Response: MRP4 gene polymorphisms and treatment response in adult ALL

We thank Brüggemann and colleagues for their interest in our paper¹  that recently reported an association between MRP4 gene polymorphisms (regulatory T-1393C and A934C leading to Lys304Asn substitution) and outcome in childhood acute lymphoblastic leukemia (ALL) patients treated on Dana-Farber Cancer Institute (DFCI) protocols. Given the preliminary character of this study, replication in an additional cohort of sufficient size was suggested. Brüggemann et al analyzed same polymorphisms in adult ALL patients who underwent treatment with German Multicenter ALL (GMALL) protocols. The cohort was enough powered to detect genotype-associated differences, but nevertheless failed to do so, further confirming little concordance that exist between pharmacogenetics findings in childhood and adult ALL.²  The authors suggest that the difference in methotrexate (MTX) dose between protocols account for observed discrepancy. While we agree with such a possibility, we would like to emphasize several other reasons possibly contributing to this finding. ALL in childhood and adulthood are distinct diseases as based on the observed difference in etiology, incidence rate, disease characteristics and survival.³  Disease-free survival in childhood ALL is as high as 80%, whereas it barely reaches 40% in adults. ALL blasts are more resistant than those from children to several drugs used in ALL treatment, suggesting that different mechanism can contribute to or modulate constitutive and acquired resistance. Adding to this is the versatile nature of MRP4, reflected in affinity for a variety of substrates that can, upon exposure, differentially regulate MRP4 expression.⁴ 

There are several differences in treatment modalities between childhood ALL DFCI protocol and GMALL adult ALL trial regarding drug type and dose, and schedule of administration.^5,6  Beside noted difference in MTX dose, high-dose MTX in GMALL trial is given at several instances during consolidation phase compared with earlier single administration during remission induction in DFCI protocol.

The association of MRP4 polymorphisms seen in childhood ALL seems to be more apparent in the standard risk group, as reflected in our article's data supplement.¹  Brüggemann et al report the analysis of standard risk patients only. However, the risk classes between 2 protocols are not comparable. Beside already mentioned age difference, the GMALL protocol does not encounter CNS disease at presentation among risk classification criteria, whereas the standard risk groups also includes the subset of T-cell leukemia.^5,6 

One possible explanation for association found in our study¹  is that higher frequency of toxicity in A934 carriers would lead to more frequent drug withdrawal or dose reduction, which might cause higher frequency of relapse. However, this reduction probably would not be sufficient to explain reduction in event-free survival, and other mechanisms contribute as well. For instance, MRP4 participates also in efflux of folate; down-regulation of MRPs might result in decreased folate efflux, thereby leading to expansion of the intracellular folate pool and antifolate resistance.⁷  This further illustrates the complexity of MRP effects and regulation.

In conclusion, we believe that the report of Brüggemann et al is important, as it highlights the applicability of pharmacogenetic findings in childhood ALL to adults, yet further studies analyzing more comparable childhood ALL populations are needed to establish the role of MRP4 polymorphisms.