Quantifying the difference in risk of adverse events by induction treatment regimen in pediatric acute lymphoblastic leukemia

An Interpretable Machine Learning Framework for Rare Disease: A Case Study to Stratify Infection Risk in Pediatric Leukemia

Background: Datasets on rare diseases, like pediatric acute myeloid leukemia (AML) and acute lymphoblastic leukemia (ALL), have small sample sizes that hinder machine learning (ML). The objective was to develop an interpretable ML framework to elucidate actionable insights from small tabular rare disease datasets. Methods: The comprehensive framework employed optimized data imputation and sampling, supervised and unsupervised learning, and literature-based discovery (LBD). The framework was deployed to assess treatment-related infection in pediatric AML and ALL. Results: An interpretable decision tree classified the risk of infection as either "high risk" or "low risk" in pediatric ALL (n = 580) and AML (n = 132) with accuracy of ∼79%. Interpretable regression models predicted the discrete number of developed infections with a mean absolute error (MAE) of 2.26 for bacterial infections and an MAE of 1.29 for viral infections. Features that best explained the development of infection were the chemotherapy regimen, cancer cells in the central nervous system at initial diagnosis, chemotherapy course, leukemia type, Down syndrome, race, and National Cancer Institute risk classification. Finally, SemNet 2.0, an open-source LBD software that links relationships from 33+ million PubMed articles, identified additional features for the prediction of infection, like glucose, iron, neutropenia-reducing growth factors, and systemic lupus erythematosus (SLE). Conclusions: The developed ML framework enabled state-of-the-art, interpretable predictions using rare disease tabular datasets. ML model performance baselines were successfully produced to predict infection in pediatric AML and ALL.

Adolescents and young adults (AYAs) vs pediatric patients: survival, risks, and barriers to enrollment

Adolescents and young adults (AYAs; ages 15-39 years) with acute lymphoblastic leukemia (ALL) have worse outcomes than pediatric patients with ALL. Multiple factors contribute to this differential survival. AYAs are more likely to have higher-risk leukemia biology than children with ALL. AYA patients have more choices for treatment facility and treatment protocol, as well as barriers to clinical trial enrollment, both of which can affect survival. AYAs must also navigate psychosocial factors inherent to their unique developmental stage. Furthermore, AYAs typically sustain more treatment-related toxicities than pediatric patients. Treatment on pediatric or pediatric-inspired ALL protocols at pediatric cancer centers has been associated with improved outcomes for AYAs with ALL, but there is still variation in the treatment that AYAs with ALL receive. Clinical trials focused on AYAs with ALL and individualized decision-making regarding choice of treatment facility and treatment protocol are needed to optimize the survival and long-term outcomes of this patient population.

Fluoroquinolone prophylaxis in patients with neutropenia at high risk of serious infections: Exploring pros and cons

BACKGROUND

The use of fluoroquinolones to prevent infections in neutropenic patients with cancer or undergoing hematopoietic stem cell transplantation (HSCT) is a controversial issue, with international guidelines providing conflicting recommendations. Although potential benefits are clear, concerns revolve around efficacy, potential harms, and antimicrobial resistance (AMR) implications.

DISCUSSION

Fluoroquinolone prophylaxis reduces neutropenic fever (NF) bloodstream infections and other serious bacterial infections, based on evidence from systematic reviews, randomized controlled trials, and observational studies in adults and children. Fluoroquinolone prophylaxis may also reduce infection-related morbidity and healthcare costs; however, evidence is conflicting. Adverse effects of fluoroquinolones are well recognized in the general population; however, studies in the cancer cohort where it is used for a defined period of neutropenia have not reflected this. The largest concern for routine use of fluoroquinolone prophylaxis remains AMR, as many, but not all, observational studies have found that fluoroquinolone prophylaxis might increase the risk of AMR, and some studies have suggested negative impacts on patient outcomes as a result.

CONCLUSIONS

The debate surrounding fluoroquinolone prophylaxis calls for individualized risk assessment based on patient characteristics and local AMR patterns, and prophylaxis should be restricted to patients at the highest risk of serious infection during the highest risk periods to ensure that the risk-benefit analysis is in favor of individual and community benefit. More research is needed to address important unanswered questions about fluoroquinolone prophylaxis in neutropenic patients with cancer or receiving HSCT.

High burden of clinically significant adverse events associated with contemporary therapy for pediatric T-cell acute lymphoblastic leukemia/lymphoma

BACKGROUND

Despite improvements in survival for children with T-cell acute lymphoblastic leukemia and lymphoma (T-ALL/LLy), morbidity remains high. However, data are lacking regarding comprehensive descriptions of clinically relevant adverse events (AEs) experienced during early intensive chemotherapy.

PROCEDURE

This single-institution retrospective study evaluated children aged 1-21 years with T-ALL/T-LLy diagnosed from 2010 to 2020. Physician chart abstraction identified and graded 20 clinically relevant AEs. AE rates were analyzed by T-ALL or LLy, minimal residual disease status, induction steroid, and use of antimicrobial prophylaxis. Statistical comparisons used the Kruskal-Wallis test (continuous variables) and Chi-square or Fisher's exact test (categorical variables).

RESULTS

The cohort included 120 patients (T-ALL: 88; T-LLy: 32). Most patients experienced AEs during induction (85 out of 120; 70.8%) and consolidation (89 out of 111; 80.2%). Nonsepsis infection was common in induction (26 out of 120; 21.7%) and consolidation (35 out of 111; 31.5%). Patients treated with dexamethasone during induction had significantly higher rates of nonsepsis infection and/or sepsis during consolidation than those who received prednisone (p < .01).

CONCLUSIONS

Clinically significant AEs are extremely common during induction and consolidation therapy for patients with T-ALL/LLy. Infectious AEs are particularly prevalent. These results can inform conversations with patients and families and aid in the development of toxicity-related aims in the next generation of, prospective clinical trials in T-ALL/LLy.

Use of the osmolal gap in diagnosing mixed physiology hyponatremia in a child with B-cell acute lymphoblastic leukemia

Hyponatremia is common among children undergoing treatment for hematologic malignancies and may be attributed to multiple underlying causes. In cases of hyponatremia due to mixed physiology, the osmolal gap, can identify pseudohyponatremia that may be masked by other causes.