Phase I pharmacokinetic and pharmacodynamic study of the prenyl transferase inhibitor AZD3409 in patients with advanced cancer.

AZD3409 is an orally active double prodrug that was developed as a novel dual prenyltransferase inhibitor. The formation of the active metabolite AZD3409 acid is mediated by esterases in plasma and cells. The aim of this phase I study was to determine the maximum tolerated dose, toxicities, pharmacokinetics and pharmacodynamics of AZD3409. AZD3409 was administered orally to patients with advanced solid malignancies using an interpatient dose-escalation scheme starting at 500 mg AZD3409 once daily. Twenty-nine patients were treated at seven dose levels. The MTD of part A was defined as 750 mg b.i.d. in the fasted state. Adverse events were mainly gastrointestinal and the severity was on average mild to moderate and reversible. The dose-limiting toxicities were vomiting, diarrhoea and uncontrolled nausea. Pharmacokinetic studies of the prodrug and the active metabolite indicated dose proportionality. Pharmacodynamic studies showed that farnesyltransferase (FTase) was inhibited at all dose levels. In conclusion, chronic oral dosing with AZD3409 is feasible and results in significant inhibition of FTase activity. Pharmacodynamic studies revealed that the maximal FTase inhibition, estimated at 49+/-11%, appeared to be reached at AZD3409 acid plasma concentrations at which the occurrence of drug-related toxicity was low. This study supports the rationale to implement biological effect studies in clinical trials with biologically active anticancer drugs to define optimal dosing regimens.

Sequence-dependent synergism between the new generation platinum agent ZD0473 and paclitaxel in cisplatin-sensitive and -resistant human ovarian carcinoma cell lines.

ZD0473 is a new generation hindered platinum agent currently undergoing worldwide Phase II clinical studies. The in vitro cytotoxicity of ZD0473 either alone or in combination with the anticancer drugs paclitaxel, gemcitabine, vinorelbine, topotecan and doxorubicin was determined using four human ovarian carcinoma cell lines and by the sulphorhodamine B assay (SRB). The lines included one model of acquired cisplatin resistance and one isogenic pair differing only in their p53 status. Notably, the simultaneous exposure to ZD0473 and paclitaxel for 96 h resulted in synergy (as defined by a median effect analysis) in all four cell lines (i.e. independent of cisplatin resistance and p53 status). In addition, synergy was observed in 3/4 lines and 2/4 lines following concomitant exposure to topotecan or gemcitabine, respectively. Sequencing studies with ZD0473 and paclitaxel revealed that, for three of the four cell lines, the combination of ZD0473 administered for 24 h prior to paclitaxel for 24 h conferred a greater growth inhibitory effect than the reverse sequential combination. This scheduling effect was particularly marked for the acquired cisplatin-resistant A2780CisR cell line; synergy being observed with ZD0473/paclitaxel, but antagonism with paclitaxel/ZD0473. This effect did not appear to be correlated with changes in drug-induced cell cycle checkpoints. These data suggest that ZD0473 may be usefully combined with various cytotoxics in the clinic, including paclitaxel, topotecan and gemcitabine.

Cellular pharmacology and in vivo activity of a new anticancer agent, ZD9331: a water-soluble, nonpolyglutamatable, quinazoline-based inhibitor of thymidylate synthase.

ZD9331 is a drug that was developed from a potent class of water-soluble, C7-methyl-substituted, quinazoline-based inhibitors of thymidylate synthase (TS) that are transported into cells via a saturable, carrier-mediated system (reduced folate carrier, or RFC) but are not substrates for folylpolyglutamate synthetase. ZD9331 is the gamma-tetrazole analogue of 2-desamino-2, 7-dimethyl-N10-propargyl-2'fluoro-5,8-dideaza folate (ZM214888), with a TS Ki of approximately 0.4 nM. ZD9331 exhibits potent growth inhibitory and cytotoxic activity; e.g., IC50 for the inhibition of human W1L2 lymphoblastoid cell line was 7 nM. The addition of thymidine to the culture medium increased the IC50 in W1L2 cells >10, 000-fold, demonstrating the high specificity of the drug for TS. ZD9331 is transported into cells predominantly via the RFC. Accordingly, it competes with methotrexate (MTX) and folinic acid for cellular uptake and has reduced activity against two cell lines with low expression of the RFC (L1210:1565 and CEM/MTX). In addition, a cell line with acquired resistance to ZD9331 displays reduced uptake of both ZD9331 and MTX. A mouse cell line (L1210:RD1694), with acquired resistance to ZD1694 due to reduced folylpolyglutamate synthetase activity, was not significantly cross-resistant to ZD9331. The flux through TS, as measured by 3H release from 5-[3H]deoxyuridine, was rapidly inhibited when cells were incubated with ZD9331. However, because ZD9331 cannot form polyglutamates, TS activity recovered rapidly once cells were placed in drug-free medium. The minimum curative dose of ZD9331 in the i.m. L5178Y TK-/- tumor model was approximately 3 mg/kg when given by 24-h continuous infusion, and it was 25-50 mg/kg when given by a single i.p. or i.v. injection. ZD9331 had antitumor activity against the L5178Y TK+/- tumor when administered by 7-day continuous infusion; growth delays of more than 5 days (and some cures) were seen at doses of 25-50 mg/kg/day. At higher doses, significant weight loss (gastrointestinal toxicity) and myelosuppression (neutropenia and thrombocytopenia) were observed, suggesting that these may be dose-limiting toxicities in the Phase I clinical studies.

Preclinical pharmacology of CB30900, a novel dipeptide inhibitor of thymidylate synthase, in mice.

CB30900 is a novel, potent thymidylate synthase inhibitor which can not be polyglutamated and may be active in cancers expressing low or defective folylpolyglutamate synthetase. Pharmacokinetics were studied in mouse tumors and tissues after bolus or infusion protocols. Elimination was triphasic after 100 mg kg-1 i.v. (T 1/2 alpha, 2.8 min; T 1/2 beta, 19.1 min and T 1/2 gamma, 4.1 hr). Peak concentrations were 716 microM; clearance, 1.19 ml g-1 hr-1; and area under the curve (AUC 0-2 hr), 131 microM hr. Biphasic elimination occurred after i.p. administration and was comparable to the i.v. route giving complete i.p. bioavailability. Kidney concentrations were similar to plasma (AUC 0-2 hr, 84.3 microM hr). CB30900 concentrations in the gut increased steadily with time (AUC 0-2 hr, 645 microM hr) and liver drug concentrations were 7-fold greater than plasma (AUC 0-2 hr, 847 microM hr). Peak tumor concentrations occurred at 30 min and were 27% of plasma concentrations, but tumor drug clearance was markedly slower than for plasma (T 1/2, 51 +/- 8.2 min, mean +/- S.E.). CB30900 was remarkably stable in vivo with 93% of an administered dose recovered unchanged after 48 hr. Plasma drug binding was concentration-dependent, ranging from 93.3 to 76% over 1 to 500 microM. During 24 hr infusion (50 mg kg-1 s.c.), steady-state plasma concentrations were 3 microM, giving an AUC 0-24 hr of 71 microM hr. Kidney drug levels were similar to plasma but liver concentrations were elevated 7-fold. By contrast, tumor drug concentrations were about 0.5 microM (AUC 0-24 hr, 14.6 microM hr). However, these low plasma drug concentrations are growth inhibitory in vitro (24-hr exposure).

ZD1694 (Tomudex): a new thymidylate synthase inhibitor with activity in colorectal cancer.

ZD1694 (Tomudex) is a new antifolate which is a specific inhibitor of thymidylate synthase (TS). Evidence suggests that ZD1694 has a spectrum of activity that only partially overlaps with 5-fluorouracil (modulated with leucovorin) against colon tumours in vitro. Potent cytotoxic activity is dependent upon active uptake into cells via the reduced folate/methotrexate cell membrane carrier (RFC) and subsequent metabolism to polyglutamated forms (tri, tetra and pentaglutamates). These polyglutamates are approximately 60-fold more active as TS inhibitors and are not effluxed readily from cells. Extensive polyglutamation also occurs in various mouse tissues (e.g. small intestinal epithelium, liver and kidney), resulting in high tissue/plasma drug ratios which persist for a prolonged period. ZD1694 has antitumour activity in mice, although the high plasma thymidine in this species complicates: (1) the interpretation of therapeutic index; (2) tumour types in which activity is likely to be observed; and (3) translation of doses and schedules for clinical evaluation. ZD1694 entered clinical study and has completed Phase I and II evaluation, with activity observed in several tumour types. Appreciable activity in the Phase II colorectal study (29% objective response rate on interim analysis) led to the current Phase III study, randomised against 5-fluorouracil/leucovorin.

Gamma-linked dipeptide analogues of 2-desamino-2-methyl-N10-propargyl-5,8- dideazafolate as antitumour agents.

Our search for water-soluble quinazoline TS inhibitors that are transported into cells via the RFC, but are not substrates for FPGS, led us to the synthesis of dipeptide analogues of ICI 198583 diglutamate. Although a number of dipeptide analogues were active against isolated TS and L1210 cells in vitro, lack of in vivo stability was a problem. This was circumvented by the synthesis of modified dipeptides where either the alpha-carboxyl of the second amino acid was removed (alpha'-COOH) e.g. -L-glu-GABA or where the second amino acid was the unnatural D-enantiomer e.g.-L-glu-D-glu. Further studies were performed with the -L-glu-D-glu and its 7-CH3, 2'F modified analogue, demonstrating that they use the RFC for cell entry but are not active through polyglutamate formation. The latter compound was tested against experimental tumour models and found to have good activity.

ICI D1694, an inhibitor of thymidylate synthase for clinical study.

The TS inhibitor, ICI D1694, is a highly potent inhibitor of tumour growth in vitro and in vivo. Uptake via the RFC and rapid metabolism to polyglutamate forms appear to be responsible for potency. Antiproliferative toxicity in mice was evident although the dose-limiting renal toxicity experienced with CB3717 was not apparent.

Split-dose and low dose-rate recovery in four experimental tumour systems.

The radiation dose-rate effect has been investigated in three murine tumour cell systems (MT carcinoma, Lewis lung tumour, B16 melanoma) and in the HX34 human melanoma xenograft taken directly from mice and irradiated in vitro. The four tumour types were remarkably similar in their radiation response characteristics, especially at low dose rate; the Lewis lung tumour tended to be the most radiosensitive at high dose rate. The data have been analysed using the Lethal-Potentially Lethal (LPL) model of Curtis and the Incomplete Repair (IR) model of Thames. The data are equally well fitted by both models. The most remarkable feature of these analyses is that both models lead to estimates for the half-time for recovery that are in the region of 0.1 h, considerably shorter than other published values. Split-dose experiments were also performed, taking care to keep the cells at 37 degrees C between exposures. In all cases the split-dose half times were longer than the values derived from dose-rate analysis and in the case of the Lewis lung and HX34 tumour lines the difference was by almost a factor of ten. The discrepancy between these estimates could be the result of biphasic cellular recovery.

Development of MeCCNU-resistance in clonally derived lines of Lewis lung carcinoma.

The development of resistance to MeCCNU in four clonal lines of Lewis lung carcinoma derived from lung colonies has been examined. Treatment of the clones with 15 mg/kg MeCCNU once in each in vivo passage resulted in a rapid reduction in tumour responsiveness to this drug. This shows that MeCCNU resistance can arise at rates which may be applicable in spontaneous tumours and their metastases. Karyotypic heterogeneity was evident very early after isolation of the clones and this has potentially important implications for the experimental use of clones and for the features of metastatic disease. Karyotype analysis of one of the clones at each passage suggested that development of resistance was associated with the selection of a single sub-population of tumour cells but the kinetics of resistance development cannot be totally explained in terms of this simple selection.

Clonal variation in the sensitivity of a murine mammary carcinoma to melphalan.

The sensitivity to melphalan of clones derived from individual lung colonies produced by i.v. injection of cells of the MT murine mammary carcinoma (caMT) and its melphalan-resistant sub-line (MTME16) has been examined. A degree of clonal heterogeneity was observed which was greater than could be explained by experimental variation. The distribution of melphalan sensitivities in both wild-type caMT and MTME16 raises questions as to the validity of a two-compartment model of drug-resistance development in tumours. A more complex model, possibly involving a continuous spectrum of drug sensitivity, is required. Differences in the sensitivity of the clonal lines of wild-type caMT in various passages were observed and this would appear to be due to phenotypic instability in these lines. This suggests that to use survival data from clones which have been passaged many times for predicting the response of the parent tumour may be misleading.

Dose-rate effects and the repair of radiation damage.

The extent of dose-sparing that occurs in a variety of cell lines and in vivo cell systems as a result of a reduction in dose-rate is reviewed. The emphasis is on the range from around 200 cGy/min down to 5 cGy/min, in which the predominant reason for dose-sparing is the repair of radiation damage. Dose-rate dependence is considered in relation to the Lethal-Potentially Lethal model of cell inactivation, which satisfactorily fits 4 sets of data that we have tested; estimates of half-time for repair varied from 0.07 to 1.4 h. The model shows that in spite of these short half-times, repair will often continue to influence response down to dose-rates below 5 cGy/min. The steepness of the dose-rate dependence varies widely among in vitro cell lines and among mouse normal tissues, indeed the ranges in vitro and in vivo are similar. Haemopoietic tissues are much less spared by a lowering of dose-rate than are other normal tissues. Uncertainties about the rate of reoxygenation preclude similar considerations in experimental tumours in vivo. There is a need for detailed studies of dose-rate dependence in human tumour cell lines, and the present review outlines the basis (including the optimum dose-rate range) for such studies.

Temporal interactions in the Lewis lung tumour between cytotoxic drugs and acute or fractionated radiotherapy.

The suggestion was examined, that acute irradiation might be combined with some cytotoxic drugs to produce a supra-additive antitumour effect by timing the second treatment to coincide with the period of repopulation by tumour cells that survived the first treatment. The possible interaction of fractionated irradiation with drugs was also studied. Lewis lung tumours were treated in vivo with acute radiation (20 Gy) and 1-(2-chloroethyl)-3-(4-methyl cyclohexyl)-1-nitrosourea (MeCCNU, 10 mg X kg-1), cyclophosphamide (CY, 120 mg X kg-1) or cis-dichlorodiammine platinum (cis-Pt, 10 mg X kg-1) with time intervals ranging from simultaneous to 9 days in either sequence. Tumour response was measured by regrowth delay and combination responses were evaluated relative to calculated "additivity envelopes". With CY and cis-Pt, tumour volume response was approximately additive, whether the drugs were administered simultaneously, or up to 7 days before or after radiation. However, with simultaneous administration, MeCCNU and radiation were supra-additive in terms of volume response, although only additive using a clonogenic cell survival assay. A transplantation experiment suggested that the supra-additive volume response may be due to retarded growth rate in regrowing tumours. In the fractionation studies, drugs were given either at the beginning or at the end of a regime of 5 daily 6 Gy doses. With CY and MeCCNU the drugs were more effective when given with the first radiation dose, but with cis-Pt either regime was equally effective. There was no evidence that repopulation could be exploited to improve therapeutic effect with any of the combination treatments used in this study.

Emergence of nitrosourea resistant sublines of Lewis lung tumour following MeCCNU treatment in vivo.

Several different drug retreatment protocols were employed to examine the emergence of resistance to MeCCNU in Lewis lung tumours. Previous studies suggested that although the majority of cells in untreated Lewis lung tumours were sensitive to MeCCNU, there was a very small proportion of resistant cells (approximately 0.001%) that limited "tumour cure' with that drug. If such cells were inherently drug resistant then it should be possible to derive highly resistant tumours by repeated drug treatment. In the first experiment tumours were treated with a single high dose of MeCCNU (35 or 40 mgkg-1) and on regrowth, transplanted into fresh mice and tested for drug sensitivity. Using both excision cell survival and growth delay endpoints, only approximately 25% of tumours were significantly resistant to the test dose, suggesting that many tumours resist the effects of the drug for reasons other than the presence of inherently drug resistant cells. One of the tumours (R4), that regrew after the initial treatment and appeared to be resistant to the test treatment, was retreated with a further 30 mgkg-1 MeCCNU and became more resistant. This line, designated R4/1, was cross-resistant to the other nitrosoureas, BCNU and CCNU, but not to cyclophosphamide, melphalan, cis-platinum or ionising radiation. The effect of treatment dose on the kinetics of MeCCNU resistance development was also studied in a retreatment regimen where the tumours were allowed to regrow and then transplanted into fresh hosts for the next treatment. Resistance developed more quickly at an intermediate dose of 15 mgkg-1 than at 7.5 mgkg-1 where the selective pressure was lower, or at 30 mgkg-1 where there was probably extinction of partially resistant cells. Resistance to MeCCNU developed even more quickly when tumours were retreated several times in the same host, although in a similar experiment with cyclophosphamide no resistance occurred.

Development of drug resistance in a murine mammary tumour.

The development of resistance to melphalan, cis-platinum and cyclophosphamide has been examined in the MT murine mammary carcinoma. A gradual decrease in therapeutic response was detected using growth delay and clonogenic cell survival during repeated drug treatment. A slow rate of resistance development, a gradual change in the slope of the dose-survival curves and the inability of 180 mg kg-1 cyclophosphamide to bring about a reduction in tumour response at a faster rate than 60 mg kg-1 cyclophosphamide suggest that resistance development was not due to the selection of a pre-existing highly drug resistant sub-population of tumour cells. Partial drug-resistance is proposed as one possible reason for the apparent inconsistency between these data and existing models of drug-resistance development. The drug-resistant lines were characterized for karyotype, DNA content and cell volume, but only the cyclophosphamide-resistant line showed any significant difference from the wild-type tumour. Cross-resistance studies revealed some inconsistencies with previous reports. Also, resistance to cyclophosphamide developed more quickly in the line which was resistant to melphalan, than in the wild-type tumour, despite the initial appearance of little cross-resistance. This increased rate of resistance development may be important in salvage chemotherapy.

Adriamycin sensitivity following "ultra" low dose rate irradiation of Lewis lung carcinoma in vivo.

As an extension of recent study on the response of the Lewis lung tumor to low dose rate continuous irradiation (CI) at 15 cGy/hr, we have gone on to investigate the effects of such irradiation on the sensitivity of tumor cells to treatment with Adriamycin (Adr). Cells from untreated tumors gave an exponential dose response curve to Adr in vitro, the D10 of which increased (sensitivity decreased) with the size of tumor (0.05 g to 0.6 g) from which the cells were obtained. After previous in vivo CI to a total dose of 28 Gy (irradiation time--186 hr), this size dependence was abolished and the cells showed an exponential response to Adr in vitro (D10 = 0.4 microgram/ml). The enhancement was also observed after equivalent doses of fractionated irradiation, but not after acute irradiation. It was difficult to characterize the proliferative status of the cells surviving irradiation, but repopulation studies showed that only after CI was there any delay before repopulation commenced. LL was relatively insensitive to Adr in vivo, however, we did observe an increased effect after previous CI.

Kinetics of cellular inactivation by fractionated and hyperfractionated irradiation in Lewis lung carcinoma.

The kinetics of cellular inactivation by fractionated irradiation in Lewis lung carcinoma was studied in the dose range of 2.3 to 6.5 Gy per fraction. Regimens of one and two fractions per day for 10 days were compared. The number of clonogenic tumor cells was determined using an in vitro soft-agar colony assay. Under the experimental conditions used, the fraction of tumor cells inactivated per day was only dependent on the total dose per day, i.e., the cellular response was the same whether the daily dose was given in one or two fractions. Thus, the two fraction per day regimen was more effective than expected from calculations based on acute in vivo radiation dose-survival curves. This result could be explained if the clonogenic tumor cells were less hypoxic during the two fractions per day regimen than the one fraction per day schedule.