Currently, the field of pediatric leukemia treatment faces numerous challenges, particularly in predicting relapse risk and developing treatment strategies for relapsed and refractory patients. At the upcoming 29th European Hematology Association (2024 EHA) Annual Meeting, several studies from Professor Xiaofan Zhu’s team from the Institute of Hematology & Blood Diseases Hospital, Chinese Academy of Medical Sciences have been selected. Two studies on pediatric leukemia delve into these critical issues, providing important academic references for further optimizing pediatric leukemia diagnosis and treatment. To better understand these latest research advances, this issue of "Oncology Frontier - Hematology Frontier" features a detailed interpretation by Professor Xiaofan Zhu.

ALL Domain – Optimizing MRD Monitoring Strategies for Ph+ ALL Patients

Abstract Number: P403

Monitoring MRD in Pediatric Ph+ Acute Lymphoblastic Leukemia: Correlation and Prognostic Significance between Multiparameter Flow Cytometry and RT-qPCR

Research Background

In pediatric Philadelphia chromosome-positive acute lymphoblastic leukemia (Ph+ ALL), real-time quantitative polymerase chain reaction (RT-qPCR) targeting the BCR::ABL1 gene and multiparameter flow cytometry (MFC) are the most commonly accepted methods for monitoring minimal residual disease (MRD). However, these methods sometimes yield inconsistent results, posing challenges for clinical decision-making.

Research Objectives

This study aims to examine the consistency of MFC and RT-qPCR in detecting BCR::ABL1 and their prognostic value in pediatric Ph+ ALL.

Research Methods

The study included untreated Ph+ ALL patients under 18 years, using the CCCG-ALL-2015 protocol. RT-qPCR followed European Against Cancer (EAC) guidelines with a sensitivity of 10^-4 (0.01%). Immunophenotyping was performed using FACSCanto™ II flow cytometer (BD Biosciences), analyzing at least 5×10^6 nucleated cells with a sensitivity of 10^-4 (0.01%). Patients with multiple inconsistent MRD results were defined as CML-like.

Research Results

The overall inconsistency rate between flow cytometry and RT-qPCR MRD detection was 25.6% (56/219), peaking at D19 (45.8%, 27/59), and gradually decreasing with treatment. All inconsistent samples fell into two categories: (1) MFC-negative and RT-qPCR-positive with at least 0.1%, (2) RT-qPCR MRD significantly higher than MFC, ranging from 2.5 log to 8.2 log (median 3.3 log).

Approximately 27.1% of patients were classified as CML-like (16/59). Reviewing a patient with BCR::ABL1 p190 fusion gene and available FISH results revealed BCR::ABL1-carrying late myelocytes, indicating myeloid involvement. There were no statistical differences in clinical characteristics between CML-like and typical ALL patients. Patients with p210 transcripts showed better prognosis. In practical and intent-to-treat analyses, MFC MRD≥0.01% at TP1 was independently associated with lower cumulative incidence (CIR) and event-free survival (EFS). RT-qPCR MRD at any assessment point did not show prognostic value.

Research Conclusion

Identifying patients with CML-like characteristics based on MFC and RT-qPCR inconsistencies is feasible. The study data suggest that only MFC MRD has prognostic significance, indicating that clinical decisions should primarily reference MFC results

Abstract Number: P419

Impact of NGS MRD and Clonal Evolution on Clinical Prognosis in Pediatric Ph+ ALL

Research Background

MRD is a crucial prognostic indicator for assessing ALL relapse risk and predicting overall survival (OS). Next-generation sequencing (NGS)-based MRD detection methods offer higher sensitivity for low-level MRD quantification, potentially improving ALL prognosis evaluation. However, the prognostic prediction of NGS MRD in pediatric Ph+ ALL has not been fully validated globally or in China.

Research Objectives

This study aims to evaluate a highly sensitive NGS MRD detection method (sensitivity 10^-6) for pediatric Ph+ ALL patients and explore its correlation with RT-PCR and FCM. Additionally, it investigates the value of NGS in detecting clonal evolution for Ph+ ALL prognosis assessment.

Research Methods

From May 2015 to April 2020, 55 Ph+ ALL patients were enrolled. All patients received prospective CCCG-ALL-2015 protocol treatment (ChiCTR-IPR-14005706). MRD was assessed at initial diagnosis and end of induction therapy (D46) using fluorescence quantitative PCR (RQ-PCR), flow cytometry (FCM), and NGS-based immune sequencing. Median follow-up was 50 months. Prognostic analysis included 48 patients with bone marrow samples at diagnosis and D46.

Research Results

Among 49 patients, 48 (98%) identified at least one malignant clonal rearrangement in the bone marrow. After induction therapy (46 days), 16 patients (33%) were negative for MRD on FCM, PCR, and NGS, and 10 patients (21%) were positive on all three methods. MRD assessment by PCR and NGS was inconsistent in 42% (20/48) of patients, with 10 PCR+/NGS- and 10 PCR-/NGS+ cases.

Patients achieving NGS MRD negativity at D46 had better relapse-free survival (RFS) trends compared to NGS MRD-positive patients (2-year RFS: 100% vs. 77%; P=0.01). Patients with FCM negativity at D46 also showed better RFS compared to FCM-positive patients (2-year RFS: 97% vs. 60%; P<0.01). One FCM-/NGS+ patient relapsed at 6 months. In contrast, PCR MRD was a poorer predictor of RFS.

NGS MRD predicted 2-year RFS more accurately than 5-year. Patients with NGS MRD negativity at D46 had better 2-year RFS (100% vs. 77%; P=0.01), but similar 5-year progression-free survival (PFS) (68% vs. 54%; P=0.3) (Fig 2A). This may be due to residual tumor cells below the detection limit causing relapse. Patients with more evolutionary clones from traceable malignant clones at D46 had higher 2-year relapse risk (RFS: 76% vs. 100%, P=0.03) (Fig 2B). Patients with lower pre-treatment malignant clone cell ratios had better 2-year (100% vs. 79%; P=0.02) and 5-year (85% vs. 36%; P<0.01) RFS (Fig 2C). We identified some risk and protective tumor clone genes affecting risk classification, such as IGHV1 expression in malignant clones being associated with lower relapse risk (P=0.02) (Fig 2D).

Research Conclusion

NGS-based MRD assessment provides accurate prognostic predictions for pediatric Ph+ ALL and offers insights into other clinical prognostic factors.

Expert Commentary

Professor Xiaofan Zhu: In pediatric ALL, Ph+ ALL accounts for about 5%. Before the advent of tyrosine kinase inhibitors (TKIs), the treatment outcomes for Ph+ ALL were often unsatisfactory. The introduction of TKIs significantly improved treatment outcomes, reducing the need for hematopoietic stem cell transplantation. Precise MRD monitoring during TKI therapy is crucial. Effective monitoring can accurately assess residual disease, guiding subsequent treatment decisions, such as whether and when to implement hematopoietic stem cell transplantation.

In clinical practice, we observed that some patients’ flow cytometry results showed MRD negativity, while PCR detection of fusion genes remained positive. This phenomenon attracted widespread attention, particularly regarding the reasons for persistent positivity and its potential impact on relapse risk. Internationally, the concept of CML-like Ph+ disease has been proposed, suggesting that besides Ph+ leukemic cells, other tissue cells may also carry this marker. Despite effective clearance of Ph+ leukemia cells, residual presence in other tissues can be detected by molecular biology methods.

Whether these residuals cause leukemia relapse remains a critical issue. Parents are concerned about how to formulate subsequent treatment strategies when flow cytometry results are negative but fusion genes remain positive, whether further transplantation or treatment enhancement is needed. Our team conducted two comparative studies on MRD monitoring methods to provide scientific and reasonable evidence for clinical decisions and patient management.

The first study (Abstract No. P403) examined the consistency between MFC and RT-qPCR in detecting BCR::ABL1 and analyzed their application value in pediatric Ph+ ALL prognosis. The results showed that detected residuals were not from ALL cells but from myeloid or other systems. Thus, molecular residuals usually had weak relevance to relapse. In another study (Abstract No. P419), we used NGS technology to monitor MRD in Ph+ ALL and compared it with flow cytometry results. The findings indicated that NGS technology had advantages over traditional flow cytometry or PCR in monitoring relapse risk, detecting relapse signs about three months earlier, and demonstrating higher sensitivity. Additionally, most relapse clones were similar to initial treatment clones, providing new perspectives on relapse mechanisms.

These studies offer robust tools for predicting treatment outcomes and monitoring relapse in Ph+ ALL patients and provide important references for clinicians in strategy formulation. As medical technology advances, we hope to further optimize MRD monitoring strategies, benefiting Ph+ ALL patients.