Tumor Cell Heterogeneity and Drug Resistance

While tumors are clonal in origin, the increasing DNA instability that accompanies the onset of neoplasia leads to increased variation of daughter cells, referred to as clonal evolution to tumor cell heterogeneity. This is associated with selection for progeny with greater survival capacity, evident as a higher proliferative capacity, resistance to apoptosis, greater metastatic or invasive potential, reduced dependence on normal cellular growth factors, and angiogenesis.⁹² Heterogeneity among tumor cells increases the number and diversity of potential target sites for chemotherapy and the need for combining therapeutic agents .

Initially, resistance was thought to be limited to the selecting agent (mono-drug resistance). The recognition of multi-drug resistance required a reexamination of this rationale for combination chemotherapy.⁹³ P-glycoprotein multi-drug resistance relates almost exclusively to natural products, but glutathione transferase, DNA repair capacity, and topoisomerase II alterations also may be associated with multi-drug resistance.⁹⁴

Recent studies of multicellular drug resistance indicate an altered set point for apoptosis.⁹⁵ Differences between in vitro and in vivo drug resistance are modifying the approach to combination chemotherapy.⁹⁶ Although prolonged drug exposure results in stable, resistant cell lines, acute exposure may induce short-term, reversible resistance. How this relates mechanistically to long-term, presumably genetic resistance is under study.

Cytokinetics

The discovery that solid tumors contained a large number of potentially clonogenic cells in G₁ or G₀, presumably because of tumor hypoxia and a low growth fraction, provided a basis for combination chemotherapy.^{16, 17, 21} Thus, cell cycle-specific agents were employed to kill mitotically active cells, and non-cell cycle-specific agents, for example alkylating agents, were added to damage the noncycling tumor cells. The use of repeated cycles allows the recovery of normal tissues, so that dose need not be compromised and G₀/G₁ tumor cells can be recruited into the proliferating fraction by increased availability of nutrients, oxygen, and vascular access.

Synchronization

Synchronization of tumor or normal cells in vitro and in vivo with drugs that inhibit DNA synthesis or arrest cells in mitosis can be achieved. Experimentally, the most impressive synchronization has been achieved with hormonal agents that affect the tumor and not essential normal cells.

Some degree of tumor-cell synchrony follows this hormonal manipulation in experimental studies. The heterogeneity of human tumors regarding the time course of recruitment and synchronization has limited this approach, and it remains investigational.^{97, 98}

Modulation

Agents that are nontoxic may still improve the therapeutic index of an established chemotherapeutic agent, either by reducing normal tissue toxicity, such as leucovorin (LCV) for methotrexate, or preferentially enhancing antitumor efficacy, such as 5-FU and LCV in metastatic and adjuvant colon cancer studies.^99–110 Biochemically, the product of FU (FdUMP) binds to the substrate site of thymidylate synthase, thus inhibiting DNA synthesis and, therefore, cellular replication. The stability and duration of this inhibition directly relate to a third agent, 5,10-methylenetetrahydrofolate, which is a metabolic product of LCV that also binds to thymidylate synthase, producing the so-called ternary complex (FdUMP-TS-5,10-methylenetetrahydrofolate). In preclinical systems, both in vitro and in vivo, LCV can favorably modulate the therapeutic index of FU. A number of clinical trials comparing FU to FU with LCV indicate an advantage for 5-FU/LCV in patients with metastatic colorectal cancer at a cost of only moderately increased mucositis and diarrhea. The combination of FU with LCV improved survival rates in two separate studies in metastatic and adjuvant colon cancers.^{109, 110} In patients with head and neck cancer and with metastatic breast cancer, FU modulated by LCV appears promising and is now under study.¹¹¹

Experimental

The properties and comparisons of combination chemotherapy and holotherapy are presented in Table 44-4. Incorporation of many more classes of agents will force more efficient experimental designs. Such designs may include, for example, a rolling Phase II/Phase III study design. Increasingly, quantitative molecular markers and “real time” pharmacology will be integrated operationally into clinical studies. The effectiveness of such related approaches should markedly improve the efficiency and effectiveness of clinical trials. Such agents are under extensive study, not only for their antitumor properties per se but also particularly for their interaction with each other and with other established antitumor agents in an effort to extend the effectiveness of combination chemotherapy. Oncogenes and tumor-suppressor genes may operate by modifying or exploiting abnormalities in the cell cycle, such as cyclin-D, which is commonly overexpressed in some of the epithelial solid tumors, and by effects on growth factors and angiogenesis, which are required in varying degree for all tumors. Proteosomes, which facilitate the ubiquination of peptides and thus their removal from the cell, telomerase, and the cell cycle provide additional attractive targets for a pharmacodynamic modulatory approach. Opportunities for using such agents in combination represent major areas for preclinical and particularly clinical research. In summary, targeted therapies provide an extension in our ability to apply and benefit from the use of cancer chemotherapeutic agents in combination.

Table 44-4

Combination Chemotherapy Versus Holotherapy.

Holotherapy—An Integrated Approach to Cancer Chemotherapy

Over a half-century has elapsed since cancer chemotherapy began. The chemotherapy agents now in use originated as biologically targeted therapy. Hitchings and Elion developed specific inhibitors of purine synthesis, such as 6-MP; Heidelberger targeted RNA synthesis with 5-FU; and Farber targeted the reduced folate pathway with aminopterin. These innovations not only provided the groundwork for cancer treatment but also became tools for discovery of the basic workings of transformed cancer cells. Natural products like the Vinca alkaloids, anthracylines, and taxanes were selected specifically for activity against cancer proliferation. These agents were combined with other classes of agents, such as hormones, alkylating agents, and irradiation, that are active against proliferating cells. The optimal use of these agents requires the integration into increasingly complex combinations such that a greater therapeutic index results. This is true for the current classical chemotherapeutic agents, which largely inhibit proliferation. Molecular biology particularly has added a number of additional classes. Thus, in addition to the classical antiproliferation compounds, there are hormones, immunotoxins, and inhibitors of invasion and metastasis. Receptors of unique structure and quantity, capable of interactions with the more heterogeneous specific molecular site targets, have recently been described.^{88, 91, 116} It became readily apparent that the vast majority of cancers would be treated successfully only with combinations of agents, which were chosen for the highest possible individual activity against a specific type of cancer. Ideally, such drugs had different dose-limiting toxicities. Empiricism was an essential component in the development of contemporary cancer therapy but rational drug discovery, analog development, preclinical modeling, precise pathologic diagnosis, careful staging of disease, and clinical trial design were the basis for the measure of success known today.

The breakthroughs in molecular biology have presented the cancer therapist with enormous opportunities and challenges. This understanding will allow a molecular diagnosis not only of where and how cancer originates but also of what processes are essential to its survival. The specific processes that initiate and propagate cancer have become the targets of unprecedented rational drug development (see Table 44-1). Pharmaceutical technology provides not only small molecules but also monoclonal antibodies, immunoconjugates, ribozymes, antisense RNA, and recombinant viruses.

The therapy of cancer now has the potential to combine agents of even more mechanisms of action than before to confront the heterogeneity of cancer with a wider array of therapeutics than ever before. Some of these combinations are now the standard of care, as improvements in response and survival demonstrate (see Table 44-4). The challenges to be overcome include the clinical development of cytostatic agents without the expectation of significant acute toxicity; combining classes of targeted agents, both molecular and biological, with regard to dose and schedule; and selection of the appropriate types of cancer and individual patients for a specific therapy.

With molecular biology playing an increasing role, the clinical and laboratory sciences that address the therapy of cancer, will continue to accelerate progress toward cancer control.

Table 44-5

Therapeutic Interaction between Agents Representing Different Classes.