Background: Detection of measurable residual disease (MRD) is an important biomarker to direct treatment for adult patients with acute lymphoblastic leukemia (ALL). Currently, multiparameter flow cytometry (MFC) is standard of care and can detect residual leukemia cells in 1 out of 100,000 cells (1 × 10−5). Routine use of next-generation sequencing (NGS) assays with higher levels of sensitivity may improve disease-free survival and overall survival in adult ALL patients. The clonoSEQ MRD assay (Adaptive Biotechnologies Corporation, USA) can detect MRD at a sensitivity of 1 × 10−6 . In this study, we compare the detection of MRD by clonoSEQ assay and MFC in adult ALL patients at a single institution. Methods: We performed a retrospective study evaluating 92 adult patients with B- or T-ALL admitted to Norris Comprehensive Cancer Center between 2014 and 2022. Patients were excluded if they did not achieve complete remission (N=9) or if they presented with prior relapse (N=27). Median time to first complete remission (CR1) was 2.38 months. Among our cohort (N=56 patients), all patients underwent pre-treatment bone marrow sampling which was used for Clonoseq assay. 52 patients had contemporaneous MFC which was used for comparative analysis. MRD on MFC was defined as residual leukemia cells at a sensitivity of 1 × 10−5. MRD on Clonoseq assay was evaluated at 1 × 10−6; residual cells under this limit were considered MRD−. Results: Demographics are summarized in the table. Median time to follow-up from MRD assessment was 18.0 months. 10 patients relapsed (17.9%), 3 patients died (5.4%), and 20 patients (35.7%) underwent allogeneic transplant during first remission. Among 52 patients analyzed by both MFC and NGS at CR, 26 (50%) were concordantly MRD− by NGS and NGF, 13 (25%) were discordantly MRD− by NGF but MRD+ by NGS, and 13 (25%) concordantly MRD+ by NGS and NGF. Residual cell levels between NGS and NGF were strongly correlated (R-Squared = 0.6744, p<0.001), however, MRD status was significantly different between NGS and NGF (p<0.001 using Fisher’s Exact Test). This difference was likely driven by disparities in the limits of detection between both methods. Conclusions: Assessment of MRD using the clonoSEQ assay over a standard MFC-based approach adds crucial prognostic information among adult ALL patients who achieved CR.

¹Kruskal-Wallis rank sum test. ²Fisher's Exact Test for Count Data.