Multidrug resistance: overcoming the good, the bad, and the ugly

In this issue, Drain et al. review the phenomenon of multidrug resistance (MDR) in lymphoid malignancies [1]. The discovery of MDR was first made in cultured solid tumor cells over three decades ago [2]. Clinically, MDR was first discovered in clinical specimens of solid tumors. As reviewed by Drain et al., there is growing evidence for the widespread occurrence of MDR, including in the chronic lymphoproliferative diseases. It is now over two decades since the cloning and sequencing of the MDR1 gene that was responsible for this initial observation [3]. MDR1 (also known as ABCB1) is a member of a superfamily of 48 genes including seven gene families. Since MDR1 was discovered, many additional cellular mechanisms have been found to cause resistance to cytotoxic chemotherapy. These may involve the activity of different types of drug transporters (including the adenosine triphosphate [ATP]-binding cassette [ABC] family, of which MDR1 is a member). Enzymes causing metabolic inactivation or degradation of drugs, DNA repair enzymes, and proteins that prevent apoptosis are among the multiple other mechanisms [4].

The sheer number of these mechanisms provokes the question: why does the cell possess these numerous factors? In many cases they are present at diagnosis, and are not generated as a result of previous therapy. Of note is that many of these proteins are evolutionarily conserved, from single cell organisms to mammals [5], suggesting, as noted by Drain and others, that they participate in vital cell functions. The crucial nature of these proteins is also implied by their redundancy, often with overlap in the function of several related proteins. Furthermore, the tissue distribution of some of these proteins led to the hypothesis that they offer protection against toxins and xenobiotics. One can conclude that, fundamentally, these proteins are necessary and good for the cell and the organism. Some of these molecules have even been found to transport antitumor drugs inside the cell. For instance, expression of the human organic cation transporter hOCT1 contributes to the success of imatinib therapy by facilitating entry of the drug into chronic myelogenous leukemia (CML) cells [6]. Thus their existence may sometimes even be good for the patient with a malignancy. However, when the expression (or overexpression) of these molecules prevents the successful activity of antitumor drugs, then they transform from beneficial to adversarial. MDR1 is the best studied of these proteins. MDR1 is widely expressed in tumor cells, because the cell of origin of these malignancies (such as kidney, liver, and intestine) has a high basal expression of MDR1 [7]. Such tumors are intrinsically resistant to many chemotherapeutic agents [7], and robust efforts must be made to thwart this mechanism of resistance. The otherwise good players in this scenario are then quite bad, and often lead to death of the patient.

Since the functions conferring drug resistance are, in many cases, essential for integrity of the cell and survival of the organism, agents that reverse drug resistance can be unduly toxic to normal tissues [4]. These challenging aspects of anticancer treatment are reviewed by Drain et al. with regard to lymphoid malignancies [1]. Furthermore, the situation can become even more complex when the chemotherapy drugs themselves cause the induction of drug resistance. There are many potential genetic mechanisms for this to occur [4]. For example, transporter efflux genes can become up-regulated, activating genes can become down-regulated, and metabolizing genes can become altered to reduce efficacy of treatment. Drug resistance mechanisms become truly ugly when exposure to cytotoxic drugs induces the very mechanisms which render the drugs ineffective. As the patient progresses through treatment, his cells may acquire drug resistance mechanisms that alter the tumor's drug-resistance phenotype with each successive line of therapy, as was demonstrated in a heavily treated patient with chronic lymphocytic leukemia (CLL) [8].

While Drain et al. review what has been done thus far to try to combat drug resistance, they do not really offer much hope for successfully overcoming this phenomenon. There are currently few practical means of testing for drug resistance prior to the use of specific agents to which the tumor may already be (or quickly become) resistant. The few existing commercially available tests for drug resistance profiles are not in widespread use and/or have not had prospective validation in clinical trials, particularly regarding their ability to predict treatment outcome.

Perhaps the best avenue for future progress on this front is the use of new biotechnological techniques [9]. Molecular profiling of tumors can be done at many levels. RNA extracted from tumor cells can be analyzed using gene expression profiles, which can accurately and simultaneously quantify the expression of large numbers of genes, such as genes related to chemotherapy resistance. Alternatively (or in fact, in parallel), the emerging techniques of proteomics can concurrently examine the production of large numbers of proteins, such as those related to drug resistance. Regarding DNA, large numbers of polymorphisms can be examined using DNA chips. The knowledge of such polymorphisms can be valuable in predicting drug resistance profiles [9]. At present, much knowledge is accumulating on the significance of these genetic and proteomic profiles in various malignancies. While at the present time there is insufficient clinical validation of these vast amounts of information, the knowledge base is constantly growing. There is hope that in the future, using bioinformatics for each tumor type and, eventually, each individual patient's tumor, it will be possible to select a treatment plan on the basis of predicted sensitivity [9]. This ‘personalized medicine’ approach is feasible using current methodologies, and offers hope for dealing with the good, the bad, and the ugly aspects of multidrug resistance.