Clinical and Laboratory Indicators of Pediatric Leukemia and Neuroblastoma in Musculoskeletal Presentations: A Retrospective Study From a Resource-Limited Setting.

BACKGROUND

Musculoskeletal (MSK) symptoms are a frequent but diagnostically ambiguous presentation in children with malignancies such as leukemia and neuroblastoma. These symptoms may precede hematologic abnormalities, complicating early diagnosis and increasing the risk of misclassification. Inappropriate treatments, such as corticosteroids, may further obscure disease features.

OBJECTIVE

To identify clinical and laboratory indicators of leukemia and neuroblastoma at initial evaluation in children presenting with MSK symptoms and to develop a simple preliminary triage tool (TRIM).

METHODS

In this retrospective study, 333 of 457 children (≤ 18 years) with MSK symptoms (2020-2024) had ≥ 6 months follow-up and were included. Clinical and laboratory data from the first encounter were analyzed. Predictors were identified using Firth logistic regression and incorporated into weighted subscores for leukemia (TRIM-L) and neuroblastoma (TRIM-N). Performance was assessed using AUC and predictive metrics with bootstrap internal validation.

RESULTS

Malignancy was diagnosed in 9/333 patients (2.7%; leukemia 1.5%, neuroblastoma 1.2%). Independent indicators of leukemia were splenomegaly, hemoglobin < 8.5 g/dL, and platelet count < 150 000/μL. Independent indicators of neuroblastoma were lymphadenopathy, weight loss, and LDH > 400 U/L. Discrimination was high (AUC: leukemia 0.996; neuroblastoma 0.963). Using a cutoff ≥ 2, TRIM-L achieved 100% sensitivity, 97% specificity, and 100% NPV; TRIM-N achieved 75% sensitivity, 97.3% specificity, and 99.7% NPV.

CONCLUSIONS

Splenomegaly, cytopenias, lymphadenopathy, elevated LDH, and weight loss were key early indicators of malignancy in children presenting with MSK symptoms. The internally derived TRIM score showed excellent negative predictive value and may assist early triage; however, external validation is required before clinical application.