The measurement of minimal residual disease (MRD) in bone marrow serves as a significant predictor of relapse in acute lymphoblastic leukemia (ALL). MicroRNAs, which are secreted in urine and saliva, typically reflect the cellular state and function, thereby demonstrating potential as non-invasive biomarkers for predicting MRD status. This study aims to evaluate the utility of circulating microRNAs in saliva and urine as MRD biomarkers in pediatric patients with ALL. The study cohort comprised 106 patients under 18 years of age diagnosed with ALL at a national pediatric referral hospital in Lima, Peru. Saliva and urine samples were collected on day 15 of induction chemotherapy. Patients were categorized into high-risk relapse (HRR) and standard/intermediate-risk relapse (SIRR) groups based on their day-15 MRD status. Small RNA sequencing was conducted on six paired samples, while RT-qPCR was performed on 23 paired samples to identify and validate differentially expressed microRNAs. A total of thirty microRNAs were differentially expressed in saliva, and two in urine, both of which were downregulated. In saliva, 22 microRNAs showed significant upregulation. miR-1246, miR-223-3p, and miR-1290 exhibited the highest levels of upregulation in HRR patients (log2 fold change = 4.00, 3.95, and 3.73, respectively). Validation confirmed the upregulation of miR-223-3p (log2 fold change = 1.15, p = 0.034). miR-223-3p, both alone and in combination with miR-1290 and miR-1246, demonstrated optimal performance in distinguishing HRR from SIRR patients (AUC = 0.68, 95% CI: 0.52-0.84; AUC = 0.69, 95% CI: 0.53-0.84, respectively). These preliminary findings indicate that miR-223-3p, whether used independently or in conjunction with miR-1246 and miR-1290, may serve as promising non-invasive MRD biomarkers in pediatric ALL. Further validation in larger patient cohorts is necessary to corroborate these results.
Pando-Caciano et al. (Thu,) studied this question.