Introduction

Acute lymphoblastic leukemia (ALL) has a varied clinical presentation, with precursor B-cell ALL (B-ALL) being one of its most common immunological subtypes and affecting almost 75% of adult ALL cases. The most frequently observed genetic aberration in B-cell ALL is the Philadelphia chromosome-positive ALL (Ph + ve ALL) [1, 2]. The American Cancer Society estimates the occurrence of 6600 new cases of ALL in their population in the year 2022 [3], whereas the incidence rate of ALL in India has been estimated to be 101.4 per million and 62.3 per million, across all age groups and genders, respectively [2]. Minimal residual disease (MRD) has emerged as a robust prognostic indicator in B-cell ALL [4]. MRD can be described as the presence of a very low number of cancerous/malignant cells after chemotherapy or following a hematopoietic stem cell transplantation (HSCT) [5, 6].

MRD detection in ALL dates to the 1980s when immunofluorescence microscopy was used [7]. The use of MRD diagnostics in clinical trials as a surrogate endpoint for evaluating the efficacy of novel agents has also been observed. It was initially used for T-cell ALL because a highly specific immunophenotype for B-cell ALL had not been identified [8, 9]. The limitation of two- or three-color immunofluorescence microscopy made the detection of minor differences in marker expressions challenging, hence many new techniques emerged [8, 9]. Polymerase chain reaction (PCR), flow cytometry (FCM), next-generation sequencing (NGS), and next-generation flow cytometry are the molecular techniques used currently for MRD assessment [10, 11].

MRD assessments are very time-point specific [10]; hence, they are a critical prognostic indicator in very high-risk B-cell ALL patients. Among these patients, the subset that fails to achieve the end-of-induction (EOI) MRD positivity displays inferior outcomes [12, 13]. In addition, patients displaying very early MRD clearance have significantly better outcomes [14]. Studies have shown a direct strong correlation between MRD and the risk of relapse in ALL patients, thereby highlighting the prognostic value of MRD [10, 15]. Results from a meta-analysis reflect that pretransplant MRD positivity is a significant negative predictor of relapse-free survival (RFS), event-free survival (EFS), and overall survival (OS) [16]. Such results emphasize the importance of MRD evaluation before transplant, especially when treatment intensification is needed.

There is a paucity of data on MRD from an Indian context, and a complete understanding of MRD assessments and their correlation with outcomes among B-cell ALL patients is also lacking [17]. The primary objective of this review was to survey the available literature that discusses different aspects of MRD testing in the Indian context, specifically focusing on adolescent and young adults (AYA) and adult B-cell ALL patients. The insights gathered from studies specific to Indian settings are presented in this systematic literature review and discussed considering the global research landscape.

Methods

Literature Search

A systematic literature search was done using the Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) reporting guidelines [18]. After the finalization of objectives, specific research questions were framed to guide the search process. Search queries were designed and reviewed independently for coverage and accuracy. The review of queries was based on the Peer Review for Electronic Search Strategies (PRESS) Guidelines. The components of the search string used in combinations are listed in Table 1.

Table 1 Components of the query included as a part of the search strategy
Full size table

Data Sources

The primary source of literature was PubMed. Additional searches were performed on Google Scholar, and the first 200 results were considered for preliminary screening [19]. From this subset, only articles not indexed in PubMed were included. Articles published within the last ten years from the date of query execution (April 3, 2022) were selected. Other data sources where the search was extended included original abstracts presented at the annual meetings of (1) the American Society of Hematology (ASH); (2) the American Society of Clinical Oncology (ASCO); (3) the European Society for Medical Oncology (ESMO); (4) the Indian Society of Hematology and Blood Transfusion (ISHBT); and the European Hematology Association (EHA) during the period 2019–2021.

Study Screening and Data Extraction

Based on objectives and research questions, the inclusion/exclusion criteria were outlined. The titles and abstracts of the articles retrieved were screened. Two independent reviewers performed the screening process, and disagreements were resolved by agreement based on discussion. Full-text versions of the articles that met the inclusion criteria were retrieved. These articles were selected for the next round of screening based on the full-text review. Articles that met the screening criteria were further considered for the extraction of study-relevant data. A predefined structured template was used for capturing data.

Screening Criteria

To minimize the risk of bias, all screening and evaluation steps were carried out independently by two individuals. Final decisions were made after resolving the disagreements based on discussion among reviewers. In addition, the inclusion/exclusion criteria were predefined based on which a template was designed to collect data. The risk of bias assessment of studies using standardized checklists (e.g., Newcastle–Ottawa scale) was not carried out as the primary goal was to survey the MRD-specific methodological aspects and application areas, and the number of articles retrieved was limited.

Results

Summary of Search

We identified and screened 205 studies from PubMed. An additional two articles were identified by searching Google Scholar for articles not indexed in PubMed. A total of 47 studies (range 2019–2021) were retrieved from annual meetings of ASH, ASCO, EHA, ESMO, and ISHBT. Of the total 254 studies, 24 articles were shortlisted according to the inclusion/exclusion criteria and after the removal of redundant articles by two independent reviewers. The detailed search strategy adopted during the systematic review is shown in Fig. 1. The list of final articles included after screening is given in Table 2.

Fig. 1
figure 1

The process and flow diagram for screening and identifying study-relevant literature. Framework adapted from the Preferred Reporting Items for Systematic reviews and Meta-Analysis guidelines. B-ALL: B-cell acute lymphoblastic leukemia

Full size image
Table 2 List of articles that met the final inclusion criteria
Full size table

MRD Assessment: Current Status

The Hematology Cancer Consortium maintains a database (Indian Acute Leukemia Research Database) that stores retrospective data from nine centers across India. Our search identified only one study exploring this database. The data published by Ganesan et al. [21] reflected that MRD assessment was available for only 47% of patients in the AYA group. Of the 1383 patients registered, 1141 (82.5%) underwent treatment, and MRD status was available for 654 patients. After induction, 76% of patients achieved complete remission (CR), and MRD was positive in 240 of 654 (37%) patients. Both univariate and multivariate analyses highlighted that inferior EFS and OS were associated with MRD positivity [21].

Timing of MRD Assessment

Regarding the timings for assessment, five studies followed EOI assessments on days 29–33 [22,23,24,25,26], except for studies by Chatterjee et al. [32] and Das et al. [33] that made assessments between days 35 and 40 and 30 and 35, respectively [27]. Individual studies (n = 2) that evaluated at mid-induction on day 21 or after phase 1a induction were also identified [27, 28]. Only three studies included end-of-consolidation (EOC) evaluations and subsequent follow-ups [24, 29, 32].

Samples and MRD Detection Methods

A majority of studies (n = 10) included in this systematic review used bone marrow aspirate, with a few (n = 3) using peripheral blood as well. Data comparing outcomes between these two samples were not available. Of the 24 studies, 22 studies used FCM using multicolored panels and multiparametric flow cytometry (MFC) for MRD assessments, whereas two used real-time quantitative PCR (RQ-PCR) [25, 29]. MFC methods with 5-color [30], 8-color [27, 42], 10-color [23, 24, 26, 28, 32, 33, 40], and 12-color [38] panels were observed in 12 studies. The panel of markers identified in these studies includes CD10, CD13, CD19, CD20, CD23, CD34, CD66c, CD123, CD200, and CD304 for the FCM analysis, and RQ-PCR was used for the BCR-ABL1 transcript characterization. A study by Chatterjee et al.[32] evaluated the expression pattern of CD304 in a cohort of adult B-cell ALL patients and reported that CD304 was found to be positive in a significant percentage of EOI (62/129 [48%]) and EOC (26/50 [52%]) MRD-positive, B-cell ALL samples. CD123 has also received consideration as a marker for residual disease assessment and response evaluation in acute myeloid leukemia and B-ALL. CD123 expression at diagnosis was shown to be associated with post-induction MRD-positive status in B-ALL (p < 0.001) [23, 33]. The Leukemia-associated aberrant immunophenotype (LAIP) and Difference from Normal (DFN) approach was found to be a reliable tool for MRD assessment at the diagnostic and MRD time points in the study by Das et al. [33]. Changes in at least one of the nine immunophenotypic markers in B-ALL post-induction was observed in 94.04% cases. The utility of other markers such as CD44, CD73, CD304, and CD200 was assessed in those cases requiring review of baseline IPT. Out of 18 cases, the information was useful in 8 cases (44.4%). The study emphasized the need for better markers for distinguishing leukemic blasts from hematogones. Additionally, MRD assessment in B-ALL is complicated by changes in IPT after induction chemotherapy, necessitating pattern recognition and simultaneous analysis of multiple IPT markers. [33]. Arunachalam et al. [29] analyzed the prognostic relevance of MRD based on BCR-ABL1 copy numbers in Ph-positive ALL patients. BCR-ABL1 copy numbers were evaluated using RQ-PCR. The cost-effectiveness of the MRD method was analyzed in three studies [25, 31, 38]. Patkar et al. [31] proposed a relatively cost-effective MRD panel applicable to over 90% of patients. Using their approach, they detected MRD in 60% and 47% of patients at mid- and end-induction time points, respectively. Another two studies focused on the optimization of MRD panels with the goal of cost-effectiveness and reduction in the number of LAIPs [22, 37].

MRD and Treatment Outcomes

Only five articles focused on evaluating specific treatment regimens with MRD as a measure of post-induction response [24, 27, 28, 35, 36]. The outcomes of patients treated with different regimens were presented as combined results in most studies. Only one study was identified, where post-induction MRD was used as one of the indications for stem cell transplantation (SCT) in the first complete remission (CR1) [34]. A phase 2 study evaluating the combination of bortezomib, rituximab, and a pediatric-inspired ALL regimen showed post-induction persistent MRD to be associated with inferior OS and EFS [24]. Outcomes among MRD-based, risk-stratified patients show that patients with poor-risk status were associated with inferior OS and EFS [29]. The impact of MRD on OS was studied in pediatric and AYA groups treated with a modified multicenter protocol (MCP) 841 [27]. The results outline that MRD-negative patients responded better than MRD-positive patients (p = 0.03). Post-induction MRD was acknowledged as a useful prognostic tool for ALL patients treated with the modified MCP 841 protocol. Furthermore, in a study assessing the outcomes associated with Berlin–Frankfurt–Münster-90 among AYA, post-induction MRD persistence emerged as the only factor predictive of poor outcomes [28].

Risk Category and MRD Status

A study conducted by Arunachalam et al. [29] assessed the prognostic relevance of MRD based on BCR-ABL1 copy numbers in Ph-positive ALL patients. In this study, the MRD status was assessed at three different time points. Patients having persistent MRD-positive status at all three measured time points or having an increasing BCR-ABL1/ABL1 copy number ratio with an increase in their MRD-positive status by the third measurement were categorized as MRD poor-risk status. Those patients having MRD negativity and decreasing BCR-ABL1/ABL1 copy number ratio with a strong MRD-negative status by the third assessment were categorized as MRD good risk. MRD poor-risk patients had adverse outcomes when compared to MRD good-risk patients in terms of OS (p = 0.031) and EFS (p ≤ 0.001). Patients with high-risk diseases and those with EOI MRD positivity are at higher risk of adverse events [22]. CD304 positivity has also been shown to be associated with BCR-ABL1 fusion, with a significant percentage of EOI and EOC MRD positivity [32]. Results from a smaller cohort indicate that patients with P2Y receptor family member 8-cytokine receptor-like factor 2 (P2RY8-CRLF2) translocation who underwent EOI MRD testing showed positivity [26].

Discussion

The primary objective of this study was to survey the available literature that discusses different aspects of MRD testing in the Indian context, specifically focusing on AYA and adult B-cell ALL patients, as B-cell ALL studies done so far have been limited to the pediatric population and have been done on smaller cohorts. A systematic search was carried out, which indicated limited data on the use and impact of MRD assessment among AYA and adult B-cell ALL patients in Indian settings. Available data on samples, timings, techniques, and outcomes were gathered. This review presents and discusses the limitations and aspects that need further investigation. The MRD assessment status was available only from one study [21].

The aspects of sensitivity of MRD assessment and its implications have not been discussed in the identified studies. Samples, sensitivity, and timings can be critical factors in leveraging maximum benefits from MRD assessment. An important point to consider is the timing of MRD measurement, which can help in taking treatment decisions. Different insights can be provided by MRD, depending upon the timing of assessment: very early, after induction/consolidation, and before and after SCT. Although early response assessment (days 8 and 15) has been discussed in the literature, the review of articles selected for this study did not reveal any discussion on such aspects. Evidence supports early MRD testing [1]. Negative MRD status at very early time points during the induction phase correlates with better outcomes both in adult and in childhood B-ALL [43, 44]. Our results also suggest that MRD was a critical prognostic indicator strongly associated with RFS and EFS [29]. The focus of most studies identified in the review was on assessment at EOI and/or EOC. The importance of EOI MRD assessment was highlighted in one study [42]. Different treatment protocols have established informative checkpoints, which aid in monitoring outcomes appropriately. However, such sequential monitoring was not observed in most studies identified in Indian settings.

The majority of studies identified from Indian settings have used bone marrow samples, with some studies using peripheral blood. However, no comparison of utility has been studied. Several clinical studies have evaluated MRD status in bone marrow samples and blood in B-cell ALL and T-cell ALL [33, 45,46,47]. The use of peripheral blood MRD can serve as a noninvasive technique to monitor systemic relapse and might have additional clinical and diagnostic value in patients with a high risk of extramedullary disease [48]. There is a paucity of data on studies evaluating the application of MRD assessment. Typically, MRD assessments are done using a single aspirate sample, which can vary due to sampling error and/or collection techniques [49]. Inaccuracies resulting from a sample that is diluted or has an unequal distribution of disease involvement in the bone marrow can pose limitations.

The use of > 10−4 or > 5 × 10−4 as a threshold has been suggested for poor MRD responders with poor prognoses [4]. The acceptable level of sensitivity of MRD assays remains unresolved [1]. MFC and quantitative PCR are the most frequently used MRD detection techniques/methods in clinical practice. MRD-based risk stratification can be further refined by using NGS like sensitive assays. Accurate identification of patients with persistent MRD who are at the highest risk of relapse will allow the design of reasonable post-remission therapies using novel agents [50]. Our results suggested that CD304 was a stable MRD marker that could be useful in detecting MFC-based MRD monitoring, especially in high-sensitivity MRD assay [32]. Even though LAIP and DFN approach combination is one of the best for MRD assessment, its utility could be affected by the immunophenotypic patterns of leukemia blasts mimicking hematogones and in CD10 dim to negative cases. Hence, every case with a hematogone pattern and dim to negative CD10 expression at diagnosis is recommended to have a statement in the diagnostic flow cytometry report so that the hemato-pathologist viewing the report is aware of this. More immunophenotypic markers should be evaluated which can help in differentiating between hematogones and leukemia blasts, thereby improving the reliability of the MFC-based MRD assays. Changes in immunophenotypic markers in B-ALL post- induction are frequent and may be useful but such changes could possibly compromise the MRD assessment in certain cases [33].

MRD assessment for monitoring treatment outcomes was one of the objectives defined for this review. The Programa Español de Tratamientos en Hematología (PETHEMA) ALL-AR03 trial used MRD to guide treatment decisions at the EOC and found that HSCT could be avoided in patients who reached MRD negativity without adversely affecting their prognosis [51]. MRD can play a role in sparing patients from risks associated with transplantation without negatively affecting survival outcomes [52]. Patients who are at a high risk of leukemia relapse after allogeneic SCT can be identified by the kinetics of MRD clearance. Patients who have not been able to achieve early molecular remission after transplantation might require prompt and appropriate preventive treatments [53]. The role of MRD in the management of Ph-positive B-cell ALL has also been established and can be important in in-patient stratification [14]. The percentage of MRD reduction corresponds with superior disease-free survival (DFS), irrespective of the tyrosine kinase inhibitor (TKI) used [54]. MRD persistence and/or reappearance can be indicative of resistant mutations (e.g., T315I). Such cases may warrant alternative approaches, including novel TKIs and/or combinations of TKI with immunotherapy [55]. The early achievement of MRD negativity in the treatment of adults with Ph-negative B-cell ALL is a strong predictor of survival [56]. Ph-negative patients in this study were classified as B-cell ALL patients who had achieved MRD-negative status at the end of induction at two different time points and were also observed to be early MRD responders. MRD also has the potential to guide the selection of patients for treatment de-intensification. However, the appropriate way to utilize MRD results for treatment de-intensification is yet to be defined.

The value of having achieved MRD negativity is significant in pediatric and adult ALL patients [15]. In adult B-ALL patients, achieving MRD negativity is consistently associated with better survival outcomes than those of patients with MRD-positive status [1]. Such results have been consistent across methods, therapies, times of MRD assessment, cutoff levels, and disease subtypes [1]. Relatively few reports are available on the significance of MRD in patients with relapsed disease. Such status was also evident from the search carried out for this study. In adults with ALL, the prognostic significance of MRD in relapsed/refractory ALL has been primarily reported in individual studies using novel salvage treatments [57]. In a study of inotuzumab as salvage therapy, achieving MRD negativity was associated with a longer remission duration [58]. A retrospective analysis of 78 patients showed a differential impact of MRD negativity according to salvage (S) status in patients with relapsed/refractory B-ALL [57]. Patients with relapsed/refractory ALL who achieved MRD negativity in S1 had long-term survival, whereas patients in S2 generally had poor outcomes regardless of MRD status. Patients in S1 who achieved MRD negativity and subsequently underwent SCT had the best outcomes, with a 2-year OS rate of 65%. Assessment of the prognostic value of MRD negativity at the end of inotuzumab treatment shows that patients in first salvage who achieved MRD negativity experienced significantly improved survival vs. that seen in MRD-positive patients. This observation was significant, particularly among those patients who proceeded to SCT. Among patients with relapsed/refractory ALL treated with inotuzumab, the MRD-negative complete remission/complete remission with incomplete count recovery (CR/CRi) group had the best survival outcomes [59]. The benefit of achieving MRD negativity highlights its relevance for assessing prognosis and measuring treatment efficacy.

Although studies identified in the search reveal minimal data on applications of MRD, there is a broader research landscape with extended scope of utility. MRD response has been considered in drug development as an early marker of efficacy in clinical studies. It has potential use as a surrogate endpoint in the registration of studies for accelerated drug approval [60, 61]. MRD status warrants consideration as an early measure of disease response for evaluating new therapies, improving the efficiency of clinical trials, accelerating drug development, and regulatory approval [15]. However, approval of such findings based on an intermediate endpoint would require confirmation using traditional efficacy endpoints.

Conclusion

The systematic search carried out as a part of this study revealed limited data on applications of MRD in the management of B-cell ALL among AYA and adult populations. The existing data suggest its applicability in facilitating improved treatment outcomes. The comparison of results from included studies with the scope of published evidence from literature databases highlights the need for more research specific to Indian settings. Aspects related to cost, resource limitations, and differences in biology have been pointed out. These may be important considerations in designing future research investigations. Current evidence suggests that MRD is an essential tool to facilitate the optimal course of management of B-cell ALL by assisting in critical clinical decisions. Such assessments can effectuate the distinctness of situations where the use of conventional options has higher chances of treatment failure and identify patients who can benefit the most from novel agents.