Progress in understanding the molecular basis of drug resistance
in cancer promises more effective treatments.
Multidrug resistance, the principal mechanism by which many cancers develop
resistance to chemotherapy drugs, is a major factor in the failure of many
forms of chemotherapy. It affects patients with a variety of blood cancers
and solid tumors, including breast, ovarian, lung, and lower gastrointestinal
tract cancers. Tumors usually consist of mixed populations of malignant cells,
some of which are drug-sensitive while others are drug-resistant. Chemotherapy
kills drug-sensitive cells, but leaves behind a higher proportion of drug-resistant
cells. As the tumor begins to grow again, chemotherapy may fail because the
remaining tumor cells are now resistant.
Resistance to therapy has been correlated to the presence of at least two
molecular "pumps" in tumor-cell membranes that actively expel
chemotherapy drugs from the interior. This allows tumor cells to avoid the
toxic effects of the drug or molecular processes within the nucleus or the
cytoplasm. The two pumps commonly found to confer chemoresistance in cancer
are P-glycoprotein and the so-called multidrug resistance−associated
protein (MRP). Because of their function and importance, they are the targets
of several anticancer efforts.
Historical perspective That cells have mechanisms to transport a variety of molecules out of the
cytoplasm has been known for decades. For example, organic cation transporters
were some of the earliest such mechanisms identified, and the kidney's
secretory capability in this regard was first demonstrated in 19471.
The first specific correlations between cell membrane transporters or pumps
and a drug-resistant phenotype were made in Chinese hamster ovary cell lines
in the mid-1970s, when it was shown that a glycoprotein of 170 kD, called
P-glycoprotein, correlated with the degree of drug resistance in several cell
lines2. A variety of cells were found that were resistant to
colchicine, vinblastine, doxorubicin, vinca alkaloids in general, etoposide,
paclitaxel, and other small molecules used in cancer chemotherapy.
P-glycoprotein was purified in 19793, and strong evidence
in support of its role in pleiotropic drug resistance came in 1982, when it
was shown that DNA from resistant cell lines that was transferred to nonresistant
cells was able to confer resistance to the latter that correlated with the
expression of the protein4. The gene for P-glycoprotein, called
MDR-1, was cloned in 19855, and the protein's putative
function as an energy-dependent pump that expels small molecules from inside
cells was postulated on the basis of sequence homologies with bacterial hemolysin
transport protein and on other studies6.
Work on a lung cancer cell line that was resistant to doxorubicin and other
chemotherapeutic agents showed that this cell line did not overexpress P-glycoprotein,
but did express another protein, namely MRP, cloned in 19927.
MRP was also found to be a pump, specifically a member of the ATP-binding
cassette transmembrane transporter superfamily, and since that time both the
MRP and P-glycoprotein have been significant targets for anticancer compounds.
Current state It was not long before companies responded to the potential that these
two multidrug resistance−conferring proteins offered for anticancer
drug discovery. Table 1 lists a selection of companies
with programs in this area. Isis, for example, is using its antisense technology
to develop drugs to block the synthesis of MRP, and its oligonucleotide combinatorial
technology to develop drugs that interfere directly with MRP function. Another
significant effort is that of Vertex, which is developing two compounds, Incel
(biricodar dicitrate, VX-710) and VX-853, to block P-glycoprotein (MDR-1)
and MRP. Incel, an intravenous compound, and VX-853, an oral compound, are
intended to be administered in combination with cancer chemotherapy agents,
since the notion is that they act by preventing cancer cells from physically
removing other anticancer drugs from their interior. The Incel multidrug resistance
inhibitor is currently being tested in combination with chemotherapy in five
phase II clinical trials targeting breast cancer, ovarian cancer, soft tissue
sarcoma, small cell lung cancer, and prostate cancer. Another program is that
of Cell Therapeutics, which is developing CT-2584. This is a small molecule
drug for the treatment of patients with chemorefractory cancers, including
prostate cancer and sarcomas.
Table 1. Selected companies with multidrug resistance programs
It is important to realize that there are two sides to multidrug resistance.
On one hand, cancer cells need to lose their chemoprotective features mediated
by MRP and MDR-1. On the other, chemotherapy-sensitive non-cancerous cells,
such as bone marrow stem cells, need to be protected from the effects of chemotherapeutic
agents. Bone marrow destruction is the single most important dose-limiting
toxicity factor in the treatment of cancer patients. One reason is that recovery
of bone marrow requires the removal of the patient from the chemotherapy regime,
thus allowing cancer cells to grow again.
An example of an approach that targets bone marrow is that of Titan Pharmaceuticals.
The company is developing a gene-based product it calls MDRx1, which can confer
multidrug resistance to blood progenitor or stem cells, thus protecting them
against chemotoxicity. MDRx1 involves the insertion of the MDR1 gene ex vivo
into stem cells that have been removed from cancer patients. The modified
stem cells are then reinfused back into the patients. It is hoped that there
they can repopulate the blood system with chemoresistant blood cells. One
advantage is that this would potentially allow patients to be given higher
doses of anticancer agents than could be given normally.
Industry challenges As mentioned before, there is a very strong correlation between the expression
of the MDR-1 gene in many cell lines or in tumors derived from patients and
the multidrug resistance exhibited by these cells. However, multidrug resistance
also occurs in cells that do not show this correlation with MDR-18. This,
in fact, represents a key challenge to the development of therapies based
on the MDR-1 and MRP targets: There are probably other multidrug resistance−inducing
molecules in cancer cells that have yet to be characterized, including ones
that belong to the MDR-1 and MRP protein superfamilies.
A related challenge is that MRP-1 and MDR have normal chemoprotective functions
in cells throughout the body. They are part of the body's defense mechanisms
against toxic small molecules, and together with other membrane transporters
and pumps, are key participants in the normal function of the liver, kidney,
gastrointestal gland, adrenal gland, and blood-brain barrier. Therefore, any
approaches that target these molecules must do so in a way that is tissue
and/or cancer specific, so as not to affect the normal function of these molecules
in healthy cells.
Finally, given the multiplicity of correlations that exist between P-glycoprotein
and MRP expression with the expression of other proteins in cancer cells,
another challenge is to find good cell-based models that enable the rapid
analysis of these relationships in a meaningful manner. Yeast is an excellent
model system, and a network of genes has been identified that confers a drug
resistance phenotype similar to that of mammalian cells9. This
system is being used extensively to study expression and functional aspects
of resistance-conferring membrane pumps and agents that are effective against
them.
Clinical status Given the importance of MDR in tumor resistance to chemotherapy, it is
not surprising that several companies are conducting clinical trials aimed
at MDR-1 and/or MRP as targets. Phase II trials investigating the activity
of Vertex's Incel in combination with other agents for the treatment
of advanced refractory ovarian cancer and small cell lung cancer are currently
underway. The compound is being tested in a range of drug-resistant cancers,
including breast, ovarian, prostate, and small cell lung cancer. BioChem Pharma
is funding the STS trial and will develop and market the product in Canada.
MDR-1 and MRP overexpression have been associated with STS that is refractory
to chemotherapy.
Preliminary phase I/II data were encouraging, showing that Incel could
restore or enhance the activity of the anticancer agent doxorubicin in STS
patients who had documented aggressive disease, and who had intrinsic or acquired
resistance to doxorubicin. Doxorubicin is the standard chemotherapy for this
disease, affecting about 7,000 new patients annually in Canada and the US.
Approximately 70% of patients do not respond to initial chemotherapy, relapse
is frequent, and the five-year survival rate for patients who are refractory
to chemotherapy is a low 10−30%. According to Vertex's data, Incel
blocks both MDR-1 and MRP, and thereby restores the sensitivity of tumor cells
to treatment by apparently raising the concentration of anticancer agents
inside the target cells.
Another ongoing trial is that of Aronex (The Woodlands, TX), using the
anthracycline known as annamycin. Annamycin is under development for the treatment
of drug-resistant breast cancer, and like Incel, it has the potential to be
used in treating a broad range of cancers. One of the potential advantages
of this compound is its better safety profile when compared with other anthracyclines
used for similar purposes.
Other key issues include, for example, the fact that agents that can reduce
MDR in vitro, such as toremifene, do not work in patients, probably because
toremifene is bound to serum proteins in blood. However, a recent study shows
how short courses of high dose toremifene in combination with vinblastine
was generally well tolerated by patients and that it was possible to achieve
concentrations of toremifene in vivo that can reverse MDR in vitro, which
opens up possibilities to counteract MDR in the clinic using such combination
therapies10. Finally, a recent report shows how significant
MDR can occur in metastases but not in the primary tumor. This is a critical
issue impacting the development of agents that are effective on primary as
well as metastatic tumors11.
The future Although P-glycoprotein and MDR protein are capable of removing a wide
variety of anticancer compounds from inside tumor cells, recent work shows
that they cannot remove them all. For example, a novel derivative of olivacine,
labeled S16020-2, has shown significant antitumor activity both in vitro and
in vivo against cells that display resistance mediated by the MDR-1 phenotype.
This may be due to the rapid rate of uptake of this compound, which effectively
bypasses P-glycoprotein, leading to its higher intracellular accumulation
and effectiveness12. Other leads with similar activity profiles
against multidrug resistant cancers are also likely to resist, and high-throughput
screening efforts are being applied to find them (see Lead validation, p. 47−49).
An important development to watch for in the future of MDR-1 and MRP-mediated
multidrug resistance has to do with establishing links between these proteins
and other cancer mechanisms, such as mutant p53 expression. Recent work suggests
that in some cells, such a non-small cell lung cancer cells, there is a correlation
between MRP and mutant p53 expression, which can be used for prognosis13. There are many other such correlations, and their potential importance
becomes apparent in the context of a pharmacogenomic analysis of cancer multidrug
resistance, where patient variability to anticancer agents could be localized
conceivably to some of these gene families (see Pharmacogenomics, pp. 40−42).
In addition, recent structural studies are beginning to reveal the actual
components of MRP that are necessary for its function14. This
work will pave the way towards targeting agents specifically to the functional
regions of these molecules, thus inactivating them. Such work is already underway.
For example, MDR-1 is being extensively analyzed by the US National Cancer
Insitute's COMPARE program for the identification of agents in the NCI
database that would be predicted to be good substrates and/or inhibitors for
the molecule15.
The future of MDR is also going to see increasing benefit from novel genetic
approaches such as transcriptional decoys. For example, a recent report shows
how targeting the promoter of the human MDR1 gene with antisense was effective
in causing leukemia cells highly resistant to vinblastine to become susceptible
to the anticancer agent16. Complementing such approaches, synthetic
combinatorial chemistry or biology is developing anticancer compounds or peptides
that target the regulatory sequences of MDR1 and block its expression17.
Finally, novel molecules that can inhibit MDR are likely to come from a
variety of sources, and the future will continue to see the increasing screening
of compounds from a huge variety of settings for this purpose. For example,
a recent study describes how the antifungal antibiotic aureobasidin A can
be an effective inhibitor of the MDR1 P-glycoprotein18, and
this is likely to become a significant addition to the arsenal of MDR1 inhibitors.
Conclusions In the fight against cancer, a number of targets are being pursued with
equal zeal. The resistance-mediating transporters discussed here represent
a significant set of clinically relevant drug targets that have therapeutic
as well as diagnostic potential. Cancer defends itself actively by using these
mechanisms, and therefore their impairment is likely to have a significant
therapeutic benefit. The science in this area is progressing very rapidly,
and corporate involvement is likely to follow suit.
Reprinted from Nature Biotechnology 17, 94−95
(1999).
