Markers of anticancer drug resistance are predictive of treatment response and outcome in patients with acute myeloid leukemia. Immunologic detection of the drug efflux pumps, P-glycoprotein (Pgp) and multidrug resistance-associated protein 1 (MRP1), correlate with functional assays of drug resistance. These accumulation defects also appear operable in acute lymphoblastic leukemia. Many of the efflux pumps identified share significant structural homology with the large superfamily of ATP-binding cassette transporters. Other markers such as lung-resistance protein, bcl-2, and breast cancer-resistance protein, have been described in acute myeloid leukemia patients although their pathophysiology and clinical relevance are less clear and the methodology for their quantification are not well standardized. Preclinical studies have shown that small molecules capable of reversing efflux can restore drug sensitivity in resistant tumor models. Although initial clinical studies were limited by both potency and specificity of the reverser, later studies with more effective reversers have in many instances been limited by pharmacokinetic interactions exacerbating the clinical toxicities of chemotherapy. Although one large randomized study has demonstrated a proven survival advantage without increased toxicity using cyclosporine, the inconsistent results with other modulators raise doubt as to the utility and overall strategy of using drug efflux blockers in patients with established Pgp overexpression. Many of these patients have additional resistance mechanisms, and achieving meaningful clinical responses will likely require more complex clinical strategies. Preventing or delaying development of drug resistance in chemosensitive patients represents another therapeutic strategy to be tested.