In vitro studies of retinoids and arsenic in non-M3 acute myeloid leukemia

Abstract

Despite advances in the treatment of patients with acute myeloid leukemia (AML), the majority of the patients will die from their disease. The current intensive therapy for AML is also complicated my substantial morbidity and mortality and, as a result, many patients cannot be given the most effective therapy. Therefore, the need for more effective as well as less toxic treatment approaches for AML is urgent. Retinoids and arsenic has recently shown impressive results in the treatment of acute promyelocytic leukemia (APL) and both drugs exhibit less toxic effects than the commonly used chemotherapeutic drugs.

In these studies we have investigated the effects of retinoids and arsenic in cells from patients with non-APL (or non-M3) AML. Furthermore, we have investigated what mechanisms that could be of importance for the sensitivity to these drugs. The results showed that there is a pronounced heterogeneity in the sensitivity to the retinoids all-trans-retinoic acid (ATRA) and 9-cis-RA but cells from the majority of the patients were sensitive and some cases exhibited a pronounced sensitivity to retinoids. There was no cross-resistance between the retinoids and daunorubicin (DNR) and cells that were more resistant to DNR tended to be more retinoidsensitive. Also, CD34-positive AML cells were more sensitive than the CD34-negative.

In order to find markers for retinoid sensitivity we correlated the expression of retinoid receptors (RARs and RXR alpha) and retinoid bindings proteins (CRBPI and CRABPs) to the sensitivity to ATRA in the patient samples. No correlation between the sensitivity to the retinoids and the basal mRNA expression of any of these proteins was found. However, upregulation of RAR beta and CRABPII after ATRA exposure did significantly correlate to retinoid sensitivity.

For arsenic, a significant dose dependent sensitivity was found in all the patient samples even at concentrations below those that are seen in plasma during APL treatment. The sensitivity to arsenic was then compared to that of common AML drugs such as anthracyclines, etoposide, AMSA and Ara-C. There was a significant correlation in the sensitivity between the different anthracyclines as well as between the anthracyclines and etoposide and AMSA. However, all correlation coefficients were negative for the comparisons between arsenic and the other drugs and most so between arsenic and idarubicin.

We then investigated the sensitivity to arsenic in P-gp-expressing cell lines showing that P-gp expression does not affect the sensitivity to arsenic. We also investigated the sensitivity in cell lines with either increased expression of MRP 1, downregulated topoisomerase II or in cells that were FAS-resistant. None of these resistant cell lines were shown to exhibit any decrease in sensitivity to arsenic compared to its non-resistant cell line. The only cell line that showed a decreased arsenic sensitivity was the lung cancer cell line GLC4/ADR. In addition to the overexpression of the lung resistance protein (LRP), these cells exhibited higher concentrations of GSH than the wild type GLG4. The multidrug resistant reverser and pyridine analogue PAK-104P exerted a powerful sensitizing effect on the cells together with arsenic, both in resistant and nonresistant cells. This sensitizing effect was accompanied by a slight decrease in GSH but this effect was moderate compared to the strong effect on cell toxicity. Therefore, we suggest that the sensitizing effect of PAK-104P should be further studied as well as its potential as a sensitizer to arsenic in vivo.