interconnected trajectories, while highlighting the importance of distinguishing variant origin for risk stratification, donor selection, and therapeutic decision-making in pediatric myeloid malignancies. Given the small cohort size, the findings remain descriptive and require validation in larger prospective studies.

Pediatric acute myeloid leukemia (AML) is biologically heterogeneous, and genomic profiling increasingly informs risk stratification and treatment. We evaluated the relationship between induction response, genomic risk, transplant allocation, and survival in pediatric AML. We retrospectively analyzed 38 pediatric patients with newly diagnosed AML, treated between 2020 and 2025. Clinical, cytogenetic, molecular, treatment, and outcome data were collected. Genomic alterations were assessed using cytogenetics, fluorescence in situ hybridization (FISH), molecular testing, and next-generation sequencing (NGS). Survival was estimated by Kaplan–Meier analysis, and prognostic factors for event-free survival (EFS) were assessed using univariable Cox regression. This study is exploratory given the limited sample size and should be interpreted accordingly. Complete remission (CR) after the first course of induction was achieved in 25/38 patients (65.8%), partial remission (PR) in 3/38 (7.9%), and refractory disease in 10/38 (26.3%). Twenty-four patients underwent allogeneic hematopoietic stem cell transplantation; 17/24 (70.8%) were alive at last follow-up, with a 2-year overall survival rate of 72.9%. Both induction response and genomic risk stratification showed suggestive associations with outcome; descriptively, induction response showed the strongest prognostic discrimination, with achievement of CR associated with markedly improved survival. High cytogenetic risk and FLT3-ITD were significantly associated with inferior EFS. Post-induction measurable residual disease (MRD) positivity was detected in 16 of 38 patients (42.1%) and was associated with suboptimal induction response; MRD negativity did not uniformly preclude adverse outcomes, particularly in the high-risk genomic subgroup. Genomic profiling refined biological risk and post-remission treatment allocation. Integrated assessment of genomic risk, induction response, and MRD status may improve therapeutic stratification in pediatric AML.

Background: Acute lymphoblastic leukaemia (ALL) is the most common paediatric haematological malignancy. Treatment-related immunosuppression compromises mucosal integrity and contributes to oral and dento-maxillary complications, which remain a clinically relevant but insufficiently characterised source of morbidity. Salivary secretory immunoglobulin A (sIgA), the principal antibody of mucosal surfaces, may provide a non-invasive surrogate measure of local immune competence; however, its longitudinal behaviour and relationship with oral morbidity in paediatric ALL remain incompletely characterised. Methods: This prospective, single-centre pilot cohort study included 21 children with newly diagnosed ALL treated at the Fundeni Clinical Institute, Bucharest (March–November 2024). Salivary sIgA was measured at diagnosis (TP1) and after completion of intravenous chemotherapy (TP3). Oral mucositis (OM) was clinically assessed and graded according to WHO criteria at induction (TP2), as well as the predominant WHO grade across the post-induction chemotherapy course. Stomatological evaluation included DMFT/dmft and PUFA/pufa indices, and microbiological screening was performed at predefined time points. Statistical analyses were exploratory and included non-parametric tests and correlation analyses. Results: All patients developed OM at induction (TP2), with 38% experiencing severe forms (WHO grades 3–4). Baseline salivary sIgA (TP1) (n = 15) showed a non-significant weak inverse association with mucositis severity (Spearman ρ = −0.35, p = 0.20); differences across severity groups did not reach statistical significance (Kruskal–Wallis p = 0.43). When grouped, lower median salivary sIgA levels were observed in patients with severe mucositis compared with mild-to-moderate forms (median 91.0 mg/L (IQR 65.0–123.0) vs. 170.5 mg/L (IQR 125.0–231.8); non-significant p = 0.13). A non-significant decline in salivary sIgA was observed from TP1 to TP3 (median 174.5, IQR 127.2–213.8 versus median 111.5, IQR 33.2–142.7; Wilcoxon signed-rank p = 0.084; n = 10 paired measurements). Pre-existing dento-maxillary morbidity was substantial (median DMFT 4.0 (IQR 2.5–5.5), PUFA 3.0 (IQR 1.5–4); dmft 4.0 (IQR 3.0–5.0), pufa 3.0 (IQR 2.0–3.5)), but showed no statistically significant association with either mucositis grade or salivary sIgA levels. No clinically meaningful correlation was observed between salivary and intestinal sIgA levels (Spearman ρ = 0.33 overall; ρ = 0.16 after exclusion of an extreme outlier; p > 0.05 in both analyses), consistent with immunological compartmental independence. Conclusions: In this exploratory pilot cohort, no statistically significant associations were observed between salivary sIgA and oral mucositis severity or dento-maxillary indices. Although trends toward lower baseline salivary sIgA in severe mucositis and declining sIgA during induction therapy were observed, larger adequately powered studies are needed to determine whether these findings represent true biological associations.

Background: Malnutrition, encompassing both undernutrition and overnutrition, is a common complication in children with cancer and is associated with impaired treatment tolerance, increased infection risk, altered pharmacokinetics, reduced quality of life, and poorer survival outcomes. Despite its importance, nutritional management in pediatric oncology lacks a unified, systematically organized clinical framework applicable to the full trajectory of the disease. Objective: This study aimed to develop expert consensus recommendations for nutritional intervention in pediatric oncology patients aged 4 to 18 years. Methods: A modified electronic Delphi (e-Delphi) process was conducted with a multidisciplinary expert panel of 22 specialists, including pediatric oncologists, pediatric gastroenterologists, clinical nutrition specialists, radiotherapy specialists, and pediatric surgeons. Statements were rated on a 9-point Likert scale across two anonymous rounds, with consensus predefined as ≥80% agreement. Results: Forty-one consensus recommendations were formulated across nine domains: nutritional screening and assessment, energy and protein requirements, micronutrient supplementation, physical activity, nutritional support escalation, refeeding syndrome prevention, treatment-specific management, survivorship, and palliative care. All recommendations achieved the predefined consensus threshold. Conclusions: This Delphi consensus provides a structured, multidisciplinary, and clinically actionable framework for nutritional management across the full trajectory of childhood cancer and is intended to reduce institutional variability and improve patient outcomes.

Intestinal microbiome dysbiosis has been associated with transplant-related mortality and graft-versus-host disease in allo-HSCT patients. We assessed how pre-transplant antibiotic and antineoplastic exposures, together with multidrug-resistant colonization, are associated with baseline gut microbiome diversity at allo-HSCT. We conducted a prospective, single-center cohort study at Fundeni Clinical Institute (Bucharest, Romania) between August 2024 and June 2025, enrolling 52 allo-HSCT recipients and 27 healthy controls. Fecal samples were collected before conditioning. Gut microbiome composition was assessed via 16S rRNA gene sequencing and analyzed using QIIME2 and R. Associations were evaluated using Wilcoxon test, multivariable linear regression, and PERMANOVA. Shannon diversity was significantly lower in patients (median 4.71, IQR 3.97–5.44) than in healthy controls (median 6.09, IQR 5.87–6.28; p < 0.001). In bivariate analyses, carbapenem (p adj = 0.02) and oxazolidinone exposure (p adj = 0.005) were associated with reduced diversity, while immunotherapy was associated with higher diversity (p adj = 0.042). Broad-spectrum penicillin (p adj = 0.062) and ESBL colonization (p adj = 0.066) did not reach significance. In the multivariable antibiotic model, although the overall model was statistically significant (model p = 0.039), no individual antibiotic class remained significantly associated with Shannon diversity after adjustment for co-exposures. Beta diversity differed modestly with carbapenem exposure (R² = 0.033, p = 0.019). Pre-transplant antibiotic exposures were associated with lower gut microbiome diversity at allo-HSCT admission, with patterns consistent with a cumulative rather than a class-specific association. These findings support antibiotic stewardship in pre-transplant care.