Search

Background: The presence of multiple, low-molecular-weight, insulinlike growth factor (IGF)-binding proteins in lung tumor cell-conditioned medium and lung cancer patient serum has been recently reported. Purpose: To begin to elucidate the genetic basis for these observations, the present study examines the expression by lung tumor cell lines of three IGF-binding protein genes, namely, IGFBP-1, IGFBP-2, and IGFBP-3. Since IGF-binding proteins are thought to modulate the biologic action of the IGFs, the relationship between the expression of IGF-binding protein genes and the genes encoding IGF-I and IGF-II also has been investigated. Methods: Gene expression was studied in four small-cell lung cancer (SCLC) and three non—small-cell lung cancer (NSCLC) cell lines using Northern blot analysis and reverse transcriptase polymerase chain reaction (RT-PCR) for IGFBP-1. Results: IGFBP-1 gene expression was detected by Northern blot analysis in one NSCLC cell line only. However, RT-PCR revealed that the IGFBP-1 gene was expressed in all four SCLC cell lines and in two of the three NSCLC lines. Northern blot analysis of IGFBP-2 gene expression demonstrated that all lung tumor cell lines expressed this gene. A low level of IGFBP-3 gene expression was detected in one SCLC cell line and in all three NSCLC cell lines. All lung tumor cell lines expressed the IGF-II gene as determined by Northern blot analysis. In marked contrast, none of the lines showed evidence of IGF-I gene expression using this method. However, RT-PCR revealed a low level of IGF-I gene expression in one SCLC and one NSCLC cell line only. Conclusions: These observations indicate 1) that IGF-binding proteins secreted by lung tumors are encoded by at least three different genes; 2) that there may be a close association between IGF-II and IGFBP-2 gene expression, such that, where there is production of IGF-II, IGFBP-2 is the principal BP; and 3) that the IGF-II gene is more widely expressed than the IGF-I gene in human lung tumor cell lines. [J Natl Cancer Institute 84: 628-634, 1992]

Single loop surface coil, often used in MR studies, was evaluated for its performance as an inductive hyperthermic applicator. The heat deposition pattern produced by the surface coil at 84 MHz and 34 MHz was mapped in muscle-mimicking agar phantoms. Temperatures were measured simultaneously at 64 points using multiple-junction thermocouples.

BACKGROUND The role of chemotherapy in the treatment of patients with primary central nervous system lymphoma (PCL) remains unclear, with no randomized trials available to aid in the interpretation of the current data. The Medical Research Council therefore conducted the current randomized trial to assess the impact on survival of postradiotherapy chemotherapy with cyclophosphamide, doxorubicin, vincristine, and prednisone (CHOP) in nonimmunocompromised adult patients with pathologically proven PCL. METHODS After surgery, patients were randomized at a ratio of 1:2 to radiotherapy alone (RT: 40 grays [Gy] in 20 fractions to the whole brain followed by a 14-Gy boost to the tumor plus a 2-cm tumor margin) or to the same radiotherapy followed by six cycles of CHOP chemotherapy given at 3-week intervals (RT-CHOP). The target sample size was 90 patients, which allowed 90% power to detect a doubling of the median survival time. RESULTS Between 1988 and 1995, 53 patients were randomized: Fifteen patients were randomized to RT, and 38 patients were randomized to RT-CHOP. The trial closed earlier than planned through poor accrual. The median patient age was 57 years, 57% of the patients were male, and 75% of the patients had unifocal disease. The median number of chemotherapy cycles received was 6 (mean, 4 cycles). Forty-three patients have died, and the median follow-up of survivors is 5 years (range, 1–9 years). There was no evidence of a benefit from RT-CHOP with respect to overall survival (hazard ratio [HR], 1.19; 95% confidence interval, 0.51–2.76) after adjustment for prognostic factors (patient age and neurologic performance status) in an analysis in which HR > 1 favored the control (RT) group. CONCLUSIONS CHOP has no clear role in the postradiotherapy treatment of patients with PCL. Chemotherapy is poorly tolerated and largely palliative in older, less fit patients. In younger patients, initial chemotherapy designed to penetrate the blood-brain barrier warrants further investigation. Cancer 2000;89:1359–70. © 2000 American Cancer Society.

31P-magnetic resonance spectroscopy (MRS) and a fiberoptic pH meter were used simultaneously to follow the changes in intra- (pHi) and extracellular pH (pHe), respectively, of murine RIF-1 tumors with hyperthermia. Hyperthermia was induced at 34 MHz using the same coil used for MR. The study was carried out until 3.5 hr after hyperthermia. In untreated tumors (n = 29), pHi was always higher than pHe. pHi was reduced after hyperthermia (30 min) at both 42 degrees C and 45 degrees C. pHe registered an increase after 42 degrees C and a decrease after 45 degrees C. The reduction in pHi was larger than the initial differential between pHi and pHe, and the change in pHe was relatively small. Hyperthermia changed the acidity of the intra- and extracellular compartments, such that pHe became more alkaline than pHi by 0.15 +/- 0.13 units after 42 degrees C [pHe (7.20 +/- 0.12) and pHi (7.03 +/- 0.05)], and by 0.12 +/- 0.14 units after 45 degrees C [pHe (6.84 +/- 0.24) and pHi (6.72 +/- 0.19)]. Simultaneous measurements of pH from the intra- and extracellular compartments demonstrated reversal in the pH gradient after hyperthermic treatment.

This study presents the development and testing of an experimental set-up for simultaneous measurements of pH (intracellular by MR and extracellular by fibre-optic pH meter), tumour bio-energetics (by MR), and core/tumour temperatures (by thermocouples) in tumour-bearing mice. Potential sources of measurement errors when using these techniques concurrently in an MR set-up are discussed. Emphasis is placed on simple practical solutions to these problems.

As a first step towards elucidating the potential role(s) of bcl-2 and bcl-2-related genes in lung tumorigenesis and therapeutic responsiveness, the expression of these genes has been examined in a panel of lung cancer cell lines derived from untreated and treated patients, and in cell lines selected in vitro for multidrug resistance. Bcl-2 was hyperexpressed in 15 of 16 small-cell lung cancer (SCLC) cell lines and two of five non-small-cell lung cancer (NSCLC) lines compared with normal lung and brain, and hyperexpression was not chemotherapy related. Bcl-x was hyperexpressed in the majority of SCLC and NSCLC cell lines as compared with normal tissues, and all lung tumour lines preferentially expressed bcl-x1-mRNA, the splice variant form that inhibits apoptosis. Bax gene transcripts were hyperexpressed in most SCLC and NSCLC cell lines examined compared with normal adult tissues. Mutant p53 gene expression was detected in the majority of the cell lines and no relationship between p53 gene expression and the expression of either bcl-2, bcl-x or bax was observed. No changes in bcl-2, bcl-x and bax gene expression were observed in multidrug-resistant cell lines compared with their drug-sensitive counterparts. Images Figure 1 Figure 2 Figure 3 Figure 4 Figure 5

As an approach to the rational design of combination chemotherapy involving the anti-cancer DNA topoisomerase II poison etoposide (VP-16), we have studied the dynamic changes occurring in small-cell lung cancer (SCLC) cell populations during protracted VP-16 exposure. Cytometric methods were used to analyse changes in target enzyme availability and cell cycle progression in a SCLC cell line, mutant for the tumour-suppressor gene p53 and defective in the ability to arrest at the G1/S phase boundary. At concentrations up to 0.25 microM VP-16, cells became arrested in G2 by 24 h exposure, whereas at concentrations 0.25-2 microM G2 arrest was preceded by a dose-dependent early S-phase delay, confirmed by bromodeoxyuridine incorporation. Recovery potential was determined by stathmokinetic analysis and was studied further in aphidicolin-synchronised cultures released from G1/S and subsequently exposed to VP-16 in early S-phase. Cells not experiencing a VP-16-induced S-phase delay entered G2 delay dependent upon the continued presence of VP-16. These cells could progress to mitosis during a 6-24 h period after drug removal. Cells experiencing an early S-phase delay remained in long-term G2 arrest with greatly reducing ability to enter mitosis up to 24 h after removal of VP-16. Irreversible G2 arrest was delimited by the induction of significant levels of DNA cleavage or fragmentation, not associated with overt apoptosis, in the majority of cells. Western blotting of whole-cell preparations showed increases in topoisomerase II levels (up to 4-fold) attributable to cell cycle redistribution, while nuclei from cells recovering from S-phase delay showed enhanced immunoreactivity with an anti-topoisomerase II alpha antibody. The results imply that traverse of G1/S and early S-phase in the presence of a specific topoisomerase II poison gives rise to progressive low-level trapping of topoisomerase II alpha, enhanced topoisomerase II alpha availability and the subsequent irreversible arrest in G2 of cells showing limited DNA fragmentation. We suggest that protracted, low-dose chemotherapeutic regimens incorporating VP-16 are preferentially active towards cells attempting G1/S transition and have the potential for increasing the subsequent action of other topoisomerase II-targeted agents through target enzyme modulation. Combination modalities which prevent such dynamic changes occurring would act to reduce the effectiveness of the VP-16 component.

The pharmacokinetics of pentoxifylline and its three major metabolites were measured after intraperitoneal administration of 10 mg/g or 100 mg/kg of drug in C3H mice. Peak concentrations of pentoxifylline were approximately 10 and 100 micrograms/ml, respectively, with elimination half-lives (+/- 2 SE) of 4.6 (4.2-5.1) and 7.5 (7.2-7.9) min, respectively. Plasma concentrations of the pharmacologically active hydroxy metabolite were approximately one-tenth those of the parent compound. In vitro evidence of the ability of pentoxifylline to increase blood cell deformability indicates that concentrations of up to 30 micrograms/ml can increase deformability of both red and white blood cells; doses between 5 mg/kg and 100 mg/kg were therefore tested 15 min after administration to test the effect of the drug on tumour and normal tissue perfusion, tumour radiosensitivity and renal function immediately after exposure to appropriate drug concentrations. Using 86Rb extraction, doses of 10-100 mg/kg pentoxifylline were shown to increase relative tumour perfusion of the RIF-1 tumour to 140-170% of control, with no effect in skin, muscle, kidney, liver or lung, but with similar increases in spleen perfusion; there was no significant effect in any tissue after 5 mg/kg. Using a clonogenic assay, this increased tumour perfusion was shown to be reflected in increased tumour radiosensitivity to 25 Gy 15 min after pentoxifylline, with the same dose threshold of 10 mg/kg, and similar lack of dose-dependence at higher doses; the response indicated reduction in hypoxic fraction by a factor of 2-3. Renal function, measured by [51Cr]EDTA and [125I]iodohippurate clearance was unaffected at doses up to 50 mg/kg, with a slight effect at 100 mg/kg. The data indicate that pentoxifylline is effective at increasing relative tumour perfusion, with minimal effects on other tissues, and this increase is reflected in improved radiosensitivity. The doses at which the drug is effective are compatible with the mechanism being modification of blood cell deformability. Pentoxifylline shows promise as a clinical radiosensitiser acting by direct increase in tumour oxygenation.

The effect of combining the oxygen-transport-modifying drug BW12C with mitomycin C was investigated in a phase 1 study of 26 patients with advanced gastrointestinal cancer. The dose of BW12C was increased from 20 mg/kg to 60 mg/kg. Dose-limiting toxicity of vomiting was experienced at doses greater than 50 mg/kg. This corresponded to whole blood levelsor = 700 micrograms/ml and to50% haemoglobin modification. Whole blood concentrations of BW12C and modification of the haemoglobin oxygen saturation curve were linearly dependent on dose. BW12C whole blood pharmacokinetics were best described by a one-compartment model and were clearly dose-dependent. The half-life increased from 2.1 h at a dose of 20 mg/kg to 7.2 h at a dose of 60 mg/kg. The AUC increased in a similar non-linear fashion with increasing dose. Mitomycin C was given at a fixed dose of 20 mg/m2 at the end of the BW12C infusion. Mitomycin C plasma pharmacokinetics fitted a two-compartment model, giving a mean beta half-life of 50 +/- 7 min and AUC of 1.1 +/- 0.08 micrograms/ml h, and were unaffected by the combined treatment. There was no evidence of increased mitomycin C toxicity.

Tumour tissue oxygenation has been measured in man during carbogen breathing (95% O2, 5% CO2) using a commercially available polarographic electrode system (Eppendorf p02 histograph). At least 200 tumour measurements in each of 17 patients with accessible tumours were taken before, and subsequently continuously after the commencement of carbogen breathing for periods of 10 to 30 min. In 12 out of 17 patients studied there was a significant increase in median tumour p02 during the first 10 min of carbogen breathing (range 9 to 1800%). There was an initial rapid increase in tumour p02 which was maintained until 8 to 12 min, but then decreased throughout the subsequent treatment period. Although there was a reduction in the proportion of point measurements < or = 10 mmHg in 11 out of 13 patients, during carbogen breathing, measured points of < or = 2.5 mmHg were only eliminated in three out of 11 tumours. The time course has implications for the planning of clinical trials utilising radiotherapy with carbogen breathing.

The effect of combining the oxyhemoglobin-modifying drug BW12C with mitomycin C was investigated in a Phase I study of 18 patients with advanced gastrointestinal cancer. The dose of BW12C was increased from 20 mg/kg to 50 mg/kg to modify the hemoglobin-oxygen saturation curve by up to 48%. The period of maximum modification was then prolonged for up to 3 hr by a maintenance infusion of 4-6 mg/kg/hr. Pharmacokinetics of BW12C and mitomycin C were performed in all patients. Peak levels of BW12C increased from 139 micrograms/ml to 378 micrograms/ml. Plasma half life was independent of dose, with an average of 3.3 hr. BW12C was well tolerated with no severe side effects. Three patients had objective tumour responses.

Tumor blood flow modification is currently under investigation as a possible means of optimizing current cancer therapies, with particular respect to improving the efficacy of bioreductive agents. A variety of calcium channel blockers have been shown to modify tumor perfusion in model systems, and may be valuable as potentiators of both bioreductive and conventional drugs. We report the effects of nifedipine, verapamil, flunarizine, and cinnarizine on renal function in C3H mice, assayed by clearance of simultaneously injected 51 Cr ethylenedlamine tetraacetate. Nifedipine at 10 mgkg −1 blocked 51 Cr ethylenediamine tetraacetate clearance for 30 min and reduced its subsequent rate of clearance by a factor (± 2 se) of 2.4 ± 0.6. At 1 mgkg −1 it reduced the rate of clearance by a factor of 1.2 ± 0.2. Verapamil at 10 mgkg −1 blocked 51 Cr ethylenediamine tetraacetate clearance by a factor of 1.2 ± 0.2. Verapamil at 10 mgkg −1 blocked 51 Cr ethylenediamine had no effect at 1 mgkg −1 Flunarizine had no effect at 50 mgkg −1 or at 5 mgkg −1 but cinnarizine at 50 mgkg −1 reduced clearance rate by a factor of 1.2 ± 0.1. The data show that some of these vasoactive agents, nifedipine and verapamil in particular, can severely compromise renal function and may, therefore, affect the plasma pharmacokinetics of co-administered drugs that are cleared by the kidney.

A novel non-immunosuppressive cyclosporin [corrected], PSC-833, has been tested for its ability to circumvent resistance to doxorubicin, vincristine and colchicine in human and murine multidrug resistant (MDR) cell lines. This compound is shown to be a highly potent resistance modifier, being 7-10-fold more potent than the parent compound, cyclosporin A, whilst approximately equal to cyclosporin A in the growth inhibitory effects of compound alone. Reversal of the P-glycoprotein-associated MDR drug accumulation defect is a major component of resistance reversal for PSC-833, as it is for cyclosporin A.

The present study examines the relationship between neuroendocrine (NE) differentiation and the clinical behaviour of non-small cell lung cancer (NSCLC). Retrospective (n = 315) and prospective (n = 44) cohorts of non-small cell tumours were obtained from surgically treated cases of lung cancer, comprising 218 squamous cell carcinomas, 65 adenocarcinomas, 51 adenosquamous carcinomas, and 25 large cell undifferentiated carcinomas. Paraffin wax embedded and fresh frozen tissue sections were stained for the NE markers neurone specific enolase, creatine kinase-BB, bombesin, neurotensin, chromogranin A, synaptophysin and UJ-13A. The expression of two or more markers was observed in 30% of cases, and was taken to identify NE-NSCLC. A statistically significant correlation between nodal status and NE differentiation (P = 0.05), and disease stage and NE differentiation (P = 0.04) was observed. However, there was no correlation between NE differentiation and survival. These findings suggest that NE-NSCLC, analogous to SCLC is more highly metastatic than non-NE-NSCLC.

During a clinical toxicity study it was possible to obtain urine samples from six patients receiving either the R-(-)- or S-(+)-stereoenantiomeric forms of the developmental 2-nitroimidazole radiosensitizer Ro 03-8799 (pimonidazole). Paired plasma samples were also obtained from four patients. The pharmacokinetic data were compared with those for the racemic mixture in the same individuals. The results revealed no major differences in the plasma pharmacokinetics, urinary clearance or N-oxidation of the individual enantiomers as compared with the racemic mixture. A similar lack of stereoselectivity with respect to the acute dose-limiting CNS toxicity syndrome suggests that this may not involve a specific CNS receptor interaction.

The hypoxic cell radiosensitizers Ro 03-8799 (pimonidazole) and SR 2508 (etanidazole) are under evaluation as single agents (Phase III) and in combination (Phase I). Ro 03-8799 produces an acute, transient central nervous system syndrome, whereas SR 2508 causes cumulative, peripheral neurotoxicity; both effects are dose-limiting. Pharmacokinetic studies have shown the importance of area under the plasma drug concentration versus time curve (AUC) in predicting the risk of peripheral neuropathy. Most patients have very similar pharmacokinetic parameters. This study reports 2/25 patients receiving 0.75 g/m2 Ro 03-8799 plus 2.0 g/m2 SR 2508 who showed significant discrepancies in drug handling. One patient exhibited a markedly elevated AUC and prolonged t1/2 beta for SR 2508 and this was associated with an unusually rapid onset of peripheral neuropathy. A second patient showed normal handling of SR 2508 but prolonged values for both t1/2 alpha and t1/2 beta for Ro 03-8799 and unusually low levels of its N-oxide metabolite. In addition a low peak Ro 03-8799 concentration combined with a very high volume of distribution was found in this patient, leading to a normal AUC value and toxicity profile. Both patients exhibited a relatively low creatinine clearance. The mechanisms which may underlie these findings are discussed, and the importance of pharmacokinetic monitoring in the use of these agents is emphasized.

Piritrexim is a lipid-soluble drug which is as effective an inhibitor of dihydrofolate reductase as methotrexate. Phase I and II studies have indicated activity in some tumour types. Because of its lipophilicity we have conducted a phase II study in recurrent high-grade malignant glioma (grades III and IV). Twenty-seven patients were treated with 25 mg p.o. three times daily for five consecutive days, repeated weekly, with provision for dose escalation or reduction according to toxicity. Five patients received less than 4 weeks' treatment because of disease progression or death. Twenty-two patients were evaluable for response. One complete and one partial response was seen (duration 262+ and 241+ weeks) and 13 patients had static disease for a median duration of 13 weeks (range 7-35). The major toxicity was myelosuppression. This response rate of 9% of evaluable patients is much lower than that seen for some conventionally used drugs and we conclude that piritrexim is unlikely to be of value in the management of high-grade gliomas.

Combining hyperthermia, an experimental/adjuvant therapeutic modality for cancer, with the non-invasive metabolic studies using Magnetic Resonance (MR) is an interesting area of research. This two parts article discusses the development and testing of a conventional RF hyperthermia applicator for MR studies and vice versa. In this first part, an inductive type applicator known as 'Magnetrode' in RF hyperthermia has been used both as an MR volume resonator and a surface coil. Its concurrent performance as an hyperthermic applicator and an MR transmit/receive coil has been evaluated.

The effects of pentobarbital anesthetic and physical restraint have been evaluated in murine RIF-1 tumors in terms of their suitability in providing minimal temporal variations in core and tumor temperatures, intracellular (pHMR) and extracellular (pHF) tumor pH, and tumor bioenergetics. pHF was measured using a fibreoptic pH probe. The implications of the changes in these parameters on the combined studies of in vivo MRS and hyperthermia have been discussed. The temporal variations of core and tumor temperatures, pHF and pHMR between the anesthetised and restrained mice were statistically significant (p < 0.0001, p < 0.005, and p < 0.0001, respectively). Differences in the temperatures between the two groups varied with time, and were maximum at about 2 h after pentobarbital. In the anesthetised mice, while there was a net increase of 0.28 pH units in pHF, there was a drop by 0.32 pH units in pHMR (p < 0.001). Temporal variations of both pHF and pHMR in the restrained mice were relatively constant. In light of these findings, wherever possible, physical restraint in a suitably designed holder is suggested for immobilizing animals for magnetic resonance spectroscopy studies.

The reduced growth factor requirements of murine fibroblasts transformed by simian virus 40 (SV 40) have been attributed to insulin-like growth factor (IGF)-I induction by T antigen and consequent activation of IGF-I receptor signaling. The present study shows that the autonomous growth of SV 40-transformed human fibroblasts also requires type-I IGF-I receptor activation but that this is not due to de novo induction of IGF-I gene expression since untransformed human fibroblasts, which fail to proliferate in the absence of serum, also showed IGF-I gene expression under serum-free conditions. DNA synthesis assays confirmed that untransformed cells were responsive to exogenous IGF and indicated that transformed cells were already maximally stimulated. In untransformed fibroblasts, IGF binding was principally to abundant membrane-associated IGFBP-5, whereas in transformed fibroblasts this protein was minimally expressed, and IGF binding was to IGF receptors. Loss of detectable membrane-associated IGFBP-5 in transformed cells was associated with diminished IGFBP-5 gene expression and with loss of IGF-II gene expression. Exogenous IGFBP-5 associated with the membranes of transformed cells and inhibited the autocrine growth of these cells. These findings suggest that loss of IGFBP-5 in SV 40-transformed fibroblasts facilitates interaction of endogenously produced IGF-I with the IGF-I receptor and increases their sensitivity to autocrine stimulation.

The time course and dose response of 600 mm3 subcutaneous RIF-1 tumours to pentoxifylline was measured in terms of relative tumour perfusion, assayed by 86Rb extraction, and pO2 distribution measured by the Eppendorf histograph. Both perfusion and pO2 distribution were maximally increased by 20 mg/kg pentoxifylline 15 min after administration, perfusion to 141 ± 15% (2 SE) of control, and median pO2 from 3 to 15 mmHg, with the percentage of values

Growth hormone (GH) regulation, glucose tolerance and serum concentrations of insulin-like growth factor (IGF) and IGF binding proteins (IGFBP) have been investigated in small cell lung cancer (SCLC) patients. Elevated serum GH was observed in the patient and smoking control groups but not in non-smoking control subjects. Glucose suppression of GH was observed in the few SCLC patients with raised basal GH but most SCLC patients exhibited a paradoxical increase in GH following oral glucose. Abnormal glucose tolerance and insulin resistance with respect to plasma glucose was observed in most patients. Patients showing GH dysregulation exhibited higher serum concentrations of IGFBP-2 than those showing no increase in GH. Abnormal glucose tolerance was associated with decreased serum concentrations of IGF-I. Given reports of elevated IGFBP secretion in SCLC and inhibition of IGF-I bioactiviry by IGFBPs, these findings may indicate that increased serum IGFBPs disrupt IGF-I regulation of GH secretion and glucose homeostasis.

Nicotinamide is an effective radiosensitiser of murine tumours, functioning by improving tumour perfusion by decreasing the proportion of intermittently closed capillaries. The effect of nicotinamide on relative tissue perfusion of RIF-1 tumour and normal skin, muscle, lung, liver, kidney and spleen were investigated using the 86 Rb extraction technique. A dose of 1000 mg/kg was shown to have transient effects on tumour, skin and lung perfusion but to have sustained effects on muscle (a drop to 80% of control), liver, kidney and spleen (with increases ranging from 165% to 280% of control) from 0.5 to 4 h after treatment i.e. during the period of maximum radiosensitisation. These increases were evident at doses as low as 100 mg/kg. The data suggest that the radiosensitisation induced by nicotinamide in the mouse may be associated with these perfusion changes. Nicotinamide was also shown to have a substantial inhibitory effect on renal function, inhibiting 51 CrEDTA clearance by a factor (± 2 SE) of 2.56 ± 0.19 and 125 I-iodohippurate clearance by a factor of 2.07 ± 0.45 at 1000 mg/kg. These effects were shown to be dose-related, and to be evident at doses from 400 mg/kg upwards. This suggests that nicotinamide potentiation of co-administered cytotoxic agents may be mediated by reduced renal clearance of the cytotoxic drug, thus increasing the plasma half-life.

This review concentrates on malignant gliomas, with brief mention of cerebral lymphomas, other primary intracranial tumors, and metastases. A recurring theme in the current literature on malignant glioma is the importance of prognostic factors other than treatment in determining outcome. Interesting results are emerging from newer strategies of stereotactic radiotherapy, brachytherapy, and photodynamic therapy, although authors who report these results acknowledge that they have often reviewed patients with intrinsically good prognoses. The year's literature has provided a consolidation of information on prognostic factors, relapse patterns, treatment volume, and dose, providing a basis for future progress.

Purpose: Many radiobiologic mechanisms may contribute to the clinical radiation resistance of Glioblastoma Multiforme. One of them is considered to be an unusually low intrinsic radiation sensitivity. This is a collaborative study between three laboratories to evaluate the intrinsic radiation sensitivity of 85 cell lines derived from human malignant gliomas as the major cause of the poor clinical results of radiation treatment to these tumors. Methods and Materials: Fifty-one cell lines were early passage. The distribution by histologic type was: 58 glioblastoma, 17 anaplastic astrocytoma, six oligodendroglioma and four astrocytoma grade 2. The intrinsic radiation sensitivity will be expressed by the surviving fraction at 2 Gy (SF 2 ). The SF 2 has been determined for single dose irradiation for cell lines on exponential phase, under aerobic conditions, growing on plastic. The patient age, Karnofski Status, histological grade, survival, dose of irradiation for 50 patients are investigated for correlation with SF 2 of the corresponding newly established cell lines. Results: The mean SF 2 of the 85 cell lines was 0.46 (0.12 – 0.87). The mean SF 2 by histologic type was 0.50, 0.34, 0.54 and 0.38 for glioblastoma, anaplastic astrocytoma, oligodendroglioma and astrocytoma grade 2 cell lines, respectively. No correlation was found between SF 2 and the patient age or Karnofski status. The difference in SF 2 between the 58 glioblastoma and 17 anaplastic astrocytoma cell lines was significant p = 0.002. The difference in actuarial survival between glioblastoma and anaplastic astrocytoma patients was borderline of significance ( p = 0.08). The difference in SF 2 of cell lines derived from these two groups of patients was of borderline significance ( p = 0.08). The difference in radiation sensitivity for anaplastic astrocytoma and glioblastoma cell lines was clearly reflected in the difference in survival for the two groups of patients from where the cell lines were derived. However, no correlation was found between SF 2 and survival within each grade. In a multivariate analysis the age, grade and Karnofski status were found to be significant prognostic values for survival with a p values of 0.032, 0.03 and 0.038, respectively, however, the In SF 2 was not significant ( p = 0.40). The mean SF 2 of the 6 oligodendroglioma cell lines (0.54) was comparable to that of glioblastoma multiforme (0.50). The high SF 2 for oligodendroglioma does not accord with the much better clinical outcome of these tumors. Conclusions: These data on 85 malignant glioma cell lines show a very broad distribution of SFZ values for irradiation in vitro . SF 2 reflected the difference in sensitivity between AA (Grade 3) and GBM (Grade 4). This may suggest that the parameter SF 2 is useful to discriminate between the sensitivity of different grades or types of histology in vitro . However, SF 2 was not a predictor of the clinical outcome on individual basis for malignant gliomas. The in vitro studies will need to be supplemented by physiologic characterization of the tumors in vivo . Such conclusions would limit the predictive value of current radiation sensitivity assays based on in vitro dose-survival measurement for at least high grade malignant gliomas.

The objective of this study was to compare pH measurements made in biological samples using 31P NMR (pHNMR) with those made with a novel, dye-based fibreoptic pH measurement system (pHF), which is compatible with use in electromagnetic fields without field perturbation. Using protein-free model solutions, pHNMR was calibrated against pHF, giving a correlation coefficient of 0.969 and a mean difference (+/- SD) between pHNMR and pHF of 0.037 +/- 0.054 over the pH range 6.8-7.7. Further calibration of pHNMR with pHF was carried out for human red blood lysates and then pHNMR was compared with pHF for whole, packed red blood cells over the pH range 7.0-7.8. Values for pHNMR, the intracellular pH, were consistently lower than for pHF, the extracellular pH, by a mean (+/- SD) of 0.15 +/- 0.02 units. A close correlation of extracellular pHNMR with pHF was demonstrated for a blood sample exhibiting two P(i) peaks, over the pH range 7.03-7.71. We conclude that concurrent use of NMR and the fibreoptic pH meter provides a reliable method of simultaneous measurement of intracellular and extracellular pH in biological systems.

The secretion of insulin-like growth-factor-binding proteins (IGFBPs) and expression of the genes encoding IGFBP-I, IGFBP-2 and IGFBP-3 have been studied in a panel of cell lines derived from breast carcinomas, Wilms' tumour, neuroblastoma, retinoblastoma, colon carcinoma, liver adenocarcinoma, Burkitt's lymphoma and a non-small-cell lung carcinoma. All cell lines, with the exception of the Burkitt's lymphoma cell line, secreted IGFBPs, as detected by affinity labelling. A 34-kDa BP was present in the conditioned media of all IGFBP-secreting cell lines, whereas BPs ranging from 18 kDa to 53 kDa were variably secreted. All IGFBP-secreting cell lines expressed the IGFBP-2 gene as determined by Northern blot analysis. The Wilms' tumour, the neuroblastoma and the retinoblastoma cell line expressed the IGFBP-2 gene only. All other cell lines, with the exception of the Burkitt's lymphoma, expressed the IGFBP-2 gene and, in addition, either the IGFBP-I gene and/or the IGFBP-3 gene. IGFBP-I gene expression could be detected by reverse transcriptase polymerase chain reaction only. IGFBP-3 gene expression was detected by Northern blot analysis, but transcripts were less abundant than IGFBP-2 mRNAs. These findings indicate that the expression of multiple BP genes and the secretion of BPs may be a common property of tumour cells. © 1992 Wifey-Liss, Inc.

Radiotherapy remains the main treatment modality for patients with malignant gliomas and is the only treatment which significantly prolongs survival. Clonogenic and tetrazolium based colorimetric assays (MTT) of early passage cultures have been performed following 2 Gy doses of x-rays in order to determine if in vitro radiosensitivity is a factor in response to treatment. Of 47 biopsies received, 39 were established in primary culture. A value of surviving fraction to 2 Gy (SF2) was obtained in 85% of growth assays and 64% of clonogenic assays. The mean SF2 value for the MTT was 0.56 which was significantly higher than the 0.42 obtained for the clonogenic assay. There was, however, reasonable qualitative agreement in assessing relative radiosensitivity/radioresistance (r = 0.7). Mean SF2 values for grade 3 tumors were 0.52 (MTT) and 0.35 (clonogenic) as against mean SF2 values of 0.63 (MTT) and 0.47 (clonogenic assay) for grade 4 tumors. In 24 patients with adequate follow-up, no direct correlation was found between SF2 and survival, although mean SF2 values for patients surviving greater than 18 months was significantly less (p = 0.01) than patients surviving less than 18 months as determined by the MTT assay.

Hydralazine has been widely used to reduce tumor blood flow in mice. It has an application in the deliberate creation of reducing environments within tumors and has been used in conjunction with both bioreductive and cytotoxic drugs. We have compared the dose-response to hydralazine of relative tissue perfusion of KHT tumor, kidney and liver, assayed by 86 Rb extraction, over the dose range 0.2 to 5.0 mgkg −1 and shown that doses of 1.0 mgkg −1 and higher cause significant reductions in perfusion of all three tissues but 0.2 mgkg −1 has no effect. Tumor perfusion (±2 se) was reduced to 80 ± 8% of control by 1.0 mgkg −1 , to 38 ± 13% by 2.5 mgkg −1 and to 35 ± 7% by 5.0 mgkg −1 . Relative kidney perfusion was reduced to 83 ± 11% of control by 1.0 mgkg −1 and to 73 ± 9% by 5.0 mgkg −1 ; relative liver perfusion was reduced to 71 ± 10% of control by 1.0 mgkg −1 and to 64 ± 10% by 5.0 mgkg −1 . This reduction in kidney and liver perfusion may indicate that there would be impairment of elimination and/or metabolism of co-administered drugs. We have therefore also measured the dose-response of the effect of hydralazine on glomerular filtration rate and effective renal plasma flow, assayed by clearance of 51 CrEDTA and 125 I-iodohippurate, respectively. 5.0 mgkg −1 hydralazine blocks clearance of EDTA for 40 min, slows subsequent clearance by a factor (±2 se) of 2.4 ± 1.2, and Slows 1211-iodohippurate clearance by a factor of 5.5 ± 0.8; 1.0 mgkg −1 hydralazine slows EDTA clearance by a factor of 1.5 ± 0.3. The time-course of the effect of 5.0 mgkg −1 hydralazine on isotope clearance was measured and this dose was shown to affect isotope clearance at times up to 4 h after administration. These data confirm that hydralazine at doses effective at reducing tumor blood flow also impairs renal function, and is therefore likely to affect the pharmacokinetics of any co-administered drug that is cleared by the kidney.

The two radiosensitizers etanidazole and pimonidazole, currently in clinical phase 3 trials, have different toxicities. The former produces a peripheral neuropathy after cumulative doses and the latter presents an accurate but transient central nervous system syndrome after each dose. A strategy for improving on the maximum radiosensitization achievable using either drug alone, has been investigated in the study reported in this paper. Escalating doses of the two drugs were given together to determine toxicity up to a maximum of 15 doses of 2.0 g/m2 etanidazole and 0.75 g/m2 pimonidazole/dose. 25 neuropathies were seen in the total of 48 patients (48%). This included 6 grade 2 neuropathies (12.5%) or 15% of those receiving 9 or more infusions. There were no significant correlations between the incidence of neuropathy and various dose and AUC parameters for the 31 patients receiving 10 or more infusions. However, in the selected group of 26 patients planned to receive multiple doses of etanidazole at 2 g/m2/dose with pimonidazole at 0.75 g/m2, significant correlations were seen with the cumulative dose, with single and cumulative AUCs for pimonidazole and also for the cumulative dose and the cumulative AUC for etanidazole, but not its single AUC. In the light of these observations and the recent reports of higher peripheral neuropathy rates than previously reported in studies in the USA, it is concluded that it is not possible at this time to determine whether there is toxicity interaction between the two drugs. There remains the possibility that, on the basis of the combined drug dosage achieved, an increased therapeutic efficacy can be reached with either drug alone.

A panel of 3 monoclonal antibodies (MAbs) directed against human lung tumour cell-surface antigens has been produced following immunizations with the established small-cell lung cancer (SCLC) cell line, NCI-H69, and with another SCLC cell line, COR-L32, recently derived from clinical material. One MAb, B10/12, reacted strongly with SCLC, immunoprecipitated a protein having an MW of 100kd and failed to react significantly with non-small-cell lung cancer (NSCLC) in radioimmunoassay and in an immunohistochemical assay. MAbs E10/5 and 2G3 reacted extensively with SCLC but also showed significant reactivity with NSCLC. MAb E10/5 immunoprecipitated a protein with an MW of 80kd but no appreciable protein was specifically precipitated by MAb 2G3. Unlike MAb 2G3, both MAbs B10/12 and E10/5 reacted strongly with selected neuroblastomas whereas only MAbs 2G3 and E10/5 reacted significantly with melanoma. All 3 MAbs reacted with breast carcinomas. Other non-pulmonary tumours thus far examined failed to react with the MAbs in radioimmunoassay or immunohistochemical assay. Immunocytochemistry and the use of viable cells in radioimmunoassay confirmed that the antigenic determinants recognized by these MAbs were surface located.

The hypoxic cell radiosensitizers Ro 03-8799 and SR 2508 have different clinical toxicities. The former produces an acute but transient central nervous system syndrome, whereas the latter produces cumulative peripheral neuropathy. Following single dose studies, an escalating multiple dose schedule using both drugs in combination showed no unexpected adverse reactions at lower doses. This study identifies the clinical tolerance and pharmacokinetics when doses in the region of the maximal tolerated dose are given to 26 patients receiving infusions of 0.75 g/m2 Ro 03-8799 and 2 g/m2 SR 2508 three times per week. At 15 doses, 3/4 patients experienced WHO grade 2 peripheral neuropathy, whereas at 12 doses 1/9 developed grade 2 and 6/9 developed grade 1 neuropathies. This represents a lower dose of SR 2508 than can be given alone suggesting that some interaction between the two drugs does exist in terms of chronic peripheral neurotoxicity. Pharmacokinetic studies show no adverse interactions between the two drugs and minimal inter-patient variation. From bivariate analysis, cumulative AUC for Ro 03-8799 has the most significant correlation with the development of peripheral neuropathy. Tumor drug concentrations normalized to the administered dose show mean values of 34 micrograms/g Ro 03-8799 and 76 micrograms/g SR 2508 30 minutes after infusion. These could be expected to produce a single dose sensitizer enhancement ratio of 1.5. The combination of the two sensitizers at the maximum tolerable dose may be expected to give an increased therapeutic efficacy over either drug alone.

The effects of sodium pentobarbitone anaesthesia, the presence of a tumour, and local hyperthermia to a tumour-bearing leg, on the pharmacokinetics of MISO in the mouse are reported. Analysis of MISO and its metabolite Ro 05-9963 was by high-performance liquid chromatography. The plasma kinetics of MISO were largely unaffected by any of these treatments, but hyperthermia substantially reduced tumour concentrations of the drug. The effects of tumour site and size on unheated-tumour drug concentrations were also studied, and an increase in tumour size was shown to decrease tumour MISO levels, but to different degrees according to whether implanted in the leg or flank. Uniformity of MISO distribution throughout heated and unheated tumours was examined, and levels were found to be constant within tumours. The presence of a temperature detector in heated tumours did not affect their drug concentration.

The HT29R colonic adenocarcinoma xenograft has been shown to be rich in the enzyme beta-glucuronidase. Experiments in rodent systems have demonstrated a marked anti-tumour effect of the drug aniline mustard (AM) on tumours with high levels of this enzyme (e.g. the plasmacytomas PC5 and PC6). We have found that AM is no more effective than its analogue paramethyl aniline mustard (PMAM) or other alkylating agents against the HT29R xenograft. Amongst the possible explanations for this may be: (1) The wide shoulder on the cell-survival curve shown for exposure to alkylating agents of HT29R in vivo. (2) Lack of correlation between physiological availability of beta-glucuronidase and the high levels measured by the standard assay. (3) Increased beta-glucuronidase levels in host mouse marrow, making the latter potentially more susceptible to AM damage.

Since the work of Moon and Richards (I), the 3’P chemical shift of inorganic phosphate (Pi) has been widely used as an indicator of the pH of biological systems. In blood, the Pi resonance often overlaps the doublet resonance of 2,3-diphosphoglycerate (2,3-DPG) and several spectral editing techniques to separate them have been suggested (2-4). However, all of those techniques require a delay of 1/2J,n(n in this method the delay between excitation and acquisition can be much less than 1/2n the 180” pulse refocuses any dephasing due to magnetic field inhomogeneity and the evolution due to chemical shift. Any evolution due to heteronuclear scalar coupling is also refocused. Consequently, both the components of the DPG doublet and Pi will have the same phase. In the second sequence, a 180” pulse is applied at the 31P and ‘H frequencies. Since the 180” proton pulse inverts the spin states of the proton spins, the evolution due to the heteronuclear scalar coupling of the DPG doublet is not refocused. Therefore at the end of the sequence, one of the resonances of the doublet would have evolved through an angle of hJpH, and the other through -rJpH (= kq5). However, the magnetization of the Pi singlet will have the same phase as before. The positions of the magnetization vectors after the two sequences are illustrated in Fig. 2 as EXP 1 a and EXP lb. The phase of the FID acquired with the sequence in Fig. 1 b is now rotated through k,$. When this FID is subtracted from that resulting from the sequence in Fig. lA, one peak of the DPG doublet cancels out, leaving the singlet from the Pi, plus the

The 2-nitroimidazole hypoxic cell radiosensitizer Ro-03-8799 has been suggested to have possible advantages over misonidazole with regard both to radiosensitization and toxicity on the basis of reported experimental work. The present work reports a Phase I escalating dose toxicity study of the drug. This has shown severe acute central neurotoxicity at high dose levels (greater than 1 g/m2). Initial results of a multiple-dose toxicity study indicate that 1 g/m2 is likely to be the maximum dose which may be given repeatedly. The plasma and tumor pharmacokinetics of the drug have been measured. The mean t 1/2 for 9 patients was 5.8 +/- 1.5 hr. Peak plasma concentration is linearly related to dose and at 1 g/m2 is 12.1 +/- 2.3 micrograms/ml (n = 6). Human tumor drug concentrations have been measured after single doses of 1 g/m2 given to 8 patients with a variety of tumors. Peak tumor concentrations of drug of 11.7-81.6 micrograms/g were found. Because of acute, dose-limiting toxicity related to individual doses it may not be possible to achieve, in human tumors, concentrations of drug that offer significant advantage over misonidazole in terms of radiosensitizing efficiency. No evidence of chronic cumulative toxicity was observed at the doses employed.

The value of neuron specific enolase (NSE) immunoreactivity as a marker for small cell lung cancer (SLC) has been assessed using a monoclonal antibody (MCAB) against NSE, MCAB specificity was confirmed using purified enolase isoenzymes, sections of human brain, a panel of lung tumours, neuroendocrine and non-neuroendocrine tumours and normal tissues. Using this MCAB in radioimmunoassay and immunohistochemistry, NSE immunoreactivity was detected in all SCLC material examined. However, considerable reactivity was also observed in a number of non-small cell lung cancer cell lines and tumour biopsy specimens. Furthermore, intratumoral heterogeneity with respect to NSE immunostaining was observed in several cases. Factors which may underlie such intratumoral phenotypic diversity were assessed using flow cytometry together with MCABs directed against both NSE and non-neuronal enolase. Such studies revealed that enolase expression in cells which were no longer actively proliferating differed markedly from that of cells in exponential growth. Furthermore, cells grown under conditions of increasing hypoxia exhibited increased enolase expression relative to those grown under oxygenated conditions. It is concluded from these studies that NSE immunoreactivity per se is an unreliable marker for the SCLC phenotype. Images Figure 1 Figure 2 Figure 3 Figure 4(a) Figure 4(b)

We have produced adriamycin (ADM)-resistant variants of the human lung cancer cell lines NCI-H69 (small cell), MOR (adenocarcinoma) and COR-L23 (large cell) but have failed to produce resistant variants of two other small cell lines. In each case, the derivation protocol took 7-9 months and included a period of drug-free growth. All three resistant lines show reduced cellular content of ADM after 1 h exposure when compared with their controls. During prolonged incubation of control and resistant NCI-H69 cells in 0.4 microgram ml-1 ADM, the ADM content of resistant cells was 6-7 times lower than that of control cells. The ratio of ADM doses to suppress growth of the two lines, however, was in the range of 40-200X. The ADM-resistant variant of NCI-H69 was also resistant to vincristine, colchicine, VP16, mitozantrone, 4' epiadriamycin and 4' deoxyadriamycin, somewhat resistant to melphalan but not resistant to aclacinomycin A, bleomycin of CCNU. The resistance to ADM could be partially overcome by the use of verapamil, an inhibitor of calcium transport.

In a randomized, double-blind study, 44 patients with recurrent high grade malignant glioma were allocated to chemotherapy of CCNU, with or without benznidazole (BENZO). Of 42 eligible patients, 23 received CCNU alone, and 19 CCNU received BENZO. Only 8 patients received the full 6 courses of treatment. The mean number of courses given was 2.8 for placebo and 3.4 for benznidazole patients. Progressive disease caused termination of treatment early for the majority of patients. There was no evidence of increased toxicity-leucopenia, anemia or thrombocytopenia-in the BENZO group, and only one patient in each treatment group had chemotherapy terminated because of toxicity. The BENZO group did not demonstrate any survival advantage: median survival time was 25 weeks in the BENZO group and 30 weeks in the placebo group. The confidence interval for the treatment difference is wide but excludes a BENZO-related addition of more than 7 months to the median survival time.

A review of misonidazole pharmacokinetics in 83 consecutive patients treated for tumours other than glioma has shown that among patients not receiving enzyme-inducing agents the plasma elimination half life is lower in patients taking steroids. Such a difference is not seen if patients already taking enzyme inducers are given steroids. Five further patients with carcinoma of the lung, treated with radiation over a period of 3 weeks, have been studied in greater detail. Misonidazole, in oral dose of 1 gm-2, was given in conjunction with the first and last radiotherapy fractions, and dexamethasone, in a divided daily dose of 8 mg, was given throughout the radiation treatment, commencing after the first treatment. Misonidazole pharmacokinetics were studied at each administration. Following the dexamethasone treatment period there was a 25% reduction in misonidazole plasma elimination half life, a 24% reduction in plasma AUC0-00, and a 38% increase in 24 h urinary excretion (all changes being statistically significant--P less than 0.005). No changes were observed in the plasma AUC0-24 and urinary excretion of the major metabolite desmethylmisonidazole. Glomerular filtration rates in one patient, before and after treatment with dexamethasone, remained unchanged. These results suggest that the effect of dexamethasone on misonidazole kinetics is not related to an enhancement of demethylation.

The concentration of drug after oral administration of 1 g misonidazole to patients undergoing neurosurgery has been studied in samples taken from intracerebral tumours, normal brain, cerebrospinal fluid and brain tumour cyst fluid. A total of 31 patients yielded samples at various times 2-8 h after drug administration. Results show a considerable range of tumour/plasma percentages but from about 2-5 h after drug administration this averaged between 65-80%. Normal brain and CSF showed similar good penetration but drug entry into cyst fluid was lower (30-50%). A further study with desmethylmisonidazole given to 11 additional patients showed a similar range of tumour penetration but lower concentrations in CSF and cyst fluid (10-20%). Low values were also seen in some but not all of the normal brain tissue samples. Necrotic tumour also had much lower concentrations than macroscopically viable tissue. These results are discussed in terms of their clinical relevance to sensitiser studies on patients with brain tumours with drugs of different lipophilicities.

The problems of the combined modality treatment of inoperable lung cancer are discussed in this paper. There is a lack of secure data on optimal radiation dose and fractionation schedules when this is the sole treatment. The problem is compounded in the combined modality situation by the possibility of increased toxicity. Whereas there is effective chemotherapy for small cell carcinoma and evidence for improved results in combined modality treatments, the results of such treatments for non-small cell carcinoma are disappointing. Strategies for improving the results are discussed.

The combination of hyperthermia and the hypoxic cell radiosensitizer Ro-07-0582 has been investigated on the EMT6 tumour implanted into the legs of BALB/c mice. Treatments at a drug dose of 1 mg/g over a range of waterbath temperatures from 37 to 45 degrees C are described. The surviving clonogenic fraction following treatment was assayed in vitro. Measurements of intra-tumour temperature have been made, and shown to be better correlated with the cytocidal effect on the tumour than the waterbath temperature. No significant effect of Ro-07-0582 was observed at 37 degrees C. However, marked cytotoxicity due to the drug was seen at intra-tumour temperatures above 42-5 degrees C for 1 h. These were in addition to the cytocidal effect of the hyperthermia. The results are discussed in relation to the distribution of temperatures and hypoxic cell populations throughout the tumour.

The hypoxic cell radiosensitising drug misonidazole, [ 1 -( 2 -nitroimidazol- 1 -yl)- 3 -methoxypropan- 2 -ol, Ro 07-0582 ] was administered to dogs at dose levels of 50–200 mg/kg on four consecutive weekly occasions by oral and intravenous routes. High-performance liquid chromatography was used to monitor the subsequent plasma concentrations of misonidazole and its 0 -demethylated metabolite [ 1 -( 2 -nitroimidazol- 1 -yl)- 2,3 -propandiol, Ro 05-9963 ]. Both misonidazole and Ro 05-9963 were also detected in the urine in the free and glucuronide-conjugated forms. The detailed pharmacokinetics of misonidazole in the dog are presented and some observations are made on the toxicity of the drug in this species. We conclude in the light of these pharmacokinetic data that the dog may prove to be a better model for the study of misonidazole than those presently used in the laboratory, particularly the rodent species.

We have compared the effects of alpha and recombinant gamma interferons (IFNs) on the growth of human lung cancer cell lines in vitro. There was a diversity of response amongst the lines studied, the most sensitive being COR-L23 (a large cell anaplastic carcinoma line) and POC (a small cell line). In these two lines, IFN-gamma was found to be more potent than IFN-alpha. During cell growth of line POC in the presence of IFN-gamma no significant shift in cell cycle distribution occurred. When lines COR-L23 and POC were grown as xenograft tumours in nude mice, daily injection of 4 X 10(5) units per mouse per day of IFN-gamma produced no discernible retardation of tumour growth.

The maximum single dose of the 2-nitroimidazole hypoxic cell radiosensitiser Ro 03–8799 is limited to 1 g/m2 by the occurrence of a well characterised acute syndrome of sweating, nausea and mental changes. In an attempt to increase the tolerable dose, the clinical toxicity of the racemic mixture was compared with that of the R- and S-enantiomers of Ro 03–8799. Twelve patients received escalating alternate doses of racemic mixture and R- or S-enantiomer, the dose levels being 0.25 g/m2, 0.5 g/m2, 0.75 g/m2 and 1.0 g/m2. Careful monitoring of the acute syndrome failed to demonstrate any consistent differences between racemic mixture and either enantiomer. This would suggest that the toxicity is not mediated via any specific central nervous system receptor. It is concluded that separation of Ro 03–8799 into its enantiomers will not enable a clinically useful increase in dosage.