© 1998 by Oxford University Press
Trials are now under way that experts say should give definitive answers about whether multidrug resistance can effectively be reversed in the clinic, making cancer cells more sensitive to chemotherapy.
Other researchers, meanwhile, are developing strategies to make MDR work for rather than against them, using gene transfer techniques to render normal cells more resistant to anticancer drugs.
The different tacks taken by these two lines of research represent the latest efforts to overcome one of the most tenacious problems in clinical oncology, the complexity of which has only recently come to be fully appreciated.
The challenges facing researchers in this field are threefold, wrote Branimir I. Sikic, M.D., director of the General Clinical Research Center at Stanford University Medical Center in California, in an October 1997 review in Seminars in Hematology. First, and most daunting, according to Sikic, is the array of alternative resistance mechanisms, controlled by different families of genes, that may thwart efforts to subdue the "classic" form of drug resistance caused by the MDR1 gene and its P-glycoprotein (P-gp) product.
Second is the lack of potency and specificity of many drugs used as P-gp inhibitors. The first generation of MDR-reversing agents consisted of drugs already in use for other indications, such as the calcium-channel blocker verapamil and the immunosuppressive agent cyclosporin A. Because these drugs lack specific binding ability to P-gp, their dose-limiting toxicities overwhelmed any potential therapeutic benefits in early studies.
Alter Pharmacokinetics
Third, P-gp modulators alter the pharmacokinetics of anticancer drugs. P-gp is expressed in many normal tissues, including parts of the liver, kidneys, intestines, and at the blood-brain barrier. As a result, drugs that affect the activity of P-gp can increase the toxicity of anticancer drugs in normal tissues, so that dosage of these drugs must be reduced when the two are given in combination.
The interpretation of early study results was also clouded by a lack of data on whether the tumors being treated actually expressed P-gp, experts said.
"There is not at this point unequivocal evidence that multidrug resistance is responsible for what we see in patients -- the failure of drugs in a host of human tumors," said Bruce A. Chabner, M.D., professor of medicine at Massachusetts General Hospital and Harvard Medical School, Boston. "There are a lot of problems with the drugs that have been used, with the way studies have been designed, and with the interpretations of studies. All of that has compounded the difficulty of understanding what's happening clinically."
As a result, "these agents have gotten bad press in the past," added William R. Friedenberg, M.D., of the Guthrie Clinic in Sayre, Pa. "They haven't worked because they've been too toxic at the levels that had to be given."
Better Aim
Current trials are testing more targeted, less toxic agents, and are also better designed to examine P-gp expression in the cancers being treated. Friedenberg is principal investigator of a phase III trial -- involving several U.S. cooperative groups, the National Cancer Institute of Canada, and the European Organization for Research and Treatment of Cancer -- of a drug called PSC 833 in patients with relapsed or refractory multiple myeloma who are being treated with the three-drug regimen known as VAD.
PSC 833 -- also known by the generic name valspodar -- is made by Novartis Pharmaceuticals, East Hanover, N.J., and is the furthest along in clinical testing of the second-generation MDR inhibitors. The drug is a derivative of cyclosporin D that does not cause the toxicities that prevented cyclosporin A from being a useful MDR inhibitor.
Early studies by Sydney E. Salmon, M.D., and colleagues at the Arizona Cancer Center, Tucson, showed that multiple myeloma is one of the diseases that most prominently displays multidrug resistance, Friedenberg said. Expression of MDR1 approaches 100% in myeloma patients treated with a variety of standard drugs, and those in the international myeloma trial have their bone marrow sampled to determine whether the gene is expressed. Results of the trial are expected in about 2 years.
First Agent
"Valspodar, in my view, is the first agent that's really been targeted at P-glycoprotein, that's been developed as an MDR modulator, and that has a chance of making a clinical difference," Sikic said in an interview. "It's more potent than cyclosporin A, it's not immunosuppressive, and lacks the kidney toxicity."
"The side effects of this drug seem to be minimal and readily reversible," Friedenberg added. "Another encouraging thing is that it seems to be absorbed well orally so it can be used as an outpatient treatment."
At Stanford, Sikic and Peter Greenberg, M.D., co-chair an Eastern Cooperative Oncology Group trial of acute myeloid leukemia patients. As in the myeloma trial, patients have relapsed or refractory disease, and are randomized to standard chemotherapy with or without PSC 833.
Other Trials
Other ongoing trials of the MDR modulator include a Cancer and Leukemia Group B phase III trial of PSC 833 as a first-line therapy in elderly patients with acute myeloid leukemia, who tend to have a higher incidence of P-gp expression than younger patients.
"That trial is very interesting because it takes patients at diagnosis, when they have had no prior therapy," Sikic said. "Any treatment tends to work better earlier in the course of disease, and these patients may be less likely to express other resistance mechanisms simply because they haven't been exposed to other chemotherapies." The Southwest Oncology Group has a similar trial in phase II.
In a large international trial sponsored by Novartis, women with stage III or IV ovarian cancer are given standard therapy of carboplatin and paclitaxel with or without PSC 833. Paclitaxel is a prime target of P-gp, which shows increased expression in ovarian cancer cells after treatment. In addition to PSC 833, several pharmaceutical companies have P-gp inhibitors in phase I trials or preclinical testing.
Promising Avenue
Another promising avenue for future studies, Sikic said, will be the prevention of drug resistance in tumors that do not yet express P-gp.
Ultimately, however, researchers acknowledged that they will have to grapple with multiple mechanisms of resistance. A flurry of discovery in the past decade has revealed an astonishing complexity on the molecular level (see part one of MDR series, News, Aug. 5, 1998), but the clinical significance of these findings is still largely unknown.
"There are now several candidate transporters that could be playing a significant role, including P-gp, MRP [multidrug resistance protein], a class of MRP-related proteins called MOAT, and a breast cancer resistance protein which has recently been isolated," Chabner said. "Then there is a whole range of biological factors -- apoptosis and cell-cycle regulatory factors. There are other very interesting findings about how upregulation of growth signals can antagonize chemotherapy and diminish the cell-killing effects of drugs. All these factors are capable of producing multidrug resistance, but we don't know the relative contribution of any of them in any one disease," he added.
If the broader MDR picture can be elucidated, Sikic said, strategies to reverse each type of resistance might be combined in the same way chemotherapy drugs are now combined to increase cytotoxicity.
Gene Therapy
Several groups of clinical researchers are looking at the MDR1 gene in a different light. Instead of abolishing its activity in cancer cells, they hope to exploit the gene's protective effects by activating it in blood and bone marrow cells.
Kenneth Cowan, M.D., Ph.D., head of the Medical Breast Cancer Section at the National Cancer Institute, and colleagues are developing a treatment that involves taking peripheral blood stem cells from metastatic breast cancer patients, transducing the cells with the MDR1 gene via a retroviral vector, incubating them with cytokines, and putting them back into the patients, who are then given several cycles of chemotherapy.
The idea is that the chemotherapy will trigger the transduced cells to overproduce P-gp, which will protect them from the cytotoxic drugs and confer a selective advantage to their growth. The result would be a hematopoietic system better able to withstand the onslaught of high-dose chemotherapy.
In their current clinical study, Cowan and colleagues transduce half the cells with the MDR1 gene and half with a neomycin resistance gene as a control vector. Patients then undergo four cycles of paclitaxel and four of Adriamycin.
All "Marked"
All six patients treated so far "marked" with MDR1, but only three maintained the marking throughout their chemotherapy. The signal from the MDR1 vector surpassed that of the neomycin vector -- a good sign that the selection process is working, Cowan said. But the falloff in MDR1 marking in half the patients, he said, reveals a major hurdle that remains: the selection for gene transfer of a pure batch of pluripotent stem cells, which make up a tiny fraction of circulating cells and are capable of giving rise to all types of blood cells.
The most specific selection researchers can now make is for cells expressing the CD34 antigen, resulting in a mixed bag that includes not only true stem cells but committed progenitor cells as well. Unlike stem cells, these latter cells have a limited half-life and are destined to die after several rounds of replication. For the gene transfer technique to work, the gene must get into enough true stem cells to keep the lineage proliferating.
Until this can be done reliably, the technique is hit-or-miss. Stem cells are notoriously slow growing, Cowan added, and the researchers are searching for the perfect cytokine cocktail to encourage their growth.
"One of the caveats is that the results in mice are much better so far than the results in humans, as far as hematopoietic reconstitution of gene-marked cells," he said.
But if the technique can be perfected, adding the protective effect of the MDR1 gene might be just the beginning of its utility. Other cancer-fighting genes might be able to tag along.
Cowan envisions three potential benefits for MDR1 gene transfer. "You could give the same amount of chemotherapy with much less toxicity. You could give higher doses of chemotherapy for a longer period of time, which might improve response rates."
"Third, and probably even more significant, you could use MDR1 as a selectable marker in order to transfer yet another gene into hematopoietic cells. For example, say you had a specific T-cell receptor targeted to a tumor antigen, and you wanted to recruit cells that would target that tumor antigen. You could engraft patients with this T-cell receptor, but it wouldn't necessarily stay around."
"Passenger" It
"So you could 'passenger' it with the MDR1 gene, select the cells in the patient, and make sure these cells stayed around because they would be protected from the anticancer drugs."
This technique has been used successfully in tissue culture and animal models, Cowan added. The two approaches to using the MDR1 gene -- modulating the gene in tumor cells and transfecting it into normal cells -- are incompatible with each other if a normal gene is used, Sikic said, because the modulator would abolish the survival advantage for the normal cells.
-- Tom Reynolds
Outwitting Drug Resistance Requires Researchers to Take Many Different Tacks
This is the second of two articles on multidrug resistance in cancer.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||