THE YALE JOURNAL OF BIOLOGY AND MEDICINE
48, 97-103 (1975)
Methotrexate and the Need for Continued Research FRED ROSENFELT Yale University School of Medicine, New Haven, Connecticut 06510 Received January 7, 1975
With the current national commitment to cancer research, new and challenging areas of investigation are being planned. Hopefully, this inevitably expanding search into the nature and treatment of malignant tumors will include a continued examination of current chemotherapeutic agents. These standard tumoricidal drugs have proved to be effective in the therapy of numerous cancers despite an incomplete understanding of their mode of action in many cases (1-4). A further unraveling of the pharmacologic basis of these drugs should undoubtedly lead to a greater therapeutic effect, perhaps producing longer remissions and increased cures. A recent example of how continued basic research on available antineoplastic agents may reveal findings pertinent to clinical chemotherapy concerns methotrexate, a drug frequently employed in the treatment of cancer (1-5, 66). Traditionally, the mode of action of this drug has assumed to be as an analog of the vitamin folic acid. In humans, folic acid is biologically inactive, requiring reduction by an enzyme termed "dihydrofolate reductase" to dehydrofolate, and then to tetrahydrofolate. This latter compound accepts 1-carbon fragments from various sources to produce the folate coenzymes. These compounds act as 1-carbon donors in numerous biochemical reactions, including the synthesis of the B-carbon of serines, the formation of the C-2 and C-8 units of the purine skeleton, and the methylation of thymidylic acid (5-7, 10, 12, 13, 15, 17, 18, 66). Methotrexate acts, then, by irreversibly binding to and inhibiting the enzyme dihydrofolate reductase thereby interfering with the maintenance of intracellular pools of reduced folates, particularly N5' '0-methylenetetrahydrofolate (1, 5, 11-13, 66). This latter compound is necessary for the conversion of deoxyuridylate to thymidylate which is an essential component of DNA. As is evident, methotrexate, as well as other antifolate agents, essentially act by inhibiting DNA synthesis (1-9, 11-12, 14-20, 27-40, 54, 66). For many years it was accepted that the biological effects of methotrexate were due entirely to its inhibition of the enzyme dihydrofolate reductase. Since free or unbound intracellular methotrexate accumulates only after all the dihydrofolate reductase binding sites are saturated, little biological activity or significance has been ascribed to this free intracellular drug component. However, evidence has been accumulating that methotrexate may have a second site of action of equal or greater importance than the inhibition of dihydrofolate reductase and furthermore, that it is the unbound intracellular drug that is involved with this alternative site. Roberts et al. (35-37), several years ago, demonstrated a lack of correlation between the inhibition of dihydrofolate reductase activity in human leukemia cells and the response of leukemia patients to methotrexate therapy. On the basis of their studies, although without substantiating data, this group theorized that free intracellular methotrexate as well as that bound to dihydrofolate reductase may have a significant role in the cytocidal effect produced by this agent. Further evidence suggestive of an additional site of methotrexate activity was next provided by Borsa and Whitmore (15, 97 Copyright © 1975 by Academic Press, Inc. All rights of reproduction in any form reserved.
98
FRED ROSENFELT
17, 18). Utilizing a subline of Earle's L-cells under conditions which permitted the examination of methotrexate induced inhibition of thymidylate and purine synthesis, Borsa and Whitmore believed that their data are consistent with the interpretation that methotrexate has sites of action in addition to its inhibition of dihydrofolate reductase. In subsequent studies, these investigators (18) demonstrated that one possible alternative site of intracellular methotrexate action was the direct inhibition of the enzyme thymidylate synthetase which is necessary for DNA synthesis. Recently, Goldman (7, 16) has confirmed and extended these findings. He found that when murine L-cells were loaded with intracellular methotrexate sufficient to bind all of the dihydrofolate reductase without accumulating in the free state, there was only a small decrease in the rate of DNA synthesis. Only with the addition of further methotrexate, resulting in the accumulation of free intracellular drug in excess of the level of dihydrofolate reductase, was Goldman able to produce complete suppression of DNA synthesis within his experimental system. He also demonstrated that methotrexate is capable of binding to and inhibiting thymidylate synthetase, although as Goldman indicated, this does not necessarily mean this is an alternate site of methotrexate activity in vivo. In a recent report concerned with this problem, Sirotnak and his co-workers (55) concluded that the target enzyme dihydrofolate reductase is the primary site of methotrexate activity at least in L1210 murine leukemia cells. They ascribed the requirement for high levels of free intracellular methotrexate (i.e., in excess of that required to completely inhibit dihydrofolate reductase) described by other investigators (7, 16-18, 35-37) to an aberrant metabolic state under the in vitro conditions utilized in these studies which does not allow normal proliferative activity (55). To achieve 100% inhibition of DNA synthesis in vivo according to Sirotnak et al. (55), levels of methotrexate only slightly above the content of dihydrofolate reductase are necessary to inhibit that fraction of enzyme which for unknown reasons is not bound tightly to the drug and, therefore, capable of continuing DNA synthesis. When the intracellular methotrexate concentration decreases below this level, enough enzyme would theoretically be freed to reinstitute DNA synthesis. Thus, Sirotnak and Donsbach believe that high drug concentrations are not required for attaining maximal inhibition of DNA synthesis, but rather to sustain this inhibition with decreasing plasma drug levels. While recognizing the controversy surrounding its acceptance, this multiple-site hypothesis of methotrexate action does prcvide a heuristic model that allows for increased speculation as to the nature of observable clinical phenomena. For example, the therapeutic value of high-dose methotrexate treatment followed by citrovorum rescue has long been recognized (23-28, 56-58). The basis for this combination drugrescue protocol is the observation that when citrovorum administration is delayed for several hours after methotrexate therapy, the antineoplastic effect of the drug is retained while host tissue toxicity is reduced (23-28, 50, 56-58). In the past, the accepted biochemical rationale for this treatment modality related the addition of citrovorum factor or N5-formyltetrahydrofolate to a circumvention of the methotrexate-induced block of thymidylate synthesis with a consequent reinitiation of DNA synthesis. Why this compound selectively rescues normal tissue with little decrease in therapeutic effect has remained problematical. However, the multiplesite theory of methotrexate activity coupled with recent, although still uncorroborated, reports demonstrating a greater persistence of this drug in malignant cells as compared to sensitive normal host proliferative tissue (55, 59) provides a speculative basis for the effectiveness of high-dose methotrexate-citrovorum rescue
METHOTREXATE AND CONTINUED RESEARCH
99
protocols (65). With high doses of methotrexate, one can hypothesize that saturation of the dihydrofolate reductase binding sites with a subsequent accumulation of high intracellular levels of free drug might occur in most cells. As suggested by the work of Sirotnak and Donsbach (55, 59), the intracellular drug concentration in normal cells may return to the level of the enzyme dihydrofolate reductase while that in the tumor cells remains elevated for a longer period. Should this hypothetical situation exist, citrovorum factor administered during this time would allow normal cells to recover proliferatively by bypassing the methotrexate-induced block of the enzyme dihydrofolate reductase. DNA synthesis in the tumor cells, on the other hand, would still be prevented despite the presence of citrovorum factor by free intracellular methotrexate inhibiting alternative sites involved in DNA synthesis (65). While this alternative site may be thymidylate synthetase as suggested by the studies of Borsa et al. (15, 17, 18) and others (7, 16), other possibilities exist. For instance, Hryniuk has proposed that cells exposed to methotrexate may die a purineless death as well as a thymineless death. Thus, a continued block in purine synthesis produced by high intracellular drug levels in malignant as compared to normal cells may be the alternative and/or additional site of methotrexate activity. An interesting therapeutic aspect of this multiple site hypothesis of (methotrexate) activity is the expectation that prolonged high intracellular concentrations of methotrexate may prove to have a greater antitumor effect, particularly if this elevation is limited to the malignant cells. In this regard, Zager et al. (67) and Fyfe and Goldman (16,66) have demonstrated that methotrexate uptake is augmented by vincristine as a consequence of inhibition of efflux from the cell or by enhanced binding of methotrexate to alternative sites. While its in vivo effect has been questioned (68), vincristine may enhance methotrexate activity and certainly should be investigated further. Despite its speculative basis, the above hypothesis indicates the possible significance of the multiple site of action theory in explaining experimentally and clinically derived data as well as possible future therapeutic manipulations deserving of further investigation. Irrespective of its specific site of action, there is ample evidence that methotrexate produces its cytoxic effects by inhibiting DNA synthesis (1-8, 16-20, 22, 29-34, 3841, 45-48, 55). As such, another important aspect of this drug, only recently appreciated, is its effects upon groups of normal and malignant cells actively traversing the cell cycle (45-49, 51-53, 63, 64). The cell cycle is defined as the interval between the midpoint of mitosis in the parent cell and the midpoint of the subsequent mitosis in one or both daughter cells (60). While the duration of the cycle may vary with different cell types, the following four phases have been uniformly identified in most systems: (1) G1, the period between completion of mitosis and the onset of DNA synthesis; (2) S, the DNA replication period; (3) G2, the time between completion of DNA synthesis and the onset of mitosis, (4) M or mitosis, the period involving prophase, metaphase, anaphase, and telophase during which the chromosomes segregate, and are equally distributed to the dividing cells (60). Ernst and Kilman (45, 46) have demonstrated that methotrexate arrests human leukemia cells in the S phase for a period of time corresponding to the length of DNA synthesis. This cytocidal effect was restricted to those cells that were in the S phase during methotrexate administration. Cells in G1, M, G2 were not directly influenced by the drug. This cell cycle phase specificity of methotrexate has been confirmed by other investigators (47, 50). Based upon these observations, treatment schedules have been proposed in an at-
100
FRED ROSENFELT
tempt to utilize the cell cycle stage specificity of methotrexate in accordance with known tumor cell kinetics (1-5, 46, 47, 52-54, 61, 64). One important observation in this regard is the fact that cells not actively involved in DNA synthesis; therefore, not in the S phase of the cell cycle, are immune to the action of this drug (1-5). Malignant cells not dividing (22) or tumor cells speculated to have a biochemically altered S phase (14) would thus be expected to demonstrate a decreased response to methotrexate. Furthermore, chemotherapeutic agents interfering with cell cycle transit; thereby preventing malignant cells from entering the S phase, would also be expected to antagonize the activity of a subsequently administered dose of methotrexate (61, 63). In fact, this mechanism has recently been proposed as one possible explanation for the antagonistic effect of prior L-asparaginase therapy upon the activity of a subsequent dose of methotrexate (61, 64). While some investigators have achieved an enhancement of methotrexate activity through the manipulation of tumor cell kinetics in this manner (53, 54, 63), this approach has not yet achieved wide clinical success. Despite this lack of consistent success, cell cycle consideration will undoubtedly play an increasingly important role in the future planning of therapeutic protocols with methotrexate as well as with other drugs. As is evident, methotrexate remains a clinically important antineoplastic agent. Fortunately, in spite of the existence of a widely recognized and accepted mode of drug action, research concerning the biochemical activity of methotrexate has continued. The new findings reviewed here, for instance, have provided a greater understanding of the biochemical mechanisms underlying this drug's action. New theories explaining empirically derived data are now possible, and these will inevitably act to stimulate further investigations. More importantly, this continuing examination of drug actions is revealing greater details of the nature of the malignant process. The importance of tumor cell growth parameters has thus been recognized and is currently being exploited for an improved therapeutic effect from cell cycle stage-specific agents as methotrexate. It should be clear, then, that in addition to the therapeutic advances from new antineoplastic agents or related treatment modalities as immunotherapy, continued research and some new thoughts about standard chemotherapeutic drugs such as methotrexate should hopefully allow for the eventual successful treatment of neoplastic diseases. REFERENCES 1. Calabresi, P., and Parks, R. E., Jr: Chemotherapy of neoplasic diseases. In (L. S. Goodman and A. Gilman, Eds.), "The Pharmacological Basis of Therapeutics" 4th ed., pp. 1344-1398. Macmillan, New York, 1970. 2. Holland, J..F., and Frei, E. III, Eds. "Cancer Medicine." Lea & Febiger, Philadelphia, 1973. 3. Sartorelli A. C., and Creasey, W. A. Cancer chemotherapy. Annu. Rev. Pharmacol. 9,51-72 (1969). 4. Brodsky, I., and Kahn, S. B., Eds. "Cancer Chemotherapy: Basic and Clinical Applications. 2nd ed. Grune & Stratton, New York. 5. Bertino, J. R. The mechanism of action of the folate antagonists in man. Cancer Res. 23, 1286-1306 (1963). 6. Borsa, J., and Whitmore, G. T. Studies relating to the mode of action of methotrexate. II. Studies on sites of action in L-cells in vitro. Mol. Pharmacol. 5, 303-317 (1969). 7. Goldman D. The mechanism of action of methotrexate. I. Interaction with a low-affinity intracellular site required for maximum inhibition of deoxyribonucleic acid synthesis in L-cell mouse fibroblasts. Mol. Pharmacol. 10, 257-274 (1974). 8. Hryniuk, W. M., Bishop, A., and Forester, J. Clinical correlates of in vitro effect of methotrexate on acute leukemia blasts. Cancer Res. 34, 2823-2829 (1974). 9. Hryniuk, W. M., and Bertino, J. R. Treatment of leukemia with large doses of methotrexate and folinic acid: Clinical-biochemical correlates. J. Clin. Invest. 48, 2140-2155 (1969).
METHOTREXATE AND CONTINUED RESEARCH
101
10. Vogler, W. R., Mingioli, E. S., and Garwook, F. A. The effect of methotrexate on granulocytic stem cells and granulopoiesis. Cancer Res. 33, 1628-1633 (1970). 11. Bischoff, K. B., Dedrick, R. L., and Zaharko, D. S. Preliminary model for methotrexate pharmokinetics. J. Pharmaceut. Sci. 59, 149-153 (1970). 12. Futterman, S. Enzymatic reduction of folic acid and dihydrofolic acid to tetrahydrofolic Acid. J. Biol. Chem. 288, 1031-1038 (1971). 13. Zahrzewski, S. F., and Nichol, C. A. Evidence for a single enzyme reducing folate and dihydrofolate. J. Biol. Chem. 235,2984-2988 (1961). 14. Hryniuk, W. M., Fischer, G. A. and Bertino, J. R. S-phase cells of rapidly growing and resting populations: Differences in response to methotrexate. Mol. Pharmacol. 5, 557-564 (1969). 15. Borsa, J., and Whitmore, G. F. Cell killing studies on the mode of action of methotrexate on L-cells in vitro. Cancer Res. 29, 737-744 (1969). 16. Goldman, I. D., and Fyfe, M. J. The mechanism of action of methotrexate II. Augmentation by vincristine of inhibition of deoxyribonucleic acid synthesis by methotrexate, in Ehrlich ascites tumor cells. Mol. Pharmacol. 10, 275-282 (1973). 17. Borsa, J., and Whitmore, G. F. Studies relating to the mode of action of methotrexate. Mol. Pharmacol. 5, 305-317, (1969). 18. Borsa, J., and Whitmore, G. F. Studies relating to the mode of action of methotrexate III. Inhibition of thymidylate synthesis in tissue culture cells and in cell-free synthesis. Mol. Pharmacol. 5, 318332 (1969). 19. Harrap, K. R., Hill, B. T., Furness, M. E., and Hart, L. I. Sites of action of amethopterin, intrinsic and acquired drug resistance, Ann. N. Y. A cad. Sci. 186, 312-324 (1971). 20. Chabner, B. A., and Young, R. C. Threshold methotrexate concentration for in vitro inhibition of DNA synthesis in normal and tumorous target tissues. J. Clin. Invest. 52, 1804-1811 (1973). 21. Margolis, S., Philips, F. S., and Sternberg, S. S. The cytotoxicity of methotrexate in mouse small intestine in relation to inhibition of folic acid reductase and of DNA synthesis. Cancer Res. 31, 2037-2046(1971). 22. Hryniuk, W., and Bertino, J. R. Growth rate and cell kill. Ann. N. Y. Acad. Sci. 186, 330-342 (1971). 23. Sartorelli, A. C., Upchurch, H. F., and Booth, B. Effects of folinic acid on amethopterin-induced inhibition of the Ehrlich ascites carcinoma. Cancer Res. 22, 102 (1962). 24. Djerassi, I., Abir, E., Roger, G. L., and Treat, C. L. Long term remission in childhood acute leukemia. The use of infrequent infusions of methotrexate; supportive roles of platelet transfusions on citrovorum factor. Clin. Pediat. 5, 502-509 (1966). 25. Goldin, A., Vendetti, J. M., Kline, I., and Mantel, N. Eradication of leukemia cells (L1210) by methotrexate and methotrexate plus citrovorum factor. Nature (London) 212, 1548-1550 (1966). 26. Hryniuk, W., Zanes, R., Guzman, P., and Bertino, J. R. Treatment of acute leukemia with large doses intermittent infusions of methotrexate followed by leucovorum, clinical and biochemical studies. Clin. Res. 15, 336 (1967). 27. Lefkowitz, E., Papac, R. T., and Bertino, J. R. Studies of head and neck cancer. III. Toxicity of 24 hour infusions of methotrexate and protection by leucovorum in patients with epidermoid carcinoma. Cancer Chemother. Rep. 51, 305-311 (1967). 28. Mitchell, M. S., Wawro, H. W., DeConti, R. C., Kaplan, S. R., Papac, R., and Bertino, J. R. Effectiveness of high-dose infusions of methotrexate followed by leucovorum in carcinoma of the head and neck. Cancer Res. 28, 1088-1094 (1968). 29. Bertino, J. R., Cashmore, A. R., and Hillcoat, B. L. "Induction" of dihydrofolate reductase; purification and properties of the "induced" human erythrocyte and leucocyte enzyme and normal bone marrow enzyme. Cancer Res. 30, 2372-2378 (1970). 30. Bertino, J. R., Booth, B. H., Cashmore, A., Beeber, A. L., and Sartorelli, A. C. Studies of the inhibition of dihydrofolate reductase of folate antagonists. J. Biol. Chem. 239, 479-485 (1964). 31. Roberts, D., Wodinsky, I., and Hall, T. The level of enzyme activity and response to methotrexate of transplantable mouse tumors. Cancer Res. 25, 1899- 1903 (1965). 32. Fischer, G. A. Increased levels of folic acid reductase as a mechanism of resistance to amethopterin in leukemic cells. Biochem. Pharmacol. 7,75-77 (1968). 33. Misra, D. K., Humphreys, S. R., Friedken, M., Goldin, A., and Crawford, E. Increased dihydrofolate reductase activity as a possible basis of drug resistance in leukemia, Nature (London) 189, 30-42 (1961). 34. Blumenthal, G., and Crunberg, D. M. Evidence for two molecular species of dihydrofolate reductase in amethopterin resistant and sensitive cells of the mouse leukemia L4946. Oncology 24, 223-229
(1970).
102
FRED ROSENFELT
35. Roberts, D., and Wodinsky, I. On the poor correlation between the inhibition by methotrexate of dihydrofolated reductase and of deoxynucleoside incorporation into DNA. Cancer Res. 28, 19551962 (1968). 36. Roberts, D., Hall, T. C., and Rosenthal, D. Coordination changes in biochemical patterns: The effect of sytosine arabinoside and methotrexate on leucocytes from patients with acute granulocytic leukemia. Cancer Res. 29, 571-578 (1969). 37. Roberts, D., and Hall T. C. Dihydrofolate reductase activity and deoxynucleoside incorporation into DNA of human leukocytes: Relation to methotrexate administration. Cancer 29,905-910 (1967). 38. Kim, J. H., Perez, A. G., and Djordjevic, B. Studies on unbalanced growth in synchronized HeLa cells. Cancer Res. 28, 2443-2447 (1968). 39. Cohen, S. S., and Barner, H. D. Studies on unbalanced growth in Escherichia coli. Proc. Nat. Acad. Sci. USA 40,885-893 (1954). 40. Maalae, O., and Hanawalt, P. C. Thymine deficiency and the normal DNA replication cycle. J. Mol. Biol. 3, 144-155 (1961). 41. Ruechert, R. R., and Meuller, C. C. Studies on unbalanced growth in tissue culture. Instruction and consequences of thymidine deficiency. Cancer Res. 20, 1184-1191 (1960). 42. Hoffbrand, A. V., and Tripp, E. Unbalanced deoxyribonucleotide synthesis caused by methotrexate. Brit. Med. J. 140, 318 (1972). 43. Hryniuk, W. M. Purineless death as a link between growth rate and cytotoxicity by methotrexate. Cancer Res. 32, 1506-1511 (1972). 44. Cohen, S. S. On the nature of thymineless death. Ann. N. Y. Acad. Sci. 186, 292-301 (1971). 45. Ernst, P., and Killman, S. A. Effect of antileukemia drugs on cell cycle of tumor leukemia blast cells in vivo. Acta Med. Scand. 186, 239 (1969). 46. Ernst, P., and Killman, S. A. Perturbation of generation cycle of human leukemic myeloblasts in vivo by methotrexate. Blood 38, 689-705 (1971). 47. Bhuyan, B. K., Fraser, T. J., Gray, L. G., Kuenteel, S. L., and Neil, G. L. Cell-kill kinetics of several S-phase-specific drugs. Cancer Res. 33, 888-894 (1973). 48. Bhuyan, B. K., Scherdt, L. G., and Fraser, T. J. Cell cycle phase specificity of antitumor agents. Cancer Res. 32, 398-400 (1972). 49. Young, R. C. and DeVita, V. The effect of chemotherapy on the growth characteristics and cellular kinetics of leukemia L1310. Cancer Res. 30, 1789-1794(1970). 50. Palen, H., Ottebro, R., and Engeset, A. The effect of methotrexate and leucovorum on cell division in Chang cells. Cancer 18,41-48 (1965). 51. Goldin, A. Factors pertaining to complete drug-induced remission of tumor in animals and man. Cancer Res. 29, 2285-2291 (1969). 52. Schabel, F. M. The use of tumor growth kinetics in planning "curative chemotherapy" of advanced solid tumor. Cancer Res. 29, 2384-2389 (1969). 53. Vendetti, J. M., and Goldin, A. Chemotherapy of advanced mouse leukemia L1210; Comparison of methotrexate alone and in sequential therapy. Cancer Res. 24, 1457-1460 (1964). 54. Straus, M. J., Mantel, N., and Goldin, A. Effects of priming dose schedules in methotrexate treatment of mouse leukemia L1210. Cancer Res. 31, 1429-1433 (1971). 55. Sirotnak, F. M., and Donsbach, R. C. The intracellular concentration dependence of antifolate inhibition of DNA synthesis in L 1210 leukemia cells. Cancer Res. 34, 3332-3340 (1974). 56. Jaffe, N., Frei, E., Traggis, D., and Bishop, Y. Adjuvant methotrexate and citrovorum-factor treatment of osteogenic sarcoma. N. Engi. J. Med. 291,994-997 (1974). 57. Djerassi, J., Rominger, C. J., Kim, J. S. Phase I study of high doses of methotrexate with citrovorum factor in patients with lung cancer. Cancer 30, 22-30 (1972). 58. Levitt, M., Mosher, M. B., DeConti, R. C. Improved therapeutic Index of Methotrexate with "Leucovorin Rescue," Cancer Res. 33, 1729-1734 (1973). 59. Sirotnak, F. M., and Donsbach, R. C. Differential cell permeability and the basis for selective activity of methotrexate during therapy of the L 1210 leukemia. Cancer Res. 33, 1290-1294 (1973). 60. Baserga, R. Biochemistry of the cell cycle: A review cell tissue kinetics. 1, 167-191 (1968). 61. Capizzi, R. L., Summers, W. P. and Bertino, J. R. Discussion paper: L-Asparaginase induced alteration of amethopterin (methotrexate) activity in mouse leukemia L5178Y. Ann. N. Y. Acad. Sci. 186, 302-311 (197 1). 62. Skipper, H. E., Schabel, F. M., and Wilcox, W. S. Experimental evaluation of potential anticancer agents. XXI. Scheduling of arabinosylcytosine to take advantage of its S-phase specificity against leukemia cells. Cancer Chemother. Rep. 51, 125 (1967).
METHOTREXATE AND CONTINUED RESEARCH
103
63. Vadlamudi, S., Krishna, B., Reddy, V. V. S., and Goldin, A. Schedule-dependent therapeutic synergism for L-asparaginase and methotrexate in leukemic (L5 178Y) mice. Cancer Res. 33, 20142019 (1973). 64. Rosenfelt, F. P., and Capizzi, R. L. Unpublished results. 65. Rosenfelt, F. P. Letter. N. Engl. J. Med. 292,432 (1974). 66. Bertino, J. R., and Johns, D. G. Folate antagonists. Annu. Rev. Med. 18,27-34 (1974). 67. Fyfe, M. J., and Goldman, I. D. Characteristics of the vincristine-induced augmentation of methotrexate uptake in Ehrlich ascites tumor cells. J. Biol. Chem. 248, 5607 (1973). 68. Zager, R. F., Frisby, S. A., and Oliverio, V. T. The effects of antibiotics and cancer chemotherapeutic agents on the cellular transport and antitumor activity of methotrexate in L1210 murine leukemia. Cancer Res. 33, 1070-1076 (1973). 69. Bender, R. A., Drake, J. C., Fisher, R., and Chabner, B. A. Letter. N. Engl. J. Med. 292,433 (1974).
48, 97-103 (1975)
Methotrexate and the Need for Continued Research FRED ROSENFELT Yale University School of Medicine, New Haven, Connecticut 06510 Received January 7, 1975
With the current national commitment to cancer research, new and challenging areas of investigation are being planned. Hopefully, this inevitably expanding search into the nature and treatment of malignant tumors will include a continued examination of current chemotherapeutic agents. These standard tumoricidal drugs have proved to be effective in the therapy of numerous cancers despite an incomplete understanding of their mode of action in many cases (1-4). A further unraveling of the pharmacologic basis of these drugs should undoubtedly lead to a greater therapeutic effect, perhaps producing longer remissions and increased cures. A recent example of how continued basic research on available antineoplastic agents may reveal findings pertinent to clinical chemotherapy concerns methotrexate, a drug frequently employed in the treatment of cancer (1-5, 66). Traditionally, the mode of action of this drug has assumed to be as an analog of the vitamin folic acid. In humans, folic acid is biologically inactive, requiring reduction by an enzyme termed "dihydrofolate reductase" to dehydrofolate, and then to tetrahydrofolate. This latter compound accepts 1-carbon fragments from various sources to produce the folate coenzymes. These compounds act as 1-carbon donors in numerous biochemical reactions, including the synthesis of the B-carbon of serines, the formation of the C-2 and C-8 units of the purine skeleton, and the methylation of thymidylic acid (5-7, 10, 12, 13, 15, 17, 18, 66). Methotrexate acts, then, by irreversibly binding to and inhibiting the enzyme dihydrofolate reductase thereby interfering with the maintenance of intracellular pools of reduced folates, particularly N5' '0-methylenetetrahydrofolate (1, 5, 11-13, 66). This latter compound is necessary for the conversion of deoxyuridylate to thymidylate which is an essential component of DNA. As is evident, methotrexate, as well as other antifolate agents, essentially act by inhibiting DNA synthesis (1-9, 11-12, 14-20, 27-40, 54, 66). For many years it was accepted that the biological effects of methotrexate were due entirely to its inhibition of the enzyme dihydrofolate reductase. Since free or unbound intracellular methotrexate accumulates only after all the dihydrofolate reductase binding sites are saturated, little biological activity or significance has been ascribed to this free intracellular drug component. However, evidence has been accumulating that methotrexate may have a second site of action of equal or greater importance than the inhibition of dihydrofolate reductase and furthermore, that it is the unbound intracellular drug that is involved with this alternative site. Roberts et al. (35-37), several years ago, demonstrated a lack of correlation between the inhibition of dihydrofolate reductase activity in human leukemia cells and the response of leukemia patients to methotrexate therapy. On the basis of their studies, although without substantiating data, this group theorized that free intracellular methotrexate as well as that bound to dihydrofolate reductase may have a significant role in the cytocidal effect produced by this agent. Further evidence suggestive of an additional site of methotrexate activity was next provided by Borsa and Whitmore (15, 97 Copyright © 1975 by Academic Press, Inc. All rights of reproduction in any form reserved.
98
FRED ROSENFELT
17, 18). Utilizing a subline of Earle's L-cells under conditions which permitted the examination of methotrexate induced inhibition of thymidylate and purine synthesis, Borsa and Whitmore believed that their data are consistent with the interpretation that methotrexate has sites of action in addition to its inhibition of dihydrofolate reductase. In subsequent studies, these investigators (18) demonstrated that one possible alternative site of intracellular methotrexate action was the direct inhibition of the enzyme thymidylate synthetase which is necessary for DNA synthesis. Recently, Goldman (7, 16) has confirmed and extended these findings. He found that when murine L-cells were loaded with intracellular methotrexate sufficient to bind all of the dihydrofolate reductase without accumulating in the free state, there was only a small decrease in the rate of DNA synthesis. Only with the addition of further methotrexate, resulting in the accumulation of free intracellular drug in excess of the level of dihydrofolate reductase, was Goldman able to produce complete suppression of DNA synthesis within his experimental system. He also demonstrated that methotrexate is capable of binding to and inhibiting thymidylate synthetase, although as Goldman indicated, this does not necessarily mean this is an alternate site of methotrexate activity in vivo. In a recent report concerned with this problem, Sirotnak and his co-workers (55) concluded that the target enzyme dihydrofolate reductase is the primary site of methotrexate activity at least in L1210 murine leukemia cells. They ascribed the requirement for high levels of free intracellular methotrexate (i.e., in excess of that required to completely inhibit dihydrofolate reductase) described by other investigators (7, 16-18, 35-37) to an aberrant metabolic state under the in vitro conditions utilized in these studies which does not allow normal proliferative activity (55). To achieve 100% inhibition of DNA synthesis in vivo according to Sirotnak et al. (55), levels of methotrexate only slightly above the content of dihydrofolate reductase are necessary to inhibit that fraction of enzyme which for unknown reasons is not bound tightly to the drug and, therefore, capable of continuing DNA synthesis. When the intracellular methotrexate concentration decreases below this level, enough enzyme would theoretically be freed to reinstitute DNA synthesis. Thus, Sirotnak and Donsbach believe that high drug concentrations are not required for attaining maximal inhibition of DNA synthesis, but rather to sustain this inhibition with decreasing plasma drug levels. While recognizing the controversy surrounding its acceptance, this multiple-site hypothesis of methotrexate action does prcvide a heuristic model that allows for increased speculation as to the nature of observable clinical phenomena. For example, the therapeutic value of high-dose methotrexate treatment followed by citrovorum rescue has long been recognized (23-28, 56-58). The basis for this combination drugrescue protocol is the observation that when citrovorum administration is delayed for several hours after methotrexate therapy, the antineoplastic effect of the drug is retained while host tissue toxicity is reduced (23-28, 50, 56-58). In the past, the accepted biochemical rationale for this treatment modality related the addition of citrovorum factor or N5-formyltetrahydrofolate to a circumvention of the methotrexate-induced block of thymidylate synthesis with a consequent reinitiation of DNA synthesis. Why this compound selectively rescues normal tissue with little decrease in therapeutic effect has remained problematical. However, the multiplesite theory of methotrexate activity coupled with recent, although still uncorroborated, reports demonstrating a greater persistence of this drug in malignant cells as compared to sensitive normal host proliferative tissue (55, 59) provides a speculative basis for the effectiveness of high-dose methotrexate-citrovorum rescue
METHOTREXATE AND CONTINUED RESEARCH
99
protocols (65). With high doses of methotrexate, one can hypothesize that saturation of the dihydrofolate reductase binding sites with a subsequent accumulation of high intracellular levels of free drug might occur in most cells. As suggested by the work of Sirotnak and Donsbach (55, 59), the intracellular drug concentration in normal cells may return to the level of the enzyme dihydrofolate reductase while that in the tumor cells remains elevated for a longer period. Should this hypothetical situation exist, citrovorum factor administered during this time would allow normal cells to recover proliferatively by bypassing the methotrexate-induced block of the enzyme dihydrofolate reductase. DNA synthesis in the tumor cells, on the other hand, would still be prevented despite the presence of citrovorum factor by free intracellular methotrexate inhibiting alternative sites involved in DNA synthesis (65). While this alternative site may be thymidylate synthetase as suggested by the studies of Borsa et al. (15, 17, 18) and others (7, 16), other possibilities exist. For instance, Hryniuk has proposed that cells exposed to methotrexate may die a purineless death as well as a thymineless death. Thus, a continued block in purine synthesis produced by high intracellular drug levels in malignant as compared to normal cells may be the alternative and/or additional site of methotrexate activity. An interesting therapeutic aspect of this multiple site hypothesis of (methotrexate) activity is the expectation that prolonged high intracellular concentrations of methotrexate may prove to have a greater antitumor effect, particularly if this elevation is limited to the malignant cells. In this regard, Zager et al. (67) and Fyfe and Goldman (16,66) have demonstrated that methotrexate uptake is augmented by vincristine as a consequence of inhibition of efflux from the cell or by enhanced binding of methotrexate to alternative sites. While its in vivo effect has been questioned (68), vincristine may enhance methotrexate activity and certainly should be investigated further. Despite its speculative basis, the above hypothesis indicates the possible significance of the multiple site of action theory in explaining experimentally and clinically derived data as well as possible future therapeutic manipulations deserving of further investigation. Irrespective of its specific site of action, there is ample evidence that methotrexate produces its cytoxic effects by inhibiting DNA synthesis (1-8, 16-20, 22, 29-34, 3841, 45-48, 55). As such, another important aspect of this drug, only recently appreciated, is its effects upon groups of normal and malignant cells actively traversing the cell cycle (45-49, 51-53, 63, 64). The cell cycle is defined as the interval between the midpoint of mitosis in the parent cell and the midpoint of the subsequent mitosis in one or both daughter cells (60). While the duration of the cycle may vary with different cell types, the following four phases have been uniformly identified in most systems: (1) G1, the period between completion of mitosis and the onset of DNA synthesis; (2) S, the DNA replication period; (3) G2, the time between completion of DNA synthesis and the onset of mitosis, (4) M or mitosis, the period involving prophase, metaphase, anaphase, and telophase during which the chromosomes segregate, and are equally distributed to the dividing cells (60). Ernst and Kilman (45, 46) have demonstrated that methotrexate arrests human leukemia cells in the S phase for a period of time corresponding to the length of DNA synthesis. This cytocidal effect was restricted to those cells that were in the S phase during methotrexate administration. Cells in G1, M, G2 were not directly influenced by the drug. This cell cycle phase specificity of methotrexate has been confirmed by other investigators (47, 50). Based upon these observations, treatment schedules have been proposed in an at-
100
FRED ROSENFELT
tempt to utilize the cell cycle stage specificity of methotrexate in accordance with known tumor cell kinetics (1-5, 46, 47, 52-54, 61, 64). One important observation in this regard is the fact that cells not actively involved in DNA synthesis; therefore, not in the S phase of the cell cycle, are immune to the action of this drug (1-5). Malignant cells not dividing (22) or tumor cells speculated to have a biochemically altered S phase (14) would thus be expected to demonstrate a decreased response to methotrexate. Furthermore, chemotherapeutic agents interfering with cell cycle transit; thereby preventing malignant cells from entering the S phase, would also be expected to antagonize the activity of a subsequently administered dose of methotrexate (61, 63). In fact, this mechanism has recently been proposed as one possible explanation for the antagonistic effect of prior L-asparaginase therapy upon the activity of a subsequent dose of methotrexate (61, 64). While some investigators have achieved an enhancement of methotrexate activity through the manipulation of tumor cell kinetics in this manner (53, 54, 63), this approach has not yet achieved wide clinical success. Despite this lack of consistent success, cell cycle consideration will undoubtedly play an increasingly important role in the future planning of therapeutic protocols with methotrexate as well as with other drugs. As is evident, methotrexate remains a clinically important antineoplastic agent. Fortunately, in spite of the existence of a widely recognized and accepted mode of drug action, research concerning the biochemical activity of methotrexate has continued. The new findings reviewed here, for instance, have provided a greater understanding of the biochemical mechanisms underlying this drug's action. New theories explaining empirically derived data are now possible, and these will inevitably act to stimulate further investigations. More importantly, this continuing examination of drug actions is revealing greater details of the nature of the malignant process. The importance of tumor cell growth parameters has thus been recognized and is currently being exploited for an improved therapeutic effect from cell cycle stage-specific agents as methotrexate. It should be clear, then, that in addition to the therapeutic advances from new antineoplastic agents or related treatment modalities as immunotherapy, continued research and some new thoughts about standard chemotherapeutic drugs such as methotrexate should hopefully allow for the eventual successful treatment of neoplastic diseases. REFERENCES 1. Calabresi, P., and Parks, R. E., Jr: Chemotherapy of neoplasic diseases. In (L. S. Goodman and A. Gilman, Eds.), "The Pharmacological Basis of Therapeutics" 4th ed., pp. 1344-1398. Macmillan, New York, 1970. 2. Holland, J..F., and Frei, E. III, Eds. "Cancer Medicine." Lea & Febiger, Philadelphia, 1973. 3. Sartorelli A. C., and Creasey, W. A. Cancer chemotherapy. Annu. Rev. Pharmacol. 9,51-72 (1969). 4. Brodsky, I., and Kahn, S. B., Eds. "Cancer Chemotherapy: Basic and Clinical Applications. 2nd ed. Grune & Stratton, New York. 5. Bertino, J. R. The mechanism of action of the folate antagonists in man. Cancer Res. 23, 1286-1306 (1963). 6. Borsa, J., and Whitmore, G. T. Studies relating to the mode of action of methotrexate. II. Studies on sites of action in L-cells in vitro. Mol. Pharmacol. 5, 303-317 (1969). 7. Goldman D. The mechanism of action of methotrexate. I. Interaction with a low-affinity intracellular site required for maximum inhibition of deoxyribonucleic acid synthesis in L-cell mouse fibroblasts. Mol. Pharmacol. 10, 257-274 (1974). 8. Hryniuk, W. M., Bishop, A., and Forester, J. Clinical correlates of in vitro effect of methotrexate on acute leukemia blasts. Cancer Res. 34, 2823-2829 (1974). 9. Hryniuk, W. M., and Bertino, J. R. Treatment of leukemia with large doses of methotrexate and folinic acid: Clinical-biochemical correlates. J. Clin. Invest. 48, 2140-2155 (1969).
METHOTREXATE AND CONTINUED RESEARCH
101
10. Vogler, W. R., Mingioli, E. S., and Garwook, F. A. The effect of methotrexate on granulocytic stem cells and granulopoiesis. Cancer Res. 33, 1628-1633 (1970). 11. Bischoff, K. B., Dedrick, R. L., and Zaharko, D. S. Preliminary model for methotrexate pharmokinetics. J. Pharmaceut. Sci. 59, 149-153 (1970). 12. Futterman, S. Enzymatic reduction of folic acid and dihydrofolic acid to tetrahydrofolic Acid. J. Biol. Chem. 288, 1031-1038 (1971). 13. Zahrzewski, S. F., and Nichol, C. A. Evidence for a single enzyme reducing folate and dihydrofolate. J. Biol. Chem. 235,2984-2988 (1961). 14. Hryniuk, W. M., Fischer, G. A. and Bertino, J. R. S-phase cells of rapidly growing and resting populations: Differences in response to methotrexate. Mol. Pharmacol. 5, 557-564 (1969). 15. Borsa, J., and Whitmore, G. F. Cell killing studies on the mode of action of methotrexate on L-cells in vitro. Cancer Res. 29, 737-744 (1969). 16. Goldman, I. D., and Fyfe, M. J. The mechanism of action of methotrexate II. Augmentation by vincristine of inhibition of deoxyribonucleic acid synthesis by methotrexate, in Ehrlich ascites tumor cells. Mol. Pharmacol. 10, 275-282 (1973). 17. Borsa, J., and Whitmore, G. F. Studies relating to the mode of action of methotrexate. Mol. Pharmacol. 5, 305-317, (1969). 18. Borsa, J., and Whitmore, G. F. Studies relating to the mode of action of methotrexate III. Inhibition of thymidylate synthesis in tissue culture cells and in cell-free synthesis. Mol. Pharmacol. 5, 318332 (1969). 19. Harrap, K. R., Hill, B. T., Furness, M. E., and Hart, L. I. Sites of action of amethopterin, intrinsic and acquired drug resistance, Ann. N. Y. A cad. Sci. 186, 312-324 (1971). 20. Chabner, B. A., and Young, R. C. Threshold methotrexate concentration for in vitro inhibition of DNA synthesis in normal and tumorous target tissues. J. Clin. Invest. 52, 1804-1811 (1973). 21. Margolis, S., Philips, F. S., and Sternberg, S. S. The cytotoxicity of methotrexate in mouse small intestine in relation to inhibition of folic acid reductase and of DNA synthesis. Cancer Res. 31, 2037-2046(1971). 22. Hryniuk, W., and Bertino, J. R. Growth rate and cell kill. Ann. N. Y. Acad. Sci. 186, 330-342 (1971). 23. Sartorelli, A. C., Upchurch, H. F., and Booth, B. Effects of folinic acid on amethopterin-induced inhibition of the Ehrlich ascites carcinoma. Cancer Res. 22, 102 (1962). 24. Djerassi, I., Abir, E., Roger, G. L., and Treat, C. L. Long term remission in childhood acute leukemia. The use of infrequent infusions of methotrexate; supportive roles of platelet transfusions on citrovorum factor. Clin. Pediat. 5, 502-509 (1966). 25. Goldin, A., Vendetti, J. M., Kline, I., and Mantel, N. Eradication of leukemia cells (L1210) by methotrexate and methotrexate plus citrovorum factor. Nature (London) 212, 1548-1550 (1966). 26. Hryniuk, W., Zanes, R., Guzman, P., and Bertino, J. R. Treatment of acute leukemia with large doses intermittent infusions of methotrexate followed by leucovorum, clinical and biochemical studies. Clin. Res. 15, 336 (1967). 27. Lefkowitz, E., Papac, R. T., and Bertino, J. R. Studies of head and neck cancer. III. Toxicity of 24 hour infusions of methotrexate and protection by leucovorum in patients with epidermoid carcinoma. Cancer Chemother. Rep. 51, 305-311 (1967). 28. Mitchell, M. S., Wawro, H. W., DeConti, R. C., Kaplan, S. R., Papac, R., and Bertino, J. R. Effectiveness of high-dose infusions of methotrexate followed by leucovorum in carcinoma of the head and neck. Cancer Res. 28, 1088-1094 (1968). 29. Bertino, J. R., Cashmore, A. R., and Hillcoat, B. L. "Induction" of dihydrofolate reductase; purification and properties of the "induced" human erythrocyte and leucocyte enzyme and normal bone marrow enzyme. Cancer Res. 30, 2372-2378 (1970). 30. Bertino, J. R., Booth, B. H., Cashmore, A., Beeber, A. L., and Sartorelli, A. C. Studies of the inhibition of dihydrofolate reductase of folate antagonists. J. Biol. Chem. 239, 479-485 (1964). 31. Roberts, D., Wodinsky, I., and Hall, T. The level of enzyme activity and response to methotrexate of transplantable mouse tumors. Cancer Res. 25, 1899- 1903 (1965). 32. Fischer, G. A. Increased levels of folic acid reductase as a mechanism of resistance to amethopterin in leukemic cells. Biochem. Pharmacol. 7,75-77 (1968). 33. Misra, D. K., Humphreys, S. R., Friedken, M., Goldin, A., and Crawford, E. Increased dihydrofolate reductase activity as a possible basis of drug resistance in leukemia, Nature (London) 189, 30-42 (1961). 34. Blumenthal, G., and Crunberg, D. M. Evidence for two molecular species of dihydrofolate reductase in amethopterin resistant and sensitive cells of the mouse leukemia L4946. Oncology 24, 223-229
(1970).
102
FRED ROSENFELT
35. Roberts, D., and Wodinsky, I. On the poor correlation between the inhibition by methotrexate of dihydrofolated reductase and of deoxynucleoside incorporation into DNA. Cancer Res. 28, 19551962 (1968). 36. Roberts, D., Hall, T. C., and Rosenthal, D. Coordination changes in biochemical patterns: The effect of sytosine arabinoside and methotrexate on leucocytes from patients with acute granulocytic leukemia. Cancer Res. 29, 571-578 (1969). 37. Roberts, D., and Hall T. C. Dihydrofolate reductase activity and deoxynucleoside incorporation into DNA of human leukocytes: Relation to methotrexate administration. Cancer 29,905-910 (1967). 38. Kim, J. H., Perez, A. G., and Djordjevic, B. Studies on unbalanced growth in synchronized HeLa cells. Cancer Res. 28, 2443-2447 (1968). 39. Cohen, S. S., and Barner, H. D. Studies on unbalanced growth in Escherichia coli. Proc. Nat. Acad. Sci. USA 40,885-893 (1954). 40. Maalae, O., and Hanawalt, P. C. Thymine deficiency and the normal DNA replication cycle. J. Mol. Biol. 3, 144-155 (1961). 41. Ruechert, R. R., and Meuller, C. C. Studies on unbalanced growth in tissue culture. Instruction and consequences of thymidine deficiency. Cancer Res. 20, 1184-1191 (1960). 42. Hoffbrand, A. V., and Tripp, E. Unbalanced deoxyribonucleotide synthesis caused by methotrexate. Brit. Med. J. 140, 318 (1972). 43. Hryniuk, W. M. Purineless death as a link between growth rate and cytotoxicity by methotrexate. Cancer Res. 32, 1506-1511 (1972). 44. Cohen, S. S. On the nature of thymineless death. Ann. N. Y. Acad. Sci. 186, 292-301 (1971). 45. Ernst, P., and Killman, S. A. Effect of antileukemia drugs on cell cycle of tumor leukemia blast cells in vivo. Acta Med. Scand. 186, 239 (1969). 46. Ernst, P., and Killman, S. A. Perturbation of generation cycle of human leukemic myeloblasts in vivo by methotrexate. Blood 38, 689-705 (1971). 47. Bhuyan, B. K., Fraser, T. J., Gray, L. G., Kuenteel, S. L., and Neil, G. L. Cell-kill kinetics of several S-phase-specific drugs. Cancer Res. 33, 888-894 (1973). 48. Bhuyan, B. K., Scherdt, L. G., and Fraser, T. J. Cell cycle phase specificity of antitumor agents. Cancer Res. 32, 398-400 (1972). 49. Young, R. C. and DeVita, V. The effect of chemotherapy on the growth characteristics and cellular kinetics of leukemia L1310. Cancer Res. 30, 1789-1794(1970). 50. Palen, H., Ottebro, R., and Engeset, A. The effect of methotrexate and leucovorum on cell division in Chang cells. Cancer 18,41-48 (1965). 51. Goldin, A. Factors pertaining to complete drug-induced remission of tumor in animals and man. Cancer Res. 29, 2285-2291 (1969). 52. Schabel, F. M. The use of tumor growth kinetics in planning "curative chemotherapy" of advanced solid tumor. Cancer Res. 29, 2384-2389 (1969). 53. Vendetti, J. M., and Goldin, A. Chemotherapy of advanced mouse leukemia L1210; Comparison of methotrexate alone and in sequential therapy. Cancer Res. 24, 1457-1460 (1964). 54. Straus, M. J., Mantel, N., and Goldin, A. Effects of priming dose schedules in methotrexate treatment of mouse leukemia L1210. Cancer Res. 31, 1429-1433 (1971). 55. Sirotnak, F. M., and Donsbach, R. C. The intracellular concentration dependence of antifolate inhibition of DNA synthesis in L 1210 leukemia cells. Cancer Res. 34, 3332-3340 (1974). 56. Jaffe, N., Frei, E., Traggis, D., and Bishop, Y. Adjuvant methotrexate and citrovorum-factor treatment of osteogenic sarcoma. N. Engi. J. Med. 291,994-997 (1974). 57. Djerassi, J., Rominger, C. J., Kim, J. S. Phase I study of high doses of methotrexate with citrovorum factor in patients with lung cancer. Cancer 30, 22-30 (1972). 58. Levitt, M., Mosher, M. B., DeConti, R. C. Improved therapeutic Index of Methotrexate with "Leucovorin Rescue," Cancer Res. 33, 1729-1734 (1973). 59. Sirotnak, F. M., and Donsbach, R. C. Differential cell permeability and the basis for selective activity of methotrexate during therapy of the L 1210 leukemia. Cancer Res. 33, 1290-1294 (1973). 60. Baserga, R. Biochemistry of the cell cycle: A review cell tissue kinetics. 1, 167-191 (1968). 61. Capizzi, R. L., Summers, W. P. and Bertino, J. R. Discussion paper: L-Asparaginase induced alteration of amethopterin (methotrexate) activity in mouse leukemia L5178Y. Ann. N. Y. Acad. Sci. 186, 302-311 (197 1). 62. Skipper, H. E., Schabel, F. M., and Wilcox, W. S. Experimental evaluation of potential anticancer agents. XXI. Scheduling of arabinosylcytosine to take advantage of its S-phase specificity against leukemia cells. Cancer Chemother. Rep. 51, 125 (1967).
METHOTREXATE AND CONTINUED RESEARCH
103
63. Vadlamudi, S., Krishna, B., Reddy, V. V. S., and Goldin, A. Schedule-dependent therapeutic synergism for L-asparaginase and methotrexate in leukemic (L5 178Y) mice. Cancer Res. 33, 20142019 (1973). 64. Rosenfelt, F. P., and Capizzi, R. L. Unpublished results. 65. Rosenfelt, F. P. Letter. N. Engl. J. Med. 292,432 (1974). 66. Bertino, J. R., and Johns, D. G. Folate antagonists. Annu. Rev. Med. 18,27-34 (1974). 67. Fyfe, M. J., and Goldman, I. D. Characteristics of the vincristine-induced augmentation of methotrexate uptake in Ehrlich ascites tumor cells. J. Biol. Chem. 248, 5607 (1973). 68. Zager, R. F., Frisby, S. A., and Oliverio, V. T. The effects of antibiotics and cancer chemotherapeutic agents on the cellular transport and antitumor activity of methotrexate in L1210 murine leukemia. Cancer Res. 33, 1070-1076 (1973). 69. Bender, R. A., Drake, J. C., Fisher, R., and Chabner, B. A. Letter. N. Engl. J. Med. 292,433 (1974).
