Is Immunoglobulin/T-Cell Receptor Measurable Residual Disease Monitoring Truly Informative in Ph+ Acute Lymphoblastic Leukemia?

Measurable residual disease (MRD) negativity represents a primary goal in adult Ph+ acute lymphoblastic leukemia (ALL) patients’ management. MRD assays must be sensitive and specific. Digital droplet PCR (ddPCR) can overcome some RQ-PCR limitations. The aims of the study were to carry out a| [/i] BCR||ABL1-based MRD quantification; i[/i] comparison of MRD concordance rate between immunoglobulin/T-cell receptor clonal gene rearrangements (IG/TR) and BCR||ABL1; ii[/i] correlation with biological features. Samples from 56 adults enrolled in the phase III frontline GIMEMA LAL2820 trial for adult Ph+ ALL were collected at the end of induction (day +70) and during the consolidation (day +133). At diagnosis, patients underwent a screening for the identification of the predominant IG/TR gene rearrangement by PCR and/or NGS, and of the IKZF1plus signature by multiplex ligation-dependent probe amplification. BCR||ABL1-based MRD was performed by RQ-PCR and ddPCR, and by ddPCR for IG/TR. The concordance rate was evaluated comparing the results obtained by RQ-PCR and ddPCR for BCR||ABL1, and by BCR||ABL1 RQ-PCR and IG/TR ddPCR. BCR||ABL1-based MRD comparison by RQ-PCR and ddPCR showed a correlation degree of R2=0.99 and a concordance of 60%, with discordances falling mostly in cases with low MRD levels. As for IG/TR monitoring, only 47/56 (83.9%) cases were evaluable (lack of a reliable marker for the others). At day +70, the overall concordance was 55.3%| 25/47 (53.1%) cases were BCR||ABL1pos| 10 were IG/TRpos, 1 IG/TRPNQ and 14 IG/TRneg; 18/47 (38.3%) cases were BCR||ABL1neg| 16 were IG/TRneg, and 2 were IG/TRPNQ; finally, 4/47 (8.5%) cases were BCR||ABL1PNQ of which 1 was IG/TRpos and 3 were IG/TRneg. The concordance rate was similar between the experimental (EA) and the control (CA) trial arm (58.1% vs 43.7%), IKZF1plus vs IKZF1 WT/IKZF1 loss (55.5% vs 51.7%), and p190 vs p210-p190/p210 (48.5% vs 64.3%). At day +133, the overall concordance was 44.4%| 13/27 (48.1%) cases were BCR||ABL1pos| 2 were IG/TRpos, 1 IG/TRPNQ and 10 IG/TRneg; 10/27 (37%) cases were BCR||ABL1neg and IG/TRneg; finally, 4/27 (14.9%) cases were BCR||ABL1PNQ and IG/TRneg. Eight/17 (47%) and 4/10 (40%) cases in the EA and in the CA were concordant, respectively| notably, 1/2 IG/TRpos in the CA experienced a hematologic relapse. A lower concordance was observed in the IKZF1plus (33.3%) vs IKZF1 WT/IKZF1 loss (53.3%); contrariwise, at this timepoint, we observed a significantly higher concordance between p190 (62.5%) compared to p210-p190/p210 (18.2%) (p=0.047). The overall good correlation for BCR||ABL1-based MRD evaluation methodologies was confirmed; IG/TR MRD monitoring was feasible in 83.9% of cases. The concordance rate between BCR||ABL1 and IG/TR, though suboptimal, is in line with previous reports; a significant correlation with the fusion type at day +133 was found. The high rate of discordances among the 2 markers suggests that IG/TR may not be a reliable MRD marker in adult Ph+ ALL.