Monitoring of post-transplant remission of childhood malignancies: is there a standard?

Abstract

Chimerism analysis has become an important tool for the peri-transplant surveillance of engraftment. It offers the possibility to realize imminent graft rejection and it can serve as an indicator for the recurrence of the underlying malignant or non-malignant disease. In addition to this analysis, the characterization of residual disease (MRD) prior to and in the course of follow-up post transplant has become an important prognostic factor to highlight patients at highest risk for relapse. Consecutive post transplant MRD monitoring, together with chimerism analysis, allows the detection of impending relapse in a substantial group of children transplanted for acute leukemia. Consequently, these investigations have become the basis for treatment intervention, for example, to avoid graft rejection, to maintain engraftment and to treat imminent relapse by pre-emptive immunotherapy.

Introduction

During the past three decades, BMT and transplantation of PBSC have become well-established treatment procedures for many malignant and non-malignant disorders in children and in adult patients. In children transplanted for acute leukemia, relapse remained the major cause for treatment failure. In relapsed patients with acute leukemia, further treatment options to achieve long-lasting remission are mainly limited to additional SCT. Donor lymphocyte infusion following chemotherapy is an instrument of increasing importance to treat or to prevent overt relapse as Kolb et al.¹ first showed its efficacy in patients with CML relapsing after allogeneic SCT. Since then, immunotherapy has shown that donor lymphocyte infusion initiated during frank hematologic relapse induces complete remission in 8% of patients with ALL and in 22% of patients with AML.² If tumor burden is reduced by chemotherapy before donor lymphocyte infusion, the rate of complete response is significantly improved (to 33% in ALL and 37% in AML).³ These results suggested that immunotherapy offers the greatest benefit to patients with acute leukemia following SCT when it is administered before hematologic relapse occurs. Therefore, it is desirable to identify patients who are at greatest risk of relapse early, so that additional immunotherapy can be delivered as prophylaxis.

Methods for post transplant monitoring

For post-transplant surveillance of disease remission, two different techniques are available in principle: (i) specific detection of minimal residual disease (MRD) or (ii) characterization of post-transplant chimerism. The MRD technique follows the malignant clone, where as chimerism assessment characterizes the origin of post-transplant hematopoiesis. During the past decade, different methods have been developed to detect MRD as well as chimerism. The techniques range from PCR to immunophenotypic analysis (flow) of aberrant cell surface molecules in both ALL and AML patients. For more information, see reviews by Szcepanski et al.,⁴ Bader et al.⁵ and Chung et al.⁶

Chimerism in malignant diseases

Molecular evidence of persisting or reappearing recipient cells may be a reflection of either survival of leukemic cells or of survival of normal host hematopoietic cells or a combination of both. Surviving host hematopoietic cells may, in turn, facilitate the re-emergence of a malignant cell clone by inhibiting immune competent donor effector cells. For patients with CML, it could be clearly demonstrated that reappearance of host hematopoietic cells in the mononuclear cell fraction preceded hematological relapse.⁷ Therefore, mixed chimerism has been considered to reduce the GVL effect in this particular group of patients.^{7, 8}

In patients with acute leukemias and myelodysplastic syndrome, several early studies have left the question unanswered whether patients with mixed chimerism do have an increased risk of relapse. In the middle of the 1990s, it was realized that evolution of chimerism is a dynamic process and that chimerism analysis should be done serially at short time intervals. By using short tandem repeat (STR)-PCR-based serial analysis of microsatellite regions at short time intervals, it could be shown that patients with rapidly increasing mixed chimerism have the highest risk of relapse.^{3, 9} These reports could be confirmed by others,^{10, 11, 12} whereas some studies did not find a correlation between chimerism and relapse.¹³ These discrepancies may partly be explained with sampling protocols used in the studies. The investigations of subpopulations in patients with acute leukemias showed that there might also be a difference between adult and pediatric patients. Guimond et al.¹⁴ demonstrated that mixed chimerism in T and natural killer cell subpopulations can frequently be found in pediatric patients with leukemia relapse but not in children in remission. In contrast, mixed chimerism in these subsets was not found in adult patients with relapse. In our own study, we showed that persistent mixed chimerism in the early post-transplant period is caused predominantly by normal recipient hematopoietic cells. Its increase precedes the reappearance of the underlying disease. These findings therefore support the hypothesis that a state of mixed hematopoietic chimerism may also reduce the clinical GVL effect of alloreactive donor-derived effector cells in patients with acute leukemias and myelodysplastic syndrome and thus facilitate the proliferation of residual malignant cells that may have survived the preparative regimen. Barrios et al.¹⁵ proved in 133 patients with acute leukemia that patients with increasing mixed chimerism have a significantly elevated risk of relapse. On the basis of these studies, several consecutive trials were initiated, evaluating the possibility of preventing relapse by pre-emptive immunotherapy on the basis of chimerism analysis in patients with acute leukemias.^{10, 12} Most recently, our group showed in 163 children with ALL that STR-based chimerism analysis at short time interval is able to define a large cohort of children with impending relapse and also that overt relapse, in principle, could be prevented by pre-emptive immunotherapy on the basis of increasing mixed chimerism.¹⁶ However, these analyses also showed, that (i) it was not possible to identify impending relapse in all patients and (ii) the time interval between the identification of chimerism and relapse can be very short.

Chimerism analysis does provide information about alloreactivity and/or tolerance induction of the graft and thereby serves more likely as a ‘prognostic factor’ than as an indirect marker for MRD. Moreover, it is, important to stress that because of its low sensitivity (approximately 1%), chimerism analysis is not a reliable procedure for the detection of MRD.

Minimal residual disease

In children transplanted for ALL, the level of MRD before transplant has a significant impact on post-transplant outcome. During the past years, two major reports have been published presenting the importance of MRD before allogeneic SCT as a prognostic marker in pediatric patients with ALL.^{17, 18} These retrospective reports have shown that MRD prior to conditioning is the strongest predictor for relapse post SCT. In the first report from the Bristol group, the authors report that patients entering the transplant with a high MRD load of >10⁻³ did not survive their disease and that patients with a low-level disease (<10⁻³) had significantly poorer outcome than those who entered the transplantation while MRD was negative.¹⁷ By using the same semiquantitative technique in our own retrospective analysis, we could largely confirm these results in a series of 45 children. Patients with high-level MRD at the time of transplant (>10⁻³ malignant cells in the background of non-malignant cells) could rarely be cured. Most likely, neither the conditioning nor the alloreactive potential of the graft could clear the disease. In this group, relapse is also occurring although patients are complete chimeras throughout the follow-up. In contrast, in patients who have a low MRD burden (<10⁻³ malignant cells in the background of non-malignant cells), residual disease can be controlled by a conversion of mixed chimerism to complete chimerism, for example, by pre-emptive immunotherapy. This is illustrated in Figures 1 and 2. MRD should be monitored using disease-specific PCR techniques as for example, TCR- or Ig-gene rearrangements for ALL and BCR/ABL fusion mRNA transcripts for CML. For review, see Gabert et al.¹⁹ and Velden et al.²⁰ There are more and more reports about the importance of MRD measurements after allogeneic transplantation in patients with leukemia. However, the results of large prospective trials within the International-Berlin-Frankfurt-Muenster-Group are pending.

Figure 1

The course of post-transplant follow-up is given of a 5-year-old boy with chronic ALL who received haploidentical transplantation in CR2. This patient has had a high level (>10⁻³) of MRD prior to transplant. The conditioning regimen could not substantially reduce the leukemia burden, and such high leukemia load could not be controlled by the alloreactive potential of the graft. The patient relapsed on day 168 post transplant. MRD=minimal residual disease; PB=peripheral blood.

Figure 2

The course of a 16-year-old girl with chronic ALL who was transplanted in CR2 from an HLA-identical unrelated donor. This patient received a low-dose DLI 1 × 10⁵/kg body weight, when increasing mixed chimerism was developed. This immunotherapy led to a conversion of mixed to complete chimerism and by then MRD was cleared. DLI=Donor lymphocyte infusion; MRD=minimal residual disease; PB=peripheral blood.

When a disease-specific marker is not available, for example, regularly in patients with AML, chimerism analysis in cell subpopulations may serve as a surrogate marker for MRD. A very elegant approach was presented by Thiede et al.²¹ showed that mixed chimerism in CD34-positive cells is predictive for relapse in patients with AML and ALL in peripheral blood. They showed that increasing autologous cells within this subset precedes relapse with a median interval of 52 days (range 12–97). To enrich this rare subpopulation in the periphery, however, 50 ml of blood is required, which limits the applicability of this procedure to adult patients. Mattsson et al.²² performed a prospective analysis in 30 patients with AML and myelodysplastic syndrome. They have investigated chimerism in CD33-, CD7- or CD45-positive cells and found significantly more relapses in patients whose subpopulations were mixed, compared with patients with complete chimerism. In ALL patients, several studies have been performed evaluating the impact of mixed chimerism after enrichment of the cell population carrying the leukemic phenotype (possible targets could be: CD10, CD19 and CD34 for precursor B-ALL and CD3, CD4, CD5 and CD8 for T-lineage ALL).²³ These studies showed a remarkable correlation between MRD and mixed chimerism in the respective subset. However, large studies in ALL patients, indicating the predictive value of mixed chimerism in different subsets for the individual patient with regard to disease recurrence, are yet missing.

Taken together, serial and quantitative analyses of chimerism in whole peripheral blood by STR-PCR allows the identification of patients at highest risk for relapse. However, not all patients can be highlighted and the time interval between onset of mixed chimerism and relapse can be very short. Therefore, it is essential to perform these analyses weekly during the first 100 days, or better, during the first 200 days, as the majority of relapses occur during this time. Performing chimerism analysis in subpopulations increases the sensitivity of the approach enormously. In this setting, chimerism analysis can be considered as a surrogate marker for MRD. Combining chimerism and MRD analysis does allow accurate documentation of engraftment and surveillance of post-transplant remission status, thus providing a rational basis for individual pre-emptive immunotherapy strategies to prevent recurrence of the underlying disease.

Conclusion

Chimerism analyses are now routinely performed for the surveillance of engraftment. These analyses are frequently used as indicators for the recurrence of the underlying malignant disease. In the most recent years, these investigations have become the basis for treatment intervention, for example, to avoid graft rejection, to maintain engraftment and to treat imminent relapse by pre-emptive immunotherapy (see Table 1).

Table 1 Procedure of routine diagnostics of chimerism and MRD in patients with hematological malignancies at the University Children's Hospital, Frankfurt

We are performing these analyses using a quantitative fluorescence-based STR-PCR with capillary electrophoresis for PCR product resolution. The investigations are performed weekly during engraftment in malignant and non-malignant diseases in whole peripheral blood. When graft failure or graft rejection is suspected clinically, analyses of subsets are performed.

If chimerism analysis on the basis of STR amplification is used for remission surveillance in malignant diseases, it is most important that the intervals are kept short. We recommend weekly intervals until day 200 post transplant. Thereafter, the intervals are lengthened to monthly investigations during the first 18 months post transplant. In patients with increasing mixed chimerism, additional immunotherapy is performed to augment the alloreactive potential of the graft. Besides chimerism investigations, MRD analyses are performed in BM and peripheral blood samples according to the underlying disease with leukemia clone-specific PCR systems.