Droplet Digital PCR Analysis for Diagnosis and Minimal Residual Disease Monitoring in Adult Ph+ Acute Lymphoblastic Leukemia

Introduction.BCR-ABL1 tyrosine kinase inhibitors (TKIs) are considered an important component of treatment for adult patients affected by Philadelphia-positive (Ph+) acute lymphoblastic leukemia (ALL). In fact, recent studies reported that treating Ph+ ALL with the combination of imatinib and multi-agent chemotherapy improved the overall outcome. Currently, no data are available on the impact of TKIs on minimal residual disease (MRD) in Ph+ ALL. In fact, although the real-time quantitative PCR (RQ-PCR) method, usually employed for monitoring the BCR-ABL1 residual transcript, is sensitive and easy to perform, it lacks a full standardization and international quality validation. Here, we describe a highly sensitive and reproducible droplet digital PCR (ddPCR) test to monitor BCR-ABL1 transcript level in Ph+ ALL.

Methods.BCR-ABL1 expression analysis by ddPCR was performed in twenty-two newly diagnosed adult Ph+ ALL patients.The diagnosis was confirmed by qualitative RT-PCR specific for the BCR-ABL1 p190 fusion gene detection. ddPCR experiments were successfully performed in all twenty-two patients at the onset; several follow-up points were evaluated in thirteen patients. ddPCR experiments were performed using primers and probes specific for BCR-ABL1 p190. GUSB was used as control gene. Fifty ng and 750 ng of cDNA templates were used for the onset and for the post-treatment samples, respectively. To increase the limit of detection (LOD), three replicates were run for the post-treatment samples. ddPCR experiments were performed by Bio-Rad's QX200 system and ddPCR data were analyzed with QuantaSoft analysis software (version 1.7.4). Target concentration was expressed as BCR-ABL1 copies/mg.

Results. First, we defined the LOD of the BCR-ABL1 p190 ddPCR system, a 10-fold dilution series (10⁰, 10^-1, 10^-2, 10^-3, 10^-4, and 10^-5) of a pool of p190 positive patients using a diluent-pool of healthy volunteers. This analysis showed remarkable linearity, trueness, and precision down to 10^-5. After converting to log-log scale, linear regression showed no concentration-dependent bias, and R² equaled 0.996. Because the negative samples showed no background, even the detection of a single droplet per well was considered a positive result. The median concentration of the BCR-ABL1 transcript at the onset was 233.8 (min 3.24 - max 1744) x 10³BCR-ABL1 copies/mg. Concerning the analysis of follow-up samples, among the thirty-four points that were negative to qualitative nested RT-PCR, twenty-three (68%) resulted to be positive by ddPCR analysis, with a median concentration of 44.95 (min 0.27 - max 573.3) BCR-ABL1 copies/mg. Follow-up points that were negative in ddPCR remained negative even when the experiments were repeated increasing the depth of the analysis, evaluating a total quantity of 4.5 mg of RNA.

Conclusions. This study indicates that, as compared to RQ-PCR, ddPCR increases the depth of the quantitative analysis of BCR-ABL1 p190 fusion transcript by allowing the evaluation of larger amounts of RNA. Moreover, our preliminary data revealed that the amount of the BCR-ABL1 fusion transcript at diagnosis is heterogeneous and that the ddPCR is much more sensitive than nested qualitative RT-PCR analysis, as the 68% of samples negative to nested PCR during the follow-up resulted to be positive by ddPCR. Therefore, we suggest that ddPCR represents a precise, sensitive and rapid method for both diagnosis and MRD monitoring of Ph+ ALL patients.