Eric K. Rowinsky, MD
There is currently an explosion in the development
of novel anticancer therapeutics. From a developmental standpoint, we are
at a critical impasse in which the abundance of molecular biological information
about malignancy ascertained over the last two decades is being rationally
used to develop anticancer therapeutics. From a commercial standpoint,
pharmaceutical companies that never dared to venture toward anticancer
therapeutics are clearly starting to get their feet wet.
We are also undergoing dramatic changes with regard
to how we gauge our interest in potential antitumor compounds based on
preclinical data. Although the National Cancer Institute has shifted its
antitumor compound screen to an automated in vitro human tumor screening
system consisting of approximately 60 different human tumor cell lines
that were initially classified according to tumor type (eg, breast, melanoma,
non-small-cell lung, small-cell lung), the system is once again evolving
from a tumor-type-based screen into a molecular target-based screen. Rapid
and simultaneous assessments of a large number of molecular targets in
tumors are now feasible using high throughput techniques that facilitate
molecular target-based screening of new compounds. For example, the importance
of paclitaxel’s preclinical antitumor spectrum is being overshadowed by
the fact that the agent is profoundly and specifically active in tumors
with mutated or functionally deficient p53. Using automated tumor screening
systems (antitumor activity), high throughput technology that can be used
to simultaneously assess drug activity against more than 100 specific molecular
targets, and sophisticated computer software enabling classification and
storage of chemical structure information, extensive data can be accumulated
on hundreds of thousands of compounds, and databanks can be searched for
agents with any complex chemical nature, antitumor spectrum, and/or molecular-targeting
profile. This system will obviously facilitate the development of natural
products as it permits rapid chemical and functional classification of
structurally complex compounds.
The development of two classes of natural product
anticancer agents, the taxanes and the camptothecins, demonstrated that
even the most difficult developmental hurdles are surmountable. In addition,
although both classes of agents were initially identified by random screening
of natural substances (eg, plants, marine organisms), these efforts resulted
in the discovery of two subcellular targets for therapeutic development
that may not have been identified otherwise -- topoisomerase I and tubulin
polymerization. These results also affirm the potential value of random
screening of natural products. At this juncture, the taxanes (paclitaxel
and docetaxel) have demonstrated the broadest clinical antitumor spectra
of any class of agents, with prominent activity in Kaposi’s sarcoma and
in carcinomas of the ovary, breast, endometrium, lung, esophagus, bladder,
and head and neck. With regard to the development of novel taxanes, efforts
are being directed at maximizing their therapeutic indices. Such efforts
include the development of novel formulations that do not contain polyoxyethylated
castor oil (Cremophor EL), water-soluble analogues that may portend less
toxicity and eliminate the need for prolonged infusions and oral taxanes.
With regard to camptothecin and its derivatives,
two new agents have been approved for the treatment of advanced malignancies
-- irinothecan (CPT-11) and topotecan for refractory and recurrent colorectal
and ovarian carcinomas, respectively. Nevertheless, it is clear that the
surface has barely been scratched with regard to the development of topoisomerase
I targeting drugs. Currently, efforts are fervent in the development of
(1) more potent camptothecin analogs with regard to topoisomerase I inhibition,
(2) camptothecin and noncamptothecin analogs that may result in less noxious
toxic effects (eg, diarrhea) yet maintain topoisomerase I targeting capabilities,
and (3) novel formulation vehicles or drug carriers that may potentially
increase the feasibility of developing water-insoluble compounds such as
the prototypical analog camptothecin and also may increase the feasibility
of developing administration schedules that simulate optimal protracted
exposure schedules. Topoisomerase I targeting agents in clinical development
at this juncture include DX-8951f, TAS-103, and NB-506. Those at the verge
of entering the clinic include ED-749, pegylated camptothecin, and a variety
of 10,11-methylenedioxy-20(S)-O-camptothecin derivatives.
The notion that nucleoside analogues are not useful
in the treatment of solid malignancies was put to rest by the successful
development of gemcitabine, which is a deoxycytadine analogue with profound
activity against advanced pancreatic, breast, non-small-cell lung, ovarian,
and bladder carcinomas. The successful development of gemcitabine has stimulated
a resurgence of interest in the development of nucleoside analogues, as
well as interest in developing agents against subcellular targets that
have been deemed as irrelevant. One of the most interesting nucleoside
analogues in clinical development is BCH-4556. BCH-4556, which is structurally
similar to the antiviral 3T3, is the first unnatural nucleoside with an
L-configuration to be developed for patients with advanced cancer. In preclinical
studies, BCH-4556 demonstrated prominent antitumor activity against a wide
variety of human tumor xenografts, particularly against renal cell carcinoma.
On a similar note, efforts are in full swing to optimize therapeutics that
target other functionally important subcellular targets, such as ribonucleotide
reductase and thymidylate synthetase. The ribonucleotide reductase inhibitor
MDL 101,731, which is currently in phase I development, has demonstrated
prominent antitumor activity against human breast, non-small-cell lung,
and colon tumor xenografts. In addition, structural and functional characterization
of the thymidylate synthetase has resulted in the development of a large
number of direct thymidylate synthetase inhibitors that have vastly different
physicochemical characteristics (ie, formation of polyglutamates, Ki
for thymidylate synthetase, lipophilicity, folate transport carrier, and
intracellular retention). Currently, the following thymidylate synthetase
inhibitors are in clinical trials: AG331, AG337, LY231514, ZD9331, and
GW1843U89. The antifol LY231514 or multitarget antifol is of particular
interest since it inhibits the folate-dependent enzymes glycinamide ribonucleotide
formyltransferase (GARFT) and dihydrofolate reductase (DHFR) in addition
to thymidylate synthetase. Preliminary clinical activity has been observed
in patients with advanced non-small-cell lung, breast, and colorectal carcinomas.
In addition, a wide variety of unique DNA-damaging natural products are
in clinical development, including ET-743 derived from a marine tunicate,
HMAF (MGI-114) derived from mushrooms, and bacterium-derived rebeccamycin
analogue that are nonclassical inhibitors of topoisomerase II.
Some of the most interesting new therapies in development
are those that specifically target tumors with specific oncogene abnormalities.
At this time, the most intense activity in this regard is being directed
against malignant cells with mutations or deficiencies of the p53 tumor
suppressor oncogene. An example of p53-targeted therapeutics is the Onyx-015
attenuated adenoviral vector, which specifically incorporates into the
genome and divides in cells with p53 mutations or deficiencies. Other viral
vectors that restore p53 suppressor function are also being evaluated.
Therapeutics are also being directed at inhibiting abnormally active proliferative
oncogenes. Abnormalities in the ras oncogene are associated with
malignant proliferation. One of several chemical steps involved in the
posttranscriptional processing of the Ras protein is farnesylation, a process
that is being currently targeted. Several highly specific inhibitors of
farnesyl transferase are currently entering clinical development.
Another novel notion guiding the development of anticancer
therapeutics is that a point of limited returns has been reached regarding
the use of cytotoxic therapies in the treatment of neoplasia. It is entirely
reasonable that a wide range of "antiproliferative" agents may produce
substantial benefit in terms of increasing time to progression and survival
without the noxious side effects associated with classical cytotoxic agents.
Some of the most exciting classes of agents in this regard include inhibitors
of tumor angiogenesis and inhibitors of various growth factor receptors.
It is likely that these agents will be optimally used in patients with
low-volume disease as chronic therapy and possibly in combination with
cytotoxic agents. There are many components of malignant angiogenesis including
local cell destruction, endothelial cell retraction, endothelial cell mitosis,
endothelial cell migration, cell-matrix interactions, and three-dimensional
restructuring, all of which are currently being targeted. Several types
of antiangiogenesis agents, such as inhibitors of matrix metalloproteinases
that degrade the extracellular matrix, decrease the metastatic potential
of malignant disease and inhibit the growth of primary malignancies. Another
potent natural products antiangiogenesis agent, squalamine, which is derived
from the liver of the dogfish shark, is undergoing early clinical development.
Other antiproliferative agents are being developed that target the transmission
of proliferative signals transmitted by growth factor receptors, particularly
epidermal growth factor receptor antagonists. At least three specific inhibitors
of epidermal growth factor receptor tyrosine kinases (ZD1839, CP358,774,
and PD153035) are on the verge of clinical development. These oral agents
appear to exert optimal antitumor activity (ie, tumor growth inhibition
and cytoreduction) when given for prolonged durations and may be devoid
of toxicities against rapidly proliferative tissues. Another interesting
growth factor antagonist in early clinical development is ABT-627, which
inhibits the transmission of signals mediated by the endothelial- derived
autocrine growth factor endothelin-I that is produced by prostate cancers.
A somewhat controversial subcellular target for anticancer
therapeutics development is the DNA telomere or the nuclear enzyme telomerase.
Telomeres are repetitive sequences of DNA (TTAGGG)n at the end
of chromosomes that are lost with each cell division. Normally, telomerase,
a reverse transcriptase with its own RNA template, attaches 50 to 200 DNA
telomere bases to the ends of chromosomes. Aging is associated with the
loss of telomerase activity and the progressive shortening of telomeres,
whereas malignant cells have high levels of telomerase and longer telomeric
sequences, which makes telomerase and telomeric DNA attractive targets
for therapeutic development. A large number of natural product and synthetic
telomerase inhibitors, including both nucleoside and nonnucleoside analogues,
are being screened for teleomerase inhibitor effects, and some interesting
candidates have been identified. Similar to the other antiproliferative
agents that have been discussed, telomerase inhibitors will likely be most
successful when administered for protracted periods in patients with low-volume
disease.
Although the availability of novel compounds has
evolved from a state of famine to feast, one of the most important challenges
that we are facing is the dwindling support and maintenance of a clinical
trials infrastructure to facilitate optimal and rapid clinical evaluations.
The negligible and declining proportions of patients entering early clinical
trials will likely thwart the development of many exciting agents, particularly
many of the "nonproliferative" compounds that must be evaluated in large
randomized, controlled trials. A national research policy to ensure the
incorporation of early clinical evaluations into the clinical practice
mainstream is desperately needed to prevent a situation in which progress
in clinical oncology winds down to a halt, particularly when the potential
for major therapeutic advances is so great.
Selected References
Weinstein JN, Myers TG, O’Connor PM, et al. An information-intensive
approach to the molecular pharmacology of cancer. Science. 1997;275:343-349.
Folkman J. Fighting cancer by attacking its blood supply. Sci Am.
1996;275:150-154.
Oliff A, Gibbs, JB, McCormick F. New molecular targets for cancer therapy.
Sci Am. 1996;275:144-149.