Epidemiology of febrile neutropenia in children with central nervous system tumor: results from a single center prospective study

Development of a Machine Learning-Based Prediction Model for Chemotherapy-Induced Myelosuppression in Children with Wilms' Tumor

Purpose: Develop and validate an accessible prediction model using machine learning (ML) to predict the risk of chemotherapy-induced myelosuppression (CIM) in children with Wilms' tumor (WT) before chemotherapy is administered, enabling early preventive management. Methods: A total of 1433 chemotherapy cycles in 437 children with WT who received chemotherapy in our hospital from January 2009 to March 2022 were retrospectively analyzed. Demographic data, clinicopathological characteristics, hematology and blood biochemistry baseline results, and medication information were collected. Six ML algorithms were used to construct prediction models, and the predictive efficacy of these models was evaluated to select the best model to predict the risk of grade ≥ 2 CIM in children with WT. A series of methods, such as the area under the receiver operating characteristic curve (AUROC), the calibration curve, and the decision curve analysis (DCA) were used to test the model's accuracy, discrimination, and clinical practicability. Results: Grade ≥ 2 CIM occurred in 58.5% (839/1433) of chemotherapy cycles. Based on the results of the training and validation cohorts, we finally identified that the extreme gradient boosting (XGB) model has the best predictive efficiency and stability, with an AUROC of up to 0.981 in the training set and up to 0.896 in the test set. In addition, the calibration curve and the DCA showed that the XGB model had the best discrimination and clinical practicability. The variables were ranked according to the feature importance, and the five variables contributing the most to the model were hemoglobin (Hgb), white blood cell count (WBC), alkaline phosphatase, coadministration of highly toxic chemotherapy drugs, and albumin. Conclusions: The incidence of grade ≥ 2 CIM was not low in children with WT, which needs attention. The XGB model was developed to predict the risk of grade ≥ 2 CIM in children with WT for the first time. The model has good predictive performance and stability and has the potential to be translated into clinical applications. Based on this modeling and application approach, the extension of CIM prediction models to other pediatric malignancies could be expected.

MMP-10 and TIMP-1 as indicators of severe sepsis in adult hematological patients with febrile neutropenia

Commonly used indicators of sepsis are nonspecific and insufficient for predicting the course of febrile neutropenia (FN) in hematological patients. We analyzed data from 91 adult FN patients who received intensive chemotherapy for acute myeloid leukemia or autologous stem cell transplantation. Compared to patients with non-severe sepsis, patients with severe sepsis had significantly higher serum levels of tissue inhibitor of metalloproteinases-1 on the day of first occurrence of fever (day 0: 172 vs. 112 µg/L, p= .002) and for the two following days (day 1: 219 vs. 128 µg/L, p< .001; day 2: 443 vs. 128 µg/L, p= .001), and significantly higher serum levels of matrix metalloproteinase-10 on day 1 (1975 vs. 876 ng/L, p= .001) and day 2 (2020 vs. 841 ng/L, p< .001). We conclude that the measurement of these biomarkers may be useful in predicting the severity of sepsis in FN patients.

Epidemiology of invasive fungal diseases in children with solid tumours undergoing autologous haematopoietic stem cell transplantation: a 10-year experience in a tertiary Italian centre

The objective of the study was to determine the incidence of invasive fungal disease (IFD) in children undergoing autologous haematopoietic stem cell transplantation (auHSCT) for solid tumours (ST). Retrospective study on auHSCT was performed in children with ST (January 2006-December 2015). Data on the number of patient-days at risk (pdr) during the first 30 and 90 days after auHSCT and cases of proven/probable IFDs were collected. Infection rate (IR, episodes/1000 pdr) and proportions and cumulative risk (CR) of IFD were evaluated. In 186 patients, 270 auHSCT were performed, for a total of 8327 pdr during the first 30 days and 24 366 up to day 90. Median age was 5 years (interquartile range 2;8), 63% were male. At day 30, seven procedures were complicated by IFD, with an IR of 0.84 (95% CI 0.66-1.02) and aCR of 2.6% (95% CI 1.4-5.4) at 18 days after HSCT. Within day 90, two further IFDs were detected with an IR of 0.37 (95% CI -0.49 to 1.23) and a CR of 3.3% (95% CI 1.7-6.3) at day 69. Children undergoing auHSCT for ST have a low incidence of IFDs in the first 90 days after the procedure.

Bloodstream infection in paediatric cancer centres--leukaemia and relapsed malignancies are independent risk factors

In a prospective multicentre study of bloodstream infection (BSI) from November 01, 2007 to July 31, 2010, seven paediatric cancer centres (PCC) from Germany and one from Switzerland included 770 paediatric cancer patients (58% males; median age 8.3 years, interquartile range (IQR) 3.8-14.8 years) comprising 153,193 individual days of surveillance (in- and outpatient days during intensive treatment). Broviac catheters were used in 63% of all patients and Ports in 20%. One hundred forty-two patients (18%; 95% CI 16 to 21%) experienced at least one BSI (179 BSIs in total; bacteraemia 70%, bacterial sepsis 27%, candidaemia 2%). In 57%, the BSI occurred in inpatients, in 79% after conventional chemotherapy. Only 56 % of the patients showed neutropenia at BSI onset. Eventually, patients with acute lymphoblastic leukaemia (ALL) or acute myeloblastic leukaemia (AML), relapsed malignancy and patients with a Broviac faced an increased risk of BSI in the multivariate analysis. Relapsed malignancy (16%) was an independent risk factor for all BSI and for Gram-positive BSI.

CONCLUSION

This study confirms relapsed malignancy as an independent risk factor for BSIs in paediatric cancer patients. On a unit level, data on BSIs in this high-risk population derived from prospective surveillance are not only mandatory to decide on empiric antimicrobial treatment but also beneficial in planning and evaluating preventive bundles.

WHAT IS KNOWN

• Paediatric cancer patients face an increased risk of nosocomial bloodstream infections (BSIs). • In most cases, these BSIs are associated with the use of a long-term central venous catheter (Broviac, Port), severe and prolonged immunosuppression (e.g. neutropenia) and other chemotherapy-induced alterations of host defence mechanisms (e.g. mucositis). What is New: • This study is the first multicentre study confirming relapsed malignancy as an independent risk factor for BSIs in paediatric cancer patients. • It describes the epidemiology of nosocomial BSI in paediatric cancer patients mainly outside the stem cell transplantation setting during conventional intensive therapy and argues for prospective surveillance programmes to target and evaluate preventive bundle interventions.

Risk factors for febrile neutropenia among patients with cancer receiving chemotherapy: A systematic review

Neutropenia with fever (febrile neutropenia [FN]) is a serious consequence of myelosuppressive chemotherapy that usually results in hospitalization and the need for intravenous antibiotics. FN may result in dose reductions, delays, or even discontinuation of chemotherapy, which, in turn, may compromise patient outcomes. It is important to identify which patients are at high risk for developing FN so that patients can receive optimal chemotherapy while their risk for FN is appropriately managed. A systematic review of the literature was performed to gain a comprehensive and updated understanding of FN risk factors. Older age, poor performance status, advanced disease, certain comorbidities, low baseline blood cell counts, low body surface area/body mass index, treatment with myelosuppressive chemotherapies, and specific genetic polymorphisms correlated with the risk of developing FN. Albeit many studies have analyzed FN risk factors, there are several limitations, including the retrospective nature and small sample sizes of most studies.