Ultra-Low BCR::ABL1 positivity by digital PCR does not improve post-transplant risk stratification beyond real-time PCR in Ph+ acute lymphoblastic leukemia post-transplant

Abstract

Background

Post-transplant relapse remains a major clinical challenge in Philadelphia chromosome–positive acute lymphoblastic leukemia (Ph + ALL). Real-time quantitative PCR (RQ-PCR) for BCR::ABL1 is the current standard for measurable residual disease (MRD) monitoring, whereas digital PCR (dPCR) offers substantially higher analytical sensitivity. Whether this increased sensitivity translates into additional prognostic value after allogeneic hematopoietic stem cell transplantation (allo-HSCT) remains unclear.

Methods

In this prospective study (NCT06211166), 270 patients with Ph + ALL were longitudinally monitored after allo-HSCT. MRD was assessed in parallel using dPCR, RQ-PCR, and MFC. Based on the first post-transplant MRD detection pattern, patients were categorized into four groups: double-negative (n = 80), dPCR–single-positive (n = 158), RQ-PCR–single-positive (n = 3), and double-positive (n = 29).

Results

The dPCR–single-positive pattern was the most prevalent MRD status, accounting for 58.5% of patients. dPCR positivity independently predicted subsequent MFC-MRD conversion (HR 9.56, P = 0.029), with a median lead time of 77 days. In addition, dPCR detected BCR::ABL1 positivity earlier than RQ-PCR, preceding subsequent hematologic relapse by a median of 64.5 and 91.5 days, respectively. However, the cumulative incidence of hematologic relapse (CIR), the primary endpoint of this study, did not differ significantly among the four MRD-defined groups (P = 0.60). Consistently, isolated dPCR positivity was not associated with inferior 2-year leukemia-free survival (LFS; P = 0.30) or overall survival (OS; P = 0.60).

Conclusions

Although dPCR detects molecular disease earlier and anticipates MFC-MRD by 2 months after allo-HSCT in Ph + ALL, isolated ultra-low–level BCR::ABL1 positivity does not impact relapse risk, LFS, or OS. Routine MRD monitoring with RQ-PCR plus MFC remains sufficient for prognostic stratification, while dPCR primarily provides an ultra-early signal to guide timely intervention rather than improving survival prediction.

Supplementary Information

The online version contains supplementary material available at 10.1007/s10238-026-02172-w.

Introduction

Philadelphia chromosome–positive acute lymphoblastic leukemia (Ph + ALL) is a high-risk hematologic malignancy defined by the presence of the BCR::ABL1 fusion transcript 1–3. The integration of tyrosine kinase inhibitors (TKIs) and allogeneic hematopoietic stem cell transplantation (allo-HSCT) has fundamentally reshaped the treatment pattern, resulting in a significant improvement in long-term survival results 4–12. However, post-transplantation recurrence is still the main cause of treatment failure, which highlights the urgent need for sensitive and reliable monitoring of measurable residual diseases (MRD) to achieve timely and pre-emptive treatment interventions 13–17.

At present, the reverse transcription-quantitative polymerase chain reaction (RQ-PCR) of BCR::ABL1 transcripts and multi-parameter flow cytometry (MFC) constitute the pillars of MRD evaluation in Ph + ALL 18–20. However, RQ-PCR is inherently limited by its non-specific and unstable amplification at its detection limit, which limits its ability to detect ultra-low molecular persistence 20, 21. At the same time, although MFC is valuable for identifying the evolution of immune phenotype diseases, it is easily affected by inter-operator variability and immune phenotype drift under the pressure of treatment selection 22–24. Overall, these restrictions may compress the effective “treatment window” between MRD detection and open hematological recurrence, thus reducing the opportunity to adapt to risk intervention in a timely manner.

Digital PCR (dPCR) has become a revolutionary technology, providing greatly enhanced analytical sensitivity for minimal residual disease (MRD) detection 25–27. Although dPCR shows incremental prognostic value in several high-risk leukemias, its clinical efficacy in Ph + ALL - especially in the environment maintained by tyrosine kinase inhibitors (TKI) after transplantation - is still controversial. The powerful inhibitory effect of TKI may induce a “clinical inertia” state, in which ultra-low-level molecular residues that can only be detected by dPCR will not be immediately transformed into obvious clinical recurrence28, 29. Therefore, a key unresolved question is whether dPCR positivity reflects short-term, clinically irrelevant molecular fluctuations, or defines a biologically meaningful and operable preemptive intervention window.

In order to solve these unresolved dynamics, we conducted a prospective study to systematically compare dPCR-based MRD monitoring with traditional RQ-PCR and MFC. Our main goal is to quantify the molecular lead time provided by dPCR and evaluate its independent prognostic value for survival outcomes within a multivariate framework. By describing different molecular response patterns, we try to improve post-transplant risk stratification and optimize the clinical management of Ph + ALL in the contemporary era of TKI-based treatment.

Method

Study design and participants

This was a prospective, single-cohort study registered at ClinicalTrials.gov (NCT06211166). Consecutive patients with Ph + ALL undergoing allo-HSCT at Peking University People’s Hospital were enrolled between January 2024 and September 2025. A total of 885 serial post-transplant bone marrow samples were collected longitudinally and analyzed in parallel for MRD assessment.Patients generally received Bu/Cy- or TBI-based conditioning, CsA–MMF–MTX–based GVHD prophylaxis, and post-transplant TKI maintenance according to Chinese consensus guidelines and as detailed in our previous reports11, 30.

Eligibility criteria included: (i) a confirmed diagnosis of Ph + ALL, defined by detection of the Philadelphia chromosome using conventional cytogenetics or FISH and identification of the BCR::ABL1 fusion gene by molecular testing; (ii) completion of allo-HSCT; and (iii) the presence of a trackable BCR::ABL1 transcript (p190 or p210 isoforms) verified by diagnostic molecular profiling and compatible with the institutional dPCR primer–probe panel.

Exclusion criteria were: (i) rare BCR::ABL1 transcript variants or cryptic rearrangements not targetable by the current dPCR surveillance panel, to minimize the risk of false-negative MRD results; (ii) patients with Ph + CML in blast crisis; and (iii) missing baseline molecular data or inadequate sample quality precluding dPCR assay development and validation.

Sample collection and processing

Bone marrow samples were collected at 1, 2, 3, 4.5, 6, 9, and 12 months after allo-HSCT and at 6-month intervals thereafter. To ensure methodological comparability, MFC, RQ-PCR, and dPCR analyses were performed simultaneously using the same bone marrow aspirate. Anticoagulated specimens were processed within 4 h in accordance with standardized institutional operating procedures, with first-pull aspirates preferentially used to minimize hemodilution. Total RNA was extracted using TRIzol reagent (Invitrogen, USA) and reverse-transcribed into complementary DNA using a commercially available kit (Invitrogen, USA).

MRD Assessment by RQ-PCR and dPCR

Primers and probes targeting the major and minor breakpoint cluster regions of BCR::ABL1 (p210 and p190 isoforms) and internal control ABL1 were referred to the report from Europe Against Cancer (EAC)31, 32 and TaqMan-based RQ-PCR was performed for longitudinal molecular MRD monitoring. cDNA which corresponding to 200-400ng RNA were added in each PCR reaction. The quantifiable limit of detection (LOD) for BCR::ABL1 by RQ-PCR is 2 copies. Each cDNA sample were performed duplicate or triplicate of PCR, and only the reproducible > = 2 copies of BCR::ABL1 were defined as positivity and the copy numbers were recorded. The ratio of BCR::ABL1 to ABL1 of a sample is the average BCR::ABL1 copies divided by the average ABL1 copies of all its PCR reaction, and MRD is defined as positivity when the ration is > 0.

Commercial one-step RT-PCR kit for dPCR detection of BCR::ABL1 rearrangement (p210 and p190 isoforms; Sniper, Suzhou, China) were used according to the manufacturer’s instructions, and ABL1 served as the internal control for quantification. dPCR was conducted on a DQ24-Dx digital PCR system (Sniper, Suzhou, China). For dPCR analysis, wells containing ≥ 20,000 accepted droplets were considered valid for subsequent quantification33. 1–1.5 ug RNA were added in each RQ-PCR reaction. Each RNA sample were performed duplicate or triplicate to ensure the summed ABL copies is higher than 200,000. The performance validation demonstrates that the limit of detection for BCR::ABL1 by dPCR is 1 copy, and > = 1 copy is defined as dPCR positivity. The ratio of BCR::ABL1 to ABL1 for dPCR is the sum of BCR::ABL1 copies divided by the sum of ABL1 copies. MRD is defined as positivity when the ratio is > = 0.0005%.

To ensure that the observed prognostic associations accurately reflected the natural history of molecular disease evolution, dPCR results were strictly blinded to treating clinicians and patients until study completion and database lock. In contrast, RQ-PCR and MFC results were accessible to clinical teams and incorporated into routine post-transplant management. RQ-PCR conversion triggered preemptive interventions, including TKI modification or escalation and adjustment of immunosuppressive therapy. In cases of MFC-MRD conversion, more intensive strategies were initiated, such as CD3-CD19 bispecific T-cell engager (BITE), inotuzumab ozogamicin (INO), chimeric antigen receptor T-cell (CAR-T) therapy, or donor lymphocyte infusion (DLI)34.

MRD assessment by MFC

Eight-color MFC was carried out on the BD FACSCanto II system. In view of the diversity of disease subtypes, MRD is evaluated using a combination of leukemia-related immunophenotype (LAIP) and different bloodline (DfN) strategies. The analytical sensitivity of this measurement reaches 0.01% 35.

Definition of molecular response patterns

Based on the concurrent longitudinal monitoring of BCR::ABL1 levels, patients were stratified into four distinct molecular response patterns according to the status of their first post-transplant molecular detection event:

1. Double Negative (DN): Persistently negative MRD detected by both dPCR and RQ-PCR throughout the follow-up period.

2. dPCR Single Positive (dPCR+/RQ-PCR−): Conversion to dPCR-MRD positivity as the initial molecular event while remaining negative MRD by RQ-PCR. This pattern represents the most common discordant profile.

3. RQ-PCR Single Positive (dPCR−/RQ-PCR+): Conversion to RQ-PCR-MRD positivity while remaining negative by dPCR-MRD. This pattern was observed as an atypical discordant profile in a minority of cases.

4. Double Positive (DP): Concurrent MRD positivity detected by both dPCR and RQ-PCR (dPCR+/RQ-PCR+) at the first occurrence of molecular detection.

For patients who progressed through these molecular stages, Lead Time was defined as the interval (in days) between the date of the first occurrence of a dPCR single positive event (dPCR+/RQ-PCR −) and the subsequent detection of: (i) RQ-PCR conversion to positivity (i.e., progression to DP status); (ii) MRD recurrence as detected by MFC; or (iii) Hematologic relapse (including morphologic bone marrow relapse or extramedullary disease involvement).

Endpoints and statistical analysis

The primary endpoint of this study was the 2-year cumulative incidence of relapse (CIR) after allo-HSCT, with non-relapse mortality (NRM) considered a competing risk. Hematologic relapse was defined by the presence of ≥ 5% blasts in the bone marrow, the reappearance of blasts in the peripheral blood, or the development of extramedullary disease (EMD). Secondary endpoints included the 2-year cumulative incidence of MFC-MRD conversion (defined as the first detection of MRD ≥ 0.01% by multiparameter flow cytometry after achieving MRD negativity), leukemia-free survival (LFS), and overall survival (OS). OS and LFS were estimated using the Kaplan–Meier method and compared using the log-rank test. CIR and MFC-MRD conversion were estimated using cumulative incidence functions and compared via Gray’s test, with subdistribution hazard ratios (HRs) calculated using Fine–Gray regression models.

For multivariable analyses (MVAs), variables with P < 0.10 in univariate analyses or those considered clinically significant (e.g., age, graft source, and molecular group) were incorporated into the final models. To address the monotone likelihood and quasi-complete separation arising from the absence of events in the double-negative (DN) or dPCR-negative groups (e.g., zero cases of MFC conversion or relapse), MVAs were performed using Firth’s penalized Cox regression (R package coxphf). This approach ensured stable HR estimations and corresponding 95% confidence intervals (CIs) despite the sparse event distribution in specific subgroups.

All statistical tests were two-sided, and a P < 0.05 was considered statistically significant. Statistical analyses were performed using SPSS version 31.0 (IBM), GraphPad Prism version 10.6.1, and R version 4.5.1.

Results

Patient characteristics and molecular stratification

A total of 270 patients with Ph + ALL were enrolled. At the time of transplantation, 247 patients (91.5%) were in first complete remission (CR1). Based on pre-allo-HSCT BCR::ABL1 transcript levels assessed by RQ-PCR, 170 patients (63.0%) were MRD-negative, 62 (23.0%) had low-level MRD positivity (0–0.1%), and 38 (14.0%) had a higher MRD burden (≥ 0.1%, corresponding to loss of MMR).

According to post-allo-HSCT molecular response patterns, 80 patients (29.6%) were double-negative, 158 (58.5%) were dPCR single-positive, 3 (1.1%) were RQ-PCR single-positive, and 29 (10.7%) were double-positive. Donor sources included matched sibling donors in 93 patients (34%), haploidentical donors in 212 patients (79%), and unrelated donors in 1 patient (0.4%). The median follow-up duration for the entire cohort was 24 months.

(Table 1).

Table 1.

Baseline clinical characteristics of patients stratified by post-transplant MRD status

Characteristic	Total (N = 270)1	Double Negative N = 801	dPCR+ N = 1581	RT-qPCR+ N = 31	Double Positive N = 291	p-value2
Age	37 (26, 49)	39 (29, 50)	35 (26, 48)	51 (22, 61)	39 (21, 49)	0.5
Patient Sex						0.4
Male	162 (60%)	49 (61%)	90 (57%)	2 (67%)	21 (72%)
Female	108 (40%)	31 (39%)	68 (43%)	1 (33%)	8 (28%)
ABL Fusion						0.035
190	173 (64%)	42 (53%)	110 (70%)	3 (100%)	18 (62%)
210	97 (36%)	38 (48%)	48 (30%)	0 (0%)	11 (38%)
Conditioning Regimen						0.2
MAC	227 (84%)	64 (80%)	134 (85%)	2 (67%)	27 (93%)
RIC	43 (16%)	16 (20%)	24 (15%)	1 (33%)	2 (6.9%)
Graft Source						0.13
PB	257 (95%)	73 (91%)	154 (97%)	3 (100%)	27 (93%)
nonPB	13 (4.8%)	7 (8.8%)	4 (2.5%)	0 (0%)	2 (6.9%)
Pre-transplant Status						0.002
CR1	247 (91%)	75 (94%)	149 (94%)	2 (67%)	21 (72%)
nonCR1	23 (8.5%)	5 (6.3%)	9 (5.7%)	1 (33%)	8 (28%)
Donor-Recipient Sex Match
Male-Male	108 (40%)	31 (39%)	60 (38%)	0 (0%)	17 (59%)
Male-Female	84 (31%)	23 (29%)	55 (35%)	1 (33%)	5 (17%)
Female-Male	53 (20%)	17 (21%)	30 (19%)	2 (67%)	4 (14%)
Female-Female	25 (9.3%)	9 (11%)	13 (8.2%)	0 (0%)	3 (10%)
Donor Relation
Sibling	93 (34%)	27 (34%)	56 (35%)	2 (67%)	8 (28%)
Parent	90 (33%)	28 (35%)	52 (33%)	0 (0%)	10 (34%)
Offspring	76 (28%)	24 (30%)	42 (27%)	1 (33%)	9 (31%)
URD	11 (4.1%)	1 (1.3%)	8 (5.1%)	0 (0%)	2 (6.9%)
HSCT Type						0.2
MSD	57 (21%)	14 (18%)	34 (22%)	1 (33%)	8 (28%)
HID	212 (79%)	66 (83%)	124 (78%)	2 (67%)	20 (69%)
URD	1 (0.4%)	0 (0%)	0 (0%)	0 (0%)	1 (3.4%)
blood_type_relation
Match	154 (57%)	42 (53%)	93 (59%)	2 (67%)	17 (59%)
Major	42 (16%)	11 (14%)	25 (16%)	0 (0%)	6 (21%)
Minor	59 (22%)	19 (24%)	34 (22%)	1 (33%)	5 (17%)
Unmatched	15 (5.6%)	8 (10%)	6 (3.8%)	0 (0%)	1 (3.4%)
Pre-HSCT BCR-ABL Levels	0.00 (0.00, 0.02)	0.00 (0.00, 0.01)	0.00 (0.00, 0.02)	0.00 (0.00, 1.80)	0.02 (0.00, 0.08)	0.079
MNC Dose (10^8/kg)	9.18 (7.76, 11.00)	9.40 (7.45, 11.75)	9.00 (7.76, 10.70)	9.20 (8.66, 9.50)	9.40 (8.47, 10.60)	> 0.9
CD34 Dose (10^6/kg)	4.00 (2.43, 5.70)	3.05 (2.18, 5.08)	4.00 (2.50, 5.90)	6.26 (3.40, 9.70)	4.31 (3.28, 5.20)	0.11
Follow-up (Months)	24 (10, 34)	27 (10, 36)	22 (11, 28)	26 (1, 34)	16 (10, 26)	0.04

Note: Data are presented as n (%) or median (interquartile range, IQR). Abbreviations: ALL, acute lymphoblastic leukemia; dPCR, digital polymerase chain reaction; RQ-PCR, real-time quantitative polymerase chain reaction; MAC, myeloablative conditioning; RIC, reduced-intensity conditioning; PB, peripheral blood; CR1, first complete remission; URD, unrelated donor; MSD, matched sibling donor; HID, haploidentical donor; HSCT, hematopoietic stem cell transplantation; MNC, mononuclear cell; BCR-ABL, breakpoint cluster region-Abelson murine leukemia viral oncogene homolog 1

¹Percentages may not total 100% due to rounding.

²P-values compare the four groups (Double Negative, dPCR+, RQ-PCR+, and Double Positive) using the Kruskal-Wallis test for continuous variables and the Chi-square test or Fisher’s exact test for categorical variables

Sensitivity and lead time of dPCR for predicting disease progression

Based on paired post- allo-HSCT MRD assessments using dPCR and RQ-PCR across 885 bone marrow samples, patients were stratified into four groups according to assay concordance (Fig. 1A). Among the 270 enrolled patients, 80 (29.6%) were classified as double-negative, 158 (58.5%) as dPCR single-positive, 3 (1.1%) as RQ-PCR single-positive, and 29 (10.7%) as double-positive. A total of 94 patients (34.8%) underwent continuous post-transplant MRD monitoring starting from month 1 after allo-HSCT. Overall, patients contributed a median of 3 post-transplant bone marrow samples (range, 1–14) for longitudinal MRD analysis.

Fig. 1.

Patient flow and MRD assessment. (A) Consort flow diagram of the study population (n = 270). (B) Lead time of dPCR positivity preceding RQ-PCR, MFC, and clinical relapse/EMD. Individual dots represent unique patient cases, with horizontal bars indicating the median and interquartile range (IQR)

This enhanced sensitivity translated into a significant clinical window for early intervention. Among patients in the dPCR single-positive group who subsequently progressed, dPCR positivity emerged as the earliest marker of disease recurrence. As shown in the longitudinal analysis, dPCR positivity preceded the detection of RQ-PCR positivity (progression to double-positive status) and MFC-MRD conversion by a median of 64.5 days (IQR 36.0–119.0) and 77 days (IQR 53.0–119.0), respectively. Most importantly, dPCR provided a substantial lead time—median 91.5 days (IQR 55.0–128.5)—prior to frank hematologic relapse or EMD (Fig. 1B).

Prognostic Impact of dPCR

Post-transplant dPCR positivity was not associated with a significantly higher CIR compared with dPCR negativity (2-year 4.60 vs. 1.60%, P = 0.293). In contrast, dPCR-positive patients exhibited a significantly higher 2-year cumulative incidence of MFC-MRD conversion (2-year 10.0 vs. 1.20%, P = 0.009). As NRM was comparable between dPCR-positive and dPCR-negative patients (P > 0.9), the similar CIR translated into comparable LFS (2-year 94.2 vs. 97.2%, P = 0.3). Similarly, OS did not differ significantly between groups (2-year 98.7 vs. 98.8%, P = 0.6) (Fig. 2).

Fig. 2.

Outcomes stratified by dPCR status. (A). LFS: Leukemia-free survival; (B). OS: Overall survival; (C). CIR: Hematologic relapse/EMD; (D). MFC-MRD Conversion: dPCR+ predicts higher conversion risk

Univariate and multivariate analyses of factors associated with progression

In univariate analyses, post-transplant dPCR positivity emerged as a significant predictor of immunophenotypic progression, rather than of clinical survival outcomes. Specifically, dPCR positivity was strongly associated with an increased risk of subsequent MFC-detected MRD conversion (hazard ratio [HR], 9.27; 95% CI, 1.23–69.85; P < 0.05; 2-year cumulative incidence, P = 0.009) (Fig. 3D). In contrast, dPCR positivity was not significantly associated with leukemia-free survival (LFS; HR, 2.35; 95% CI, 0.50–11.01; P = 0.278; Kaplan–Meier analysis, P = 0.30) (Fig. 3A), overall survival (OS; HR, 1.78; 95% CI, 0.18–17.87; P = 0.623; Kaplan–Meier analysis, P = 0.60) (Fig. 3B), or hematologic relapse (HR, 2.94; 95% CI, 0.35–24.43; P = 0.319; 2-year cumulative incidence, P = 0.293) (Fig. 3C).

Fig. 3.

Univariate Cox regression analysis for clinical outcomes (n = 270). (A). LFS: Impact of clinical variables on leukemia-free survival; (B). OS: Impact of clinical variables on overall survival; (C). Hematologic Relapse: Univariate factors associated with hematologic relapse risk; (D). MFC-MRD Conversion: dPCR status is the significant predictor for MFC-MRD conversion (P = 0.031). Note: HR, hazard ratio; CI, confidence interval

Other clinical variables, including age > 40 years (P = 0.141–0.767) and conditioning regimen (P = 0.319–0.959), did not reach statistical significance for any evaluated endpoint in univariate models. Likewise, BCR::ABL1 isoform (P = 0.194 for LFS) and mononuclear cell dose (P = 0.195–0.541) showed no significant prognostic impact. Donor sex was associated with an increased risk of MFC-detected MRD conversion (HR, 0.37; 95% CI,0.18–0.79; P = 0.010), whereas donor relationship (P = 0.552–0.892) were not significantly associated with clinical outcomes. (Fig. 3).

In multivariable analyses, post-transplant dPCR positivity was not independently associated with survival outcomes or clinical relapse but emerged as a strong independent predictor of immunophenotypic progression. Specifically, after adjustment for clinical covariates, dPCR positivity did not retain independent prognostic significance for leukemia-free survival (LFS; hazard ratio [HR], 2.84; 95% CI, 0.59–13.56; P = 0.192) (Fig. 4A), overall survival (OS; HR, 2.58; 95% CI, 0.23–28.93; P = 0.443) (Fig. 4B), or hematologic relapse (HR, 3.36; 95% CI, 0.40–28.34; P = 0.266) (Fig. 4C).

Fig. 4.

Multivariate analysis of clinical outcomes in Ph + ALL (n = 270). (A). LFS: dPCR status is not an independent predictor for LFS; (B). OS: No significant independent association between dPCR status and OS; (C). Hematologic Relapse: Multivariate analysis (Firth Model) for relapse risk; (D). MFC-MRD Conversion: dPCR status is a significant independent predictor (HR 9.56, P = 0.029)

In contrast, post-transplant dPCR positivity was the variable independently associated with a significantly increased risk of subsequent MFC-detected MRD conversion in the Firth penalized model (HR, 9.56; 95% CI, 1.26–72.54; P = 0.029) (Fig. 4D). Donor sex (female vs. male) was the variable associated with a significantly decreased risk of subsequent MFC-detected MRD conversion (HR, 0.37; 95% CI,0.17–0.80; P = 0.012). Although several clinical variables met inclusion criteria in univariate screening, none retained independent significance across the evaluated endpoints. Age > 40 years demonstrated a numerical trend toward increased risk for OS (HR, 5.35; P = 0.154) and LFS (HR, 2.81; P = 0.105) without reaching statistical significance. Other covariates, including sex and conditioning regimen, were not independently associated with clinical or molecular outcomes (all P > 0.05). (Fig. 4).

Clinical outcomes according to dPCR and RQ-PCR concordance

Eighty patients (29.6%) remained persistently negative by both assays (double-negative group). Most patients (n = 158, 58.5%) demonstrated a discordant molecular pattern, characterized by conversion to dPCR positivity (median, 0.009%; range, 0.0005–0.0688%) while remaining negative by RQ-PCR (dPCR single-positive group). A small, atypical subset (n = 3, 1.1%) was positive exclusively by RQ-PCR, whereas 29 patients (10.7%) exhibited concurrent positivity by both assays (double-positive group; dPCR median, 0.0153%, range, 0.0005–34.27%; RQ-PCR median, 0.016%, range, 0.0019–53.9%).

Baseline characteristics—including age (P = 0.50), sex (P = 0.40), conditioning regimen (P = 0.20), HSCT type (P = 0.20), graft source (P = 0.13) and CD34⁺ cell dose (P = 0.11)—were well balanced across the four molecular response groups. In contrast, significant differences were observed in pre-transplant disease status (P = 0.002), and BCR::ABL1 isoform (P = 0.035). Pre- allo-HSCT BCR::ABL1 transcript levels showed a nonsignificant trend toward variation among groups (P = 0.079).

Compared with the double-negative group, patients in the dPCR single-positive (discordant) group exhibited comparable 2-year cumulative incidence of relapse (3.44% vs. 1.69%; global P = 0.30) but a modestly inferior leukemia-free survival (95.1% vs. 98.3%; global P = 0.026). Notably, the double-positive group experienced the poorest outcomes across nearly all evaluated endpoints, with a markedly higher cumulative incidence of MFC-detected MRD conversion (2-year, 38.6% vs. 0%; global P < 0.001). Although differences in LFS reached statistical significance (P = 0.026), the cumulative incidence of hematologic relapse did not differ significantly among the four groups (P = 0.30). (Fig. 5).

Fig. 5.

Clinical outcomes by dPCR and RQ-PCR status. (A). LFS: Leukemia-free survival; (B). OS: Overall survival; (C). CIR: Hematologic relapse/EMD; (D). MFC-MRD Conversion: Double positive group shows highest risk of conversion (P < 0.001)

Discussion

To our knowledge, this represents the largest prospective study to systematically evaluate the incremental prognostic value of dPCR-based MRD monitoring beyond conventional RQ-PCR and MFC in patients with Ph + ALL undergoing allo-HSCT. Our findings demonstrate that dPCR surveillance identifies a substantial subgroup of patients harboring occult molecular disease that remains undetectable by standard assays. Importantly, unlike observations in other high-risk leukemias such as KMT2A-rearranged disease36, ultra-low–level molecular persistence in Ph + ALL appears to be characterized by a distinct state of clinical inertia. In this setting, dPCR positivity serves as an early indicator of subsequent immunophenotypic progression; however, under contemporary post-transplant management—including continuous TKI therapy—it does not independently translate into adverse survival outcomes.

Consistent with the 2024 ELN recommendations, molecular MRD monitoring remains a cornerstone of relapse risk assessment in Ph+ ALL^1,4. While RQ-PCR continues to serve as the clinical reference standard, dPCR is increasingly recognized for its superior analytical sensitivity33, 37. In this study, we identified a substantial discordant subgroup characterized by dPCR positivity in the absence of RQ-PCR–detectable disease—patients who would otherwise be classified as MRD-negative under current criteria.

Importantly, this ultra-low–level molecular persistence was not clinically silent. dPCR positivity was strongly associated with subsequent immunophenotypic progression, indicating that it captures an earlier phase of molecular evolution before conventional MRD detection. However, despite this clear biological signal, it did not translate into an increased risk of hematologic relapse or inferior survival outcomes. This apparent paradox highlights a key feature of Ph + ALL in the contemporary treatment era: the clinical impact of residual disease is substantially modulated by effective post-transplant interventions, including switching TKIs, GVL effects, and timely immunotherapeutic strategies following MFC-MRD conversion4, 15, 38, 39.

Within this framework, it is critical to recognize that MRD-directed clinical interventions in our cohort were driven exclusively by RQ-PCR and MFC results, while dPCR findings were blinded. As a result, molecular progression identified by conventional assays triggered preemptive treatment escalation, effectively interrupting disease evolution. This dynamic likely explains why even patients with higher molecular burden, including those in the double-positive group, did not uniformly experience clinical relapse, and why survival outcomes remained comparable across molecular subgroups9, 40, 41.

Notably, the principal value of dPCR lies in its ability to detect this molecular evolution at a substantially earlier stage. In patients who eventually progressed, dPCR positivity consistently emerged as the earliest detectable signal, preceding RQ-PCR positivity and MFC-MRD conversion by a median of 64.5 and 77 days, respectively, and anticipating overt hematologic relapse or EMD by approximately three months. This extended lead time represents a clinically meaningful window for preemptive intervention.

Taken together, these findings support a conceptual shift in the role of dPCR: rather than serving as a direct predictor of imminent treatment failure, dPCR functions as an ultra-early warning signal of molecular progression. In the context of effective TKI maintenance and responsive clinical intervention, this early signal is frequently contained before translating into overt relapse. Therefore, the primary clinical utility of dPCR lies in enabling earlier therapeutic intervention, rather than refining prognostic stratification for survival outcomes.

Finally, several limitations should be acknowledged. This was a single-center study with variable longitudinal sampling intensity and relatively limited follow-up, resulting in a low number of relapse events. In addition, IgH rearrangement–based NGS-MRD monitoring was not systematically performed, and distinction between de novo Ph + ALL and CML-LBC was not uniformly established in all cases. To further elucidate the biological nature of dPCR-detected signals, future studies should integrate BCR::ABL1 monitoring with IgH rearrangement–based NGS-MRD, which may help distinguish true clonal relapse from multilineage stem cell–derived disease evolution and thereby refine risk stratification and guide more precise early intervention strategies42–45.

Conclusion

In this largest prospective cohort of patients with Ph + ALL undergoing allo-HSCT, dPCR-based MRD monitoring demonstrated superior analytical sensitivity and provided a meaningful molecular lead time beyond conventional assays. Our findings indicate that ultra-low–level BCR–ABL1 persistence detected by dPCR is characterized by a distinct state of clinical inertia, in which molecular positivity does not translate into an immediate survival disadvantage under contemporary post-transplant intervention. Importantly, dPCR positivity consistently preceded RQ-PCR positivity, MFC-MRD conversion, and overt relapse, thereby creating a clinically actionable window for early intervention. In this context, dPCR serves as a robust and independent indicator of subsequent molecular and immunophenotypic progression, functioning primarily as an ultra-early warning signal to guide preemptive therapeutic strategies rather than as a determinant of survival outcomes.

Electronic Supplementary Material

Below is the link to the electronic supplementary material.

Author contributions

Author Contributions Conceptualization: YL, Y-ZQ, and ML; Methodology: YL and W-MC; Software: YL; Validation: YL, W-MC, and X-SZ; Formal analysis: YL; Investigation: YL, W-MC, X-SZ, Y-JC, TZ, YW (Ying Wu), Y-FC, X-DM, Y-QS, YW (Yu Wang), L-PX, and X-HZ; Resources: X-JH, Y-ZQ, and ML; Data curation: YL and W-MC; Writing—original draft preparation: YL; Writing—review and editing: Y-ZQ and ML; Visualization: YL; Supervision: Y-ZQ, ML, and X-JH; Project administration: Y-ZQ and ML; Funding acquisition: Y-ZQ, ML, and X-JH. All authors read and approved the final manuscript.

Data availability

The datasets generated during and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Conflict of Interest

The authors declare no competing interests.

Ethics approval

This study was conducted in accordance with the Declaration of Helsinki and approved by the Institutional Review Board of Peking University People’s Hospital (Ref No. 2021PHA154-001).

Informed consent

Informed consent was obtained from all individual participants included in the study.