Selective Destruction of Cancer Cells

SELECTIVE DESTRUCTION OF CANCER CELLS ROGER G. HART* Introduction It has been proposed [?] that some or all cancers could be cured by a combination of(^4) treatment that reversibly suppresses mitosis ofnormal cells, but not of cancer cells, and (B) treatment that kills dividing cells. In such a strategy, B kills cells entirely on the basis oftheir phase, while A provides the necessary discrimination between normal and cancer cells by allowing only cancer cells to reach the phase vulnerable to B. It is argued that the A-B combination could achieve a highly selective destruction of cancer ceils without needing drugs offantastic specificity, that agents now available may, in fact, meet the requirements for A and B. This is a further attempt to define those requirements in relation to the proposed therapeutic strategy. It will be convenient to consider B first. Requirementsfor B Dustin [2] defines mitotic poisons as "those substances nearly specifically affecting dividing cells, while leaving unharmed the resting cells (i.e., postmitotic, differentiating cells or intermitotic cells not preparing for division)." This seems a good statement of the requirements for a satisfactory B, if A is understood to place normal cells in the status of resting cells, unharmed by B. Among the drugs called mitotic poisons, many ofwhich have been considered for cancer chemotherapy, we should eliminate as B candidates those agents that may actually cause latent damage to resting cells, though their effects may become manifest only when the damaged cells prepare to divide. Thus, alkylating agents [3] and other compounds that can attack DNA probably should not be considered . We should also eliminate as B candidates drugs that have pro- * Bio-Medical Reseach Division, Lawrence Radiation Laboratory, University of California, Livermore. This work was performed under the auspices ofthe U.S. Atomic Energy Commission. 592 Roger G. Hart · Destruction ofCancer Cells Perspectives in Biology and Medicine · Summer 1967 duced serious side effects unrelated to their antimitotic activities. The neurological effects ofcolchicine [4, pp. 340-341] and ofthe vinca alkaloids [4, pp. 1374-1375], for example, would probably preclude the use of these drugs in the high doses and frequencies required for B. On the other hand, reports ofside effects due to the destruction ofrapidly dividing normal cells in the marrow and gut are to be regarded as necessary credentials for a good B candidate. These effects, which usually have limited the tolerance ofcancer chemotherapeutic agents, must now be controlled by A Among the biochemical processes known to be necessary for cell division , the most attractive target for manipulation by B is the synthesis of DNA. This occurs during the well-defined S-phase [5], prior to visible prophase. Two kinds of interference are easily imagined: incorporation of bogus precursors into the DNA, to an extent that subsequent replication or mitosis is impaired, or deliberate blockage of enzymes needed to synthesize DNA from available metabolites. The latter is the more familiar approach in cancer treatment. Enzyme blocking, either by precursor analogues, notably fluorouracil [6], orby folic acid antagonists [7], notably methotrexate, has produced temporary remissions of many cancers and five-year cures ofchorio-carcinomas [8]. Before assigning such a function for B, however, we should ascertain whether the blocking agents block the enzymes of cells only in some particular phase, perhaps S-phase, or whether the blocking can occur at any phase but only becomes apparent when the enzymes are needed, at S-phase. In the latter case, which unfortunately seems more likely, enzyme blocking would be a poor choice for the function of B. It is conceivable that evidence will be found to exonerate one or more of the alternatives treated suspiciously above; at present, however, the most promising role for B seems to be the incorporation of bogus precursors into cellular DNA. Thymine, a major constituent of DNA, but not ofRNA or protein, provides an obvious chemical site for substitution. The substitute precursor could be radioactive thymidine or a chemical analogue. Although, as noted above, the thymine analogue 5-fluorouracil destroys cells mainly by blocking enzymes, other analogues, particularly 5-iododeoxyuridine [9], are known to be incorporated into the DNA of mammalian cells. The latter compound has been tested clinically [10] with results that...