Definition
Minimal residual disease(MRD) in leukemia refers to disease that is present in patients following therapy below the level of conventional morphologic detection. Because morphologic remission is conventionally defined as less than 5% blasts, patients with remission marrows can still harbor a burden of more than 10¹⁰ leukemic cells. In patients with detectable blasts below the level of 5%, karyotyping, fluorescence in situ hybridization, or conventional flow cytometry can often help to distinguish regenerating marrow blasts from leukemic cells. However, as typically used, minimal residual disease customarily means levels of disease considerably below this.

Methods for Detection of Minimal Residual Disease
Minimal residual disease in ALL can be detected either by molecular techniques using the polymerase chain reaction, or by multiparameter flow cytometry. The former has been most widely employed.

Molecular MRD analysis
There are two broad strategies that can be employed to detect MRD in ALL. The first of these is technically the most straightforward, and involves RT-PCR detection of chimeric transcripts encoded by several leukemia-specific translocations. In childhood, the 4 translocations BCR-ABL;AF4-MLL;E2A-PBX1; and TEL-AML1 account for about 30% of childhood ALL, though this figure is misleadingly high as TEL-AML1 is by far the most common translocation, accounting for 20-25% of ALL. While technically, these aberrant transcripts can be detected at high sensitivity, and can be readily quantified using real time quantitative PCR methods, in practice they have not proved that valuable because of their limited applicability; this is especially true because TEL-AML1 patients have a particularly good prognosis, and a very low frequency of MRD positivity.

For these reasons, most work has involved the use of antigen receptor targets. Immunoglobulin heavy and light chain genes, and T cell receptor genes are rearranged in a high proportion of cases of ALL. Specifically, IgH rearrangements are found in greater than 95% of patients, with TCR-delta in approximately 90%; and TCR-gamma and Igkappa in about 60%. Suitability of antigen receptor rearrangements as targets for MRD detection depends upon more than just their presence. Sensitivity of detection can be limited based on nonspecific amplification of background normal cells, particularly when T cell targets are used. The most significant limitation occurs because of clonal evolution. Ongoing rearrangements during treatment will often result in the loss of a particular target; clonal evolution is more common when oligoclonal rearrangements are seen at diagnosis, something that occurs in 20-40% of cases. Stability of different targets has been extensively studied. It has been shown that IgK rearrangements involving kappa deleting elements are the most stable targets, with 90% of targets persisting after therapy; as these rearrangements are found in about 50% of precursor B-ALL, they are the single most useful target for MRD studies. However, it is recommended that in all cases at least two targets be used.

Assay of MRD using antigen receptor targets is more cumbersome than detection of chromosomal translocations because of the need to synthesize allele-specific oligonucleotides to use either as probes or primers. At diagnosis, consensus primers are used to amplify the rearrangements; there are now well-characterized multiplex reactions that can efficiently amplify all known targets. However, once the target is amplified, it is necessary to sequence the breakpoint and synthesize the relevant clone-specific oligonucleotide. This is a time-consuming and expensive process that has limited the use of this technology to a few well-funded clinical trials.

Detection of MRD traditionally relied on semiquantitative assays such as limiting dilution, or competitive PCR analysis. More recently, this has been adapted to "Real-Time" quantitative PCR analysis, using either the ABI system, or Light Cycler technology; both of these technologies appear to give equivalent results. These methodologies can either use consensus primers and a fluorogenic allele specific probe, or else use an allele-specific oligonucleotide as the forward primer; the latter technology has been suggested to be more sensitive.

Advantages and Disadvantages of Molecular Testing for MRD
The advantages of using molecular methods to detect chromosomal translocations are that the methods are straightforward, highly sensitive, and easy to quantitate. However, as noted, these are only applicable to a small fraction of children with ALL. Nevertheless, this is probably the method of choice for assessing MRD in Ph+ ALL. Antigen receptor PCR is applicable to nearly all children with ALL. It too, is highly sensitive, with detection limits of 10^-4 to 10^-5. Another advantage is that the procedures are easy to standardize. The outstanding work of the European BIOMED initiative, under the leadership of Jacques van Dongen, has specified primer sets and reaction conditions, and have show that these studies can be reproducibly carried out at a number of different laboratories.

The major disadvantage of this technology, as noted above, is cost, both in materials and in sheer time required to synthesize allele-specific probes and primers. It has been estimated that costs of these studies range up to $3000 per patient. Moreover, most studies to date have been retrospective. The logistical complications involved in producing a reagent and having it available to assess samples early on in therapy, so that therapy can be changed based on results, is now only beginning to be evaluated in prospective trials.

Flow Cytometric Detection of MRD
The other major technique used in assessment of MRD is flow cytometry. This technique is based on the principle that leukemic cells have signature phenotypes that differ from those of normal cells. This phenotypic aberrancy can take one of two forms. In some cases, leukemic cells express novel antigens not seen in normal differentiation. For example, many cases of precursor B-ALL will coexpress CD10 and CD66b(KORSA-3544), a phenotype not seen in any normal cells. Similarly, the great majority of T-ALL cases will coexpress the aberrant combination cytoplasmic CD3/TdT.

However, in more cases, abnormal cells can be recognized because they express normal antigens at intensities that are inappropriate for normal maturation. This is particularly true in B cell maturation, where normal B cell precursors express a precise and reproducible sequence of maturation antigens. With a relatively limited number of 3- or 4- color combinations, leukemic cells can be seen to occupy regions of so-called "empty space" on bivariate displays, where normal B cell precursors do not reside. Particularly useful markers include overexpression of CD10 (particularly in combination with CD20); underexpression of CD45; overexpression of CD34; underexpression of CD38; overexpression of CD58; overexpression of CD34; overexpression of CD9; overexpression of TDT. Combinations of markers will accentuate differences between normal and leukemic cells. For example, coexpression of CD34 and CD20,

though not a common combination in ALL, is an aberrant phenotype not seen in normal. With a relatively limited panel of antibodies, at least 90% of cases of ALL will have at least one phenotypically aberrant combination that can allow detection of MRD. The sensitivity of detection is on the order of 10^-4. However, sensitivity depends to a degree on how aberrant the phenotype is, and how many background normal cells of a similar phenotype are present in the sample. This in turn is a function of the time at which the assay is performed; fortunately, at end of induction, where MRD assessment appears most useful (see below) there are rarely significant numbers of normal B cell precursors present to interfere with the analysis.

Advantages and disadvantages of flow cytometry
The major advantage of flow cytometry is the rapidity of assay. An MRD result can be available within a few hours of receipt of sample. It is also far less costly than molecular results that rely on sequencing and synthesis of allele-specific probes or primers. High sensitivity requires assaying 500,000-1,000,000 cells, which in turn requires higher volumes of reagents than routine flow cytometry, and longer acquisition times, but even so, flow cytometry assessment of MRD is on the order of ten fold less expensive than molecular testing.

Although sensitivities of 10^-5 can be approached in rare cases with very aberrant phenotypes and clean backgrounds, in general this technique is not as sensitive as molecular methods. The other major disadvantage is that there is far less standardization of flow cytometry. Most studies have been done in single laboratories, so there is little information on interlaboratory reproducibility. Something that is not often discussed, but which creates a problem, is that of the denominator. Acquiring large numbers of events also means introducing some debris that cannot always be reproducibly excluded. Older samples, with degenerating granulocytes, can be particularly problematic. Some flow studies have relied on ficoll hypaque separation to clean up their specimens, while others have used nucleic acid binding dyes to correct their percentages. However, to date no standard method exists.

The other limitation to flow studies is the possibility of changes in phenotype during progression of the disease. This phenomenon clearly occurs, but the extent to which it creates problems for MRD monitoring is uncertain. Persistent leukemic cells can show loss of intensity of expression of CD34, of CD10, and of CD66b, among other markers. Moreover, phenotypically heterogeneous populations can be more homogeneous in persisting disease; this is particularly true of CD20, which is often heterogeneously expressed in ALL, with the brighter CD20+ subpopulation often persisting. For these reasons, it is important to monitor more than one aberrant phenotype in assessing ALL.

Comparison of flow and molecular methods
Several studies have performed direct comparison of flow and molecular methods for MRD detection. In general, these have been highly correlated. However, in all series there are small numbers of cases that show significant discrepancies. Reasons for these discrepancies have never been adequately explained. Obviously there can be differences in sampling; it has been shown that in some cases, there can be a log difference in the presence of B cell precursors depending upon whether the first or second pull marrow is used. In addition, it is important to note that flow cytometric methods detect abnormal viable cells, while DNA-based methods could, in principle, be detecting DNA from apoptotic cells not present in the flow assays. In addition, it has been pointed out that agreement of the two assays around critical action values (generally 0.1% see below) is not all that good. Thus, acting on MRD results outside the context of a clinical trial with defined methodology is probably not appropriate at this time. Clinical studies of MRD in childhood ALL. With the exception of a few early studies where methodology might be questioned, virtually all studies have shown that the presence of MRD in ALL is of adverse prognostic significance. This appears to be true throughout therapy. In addition, quantitation is important; patients with higher levels of MRD fare worse than those with low levels. Most studies have now focused on detection of MRD relatively early in therapy. There are several reasons for this. First, the number of patients with persistent MRD late in therapy is very small, so that even though their prognosis appears poor, there have not been large numbers of such patients studied. More significantly, however, it is apparent that early intervention in therapy has the greatest chance of potential cure. For this reason, most of the effort in MRD detection has focused on end-of-induction, or shortly thereafter in consolidation, typically around week 12.

There is still variability in the timing and level of MRD that is considered most significant. Frequency of MRD positivity varies from 25-75% depending upon exact definition of end-induction, on method, and on therapeutic protocol. In addition, the level of clinical significance of MRD has varied from 10^-2 in one PCR-based study, to 10^-3 in another flow based study. Still other studies have shown a continuous increase in risk at all levels. The largest molecular series to date defines 3 risk groups: patients who are negative at both end induction and week 12 have an excellent prognosis, while those who are positive (>0.1%) at both time points have a very poor one; other patients had an intermediate prognosis.

Interpretation of prognostic significance is further complicated by underlying prognostic factors. Age, white count, chromosomal abnormalities, and initial response to therapy, measured either by decrease in peripheral blood blast count or by day 8 or day 15 marrow morphology all have prognostic significance, and the frequency of MRD positivity is different in these different groups. For example, patients with the Ph chromosome, a very poor risk group, have a very high incidence of MRD, while those with Tel-AML1, a good prognosis lesion, have very low levels. Curiously, patients with hyperdiploidy, also a good prognosis group, have a relatively high frequency of MRD positivity, suggesting that the prognostic significance of MRD may be subgroup dependent. Nevertheless, most studies suggest that MRD is an important risk factor even when other prognostic factors are accounted for.

Finally, although it is hoped that identifying patients with MRD may be useful, there are no studies to date that have demonstrated that these poor prognosis patients can be rescued by changing therapy. However, it is well known that patients with slow initial response to chemotherapy can improve their prognosis if therapy is intensified shortly after induction. Because there is a significant correlation between end-induction MRD and conventional measures of response, it is attractive to think that intensifying patients with MRD will result in an improved treatment outcome. Several such studies are underway.

References

Molecular studies

1. van Dongen JJ, Seriu T, Panzer-Grumayer ER, et al. Prognostic value of minimal residual disease in acute lymphoblastic leukaemia in childhood. Lancet. 1998 352:1731-
2. Cave H, van der Werff ten Bosch J, et al. Clinical significance of minimal residual disease in childhood acute lymphoblastic leukemia. European Organization for Research and Treatment of Cancer--Childhood Leukemia Cooperative Group.N Engl J Med. 1998 339:591-8.
3. Gameiro P, Moreira I, Yetgin S, et al. Polymerase chain reaction (PCR)- and reverse transcription PCR-based minimal residual disease detection in long-term follow-up of childhood acute lymphoblastic leukaemia. Br J Haematol. 2002 119:685-96.
4. Pongers-Willemse MJ, Verhagen OJ, et al. Real-time quantitative PCR for the detection of minimal residual disease in acute lymphoblastic leukemia using junctional region specific TaqMan probes.Leukemia. 1998 12:2006-14.
5. Szczepanski T, Willemse MJ, Brinkhof B, et al. Comparative analysis of Ig and TCR gene rearrangements at diagnosis and at relapse of childhood precursor-B-ALL provides improved strategies for selection of stable PCR targets for monitoring of minimal residual disease. Blood. 2002;99:2315-23.
6. van der Velden VH, Hochhaus A, Cazzaniga G, et al. Detection of minimal residual disease in hematologic malignancies by real-time quantitative PCR: principles, approaches, and laboratory aspects. Leukemia. 2003 17:1013-34.
7. Li A, Zhou J, Zuckerman D, Rue M, et al. Sequence analysis of clonal immunoglobulin and T-cell receptor gene rearrangements in children with acute lymphoblastic leukemia at diagnosis and at relapse: implications for pathogenesis and for the clinical utility of PCR-based methods of minimal residual disease detection. Blood. 2003 102:4520-6.
8. van der Velden VH, Willemse MJ, van der Schoot CE, et al. Immunoglobulin kappa deleting element rearrangements in precursor-B acute lymphoblastic leukemia are stable targets for detection of minimal residual disease by real-time quantitative PCR. Leukemia. 2002 16:928-36.
9. Pine SR, Moy FH, Wiemels JL, et al. Real-time quantitative PCR: standardized detection of minimal residual disease in pediatric acute lymphoblastic leukemia. Polymerase chain reaction. J Pediatr Hematol Oncol. 2003 25:103-8.
10. Eckert C, Scrideli CA, Taube T, et al. Comparison between TaqMan and LightCycler technologies for quantification of minimal residual disease by using immunoglobulin and T-cell receptor genes consensus probes.Leukemia. 2003 17:2517-24.
11. Cazzaniga G, Lanciotti M, Rossi V, Prospective molecular monitoring of BCR/ABL transcript in children with Ph+ acute lymphoblastic leukaemia unravels differences in treatment response. Br J Haematol. 2002 119:445-53.
12. Seyfarth J, Madsen HO, Nyvold C, et al. Post-induction residual disease in translocation t(12;21)-positive childhood ALL. Med Pediatr Oncol. 2003 40:82-7.
13. Nyvold C, Madsen HO, Ryder LP, et al. Precise quantification of minimal residual disease at day 29 allows identification of children with acute lymphoblastic leukemia and an excellent outcome. Blood. 2002 99:1253-8.
14. Eckert C, Biondi A, Seeger K, et al. Prognostic value of minimal residual disease in relapsed childhood acute lymphoblastic leukaemia. Lancet. 2001;358:1239-41.
15. van der Velden VH, Joosten SA, Willemse MJ, et al. Real-time quantitative PCR for detection of minimal residual disease before allogeneic stem cell transplantation predicts outcome in children with acute lymphoblastic leukemia. Leukemia. 2001 15:1485-7
16. Schmiegelow K, Nyvold C, Seyfarth J, et al. Post-induction residual leukemia in childhood acute lymphoblastic leukemia quantified by PCR correlates with in vitro prednisolone resistance. Leukemia. 2001 15:1066-71.
17. Willemse MJ, Seriu T, Hettinger K, et al. Detection of minimal residual disease identifies differences in treatment response between T-ALL and precursor B-ALL. Blood. 2002; 99:4386-93.
18. van der Velden VH, Jacobs DC, Wijkhuijs AJ et al. Minimal residual disease levels in bone marrow and peripheral blood are comparable in children with T cell acute lymphoblastic leukemia (ALL), but not in precursor-B-ALL. Leukemia. 2002; 16:1432-6.

19. Coustan-Smith E, Sancho J, et al. Clinical importance of minimal residual disease in childhood acute lymphoblastic leukemia. Blood. 2000 96:2691-6.
20. Campana D, Neale GA, Coustan-Smith E, et al. Detection of minimal residual disease in acute lymphoblastic leukemia: the St. Jude experience. Leukemia. 2001 15:278-9.
21. Dworzak MN, Froschl G, Printz D, et al. Prognostic significance and modalities of flow cytometric minimal residual disease detection in childhood acute lymphoblastic leukemia. Blood. 2002;99:1952-8.
22. Bjorklund E, Mazur J, Soderhall S, et al. Flow cytometric follow-up of minimal residual disease in bone marrow gives prognostic information in children with acute lymphoblastic leukemia. Leukemia. 2003 17:138-48.
23. Lucio P, Gaipa G, van Lochem EG, et al.. BIOMED-I concerted action report: flow cytometric immunophenotyping of precursor B-ALL with standardized triple-stainings. BIOMED-1 Concerted Action Investigation of Minimal Residual Disease in Acute Leukemia: International Standardization and Clinical Evaluation. Leukemia. 2001 15::1185-92.
24. Porwit-MacDonald A, Bjorklund E, Lucio P, et al. BIOMED-1 concerted action report: flow cytometric characterization of CD7+ cell subsets in normal bone marrow as a basis for the diagnosis and follow-up of T cell acute lymphoblastic leukemia (T-ALL).Leukemia. 2000 14:816-25.
25. Weir EG, Cowan K, LeBeau P, et al. A limited antibody panel can distinguish B-precursor acute lymphoblastic leukemia from normal B precursors with four color flow cytometry: implications for residual disease detection. Leukemia 1999; 13:558-567.
26. Coustan-Smith E, Sancho J, Hancock ML, et al. Use of peripheral blood instead of bone marrow to monitor residual disease in children with acute lymphoblastic leukemia. Blood. 2002;100:2399-402.
27. Prognostic importance of measuring early clearance of leukemic cells by flow cytometry in childhood acute lymphoblastic leukemia. Blood. 2002;100:52-8.
28. Borowitz MJ, Pullen DJ, Shuster JJ, et al. Minimal residual disease detection in childhood precursor-B-cell acute lymphoblastic leukemia: relation to other risk factors. A Children's Oncology Group study. Leukemia. 2003; 17:1566-72.
29. Chen JS, Coustan-Smith E, Suzuki T, et al. Identification of novel markers for monitoring minimal residual disease in acute lymphoblastic leukemia. Blood. 2001 97:2115-20.
30. Veltroni M, De Zen L, Sanzari MC, et al. Expression of CD58 in normal, regenerating and leukemic bone marrow B cells: implications for the detection of minimal residual disease in acute lymphocytic leukemia. Haematologica. 2003; 88:1245-52.
31. Tomova A, Babusikova O. Shifts in expression of immunological cell markers in relapsed acute leukemia. Neoplasma. 2001;48:164-8.
32. Malec M, Bjorklund E, Soderhall S, Flow cytometry and allele-specific oligonucleotide PCR are equally effective in detection of minimal residual disease in ALL. Leukemia. 2001;15:716-27.