An artificial intelligence-driven predictive model for pediatric allogeneic hematopoietic stem cell transplantation using clinical variables.

BACKGROUND: Hematopoietic stem cell transplantation (HSCT) is a procedure with high morbidity and mortality. Identifying patients for maximum benefit and risk assessment is crucial in the decision-making process. This has led to the development of predictive risk models for HSCT in adults, which have limitations when applied to pediatric population. Our goal was to develop an automatic learning algorithm to predict survival in children with malignant disorders undergoing HSCT. METHODS: We studied allogenic HSCTs performed on children with malignant disorders at a third-level hospital between 1991 and 2021. Survival was analyzed using the Kaplan-Meier method, log-rank test for the univariate analysis, and Cox regression for the multivariate analysis. A prognostic index was constructed based on these findings. Lastly, we constructed a predictive model using a random forest algorithm to forecast 1-year survival after HSCT. RESULTS: We analyzed 229 HSCTs in 201 patients with a median follow-up of 1.64 years. Variables that impacted on the multivariate analysis were older age (hazard ratio [HR] 1.40, 95% confidence interval [CI] 1.12-1.76, p = .003), oldest period of HSCT (HR 0.46, 95% CI 0.29-0.73, p < .001), and mismatched donor (HR 2.65, 95% CI 1.51-4.65, p = .001). Our prognostic index was associated with 3-year overall survival (OS; p < .001). A random forest was developed using as variables: diagnosis, age, year of HSCT, time from diagnosis to HSCT, disease stage, donor type, and conditioning. This achieved 72% accuracy in predicting 1-year OS. CONCLUSIONS: Our index and random forest was effective in predicting 1-year survival. However, further validation in diverse populations is necessary to establish their generalizability.

Treosulfan-Based Conditioning Regimen In Pediatric Hematopoietic Stem Cell Transplantation: A Retrospective Analysis on Behalf of the Spanish Group for Hematopoietic Transplantation and Cellular Therapy (GETH-TC).

Increasing data on treosulfan-based conditioning regimens before hematopoietic stem cell transplantation (HSCT) demonstrate the consistent benefits of this approach, particularly regarding acute toxicity. This study aimed to describe the results of treosulfan-based conditioning regimens in children, focusing on toxicity and outcomes when used to treat both malignant and nonmalignant diseases. This retrospective observational study of pediatric patients treated in Spain with treosulfan-based conditioning regimens before HSCT was based on data collection from electronic clinical records. We studied a total of 160 treosulfan-based conditioning HSCTs to treat nonmalignant diseases (n = 117) or malignant diseases (n = 43) in 158 children and adolescents. The median patient age at HSCT was 5.1 years (interquartile range, 2 to 10 years). The most frequent diagnoses were primary immunodeficiency (n = 42; 36%) and sickle cell disease (n = 42; 36%) in the nonmalignant disease cohort and acute lymphoblastic leukemia (n = 15; 35%) in the malignant disease cohort. Engraftment occurred in 97% of the patients. The median times to neutrophil engraftment (17 days versus 14 days; P = .008) and platelet engraftment (20 days versus 15 days; P = .002) were linger in the nonmalignant cohort. The 1-year cumulative incidence of veno-occlusive disease was 7.98% (95% confidence interval [CI], 4.6% to 13.6%), with no significant differences between cohorts. The 1-year cumulative incidence of grade III-IV acute graft-versus-host disease (GVHD) was higher in the malignant disease cohort (18% versus 3.2%; P = .011). Overall, the malignant cohort had both a higher total incidence (9% versus 3%; P < .001) and a higher 2-year cumulative incidence (16% versus 1.9%; P < .001) of total chronic GVHD. The 2-year cumulative transplantation-related mortality was 15%, with no difference between the 2 cohorts. The 5-year overall survival was 80% (95% CI, 72% to 86%) and was higher in the nonmalignant cohort (87% versus 61%; P = .01). The 2-year cumulative incidence of relapse was 25% in the malignant cohort. The 5-year cumulative GVHD-free, relapse-free survival rate was 60% (95% CI, 51% to 70%) and was higher in the nonmalignant cohort (72% versus 22%; P < .001). A treosulfan-based radiation-free conditioning regimen is feasible, achieving a high engraftment rate and 5-year overall survival, and is an emerging option for the first HSCT in nonmalignant diseases.

Phase 2 Clinical Trial of Infusing Haploidentical K562-mb15-41BBL-Activated and Expanded Natural Killer Cells as Consolidation Therapy for Pediatric Acute Myeloblastic Leukemia.

BACKGROUND: Acute myeloid leukemia (AML) accounts for approximately 20% of pediatric leukemia cases; 30% of these patients experience relapse. The antileukemia properties of natural killer (NK) cells and their safety profile have been reported in AML therapy. We proposed a phase 2, open, prospective, multicenter, nonrandomized clinical trial for the adoptive infusion of haploidentical K562-mb15-41BBL-activated and expanded NK (NKAE) cells as a consolidation strategy for children with favorable and intermediate risk AML in first complete remission after chemotherapy (NCT02763475). PATIENTS AND METHODS: Before the NKAE cell infusion, patients underwent a lymphodepleting regimen. After the NKAE cell infusion, patients were administered low doses (1 × 106/IU/m2) of subcutaneous interleukin-2. The primary study endpoint was AML relapse-free survival. We needed to include 35 patients to demonstrate a 50% reduction in relapses. RESULTS: Seven patients (median age, 7.4 years; range, 0.78-15.98 years) were administered 13 infusions of NKAE cells, with a median of 36.44 × 106 cells/kg (range, 6.92 × 106 to 193.2 × 106 cells/kg). We observed chimerism in 4 patients (median chimerism, 0.065%; range, 0.05-0.27%). After a median follow-up of 33 months, the disease of 6 patients (85.7%) remained in complete remission. The 3-year overall survival was 83.3% (95% confidence interval, 68.1-98.5), and the cumulative 3-year relapse rate was 28.6% (95% confidence interval, 11.5-45.7). The study was terminated early because of low patient recruitment. CONCLUSION: This study emphasizes the difficulties in recruiting patients for cell therapy trials, though NKAE cell infusion is safe and feasible. However, we cannot draw any conclusions regarding efficacy because of the small number of included patients and insufficient biological markers.