Haploidentical Hematopoietic Cell Transplantation Using Post-transplant Cyclophosphamide for Children with Non-malignant Diseases.

Haploidentical hematopoietic cell transplantation (HCT) is a valuable curative option for children with non-malignant diseases. Haploidentical HCT using post-transplant cyclophosphamide (PTCy) is a readily available option in the absence of an HLA-matched donor. We conducted a retrospective single-center study on the outcome of haploidentical HCT in children with non-malignant diseases. We gathered data from 44 patients underwent HCT in the period 2015 to 2020. The indications for HCT were bone marrow failure, primary immunodeficiency, metabolic disorders, and hemoglobinopathy. Median age at HCT was 4 years (range 0.7-20). The conditioning regimens were myeloablative (n = 17) or reduced intensity (n = 27). After a median follow-up of 20 months (range 4-71), 2-year overall survival was 89% and 2-year GvHD-free relapse-free survival (GRFS) was 66%. Incidence of primary graft failure was 13.6%. Cumulative incidence of grade II-IV acute and moderate/severe chronic GvHD were 20% and 6.4%, respectively. Younger age at HCT (< 4 years) and primary immunodeficiency were significantly associated with better GRFS (p < 0.05). In conclusion, haploidentical HCT using PTCy is feasible and curative in children with non-malignant diseases lacking an HLA-matched donor. Early diagnosis and referral in addition to timely treatment can further improve outcomes.

Using ChatGPT to Predict Cancer Predisposition Genes: A Promising Tool for Pediatric Oncologists.

BACKGROUND: Determining genetic susceptibility for cancer predisposition syndromes (CPS) through cancer predisposition genes (CPGs) testing is critical in facilitating appropriate prevention and surveillance strategies. This study investigates the use of ChatGPT, a large language model, in predicting CPGs using clinical notes. METHODS: Our study involved 53 patients with pathogenic CPG mutations. Two kinds of clinical notes were used: the first visit note, containing a thorough history and physical exam, and the genetic clinic note, summarizing the patient's diagnosis and family history. We asked ChatGPT to recommend CPS genes based on these notes and compared these predictions with previously identified mutations. RESULTS: Rb1 was the most frequently mutated gene in our cohort (34%), followed by NF1 (9.4%), TP53 (5.7%), and VHL (5.7%). Out of 53 patients, 30 had genetic clinic notes of a median length of 54 words. ChatGPT correctly predicted the gene in 93% of these cases. However, it failed to predict EPCAM and VHL genes in specific patients. For the first visit notes (median length: 461 words), ChatGPT correctly predicted the gene in 64% of these cases. CONCLUSION: ChatGPT shows promise in predicting CPGs from clinical notes, particularly genetic clinic notes. This approach may be useful in enhancing CPG testing, especially in areas lacking genetic testing resources. With further training, there is a possibility for ChatGPT to improve its predictive potential and expand its clinical applicability. However, additional research is needed to explore the full potential and applicability of ChatGPT.

Early post-induction augmented therapy improves outcome in children and adolescents with T-cell acute lymphoblastic leukemia.

INTRODUCTION: T-cell acute lymphoblastic leukemia (T-ALL) accounts for approximately 15% of all newly diagnosed ALL in children and adolescents and is associated with worse outcomes compared to pre-B ALL. We aimed to decrease T-ALL relapses by intensifying our regimen. METHODS: Patients with T-ALL were treated using two different regimens; before September 2014, patients were treated per St. Jude Total XV protocol; subsequently, a major change was adopted by adding two intensive blocks: FLAG and Reintensification. Cranial radiation was limited to patients with WBC ≥ 100 k/µl at diagnosis and/or patients with CNS2/CNS3 status. RESULTS: Between June 2005 and April 2020, a total of 100 patients (76 males) were treated and followed up for a median of 70 months (range 14-181). Median age at diagnosis was 9 years (range 0.5-17.8). Forty-eight patients were diagnosed after September 2014 and received the augmented regimen; their median follow up was 46 months (range 14-74). The 5-year-EFS estimates for patients who received the augmented regimen versus standard regimen were 87% ± 4.9% versus 67% ± 6.8% (p = .03); and the 5-year-OS estimates were 87% ± 5.1% versus 71% ± 6.3% (p = .06), respectively. Treatment related mortality (TRM) was reported in two patients treated per standard regimen but none for patients who received the augmented regimen. CONCLUSIONS: We implemented a novel approach with early intensification added to a backbone of modified St. Jude Total-XV regimen for patients with T-ALL that resulted in improved outcome with no treatment related mortality.

Factors that predict severity of infection and seroconversion in immunocompromised children and adolescents with COVID-19 infection.

Objectives: We aimed to study the outcomes, severity, and seroconversion post SARS-CoV-2 infection in immunocompromised children and adolescents treated at our center. Method: For this observational study, all pediatric patients who had COVID-19 infection from Sep-22-2020 to Nov-10-2021were identified by reviewing our laboratory records. Their charts were reviewed to determine clinical severity and outcome. Blood samples were drawn for anti-SARS-CoV-2 antibody assay. Serious COVID-19 infection (SVI) was defined if the patient had moderate, severe, or critical illness. A cutoff of 100 U/mL anti-SARS-CoV-2 antibodies was used to categorize low and high titer seroconversion. Results: We identified 263 pediatric patients with COVID-19; most (68%) were symptomatic: 5% had severe or critical infection, 25% were hospitalized, 12 required respiratory support, 12 were admitted to the ICU, and five patients (2%) died. Multivariable analysis revealed several factors that predict SVI: Age above 12 years (p=0.035), body mass index above 95th percentile (p=0.034), comorbid conditions (p=0.025), absolute neutrophil count ≤500(p=0.014) and absolute lymphocyte count ≤300 (p=0.022). Levels of anti-SARS-CoV-2 spike antibodies were obtained for 173 patients at a median of 94 days (range, 14-300) after PCR diagnosis; of them 142 (82%) patients seroconverted; the lowest seroconversion rate was observed in patients with hematological malignancies (79%). Our univariable model showed that the following factors were predictive of low titer: lower ANC, p=0.01; hematologic malignancy, p=0.023; receiving steroids in the last 14 days, p=0.032; time since last chemotherapy or immunosuppressive therapy less than 30 days, p=0.002; and being on active chemotherapy in the last 3 months prior to infection, p<0.001. Conclusions: SARS-CoV-2 antibodies developed in most immunocompromised patients with COVID-19 infection in our study. Mortality was relatively low in our patients. Our univariable and multivariable models showed multiple variables that predict severity of infections and antibody response post COVID-19 infection. These observations may guide choice of active therapy during infection and the best timing of vaccination in this high-risk population.