Kinetics of minimal residual disease and chimerism in patients with chronic myeloid leukemia after nonmyeloablative conditioning and allogeneic stem cell transplantation

Allogeneic hematopoietic stem cell transplantation (SCT) is the treatment of choice for patients with chronic myeloid leukemia (CML) who have a suitable donor.1 In older patients and among those with comorbidity, however, there is a high risk of regimen-related toxicity that makes conventional stem cell transplantation (CST) unsuitable for these patients. In recent years, nonmyeloablative stem cell transplantation (NST) has been studied as a safer approach for older patients.2-4 Such transplantations are more dependent on a graft-versus-leukemia effect, which is known to be powerful in CML patients.1,5 

Molecular techniques for chimerism and minimal residual disease (MRD) analysis after SCT are routinely used in many laboratories to follow engraftment and predict a threatening relapse of leukemia. In the conventional SCT setting, such analyses have been well evaluated in CML patients.6-9 

With the introduction of nonmyeloablative transplantations, the significance of chimerism and MRD results to clinical outcome after allogeneic SCT needs to be reevaluated.10,11 

In the present study, we determined the incidence and kinetics of chimerism and MRD in 15 CML patients receiving NST and compared them with 10 patients receiving CST.

Patient characteristics for the 25 CML patients included in this study are shown in Table 1. Fifteen patients received a nonmyeloablative conditioning regimen and allogeneic SCT (NST) between April 1999 and April 2001. Of these 15 patients, 8 were either too old or considered unfit for conventional SCT (CST). Seven patients requested NST after shared decision making. Thirteen patients received CST between January 1999 and March 2001. Three patients were excluded because samples were not available. Of these, 1 patient relapsed 2 years after SCT and 2 patients are alive in clinical remission 18 and 19 months after SCT, respectively. The ethics committee at Karolinska Institute, Huddinge University Hospital, approved this study. Informed consent was provided according to the Declaration of Helsinki.

The NST patients received fludarabine (Flu) 30 mg/m²/d for 6 days, busulfan (Bu) 4 mg/kg/d for 2 consecutive days, and antithymocyte globulin (ATG) 2 to 10 mg/kg for 4 days.4 

The CST group was conditioned with cyclophosphamide (Cy) 120 mg/kg with 10 to 12 Gy total body irradiation (TBI) (n = 5) or busulfan 16 mg/kg (n = 5).12 Details regarding supportive care have been described elsewhere.13 

Samples for chimerism and MRD analyses were 5 to 10 mL of peripheral blood.

A median of 8 (range, 3-35) chimerism analyses and 10 (range, 3-30) MRD analyses was performed for each patient.

Polymerase chain reaction (PCR) amplification of variable number of tandem repeats (VNTRs) was used to evaluate various degrees of donor and recipient chimerism in CD3⁺, CD19⁺, and CD45⁺ cells as previously described.14 

Quantification was done by competitive PCR using plasmid constructs containing a modified BCR-ABL fusion gene.15 Dr N. C. P. Cross, Hammersmith Hospital, London, kindly provided pNC210/G (p210) and pNC190/G (p190) competitor plasmids.

BCR-ABL and ABL transcript numbers were estimated by comparing the competitor and sample band intensity to find the equivalence point. Results were expressed as the ratio betweenBCR-ABL and ABL transcript numbers(BCR-ABL/ABL).

RNA from K562 cells was serially diluted in RNA from HL-60 cells in a total amount of 20 μg RNA. After cDNA synthesis and 40 cycles of PCR amplification, a sensitivity of 10⁻⁶ was obtained.

In the NST group, 13 patients are alive with a median follow-up of 20 months (range, 6-29 months). Two patients died of graft-versus- host disease (GVHD) and progressive disease 5 and 17 months after SCT, respectively (Table 1). One patient (N1) relapsed 7 months after SCT.

In the CST group, 9 patients are alive with a median follow-up of 25 months (range, 9-34 months). One patient died of GVHD 7 months after SCT. Two patients relapsed 9 and 16 months after SCT, respectively.

T-cell mixed chimerism (MC) was detected in all 15 patients, and all but 2 (N7 and N15) converted to donor chimerism (DC) at the end of this study (Figure 1). The median time for T-cell DC to occur was 87 days (range, 28-145 days) and was significantly delayed (21 days [range, 14-60 days]) compared with the CST group (P < .01). Granulocyte and B-cell MC were found in 12 (80%) and 9 (60%) of the patients, respectively. Median times to granulocyte and B-cell DC were 29 days (range, 14-128 days) and 30 days (range, 14-190 days), respectively.

Five patients showed T-cell MC after SCT. Three patients had MC only the first month, whereas 2 patients (C9 and C10) had increasing levels of recipient cells in all cell fractions; both relapsed.

In accordance with other NST studies, we found a high incidence of MC.10,14,16 The engraftment of T cells lagged behind granulocytes and B cells, which is in agreement with some but not other studies.10,17 

All patients in this group had detectable BCR-ABLtranscripts after SCT. The BCR-ABL/ABL ratio during the first 3 months was, with a median of 0.2%, significantly more compared with 0.01% in CST patients (P < .01) (Figure 1). This probably reflects the lower antitumor effect of nonmyeloablative conditioning. Eleven patients became MRD negative within 7 months (median, 3.5 months; range, 1-7 months). Three patients (N6, N12, N15) were BCR-ABL positive during the entire posttransplantation period. However, in 2 patients, the follow-up was less than 9 months.

After SCT, 7 of 10 patients showed at least 1 positive PCR assay forBCR-ABL. The percentage of MRD-positive patients was 50% at 1 year (Figure 1). Long persistent expression of BCR-ABL in the absence of clinical relapse has been observed in many studies.6,18-21 For these patients, it is important to monitor the change in transcript levels with quantitative analysis.15,22 

A switch from T-cell MC to DC was usually seen before (n = 5) or at the time (n = 4) of MRD negativity. In the former group, the median interval between T-cell DC and MRD negativity was 40 days (range, 24-90 days). One patient (N3) had T-cell MC with no signs of theBCR-ABL transcript for 1 month. The short interval between T-cell DC and MRD negativity suggests that complete DC is not necessary for disease response.11 

Twelve patients with MC in the granulocyte cell population were all PCR positive for BCR-ABL. Among those, a BCR-ABL/ABLratio of at least 0.1% was found in 10.

MC beyond 1 month after SCT was seen in only 2 patients (C9 and C10). Both had increasing BCR-ABL levels, and MC in B cells and granulocytes appeared only at BCR-ABL/ABL ratios of more than 1%. Three patients (C2, C5, and C8) with persistentBCR-ABL transcript (ratio less than 0.1%) more than 1 year after SCT had DC in all cell fractions.

The incidence of acute GVHD was 73% and 90% in the NST and CST groups, respectively (Table 1). Interestingly, 4 NST patients without acute GVHD became MRD negative. This suggests that an allogeneic graft-versus-leukemia (GVL) effect may be seen even in the absence of GVHD, which is in accordance with a report in patients with acute leukemia.23 

In conclusion, despite high BCR-ABL levels during the early posttransplantation period and a high incidence of mixed chimerism, nonmyeloablative transplantation for CML patients may induce molecular remission in most patients. Close monitoring of chimerism and MRD is needed to guide the timely introduction of immune-therapeutic interventions.

We thank all staff at the departments for excellent and skillful patient care.