Principles Of Targeted Drug Therapy

One strategy to circumvent the problems associated with conventional chemotherapy (^22.2) is to develop drugs against more specific targets in the cancer. This is not a fundamentally novel idea. Many drugs in current use interact with highly specific targets such as microtubular proteins or topoisomerase enzymes (^■22.2). Their targets are not specific to cancers, though. Those drugs called 'biological agents' in the previous section come closer to the ideal, since they act on specific receptor proteins which may occur preferentially in certain tissues, but, more importantly, are essential for the growth of specific cancers. These, then, are the forerunners of a novel drug generation.

All-trans retinoic acid, e.g., binds to receptors that are more or less ubiquitous in the body (^8.5). Indeed, synthetic analogues of this tissue hormone are also used for the treatment of benign skin diseases like acne, because retinoic acid promotes cell differerentiation in the skin, as in many other epithelia. Retinoids therefore have been tried as anticancer drugs in almost every type of cancer, usually with detectable, but limited effects on tumor growth. In contrast, retinoic acid is highly active in most cases of acute promyeolocytic leukemia (^10.5), out of all acute leukemias. What makes the difference towards all other cancers is that in this particular type of leukemia the causative genetic change involves the retinoic acid receptor a, whereas in other cancers changes in the response to retinoids may well occur, but are non-essential for their growth and survival.

This case, then, comes close to the ideal of target-oriented cancer therapy. Elucidation of crucial events that drive cancer growth should provide targets for therapy. Targets for rational therapy ought to be at least specific to the tumor, but better essential for its growth and survival. Elucidating these crucial events and identifying suitable target molecules are however no simple tasks.

In many hematological cancers, the presence of a characteristic chromosomal translocation points to an essential genetic event (^10.2). Yet, even in leukemias and lymphomas developing a therapy from that knowledge can be difficult. Acute promyeolocytic leukemia (APL) is exceptional in so far as the fusion protein formed by the causative chromosomal translocation (^10.5) contains a receptor protein whose ligands are well characterized. Developing a therapy for chronic myelocytic leukemia (CML) by targeting the causative BCR-ABL fusion protein was still quite straightforward, since it contains an essential protein kinase activity (^10.4).

Unfortunately, not all fusion proteins display functions that lend themselves to inhibition or activation by small molecule drugs. In the jargon of pharmaceutical research, they are not easily 'drugable'. Furthermore, APL and CML are untypical in constituting essentially homogeneous diseases, whereas other hematological cancers may be caused by a variety of different translocations and gene fusions (cf. Figure 10.2).

As carcinomas are characterized by multistep development with accumulation of a larger number of various genetic and epigenetic alterations (cf. 13.3), it is generally even less clear which targets are optimal. Optimists assume that many of these alterations are essential for the growth and survival of the cancer and conclude that the multitude of changes in carcinomas offer a wide choice of targets for therapy. Pessimists point out that the more alterations have already occurred, the higher the chance that some of them may be passenger alterations. Worse, a cancer with many genomic alterations is likely to develop further ones allowing escape from therapy. Likewise, optimists suggest that cancer pathways activated in specific cancers (cf. 6) present excellent targets, as they are crucial for driving tumor growth. Pessimists point out that these same pathways are also important for normal cells which might mean a narrow 'therapeutic window'.

In practice, potential targets for cancer-specific therapy are defined based on a variety of considerations (Table 22.3).

Table 22.3. Molecular targets for cancer therapies

Type of target


Ectopic proteins Overexpressed proteins viral proteins, cancer-testis antigens oncogene products, particularly receptor tyrosine kinases mutated products of oncogenes MAP kinases, CDKs

Altered proteins

Cancer pathway components

Ectopic targets: An ideal drug target in a cancer would never occur in a normal tissue. Since cancer cells are derived from normal cells, one would think that such targets might be rare, but some do exist. (1) In cancers induced by viruses or cancers harboring viruses, viral proteins can be targeted. The E6 and E7 proteins of HPV (^Box 5.1) are involved in carcinogenesis in several tissues. Other viruses like EBV and HBV, while not necessarily driving tumor growth, are at least present in many Burkitt lymphomas (^10.3) and hepatocellular carcinomas (^16.3), respectively. (2) Fusion proteins in leukemias and lymphomas are composed of proteins which are also present in normal tissues, but always separately. Their fusion confers novel properties which can be exploited to target them selectively. (3) Many cancers express proteins (^12.5) which are otherwise only found in fetal tissues ('oncofetal proteins') or in a very small range of other tissues, e.g. in the testes ('cancer testis antigens'). These are often not essential for the growth of the cancer, but they can be used for the targeting of toxins or for immunotherapy.

Overexpressed proteins: A second class of targets is provided by proteins over-expressed in cancer cells. Some oncofetal proteins and cancer testis antigens actually belong to this class, because they are expressed at very low levels in normal tissue. The most important group, however, of such proteins are the products of oncogenes that have become activated by overexpression, e.g. as a consequence of gene amplification.

Several strategies have been developed, e.g., to exploit the overexpression of the EGFR or ERBB2 receptor tyrosine kinases in many advanced carcinomas for therapy. An evident disadvantage of using such proteins as drug targets, though, is the very fact that they are overexpressed. This point is illustrated by the amplification of the androgen receptor gene in prostate carcinomas which have become unresponsive to anti-androgenic treatment (^19.2). Accordingly, targeting an amplified protein kinase by an inhibitory drug may invite further amplification as a mechanism of resistance. Nevertheless, as the case of trastuzumab shows, overexpressed cell-surface receptor tyrosine kinases can be used for targeting by antibodies (^18.4). Antibodies can also be directed at proteins that are not as essential for the growth and survival of the tumor cell as ERBB2 is for many breast cancers. Modern high-throughput proteomics and expression profiling approaches are excellently suited for the identification of proteins overexpressed in cancer cells. There is therefore no shortage of candidates for this approach.

Proteins with altered structures: The third class of targets are proteins whose structure is altered in cancer cells. Fusion proteins could also be assigned to this class. They are excellent targets, because their structure is different in tumors compared to normal cells and they are essential for cancer growth. The same is true for oncogenic proteins activated by point mutations, such as KRAS in colon cancer (—>13.3) or P-Catenin in the same cancer and more often in hepatocellular carcinoma (^16.2).

Even tumor suppressor proteins inactivated by point mutations deserve consideration. They could provide targets for immunotherapy, but drug therapy is not inconceivable. A favorite candidate in this respect is TP53, since it is more often inactivated by missense mutations than by deletions or promoter hypermethylation.

Most missense mutations in TP53 appear to interfere with the conformational activation of the protein, which strongly accumulates in cancer cells because the mutated protein is more slowly degraded (^5.3). A drug pushing the protein into an active state would therefore activate a comparatively huge amount of TP53 protein and likely elicit apoptosis.

In addition to these clear-cut cases of mutated oncoproteins, there is some evidence that cancer cells in general may harbor a larger proportion of misfolded and altered proteins than normal cells. It is not quite clear what causes this defect, but it may make cancer cells more susceptible to inhibition of chaperones and of proteasomal degradation. Inhibition of either sort of target may overload the cell with misfolded proteins, like during a heat-shock. Indeed, inhibitors of heat-shock proteins acting as chaperones have emerged as surprisingly good inhibitors of cancer growth, with few side effects. Likewise, inhibitors of proteasome function, e.g. of threonine proteases, have turned out to be surprisingly specific for cancer cells.

Cancer pathway signaling: The fourth category of targets comprises molecules that regulate 'cancer pathways' (^6). The proliferation and survival of cancer cells depend on a relatively restricted number of signal transduction pathways. In different cancers, one or the other of these are overactive or inactive. Inhibition of overactive pathways or restoration of inactivated pathways is a major goal of many current drug development.

However, the designation 'cancer pathways' is in so far imprecise, as the same pathways also control the proliferation, differentiation, survival and function of normal tissues. So, differences between normal and cancer cells are expected to be quantitative rather than qualitative. Hope that these differences may still be sufficient to allow improved cancer therapy is based on observations and ideas summarized by the 'addiction hypothesis'.

Compared to normal cells, signaling pathways in cancers are thought to be 'rewired'. For instance, overactivity of pro-proliferative signals relayed through the canonical (ERK) MAPK pathway would in normal cells be counteracted by increased apoptosis (^6.4). In cancer cells, this increase in apoptosis is impeded by over-activity of other pathways such as the PI3K or the NFkB pathway (^6.4) or by overexpression of anti-apoptotic proteins (^7.3). Therefore, the survival of cancer cells is much more dependent on these anti-apoptotic activities than that of normal cells, which display more moderate and transient activities of MAPK pathways (Figure 22.7). In other words, the cancer cells have become 'addicted' to the activity of the anti-apoptotic cancer pathway.

Of note, the addiction hypothesis predicts that inhibiting the MAPK pathway which actually drives proliferation could be less efficient than inhibiting the PI3K pathway that influences proliferation only indirectly but allows cell survival. This hypothesis may also provide an alternative explanation why cancer cells react more sensitively to inhibitors of heat-shock proteins. They may have become dependent on mutant proteins that could not be assembled in the absence of these molecular chaperones.

Targets defined by such considerations can be exploited by different kinds of therapy. Development of pharmacological therapy with small molecules is the most obvious approach. It is best suited, but not restricted to proteins with enzymatic activities. This strategy has the important practical advantage that it can build on established procedures. Nowadays, pharmaceutical companies possess compound libraries comprising ten thousands of synthetic and natural chemicals that can be screened by high-throughput methods for activating or inhibitory activity against a specific target enzyme. Even protein-protein or protein-DNA interactions can be influenced. A molecule with activity is considered a 'lead' compound. Lead compounds can be chemically modified by a host of well established techniques and procedures to achieve increased specifity and better general pharmacological properties. Determination of the structure of the target protein by modern biophysical methods and drug design using sophisticated computer methods have further facilitated this strategy.

Figure 22.7 The 'addiction hypothesis' illustrated by the MAPK and PI3K pathways The width of the arrows indicates the activity of the pathways. The lightning symbolizes an upstream mutation activating primarily the MAPK pathway, such as a receptor tyrosine kinase or a RAS mutation. See text for a detailed exposition of the hypothesis.

Figure 22.7 The 'addiction hypothesis' illustrated by the MAPK and PI3K pathways The width of the arrows indicates the activity of the pathways. The lightning symbolizes an upstream mutation activating primarily the MAPK pathway, such as a receptor tyrosine kinase or a RAS mutation. See text for a detailed exposition of the hypothesis.

However, small molecule drugs are not the only option anymore. In fact, some of the most successful 'novel' cancer drugs are antibodies against growth factor receptors. Their development, likewise, has benefited from the availability of a wide range of sophisticated molecular biology methods. For instance, therapeutic antibodies can now be detected and optimized not only in animals, but also in bacteria and phages. If an antibody is initially developed in an animal, it can be 'humanized', i.e. the constant chains can be replaced by a human immunoglobulin sequences using standard methods of genetic engineering. This adaptation impedes the development of an immune response towards the therapeutic antibody in the patient. Without humanization, antibodies become inactive upon repeated administration or may even cause serious adverse reactions such as an allergic shock.

Application of antibodies can be regarded as a type of immune therapy, although antibodies can also be used in a similar fashion as small molecule drugs. In contrast, a suitable target molecule can be also used as the basis for a true cancer vaccine. Some modern approaches at cancer immunotherapy do indeed use defined targets 022.5).

Defined targets in cancer can also be exploited for gene therapy. In theory, gene therapy is a more straightforward approach than drug or immune therapy. However, the development of new drugs and vaccines can be pursued on a strong fundament of established procedures and long-term experience, whereas in gene therapy almost everything has to be developed from scratch (^22.6). Of all the strategies contemplated and tried in gene therapy, the use of antisense oligonucleotides or siRNA against overexpressed or altered proteins in cancer most closely resembles the approach in the development of drugs. It therefore runs a good chance of becoming established in the clinic first among gene therapy approaches. There are, however, strategies that are unique to gene therapy and may in the long run prove superior. For instance, gene therapy can be used to re-introduce a tumor suppressor that has been inactivated in cancer cells or to exploit the presence of an oncogenic change to allow the replication of a cytolytic virus (^22.6).

Last not least, combinations of novel therapies, drug, immune, and gene therapy, are pursued, and either type of 'novel' therapy can be used in conjunction with established chemotherapies or radiotherapies directed at less cancer-specific targets.

10 Ways To Fight Off Cancer

10 Ways To Fight Off Cancer

Learning About 10 Ways Fight Off Cancer Can Have Amazing Benefits For Your Life The Best Tips On How To Keep This Killer At Bay Discovering that you or a loved one has cancer can be utterly terrifying. All the same, once you comprehend the causes of cancer and learn how to reverse those causes, you or your loved one may have more than a fighting chance of beating out cancer.

Get My Free Ebook

Post a comment