Background: NGS for IG/TCR rearrangements is a highly sensitive tool for detecting MRD in ALL. NGS MRD negativity with low-level detectable PCR for BCR::ABL1 is frequently observed in patients (pts) with Ph+ ALL. However, the relationship of discordant NGS and PCR with fusion transcript type (p190 vs. p210) and its long-term clinical implications have not been fully elucidated.

Methods: We retrospectively analyzed pts with Ph+ ALL who underwent paired NGS for IG/TR and PCR for BCR::ABL1. Pts with known or suspected CML lymphoid blast phase were excluded. Only pts with negative NGS MRD (defined as absence of any detectable residual sequences) with a simultaneous paired PCR for BCR::ABL1 were included. Negative PCR was defined as absence of quantifiable transcripts at a level of 10-4. Frequency and patterns of NGS-/PCR+ discordance by BCR::ABL1 transcript type were evaluated. Cumulative incidence of relapse (CIR), relapse-free survival (RFS) and overall survival (OS) were also evaluated among newly diagnosed pts.

Results: We identified 187 pts with Ph+ ALL with paired NGS for IG/TR and PCR for BCR::ABL1 (896 total paired samples). The median age at diagnosis was 49 years (range, 12-85 years); 74% of pts had ≥1 additional chromosomal abnormalities. Overall, 134 pts (72%) were p190, 39 pts (21%) were p210, and 14 pts (7%) had unknown transcript type.

Fifty-two pts (28%) had at least 1 discordant MRD assessment (i.e. NGS-/PCR+), suggesting multilineage involvement of BCR::ABL1. Overall, 27 pts (14%) were NGS-/PCR+ with a PCR value of 0.01-0.1%, 17 pts (9%) at a level of 0.1-1%, and 7 pts (4%) at a level of ≥1%; an additional 1 pt (1%) had non-quantifiable PCR transcript (sensitivity of 0.1%). Among 14 total samples that were NGS-/PCR+ at a level of ≥1%, all were MRD-negative by flow cytometry at the same timepoint; 3 of these samples (21%) had detectable FISH for BCR::ABL1 (concomitant PCR values of 26%, 4% and 2.5%).

The rate of NGS-/PCR+ discordance was higher in pts with p190 vs p210 transcripts (34% [45/134] vs 18% [7/39], respectively; P=0.09). NGS-/PCR+ at a level of ≥0.1% was observed in 22 pts (16% [22/134]) with p190 and 3 pts (8% [3/39]) with p210. All 7 pts with PCR values ≥1% had p190 Ph+ ALL.

Among the 52 NGS-/PCR+ pts, 3 discontinued their TKI during the study period in the setting of low-level PCR-positivity (all of whom had the p190 transcript). In pt #1, the PCR increased from 0.03% to 17.68% with TKI discontinuation; after resumption of the TKI, the PCR declined to 0.38%. In pt #2, the PCR increased from 0.03% to 28.94% with TKI discontinuation, and declined to 1.80% after resumption of the TKI. Both pts remained NGS MRD negative, and neither had laboratory or morphological evidence of CML. PCR values >10% in the setting of NGS MRD negativity were only observed in these 2 pts and only while off TKI therapy. In pt #3, the PCR increased from 0.03% to 0.97% with TKI discontinuation; after resumption of the TKI, the PCR declined to 0.23%.

To better understand the clinical impact of NGS/PCR discordance, we evaluated the clinical outcomes among 80 pts with newly diagnosed Ph+ ALL who had paired NGS and PCR assessment within the first 6 months of frontline therapy. With a median follow-up of 26 months (range, 4-113 months), the estimated 3-year CIR for pts who were NGS-/PCR- (n=52) vs NGS-/PCR+ (n=28) were 20% and 19%, respectively (P=0.81). 3-year RFS and OS rates were also similar in these 2 groups (RFS: 74% vs 78%; P=0.59 and OS: 88% vs 91%; P=0.68, respectively). When further stratified by degree of NGS/PCR discordance, similar outcomes were observed in those who were PCR-positive at a level of 0.01-0.1% vs ≥0.1% (3-year OS: 93% vs 88%, respectively; P=0.65).

Conclusion: Multilineage involvement of Ph+ ALL (i.e. NGS-/PCR+) was observed in 28% of pts with Ph+ ALL, with higher frequency in pts with p190 transcripts. While most discordant cases have very low-level PCR values, BCR::ABL1 transcripts ≥0.1% were observed in 14% of pts overall. PCR values ≥1% were only seen in pts with p190 Ph+ ALL, arguing against the presence of underlying chronic myeloid leukemia. In newly diagnosed pts who achieved NGS MRD negativity, detectable PCR for BCR::ABL1 (regardless of the PCR value) had no impact on relapse or survival outcomes. However, rapid rise in PCR values could be seen in NGS-/PCR+ pts who underwent TKI discontinuation, arguing that indefinite TKI duration may be appropriate for pts with this finding.

This content is only available as a PDF.
2025
Sign in via your Institution