ORAL PRESENTATIONS

Results: Do patients want to connect with others? One-third of participants reported not having someone to turn to about how to deal with a problem (34%) and not having someone to share their worries or fears with (30%). Two-thirds (62%) of participants wanted information about cancer-related peer support available to them.Who do patients want to connect with? Participants reported that peers who had received the same cancer treatment (56%) and were diagnosed with the same type of cancer (54%) were more important than connecting with those who were similar in age (42%), gender (38%) or culture (32%). How do patients want to connect with others? Peer support, which was provided one-to-one, was more preferable than a group-based format. Face-to-face modes of delivery were more preferred than telephone-based or web-based approaches. Factors associated with cancer patients’ preferences for involvement in peer support will be reported.


Introduction
The laboratory phenomenon of "low dose hyperradiosensitivity" (LDHRS) describes an excess of cell kill at doses < 1Gy relative to that predicted by the LQ model. These biological data have stimulated investigation into the existence of LDHRS in vivo. Methods Following ethical approval, 2 studies were initiated. Skin was used as a model of normal human tissue. Study 1: 24 patients receiving pelvic radiotherapy (RT) were assessed. Once daily doses of ~0.5Gy and >1.0Gy were compared. Biopsies were taken before & during RT & analysed to assess the changes in basal cell density (BCD). Study 2: 8 patients with metastatic tumour nodules to skin were evaluated (40 nodules). The nodules were randomised to receive conventionally fractionated RT (1.5Gy/day), or ultrafractionated RT (0.5Gy TDS). Both groups were treated for 12 days (18Gy). Measurements were made on days 0,5,8,12,26 & monthly until regrowth. Biopsies of surrounding skin were also taken to assess changes in BCD & proliferation. Results: 1:Skin (Study 1): 19 patients demonstrate a reduction in BCD in the low-dose side compared to the high-dose side when BCD is plotted against dose (significant in 14). Analysis of the whole data set demonstrates a significant reduction in basal cell density in the low dose group. If, however, BCD is plotted against time there is a significant reduction in BCD in the higher dose group. Study 2: When BCD is plotted against time, 7/8 patients demonstrate a greater reduction in BCD in favour of the conventional RT arm. A significant reverse fractionation effect was seen in only 1/8 patients. Proliferative response was similar in all patients irrespective of regime. 2: Tumour Nodules: To date 30 paired nodules have regrown. Analysis of the "radio-resistant" tumoursmelanoma & sarcoma -demonstrates significant growth delay in favour of the "ultrafractionated" regime (p = 0.002). Conclusions: There is no evidence of LDHRS in normal human skin when regimes of equal dose intensity are compared. LDHRS occurs in "radioresistant" tumours.
Tumours have an increased rate of glucose uptake relative to benign tissue,and tumour hypoxia leads to poor outcome. The increase in glucose uptake observed particularly in hypoxic tumours enables a compensatory switch to anaerobic glycolysis.This effect is mediated chiefly by the HIF-1 regulated glucose transporter Glut-1, which predicts poor prognosis in a wide variety of solid tumours, including cervix, lung and colorectal tumours. Glut-1 is also an intrinsic marker of hypoxia, correlating with PO 2 and pimonidazole adduct formation. The prognostic effects of Glut-1, and their relevance to tumour hypoxia, make Glut-1 an interesting target for novel anticancer strategies. For example, it has been found that certain flavonoid tyrosine kinase inhibitors e.g. quercetin, directly bind Glut-1, resulting in a sizeable decrease in glucose uptake that may induce cytotoxicity. In an effort to find possible lead drugs that may mediate their toxicity through Glut-1 binding and inhibition, the level of Glut-1 expression was investigated in the NCI panel of 60 cell lines, using immunohistochemical staining of microarrays containing samples of each tumour cell line. There was a wide variation of Glut-1staining (scored 0-5 according to staining intensity) between cell lines of the panel. Data is currently undergoing analysis, where any correlation between Glut-1 score and cytotoxicity of NCI compounds will be highlighted and investigated further. This work represents a departure from the traditional use of glucose analogues as glycolytic antimetabolites, and instead offers an alternative means of exploiting tumour hypoxia by directly targeting and binding to hypoxia-regulated proteins.
MNG719, is an indolequinone-based prodrug designed to undergo reductive fragmentation in hypoxia to liberate the anti-vascular agent combretastatin. The impact of MNG719 versus 'free' combretastatin on human umbilical vein endothelial cell (HUVEC) morphology and cytotoxicity were assessed. Target tubulin polymerisation was evaluated by a commercially available assay kit. Substrate specificity and inhibition of recombinant human DT-diaphorase was investigated and results interpreted by parallel molecular modelling. Metabolic studies were performed with HUVEC cell lysates and supersomal cytochrome P450 reductase. Proliferating endothelial cell area is unaffected by MNG719 in air exposed to doses of the prodrug up to 10 µmol dm -3 for 30 min whereas under anoxia cell shape reduction or 'shrinkage' is equivalent to doses of combretastatin of less than 1 µmol dm -3 . Combretastatin completely inhibits tubulin polymerisation at 0.5 µmol dm -3 but the prodrug is inactive even at a 40-fold higher concentration. Reductive metabolism of MNG719 by HUVEC cells and cytochrome P450 reductase under anoxia liberates combretastatin which is inhibited by air and is reflected in the comparative anoxia versus air cytotoxicities. Despite the 'bulkiness' of MNG719 the prodrug can dock in the active site of DT-diaphorase where reductive fragmentation results in enzyme inhibition. The results are consistent with MNG719 'redox-cycling' in air but undergoing reductive fragmentation under anoxia, mainly by cytochrome P450 reductase, to release combretastatin. Prodrugactivation is 'oxygen-sensitive' and future work will seek to modify the rate of reductive fragmentation to control hypoxia-selective drug delivery. Medulloblastoma (MB) is the most common malignant brain tumour of childhood, however the molecular pathogenesis of the majority of MBs is not understood. The epigenetic silencing of tumour suppressor gene (TSG) expression by promoter hypermethylation has emerged recently as an important mechanism of TSG inactivation. To assess the significance of aberrant promoter methylation events in MB, and to identify critical genes in its pathogenesis, we profiled the methylation status of eleven candidate TSGs (p14 ARF , p15 INK4b ,p16 INK4a ,Caspase 8,EDNRB,TP73,TSLC1, in MBs (11 cell lines, 44 primary tumours) and the normal cerebellum. Our data show that extensive hypermethylation of the RASSF1A promoter is a tumour-specific event, which occurs in the majority (~90%) of MBs. In MB cell lines, RASSF1A methylation was associated with epigenetic transcriptional silencing. Moreover, total RASSF1A methylation was observed in the absence of gene deletion or mutation, indicating that RASSF1A inactivation in MB occurs by bi-allelic epigenetic mechanisms. Increased tumourspecific methylation was also observed for HIC-1 (39%) and Caspase 8 (32%), in both cases occurring against a background of normal tissue-specific methylation in the cerebellum. These data demonstrate that aberrant promoter methylation events are a significant feature of MB development, and identify putative MB TSGs. Epigenetic RASSF1A inactivation represents the most common defect detected to date in MB, emphasising the importance of its further investigation in this disease. Raf-1 is a member of the Raf serine/threonine kinase family which consists of Raf-1 (c-Raf), A-Raf and B-Raf. It is a key intermediate in the Ras/Raf/MEK/ERK pathway which relays extracellar signals to the nucleus. Over 90% of ovarian carcinomas express Raf-1 protein, with high levels correlating with poor survival (McPhillips et al, Br J Cancer, 85, 1753,2001. In this study we have used a series of 15 ovarian cancer cell lines to help identify determinants of sensitivity to anti-Raf-1 antisense oliogonucleotides (ASOs). Growth inhibition (at 72h) varied from 100% down to 12% across this series of cell lines after treatment (200 nM for 3h) with either fully phosphorothioated ASO, ISIS 5132, (TCCCGCCTGTGACATGCATT) or a further (2'methoxyethyl) modified ASO, ISIS 13650. The degree of growth inhibition was not associated with level of Raf-1 protein expression, intracellular uptake of ASO or extent of Raf-1 protein inhibition after ASO treatment (61-99% inhibition). However, growth inhibition was significantly (p=0.019, Mann-Whitney) higher in cell lines with a higher proportion of Raf-1 mRNA (relative to total Raf mRNA i.e. Raf-1 + A-Raf + B-Raf) suggesting that greater utilisation of Raf-1 rather than other Raf isoforms was important. Consistent with this, growth inhibition was also significantly (p=0.04, Mann-Whitney) higher in cell lines which demonstrated greater (>2-fold) Raf-1 kinase activation after growth factor (1nM TGF α) stimulation. These results indicate that ovarian cancer cell lines demonstrate variable growth sensitivity to anti-Raf ASOs but sensitive tumors can feasibly be identified. Fibroblast growth factor-2 (FGF2) is a potent angiogenic factor, which requires heparan sulfate proteoglycans (HSPGs) for its biological activity. HSPGs mediate stable high affinity binding of FGF2 to its receptor tyrosine kinases (FR). The aim of this study was to investigate the biological activity of epithelial ovarian cancer (EOC) endothelial HS with respect to the activation of FGF2. We examined the ability of HSPGs to promote FGF2/FR binding using a soluble FR fusion construct (FR1-AP) (1). This encodes the extracellular domain of FR1 linked to alkaline phosphatase (AP A putative pathway for cervical glandular carcinogenesis exists from normal glands through cervical glandular intraepithelial neoplasia (CGIN) to invasive adenocarcinoma. Cyclins and CDK-inhibitors are known to play a crucial role in regulating cell cycle. We aimed to identify biological markers that define early cervical glandular neoplasia and to differentiate CGIN from their benign mimics. Paraffin-embedded sections of normal cervix (Group: 1, n=11) edometriosis/TEM (Group: 2, n=19), CGIN (Group: 3, n=33) and invasive adenocarcinoma (Group: 4, n= 28) were stained with p16, cyclins A, B, D and E. A high level of p16 expression was found in CGIN (mean: 81.4%) and invasive adenocarcinoma (84.4%) when compared with normal endocervix (2.3%) . Levels of cyclin A, cyclin B and p16 in CGIN were significantly higher than those in group 2 (P= 0.05, <0.001 and <0.001 respectively). A progressive increase in cyclin A and B expression occurred from normal cervix to invasive adenocarcinoma. The latter expressed higher levels of cyclin A and B when compared with CGIN (P<0.001). Cyclin E expression increased from group 1 to 2 (P=0.03) with no significant further increase. Changes in Cyclin D expression were unremarkable. Our data highlight that p16 could be used as a marker for cervical glandular neoplasia. Concurrent expression of cyclin A, cyclin B and p16 can help distinguish between CGIN and TEM/endometriosis.

Aims:
The steroid hormone vitamin D is thought to protect against breast cancer (BC). Vitamin D exerts its cellular effects by binding to the vitamin D receptor (VDR). The VDR has several polymorphic sites, resulting in differing genotypes that may alter susceptibility to BC. We have investigated whether specific polymorphisms in the VDR gene are associated with BC risk in a UK Caucasian population.
Methods: Female BC patients (n=411), and control women with a negative screening mammogram (n=427) were recruited, and their VDR genotypes were determined.
Results: The 3′ VDR polymorphism BsmI and a variable length Poly A sequence were both significantly associated with BC risk; odds ratio bb vs BB genotype = 1.74, (P=0.0096); odds ratio LL vs SS = 1.70, (P=0.0146). A 5′ VDR gene variant, FokI was not associated with BC risk when analysed in isolation (P=0.31). However, FokI modulated the increased risk associated with the bb/LL genotype such that possession of one or more F alleles together with the bb/LL genotype augmented the risk of BC. Furthermore, the highest proportion of bb (53%) and FFLL/FfLL (52%) genotypes occurred in women with distant metastatic BC, and 73% of established cell lines derived from BC metastases (n=15) displayed the bb/LL genotype.
Conclusions: VDR polymorphisms are associated with BC risk and progression. This research has implications for the monitoring and treatment of women at risk of BC, and for those likely to develop metastasic disease. Future research will investigate the effect of genotype on VDR function. In breast cancer, steroids in particular estrogen, functioning through its receptors (ER) contribute to tumour progression by regulating the transcription of target genes. Recent studies however suggest that estrogen may also upregulate the mitogen-activated protein kinase (MAPK) pathway, mimicking the effect of growth factors such as epidermal growth factor (EGF). This clinical observation suggests that estrogen may be able to mediate some of cell survival properties through an ER-independent mechanism.
The ER-positive MCF-7 and the ER-negative SKBR3 human breast cancer cell lines were incubated in the presence and absence of EGF and estrogen with or without G-protein coupled receptor (GPCR) inhibitor, pertussis toxin and EGF receptor inhibitor, AG 1478. Phospho-c-raf, phospho-Erk, phospho-cdc2 and survivin expression were detected using western blotting techniques.
In MCF-7 and SKBR3 human breast cancer cells, we have found that EGF and estrogen stimulation rapidly increased the phosphorylation of c-raf, Erk and subsequently an increase in cdc2 phosphorylation and an up-regulation of survivin expression. A further increase was noted in the presence of EGF in combination of estrogen. Moreover, estrogen induced a translocation of phospho-cdc-2 from the cytosol to the nucleus indicating activation of the phospho-cdc-2/survivin complex. The effects of EGF on phosphorylation of craf, Erk, cdc2 and survivin were attenuated by AG1478. The effects of estrogen on phosphorylation of c-raf, Erk, cdc2 and survivin were attenuated by pertussis toxin and to the lesser extent by AG1478.
These observations implicate estrogen in the MAPK cell survival mechanism and survivin anti-apoptotic pathway. Elucidation of ER-independent estrogen survival pathways may in part explain clinical observations of steroid therapy resistance in breast cancer.

HER2 OVEREXPRESSION CONFERS RELATIVE RESISTANCE TO ZOLEDRONIC ACID IN VITRO
Lisa M Pickering*, Jon P Coxon, Janine L Mansi, Kay W Colston St. George's Hospital Medical School, London. UK. SW17 0RE.
Her2 receptor overexpression confers an increased risk of breast cancer recurrence and is associated with relative resistance to many endocrine and chemotherapeutic agents. Bisphosphonates, particularly potent third generation compounds such as zoledronic acid (ZA), limit skeletal morbidity in patients with advanced breast cancer. We have previously shown that ZA induces apoptosis in her2 non-overexpressing breast cancer cell lines and limits their capacity to adhere to mineralised matrices. This study aimed to investigate whether her2 receptor overexpression attenuates sensitivity to these ZA-induced anti-cancer effects.
Adhesion was assessed by seeding viable cells onto extracellular matrices for 24 hours. Adherent cells were then washed, fixed, stained and counted. Induction of apoptosis was assessed by the cell death ELISA and MTS cell viability assays. Forced her2 receptor overexpression in the non-malignant breast epithelial cell line Hb4a resulted in a 2-fold increase in their capacity to adhere to mineralised matrix. Treatment with 1nM-50µM ZA for 24h resulted in a significant reduction in adhesion of two non-her2 receptor overexpressing cell lines cells, with a reduction in adhesion of 50-85% in MCF-7 cells and 35-80% in MDA-MB-231 cells. However the her2 receptor overexpressing SKBr3 cell line experienced only 16% reduction in adhesion with ZA treatment. Furthermore, whilst ZA induced apoptosis in all three cell lines, a significantly lower percentage of SKBr3 cells than either MCF-7 or MDA-MB-231 cells underwent apoptosis.
In conclusion, these results support the observation that her2 receptor overexpression is associated with increased metastatic potential and suggest that it confers relative resistance to ZA. ER-α and ER-β function as transcription factors to modulate genes relevant to breast cancer progression. Both interact with nuclear regulatory proteins to enhance or inhibit transcription. Following ethical approval ER-α, ER-β, the co-activator SRC-1, and the corepressor SMRT were localized within breast tissue by immunohistochemistry, and their spatial co-expression assessed by immunofluoresence.
The expression of ER-α, ER-β and the coregulators was correlated with clinicopathological features. The ability of β-estradiol and 4-hydroxytamoxifen (4-OHT) to modulate protein expression of ER-α , ER-β , SRC-1, and SMRT in primary breast cancer cell cultures was assessed by immuno-blotting. SRC-1 was found to be significantly associated with nodal positivity and resistance to endocrine treatment. There was up-regulation in ERβ and the co-regulator protein expression in the presence of βestradiol. 4-OHT induced an increase in ER-α and SMRT, whereas it decreased or unaltered the protein expression of ER-β and SRC-1. Associations between SRC-1 expression and recurrence suggest that co-activators may impact on the clinical phenotype of the tumour. The differential regulation of the ER-isoforms and the co-regulators by βestradiol and 4-OHT provides evidence that co-regulatory proteins may have a role in determining the response of breast cancer cells to estrogen and tamoxifen. IL-7 is known to induce the differentiation and proliferation of some leukaemias and lymphomas. Although it has been implicated in the process of lymphangiogenesis, Little is known about its role in solid tumours including breast cancer. We studied the expression of IL-7 and its receptor, IL-7R, in a cohort of patients with breast cancer.
Tumour (n=122) and normal (background) tissues (n=41) from patients with breast cancer (median follow up 72.2 months) were analysed for the level of mRNA of IL-7 and IL-7R, using quantitative RT-PCR, and protein using Western blotting and immunohistochemistry. The results were analysed according to tumour grade, nodal involvement, patients survival rate and prognosis (using the Nottingham Prognostic Index). The expression of IL-7 was higher in the cancer tissues compared to the background tissues. IL-7 was significantly higher in grade 3 tumours compared to grades 2 and 1 (P=0.03). Ductal carcinomas highly expressed IL-7 compared with lobular tumours (P=0.08). Furthermore, IL-7 expression was significantly higher in the node positive group compared to the node negative patients (P=0.03). Number of copies of IL-7 was higher in the worst prognosis group (NPI3) compared to the good prognosis group (NPI1) (P=0.004). Additionally, patients who died from breast cancer had significantly higher IL-7 expression in their tumours compared with the survivals (P=0.03). The expression of IL-7R was also significantly elevated in node positive tumours and in patients with NPI3. This is the first study to show that IL-7 expression is associated with tumour grade, nodal involvement and prognosis in breast cancer. Aberrant expression of IL-7 and IL7R in breast cancer may be involved in the progression and spread of breast cancer. A series of novel polycyclic acridines inhibit telomerase in the TRAP assay ( tel IC 50 ~ 0.3µM) and are highly promising potential new anti-cancer agents. The lead compound, RHPS4 (4,11-difluoro-6,8,13-trimethyl-8H-quino[4,3,2kl]acridinium methosulfate) exhibits anti-proliferative effects, induction of senescence and reduction in telomere length; cell extracts incubated with RHPS4 show reduced telomerase activity. Modelling, biophysical and NMR data suggest the compounds stabilise G-quadruplex structures such as those formed in the telomeres and possibly the c-myc oncogene promoter.
The c-myc gene is capable of activating telomerase by increasing expression of the catalytic subunit of telomerase (TERT). We investigated whether the polycyclic acridines might stabilise a quadruplex structure in the c-myc promoter and reduce cellular expression of c-myc and possibly in turn hTERT. A cell-free, polymerase stop assay revealed potent stabilisation by the compounds of a G-quadruplex structure in the c-myc promoter sequence. Effects on c-myc expression in vitro were evaluated by RT-PCR with members of the series eliciting a modest reduction in c-myc and hTERT mRNAexpression in cancer cell lines. However, the reduction in hTERT expression does not appear to be mediated exclusively through a reduction in c-myc and the pattern of c-myc down-regulation in cells does not correlate with G-quadruplex stabilisation in the stop assay. Thus we conclude that the anti-proliferative and telomerase inhibitory effects of these compounds are not significantly mediated through a reduction in c-myc expression, but more likely through direct interaction with the telomeres. von Hippel-Lindau disease (VHL) is a dominantly inherited familial cancer syndrome caused by mutations of the VHL tumour suppressor gene (TSG). VHL predisposes to the development of tumours including renal cell carcinoma (RCC), hemangioblastoma (HAB) and pheochromocytoma (PHE). Distinct patterns of tumour development and genotype-phenotype correlations exist in VHL disease, suggesting multiple and tissue-specific functions for pVHL, the VHL TSG protein product. pVHL plays a critical role in the regulation of hypoxia inducible factor-1α (HIF-1α) and cellular oxygen sensing. We recently showed that disruption of HIF-1α regulation by disease-associated pVHL mutants correlated with RCC and HAB predisposition in VHL disease, however HIF-1α de-regulation was not associated with PHE susceptibility.

IDENTIFICATION OF pVHL-INTERACTING PROTEINS AND
To investigate further pVHL functions that may play a role in VHLassociated PHE development, we have performed a yeast-2-hybrid screen against a PHE-representative cDNA library to identify putative pVHL-interacting proteins. This assay initially identified 100 potential pVHL-interacting clones. Twenty of these were characterised in detail, and 10 unique binding species were identified which represented inframe protein coding sequences. These included 6 uncharacterised cDNAs and 4 known proteins including the zinc finger protein, Zbrk-1, and eNOS-interacting protein. The specificities of all ten interactions were confirmed in yeast using small-scale yeast-2-hybrid assays.
The ten pVHL-interacting proteins identified are now the focus of further investigations to establish their in vitro and in vivo significance, and any role they may play in PHE development. The diagnosis and treatment of oligodendroglial neoplasms is currently highly controversial. Loss of chromosomes 1p and 19q may be predictive of chemosensitivity, while tumour physiology influences drug delivery and therapeutic responsiveness.
In a prospective study of 65 gliomas treated with PCV chemotherapy at a single treatment centre, laser capture microdissection and microsatellite analysis was used to determine allelic losses. 11/18 OII, 5/18 OAII, 9/10 OIII, 6/18 OAIII and 0/1 GBMO had loss of both 1p36 and 19q13. 3 OII, 7 OAII, 3 OIII, 9 OAIII and 1 GBMO had loss of 17p13 and 5 OIII and 7 OAIII had losses in chromosome 10. Inter or intratumoral heterogeneity of histological subtype seen in 9 cases, was not reflected in genetic heterogeneity. Comparison of pre-treatment molecular genetics and physiology data suggests that tumours with 1p36/19q13 loss are more likely to have increased uptake of 201 Tl. For a subset of cases, pre and post therapy in vivo imaging data, including 1 H MR Spectroscopy, rCBV, 18 FDG-PET and 201 Tl-SPECT is being used to assess radiological responses. For example, a hypermetabolic anaplastic tumour with loss of 1p36/19q13 responded completely to PCV, while a hypermetabolic anaplastic tumour with loss of 17p13 and mutant p53 progressed. Continued clinical follow-up will elucidate the complex physiological and molecular factors that may influence chemosensitivity in oligodendroglial neoplasms.

Background:
The prognosis for children with Wilms tumour has improved steadily over the past four decades so that it is now possible to focus on treatment refinement with a view to reduction of short and long-term side effects. One remaining area of contention is the timing of surgical resection. It has been suggested that the use of preoperative chemotherapy might reduce the risks of surgical intervention and might, by identifying tumours that are chemosensitive, reduce the total burden of therapy needed by patients fortunate enough to have a responsive tumour.
Methods: Children with nonmetastatic renal tumours that were judged to be surgically resectable were randomly assigned either to immediate nephrectomy or to percutaneous biopsy followed by six weeks of chemotherapy with actinomycin D and vincristine followed by delayed nephrectomy. In both arms postoperative therapy was dictated by the tumour stage assigned at the time of nephrectomy.
Results: During the period October 1991 to March 2001) of the 607 eligible patients, 205 were randomised. Of this group 21 were subsequently found to have non-Wilms histology. There were 94 patients who had immediate nephrectomy and 90 patients who had preoperative chemotherapy. Analysis of the stage distribution in the two arms shows a highly significant improvement in the stage distribution for the preoperatively treated group (stage I: 66.3 vs 55.3, stage II: 23.3 vs 14.9, stage III: 10 vs 29.8%). EFS and OS were equivalent between the two arms. There was also a reduction in peroperative tumour rupture and other surgical complications.
Conclusions: Six weeks of preoperative therapy with vincristine and actinomycin D for patients with Wilms tumour identifies patients with chemoresponsive tumours and leads to a more favourable stage distribution, permitting a reduction in the overall burden of their therapy without compromising the chances of cure.
Allogeneic whole cell vaccines are a practical approach to immunotherapy. They do, however, present unique challenges for clinical trial monitoring since their multivalent antigenic nature makes it impractical to examine specific immune responses. We have examined cytokine secretion by peripheral blood lymphocytes (pbl) gathered from a Phase I/II study in hormone refractory prostate cancer. Patients were shown to be free of bone metastases and immunocompetent on entry. Each individual received a mixture of three allogeneic prostate lines plus BCG adjuvant on weeks 0 and 2. Patients received cells alone on week 4 and monthly thereafter for one year. Blood samples were taken at visit and weekly on weeks 12-16 and 32-36.
Pbl were thawed and stimulated overnight with either PHA/ionomycin or LPS. Cytokine levels were measured in the supernatants by cytometric bead array (CBA) and the producing cells' RNA used for real-time PCR with absolute quantitation. Data from 15 subjects initially showed significant elevations in T H 1 cytokines in patients with marked increases in PSA doubling times. As the trial progressed, this phenotype became mixed T H 1 and T H 2. Patients with no PSA response showed either a complete lack of cytokine detection or elevated T H 2 cytokines. Interestingly, protein and cDNA levels correlated well up to week 16 and then cDNA levels appeared depressed compared to protein.
The major goal of cancer research is the development of therapeutic agents specifically aimed at tumour cells. One mechanism potentially amenable to chemotherapeutic intervention involves cytochrome P450 CYP1B1. Our research has established the concept of over-expression of individual forms of P450 in particular CYP1B1 ina range of solid tumours. Moreover our invitro studies have demonstrated the presence of metabolically active CYP1B1 and cytochrome P450 reductase in tumour samples. We have previously identified several anti-cancer drugs (docetaxel, paclitaxel, mitoxantrone and flutamide) as substrates for CYP1B1. Furthermore, our in vitro studies have shown that the presence of CYP1B1 reduces the efficacy of docetaxel. In this study we used expressed human CYP1B1 in a competitive microassay involving the deethylation of ethoxyresorufin to resorufin to extended our screen of anti-cancer drugs and identify which ones interact with CYP1B1, by the change in resorufin production over time. Our findings to date indicate that a range of structurally diverse anti-cancer drugs interact with CYP1B1 (melphalan, bleomycin, methotrexate, altretamine, ellipticine, resveratrol, carmustine, dactinomycin, raltitrexed, epirubicin, and mitomycin C). Several of these drugs have been further characterised to determine their mechanism of interaction with CYP1B1. The overexpression of CYP1B1 in tumour cells and interaction of this P450 with anti-cancer drugs highlight CYP1B1 as an important P450 in tumour cells.
Acknowledgement Aim: To ascertain the frequency of constitutional WT1 mutations in unselected UK patients with sporadic Wilms tumour (WT) and investigate heritability.
Methods: Constitutional DNA from 294 unselected patients with sporadic WT treated at 7 UKCCSG centres was screened for mutations in WT1 by heteroduplex analysis. 11 samples in which >20% of PCR fragments failed were excluded. Suspected mutations were confirmed by sequencing and analysed in parental samples to investigate inheritance. The cohort included 264 unilateral WT, 18 bilateral and 1 extrarenal.
Results: Six children had constitutional WT1 mutations: a nonsense and a splice site mutation in exon1, a frameshift in exon3 and three nonsense mutations in the zinc finger region (R301X and two identical R390X mutations), the latter are recognised somatic and constitutional mutations in Wilms tumour. These mutations were not present in any of the 9 parents tested. The WT1 mutation group had a young median age at diagnosis (13.8m), 4 were female, 2 were male. Only 1/6 had bilateral tumours, all were favourable histology. Only 1/6 had congenital abnormalities (ptosis, bilateral cryptorchidism). None had family history of renal disease or early onset cancers.
Conclusions: This study demonstrates that a small proportion (2.12%: 95%CI 0.44%-3.80%) of children with apparently sporadic Wilms tumour carry constitutional WT1 mutations. Only 2/6 had features suggestive of genetic predisposition. These findings have implications for the counselling of patients with Wilms tumour and their parents.
We thank all UKCCSG centres who contributed to this study and the families for their participation. Aims: In vivo 31 P MRS can be used to obtain information non-invasively on tumour bioenergetics (NTP/inorganic phosphate (Pi) ratio), intracellular pH, and membrane metabolism (measuring phosphorylcholine (PC) and phosphorylethanolamine (PE)). The aim of this study was to investigate these as surrogate markers for tumour response after ifosfamide (Ifos) therapy in Rd xenografts with a view to translating these pharmacodynamic markers to clinical studies.
Methods: Rd xenografts were grown in nude mice. When tumour size reached 582 ± 47 mg, in vivo localised 31 P MRS was performed before and 7 days after treatment with saline (control, n=5) or Ifos (250mg/kg, n =11). Tumours were then extracted and analysed using complementary high resolution 31 P MRS in vitro. Results: At 7 days, tumour volume changed by 7±7% after Ifos compared with an increase of 102± 19% after saline (p = 0.006). (PE+PC)/Pi (p = 0.03) and β-ATP/Pi (p=0.04) ratios were significantly increased in Ifostreated tumours, whereas there were no significant changes in controls. PE was also increased in the extracts of Ifos-treated tumours compared with control (p = 0.005) The in vitro finding confirmed the in vivo data and showed that the rise of the (PC+PE)/Pi ratio in vivo was due to increased PE, with PC and Pi unchanged.
Conclusions: Significant changes in tumour bioenergetics and membrane metabolism were found after chemotherapy. This suggests measurements with 31 P MRS could be used as a useful non-invasive surrogate marker in clinical studies. Anonymised individual level data for the years 1998-2000 were obtained from Thames Cancer Registry. For registered cancers in the age range 10-24 years frequency tables were produced. Tumours were categorised using the new Birch-Alston (Manchester) classification. A total of 1,394 tumours were identified over approximately 7.7million person years at risk. Annual incidence rates per million were calculated based on mid-year population estimates for the study region. Age specific incidence rates for each 5-year age group: 10-14, 15-19 and 19-24 years were calculated. Direct standardisation to the world population was used to calculate an age-standardised rate. A crude analysis of service requirement was formulated using registryderived figures for uptake of chemotherapy, radiotherapy and surgery.
Overall rates in the three age groups were 97.3, 167.8 and 268.2 per million person years respectively. Lymphoma was the commonest diagnosis in all age groups. Our results closely mirror nationally published data. We have described a population of teenagers and young adults, using a novel classification scheme that helps target studies in young cancer patients. The aim of this study was to further characterise the role of death receptors in bFGF-induced cell death in Ewing's sarcoma family tumour (ESFT) cell lines, and examine the hypothesis that expression of these receptors may be regulated by sustained activation of the MAPK p38 and/or ERK. The ESFT cell lines studied differentially expressed the death receptor p75 NTR , however, there was no correlation between the level of p75 NTR expression and induction of cell death by bFGF. Up-regulation of p75 NTR occurred 48h after exposure to bFGF in ESFT cells that died following exposure to bFGF, but was unchanged in cells that did not die. To identify whether other apoptotic genes are upregulated in response to bFGF, a commercially available apoptotic cDNA microarray was probed using total RNA isolated from bFGF-treated and untreated TC32 cells. Three additional death receptors were upregulated in bFGF-treated TC-32 cells compared to the control: CD95/Fas, DR5/KILLER and the TRAIL decoy receptor DcR2. To determine if p38 or ERK signalling is required for up-regulation of death receptors following exposure to bFGF, TC-32 cells were pre-treated with either SB202190 (p38 inhibitor) or PD98059 (ERK inhibitor) and then exposed to bFGF or were left untreated. Western analysis demonstrated that both CD95/Fas and p75 NTR are upregulated post-treatment with bFGF when compared to untreated control, and pre-treatment with the p38-inhibitor SB202190 prevented upregulation of these two death receptors. However pre-treatment with the ERK inhibitor PD98059 had no effect on death receptor expression following exposure to bFGF. Chemotherapy has dramatically improved the prognosis for osteosarcoma but a proportion of patients still develop drug resistance. Cisplatin (cDDP) is reported to be the most active agent but the mechanisms by which it induces cell death are not clearly defined. Caspases 3 and 9, play an important role in cDDP-induced apoptosis in other cell types. Caspase 9 activation occurs following release of cytochrome c (Cyt c) from mitochondria. We investigated mitochondrial apoptotic pathway activation by cDDP in the HOS cell line. HOS cells stained with the mitochondrial membrane potential(∆ϕ) sensitive probe JC-1 demonstrated a reduction in within 45 min of cDDP (1-10µg/ml) addition. Cyt c release and pro-caspase 9 and 3 activation were determined by Western blotting. Low levels of Cyt c were found in only the mitochondria of control cells. In cDDP (10µg/ml) treated cells, Cyt c was present in the cytosol and at higher levels in the mitochondria after 1 and 0.5 hours respectively. Pro-caspase 9 was present in cytosolic extracts of control and cDDP treated cells at all time points but was present in the mitochondrial fraction of only cDDP treated cells. These cells showed low levels of cleaved active caspase 9, 0.5 hours after treatment, which significantly increased by 2 hours. High levels of caspase 3 were present in the cytosol of cDDP treated cells at all time points but was virtually undetectable in controls.

CELL DEATH FOLLOWING EXPOSURE OF ESFT CE LLS TO bFGF IS EFFECTED BY
These results show activation of a mitochondrial apoptotic pathway by cDDP in human osteosarcoma cells. Greater understanding of the mechanisms involved will assist in efforts to over come resistance. Reciprocal chromosomal translocations are associated with acute leukaemias. 5-10% of these translocations result in the fusion of the MLL on 11q23 to a number of other genes, including MLLT10. MLLT10 encodes a 1084aa protein and contains a LAP/PHD zinc finger, together with a functional nuclear localisation signal, AT hook motif and a leucine zipper.
Two AF10 "like ,, genes have been identified in S. pombe. Deletion of both genes PAL1 and PAL2 (pombe AF10 Like)indicates that the genes are non-essential in yeast. In addition, Over-expression of human MLLT10 is deleterious for fission yeast. Cells in which MLLT10 is overproduced are highly elongated, reminiscent of cell cycle delay. Nuclear staining with DAPI shows an altered "halo" like chromatin structure, and FACS analysis shows abnormal ploidy. Electron microscopic analysis demonstrates extensive cytosolic vacuolisation and increased numbers of mitochondria and peroxisomes. Immunogold labelling of anti-AF10 antibody localises MLLT10 to the nucleus, and specifically the condensed chromatin. These results may indicate that overproduction of MLLT10 results in the disturbance of pathways involved in chromatin DNA structure and vesicle trafficking by interacting with endogenous yeast proteins. Activation of the Ets transcription factors, Ets-1, Ets-2 and PEA3 are implicated in the transcriptional regulation of HER2/ neu. The transcriptional activity of Ets is thought to be modulated by nuclear regulatory proteins, including amplified in breast cancer 1 (AIB1).

THE INTERACTION OF THE ETS TRANSCRIPTION FACTOR FAMILY AND ITS CO-REGULATORY
Our aims were to localise the transcription factors and AIB-1 in breast cancer using immunohistochemistry and to co-localise them using immunofluoresence. The ability of the growth factors bFGF and EGF to modulate the protein expression of the Ets transcription factors in primary breast cultures and breast cancer cell lines was assessed using western blotting.
PEA3, Ets-1, Ets-2 and AIB-1 were localised within breast cancer tissue. The transcription factors and AIB-1 were co-localised to the same cell. Growth factors induced an up-regulation in the protein expression of the Ets transcription factors in a dose dependant manner.
The growth factors, bFGF and EGF regulate Ets transcription factors. Co-localisation with AIB1, implicates nuclear regulatory protein interaction in the modulation of Ets activity in human breast cancer. Between 1987 and 2000, 250 patients were treated with intramuscular methotrexate 50mg on alternate days (1,3,5,7) with four doses of oral folinic acid 7.5mg rescue on alternate days (2,4,.6,8), with 7 days between courses. The overall complete response rate without recurrence was 72% for first line treatment and 95% for those who required second line chemotherapy.
Eight women (3.2%) had recurrence following remission and 2(0.8%) had new moles. Two died of their disease giving an overall cure rate of 99%. Only 10 women (4%) experienced grade 3 or 4 toxicity during the first course of treatment and 14 women (5.2%) subsequently. The main toxicities were mucositis and pleuritic chest pain. 141 women (56.4%) became pregnant following treatment; 128 (90.7%) of these were full term normal pregnancies.
No second malignancies have been observed. Methotrexate with folinic acid rescue is an effective treatment for low risk persistent trophoblastic disease. It has minimal severe toxicity, excellent cure rates and does not appear to affect overall fertility. Taking the Charing Cross, London [McNeish et al. J Clin Oncol 2002, 20:1838, and Weston Park, Sheffield, joint experience for all low risk patients treated in the UK (735 over 14 years) the cure rate approaches 100% -a testament to the effectiveness of the national surveillance and treatment scheme.  (1) and shown no difference in overall survival (OS) or relapse free survival(RFS). However, those patients in the neoadjuvant arm who responded well to treatment had an improved RFS but not a significantly improved OS compared to those patients who failed to respond.We now update these results with a further 3 years of follow-up data. Methods 309 women were randomised to either primary surgery followed by 8 cycles of adjuvant mitoxantrone, methotrexate with tamoxifen (2MT) with or without mitomycin-C(3MT) or the same chemoendocrine regimen for 4 cycles before and 4 cycles after surgery. Radiotherapy was prescribed according to standard criteria. Tamoxifen was continued for 5 years for estrogen receptor positive patients. 142 patients were treated in the adjuvant arm and 144 patients were treated in the neoadjuvant arm. Median age was 56 years (27-70). Patient demographics were balanced between the two arms. Median follow-up of patients is 111 months (range 10 months -131 months). Results There is still no significant difference in OS or RFS when comparing women receiving adjuvant versus neoadjuvant treatment. As in previous reports the 73 patients in the neoadjuvant arm who achieved a complete or near complete response with "minimal residual disease" (CR,MRD) had an improved RFS compared with the 69 patients who achieved a partial response or who demonstrated no change (PR, NC The therapeutic benefit of adding vinorelbine to epirubicin was evaluated in patients with metastatic breast cancer. A total of 387 patients were randomized to receive intravenous vinorelbine 25 mg/m² days 1 and 8 plus epirubicin 90 mg/m² day1 (EV) or epirubicin 90 mg/m² day1 (E). Both study regimens were given every 3 weeks until progressive disease, excessive toxicity or a maximum of one year. Cumulative dose of epirubicin should not exceed 950 to 1000 mg/m². Prior anthracycline-based adjuvant chemotherapy and prior chemotherapy for metastatic breast cancer was not allowed. The primary endpoint was progression-free survival (PFS).
The 2 study arms were well balanced for patient and tumour characteristics. Median number of cycles (9) and median cumulative doses of epirubicin (758 and 729 mg/m²) were similar. Overall response rates (RR) to EV and E were 50% and42%, respectively. The complete RR was significantly higher in the combination arm (17% versus 10%, p =0.048). Median PFS was significantly longer with EV compared to E (10.1 versus 8.2 months, p = 0.019). Addition of vinorelbine to epirubicin reduced the risk of progression by 25% (p= 0.006). Median survival was 19.1 months with the combination and 18 months with epirubicin alone (p = 0.50). Leucopenia-related complications, stomatitis and peripheral neuropathy were more frequent with the combination. Incidences of cardio-toxicity, constipation and injection site reactions were not significantly different between the two arms.
In conclusion, addition of vinorelbine toepirubicin conferred a significant advantage in terms of complete response rate and progression-free survival. Background: Previous studies have shown an association between vitamin D deficiency and breast cancer risk. Vitamin D is also known to inhibit the mitogenic and anti-apoptotic effects of IGF-1 in breast cancer cells; the bioavailability of IGF-1 is regulated by insulin-like growth factor binding protein-3(IGFBP-3). The local expression of vitamin D in the breast, however, is unknown. Furthermore, it remains to be determined whether these effects involve IGFBP-3.
Aims: i)To quantify the expression of 25-hydroxyvitamin D-1-alpha-hydroxylase (1 alpha OHase), the enzyme which converts 25 hydroxyvitamin D to its active metabolite, in paired normal and malignant breast tissue and to correlate this expression with that of IGFBP-3 mRNA and tumour phenotype(histology, oestrogen receptor (ER) status, vascular invasion and lymph node metastases).
Methods: Total RNA was extracted from 21 paired samples of normal and malignant breast tissue. Local Ethical Committee Approval was obtained. mRNA transcription was quantified using a real time, hydrolysis probe-dependant RT-PCR assay ('Taqman').
Conclusions: The significant down regulation of 1 α-OHase in normal compared to malignant breast tissue supports a role for vitamin D in early breast tumourigenesis. The significant correlation in tumours between 1 α-hydroxylase and IGFBP-3 expression supports the view that IGFBP-3 might influence the regulatory effects of vitamin D in breast cancer.

CORRELATION BETWEEN BCL-2 EXPRESSION AND DIFFERENT CLINICO-PATHOLOGICAL VARIABLES IN HUMAN BREAST CANCER
Alaa T Abdel-Rahman*, E Husain, C Wells, and R Carpenter. St Bartholomew's Hospital, Breast and Endocrine Unit, London.
Introduction Bcl-2 plays an important role as an inhibitor of apoptosis that may extend the viability of cells containing genetic alterations and facilitate tumour progression. Inspite of this antiapoptotic effect, Bcl-2 expression is associated with good prognosis in breast cancer. The aim of this study was to relate Bcl-2 expression to differing clinico-pathological variables in breast cancer.
Methods We reviewed fifty-four patients with breast cancer. Immuno-histochemical methods were used together with modified quick score of archival paraffin-embedded sections of tumour. Bcl-2, HER-2/neu, oestrogen receptor, androgen receptor and progesterone receptor were assessed. Expression of Bcl-2 was also related with age, menopausal status, lymph node status, tumour size, tumour grade and tumour types.
Conclusion We confirm that Bcl-2 is related to favourable pathological variables. Its assessment in combination with these variables might allow selection of patients with good prognosis who might avoid aggressive systemic therapy. The incidence of hormone dependent cancers is lower in Asia as compared to Western nations. The Asian diet is traditionally high in phytoestrogens (PEs), which exhibit putative anti-cancer activities.This research aims to determine the growth effects of the dietary PEs daidzein, genistein and coumestrol in human breast cancer cells, also investigating potential mechanisms of ER-mediated signalling. The characterisation of ER expression using flow cytometry revealed ER as the predominant receptor subtype in MCF-7 cells, whilst equal levels of ERα and ERβ protein were detected in T47-D. Both ERpositive cell lines demonstrated biphasic growth responses following 0.1-50µM PE treatment, as determined by proliferation assays and cell cycle analysis. Physiological doses of 0.1-10µM genistein and coumestrol stimulated MCF-7 and T47-D growth and S-phase activity, whilst concentrations in excess of 25µM induced G 2 /M phase arrest. A dose-dependent stimulatory effect of daidzein was observed in MCF-7, however T47-D growth was inhibited only at 25µM. Reporter gene assays suggest the involvement of ER-dependent signalling via the activation of ERE and AP-1 response elements.
We have additionally assessed the effects of PEs on novel primary breast epithelial cultures established from tumour tissue. Tumour cell proliferation was similarly stimulated by PE doses of 0.1-10µM. Our data therefore suggests that the ingestion of dietary PEs may exacerbate the growth of ER-positive breast tumours. Thismay be a cause for concern for post-menopausal subjects using PEs as a natural alternative to HRT. We carried out a retrospective analysis of patients with tumours <10mm seen at the Royal Marsden Hospital between 1990 and 2002 from our perspective database. All patients had axillary node dissection and we have full histological information on lymphovascular invasion, ER/PR status, number of involved nodes and the tumour grade. Patients who received neoadjuvant chemo/endocrine therapy or preoperative irradiation were excluded.
Results: 452 patients met the analysis criteria. Discussion: Ninety-eight out of the 452 patients (21.7%)with tumours under 10mm had axillary lymph node involvement. Even in grade I tumours, there is a 12.8% incidence of nodal involvement. With holding axillary sampling in these patients may result in understaging of a significant number of patients.
We found vascular invasion and grade of the tumour to be independent predictors for nodal involvement. 58% of the node +ve patients had vascular invasion while only 41%of the node +ve patients had no vascular invasion. The incidence of nodal involvement increased with grade, being 12.8% for grade I tumoursand 29% for grade III tumours. Higher grade tumours tended to have a higher incidence of vascular invasion.
We found that 56% of the node +ve patients were ER +ve while only 15.3% of node +ve patients were ER-ve. This is thought to reflect the higher frequency of ER +ve patients (276/452) skewing the data as 19.9% of ER +ve patients were node +ve compared with 26.7% of ER-ve patients. Therefore, a higher percentage of ER-ve patients seem to have nodal involvement. Conclusion: Vascular invasion, grade and ER -ve status are all predictive factors for lymphnode metastasis. Purpose: To evaluate neoadjuvant cap/ox prior to CRT and TME in newly diagnosed patients (pts) with MRI defined poor risk/locally advanced rectal cancer. Methods: Pts received 12 weeks of neoadjuvant cap (2000mg/m 2 /day po for 14 days every 3 weeks) and ox (130mg/m 2 iv every 3 weeks). Starting on week 13, cap was continued at 1650mg/m 2 /day continuously with concomitant radiotherapy. TME was planned 6 weeks after CRT. Post-operatively, pts received 12 weeks of cap at 2500mg/m 2 /day. MRI was repeated after chemotherapy and CRT. Results: Between November 01 and November 02, 22 eligible pts were recruited. Median age was 62 (range=38-80). So far, 17 patients are evaluable for radiological response and 15 patients have proceeded to TME. Following neoadjuvant cap/ox, all pts had partial responses (RECIST criteria). In addition, 80% of pts had symptomatic responses in a median of 22 days (i.e. after one cycle of chemotherapy) including reduced rectal bleeding (100%), improvement in diarrhoea/constipation (79%), diminished pelvic pain/tenesmus (64%) and weight gain/stabilisation (100%). Following CRT, tumour response was sustained in all pts. Of 20 pts with initial threatened circumferential resection margin (CRM), tumour regression away from the CRM occurred in all pts who have completed treatment so far allowing R0 resection in 100% of these pts. Pathological CR was found in 5 pts. Conclusion: Capecitabine and oxaliplatin prior to synchronous CRT and TME produces almost universal tumour regression, rapid symptomatic response and allows R0 resection to be achieved. Introduction Late side-effects following short course preoperative radiotherapy (SCPRT) and total mesorectal excision (TME) for operable rectal cancer have not been adequately studied. Acute toxicity in a series of 176 consecutive patients has been previously reported. In this study, late effects in the same cohort are presented.
Method Side effects occurring more than 3 months after the start of SCPRT were graded using the EORTC/RTOG late radiation toxicity system. Multivariate analysis was performed to identify associated factors Results Of 176 patients, 15 died within three months of SCPRT and 5 patients were lost to follow up. At a median follow up of 40 months severe (grade 3-4) toxicity was seen in 20 (13%) of 156 assessable patients: Gastrointestinal 13 (8%); Other 7 (5%). On multivariate analysis, abdominoperineal (AP) resection (p<0.001) was associated with less severe toxicity.
Conclusions In this retrospective series the rate of late grade 3-4 toxicity following SCPRT and TME was 13%. Although AP resection is associated with a lower incidence of late side-effects this may be counterbalanced by the impact of a stoma on quality of life. These factors should be considered when deciding on the optimal management of operable rectal cancers. Introduction: Chemoradiation prior to surgery is now a common approach in the treatment of locally advanced rectal cancer. This report describes the resection, survival and local relapse rates of patients treated over an 8-year period at Mount Vernon Hospital.
Methods: Between 1994 and 2002, 155 patients with a locally advanced adenocarcinoma of the rectum were treated. T stages were: T2:2, T3:59, T4:94. Median age 67 years, male: female ratio 73:27%. The radiotherapy dose was 45Gy in 25 fractions over 5 weeks. 5-FU was given as a 60 minutes infusion on days 1-5 and 29-33, at 350mg/m 2 , following low dose folinic acid. Surgery was performed in 114 patients. Actuarial survival, local control and toxicity rates according to CTC criteria, were assessed.
Conclusion: Chemoradiation with 5-FU can produce effective downstaging in patients with locally advanced rectal carcinoma and toxicity is manageable. Survival appears poor in this group and the results of the Phase III EORTC trial with 5FU/folinic acid are awaited. Phase two studies are in progress to investigate the use of newer agents in combination with 5FU, which may not only improve LR and R0 resection rates but also impact upon survival.

Introduction:
The lymph node (LN) status of rectal cancers is a key prognostic indicator. UICC guidance suggests that at least 12 LNs should be examined before a tumour can be pronounced nodenegative. Others suggest that 7 or 8 LNs might be sufficient, but with increasing usage of neoadjuvant chemoradiation (NCRT) the effacement of LNs means that often far fewer are found.
Patients and methods: 185 patients with T3 or T4 rectal cancer treated with NCRT between 1994 and 2002 have been entered into a prospective database. All patients were deemed inoperable on initial staging. Treatment was with concurrent radiotherapy (45 Gy in 25 fractions over 5 weeks) and leucovorin/5FU chemotherapy.
Results: 134 males and 51 females had a mean age of 63. A total of 161 patients (pts) proceeded to surgery. The mean number of LNs found in the resection specimens was 5.47 (range 0-21). Complete pathological response was seen in 19 pts (12%). Overall 3-year survival figures were 54%. For node positive pts this fell to 43%, with a disease free survival (DFS) of 16%. In node negative pts with 2 or less nodes examined, 3-year survival figures were 47% (DFS = 26%). Where 3 or more negative nodes were found, overall survival was 66% (DFS = 58%).
Conclusions: LN status following chemoradiation is predictive of survival. There is considerable effacement of LNs, resulting in a yield lower than would be traditionally required for prognostic information. We demonstrate that if more than 2 negative nodes are found, the outcome is more favourable than if 2 or less negative nodes are identified. It seems likely that this is due to understaging of pts in whom insufficient nodes are recovered, and we suggest that following NCRT at least 3 nodes are required before a specimen can be considered truly N0. This large prospective, multicentre, multidisciplinary, pragmatic, phase III trial was designed to compare conventional (Open) versus laparoscopic-assisted (Lap) surgery in colorectal cancer in terms of pathological, clinical and health economic outcomes. Patients were randomised in an unbalanced allocation of 2:1 Lap: Open resection. Recruitment closed on 31 July 2002 with a total of 794 patients (526 Lap, 268 Open).
Primary short-term endpoints were pathological resection margins and 30-day operative mortality. Secondary short-term endpoints were blood transfusion requirements, operative complication rates and short-term quality of life. All pathological data were centrally reviewed. Longer-term endpoints are defined as recurrence and metastases, disease-free and overall survival.
This trial provides an early indication of the role of laparoscopic resection in colorectal cancer. The short-term results suggest that, in the UK, the laparoscopic-assisted technique for colorectal cancer appears to be no different from open surgery. It is the only trial considering rectal cancer, and is unique in its high-quality pathological data collection and pathological endpoints. The targeting of anticancer drugs to tumours seeks to exploit a difference in the biology of tumours compared with that of normal tissues. One such difference is the high expression of the α-FR in some epithelial tumours e.g. 90% of ovarian. The α-FR, linked to increased susceptibility to malignant transformation and regulation of tumour proliferation, also is a high affinity, low capacity transporter of folates. The reduced-folate carrier (RFC) is the primary folate/antifolate transporter and its ubiquitous expression leads to antifolate cytotoxic effects in tumours and normal tissues. TS inhibitors such as raltitrexed, pemetrexed and ZD9331 are also co-transported via the α -FR in some preclinical models. Thus, we hypothesised that a TS inhibitor displaying high and low affinity or the α -FR and RFC respectively could be an α-FR-targeted drug. Two cyclopenta[g]quinazoline-based compounds, CB300638 and CB300945, possess this profile (TS K i = 0.24nM and 1.4nM respectively). Following rapid binding to the α -FR they are slowly trafficked via an endosomal pathway. The 72h IC 50 in A431-FBP (α-FR transfected) and KB (α -FR constitutive) cells is ~3nM while in α -FR negative lines is ~1 to 10µM. Both compounds localised to the KB xenograft in mice (100mg/kg i.p. bolus) compared to normal tissues. Target inhibition in tumour, but not in normal tissues, was demonstrated using 125 IdUrd metabolism as a surrogate marker of TS inhibition. By contrast, selective targeting in this model was not seen with ZD9331. CB300638 and CB300945 have been selected as potential drug candidates that are predicted to display efficacy in α -FR + tumours and very low toxicity. Supported by Cancer Research UK. This paper draws on data drawn from two studies undertaken to assess the health care needs of patients with primary lung cancer. In the first of the two studies (a national mail questionnaire survey), 209 (45%) patients accessed from 24 randomly selected hospitals throughout the UK, returned completed questionnaires. Participants responded to questions relating to their experiences of care provision from time of first presenting to their GP, on through treatment and into follow-up. The second of the two studies, draws on in-depth interview data gathered from a cohort of 60 patients with lung cancer who agreed to take part in interviews at one, three and six months post diagnosis.
Although the considerable impact of the diagnostic process and of its "telling" on cancer patients' construction of their experience of care are well documented, little is known of the process and associated outcomes of the diagnostic pathway for patients with lung cancer. Given that the median survival for patients with lung cancer after diagnosis is less than four months, with about 80% of patients dying within one year 1 , the need to ensure that the diagnostic process is swift, well co-ordinated and therapeutic in its own right, is imperative.
Data from our studies suggest that work needs to be undertaken, and its effect evaluated, to better inform the public of symptoms that warrant urgent attention. This study aimed to assess the palliative benefit of MVP chemotherapy in patients (pts) with mesothelioma and assess prognostic factors. Pts with histologically confirmed mesothelioma were treated with mitomycin-C 8mg/m 2 (courses 1,2,4 &6), vinblastine 6mg/m 2 (maximum 10mg) and cisplatin 50mg/m 2 IV q 21 days for up to 6 cycles. 150 pts were treated and analysed for objective and symptomatic tumour response, toxicity and survival. The overall response rate was 15.3% (95% CI 9.3-21.4%) and 68.6% demonstrated no objective change. 147 pts reported symptoms at the start of chemotherapy and 69% had a symptomatic response to MVP, including 2 pts with a complete resolution of all symptoms. There were no toxic deaths. The most frequent grade 3/4 non-haematological toxicities were infection (12%), nausea/vomiting (9%) and constipation (5%). Grade 3/4 haematological toxicities were neutropaenia (22%), anaemia (6%) and thrombocytopaenia (5%). Median survival was 7 months and at 1-year 31.6% (95% CI:24.0-39.5%) were alive. Univariate analyses demonstrated that poor performance status (WHO 2/3) (p=0.003), weight loss (p=0.009), non-epithelioid histology (p=0.003), low haemoglobin (<13 g/dl in males and <11.5g/dl in females) (p=0.008), low white cell count (<11x10 9 /l) (p=0.008) and low absolute neutrophil count (<5.0 x 10 9 /l)(p=0.02) were all significant prognostic factors. Multivariate analysis demonstrated that weight loss (relative risk [RR] 3.34,p=0.001), non-epithelioid histology (RR 2.69,p=0.01), low haemoglobin (RR 2.41, p=0.01), low white cell count (RR 2.94,p=0.001) were all independent prognostic factors. We conclude that MVP is a palliative regimen which can be given safely even to pts with poor performance status. In addition, we have confirmed that weight loss, non-epithelioid pathology, and anaemia are poor prognostic factors. To improve outcomes treatment strategies will need to address symptoms such as weight loss and anaemia. Dr University Department of 1 Surgery, 2 Histopathology, 5 Immunology Royal Free and University College Medical School, London, U.K, 3 HPV Laboratory, Institut Catala d'Oncologia, L'Hospitalet, Barcelona,Spain,and 4 Delft Diagnostic Laboratory, The Netherlands.
Most cases of anal carcinoma develop from high grade anal intraepithelial neoplasia (AIN) caused by persistent human papillomavirus (HPV) infection. We studied the host's local cellular immune response in HPVrelated anal lesions and compared the results between HIV + and HIVgroups. This study assessed the degree of local T lymphocytic infiltrate (CD-3, CD-4 and CD-8) in the stroma and epithelium of HPV related anal lesions in 75 patients (41 HIV -and 34 HIV+) with immunohistochemistry, and the HPV serotype/s by PCR detection of carcinogenic HPV types (16, 18, 31, 33, 35, 39, 45, 51, 52, 56, 58, 59 and 68).
There was a local lymphocytic infiltrate in the stroma and epithelium, in all groups of lesions. The T cell infiltrate (CD4 and CD8) showed a significant (p<0.05) increase as the lesions became gradually dysplastic and invasive both in HIV -and HIV + groups. The cell infiltrate was significantly lower in the HIV + group for each group of lesions, mostly at the expense of CD4 cells. 74% of lesions in the HIV -group and 88% in the HIV + group contained carcinogenic HPV types. This difference was not statistically significant. This is the first study that shows a local cellular immune response in anal HPV lesions. These data also suggest that the poorer prognosis in HIV + patients is likely to be due to their inherent cellular immunosupression, rather than a higher exposure to carcinogenic HPV types.