Biomarker of sensitivity to PR-104 in leukemia

In this issue of Blood, the paper by Moradi Manesh et al¹  reports that PR-104 represents a potential novel treatment of relapsed/refractory T-lineage acute lymphoblastic leukemia (T-ALL) and that aldo-keto reductase 1C3 (AKR1C3) expression could be used as a biomarker to select patients who may respond to PR-104 in prospective clinical trials.

Acute lymphoblastic leukemia (ALL) represents the most common childhood malignancy.²  Because patients with T-ALL are more likely to experience induction therapy failure and early relapse compared with the B-cell precursor subtype (BCP-ALL), there is a high medical need to develop novel treatment strategies for these patients. PR-104, a pre-prodrug currently tested in adult early clinical trials for the treatment of cancer, is converted in vivo to the nitrogen mustard prodrug PR-104A and subsequently activated via hypoxia-dependent reductases or independently of hypoxia by AKR1C3.^3,4  Because AKR1C3 is frequently overexpressed in human cancers, including leukemia, PR-104 is considered a promising targeted drug for cancers with high AKR1C3 expression. Testing of PR-104 in a panel of primary pediatric cancer xenografts in immune-deficient mice has previously revealed high in vivo efficacy in childhood leukemia, in particular against T-ALL compared with BCP-ALL, when tested at doses providing plasma pharmacokinetics achievable in humans.^5,6  However, the underlying mechanism for this differential sensitivity of ALL subtypes remained elusive.

Against this background, the aim of this study was to test whether AKR1C3 is a predictive biomarker of in vivo PR-104 sensitivity. To investigate whether PR-104 exhibits lineage-specific in vivo efficacy against T-ALL, Moradi Manesh et al extended the evaluation of PR-104 to a panel of patient-derived pediatric ALL xenografts.¹  PR-104 exerted significantly greater antileukemic efficacy against T-ALL xenografts than BCP-ALL.¹  Comparison of PR-104 with an induction-type regimen consisting of vincristine, dexamethasone, and l-asparaginase revealed that single-agent PR-104 was more efficacious against T-ALL xenografts than this standard-of-care regimen.¹  To identify a biomarker for PR-104 sensitivity the authors performed gene expression profiling comparing PR-104A–sensitive and PR-104A–resistant xenografts.¹  This analysis revealed AKR1C3 as one of the top 2 differentially expressed genes, and AKR1C3 expression correlated with PR-104/PR-104A sensitivity in vivo and in vitro.¹  A causal relationship between AKR1C3 expression and sensitivity to PR-104 was then demonstrated by overexpression of AKR1C3 in a resistant BCP-ALL xenograft, which led to increased sensitivity to PR-104 in vivo.¹  To validate their results, the authors went on to test ex vivo sensitivity to PR-104A against patient-derived leukemic blasts.¹  Primary T-ALL cells proved to be more sensitive than BCP-ALL to PR-104A in vitro, and this sensitivity correlated with AKR1C3 expression.¹  Together, these findings indicate that PR-104 represents a promising novel therapeutic for refractory and relapsed T-ALL and that AKR1C3 expression could serve as a biomarker to select patients who will most likely benefit from treatment with PR-104 in prospective clinical trials.

This study has important clinical implications. In line with the present study showing the safety of PR-104 in preclinical mouse xenograft models with no toxicity-related events,¹  PR-104 proved to be well tolerated in early-phase clinical trials in adult solid tumors and hematologic malignancies.^7,8  PR-104 may be of particular interest for refractory and relapsed cases of T-ALL, as PR-104 has been shown to specifically target hypoxic regions of leukemia infiltration (eg, in the bone marrow),⁶  which are considered as important contributors to chemoresistance and relapse in ALL.

Nevertheless, there are also a number of open questions. Although in the study by Moradi Manesh et al, AKR1C3 expression levels proved to be a strong predictor of response to PR-104 independently of hypoxia in ALL,¹  AKR1C3 expression levels did not significantly correlate with overall tumor responsiveness to PR-104 across a panel of solid and hematologic pediatric cancers in another study.⁵  Furthermore, ex vivo testing of primary T-ALL blasts in the present study revealed 1 case with low AKR1C3 expression but high sensitivity to PR-104.¹  Although the molecular basis of this outlier is currently not well understood, this observation suggests that AKR1C3 may not universally serve as a suitable biomarker to select patients who may benefit from PR-104. Moreover, the underlying mechanisms responsible for the reported increased expression levels of AKR1C3 in T-ALL compared with BCP-ALL remain subject to further investigation. Also clinically relevant is the question as to whether or not PR-104 is active against biologically distinct subgroups of T-ALL at high risk of treatment resistance and relapse that have recently been identified, such as early T-cell precursor ALL.⁹  Another issue relates to PR-104–based drug combinations to exploit synergistic drug interactions or to overcome resistance to PR-104. Because DNA interstrand cross-link repair proficiency, in addition to hypoxia and reductase activity, has been reported to determine sensitivity to the alkylating agent PR-104,¹⁰  evaluation of PR-104 together with DNA repair inhibitors might be of special interest.

In sum, the present study supports the further investigation of PR-104 for refractory or relapsed T-ALL patients that are selected by high AKR1C3 expression in their leukemic blasts as a biomarker.