A Novel 12-Color 2-Tube Next Generation Flow Approach Enables Sensitive Detection of Measurable Residual Disease in Virtually All T-Cell Acute Lymphoblastic Leukemia Patients

T-cell acute lymphoblastic leukemia (T-ALL) accounts for approximately 15% of childhood ALL and 25% of adult ALL. At present, measurable residual disease (MRD) by conventional flow cytometry and/or molecular methods is the most important single prognostic factor used for risk-stratification of T-ALL patients after starting therapy. Currently however, molecular MRD quantification is not always possible due to the lack of clonal IG/TR rearrangements in a substantial fraction (up to 25%) of T-ALL patients (e.g., early T cell precursor ALL). Therefore, there is an urgent clinical need for standardized and sensitive flow cytometric (MFC) MRD assessment in T-ALL.

Here we describe a novel 12-color 2-tube T-ALL MRD panel and next-generation flow (NGF) approach for highly-sensitive MRD monitoring in T-ALL and its performance in a multicenter study on 96 pediatric and adult T-ALL patients (ETP-ALL, n=18; non-ETP-ALL, n=76; undetermined, n=2) involving 7 centers of the EuroFlow Consortium.

Follow-up bone marrow (BM, n=109) and peripheral blood (PB, n =17) samples from T-ALL patients were analyzed for MRD by NGF based according to EuroFlow SOPs for cytometer set-up, calibration and sample staining, available at www.EuroFlow.org. For the staining of samples an innovative, 2-tube, 12-color EuroFlow antibody panel was used, consisting of 15 and 13 markers, including a backbone of 8 shared antibodies: CD16+NKp80, cytoplasmic (Cy), the surface membrane (Sm) CD3, CD5, CD7, CD34, CD45 and CD45RA, plus CD6, CD38, and TdT, supplemented with a single-fluorochrome CD117+CD13+CD33 combination in tube 1 and CD2, CD4, CD8 and CD99 in tube 2, respectively. Stained samples were measured in FACS Lyric or LSR Fortessa flow cytometers (BD, San Jose, CA) at a limit of detection (LOD) level of <10^‑5 to confirm MRD-negativity.

The new NGF method solved all major challenges in MCF-based MRD detection in T-ALL, related to the discrimination of leukemic cells (blasts) from: 1) normal T-cells and NK-cells, and particularly the NK-cell subset that expresses CyCD3, using the combination of CD16 and NKp80 as NK-cell exclusion markers, not expressed in T-ALL, 2) as well as discrimination from some myeloid cell types, such as dendritic cells, CD5⁺ naïve B-cells and/or CD7⁺CD34⁺ hematopoietic precursors. For ETP-ALL patients, discrimination of T-ALL blasts from normal T-cells was mostly based on their differential expression of SmCD3, CD5, CD34 and CD45 in both tubes 1 and 2, followed by the combination of CD117+CD13+CD33 and CD8 in their respective tubes. For non-ETP-ALL patients, CD6 and CD99 were of most relevance. For the discrimination between blasts and NK cells, CD16+NKp80, followed by CD45, CD5, CD45RA in all patients were the most powerful markers, in addition to TdT and CD8 in ETP-ALL patients, and CD34 and CD99 in the non-ETP-ALL group.

Once we compared the MRD levels detected with both tubes, virtually identical results were observed (R² = 0.99). For further validation of the new NGF approach, NGF-MRD results were compared with those obtained in routine diagnostics by the reference ASOqPCR MRD assay. Samples with insufficient cells acquired by MFC (below the required limit of detection, LOD) and those with MRD levels ≥10^-1 were excluded from this analysis. Overall, in a subset of 39 corresponding cases, our results showed a high concordance rate between NGF and ASOqPCR of 97.5% with only one -NGF-MRD-/ASOqPCR+ discrepant case presenting an MRD level at <5x10^-4. The correlation between NGF-MRD and PCR-MRD data expressed as R² equaled 0.81 and 0.72, respectively.

In summary, here we described and validated a new 12-color, 2-tube NGF-MRD EuroFlow approach that allows for highly-sensitive (<10^-5) standardized MRD detection in T-ALL which for the first time is applicable to virtually every pediatric and adult T-ALL patient. Importantly, the newly proposed markers and marker combinations significantly improved the reliable discrimination of the T-ALL blasts from normal cells both in ETP-ALL and non-ETP-ALL patients.