Validating Low Flow Cytometry Events with Next-Generation Sequencing for Precise Detection of Central Nervous System Involvement in Pediatric Acute Lymphoblastic Leukemia

Backgrounds

Acute lymphoblastic leukemia (ALL) is the most common pediatric cancer, and central nervous system (CNS) involvement at diagnosis is a critical prognostic factor. Although flow cytometry has been utilized for the detection of CNS involvement at ALL diagnosis, the gold standard remains morphological analysis by traditional cytology due to the potential false positives from flow cytometry. However, cytology may be associated with false negatives, potentially leading to inadequate CNS-directed treatments and an increased risk of relapse. Our study aims to achieve more sensitive CNS involvement detection by utilizing both 8-color fluorescent flow cytometry and next-generation sequencing (NGS) techniques to detect immunoglobulin heavy chain (IgH) rearrangement clonotypes. The NGS technique will also be used to validate the results from flow cytometry and to overcome the issue of false positives.

Methods

Patients with newly diagnosed or relapsed B-cell precursor ALL were enrolled in the study. Their bone marrow (BM) samples at diagnosis or relapse were sent for leukemia-associated immunophenotypes (LAIPs) analysis by flow cytometry and IgH rearrangement clonotypes detection by NGS. The patient-specific LAIPs and IgH rearrangement clonotypes were used to analyze the patients' first cerebrospinal fluid (CSF) samples to define CNS involvement. Clinical information, including CNS cytology reports, was collected prospectively.

Results

A total of 170 patients with newly diagnosed or relapsed B-cell precursor ALL were enrolled in the study. Among them, 101 patients were male, and 69 were female. The median age was 5.2 years old (range: 0.4-23.0 years old). According to traditional CSF cytology reports, 31 patients (18.3%) had positive cytology results, while 139 patients (81.7%) had negative cytology results.

If a positive flow cytometry result is defined as having more than 10 events, 27 patients (15.9%) had positive CSF flow cytometry reports, while 143 patients (84.1%) had negative CSF flow cytometry reports. The concordance rate between cytology and flow cytometry was 88.2%, with no statistically significant difference (P = 0.99). If a positive flow cytometry result is defined by the presence of any events, 68 patients (40.0%) had positive CSF flow cytometry reports, while 102 patients (60.0%) had negative CSF flow cytometry reports. The concordance rate between cytology and flow cytometry was 72.3% under this definition. Flow cytometry was found to be statistically more sensitive than traditional cytology (P < 0.01).

BM NGS was performed on 127 patients, and IgH rearrangement clonotypes were identified in 115 patients (90.6%). Among these 115 patients, 10 had inadequate CSF samples for further analysis, 31 had failed library construction due to low sample DNA, 2 were pending analysis, and 72 had IgH rearrangement data obtained from their CSF samples. The CSF clonotype results matched the BM clonotype results in 20 patients, indicating CNS involvement detected by NGS. The remaining 52 patients did not share the same clonotypes among their BM and CSF samples, indicating no CNS involvement. The concordance rate between flow cytometry and NGS in these 72 patients was 84.7%, with no statistically significant difference (P = 0.23).

Among 72 patients having their CSF samples examined by cytology, flow cytometry and NGS, 17 patients (23.6%) had discordant reports. Thirteen had negative CSF cytology and positive CSF flow cytometry reports, while 4 had positive CSF cytology and negative CSF flow cytometry reports. Among 13 patients with negative cytology and positive flow cytometry reports, all had fewer than 10 events on flow cytometry, and 8 (61.5%) had positive NGS reports. Among 4 patients with positive cytology and negative flow cytometry reports, only one (25.0%) had positive NGS report.

Conclusion

The sensitivity of flow cytometry and traditional cytology is similar if a positive flow cytometry result is defined by events of more than 10. However, if a positive flow cytometry result is defined by any events, the sensitivity of flow cytometry is significantly higher than that of traditional cytology and similar to that of the NGS technique. In patients with discordant cytology and flow cytometry reports, NGS helps in validation and supports the flow cytometry results in over 60% of cases, even when the flow cytometry events are fewer than 10.