The combination of N-(phosphonacetyl)-L-aspartate (PALA), 6-methylmercaptopurine riboside (MMPR), and 6-aminonicotinamide (6AN) has been shown to be an effective antineoplastic regimen and also to enhance the effects of their other antineoplastic agents (1-4). To further enhance the effect of this combination, we investigated the effects of adding adriamycin, at its maximally tolerated dose, to this regimen. The response rate (complete regression + partial regression) for the four-drug regimen was higher than for the three-drug regimen, and the tumor growth delay was also significantly higher than for treatment with PALA, MMPR, 6AN, or after treatment with maximally tolerated doses of adriamycin alone (11 mg/kg). The addition of adriamycin to PALA, MMPR, 6AN did not result in enhancement of the effect of radiation, as measured by tumor growth delay studies and tumor control (complete and partial regression rate). The mechanism of action of the combination of PALA, MMPR, and 6AN is not known definitively, but a possible mechanism previously suggested is biochemical modulation of energy metabolism and inhibition of production of tumor ATP. Treatment with PALA< MMPR, 6AN, and adriamycin (at 2.5 hr post MMPR, 6AN) resulted in a nadir NTP/Pi value, as determined by P-31 NMR spectroscopy, at approximately 10 hr post MMPR + 6AN (7.5 hr post adriamycin), which was not significantly different from the NTP/Pi value determined after treatment with the three-drug combination.

Treatment with a combination (PMA) of (N-phosphonacetyl)-L-aspartic acid (PALA), methylmercaptopurine riboside (MMPR), and 6-aminonicotinamide (6AN) induced partial regressions of CD8F1 murine mammary tumors and provided for tumor growth inhibition without regression of Colon 38 tumors. HPLC-nucleotide pool analysis of CD8 mammary tumors obtained at various times after treatment with PMA revealed that MMPR-5′-phosphate, which inhibits de novo purine nucleotide biosynthesis, was constant at levels of approximately 2.5 nmol/mg protein for 72 hr after treatment. In contrast, the MMPR-5′-phosphate levels of C38 tumors decreased from 24-hr levels at 1.5 nmol/mg protein with a half-time of about 24 hr. Treatment of CD8 tumor-bearing mice with iodotubercidin, a potent inhibitor of adenosine/MMPR kinase, at various times after PMA, reversed both the accumulation of high levels of MMPR-5′-phosphate and the number of partial tumor regressions These data demonstrate that a cycle of MMPR rephosphorylation is active in the CD8 mammary tumor and suggest that this recycling of MMPR is important for the optimal effect of PMA treatment.

DNA-damaging agents, e.g. Adriamycin® (ADR), are reported to cause tumor regression by induction of apoptosis. A reduction in the intracellular content of ATP is part of the biochemical cascade of events that ultimately results in programmed death of the cell, or apoptosis. A chemotherapeutic three-drug combination (PMA) consisting of N-(phosphonacetyl)-L-aspartate (PALA) + 6-methylmercaptopurine riboside (MMPR) + 6-aminonicotinamide (6AN) significantly lowers levels of ATP in CD8F1 murine breast tumors in vivo and produces tumor regression by apoptosis. Addition of the DNA-damaging antitumor agent ADR to PMA was found to further significantly deplete ATP in CD8F1 murine breast tumors in vivo with a concomitant significant increase in the number of tumor regressions. The correlative biochemical and therapeutic results are consistent with, and support, the hypothesis that ATP depletion is a significant factor and, therefore, is a worthy therapeutic target in the production of apoptosis.

Initial clinical trials have demonstrated that uridine (Urd) rescue given i.v. over at least 3 days can ameliorate 5-fluorouracil (FUra) toxicity; to avoid Urd-induced phlebitis in the peripheral veins of patients, a central vein is used. The latter necessity, along with the need for 3 days of i.v. administration, makes Urd rescue by parenteral means a cumbersome and complicated clinical procedure. It would appear preferable to use oral Urd; however, the oral Urd dose in the clinic is limited, as high doses cause diarrhea. Therefore, using a tumor-bearing murine model we investigated as to whether low doses of oral Urd coupled with a Urd phosphorylase inhibitor benzylacyclouridine (BAU), would effect safe rescue of FUra toxicity with preservation of antitumor activity. A high-dose FUra-containing drug combination that included parenteral Urd rescue was used as a control; other groups of tumor-bearing mice received the same drug combination, except that p.o. Urd was substituted for i.p. Urd. In the absence of BAU, p.o. Urd could effect rescue while maintaining an antitumor effect comparable to that obtained with i.p. Urd. When given concomitantly with BAU, a 50% reduction in the oral Urd dose (i.e., from 4,000 to 2,000 mg/kg) enabled the achievement of a comparable therapeutic index. Intraperitoneal Urd produces very high (6–8 mM) plasma and tissue Urd levels, which remain above 100 μM for at least 6 h. In contrast, neither oral Urd nor oral BAU alone raised plasma Urd concentrations above about 50 μM. However, the combination of oral Urd plus oral BAU gave a peak plasma Urd level of about 300 μM, and the level was maintained above 100 μM for 6 h. Following oral Urd administration, gut tissue levels of Urd were in the mM range and those of BAU were in the range of 10–20 μg/g tissue, a level sufficient to result in substantial inhibition of Urd phosphorylase. Oral Urd plus oral BAU appears to be a promising clinical alternative to parenteral administration of Urd for selective rescue of FUra toxicity.

Chronic weekly administration of FUra to CD8F1 female mice bearing spontaneous mammary tumors produced body weight loss during the first 2 weeks of treatment, which became less severe during subsequent weeks of therapy. To our knowledge, the development of such a decrease in FUra toxicity in vivo during chronic treatment with the drug has not been described previously, and a study of this phenomenon was therefore underfaken in tumor-free CD8F1 female mice. Weekly administration of FUra at 85 mg/kg resulted in toxicity expressed in body weight loss and in depressed peripheral WBC levels; however, the magnitude of these toxic effects decreased significantly by the 5th week of treatment. Pretreatment of normal mice with FUra for 7 weeks resulted in a dose-related shift in the LD50 of FUra administered as a subsequent challenge. Compared with an LD50 of 240 mg/kg for FUra in normal mice, the LD50 in mice pretreated with FUra at 50 or 85 mg/kg per week was found to be significantly elevated to 370 and 460 mg/kg, respectively. Pretreatment with FUra at 85 mg/kg for 7 weeks did not alter the activity of the enzymes responsible for the activation of FUra, namely uridine kinase or orotate phosphoribosyltransferase, in the intestinal epithelium or bone marrow, but it did decrease the 24-h urinary excretion of intact [3H]FUra by almost 40% (P<0.01). In addition, the FUra pretreatment schedule resulted in a 31% (P=0.14) increase in the activity of dihydrouracil dehydrogenase in the liver. These results suggest that increased degradation of FUra can be induced by chronic treatment with the drug. Finally, knowledge of the development of increased drug catabolism was used to increase the therapeutic effectiveness, of FUra by its incorporation into an increasing-dose regimen. Mice bearing 24-h transplants of the murine breast tumor were treated with a constant dose of FUra for 12 weeks or with a dose that was increased, after 7 weeks, to a dose normally causing a high degree of drug-related mortality. The group receiving the incremented FUra dose had a significantly slower tumor growth rate without an increase in drug-related toxicity. These results are discussed in light of their obvious clinical implications.

The effect of uridine on the incorporation of 5-fluorouracil into RNA and the inhibition of DNA synthesis by the FdUMP block of thymidylate synthetase was studied in the CD8F1 murine mammary carcinoma system. The administration of exogenous uridine resulted in about a one third reduction of 5-fluorouracil in RNA of tumor and normal tissues. However, unlike thymidine, uridine was unable to reverse the early, partial inhibition of DNA synthesis. The amount of fluorouridine nucleotides and (5-fluorouracil)RNA formed in various tissues correlates with the level of orotate phosphoribosyl transferase activity suggesting that the major pathway for activation of 5-fluorouracil to nucleotide form in these tissues is via phosphoribosyl transferase. Enzyme preparations from three different murine tumors convert about 15 times as much 5-fluorouracil to FUMP as they do uracil to UMP. In contrast, the ratio of FUMP to UMP formed in enzyme preparations from gut and bone marrow is lower, 2–6 fold. However, in none of these tissues was the in vitro conversion of 5-fluorouracil to FUMP or incorporation into RNA substantially inhibited by uracil. Examination of tumor, gut and bone marrow uridine nucleotide pools showed that the thymidine-uridine-5-fluorouracil schedule does increase uridine nucleotide pools. Thus, the reduction in 5-fluorouracil in RNA is probably not due to inhibition of the conversion of 5-fluorouracil to FUMP by uracil (derived from phosphorylase cleavage of uridine) but, rather, is probably due to the elevated levels of UTP. We conclude that the protection from 5-fluorouracil toxicity afforded by the addition of uridine is due to the reduction in 5-fluorouracil in RNA rather than by reversal of the FdUMP block on thymidylate synthetase.

This review summarizes a body of information suggesting that proper metabolic modulation with certain metabolites can sensitize tumor cells to anti-metabolites, and others can de-sensitize (i.e. protect) normal cells from the toxicity of anti-metabolites. This new approach offers the possibility of increasing the selectivity of drug therapy, with the promise of a real advance in cancer chemotherapy. The metabolite thymidine (TdR), long used as a cell synchronizing agent, is known to exert this effect in vitro by metabolic modulation of a number of enzymes in the salvage pathway to DNA synthesis. Against this biochemical background, in vivo effects of TdR employed as an agent for cancer therapy are reviewed as follows: 1) TdR alone, and in combination with, 2) Methotrexate (MTX), or 3) 5-Fluorouracil (FU), or 4) Cytosine arabinoside (ara-C). TdR is shown in all instances either to protect against host toxicity (eg. MTX), or to potentiate the anti-tumor effect (eg. FU and ara-C). Findings are also presented that a sequential schedule of MTX prior to TdR prior to FU is important for the optimal therapeutic activity of these drugs. The biochemical basis for the MTX leads to FU augmentation is reportedly due to increased activation of FU by MTX (acting indirectely). On the basis of this biochemical insight, a completely different chemotherapeutic agent methyl-mercaptopurine raboside (MMPR) was substituted for MTX, resulting in a dramatic potentiation of anticancer activity. Metabolic modulation with still other metabolites (UR) and a hormone (testosterone) was demonstrated to protect from host toxicity due to certain anti-cancer agents without offsetting anti-tumor activity. The ability to prevent leukopenia by these means was particularly impressive. Clinical trials have been initiated with TdR alone, TdR + MTX, and TdR + FU; the available clinical data are summarized.

Because it had been reported that splenectomy produces a tumor-inhibitory effect in several transplantable tumor systems when the surgery is performed before tumor challenge, we attempted to examine this putative immunological manipulation in a therapeutic situation.