For many years, hematologic response was the goal of therapy with agents such as hydroxyurea and busul-fan. The most important legacy of the IFN-a era was to demonstrate that achieving a cytogenetic response would prolong the survival of patients with CML.834 The goal of therapy became achieving a major, and particularly a complete cytogenetic response. With most patients achieving a complete cytogenetic response on imatinib, the goal of therapy is shifting again now toward achieving a molecular response. Eliminating all evidence of minimal residual disease (at least according to currently available techniques) is associated with improved probability of long-term remissions in many tumors. This is clearly true also in CML.

The clinical significance of molecular monitoring in CML is best documented in patients treated with an SCT. Patients who have BCR-ABL detectable by PCR, particularly if later than 6 months after transplant, have a significantly higher probability of relapse.35 36 Furthermore, Q-PCR has become an important tool to determine the risk of relapse. Patients with the highest levels of BCR-ABL transcripts have the highest risk of relapse.35,36 Molecular monitoring with Q-PCR is thus routine after SCT. Indeed, early intervention (e.g., donor lymphocyte infusion) is frequently indicated based on increasing levels of BCR-ABL transcripts, with the best results obtained when intervention occurs at the first evidence of relapse. Among patients treated with IFN-a, there is a considerable heterogeneity in the levels of BCR-ABL transcripts detectable among those who achieve a complete cytogenetic response. However, the risk of relapse is minimal for patients with levels below the median, while most of those with higher levels lose their response.9 Indeed, approximately one-third of patients who achieve a complete cytoge-netic response have undetectable BCR-ABL transcripts by nested PCR, and none of these patients has lost this response after 10 years.8

Despite the relatively brief follow-up available for patients treated with imatinib, there is mounting evidence that molecular monitoring has important clinical implications. The IRIS trial demonstrated a significantly improved molecular response with imatinib compared to IFN-a at each time point from the achievement of complete cytogenetic response.37 Overall, after a median follow-up of 18 months, nearly 40% of the patients treated with imatinib had a 3-log reduction of BCR-ABL transcripts, and 4-10% had unde-tectable BCR-ABL levels. This magnitude of response was rare among patients treated with IFN-a. This may have been due, in fact, to the short follow-up for patients treated with IFN-a, as over 80% of patients crossed over to the imatinib arm after a short period of time. Still, with the significantly higher rate of complete cytogenetic responses achieved with ima-tinib, it is likely that a significant difference would have been observed even if patients had continued IFN-a therapy. Furthermore, obtaining a molecular response earlier has important clinical implications. Among patients who achieved a complete cytoge-netic response with imatinib, those who had at least a 3-log reduction in BCR-ABL transcripts at 12 months from the start of therapy had a significantly better probability of progression-free survival compared to those who had a less pronounced reduc-tion.37 In another study of patients in chronic phase, who had achieved a complete cytogenetic response after imatinib therapy, those who achieved levels of BCR-ABL/ABL ratio of <0.05% (i.e., major molecular response) had a significantly lower probability of losing their cytogenetic response compared to those who did not reach these levels.38 The difference was even more significant when considering only patients who achieved a complete molecular response (i.e., undetectable BCR-ABL). Similarly, those who had achieved a major molecular response after 12 months of therapy had a significantly better predicted probability of a sustained complete cytogenetic remis-sion.38 A third study with a relatively smaller cohort of patients confirmed a longer duration of complete cytogenetic response for patients who had a maximum reduction of BCR-ABL transcripts of at least 2 logs.39 It has also been suggested that patients who have lower levels of transcripts detectable after the first few months of therapy with imatinib have a significantly lower probability of developing point mutations conferring resistance to imatinib. Thus, it is evident from these studies that the objective of therapy in the imatinib era should be to achieve at least a major molecular response, and that achieving this response early increases the probability of a long-term durable response.

A recommended algorithm for monitoring patients receiving therapy with imatinib is presented in Table 18.1. At the time of diagnosis, it is recommended that all patients have a regular karyotype as well as Q-PCR

Table 18.1 Recommendations for monitoring of patients treated with imatinib

Timing Recommended tests and frequency

At diagnosis • Cytogenetics, Q-PCR, ?FISH

First year • Cytogenetics every 3-6 months

After first year • Cytogenetics every 6-12 months

Q-PCR = quantitative polymerase chain reaction; FISH = fluorescent in situ hybridization.

(real-time PCR) done. As will be discussed later in this chapter, the first 12 months of therapy may be the most critical to establish the long-term prognosis of the patient, and therefore to make treatment decisions. Thus, during these first months, repeat monitoring is recommended every 3-6 months. This should include Q-PCR and cytogenetics. The major inconvenience of monitoring with cytogenetics is the need for a bone marrow aspiration. However, this is the only test to date that gives reliable information regarding other chromosomal abnormalities. The presence of additional chromosomal abnormalities may reduce the probability of response to imatinib and the overall survival.40-42 In addition, 6-8% of patients who respond to imatinib may develop chromosomal abnormalities in the Ph-negative metaphases.42-44 Although the long-term implications of these abnormalities are still uncertain, it is important to recognize and follow this phenomenon. After the first 12 months of therapy, Q-PCR should be performed at least every 6-12 months and a routine cytogenetic analysis every 12 months. Fluorescent in situ hybridization can be used to monitor the cytogenetic response between cytoge-netic analysis, as it can be done in peripheral blood. However, even with the newest probes, there is a small percentage of false positivity. For the determination of molecular response, a Q-PCR (i.e., real-time PCR) is needed. Unfortunately, this test is not widely available. Furthermore, there is considerable heterogeneity between the reports from different laboratories, and the results from many laboratories that offer this test have not been clinically validated. In addition, the analysis of the results from the IRIS trial, the largest to date to report on molecular monitoring, initiated molecular monitoring only after the patients achieved complete cytogenetic response. To determine the log reduction at different time points, a baseline was derived from results obtained from 30 patients. Thus, a 3-log reduction in fact corresponds to levels of <0.036%, which is a 3-log reduction from the standardized base line of 36% for the sample cohort. Using each patient's individual baseline value to determine the 3-log reduction has not been validated as an important endpoint.

0 0

Post a comment