Background:Measurableresidual disease (MRD) is an established prognostic biomarker in B-cell acute lymphoblastic leukemia (B-ALL) with implications for decisions around consolidation, hematopoietic stem cell transplant (HSCT), and MRD-directed therapies. However, MRD utility is influenced by test timing, methodology, and sensitivity. While national guidelines have provided recommendations regarding each of these factors, the real-world patterns of MRD testing have not been well described. This study examined real-world MRD testing modalities and patterns, as well as associated outcomes, among US patients (pts) in Philadelphia-negative (Ph-ve) B-ALL.

Methods: This longitudinal, retrospective cohort study used patient-level data from the US-based, electronic health record-derived, deidentified Flatiron Health Research Database and included pts with a diagnosis of Ph-ve B-ALL diagnosed between January 1, 2014, and September 30, 2024 who had blinatumomab exposure. MRD testing details were collected serially from date of diagnosis onward. Descriptive analyses are reported on baseline pts and clinical characteristics by MRD testing results and timing, early testing occurring <3 months from first-line (1L) start date, or late testing occurring >3 months from 1L start date. Associations of MRD testing results with real-world overall survival (rwOS ) were assessed using Kaplan-Meier methods and adjusted Cox proportional hazards models controlling for age at 1L, race/ethnicity, socioeconomic status, stage, ECOG status, cytogenetic/molecular disease risk, insurance status, and allogeneic stem cell transplant (allo-HSCT) receipt.

Results: Among 585 pts with Ph-ve B-ALL, 362 (61.8%) underwent MRD testing. A total of 1,666 MRD tests were performed, with a median (interquartile range [IQR]) of 3 (2-6) tests per pts. The median time (IQR) from initiation of 1L therapy to the first MRD test was 78 (41-165) days. Testing methods included flow cytometry (45%) and PCR/NGS (43%), with 44% of tests having a documented sensitivity of ≤10⁻⁶.

The majority of MRD-tested pts initiated 1L therapy after 2018, with 42% treated between 2018 and 2021 and 33% from 2022-2024. 1L therapies included multidrug chemotherapy in 45% (162), blinatumomab combinations in 41% (150), inotuzumab combinations 2.5% (9), and other or clinical trial regimens in 11% (41). Allo-HSCT and chimeric antigen T-cell therapy were seen in 44% (161) and 6.9% (25) of pts respectively during the course of their treatment.

Overall, 241 (66.6%) pts ever had an MRD+ve result, while 121 (33.4%) had only undetectable MRD (uMRD). MRD status was associated with overall survival (OS). Pts with persistently MRD+ve disease had significantly shorter OS (25.5 months; 95% CI: 14.4-47.9; P = 0.001) compared to those with uMRD (49.0 months) or those with any MRD+ve test (49.5 months). Pts who converted from MRD-ve to MRD+ve had a shorter rwOS (44.8 months) compared to those who converted from MRD+ve to MRD-ve (56.0 months), although this difference was not statistically significant. Early vs late MRD testing showed no difference in rwOS (95% CI), 44.0 months (38.7, _) vs 41.1 months (35.6, 52.3) (P = 0.9) respectively.Conclusions: This real-world analysis demonstrates that while most pts with Ph-ve B-ALL receive MRD testing in accordance with clinical guidelines, testing patterns remain variable. Differences in timing, frequency, methodology, and sensitivity highlight the lack of standardization in practice. Persistent MRD positivity was associated with significantly worse survival, reinforcing its role as a critical prognostic marker, with a trend suggesting a prognostic impact of MRD conversion patterns. Timing of MRD testing (early vs late) was not associated with OS, suggesting a need for deeper evaluation of the optimal testing window. These findings support further investigation into whether targeted interventions in persistently MRD+ve pts can improve survival.

This content is only available as a PDF.
2025
Sign in via your Institution