Clinical Decision Making in Chronic Myeloid Leukemia Based on Polymerase Chain Reaction Analysis of Minimal Residual Disease

To the Editor:

In a recent commentary, Faderl et al1 posed the question “Should polymerase chain reaction analysis to detect minimal residual disease in patients with chronic myelogenous leukemia be used in clinical decision making?” We believe that the answer to this question is indubitably “yes.” Although we agree that nonquantitative assays that only indicate the presence or absence of BCR-ABL transcripts are of limited value, the results of serial quantitative reverse transcription-polymerase chain reaction (RT-PCR) studies may be highly informative. For chronic myeloid leukemia (CML) patients in cytogenetic remission after treatment by allografting, the results of serial quantitative RT-PCR analysis can alert the clinician to the need for further antileukemia therapy. For CML patients treated with interferon-α (IFN-α), BCR-ABL transcript numbers very seldom fall below the level of detection, but their actual level correlates with the probability that remission will be maintained.

The principal aim of residual disease analysis in patients with CML is to assess a patient’s response to treatment and to recognize early signs of relapse. RT-PCR for BCR-ABL mRNA is by far the most sensitive assay in this context and can detect a single leukemia cell in a background of 10⁵ to 10⁶ normal cells. However, it is important to note that this high sensitivity may not be achieved routinely, and variations in the actual sensitivity for each specimen almost certainly account for many of the apparent discrepancies in the literature.2 Because nonquantitative RT-PCR analysis gives only limited information, several groups have developed quantitative or semiquantitative RT-PCR assays that enable the kinetics of residual disease to be monitored over time. The number of BCR-ABL transcripts in a specimen is typically normalized to the number of transcripts of a housekeeping gene, which serves to control for the integrity and quantity of patient cDNA. For RT-PCR–negative specimens, the control gene gives a clear indication of the sensitivity with which residual disease can be excluded for that particular specimen and thus provides meaning to a result that previously had been difficult to interpret. Because of the technical complexities of these assays, quantitative RT-PCR has only been performed in a limited number of centers. Nevertheless, several studies have clearly indicated the clinical utility of this approach.

After allogeneic bone marrow transplantation (BMT), serial quantitative RT-PCR analysis of peripheral blood specimens can effectively distinguish those patients who are destined to remain in remission from those who are destined to relapse. Patients who remain in remission after BMT have persistently undetectable, low, or falling BCR-ABL levels on sequential analysis. After 6 to 9 months, BCR-ABL is undetectable in most cases and remains so indefinitely. Such patients may never require any further examination of their bone marrow, a fact for which they are very grateful. Other patients may remain intermittently or persistently RT-PCR positive for prolonged periods of time without evidence of cytogenetic relapse. The level of detectable BCR-ABL transcripts in these individuals is usually very low. In contrast, for patients destined to relapse, increasing or persistently high levels of BCR-ABL mRNA can be detected on sequential analysis, often several months before the cytogenetic detection of the Philadelphia (Ph) chromosome in bone marrow metaphases. The recognition of relapse at the molecular level provides a window for therapeutic intervention while the burden of disease is still relatively low.3-5 To take account of the fact that low-level RT-PCR positivity after allografting may not herald clinically relevant disease recurrence, we have established specific criteria for molecular relapse.2 Patients who meet these criteria receive treatment by donor lymphocyte infusions (DLI), if possible before the onset of hematologic relapse, because the evidence suggests that early use of DLI is more effective than if the treatment is delayed.6 The great majority of patients who respond to DLI achieve durable molecular remission (RT-PCR negativity), with a median follow-up of more than 2 years.7 

After treatment with IFN-α, virtually all patients remain RT-PCR positive despite the fact that many achieve complete cytogenetic remission. For patients in continuing cytogenetic remission, quantitative RT-PCR has demonstrated that the levels of detectable residual disease may vary between patients by as much as 10,000-fold.8-10 The actual level of residual BCR-ABL transcripts is related to the probability of relapse, opening up the possibility that molecular monitoring may identify a subset of patients for whom treatment may be safely withdrawn.11 

Based on current evidence, we believe that quantitative RT-PCR has clear utility for early detection of relapse after allogeneic BMT and is useful to gauge patient response after other treatments. Some of the theoretical arguments that have been used1 against the use of RT-PCR are, in fact, specious. For example, although variant BCR-ABL fusions are seen in a small proportion of CML patients, there is no evidence for the evolution of subclones expressing different fusion products from the original clone. Similarly, the evidence for a neoplastic clone before the acquisition of the BCR-ABL fusion, although provocative, is weak and not widely accepted. Moreover, although it is true that very low levels of BCR-ABL can be detected in some normal individuals using modified and extremely sensitive PCR assays, this is not expected to impinge significantly on the routine analysis of minimal residual disease.12 

New real time quantitative RT-PCR procedures promise to greatly simplify the cumbersome protocols that are currently in use. They also offer a unique opportunity to standardize the assay and to develop rigorous standards and controls. We believe that quantitative RT-PCR will shortly become a routine and robust basis for clinical decision making in CML and other malignancies.