Nucleosides. CXXXV. Synthesis of some 9-(2-deoxy-2-fluoro-beta-D-arabinofuranosyl)-9H-purines and their biological activities.

Seven purine nucleosides containing the 2'-deoxy-2'-fluoro-beta-D-arabinofuranosyl moiety were synthesized and tested for their antitumor activity. Direct condensation of 3-O-acetyl-5-O-benzoyl-2-deoxy-2-fluoro-D-arabinofuranosyl bromide (1) with N6-benzoyladenine in CH2Cl2 followed by saponification of the product afforded the adenine nucleoside (I, 2'-F-ara-A). Deamination of I with NaNO2 in HOAc gave the hypoxanthine analogue (II, 2'-F-ara-H). The 6-thiopurine nucleoside (III, 2'-F-ara-6MP) was prepared by condensation of 1 with 6-chloropurine by the mercury procedure followed by thiourea treatment and saponification of the product. Methylation of III gave the 6-SCH3 analogue (IV). Raney Ni desulfurization of III afforded the unsubstituted purine nucleoside (V, 2'-F-ara-P). Condensation of 1 with 2-acetamido-6-chloropurine by the silyl procedure afforded the protected 2-acetamido-6-chloropurine nucleoside which served as the precursor for both the guanine and 6-thioguanine nucleosides (VI, 2'-F-ara-G and VII, 2'-F-ara-TG, respectively). Thus, alkaline hydrolysis of the precursor gave VI. Thiourea treatment prior to alkaline hydrolysis gave VII. The new nucleoside, 2'-F-ara-G (VI) is found to be selectively toxic to human T-cell leukemia CCRF-CEM.

Phase I trial of alpha-difluoromethyl ornithine (DFMO) and methylglyoxal bis (guanylhydrazone) (MGBG) in patients with advanced prostatic cancer.

Both alpha-difluoromethyl ornithine (DFMO) and methylglyoxal bis (guanylhydrazone) (MGBG) inhibit sequential enzymatic reactions in the pathway of polyamine biosynthesis. Since polyamines may be important factors in proliferation of cancer cells and DFMO combined with MGBG has shown synergistic cytotoxicity in an experimental prostatic tumor, we evaluated these agents in phase I clinical trial involving 5 patients with advanced, hormone-resistant prostatic cancer. Toxic reaction to combined DFMO and MGBG was dose-related and included nausea, fatigue, and diarrhea especially with the higher doses of MGBG. No therapeutic responses of significance were seen, but toxicity precluded adequate evaluation. Future Phase II studies of combined DFMO and MGBG should employ low, nontoxic doses of MGBG combined with evaluation of polyamine levels and inhibition of polyamine enzymatic activity to minimize toxicity.

Synthesis and biological activities of 5-deaza analogues of aminopterin and folic acid.

N-[p-[[(2,4-Diaminopyrido[2,3-d]pyrimidin-6-yl)methyl] amino]benzoyl]-L-glutamic acid (1a, 5-deazaaminopterin) and the 5-methyl analogue (1b) were synthesized in 14 steps from 5-cyanouracil (4a) and 5-cyano-6-methyluracil (4b), respectively, by exploitation of the novel pyrimidine to pyrido[2,3-d]pyrimidine ring transformation reaction. The 5-cyanouracils 4 were treated with chloromethyl methyl ether to the 1,3-bis(methoxymethyl)uracils (5, which were treated with malononitrile in NaOEt/EtOH to give the pyrido[2,3-d]pyrimidines 6. Diazotization of 6 in concentrated HCl afforded the 7-chloro derivatives 8 in high yield. After reduction of 8, the 7-unsubstituted products 9 were reduced in the presence of Ac2O and the products, 6-(acetamidomethyl)pyridopyrimidines 10, were converted into the 6-acetoxymethyl derivatives 12 via nitrosation. After removal of the N-methoxymethyl groups from 12, the 6-(acetoxymethyl)pyrido[2,3-d]pyrimidine-2,4(1H,3H)-diones 14 were converted into 2,4-diamino-6-(hydroxymethyl)pyrido[2,3-d]pyrimidine (15a) and its 5-methyl analogue 15b by the silylation-amination procedure. Compounds 15 were brominated to the 6-bromomethyl derivatives 16, which were treated with diethyl (p-aminobenzoyl)-L-glutamate, and the products 17 were saponified to afford 5-deazaaminopterin (1a) and its 5-methyl analogue 1b. Compound 1b was also prepared by an alternative procedure in 10 steps from cyanothioacetamide and ethyl beta-(ethoxymethylene)acetoacetate via 2,4-diamino-6-(hydroxymethyl)-5-methylpyrido[2,3-d]pyrimidine (15b). 5-Deaza-5-methylfolic acid (2) was also prepared in four steps from 15b. The aminopterine analogues 1 showed significant anticancer activity in vitro and in vivo, whereas the folic acid analogue 2 did not exhibit any significant toxicity.

Phase I trial of 1-(2'-deoxy-2'-fluoro-1-beta-D-arabinofuranosyl)-5-methyluracil (FMAU).

1-(2'-Deoxy-2'-fluoro-1-beta-D-arabinofuranosyl)-5-methyluracil (FMAU), a new pyrimidine nucleoside, is of potential clinical interest both as an anticancer and as an antiviral drug. FMAU is active in vitro and in vivo against P815 and L1210 cell lines resistant to cytarabine. Moreover, in mice inoculated ic with herpes simplex virus Type II, FMAU is 100-fold more potent than vidarabine or acyclovir. We have conducted a phase I trial of FMAU in 17 patients with advanced cancer. The dose levels studied were 2, 4, 8, 16, 32, 64, and 128 mg/m2/day iv for 5 days. The dose-limiting toxic effect was drug-induced central nervous system dysfunction. Although 32 mg/m2/day for 5 days produced only transient, mild symptoms, severe encephalopathy with extrapyramidal dysfunction occurred at 64 and 128 mg/m2/day for 5 days and contributed to two deaths. No toxicity was observed at less than 32 mg/m2. A dose of 32 mg/m2/day for 5 days is suggested for phase II study. Because of its potent and selective antiviral activity, future trials of low doses of FMAU in immunosuppressed patients with herpes virus infections are under consideration.

Effects of alpha-difluoromethylornithine and methylglyoxal bis(guanylhydrazone) on the growth of experimental renal adenocarcinoma in mice.

alpha-Difluoromethylornithine (DFMO) and methylglyoxal bis-(guanylhydrazone) (MGBG) were tested against a murine renal adenocarcinoma, because polyamines are necessary for neoplastic cell growth and because human renal adenocarcinomas contain higher levels of spermidine than do normal renal cells; MGBG inhibits spermidine synthesis and has some activity against human renal tumors; DFMO irreversibly inhibits ornithine decarboxylase, the first rate-limiting enzyme controlling polyamine biosynthesis; and DFMO promotes intracellular accumulation of MGBG in experimental tumor models and human leukemia. DFMO (2%) in drinking water, MGBG (15 mg/kg i.p.), or a combination of DFMO and MGBG was administered daily to BALB/c mice (n = 80) with intrarenal transplants of renal adenocarcinoma cells. At 28 days, renal carcinomas weighed 64 and 73% less, respectively, in DFMO- and DFMO-MGBG-treated mice than in control animals (p less than 0.01). MGBG alone had no antigrowth effect. DFMO-MGBG reduced the total metastatic index (total number of metastases/total number of animals) to 1.2 versus 3.6 in control animals (p less than 0.01) and increased survival by 12.3 +/- 1.5 (S.E.) days, from 30.8 to 42.5 days (p less than 0.05). Compared with control, DFMO-, or MGBG-treated animals, DFMO-MGBG exposure reduced tumor growth and the number of metastases, prevented metastases in some animals (47%), and increased survival of mice bearing renal adenocarcinomas. DFMO also appeared to selectively increase the uptake of [14C]MGBG by tumor tissue, which may help to explain the enhanced synergistic antigrowth effect of DFMO and MGBG against this murine renal adenocarcinoma.

Methylglyoxal-bis(guanylhydrazone) (Methyl-GAG): current status and future prospects.

Initial clinical trials of methyl-GAG (MGBG) showed that repetitive daily administration produced severe, occasionally fatal, toxic reactions. After two decades of neglect, recent studies have shown that doses of 500-600 mg/sq m administered every 7-14 days are very well tolerated. Moreover, current results indicate that MGBG has useful antitumor activity in patients with advanced malignant lymphoma and carcinomas of the head and neck, esophagus, and lung (non-small cell). The drug's mechanism of cytotoxic action and its toxic effects are not shared by most other cancer chemotherapeutic drugs. Furthermore, Phase II and Phase III evaluation are required to determine the therapeutic potential of this unique agent.

Sequential inhibition of polyamine synthesis. A phase I trial of DFMO (alpha-difluoromethylornithine) and methyl-GAG [methylglyoxal-bis(guanylhydrazone)].

Both DFMO and methyl-GAG inhibit sequential enzymatic reactions in the pathway of polyamine biosynthesis. Since polyamines may be important factors in proliferation of cancer cells, we initiated a phase-I study of these agents in patients with advanced cancer. DFMO was given by mouth at a constant daily dose of 4 g/m2 starting on day 1 of the treatment protocol. The dose of methyl-GAG ranged from 200 to 700 mg/m2 administered IV every 2 weeks beginning on day 4. Twenty-two patients were entered into the protocol. Toxic reactions to this therapy were dose-related and included nausea, fatigue, diarrhea, and myelosuppression. One patient with colon cancer experienced a greater than 50% decrease in measurable disease but developed severe myelotoxicity. While DFMO was well tolerated, the combination of drugs appeared to cause substantially more hematologic and gastrointestinal toxicity than encountered during our recent experience with methyl-GAG used alone. We suggest that future studies of this drug combination carefully evaluate levels of polyamines and inhibition of enzymatic activity to minimize toxicity.

Phase I clinical trial and pharmacokinetics of 4'-carboxyphthalato(1,2-diaminocyclohexane)platinum(II).

4'-Carboxyphthalato(1,2-diaminocyclohexane)platinum(II) is a new, second generation platinum analog which had demonstrated in vitro activity in L1210 cell lines resistant to cisplatin and had less nephrotoxicity than did cisplatin in preclinical animal testing. A Phase I trial with this agent has been performed in 45 patients with advanced refractory cancers. Nine dosage levels, ranging from 40 to 800 mg/sq m, were studied. Major toxicities seen were myelosuppression, nephrotoxicity (which was generally mild), nausea and vomiting (which was quantitatively less than that seen with cis-platin), allergic reactions, and a peripheral neuropathy. The dose-limiting toxicity was thrombocytopenia. Pharmacokinetics performed at three dosage levels indicates that 4'-carboxyphthalato-(1,2-diaminocyclohexane)platinum(II) has a long t1/2 of 20 to 30 hr (total platinum) and is only partially excreted in the urine and that a high proportion of the drug is nonfilterable within 30 to 60 min of administration. Therapeutic responses were seen in nasopharyngeal carcinoma, adenocarcinoma of the cervix, and lung and gastric cancer. As a starting dose for Phase II studies, which are planned for patients with ovarian, testicular, lung, gastric, and esophageal cancers, 640 mg/sq m given every 3 to 4 weeks is recommended.

Synthesis and antitumor activity of new platinum complexes.

A new type of antitumor platinum complex has been prepared and examined for antitumor activity against L1210 leukemia both in vitro and in vivo. The coordination environment of platinum in these complexes consists of three anionic chloride ions and a positively charged amine. The positive charge is introduced by monoprotonation or monoalkylation of a diamine. Platinum(IV) derivatives have been prepared for several of the complexes, and a water-soluble sulfate derivative has been prepared for one of them. Several of these complexes exhibit significant in vitro activity, and trichloro(3-aminoquinuclidinium)platinum(II) (QTP) exhibits significant in vivo activity as well. An increase in life span of approximately 40% has been observed using QTP. QTP is toxic at doses slightly in excess of effective doses.

Activity of 2-fluoro-5-methylarabinofuranosyluracil against mouse leukemias sensitive to and resistant to 1-beta-D-arabinofuranosylcytosine.

A new pyrimidine nucleoside, 2'-fluoro-5-methyl-1-beta-D-arabinofuranosyluracil, previously has been shown to be active against the herpes group of viruses in vitro and in vivo. It is also active against mouse and human leukemic cells in culture and against mouse leukemias L1210, P388, and P815 in vivo. In contrast to other 1-beta-D-arabinofuranosylcytosine (ara-C) derivatives, 2'-fluoro-5-methyl-1-beta-D-arabinofuranosyluracil, when given either i.p. or p.o., is highly active against lines of leukemias P815 and L1210 made resistant to ara-C. Against P815/ara-C and L1210/araC, it is more effective than is 5-azacytidine, a drug which has shown definite effectiveness in patients with acute leukemia whose disease has become resistant to ara-C. For these reasons, 2'-fluoro-5-methyl-1-beta-D-arabinofuranosyluracil would seem to merit clinical trial in patients with acute nonlymphocytic leukemia whose disease has become resistant to ara-C.

Treatment of acute nonlymphocytic leukemia in adults: response to 2,2-anhydro-1-B-D-arabinofuranosyl-5-fluorocytosine and thioguanine on the L-12 protocol.

Fifty-one adult patients with acute nonlymphocytic leukemia (excluding acute promyelocytic leukemia) were treated on the L-12 protocol. The L-12 differed from the preceding L-6 in that 2,2-anhydro-1-B-D-arabinofuranosyl-5-fluorocytosine (AAFC), replaced arabinosylcytosine (ara-C) together with 6-thioguanine (TG) for remission induction. Achievement of remission was followed by an extended 14-week multi-drug consolidation program. With this more intense regimen, an overall complete remission rate of 49% and a median remission duration of 23.7 months were achieved; these results were not significantly better than the 57% complete remission rate and 8.6 months median remission duration obtained with the L-6 regimen. Four year disease-free survival was 22% on the L-12 compared with 16% on the L-6 protocol. No relationship between prognosis and FAB classification was found on either the L-6 or the L-12 protocol.

Pharmacokinetics of Malonato (1,2 diaminocyclohexane) platinum.

Malonato-(1,2 diaminocyclohexane) platinum (MP) is a new platinum analog currently undergoing phase I clinical trials. Using flameless atomic absorption spectrophotometry, the pharmacokinetics of MP were studied at five dosage levels. The drug was given as a prolonged intravenous infusion, lasting from 6 to 24 hours. Peak plasma platinum concentrations (Pt) were seen at the end of the infusion, and ranged from 1.1 microgram/ml when 3 mg/kg was given to 14-20.5 micrograms/ml at the 24-mg/kg level. Following completion of the infusion, a prolonged T1/2 beta (mean 63.5 hours) was noted. The percentage of free:total platinum was high (90-95%) at the beginning of the infusion but fell rapidly, to only 15-21% at the end of the 24-hour infusions. Urinary excretion accounted for 16-37.5% of the total administered dose. MP appears to have several pharmacokinetic features in common with cisplatin: rapid binding to protein, a prolonged terminal phase half-life involving primarily bound platinum, and incomplete excretion by the kidney.

Biochemical, pharmacological, and phase I clinical evaluation of pseudoisocytidine.

Pseudoisocytidine (psi ICyd) is a C-nucleoside with enhanced stability and resistance to enzymatic deamination when compared to 5-azacytidine and 1-beta-D-arabinofuranosylcytosine. Elimination kinetics in plasma using [14C]psi ICyd showed a beta-phase for t1/2 for 14C of 2 hr and a beta-phase t1/2 of unchanged psi ICyd of 1.5 hr. Net recovery of radioactivity in urine over 24 hr varied between 40 and 80% of the administered dose; 50 to 90% was unchanged drug and the rest was pseudouridine. Human leukemic cells in vitro deaminated psi ICyd very slowly, formed appreciable quantities of pseudoisocytidine triphosphate, and incorporated small amounts into RNA and DNA. Clinical trials were done using a daily i.v. injection for 5 consecutive days. Hematological or intestine toxicities were not seen, nor was depression of white blood cell count observed in leukemic patients. Hepatic toxicity proved to be dose limiting; this was characterized by an early phase with elevation of prothrombin time and aspartate aminotransferase. A later phase with cirrhosis was observed in two patients. Autopsy showed massive hepatic necrosis in patients dying of acute toxicity and micronodular cirrhosis in one patient dying with the chronic form.

Preclinical and phase I studies of malonatoplatinum.

After preclinical toxicologic study in baboons, we are conducting a phase I trial of malonatoplatinum, starting with 3 mg/kg and now reaching 1 mg/kg. Toxicity, mainly hematologic, was mild and our study was mainly limited by poor solubility. Regressions, which have been rare in the advanced-tumor patients, have been observed in three patients considered as clinically resistant to cisdichlorodiammino-platinum (DDP). Malonatoplatinum pharmacokinetics appeared similar to those of DDP as far as total platinum is concerned.

Rationale for development of platinum analogs.

There is a great need for clinical trial of new second-generation platinum coordination compounds that might demonstrate greater clinical activity in a broader spectrum of tumors, decreased renal toxicity and emesis, improved solubility, synergism in combination therapy, and lack of cross-resistance to cis-dichlorodiammineplatinum(II) (cis-platinum). Lack of cross-resistance to cis-platinum is shown by certain 1,2-diamino-saturated cyclic platinum derivatives which also have a high degree of activity against transplanted mouse leukemias. cis-Platinum and these cyclic compounds combine synergistically with derivatives of cytosine arabinoside, VP-16-213, and Adriamycin. These 1,2-diamino cyclic compounds appear to have less renal toxicity than cis-platinum. The toxic and therapeutic effects of both cis-platinum and the diamino cyclic compounds can be blocked by massive doses of thiourea. Varying doses and time intervals of thiourea rescue are being studied in the hope of improving the therapeutic index of the platinum derivatives.

Phase I evaluation of tetrahydrouridine combined with cytosine arabinoside.

In conventional clinical use, cytosine arabinoside (ara-C) is rapidly deaminated by pyrimidine nucleoside deaminase to the nontoxic compound uracil arabinoside. Tetrahydrouridine (THU) effectively inhibits this enzymatic degradation but is by itself nontoxic. This study demonstrates that concomitant administration of THU markedly increases the myelosuppressive potency of ara-C. When 25 or 50 mg/kg of THU iv and 0.1--0.2 mg/kg of ara-C iv are given daily x 5 days, they produce moderate-to-severe leukopenia and mild-to-moderate thrombocytopenia. A dose of 25 mg/kg of THU with 0.1 mg/kg of ara-C iv daily x 5 days appears appropriate for phase II studies; it produces myelosuppression equivalent to that produced by 3 mg/kg/day x 5 days of ara-C alone. No toxicity occurred with this combination that would not have been expected from ara-C given alone in an equitoxic dose. Although THU and ara-C produced a reduction in peripheral blood and bone marrow blast cells in eight of nine patients with acute leukemia, bone marrow remission did not occur in any of these heavily pretreated patients.