Minimal residual disease in adult ALL: technical aspects and implications for correct clinical interpretation

Abstract

Nowadays, minimal residual disease (MRD) is accepted as the strongest independent prognostic factor in acute lymphoblastic leukemia (ALL). It can be detected by molecular methods that use leukemia-specific or patient-specific molecular markers (fusion gene transcripts, or immunoglobulin/T-cell receptor [IG/TR] gene rearrangements), and by multi-parametric flow cytometry. The sensitivity and specificity of these methods can vary across treatment time points and therapeutic settings. Thus, knowledge of the principles and limitations of each technology is of the utmost importance for correct interpretation of MRD results. Time will tell whether new molecular and flow cytometric high-throughput technologies can overcome the limitations of current standard methods and eventually bring additional benefits. MRD during standard ALL chemotherapy is the strongest overall prognostic indicator and has therefore been used for refining initial treatment stratification. Moreover, MRD positivity after the maintenance phase of treatment may point to an impending relapse and thus enable salvage treatment to be initiated earlier, which could possibly improve treatment results. The prognostic relevance of pretransplantation MRD was shown by several studies, and MRD high-risk patients were shown to benefit from stem cell transplantation (SCT). Also, MRD positivity after SCT correlates with worse outcomes. In addition, MRD information is very instructive in current clinical trials that test novel agents to evaluate their treatment efficacy. Although conventional clinical risk factors lose their independent prognostic significance when combined with MRD information, recently identified genetic markers may further improve the treatment stratification in ALL.

Figure 1.

MRD detection in ALL. Schematic diagram for detection of MRD. The red dashed line indicates the detection limit of cytomorphology (5%). Note the difference between nondetectable and truly negative MRD.

Figure 2.

General considerations for correct interpretation of MRD results. This schematic illustrates that MRD is a time point (TP)–dependent variable. Therefore, the prognostic meaning of MRD negativity is different when measured very early during therapy (Treatment Element A) compared with late time points (Treatment Element B). For example, MRD assessment at the end of induction therapy is useful for recognizing patients with low risk of relapse, and MRD measurement at the end of consolidation therapy is useful for identifying patients at high risk of relapse. In addition, the prognostic significance of MRD might be influenced by subsequent treatment elements (Treatment Elements C and D). MRD negativity does not necessarily indicate the eradication of the disease but does indicate a decrease to a level below the detection limit of the respective assay. Therefore, the knowledge of the sensitivity of the method is important for correct interpretation of the results.

Figure 3.

Comparison of MRD results assessed with different techniques. (A) MFC vs IG/TR RT-qPCR. The sensitivity of MFC for MRD detection relative to RT-qPCR (y-axis) is shown for certain time points (days 15, 33, and 78) and for all samples together. Data are presented for variable numbers of acquired cells (x-axis): all samples (independent of cell counts; n = 377) and samples with at least 1 (n = 330), 2 (n = 287), 3 (n = 255), 4 (n = 227) and 5 (n = 191) million cells acquired. Sensitivity is calculated as the number of samples positive by both FCM and PCR divided by the total number of samples positive by PCR. The sensitivity significantly increases when larger numbers of cells are acquired. (B) Immunoglobulin heavy chain RT-qPCR vs immunoglobulin heavy chain NGS. The comparison of MRD as detected by RT-qPCR and NGS is shown for different follow-up time points (days 15, 33, and 78) by qPCR (x-axis) and NGS (y-axis). The correlation of both methods is good (R² = 0.72) with the majority of the noncorrelating samples below the sensitivity thresholds of the methods. (C) IG/TR RT-qPCR vs BCR-ABL1 RT-qPCR. Comparison of MRD levels in ALL patients as detected by IG/TR RT-qPCR (x-axis) vs BCR-ABL1 genomic transcript quantification (y-axis). A significant number of samples (23%) have quantifiable BCR-ABL1 levels, whereas IG/TR MRD is negative. Panel C adapted from Hovorkova et al.

Figure 4.

Prognostic value of MRD in Ph^–adult ALL. (A) Probability of OS for patients in the standard-risk and high-risk groups according to molecular response (MR) status (leukemia-associated IG/TR rearrangements detected by RT-qPCR) after induction or early consolidation chemotherapy (week 16) excluding SCT in CR1. Results of the GMALL 06/99 and 07/03 trials. Complete molecular remission (MolCR) is defined as MRD negativity with an assay sensitivity of at least 10^–4. Molecular failure (MolFail) is defined as quantifiable MRD positivity ≥10^–4. (B) OS according to different molecular levels of postinduction MRD (detected by IG/TR RT-qPCR). Results of the NILG ALL 09/00 trial. Complete molecular remission (CMR) is defined as MRD negativity at weeks 10, 16, and 22. MR is defined as MRD <10^–4 at week 10 and/or week 16 and/or week 22. Molecular resistance type 1 (MR1) is defined as measurable MRD ≥10^–4 and <10^–3. MR2 is defined as measurable MRD ≥10^–3. Panel B adapted from Bassan et al with permission. (C) Effect of SCT on OS. Results of the GRAALL-2003/2005 trials. Simon-Makuch plots of SCT time-dependent analysis of OS according to molecular MRD response (≥10^–3or <10^–3) and type of postremission therapy (SCT vs no SCT) in high-risk ALL. MRD was detected by IG/TR-based RT-qPCR. Panel C adapted from Dhédin et al. (D) OS for patients with relapsed or refractory BCP-ALL receiving salvage immunotherapy (inotuzumab ozogamicin or blinatumomab) by MRD response (MRD positive or MRD negative by 6-color MFC) and salvage status (Salvage 1 [S1] vs Salvage 2 [S2]). Retrospective analysis at MD Anderson Cancer Center. Includes data from 4 Ph⁺ patients. Panel D adapted from Jabbour et al with permission.