Poster presentations

Background: Purpose: The aim of the study was showing the transplant activity in aspect local donors and donors received from Center of Coordination Poltransplant. We showed the aspects of utilization donors, reason for refusal, educational program. We analyzed 8 years of activity. Educational program: From June 2005 to December 2006 we want to win people over transplantation and donation. We would specifically cover procurement promotions including the production and distribution of brochures, advertising, public education programs.

The issue of gene delivery has recently been identified as the bottleneck for human gene therapy in general [1]. For cancer gene therapy it is namely the lack of transfection efficiency and specificity, of synthetic gene delivery agents which hamper transfer of new therapeutic strategies into the clinic [2]. Viral vectors, while more efficient, have raised serious safety concerns. While numerous compounds have been tested as potential delivery agents this approach was historically based on trial and error. Rational design of molecules which show optimum interaction with DNA would greatly accelerate the development of more efficient systems. The use of dendrimers as gene delivery agents has been largely focused on the use of the polyamidoamine molecules and although polypropylenimine dendrimers contain 100% protonable nitrogen and bind DNA they have been relatively poorly studied as gene transfer agents. Here we demonstrate how insights from computer based molecular modelling can be used to select suitable molecules from a series of polypropylenimine dendrimers. Methods: DNA binding was evaluated by measuring the reduced fluorescence of ethidium bromide and molecular modelling of dendrimer -DNA complexes was also carried out. Cell cytotoxicity was evaluated against the A431 cell line using the MTT assay. In vitro transfection was evaluated against the A431 cell line using the b-galactosidase reporter gene and DOTAP, served as a positive control. Results: Molecular modelling revealed that the number of dendrimer -DNA contact points per dendrimer and the simulated dendrimer -DNA intermolecular energies increase with dendrimer generation. DNA condensation (measured by the exclusion of ethidium bromide), and experimentally determined dendrimer -DNA positive zeta potential also increased with dendrimer generation. Cell cytotoxicity was largely but not exclusively generation dependant and cytotoxicity followed the trend DAB 64 > DAB 32 > DAB 16 > DOTAP > DAB 4 > DAB 8 while transfection efficacy followed the trend DAB 8 = DOTAP = DAB 16 > DAB 4 > DAB 32 = DAB 64. Conclusion The generation 2 polypropylenimine dendrimer combines a sufficient level of DNA binding with a low level of cell cytotoxicity to give it optimum gene transfer activity in vitro. 1. Anderson WF. Human gene therapy. Nature 1998; 392: 25-30. 2. Schätzlein AG. Non-viral vectors in cancer gene therapy: Principles and progress. Anti-cancer Drugs 2001; 12: 275-304.

Introduction and Objectives:
A novel gene therapy system has been developed using adenoviruses encoding E. Coli nitroreductase (NR), which converts the prodrug CB1954 into a potent alkylating agent. To elicit additional systemic antitumour effects single and combination adenoviral vectors expressing NR and mouse granulocyte-monocyte colony stimulating factor (mGMCSF) have also been generated. The syngeneic TRansgenic Adenocarcinoma Mouse Prostate (TRAMP) model of prostate cancer was utilised both in vitro and in vivo to characterise the effects of these vectors in prostate cancer. Methods: UV microscopy and FACS analysis were used to assay infectivity of the cell lines by adenoviruses in vitro with a vector expressing the marker gene Green Fluorescent Protein (GFP). Agalactosidase expressing vector was used to assay infectivity and transgene expression levels in vivo. Subcutaneous flank tumours were established in syngeneic C57/Bl6 mice for in vivo studies of both intratumoural injection and ex-vivo gene therapy. Results: Approximately 100% of TRAMP cells were infected at an adenoviral multiplicity of infection (MOI) of 100. In vitro enhancement of toxicity to CB1954 was similar in cells infected with the dual virus or a virus expressing NR on its own. Studies of established subcutaneous TRAMP C2 tumours injected with Ad5-b-Galactosidase showed 1000 fold increases in transgene expression compared with mock infections. Injection of the dual expressing vector Ad5-mGMCSF-NR produced a survival benefit in animals with TRAMP C2 tumours, but no additional benefit in survival was produced by injection of CB1954. Ex-vivo gene therapy of TRAMP C2 cells with Ad5-mGMCSF ablated tumour development compared with control adenovirus. Conclusions: Efficient infection and expression of transgenes from adenoviral gene therapy vectors has been shown in the TRAMP model. Subcutaneous tumours grown in syngeneic mice respond to intratumoural Ad5-mGMCSF/NR but addition of prodrug conferred no additional survival advantage. Ex-vivo treatment of TRAMP C2 cells with Ad5-mGMCSF ablates tumourigenesis, suggesting that mGMCSF is the therapeutically active transgene.
Introduction and Objectives: An adenovirus dl1520, which selectively replicates in & kills cancer cells, is produced by deletion of the adenoviral E1B sequence. Variants of dl1520 containing the gene for the prodrugconverting enzyme E. Coli nitroreductase (NR) were generated in an attempt to enhance cancer cell killing. The transgene is placed under the control of a prostate specific transcriptional regulatory region. Methods: The gene for NR was cloned downstream from a 632bp PSA promoter sequence and a single 1469bp PSA enhancer sequence into the E1 region of the adenoviral genome. Replication of the virus was analysed by dot blotting of viral DNA preparations. LNCaP cells were infected with a similar marker virus encoding GFP and transgene expression analysed by UV microscopy and FACS analysis. NR expression was analysed by Western blotting of protein lysates of LNCaP cells infected with adenovirus encoding NR Results: Replication of the adenoviral vector within LNCaP cells was demonstrated on dot blot analysis of viral DNA preparations. Infection of LNCaP cells with virus encoding GFP demonstrated high levels of transgene expression by both UV microscopy and FACS analysis. Western blot analysis confirmed rising NR expression levels over a 5-day time course. Infection of LNCaP cells with NR encoding adenovirus and addition of the prodrug CB1954 produced additive cell killing. Conclusions: Novel adenoviruses capable of replicating and expressing marker and therapeutic transgenes in the prostate cancer cell line LNCaP have been produced. Transgene expression levels from the PSA TRR can be increased by viral replication. Additive cell killing effects have been achieved using this vector in its capacity as gene therapy vector and viral oncolytic agent. Three-and four-stranded DNA structures are important targets in drug design. Antigene strategies for protein down-regulation require adjuvant ligands to stabilise triplex structures; four-stranded DNA is implicated in c-myc expression and the abnormal genomic DNA of Werner's, Bloom's and Fragile-X syndromes, besides its role in the regulation of telomerase. Ligands with precisely-defined structural selectivity could therefore have a range of potential therapeutic applications. Furthermore, the avoidance of undesirable cytotoxic effects dictates that any competing affinity for duplex structures must be eliminated in the course of the design. Pyrimidines bearing (w-aminoalkyl)aryl substituents (e.g. 1) are intriguing ligands for high-order DNA structures. Preliminary studies 1 indicate that thioether-linked compounds are remarkable stabilisers of triplex DNA, whilst minor alteration of the functional groups in the linker switches the structural preference toward four-stranded DNA structures.

P237 NOVEL LIGANDS FOR THREE-AND FOUR-STRANDED
Novel 4,6-diarylpyrimidines and related chromophores have been designed and synthesised to explore how the linker functional groups and length, the disposition of charges and amine pKa influence the SAR for interaction with relevant nucleic acid secondary structures. Details of the synthesis, biophysical studies and in vitro cytotoxicity will be presented.
(1) Wheelhouse et al., 2001, Clin, Cancer Res. 7 711 Suppl. S. Murphy et al., 2001 Many tumour cell lines develop resistance towards alkylating chemotherapeutic agents due to the active status of the DNA repair protein O6-alkylguanine-DNA alkyltransferase (ATase). This protein recognises and repairs the alkylated lesions formed at the O6 position of guanine bases in DNA, through which the cytotoxic effect would otherwise be derived. Hitherto, ATase inhibitors have been designed to covalently modify the active site of the protein, e.g. benzylguanine. Herein we present the design, synthesis and preliminary evaluation of a different kind of substrate analogue that may interact non-covalently with the ATase protein.
Careful examination of X-ray crystal structures of the active site of the human form of the repair protein -in its native form and covalently modified by known inactivators of the protein -coupled with consideration of the reaction mechanism of the repair process has allowed the design of guanine derivatives which might bind in the active site by electrostatic and hydrogen bonding interactions. Furthermore, these novel derivatives bear acidic or basic functional groups which might interfere with catalysis of the DNA repair reaction. Such substrate analogues may be useful probes for crystallographic studies of purine-protein complexes and have application to the potentiation of cytotoxic chemotherapy. Details of the design and synthesis of novel 6-substituted purines will be presented, accompanied by preliminary data on their ability to bind and inhibit the ATase protein.  Rev Pharmacol Toxicol. 2001;41:297-316). As part of the target validation process CYP1B1 protein expression has been characterised in clinical head and neck squamous cell carcinomas (HNSCCs) and in a series of human tumour xenografts from low passage head and neck cancer cell lines. The latter to be evaluated as a clinical model to test a novel CYP1B1/prodrug combination designed to target the delivery of the radiosensitiser/cytotoxin nitric oxide to tumours. Immunhistochemistry was performed using a monoclonal antibody specific for CYP1B1 which did not react with other cytochrome P450s including CYP1A1 and CYP1A2 (McFadyen et al., J Histochem Cytochem. 1999;47:1457-1464. HNSCCs taken from a range of anatomical sites were of variable grade (n = 23). The intensity of immunoreactivity was assessed on a semi-quantitative scale as strong (3+), moderate (2+), weak (1+), negative (-). The majority of tumours showed positive immunoreactivity for CYP1B1 but both inter-and intra-patient variability in CYP1B1 protein expression was evident. Only three (13%) of tumours did not exhibit any immunoreactivity for CYP1B1, two tumours (9%) showed weak CYP1B1 immunoreactivity, six (26%) tumours displayed moderate immunoreactivity while twelve (52%) showed strong CYP1B1 immunoreactivity. In each case CYP1B1 expression was localised in the cytoplasm of the tumour cells and was absent from the associated normal stromal tissue. Spectral image analysis technology developed at the Gray Cancer Institute, which exploits the characteristic absorptions of individual stains (to distinguish stained from non-stained areas or co-localisation of stains), was used to confirm tumourspecific localization of CYP1B1. Similar patterns of expression have been observed previously in a range of malignancies and emphasizes the importance of CYP1B1 as a target for cancer therapeutics. Previously, the inability to identify suitable experimental tumour models, which constitutively express CYP1B1 (with the frequency and intensity observed in clinical tumours), has proven a hinderance to understanding the underlying mechanisms of CYP1B1 regulation. In this study human tumour xenografts from low passage head and neck cell lines (including the UT16A, UT24A, UT15, UT33) have been found to contain the CYP1B1 phenotype. The UT16A and UT24 xenografts in particular exhibited strong CYP1B1 protein expression compared to many human tumour xenografts derived from established cell lines.

P239
In conclusion, HNSCC expresses the CYP1B1 protein with high frequency (n = 23, 90%). A suitable in vivo model for studying both CYP1B1 regulation and testing CYP1B1-activated cancer therapeutics has been identified. Introduction: New treatments are needed for patients with hormone refractory prostate cancer. We have investigated its effectiveness of the CD/5-FC suicide gene therapy system in a rat prostate cancer model. Method: PA3 cells were stablely transfected with cytosine deaminase using a doublecopy recombinant retrovirus under the control of the ERBB2 promotor. Transfection was confirmed using immunohistochemistry. Cell cytotoxicity with 5-FC was assessed using the MTS assay. Apoptosis was assessed using the commercially available Apoptag kit. Bystander effect was assessed by plating cells at different concentrations of wild type and transfected cells. Cell viability was assessed using the MTS assay. Results: Cytosine deaminase expression was confirmed by immunohistochemistry. Six days treatment with 10ug/ml 5-FC resulted in 40% cell death in the PCD cells compared to only 5% on the wild-type. Apoptosis was shown in 30% of PCD cells compared to 0% in PA3-WT after 6 days treatment with 5-FC. Conclusions: In our studies cell death by apoptosis can be shown in transfected cells in vitro. We have shown a significant bystander effect in vitro and are currently using this system in an animal model.

P242 CYTOSOLIC HUMAN NOS II ACTIVITY IN MDA 231 BREAST TUMOUR TRANSFECTED CELLS ENHANCES SR4233 ACTIVATION AND TOXICITY.
Edwin Chinje*, Jian Feng, Sanjeev Paul Sharma, Natasha Wind, Rachel Cowen and Ian Stratford. Experimental Oncology Group, School of Pharmacy, University of Manchester, Oxford Road, Manchester, M13 9PL, UK.
SR4233 (tirapazamine, TPZ) is a lead bioreductive drug currently in clinical trials. It is selectively toxic to hypoxic cells commonly found in solid tumours and toxicity results from the intracellular metabolism of the drug to a highly toxic radical under hypoxia. Nitric oxide synthase (NOS) is an endogenously up-regulated enzyme that could provide tumour selective metabolism of chemotherapeutic prodrugs. Significantly, elevated levels have been found in several neoplastic tissues. Its dimeric nature provides two distinct catalytic domains (a reductase and an oxygenase) that will bioactivate a broad repertoire of chemotherapeutic agents including TPZ. The enzyme is a haem-based protein similar to members of the cytochrome P450 family and the reductase domain shares a high degree of sequence homology with cytochrome P450 reductase (P450R). However, unlike the former two enzymes, NOS II (inducible NOS) is soluble and exclusively expressed within the cytoplasm. Since, some reports have suggested that only metabolism of TPZ in the nucleus is likely to contribute to DNA damage [Evans JW et al., 1998, Cancer Res. 58: 2098-2101, we have exploited this differential cellular localization to evaluate the contributions of cytosolic metabolism of TPZ to its overall toxicity. To achieve this, we have constitutively over-expressed NOS II by employing an optimised expression vector in which the human elongation factor 1a promoter produces a bicistronic message containing the genes for human NOS II and puromycin resistance. Following successful transfection of the human breast cell line, MDA231, clones have been selected that stably express varying levels NOS II activity. NOS reductase activity as measured by the NADPH-dependent reduction of cytochrome c ranged from 7.5-22.0 nmol cyt. c reduced/min/mg. The highest activity represented about 6-fold increase over parental activity. Similarly, the catalytic activity of the oxygenase domain, measured by the conversion of 14 C-labelled L-arginine to L-citrulline, ranged from 20-66 pmol citrulline / min/mg and the highest activity was about 110-fold over parental cell activity. NADPH-dependent metabolism of TPZ to the stable two-electron metabolite, SR4317, was determined by HPLC analysis. Rate of metabolism was enhanced by the increase in NOS II expression. In addition, direct detection of the TPZ radical formation by the dihydrorhodamine (DHR) 123 assay, correlated with HPLC metabolism data (r 2 = 0.92). DT-diaphorase activity (can metabolise TPZ in a single 2-electron step by-passing the free radical) did not alter as a result of the clonal process. Sensitivity to TPZ following 3 hr hypoxic exposure also increased with elevated NOS II levels.
Taken collectively, our data shows that cytosolic NOS II may contribute significantly to the bioactivation of tirapazamine. Background The development of cisplatin resistance has seen the need for novel anti-cancer agents that can overcome this resistance. BBR 3464 is the first trinuclear platinum drug to enter clinical trials. It differs from cisplatin in its length and its +4 charge. These differences are believed to enable BBR 3464 to form adducts markedly different in structure to those formed by cisplatin Methods Purified DNA was reacted with BBR 3464 or cisplatin to give an adduct frequency of 1 per kilobase. DNA was also extracted from drug treated human ovarian A2780 cells. Purified DNA was digested using nuclease P1 and DNase 1 and analysed by ionexchange chromatography with elution using NaCl concentration gradients. BBR 3464-DNA was also digested with a higher concentration of the above enzymes (enhanced digestion). Platinum content in fractions were analysed using inductively coupled plasma mass spectrometry (ICP-MS). Purified DNA Samples were analysed using a sensitive competitive ELISA procedure (1) that detects low levels of adducts formed by cisplatin. Results Analysis of cisplatin-DNA adducts using the Mono Q anionexchange column gave the expected pattern of platinum containing peaks (2). None of the BBR 3464 derived material from enhanced digestion was retained on this column. Analysis using the Macro-prep CM column revealed two main peaks of BBR 3464 derived material that eluted at 6 minutes (1.5M NaCl) and 10.5 minutes (2.6M NaCl). BBR 3464 adducts showed markedly lower immunoreactivity than cisplatin adducts (> 2000 fold). Three to four times the levels of adducts were detected on extracted DNA treated with BBR 3464 as compared to cisplatin at equivalent drug doses. Conclusion The chromatographic behaviour and immunological properties of DNA adducts formed by BBR 3464 indicate markedly different properties compared to cisplatin adducts. The enhanced digestion needed for BBR 3464-DNA samples may indicate that BBR 3464 is distorting the DNA and restricting access of digestive enzymes. It is anticipated that the strong affinity of BBR 3464 adducts to the cation-exchange column will permit maximum sensitivity for quantification by ICP-MS of adducts in DNA from blood cells collected during a recent phase II trial. Attenuated salmonellae have been extensively studied as live vectors for delivery of heterologous antigens. They are safe, highly immunogenic and can elicit long-lasting protective systemic and mucosal immunity. Oral immunization with S. typhimurium harboring plasmid DNA has resulted in gene delivery to gut lymph nodes, and induction of immunity to antigens encoded by the plasmid. There are several types of Salmonella mutants currently used for gene delivery. The most common, harbour mutations in aromatic amino acids or purin biosynthesis pathway (e.g. aroA, aroC, aroD and purA, purE), deficient production of adenylate cyclase (cya) or cyclic AMP receptror protein (crp) or mutations affecting the global regulatory system (phoP/phoQ). We are evaluating two new S. thyphimurium mutants, SPI2-sifA and ssav, as gene therapy vectors. In vitro gene transfer experiments using the plasmid pIRES2, which contains the Green Fluorescent Protein (GFP) reporter gene cDNA under a eukaryotic promoter have been completed in mouse macrophages and human monocyte cell lines-J7742 and U937 respectively. The expression of GFP using fluorescent microscope, was detected in 2% of cells. The same protocol was applied to primary mouse macrophages where expression of GFP was seen in up to 60% of cells. We plan to use SPI2 mutants as gene therapy vectors using an erbB2 DNA vaccine in mouse mammary tumour model. We have cloned truncated versions of the erbB2 genes representing peptide epitopes for the intracelleular and extracellualr domains of the erbB2 oncoprotein. We will use S. typhimurium to deliver these plasmid constructs as DNA vaccines by the intraperitoneal and oral routes, and compare the efficacy and immunogenicity of this approach with intramuscular delivery of the same constructs. All in vivo preclinical studies comply with UKCCR " Guidelines for the Welfare of Animals in Experimental Neoplasia."

IN VITRO ANTI-TUMOUR AND ANTI-ANGIOGENIC ACTIVITY OF AW464 (NSC 706704), A NOVEL QUINOL
A Mukherjee* 1 , SG Martin 1 , TD Bradshaw 2 , AD Westwell 2 , MFG Stevens 2 , J Carmichael 1 Cancer Research UK Academic Unit of Oncology 1 , University of Nottingham, Nottingham, UK, Cancer Research Laboratories 2 , University of Nottingham, Nottingham, UK Introduction/Aims: AW464 (NSC 706704) is a novel benzothiazole substituted quinol compound with potent and selective activity concentrated in certain colon and renal cancer cell lines. The mechanism of action of AW464 however is not well understood. COMPARE analysis suggested possible interaction with thioredoxin/ thioredoxin reductase signalling. It is unknown whether it is likely to have any significant anti-angiogenic activity. In this study, the effects of AW464 were compared with those of Paclitaxel, which in addition to having direct tumour cell toxicity, also has a proven antiangiogenic action (Belotti D et al, 1996).

Materials and Methods:
In vitro experiments were performed on HT29, SW620 and SW480 colorectal cancer cell lines and on human umbilical vein endothelial cells (HUVEC). The effects of both drugs on cell growth were assessed by the MTS assay and growth curves. Effects on HUVEC differentiation were assessed by the Matrigel assay. In some experiments, HUVEC were exposed to tumour conditioned media to mimic the tumour environment. Results: The MTS assay and growth curve experiments demonstrated that AW464 exerted anti-proliferative effects on tumour cell lines as well as HUVEC at concentrations between 10 -5 M and 10 -7 M. This activity was comparable with Paclitaxel. HUVEC proliferation was significantly inhibited by exposure to tumour conditioned medium alone. Addition of AW464 further suppressed proliferation slightly. Using the Matrigel assay, endothelial tube formation was partly inhibited within 24-48 hours by AW464 although this effect was less pronounced than with Paclitaxel. We conclude that AW464 may have an effect on endothelial cell proliferation and differentiation in addition to its tumour cell toxicity. Ref. Belotti D et al (1996)  Cancer treatment is the main application of gene therapy, initially devised to correct inherited monogenic disorders. Gene directed enzyme prodrug therapy (GDEPT) involves the transfer of a gene encoding an enzyme to cells, and the subsequent administration of a prodrug, which is converted to a cytotoxin by the enzyme. Several GDEPT systems have been devised, including HSV thymidine kinase/gancyclovir and cytosine deaminase/5fluorocytosine. A recently developed system is the horseradish peroxidase/indole-3-acetic acid (HRP/IAA) combination (Greco et al., 2000, Cancer Gene Ther. 7, 1414. When expressed in mammalian cells in vitro, the plant enzyme HRP is able to convert IAA, a plant hormone, to a cytotoxin. In order to evaluate this combination further, we utilised the nasopharyngeal squamous cell carcinoma cell line FaDu, transfected with the HRP gene, in tissue culture, in tumour spheroids and in a xenograft tumour model. We show that HRP expression is able to selectively sensitise FaDu cell monolayers to IAA and 2 of its derivatives, 5bromoindole-3-acetic acid and 1methylindole acetic acid, after 4 and 24 hour exposure. 5bromoindole-3acetic acid was active at lower concentrations than IAA, with 2-log cell kill at 1 mM in HRP transfectants. As an intermediate between in vitro monolayers and in vivo experimental tumours, a 3-dimensional multitumour cell spheroidal model has been developed. IAA and its derivatives were also active in this more complex in vitro system. Xenografted tumours grown from HRP-transfected FaDu cells were shown to retain enzyme activity in vivo. Initial pharmacokinetic studies with a single bolus intraperitoneal injection of IAA indicate first order kinetics. With a non-toxic 250 mg/kg dose, plasma, liver and skeletal muscle concentrations peaked 10 minutes after administration. In the tumour, concentrations reached 1 mM, and remained above 0.5 mM for 50 minutes. These data provide promise for the ongoing evaluation of the HRP/IAA combination for GDEPT. This work was funded by Cancer Research UK. Immunotherapy as a treatment modality for hormone-refractory prostate cancer (HRPC) has shown promise in recent trials using whole cell vaccines and dendritic cell (DC) approaches. However, DC-based strategies have been limited by the need for repeated venesection or leukapheresis to provide sequential supplies of DCs for modification and vaccination. We have developed a method of culturing then cryopreserving monocyte-derived DCs after pulsing with irradiated allogeneic prostate cancer cell lysate. The freezethaw process resulted in minimal loss of cells, increased DC maturation and thus enhanced antigen-presenting capability. This method has provided numerous identical aliquots of DCs; one aliquot is thawed, tested for phenotype, function and microbial contamination prior to administration. In this way we ensured patients received identical sequential vaccines. We have completed a phase I/II, local research ethics committee approved clinical trial of 14 patients with metastatic HRPC who received 1 x 10 6 DCs intradermally or intranodally at 2 weekly intervals for 3 months then monthly. Keyhole limpet haemocyanin (KLH) was used as the adjuvant. This protocol was safe and non-toxic with only 2 reports of slight fever. Initial results indicate a PSA response in 4 patients, 2 with an arrest in PSA levels over 4 months and 2 showing reduction in PSA velocity over the same time period. All patients showed a positive delayed-type hypersensitivity test (DTH) prior to commencing treatment confirming their immunocompetence. We intend to assess response to vaccine clinically, radiologically, DTH response to intradermal injection of the tumour lysate, ELISPOT, cell-specific cytokine production using CBA array and proliferative lymphocyte responses to vaccine DCs or recall antigens. Flt3 ligand is a novel haematopoietic growth factor. Its administration has been shown to increase dendritic cell and natural killer cell numbers and activity in vivo by activating and expanding Dendritic Cell and Natural Killer cell progenitors. Previous studies used the recombinant protein have shown increased tumour rejection and survival in melanoma and lymphoma. However, this is of limited availability and high cost. The use of Polynucleotide DNA using cytokine cDNA cloned into eukaryotic expression vectors has shown efficacy as a vaccine in infectious disease and cancer models. Plasmid DNA expressing the extracellular domain of murine flt3 ligand was produced using a standard DNA extraction kit. Its expression was confirmed by COS cells transfection. We studied the effect of vaccinating with flt3 ligand alone, or in combination with the melanoma antigens TRP1 and TRP2. The cytokine GMCSF was also included in the vaccination protocol. We were unable to reproduce the results demonstrated using the recombinant flt3 ligand with no significant survival or immune responces. Combination of flt3 ligand with antigen did confer a slight protection. GMCSF did produce a survival advantage that appeared independent of tumour antigen. We are unaware of any published studies using flt3 ligand as a DNA vaccine prior to this. All in vivo work was carried out in strict accordance with Home Office and Institutional Guidelines Syngeneic tumour cell vaccines are difficult to produce, due to the inherent problems associated with bespoke vaccine production. Stable, well characterised allogeneic tumour cell vaccines containing cross-reactive tumour antigens circumvent these problems, and are an attractive proposition. Using the B16-F10 murine melanoma system, we have shown vaccination with irradiated, allogeneic tumour cells to protect from challenge with live, syngeneic tumour. The introduction of cytokine encoding sequences into syngeneic vaccine cells has been shown to increase vaccine efficacy by several groups (Dranoff, Jaffee, Lazenby et al., 1993, PNAS USA, 90:3539). In this study, we introduced the following murine cytokine sequences into B16-F10 cells using the pLXIN (Clontech) retroviral system; GM-CSF, IL-2, IL-4, IL-7. Single clones were selected, rendered replication incompetent by irradiation, and used to immunise both syngeneic (C57BL/6j) and allogeneic (C3H/HeN) female mice sub cutaneously. Following 2 doses of vaccine, mice were challenged on the opposing flank with a tumourigenic dose of syngeneic tumour. Immunisation with allogeneic vaccine expressing cytokine did not confer any real survival benefit when compared to the wild type vaccine. However, transfection with IL-2 did produce a small, but significant, increase in survival time. In contrast, only IL-4 transfection increased the efficacy of wild type vaccine within the syngeneic mice. Splenocytes from allogeneic immunised mice were able to produce significant levels of CTL killing specific for the syngeneic tumour challenge cells. In addition, cytokines such as TNF-alpha, IFN-gamma, IL-2 and IL-5 were also induced by these splenocytes following in vitro incubation with both syngeneic whole tumour cells and lysate. Thus, vaccination with allogeneic cells induces a cellular immune response which can cross-react and protect against challenge with syngeneic tumour. The potency of the host's anti-allogeneic response appears to be such that the introduction of additional cytokine genes into the vaccine does not readily improve vaccine efficacy. The aim of this study is to validate the use of radiolabelled thymidine ([ 3 H]TdR) uptake as a measure of TS inhibition. The working hypothesis is that TS inhibition causes a depletion in thymidine monophosphate and to overcome this blockade, cells undergoing DNA synthesis will increase the amount of labelled thymidine taken up via the salvage pathway. RIF-1 cells growing in culture showed an increase in [ 3 H]TdR uptake at 0, 2 and 6 h after a 2-h treatment with 5-fluorouracil (5-FU; 1-100 mg/ml). A decrease in [ 3 H]TdR uptake was observed at 24 and 48 h. Both effects were dose dependent but whereas the maximum increase in [ 3 H]TdR was only 25 ± 2% of controls, the decrease in [ 3 H]TdR uptake at 48 h for instance was >95% at high concentrations. A similar effect was seen in HT29 cells (slower growing) treated with 5-FU, although the decrease in [ 3 H]TdR at the latter time points was less marked. In contrast, treatment of the cells with cisplatin (non-TS inhibitor; negative control) led to a decrease in [ 3 H]TdR uptake at 2, 6, 24 and 48 h. To further confirm that the early increase in [ 3 H]TdR uptake was dependent on TS inhibition, similar studies with the non-classical TS inhibitor AG337 were performed. In RIF-1 cells, AG337 treatment (0.05-5 mg/ml; continuous exposure) resulted in a large increase in [ 3 H]TdR uptake (up to 152 ± 5% of controls at 2 h). The increase in [ 3 H]TdR peaked at 2 h, was still high at 6 h but as with 5-FU, decreased at 24 and 48 h. In all cases MTT assays were performed in parallel with the [ 3 H]TdR uptake studies. These studies showed that the early increase in [ 3 H]TdR uptake observed was not due to an increase in viable cell number. Although a decrease in formazan production was observed in all cases where there was a decrease in [ 3 H]TdR uptake, this was less pronounced compared to the decrease in [ 3 H]TdR uptake. This implies that the decrease in [ 3 H]TdR uptake at the latter time points could be explained in part by a decrease in cell number. The above studies support the clinical use of [ 11 C]thymidine as a noninvasive positron emission tomography probe for imaging TS inhibition in patients and indicate the appropriate timing for such studies. Future experiments will assess whether the changes in [ 3 H]TdR uptake correlate with direct measures of TS inhibition.

Introduction:
The recombinant human monoclonal antibody trastuzumab targets cellular proliferation by blocking HER2 signaling. However the majority of patients who overexpress HER2 fail to respond to single agent trastuzumab. Farnesyltransferase regulates the post-translational modification of a number of important proteins including the cell signaling protein ras. Farnesyltransferase inhibition can thereby block ras signaling. Concurrent farnesyltransferase and HER2 blockade result in synergistic antitumor activity in HER2 positive cell lines. R115777 is a potent, selective and nonpeptidomimetic inhibitor of farnesyltransferase that can inhibit the growth of H-, K-, and N-ras transformed as well as wild-type ras xenograft tumors. It has single agent clinical activity in breast cancer. Methods: A phase I study in patients with HER2 positive disease of oral R115777 and trastuzumab was pursued. Trastuzumab doses were fixed at 4 mg/kg IV day 1, then 2 mg/kg IV weekly. R115777 dose levels studied were 200 mg, 300 mg, and 400-mg po bid for 21-days every 28 days. Results: Twenty-one patients (8 breast, 5 NSCLC, 4 colon, and 1 each of ovary, pancreas and leiomyosarcoma) have received 78 courses (range: 1-13; median: 4) of R115777 and trastuzumab. Protracted grade 4 neutropenia lasting more than 5 days, fever, and grade 3 thrombocytopenia were dose limiting at an R115777 dose of 400 mg bid. Non-hematologic toxicities have been mild and include nausea/vomiting), headache and fatigue. No clinically significant decrements in LV ejection fraction have been observed. One patient with colorectal cancer has had a partial response. Overall 4 patients have remained on study for more than 8 months (2 colorectal, 1 breast, 1 leiomyosarcoma). Surrogate studies in peripheral blood mononuclear cells have demonstrated inhibition of hDJ chaperone protein farnesylation at all R115777 doses evaluated. R115777 and trastuzumab can be safely administered at full doses for both agents. Conclusion: This combined signal transduction inhibition approach warrants further investigation and a Phase II evaluation of this combination in HER2 positive breast cancer patients is now underway. Introduction: Epothilone B is a non-taxane microtubule-targeting agent that induces tubulin polymerization and stabilizes microtubules. Similar to paclitaxel epothilone B blocks the cell cycle at the G2-M transition, thereby inducing apoptosis. Epothilone B and its lactam analogue BMS-247550 are active against paclitaxel-resistant tumor cells. Methods: Since weekly administration of other anti-microtubule agents has afforded a favorable toxicity profile BMS-247550 was administered continuously once weekly by a 60-minute intravenous infusion utilizing a 28-day regimen. Toxicity endpoints were sought separately in heavily (HP) and minimally pretreated (MP) patients. Results: Twenty patients (M:F=9:11) with a median age of 52 years and a median ECOG PS of 1 (6 colorectal; 3 ovarian; 3 NSCLC; 2 each of prostate, melanoma and breast; and 1 each uterus and cervix) have received 49 four-week courses (range 1 to 5; median 2) of BMS-247550. Dose-limiting toxicities include grade 3 fatigue in 1 HP pt at 20 mg/m 2 , protracted grade 4 neutropenia (>5 days) in 2 HP patients at 30 mg/m 2 and neutropenic sepsis in 1 MP patient at 30mg/m 2 . Neutropenia also precluded weekly treatment on-time in 2 of 3 MP patients at 30 mg/m 2 during course 1. One patient with hormone-, mitoxantrone-and paclitaxel-refractory prostate carcinoma had a greater than 50% PSA decrement (from 5100 to 810) with complete resolution of his tumor related bony pain. This patient developed grade 3 peripheral neuropathy and grade 2 fluid retention necessitating his withdrawal from the study after 6 months of treatment. A minor response was observed in a patient with melanoma, and tumor marker decrements were noted in 3 patients with taxane-refractory ovarian carcinoma. Neuropathy, assessed clinically, by nylon monofilament testing and nerve conduction studies, was observed in 6 patients (NCI-CTC grade 1 or 2 in 5; grade 3 in 1) and seen mainly at doses above 20 mg/m 2 with repeated dosing. All other drug-related toxicities have been mild to moderate (£ grade 2) including anemia, fatigue, myalgias, arthralgias, alopecia, anorexia, nausea, diarrhea, vomiting and thrombocytopenia. PK studies indicate dose-proportional kinetics (Cmax and AUC(0-Inf)) over the dose-range evaluated. Estimated PK parameters: t1/2=50.4±42.7 hour; Vss=795.4±331.5 L/m 2 ; Cl=23.9±15.2 L/hr/m 2 . The erythromycin breath test did not predict drug clearance. Accrual continues at 25 mg/m 2 /wk (HP) and 30 mg/m 2 /wk (MP) to define safe dose schedules for MP and HP patients. Conclusion: This weekly schedule of BMS247,550 has clinical anticancer activity and warrants broad diseaseoriented testing in phase II studies Cytochrome P450 CYP1B1 is a member of a superfamily of haemoproteins which are central to the oxidative metabolism of a wide variety of endogenous and exogenous compounds. Several of these enzymes have an established role in the metabolic bio-transformation of a variety of anticancer drugs. We have previously shown that CYP1B1 is a tumour-specific form of cytochrome P450 highly expressed in a variety of malignant tumours localised specifically to tumour cells 1 . In contrast CYP1B1 protein is undetectable in corresponding normal tissue. Recently we have identified several anti-cancer drugs that are substrates for CYP1B1 2 . Moreover, our in vitro studies have shown that the presence of CYP1B1 reduces the efficacy of some of these neoplastic agents 3 . Cytochrome P450 enzymes require the presence of cytochrome P450 reductase (CPR) for optimal metabolic activity. Since CYP1B1 metabolises anti-cancer drugs in tumour cells it is important to determine the level of active CYP1B1 and CPR in tumours. In this study we demonstrated both CYP1B1 and CPR activity in the microsomal fraction of ovarian (n=9) and kidney tumours (n=24). The Ortho-deethylation of ethoxyresorufin to resorufin was used to measure CYP1B1 activity. We have found CYP1B1 activity in the 100,00g fraction in both ovarian and kidney tumours (100-800 fmol/min/mg of protein). Co-incubation with the CYP1B1 inhibitor alpha-naphthoflavone inhibited this activity. No CYP1B1 activity was observed in any of the normal kidney samples examined (n=8). CPR activity was determined spectrophotometrically by measuring the reduction of cytochrome c at 550nM. CPR activity was detected in all normal and tumour samples (0.04-3nmol/min/mg protein). The presence of CYP1B1 which is capable of metabolising anti-cancer drugs and which is active only in tumour cells highlights a novel target for chemotherapeutic intervention. Pharmacokinetic analysis is often a bottleneck in the drug discovery process. The aim of this study was to evaluate cassette dosing, the simultaneous administration of several compounds to a single animal, and in vitro metabolic stability screening as methods to predict the pharmacokinetics of a series of novel DNA-dependent protein kinase (DNA-PK) inhibitors in high throughput. DNA-PK is involved in the repair of double strand breaks, thus inhibition of this enzyme may potentiate the effects of radiotherapy or chemotherapy by DNA-damaging agents. NU7053, NU7059, NU7062, NU7119 and NU7163 were administered intravenously to Balb Cmice alone at 5mg/kg and in combination at 1mg/kg each. Mouse liver microsomal incubations of the five compounds were carried out at a concentration of 10mM for 20 minutes. Plasma and microsomal concentrations were measured by HPLC-MS/MS with multiple reaction monitoring. Pharmacokinetic parameters were evaluated by non-compartmental analysis. The compounds displayed a linear increase in maximum concentration (Cmax) and area under the curve (AUC) with a 5-fold increase in dose from 1mg/kg cassette administration to 5mg/kg single administration. The clearance and half-lives of the compounds were similar following cassette and single dosing. For example, the clearance and half-life of NU7059 following cassette and single dosing was 0.091L/hr and 0.17hr, and 0.081L/hr and 0.16hr respectively. The compound which showed the lowest clearance (0.091L/hr), NU7059, also showed the lowest extent of metabolism (28%). Conversely, NU7053 displayed the highest clearance (0.46L/hr) and metabolism (98%) of the five compounds. The rank order of the compounds from lowest to highest clearance was similar whether they were dosed as single agents (NU7059 > NU7163 > NU7119 > NU7062 > NU7053) or in combination (NU7059 > NU7119 > NU7163 > NU7062 > NU7053). This ranking was maintained in terms of metabolic stability (NU7059 > NU7119 > NU7163 > NU7062 > NU7053). In conclusion, both in vivo cassette dosing and in vitro metabolic stability screening are suitable methods to evaluate the pharmacokinetics of novel DNA-PK inhibitors. These methods will be used to assess a larger number of compounds from this series in order to increase throughput whilst reducing the number of animals used.

P256 STRUCTURES OF ATYPICAL PLATINUM-DNA ADDUCTS IN CERTAIN RESISTANT LUNG CANCER LINES
A Azim-Araghi* 1 , CJ Ottley 2 , DG Pearson 2 , MJ Tilby 1 . 1 Cancer Research Unit, The Medical School, University of Newcastle, Newcastle upon Tyne, NE2 4HH, UK, 2 Department of Geological Sciences, Durham University, Durham, UK Aims Cisplatin-DNA adducts, formed in inherently drug resistant adenocarcinoma lung cancer cell line (Mor), and drug sensitive small cell lung cancer line (H69) appear to be markedly different in structure as indicated by differences in immunoreactivity using CP9/19 monoclonal antibody (1). The antibody does not recognise Pt-pApG adducts and the initial aim of the present work was to test the hypothesis that the difference between the cell lines was in the ratio of major DNA adducts. Methods The Pt-DNA adducts present in DNA extracted from the two cell lines exposed to cisplatin were analysed, using anion exchange chromatography (Mono Q), after enzymatic digestion of the Pt-DNA (2). Fractions were collected every 30 seconds, and the total Pt in each fraction was determined, using the sensitive Inductively Coupled Plasma Mass Spectrometry (ICP-MS) technique. This permits analysis of DNA containing relatively low adduct levels (3).

Results
The high sensitivity of ICP-MS permitted, for the first time, direct analysis of Pt in chromatography fractions from the analysis of DNA extracted from drug treated cells. The ratios of Pt-pGpG to Pt-pApG crosslinks were similar in both lines and were consistent with established data (2). However, additional Pt containing peaks were detected compared to those resulting from analysis of reaction products of cisplatin with purified DNA. Conclusion The difference in immunoreactivity, in Pt-DNA adducts of the two cell lines, was not due to differences in relative levels of the major adducts. The additional types of adducts formed in cells do not appear to account for the differences between the cell lines but these need further investigation. The base sequence, where the platinum binds to the DNA is currently being investigated, as these appear to affect the immunoreactivity of Pt-DNA adducts. 1) Tilby et al. 1991, Cancer Res. 51: 123. 2) Fichtinger-Schepman et al. 1985 The effect of the anti-vascular agent combretastatin A-4 phosphate (CA4P) on tumour biology is not fully understood. It has previously been reported that CA4P, a tubulin binding agent, causes vascular shutdown and necrosis in tumours in vivo. To further analyse the role of CA4P, we are using a Zeiss microscope with motorised stage and accompanying KS300 image analysis software, to investigate its effect on selected tumour parameters over time using a human colorectal xenograft in nude mice. These include vascular distribution (CD31), perfusion (Hoechst 33342), hypoxia (Pimonidazole), and cell proliferation (BrdU). Mice received either no treatment or a single IP dose of CA4P (200 mg/kg), the latter being sacrificed at either 1 or 24 hours after treatment. Quantitative analysis of relative changes in these parameters following therapy will be discussed. Control tumours: CD31 and Hoechst staining demonstrated perfused blood vessels throughout the whole of the tumour. Similarly, BrdU positive cells were also found in both central and peripheral parts of the tumour. In contrast, pimonidazole staining revealed some areas of hypoxia that tended to be distant from blood vessels and/or surrounding necrosis. Treated tumours: Vascular shutdown studies demonstrated that treatment with a single dose of CA4P resulted in almost complete vascular shutdown 1 hour after treatment. Simultaneous fluorescent staining of both Hoechst and CD31 revealed that although blood vessels were distributed throughout the whole of the tumour, perfused vessels were predominantly in the outer rim with only a few found in the central part of the tumour. Interestingly, BrdU staining revealed that proliferation corresponded to the well perfused outer rim, while only a few nests of cells surrounding blood vessels within the poorly perfused central part of the tumour stained positive for BrdU. These data indicate that a 1 hour treatment with a single dose of CA4P resulted in both vascular shutdown as well as a block in proliferation. Pimonidazole staining, on the other hand, revealed that at this time point, hypoxia was most commonly found in the tumour centre with little or no staining at the outer rim of the tumour. At 24 hours after treatment, when most of the tumour was necrotic, both perfusion and proliferation were restricted to the viable region at the edge of the tumour. Furthermore, pimonidazole staining was also seen in the viable rim, demonstrating hypoxic cells bordering regions of tumour necrosis. Taken together, the data in this study provide further insight into the mechanisms by which CA4P may exert its effects on tumour cells. Supported by the Association for International Cancer Research, Cancer Research UK, and Trusthouse Charitable Foundation. CA4P was kindly supplied by OXIGENE.

P258 COMPARISON OF THE CHEMOPREVENTIVE EFFICACY OF CURCUMIN AFTER LONG-OR SHORT-TERM ADMINISTRATION IN MIN/+ MICE.
Sarah Perkins*, William P Steward and Andreas Gescher. Department of Oncology, University of Leicester, Leicester LE1 9HN Curcumin, the major yellow pigment extracted from the spice turmeric, possesses a broad spectrum of chemopreventive efficacy. It has been shown to prevent carcinogen-induced malignancies and premalignancies in rodents. It is also active in the Min/+ mouse, a genetic model of human hereditary familial adenomatous polyposis. The aim of this study was to determine an efficacious dose of curcumin and an optimal time period of administration in the Min/+ mouse. We also determined curcumin levels by HPLC analysis commensurate with potential efficacy. C57BL/6J Min/+ mice were divided into 5 groups (n=10-14 in each). Control mice were kept on a high-protein (RM3) diet, to reflect western-style human diets. Mice received curcumin (0.2% in RM3) for the following time periods: i. perinatally and up to day 30, ii. from day 30 to day 75, iii. from day 75 to day 120, iv. from day 30 to day 120. At 120 days the study was terminated, and small intestinal adenomas were scored. To assess the effect of the high protein basal diet on its own, C57BL/6J Min/+ mice received RM3 or a standard AIN76A diet from day 30 to day 120. Curcumin administered from day 30 to day 120 significantly reduced small intestinal tumour formation by 39% compared to controls (p<0.05), whilst administration for shorter periods of time failed to alter tumour burden. The steady state level of curcumin in intestinal mucosa tissue of mice on 0.2% curcumin was 110±23 nmoles/g. Comparison of the basal diets showed that tumour burden is significantly lower when mice are fed a low-protein (AIN76A) diet, decreasing total intestinal adenoma number from 110±8 to 61±3.3 (p< 0.05; ± SEM). The results highlight the potential chemopreventive activities of a low-protein diet and of curcumin, and hint at the possible clinical relevance of curcumin as a colorectal cancer chemopreventive agent in humans. NU:UB 31 is a spacer-linked anthraquinone-L-proline conjugate which belongs to a series of compounds designed to inhibit DNA topoisomerase enzymes. Unlike most topoisomerase inhibitors in current clinical use, NU:UB 31 has been shown in vitro to be a dual inhibitor of DNA topoisomerase I and II 1 . The aim of this work was to study the in vitro and in vivo antitumour activity and pharmacokinetics of the compound. IC50 values for the compound calculated against MAC 15A (murine adenocarcinoma) cells using a MTT assay were 5, 8, and 38 mM for 1, 24 and 96-hour exposures respectively. Previous work has shown that treatment with NU:UB 31 at its maximum tolerated dose of 100 mg kg -1 led to a significant delay in tumour growth (P<0.01) 2 . The activity of doses below the MTD was assessed to look at the therapeutic range of the compound. NU:UB 31 was administered to NMRI mice bearing subcutaneous MAC 15A tumours i.p. (in DMSO/oil) at single doses of 67, and 33 mg kg -1 and tumours were measured daily. The growth delay for each dose was 4.3, and 3.2 days respectively. In both cases, treatment resulted in a statistically significant delay in tumour growth (p<0.01) compared to controls. A method was developed to detect and analyse NU:UB 31 in biological samples using LC-MS. Initial work assessed the stability of the compound in vitro with NU:UB 31 having t1/2 values of 75.1 and 25.7 hours in tissue culture medium and whole murine blood respectively at 37 o C. The compound was very stable in murine plasma with t1/2 >100 hours. Pharmacokinetic analysis of NU:UB 31 in vivo was assessed in MAC 15A tumour bearing mice. NU:UB 31 was given i.p. at 100mgkg -1 and plasma and tumour levels monitored over an 8 hour period. Drug concentration peaked in the plasma after 30 minutes (Cmax = 12 mg ml -1 ) with a t1/2 value of 2.2 hours and AUC of 44.5 mg h ml -1 . Peak tumour concentrations exceeded those seen in the plasma by approximately 3-fold peaking at 2 hours (Cmax = 39.5 mg g -1 ) with a t1/2 of 6 hours and AUC of 371 mg h g -1 (8-fold greater than plasma). IC50 values for the compound in vitro were converted to CxT values (taking into account the stability of the drug in tissue culture medium) giving values of 18.8, 82, and 176.2mg h ml -1 for 1, 24 and 96 hour exposures respectively. The concentration of NU:UB 31 measured in MAC 15A tumours in vivo therefore exceeds those of the in vitro IC50 values. The concentration of NU:UB 31 in tumours and the long half life may go some way to explaining the good anti-tumour activity seen against MAC 15A tumours.

Introduction
Ewing's sarcoma and primitive neuroectodermal tumour are members of the Ewing's family of tumours. This is consistently associated with chromosomal translocation and functional fusion of the EWS gene to transcription factor genes, such as Fli1 (95% of cases). The resulting chimaeric proteins are believed to contribute to tumour biology by aberrant regulation of gene expression altering controls of cell cycle regulation. These tumour-specific molecular rearrangements are useful as potential therapeutic targets. For this to be effective a detailed understanding of the difference between EWS-Fli1 protein at the level of gene expression is required. Methods Fli1 and EWS-Fli1 were cloned into the pET32a vector system. Expression of protein was achieved by growth at 37°C until an A600 of 0.3 was achieved, at which stage the bacteria were induced with 0.1mM IPTG and grown overnight at 25°C. Cells were harvested, sonicated and pelleted. The supernatant was collected as soluble protein.
Purification involved incubation of nickel NTA-agarose with protein sample for 3 hours. Sample was eluted with 1M Imidazole elution buffer. Sequencing was conducted with the use of ABI sequencing manual.

Results
We have demonstrated expression of Fli1 and EWS-Fli1 proteins by the use of western blot analysis. We have optimised soluble expression of the Fli1 and EWS-Fli1 proteins. 90% purity has been achieved, with the use of affinity chromatography. Following optimisation of purification, achieved by Ion exchange chromatography, future work would entail analysis of structural similarities between Fli1 and EWS-Fli1.

Conclusion
We have managed to achieve high expression of soluble Fli1 and EWS-Fli1. These proteins can therefore be used for detailed three dimensional structural analysis to elucidate differences in DNA:protein binding between the two proteins thus allowing the development of novel therapeutic strategies. Background The tumour suppressor p53 is a stress-inducible transcription factor which modulates the expression of genes involved in cell growth or apoptosis. This damage-inducible characteristic of the p53 pathway is linked to the death induced in tumour cells after being exposed to chemotherapeutic drugs or radiation. In order to improve the efficiency of anti-cancer drugs as single agents or in combination, p53 pathway modulation is an obvious choice. In this study we investigated the effects of the anthracycline doxorubicin and the vinca alkaloid vinorelbine on the induction and phosphorylation status of p53 in the breast cancer cell line MCF7. Materials and Methods MCF7 cells were grown in DMEM containing 10% FCS and 1% P/S at 37 o C and 5% CO2. At 70% confluence, medium containing 50 nM doxorubicin or 5 nM Vinorelbine as single agents or a combination of both drugs was added. After 4 and 24 h treatment the cells were harvested by scraping and lysates produced using urea buffer. p53 expression was determined using the antibody DO-12 (specific for p53 core domain). The phosphorylation status at Ser15, Thr18, Ser20 and Ser315 was investigated using phospho-specific monoclonal antibodies. Results Induction of p53 expression was observed after treatment with doxorubicin (single or combination) but not with vinorelbine. A key regulatory step in the control of p53 function implicates the mdm2 ubiquitindependent degradation pathway. Ser15 phosphorylation is thought to simultaneously reduce mdm2 binding (and subsequent p53 degradation) and promote interaction of p53 with components of the transcriptional machinery. Ser15 phosphorylation was observed after doxorubicin but not vinorelbine treatment. It has been shown that phosphorylation of p53 at Ser315 primes and stimulates its DNA binding function in vitro. Doxorubicin treatment but not vinorelbine resulted in phosphorylation of Ser315. However, when both drugs were added together a decrease in Ser315 phosphorylation was observed compared to the results obtained with addition of doxorubicin on its own. Conclusion This data suggest that chemotherapeutic drug interactions may affect doxorubicin anticancer activity on the p53 pathway. 85-90% of cancers express the enzyme telomerase which is contrastingly absent in benign tumours and most normal somatic cells: telomerase has therefore become a target for cancer treatment. Small molecules capable of stabilising the G-quadruplex structure thought to exist in telomeric repeat sequences, inhibit telomerase. RHPS4 inhibits 50% of telomerase activity at