Search

Neovasculature development is a crucial step in the natural history of a cancer. While much emphasis has been placed on proangiogenic growth factors such as VEGF, it is clear that endogenous angiogenesis inhibitors also have critical roles in the regulation of this process. Recent research has identified several cryptic fragments of extracellular matrix/vascular basement membrane proteins that have potent antiangiogenic properties in vivo. It has become apparent that many of these fragments signal via interactions with endothelial integrins, although multiple downstream effector pathways have been implicated and endostatin, the first non-collagenous domain of collagen XVIII, influences an intricate signalling network. The activity of these molecules in animal models suggests that they may have significant clinical activity; however, results of phase I/II trials with endostatin were disappointing. Many possible reasons can be found for the failure of these studies. Weaknesses in trial design, endostatin administration regimen and patient selection are identifiable, and importantly the lack of a clearly defined antiangiogenic mechanism for endostatin hindered assessment of biologically effective dose. Additionally, in vivo immunological and proteolytic function-neutralising mechanisms may have negated endostatin's actions. Lessons learned from these studies will aid the future clinical development of other antiangiogenic extracellular matrix protein fragments.

Endocrine therapy is a recognised option in the treatment of chemo-resistant ovarian cancer. We conducted a nonrandomised phase II evaluation of combination endocrine therapy with tamoxifen and goserelin in patients with advanced ovarian cancer that had recurred following chemotherapy. In total, 26 patients entered the study, of which 17 had platinum-resistant disease. The median age was 63 years and enrolled patients had received a median of three chemotherapy regimens prior to trial entry. Patients were given oral tamoxifen 20 mg twice daily on a continuous basis and subcutaneous goserelin 3.6 mg once a month until disease progression. Using the definition of endocrine response that included patients with stable disease (SD) of 6 months or greater, the overall response rate (clinical benefit rate) was 50%. This included one complete response (CR) (3.8%), two partial responses (PR) (7.7%) and 10 patients with SD (38.5%). The median progression-free interval (PFI) was 4 months (95% CI 2.4-9.6) while the median overall survival (OS) was 13.6 months (95% CI 5.5-30.6). Four patients received treatment for more than 2 years (range 1-31) and one of them is still on treatment. In none of the four patients was there any evidence of recurrent or cumulative treatment related toxicity. Treatment-limiting toxicity was not seen in any of the study population. Endocrine data demonstrated a marked suppression of luteinising hormone (LH) and follicle-stimulating hormone (FSH) to less than 4% of baseline values. No consistent correlation could be established between LH/FSH suppression and tumour response. Likewise no relationship was observed between Inhibin A/B and pro-alpha C levels and tumour response. Inhibin is unlikely to be a useful surrogate marker for response in locally advanced or metastatic ovarian cancer. Combination endocrine therapy with tamoxifen and goserelin is an active regimen in platinum-resistant ovarian cancer patients. Hormonal therapy is advantageous in its relative lack of toxicity, ease of administration and tolerability, thus making it suitable for patients with heavily pretreated disease, compromised bone marrow function and other comorbid conditions that contraindicate cytotoxic therapy as well as in patients with indolent disease.

Aims: Vascular endothelial growth factor (VEGF)-C and VEGF-D are angiogenic and lymphangiogenic members of the VEGF family of growth factors. Increased VEGF-C or VEGF-D expression in human tumours may be associated with lymph-node metastasis and lymphatic invasion. Circulating plasma levels of VEGF-A, VEGF-C and VEGF-D were measured in patients with colorectal cancer, and assessed for their usefulness as a diagnostic tool for determining lymph-node metastasis., Materials and Methods: One hundred and twenty patients with colorectal cancer and 50 healthy control patients were included in the study. Plasma growth-factor levels were assessed by enzyme-linked immunosorbent assays., Results: No significant differences in plasma VEGF-C or VEGF-D levels were seen between patients subgrouped by clinicopathological variables. In particular, there were no differences in median plasma VEGF-C or VEGF-D level in patients with and without lymph-node involvement (VEGF-C: 11.2 U/ml [range, 4.9-51.9] vs 9.9 U/ml [4.4-93.4 U/ml]; P = 0.90; VEGF-D: 335 pg/ml [113-1102] vs 316.5 pg/ml [0-1343]; P = 0.68)., Conclusions: Circulating plasma levels of VEGF-C and VEGF-D do not allow pre-operative identification of lymph-node status in patients with colorectal cancer.

Fibroblast growth factor-2 (FGF-2) is a potent angiogenic cytokine that is dependent on heparan sulfate for its biological activity. We have investigated the relationship among heparan sulfate, FGF-2, and the signal-transducing receptors in human, advanced-stage, serous ovarian adenocarcinoma. Using a unique molecular probe, FR1c-Ap, which consisted of a soluble FGF receptor 1 isoform IIIc covalently linked to an alkaline phosphatase moiety, the distribution of heparan sulfate that had the ability to support the formation of a heparan sulfate/FGF-2/FGFR1 isoform IIIc alkaline phosphatase heparan sulfate construct complex was determined. This may be taken as a surrogate marker for the distribution of biologically active heparan sulfate and was distributed predominantly in endothelial cells and stroma but was absent from adenocarcinoma cells. In situ hybridization revealed the expression of FGFR1 mRNA in the endothelium and reverse transcription-PCR confirmed the presence of FGFR1 isoform IIIc but not isoform IIIb. The presence of FGF-2 around tumor endothelium was detected through immunohistochemistry. Double-staining techniques showed that heparan sulfate was found predominantly at the basal aspect of the endothelium and suggested that syndecan-3 might function as one of the proteoglycans involved in FGF-2 signaling in the endothelium. The data suggest that the entire extracellular signaling apparatus, consisting of FGF-2, biologically active heparan sulfate, and FGFRs capable of responding to FGF-2, is present in ovarian cancer endothelium, thereby highlighting the cytokine and its cognate receptor as potential targets for the antiangiogenic treatment of this disease.

Vascular and angiogenic processes provide an important target for novel cancer therapeutics. Dynamic contrast-enhanced magnetic resonance imaging is being used increasingly to noninvasively monitor the action of these therapeutics in early-stage clinical trials. This publication reports the outcome of a workshop that considered the methodology and design of magnetic resonance studies, recommending how this new tool might best be used.

Background: Previous studies have shown extensive vascularisation surrounding subcutaneously implanted fibres when the duration of the US National Cancer Institute (NCI) hollow fibre assay was prolonged., Materials and Methods: The feasibility of adapting the NCI assay for evaluating agents targeting the tumour vasculature was investigated in vitro and in vivo. Finally, in the optimised assay, changes in neovasculature formation around the fibres following treatment with the anti-vascular agent paclitaxel were quantified by immunohistochemistry., Results: Correlations between cell number seeded, time in culture and vascular endothelial growth factor (VEGF) secretion were seen. In vivo studies showed that transplanting single rather than 3 fibres at a site reduced inflammation, reducing the length of the fibre transplanted, as did without any significant loss in cell growth over 21 days. A statistically significant reduction in neovascularisation surrounding the fibres was seen accompanying paclitaxel treatment., Conclusion: Modifications made here to the NCI hollow fibre assay demonstrate its potential for analysing anti-tumour vasculature agents.

Purpose: The purpose of our study was to investigate the cellular accumulation, DNA cross-linking ability, and cellular toxicity of RH1 (2,5-diaziridinyl-3-[hydroxymethyl[-6-methyl-1,4-benzoquinone), a novel DNA alkylating agent currently in clinical trials. In addition, the in vivo efficacy of RH1 formulated in different vehicles was also compared., Experimental Design: RH1 is activated by the two-electron reducing enzyme NQO1 [NADPH:quinone oxidoreductase] forming a potent cytotoxic agent that cross-links DNA. We have used whole blood, cell lines, and primary explanted tumor cultures to measure both the cellular accumulation, DNA cross-linking, and cytotoxicity of RH1. Furthermore, the pharmacokinetic and pharmacodynamic characteristics of RH1 formulated in different vehicles were measured in vivo using the validated comet-X assay in mice bearing human tumor xenografts., Results: Accumulation of RH1 was shown to be both time and concentration dependent, reaching a maximum after 2 hours and correlated well with DNA cross-linking measurements. DNA cross-linking in vitro could be detected at low (1-10 nmol/L) concentrations after as little as 2 hours exposure. In primary tumor cultures, RH1 induces much higher levels of DNA cross-links at lower doses than either mitomycin C or cisplatin. In vivo efficacy testing using polyvinyl pyrrolidone, saline, or cyclodextrin as vehicles showed DNA cross-links readily detectable in all tissues examined and was enhanced when given in cyclodextrin compared with polyvinyl pyrrolidone or saline., Conclusions: RH1 represents a potent bioreductive anticancer drug, which may prove effective in the treatment of cancers, particularly those that overexpress NQO1. DNA cross-linking can be reliably measured in tissue using the validated comet-X assay.

We assessed the tolerability, safety, pharmacokinetics and dose-limiting toxicity (DLT) of the recombinant humanized IgG4 anti-vascular endothelial growth factor (VEGF) monoclonal antibody, HuMV833, in patients with advanced cancer. Cohorts of patients with progressive solid tumours received escalating doses of HuMV833 as a 1-h intravenous (I.V.) infusion on days 1, 15, 22, and 29. Twenty patients (median Eastern Cooperative Oncology Group (ECOG) score 1) were accrued. HuMV833 infusions were well tolerated and there were no grade III or IV toxicities definitely related to the antibody. Grade I or II toxicities probably related to the antibody included fatigue, dyspnoea and rash. There were two episodes of asymptomatic hypocalcaemia, one at grade III and one grade IV, which were recorded in early follow-up. There were eight grade I episodes of asymptomatic elevation of activated partial thromboplastin time (APTT) and two grade III events; one in a patient receiving 1 mg/kg and the other receiving extended doses of 10 mg/kg. Pharmacokinetic analysis revealed a non-linear kinetic and an elimination half-life of between 8.2 (0.3 mg/kg) and 18.7 (10 mg/kg) days. One patient with ovarian cancer experienced a partial response (PR) of 9 months duration and eight had disease stabilisation (SD) including one patient with colorectal carcinoma whose disease was stable for 14 months. In 13 of the 14 samples taken from 12 patients, the plasma concentration of hepatocyte growth factor (HGF) was reduced 24 h after drug administration. HuMV833 is safe and lacked DLT at doses up to 10 mg/kg on this schedule. Multiple doses were well tolerated, despite occasional asymptomatic elevations in APTT. By combining pharmacokinetic, pharmacodynamic and toxicity data, we can identify doses of 1 and 3mg/kg for further investigation. HuMV833 appears to possess some clinical activity.

Purpose: CDP860 is an engineered Fab' fragment-polyethylene glycol conjugate, which binds to and blocks the activity of the beta-subunit of the platelet-derived growth factor receptor (PDGFR-beta). Studies in animals have suggested that PDGFR-beta inhibition reduces tumor interstitial fluid pressure, and thus increases the uptake of concomitantly administered drugs. The purpose of this study was to determine whether changes in tumor vascular parameters could be detected in humans, and to assess whether CDP860 would be likely to increase the uptake of a concurrently administered small molecule in future studies., Patients and Methods: Patients with advanced ovarian or colorectal cancer and good performance status received intravenous infusions of CDP860 on days 0 and 28. Patients had serial dynamic contrast-enhanced magnetic resonance imaging studies to measure changes in tumor vascular parameters., Results: Three of eight patients developed significant ascites, and seven of eight showed evidence of fluid retention. In some patients, the ratio of vascular volume to total tumor volume increased significantly (P < .001) within 24 hours following CDP860 administration, an effect suggestive of recruitment of previously non-functioning vessels., Conclusion: These observations suggest that inhibition of PDGFR-beta might improve delivery of a concurrently administered therapy. However, in cancer patients, further exploration of the dosing regimen of CDP860 is required to dissociate adverse effects from beneficial effects. The findings challenge the view that inhibition of PDGF alone is beneficial, and confirm that effects of PDGFR kinase inhibition mediate, to some extent, the fluid retention observed in patients treated with mixed tyrosine kinase inhibitors.

Platelet-derived growth factors (PDGFs) and their tyrosine kinase receptors (PDGFRs) have been implicated in the pathogenesis of a number of tumor types and play an important role in angiogenesis. Tumor growth can be promoted by PDGF via autocrine stimulation of malignant cells, by overexpression or overactivation of PDGFRs, or by stimulation of angiogenesis within the tumor. These mechanisms could provide possible therapeutic targets. PDGFR blockade may also lower the interstitial fluid pressure within solid tumors and enhance drug delivery. Here we discuss the possible therapeutic roles of PDGFR antagonists in the treatment of cancer, alone and in combination with chemotherapy or other targeted agents. Extensive experimental data highlight the potential therapeutic advantage of targeting PDGFR. However, recent clinical data suggest that antagonism of this growth factor is associated with fluid accumulation that could obscure any clinical benefit. Further clinical research is required to optimise inhibition of this cytokine-receptor system.

Many novel antiangiogenic agents are currently in various phases of clinical testing. These agents tend to be cytostatic, and therefore few responses are observed with conventional imaging by computerized tomography. Furthermore, toxicity with these agents is seen when the maximum-tolerated dose is combined with chemotherapy. Hence, there is a need to develop imaging strategies that can determine the minimum and optimum biologically active doses. There is increasing awareness of the need to obtain evidence of drug activity through the use of surrogate markers of the biologic mechanism of action during early clinical trials, in addition to determining the pharmacokinetics, toxicity profile, and maximum-tolerated dose. One of the major impediments to the rapid development of antiangiogenic agents in the past has been the lack of validated assays capable of measuring an antiangiogenic effect directly in patients. Recently, dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) has emerged as a useful technique for noninvasive imaging of tumor vasculature in preclinical and clinical models. The problem of tumor heterogeneity remains to be addressed. The major challenge is the standardization of the technique worldwide for the purpose of early clinical studies that are likely to be multicenter. Convincing data on correlations between changes observed through molecular imaging and changes in tumor angiogenesis, and hence tumor biology, are still lacking. Whether this would translate into a survival advantage remains to be seen. The ultimate test of the surrogate biological end points determined by molecular imaging will occur in randomized phase III trials. Results of the first randomized trial that showed a survival advantage in favor of antiangiogenic agents were released at the American Society of Clinical Oncology meeting in 2003. There it was reported that the combination of 5-fluorouracil, leucovorin, and irinotecan (Camptosar; Pfizer Pharmaceuticals; New York, NY) with anti-vascular endothelial growth factor antibody (bevacizumab-Avastin; Genentech, Inc.; South San Francisco, CA) was superior to the chemotherapy regimen alone when used to treat patients with metastatic colorectal cancer. However, until further phase III clinical trials confirm these results, surrogate end points of clinical efficacy of the newer agents are urgently needed so that development of ineffective drugs can be halted early. This review briefly discusses the role of molecular imaging in general, and DCE-MRI in particular, in relation to treatment with antiangiogenic agents and highlights some of the difficulties encountered in this area.

Motion during time-series data acquisition causes model-fitting errors in quantitative dynamic contrast-enhanced (DCE) MRI studies. Motion correction techniques using conventional registration cost functions may produce biased results because they were not designed to deal with the time-varying information content due to contrast enhancement. We present a locally-controlled, 3D translational registration process driven by tracer kinetic modeling that successfully registers abdominal DCE-MRI data at high temporal resolution and compare this method to a similar approach based on registration to the time series mean image in data from 8 patients. When the registration is driven by an appropriate model, we find significant improvements in model-fitting. Also, model-driven registration influences parameter estimates and reduces repeat study variability in measurements of blood volume.

Background: Chemotherapy with a platinum agent and a taxane (paclitaxel) is considered the standard of care for treatment of ovarian carcinoma. We compared the combination of docetaxel-carboplatin with the combination of paclitaxel-carboplatin as first-line chemotherapy for stage Ic-IV epithelial ovarian or primary peritoneal cancer., Methods: We randomly assigned 1077 patients to receive docetaxel at 75 mg/m2 of body surface area (1-hour intravenous infusion) or paclitaxel at 175 mg/m2 (3-hour intravenous infusion). Both treatments then were followed by carboplatin to an area under the plasma concentration-time curve of 5. The treatments were repeated every 3 weeks for six cycles; in responding patients, an additional three cycles of single-agent carboplatin was permitted. Survival curves were calculated by the Kaplan-Meier method, and hazard ratios were estimated with the Cox proportional hazards model. All statistical tests were two-sided., Results: After a median follow-up of 23 months, both groups had similar progression-free survival (medians of 15.0 months for docetaxel-carboplatin and 14.8 months for paclitaxel-carboplatin; hazard ratio [HR] docetaxel-paclitaxel = 0.97, 95% confidence interval [CI] = 0.83 to 1.13; P = .707), overall survival rates at 2 years (64.2% and 68.9%, respectively; HR = 1.13, 95% CI = 0.92 to 1.39; P = .238), and objective tumor (58.7% and 59.5%, respectively; difference between docetaxel and paclitaxel = -0.8%, 95% CI = -8.6% to 7.1%; P = .868) and CA-125 (75.8% and 76.8%, respectively; difference docetaxel-paclitaxel = -1.0%, 95% CI = -7.2% to 5.1%; P = .794) response rates. However, docetaxel-carboplatin was associated with substantially less overall and grade 2 or higher neurotoxicity than paclitaxel-carboplatin (grade > or =2 neurosensory toxicity in 11% versus 30%, difference = 19%, 95% CI = 15% to 24%; P<.001; grade > or =2 neuromotor toxicity in 3% versus 7%, difference = 4%, 95% CI = 1% to 7%; P<.001). Treatment with docetaxel-carboplatin was associated with statistically significantly more grade 3-4 neutropenia (94% versus 84%, difference = 11%, 95% CI = 7% to 14%; P<.001) and neutropenic complications than treatment with paclitaxel-carboplatin, although myelosuppression did not influence dose delivery or patient safety. Global quality of life was similar in both arms, but substantive differences in many symptom scores favored docetaxel., Conclusions: Docetaxel-carboplatin appears to be similar to paclitaxel-carboplatin in terms of progression-free survival and response, although longer follow-up is required for a definitive statement on survival. Thus, docetaxel-carboplatin represents an alternative first-line chemotherapy regimen for patients with newly diagnosed ovarian cancer.

Objectives: There is an urgent need for new agents with activity in platinum- and taxane-resistant epithelial ovarian cancer. Exatecan mesylate is a novel topoisomerase I inhibitor with potent activity against ovarian cancer in vitro. A multicentre phase IIA study was conducted in patients with platinum- and taxane-resistant epithelial ovarian cancer., Patients and Methods: Fifty-seven patients with bidimensionally measurable ovarian cancer, previously exposed to platinum and taxanes, whose disease had relapsed within 6 months of platinum-containing chemotherapy were randomised to one of two intravenous schedules of exatecan mesylate; 0.3 mg/m(2) daily for 5 days every 3 weeks (Arm A) or 2.1 mg/m(2) weekly for 3 weeks out of 4 (Arm B)., Results: There were no responses in the weekly arm and a radiological response rate of 5.3% (95% CI 0.3-21.8%) in the daily arm. Principal toxicities were myelosuppression and emesis. Grade 3/4 neutropenia occurred in 29% of patients in Arm A and 6% patients in Arm B. Seventy-one percent of patients in Arm A required red cell transfusions while on treatment., Conclusions: Exatecan is well tolerated in this poor prognosis group of patients but only has modest single agent activity when administered in a daily regimen.

Purpose: Heparan sulfate proteoglycans have been implicated in cancer cell growth, invasion, metastasis, and angiogenesis. This study was designed to compare their expression in normal ovary and ovarian tumors and then to examine their prognostic significance in ovarian cancer., Experimental Design: The expression of syndecan-1, -2, -3, and -4, glypican-1, and perlecan was assessed by immunohistochemistry in 147 biopsies that included normal ovary and benign, borderline, and malignant ovarian tumors. Clinical data, including tumor stage, performance status, treatment, and survival, were collected. Univariate and multivariate analyses were performed to evaluate prognostic significance., Results: The expression patterns of syndecan-1 and perlecan were altered in ovarian tumors compared with normal ovary. Syndecan-1 was not detected in normal ovary but was present in the epithelial and stromal cells of benign and borderline tumors and in ovarian adenocarcinomas. Perlecan expression was decreased in basement membranes that were disrupted by cancer cells but maintained in the basement membranes of blood vessels. Syndecan-2, -3, and -4, and glypican-1 were expressed in normal ovary and benign and malignant ovarian tumors. Stromal expression of syndecan-1 and glypican-1 were poor prognostic factors for survival in univariate analysis., Conclusion: We report for the first time distinct patterns of expression of cell surface and extracellular matrix heparan sulfate proteoglycans in normal ovary compared with ovarian tumors. These data reinforce the role of the tumor stroma in ovarian adenocarcinoma and suggest that stromal induction of syndecan-1 contributes to the pathogenesis of this malignancy.

The beta-endoglucuronidase heparanase plays an important role in tumor invasion, a process that is significantly enhanced by hypoxia. We have used a strategy of stable transfection with antisense to derive ovarian carcinoma cell lines that express different levels of heparanase and used these to demonstrate that invasion correlates with heparanase activity. Secreted heparanase activity was increased by reduction, hypoxia, and growth of cells under reduced oxygen (1%) augmented heparanase activity and invasion, both of which are inhibited by treatment with antiheparanase antibodies. This is the first demonstration that heparanase activity may be regulated by microenvironmental redox conditions, which influence invasion, and that invasion can be blocked with specific heparanase-neutralizing antibodies.

The number of anti-angiogenic agents developed for clinical use has risen greatly over the past decade. Currently, more than 80 are in trials ranging from phase I through to phase III studies and many more are in preclinical evaluation. Much hope was envisaged for these new agents to become the panacea of anti-tumoural treatment. Unfortunately the single agent activity to date has proven to be disappointing although one trial has recently reported a survival advantage when chemotherapy was administered with anti-VEGF antibodies in the setting of advanced colorectal cancer. To an extent, this may be due to great expectations of cytostatic compounds, but recently many factors have been examined to explain the differences between clinical and experimental findings. In this review, some of the factors responsible for the discrepancy are examined, with a specific focus on inhibitors of VEGF. The key factors responsible for the lack of activity are tumour heterogeneity and redundancy in the VEGF signalling system. An increased understanding of these factors is critical to the development of effective anti-angiogenic agents and need to be taken into account as new generations of drugs emerge.

The development of agents that target tumour vasculature is ultimately dependent on the availability of appropriate preclinical screening assays. Several quantitative angiogenesis assays exist, each with its own unique characteristics and disadvantages. In this review we discuss some of the commonly used assays, their methodological pitfalls and current use. The corneal micropocket and the CAM assay are well established. However, the matrix-implant assays have the potential advantage of replicating the hypoxic tumour microenvironment, thus making them suitable for the study of tumour angiogenesis. The ideal quantitative angiogenesis assay does not exist and the use of two complimentary quantitative assays, such as a matrix implant assay and a microcirculatory preparation like the CAM or corneal micropocket assay, provides the best compromise. Newer models like the hollow-fibre assay are being developed and older ones refined. Assay systems should reflect distinct disease processes. Thus it is appropriate to develop assays that study exclusively pro- or anti-angiogenic compounds or anti-vascular agents. Criticisms of currently available screening systems are that the predictive value of current screening systems remains to be established as anti-angiogenic agents are still in clinical development. Anti-angiogenic agents are likely to be most effective as chronic therapy for remission maintenance in the metastatic setting or as adjuvant therapy in patients at high risk of relapse, an important clinical aspect not addressed in animal models of tumour angiogenesis. Histological analysis still provides the most detailed information on in vivo angiogenesis. However, angiogenesis is a dynamic process and assays that permit continuous monitoring of the angiogenic response and provide information on the physiological characteristics of new vessels will be distinctly advantageous over older systems. The development of non-invasive techniques for quantitation of angiogenesis will greatly facilitate this process.

Bryostatin-1 is a macrocyclic lactone whose main mechanism of action is protein kinase C modulation. We investigated its activity as a weekly 24-h infusion in recurrent ovarian carcinoma. In all, 17 patients were recruited and 11 had chemotherapy-resistant disease as defined by disease progression within 4 months of last cytotoxic therapy. All were evaluable for toxicity and 14 for response. There were no disease responses and the main toxicity was myalgia.

Endostatin is a naturally occurring proteolytic fragment of the C-terminal domain of collagen XVIII. It inhibits angiogenesis by a mechanism that appears to involve binding to HS (heparan sulphate). We have examined the molecular interaction between endostatin and HS from micro- and macrovessel endothelial cells. Two discrete panels of oligosaccharides were prepared from metabolically radiolabelled HS, using digestion with either heparinase I or III, and then examined for their endostatin affinity using a sensitive filter-binding assay. Two types of endostatin-binding regions were identified: one comprising sulphated domains of five or more disaccharides in length, enriched in 6-O-sulphate groups, and the other contained long heparinase I-resistant fragments. In the latter case, evidence from the present study suggests that the binding region encompasses a sulphated domain fragment and a transition zone of intermediate sulphation. The contribution to binding of specific O-sulphate groups was determined using selectively desulphated HS species, namely HS from Hs2st-/- mutant cells, and by comparing the compositions of endostatin-binding and non-binding oligosaccharides. The results indicate that 6-O-sulphates play a dominant role in site selectivity and 2-O-sulphates are not strictly essential.

Vascular endothelial growth factor-C (VEGF-C) and VEGF-D are members of the VEGF family of cytokines and have angiogenic and lymphangiogenic actions. In gastric adenocarcinoma, VEGF-C mRNA and tissue protein expression correlate with lymphatic invasion, lymph node metastasis and in some reports, venous invasion and reduced 5-year survival. Patients with gastric adenocarcinomas containing high levels of VEGF-C expression have significantly reduced 5-year survival rates, and VEGF-C expression is an independent prognostic risk factor for death. The role of VEGF-C in oesophageal squamous and colorectal cancer and VEGF-D in colorectal cancer is not clear, with conflicting reports in the published literature. In order to exploit potential therapeutic applications, further research is necessary to define the precise roles of these cytokines in health and disease.

A large number of drugs have been used to treat recurrent ovarian cancer, yet there are few data that guide the physician's choice. Typically, the decision to re-treat with platinum-based therapy depends on the progression-free interval. However, the optimum agent for the treatment of platinum-resistant or refractory disease is not defined. In this study, we investigated the efficacy of oral melphalan in patients who have platinum refractory or resistant disease. A retrospective analysis was performed on 22 patients with ovarian carcinomas who had relapsed within 6 months of their platinum-based chemotherapy and were treated with oral melphalan. No objective responses were seen and the median overall survival was 3 months from commencement of therapy. Although the treatment was generally well tolerated, only two of the 22 patients managed to complete the planned six cycles of treatment. At the time of analysis, only two patients were alive. Other nonplatinum compounds have demonstrated response rates in the region of 20% in similar patient populations and it is unlikely that any positive responses could have been missed by chance (95% CI 0-15.4). The results of this study serve to eliminate oral melphalan as a treatment option in patients with platinum-resistant or refractory ovarian carcinoma.

Malignant ascites occurs in association with a variety of neoplasms. It is a frequent cause of morbidity and presents significant problems for which there are no clear management guidelines. In this article we discuss various modalities which are available including diuretic therapy, paracentesis, peritoneovenous shunts and intraperitoneal chemotherapy. There are no randomized trials of diuretic drugs to assess their efficacy in malignant ascites. Phase II data suggest that they are effective in approximately one-third of patients with malignancy, and their efficacy may be determined by plasma renin/aldosterone concentrations. Paracentesis provides relief in up to 90% of patients; because of varying reports of hypovolaemia, some advocate simultaneous intravenous fluid infusion. Permanent percutaneous drains may prevent the need for repeated paracentesis, although there is potential for infection. A peritoneovenous shunt also prevents the need for repeated paracenteses, whilst maintaining normal serum albumin concentrations. Blockage occurs in 25% of shunts, which are contraindicated in the presence of heavily bloodstained ascites because of the risk of occlusion. The preclinical and clinical experience with anti-angiogenic agents such as the matrix metalloproteinase inhibitors and the VEGF antagonists suggests that these agents may have a role in the treatment of malignant ascites.

We report our experience in the management of patients with carcinosarcoma of the ovary, a rare but aggressive variant of ovarian cancer. Forty patients were treated at a single centre, which is the largest reported series. The median age at diagnosis was 65 years (range 45-86) and the median Karnofsky performance (KP) status was 70. Thirty-two patients (80%) presented with FIGO stage III or IV disease. Twenty-four had heterologous and 14 homologous carcinosarcoma on review of histopathology, but there was no significant difference in survival between these groups (P=0.28). Twenty-seven of the 40 patients had bulk residual disease present after surgery and this was associated with a worse prognosis (P=0.045). Chemotherapy was given to 32 patients (80%) of whom 26 (81%) received platinum-based regimens. Of these 32 patients, three (9.4%) achieved a complete response (CR), 10 (31%) a partial response (PR), five (16%) had stable disease, 10 (31%) had progressive disease and four were not assessable. Of the 19 patients who had a CR, PR or stable disease after chemotherapy or were unevaluable (stage Ic), the median survival was 29.6 months. Currently, seven patients are still alive although one has cancer. The overall censored median survival was 8.7 months after a median follow-up of 34 months, and the 1- and 5-year survival were 40 and 7.5%, respectively.

We have investigated the reproducibility of dynamic contrast enhanced imaging techniques in nine patients with cerebral glioma. Patients were imaged twice with a 2 day interval between scans. Maps were produced of the time taken to achieve 90% enhancement (T90), the maximal intensity change per time interval ratio (MITR), the volume transfer coefficient between plasma and the extravascular extracellular space (K(trans)) and the extravascular extracellular contrast distribution volume, v(e). Measurements of K(trans) greater than 1.2 min(-1) were used to exclude pixels where first pass perfusion effects dominated the measurement. Measures of the test-retest coefficient of variation (CoV) and intraclass correlation coefficients were used to assess reproducibility for measurements from a volume of interest containing enhancing tissue from the whole tumour. MITR showed poor reproducibility (mean CoV 17.9%, 95% confidence limits for group comparisons 20.2%). T90 showed good reproducibility (mean CoV 7.1%, 95% confidence limits for group comparisons 5.2%). Calculated values of K(trans) and v(e) also showed good reproducibility (mean CoV 7.7% and 6.2% respectively, 95% confidence limits for group comparisons 6.2% and 4.8%, respectively). We conclude that the measurements of K(trans) and v(e) derived from pharmacokinetic analysis are sufficiently reproducible to support their use as a biological markers in therapeutic trials.

Background: Vascular endothelial growth factor (VEGF) is a potent angiogenic cytokine, and various inhibitory agents, including specific antibodies, have been developed to block VEGF-stimulated angiogenesis. We developed HuMV833, a humanized version of a mouse monoclonal anti-VEGF antibody (MV833) that has antitumor activity against a number of human tumor xenografts, and investigated the distribution and biologic effects of HuMV833 in patients in a phase I trial., Methods: Twenty patients with progressive solid tumors were treated with various doses of HuMV833 (0.3, 1, 3, or 10 mg/kg). Positron emission tomography with (124)I-HuMV833 was used to measure the antibody distribution in and clearance from tissues. Magnetic resonance imaging was used to measure the vascular permeability surface area product with a first-pass pharmacokinetic model (k(fp)) to determine tumor vascular permeability., Results: The antibody was generally well tolerated, although the incremental dose, phase I study design, and pharmacodynamic endpoints could not identify the optimum biologically active dose. Antibody distribution and clearance were markedly heterogeneous between and within patients and between and within individual tumors. HuMV833 distribution to normal tissues also varied among patients, but the antibody was cleared from these tissues in a homogeneous fashion. Permeability was strongly heterogeneous between and within patients and between and within individual tumors. All tumors showed a reduction in k(fp) 48 hours after the first treatment (median = 44%; range = 4%-91%)., Conclusions: Because of the heterogeneity in tumor biology with respect to antibody uptake and clearance, we suggest that either intrapatient dose escalation approaches or larger, more precisely defined patient cohorts would be preferable to conventional strategies in the design of phase I studies with antiangiogenic compounds like HuMV833.

Over the last decade assessment of angiogenesis has emerged as a potentially useful biological prognostic and predictive factor in human solid tumours. With the development of highly specific endothelial markers that can be assessed in histological archival specimens, several quantitative studies have been performed in various solid tumours. The majority of published studies have shown a positive correlation between intra-tumoural microvessel density, a measure of tumour angiogenesis, and prognosis in solid tumours. A minority of studies have not demonstrated an association and this may be attributed to significant differences in the methodologies employed for sample selection, immunostaining techniques, vessel counting and statistical analysis, although a number of biological differences may account for the discrepancy. In this review we evaluate the quantification of angiogenesis by immunohistochemistry, the relationship between tumour vascularity and metastasis, and the clinicopathological studies correlating intra-tumoral microvessel density with prognosis and response to anti-cancer therapy. In view of the extensive nature of this retrospective body of data, comparative studies are needed to identify the optimum technique and endothelial antigens (activated or pan-endothelial antigens) but subsequently prospective studies that allocate treatment on the basis of microvessel density are required., (comCopyright 2002 Cancer Research UK)

Heparan sulfate proteoglycans (HSPGs) are widely distributed in mammalian tissues and involved in a number of processes related to malignancy. They are composed of a core protein to which chains of the glycosaminoglycan, heparan sulfate (HS), are attached. The existence of various classes of core protein, in addition to highly polymorphic HS chains, creates a superfamily of macromolecules with considerable diversity of structure and function. HSPGs interact with many proteins including growth factors, chemokines and structural proteins of the extracellular matrix to influence cell growth, differentiation, and the cellular response to the environment. The recent identification of two inherited syndromes that are associated with an increased cancer risk, and caused by mutations in HSPG-related genes, has intensified interest in these molecules. This review describes our current understanding of HSPGs in cancer and highlights new possibilities for therapeutic control., (Copyright 2001 Cancer Research Campaign http://www.bjcancer.com.)

The majority of cancer have an absolute requirement for angiogenesis, the process by which new blood vessels are formed. The most potent angiogenic cytokine is vascular endothelial growth factor (VEGF) and there has been substantial research into the development of VEGF/VEGF receptor (VEGFR) antagonists. To date these strategies have included gene therapy techniques that deliver antisense oligonucleotides, soluble VEGFRs that function in a dominant negative fashion and ribozymes. Additional strategies have included the development of receptor tyrosine kinase (RTK) inhibitors and monoclonal antibodies (mAbs) directed against VEGF or the signalling receptor. The most promising agents appear to be the monoclonal anti-VEGF antibodies and the RTK inhibitors as these have demonstrated broad spectrum antitumour activity in vivo and single agent activity in early phase clinical trials in patients with advanced pre-treated breast and colorectal carcinoma and Kaposi's sarcoma. The RTK inhibitors are of particular interest as they can be administered by mouth. Collation of the early clinical trial data suggests that VEGF antagonists are largely well-tolerated but may be associated with vascular toxicities such as haemorrhage and thromboembolic events. Combination studies of chemotherapy and VEGF antagonists are underway but the benefit of these regimens will need to be established in adequately powered Phase III studies. Potentially these agents may play a role in the treatment of both early (adjuvant) and advanced cancer. The efficacy of the drugs will be explored in a number of non-malignant conditions including rheumatoid arthritis (RA), psoriasis, diabetic retinopathy and possibly as non-steroidal contraceptives but the overall clinical development of these agents can only be optimised if appropriate biological end points are identified and incorporated into clinical trials.

Bryostatin 1 is a naturally occurring macrocyclic lactone with promising antitumour and immunomodulatory function in preclinical and phase I clinical investigations. In this phase II study, 17 patients with progressive non-Hodgkin's lymphoma of indolent type (NHL), previously treated with chemotherapy, received a median of 6 (range 1-9) intravenous infusions of 25 microg/m(2) bryostatin 1 given once weekly over 24 hours. In 14 evaluable patients no responses were seen. Stable disease was attained in one patient for 9 months. The principal toxicities were myalgia and phlebitis. Treatment was discontinued early because of toxicity alone (phlebitis) in 2 patients, toxicity in addition to progressive disease in 3 patients (myalgia and phlebitis n = 2; thrombocytopenia n = 1) and progressive disease in 5 patients. The results fail to demonstrate efficacy of this regimen of bryostatin 1 in the treatment of NHL. In light of preclinical data that demonstrate synergy between bryostatin 1 and several cytotoxic agents and cytokines, clinical studies to investigate bryostatin 1 in combination are warranted. We also present data to demonstrate that central venous lines may be used in future studies to avoid phlebitis.

Background and Purpose: Regimens containing platinum drugs continue to be amongst the most effective therapies for ovarian cancer. However, despite high initial response rates most patients relapse and die of their disease. Elevation of metallothionein has been implicated as a mechanism by which tumour cells become resistant to platinum anticancer drugs, although most of these studies have been carried out in vitro. This study was carried out to determine whether metallothionein expression was associated with response or survival in patients with epithelial ovarian cancer., Methods: Metallothionein was determined by radioimmune assay using frozen ovarian tumour tissue taken either before or following cytotoxic chemotherapy., Results: An increase in expression of metallothionein was seen in tumour tissue from patients who had undergone cytotoxic chemotherapy, although this did not attain significance. However, a preliminary study using biopsy material from the same patient, taken both before and after chemotherapy, showed a statistically significant increase in metallothionein. An analysis of these data showed that the level of metallothionein expression was not associated with survival or response., Conclusion: These data do not support the hypothesis that metallothionein expression is a determinant of response in ovarian cancer. There is some preliminary evidence from the study of paired samples which indicates that cytotoxic chemotherapy may increase metallothionein expression. An increase in metallothionein was also seen in the study using unmatched biopsies although this did not attain statistical significance, due in part to the large inter-patient variability in expression of this protein.

We conducted a retrospective immunohistochemical evaluation of the prognostic significance of the expression of p53 and the related proteins Bax, Bcl-2, growth arrest and DNA damage (Gadd45), murine double minute 2 (Mdm2) and p21(WAF1/CIP1) in chemonaive tumours taken from 66 patients with ovarian cancer. Ki-67 expression (a marker of cell proliferation) was also evaluated immunohistochemically, while apoptosis within malignant cells was determined with the terminal deoxynucleotidyltransferase-mediated dUTP nick-end labelling (TUNEL) assay. The expression of each of the following proteins was significantly associated in the tumours (P < 0.05 unless otherwise stated): Bax with Bcl-2 (P < 0.01); Bax with Mdm2; p21(WAF1/CIP1) with Gadd45 (P < 0.01); p21(WAF1/CIP1) with p53; p53 with Mdm2. Univariate analysis showed that expression of p53, Bax, bulk residual disease and International Federation of Gynecology and Obstetricians (FIGO) stage were all strongly correlated with response to chemotherapy (P < 0.01). Similarly, the FIGO stage and Ki-67 expression (P < 0.01), as well as pathological subtype and bulk residual disease (P < 0.05), were prognostic factors for disease progression. The FIGO stage and Ki-67 expression were significant prognostic factors for overall survival (P < 0.01), with Gadd45 expression and pathological subtype also significant (P < 0.05) in a univariate analysis. Multivariate analysis for response to chemotherapy showed that expression of p53, Bax and FIGO stage were all independent prognostic factors (P < 0.01). The FIGO stage was the most important independent prognostic factor for progression and survival on multivariate analysis (P < 0.01). However, Ki-67 expression was also an independent prognostic factor for disease progression (P < 0.05) and approached significance for survival (P = 0.055). Taken together, these data suggest that determination of Ki-67 expression could supplement established prognostic factors.

Four hundred and twenty-nine patients received myeloablative chemotherapy for solid and haematological malignancies in a bone marrow transplantation unit. Regimens appropriate to the tumour type were administered and haemopoietic reconstitution was achieved with peripheral blood progenitor cells (PBPC; n = 275), autologous bone marrow (auto-BMT; n = 69) or allogeneic bone marrow (allo-BMT; n = 85). World Health Organization (WHO) oral mucositis scores were collected prospectively from the start of chemotherapy (d 1) until d 28 or discharge. Oral mucositis (OM) was experienced by 425 (99%) patients and in 289 (67.4%) this was grade III or IV. Strong opiate analgesia was prescribed for a median of 6 d to 47% of patients. Univariate analysis suggested that the area under the OM curve (AUC; sum of daily mucositis grades, d 1-28) was associated with the myeloablative regimen, haemopoietic progenitor source (PBPC > allo-BMT > auto-BMT), use of myeloid growth factors and age. Multivariate analysis showed that the only independent risk factor for mucositis was the conditioning regimen (P < 0.00005). The mean OM AUC for high-dose melphalan (HDM) regimens (52 grade-days) exceeded busulphan (41), busulphan-cyclophosphamide (35), cyclophosphamide-total body irradiation (TBI) (34), cyclophosphamide-carmustine (BCNU) (20) and cyclophosphamide-etoposide-carmustine (CVB) (19). HDM regimens resulted in the highest mean peak OM (3.6), followed by busulphan regimens (2.6), cyclophosphamide/TBI (2.3) and cyclophosphamide-carmustine and CVB (1.4). Busulphan produced significantly delayed OM (median 3 d; P < 0.00005). There was a linear association between the area under the OM curve for each treatment group and the time to reach grade 3 OM (P < 0.00005), but no association with the time to reach grade 4 neutropenia (P = 0.24) or thrombocytopenia (P = 0.73), implying that haematological and mucosal toxicity are not associated. The cytotoxic regimen is the most significant determinant of OM. Studies investigating agents to ameliorate mucosal toxicity should be stratified according to cytotoxic regimen.

Background: A randomised phase II study was performed to compare standard combination chemotherapy containing cisplatin and etoposide with infusional carboplatin., Patients and Methods: One hundred twenty patients with locally advanced/metastatic non-small-cell lung cancer or mesothelioma were enrolled. All were chemotherapy-naïve and had a Karnofsky performance status of > or = 50. Patients were randomised to either four cycles of bolus therapy of cisplatin 80 mg/m2 day 1, etoposide 120 mg/m2 day 1-3, or continuous infusion of carboplatin 100/mg/m2/week for six weeks., Results: No patients on infusional therapy incurred grade 3-4 toxicity while in the bolus arm, grade 3 and grade 4 leucopenia occurred in 17% and 35% of patients, respectively. Grade 4 thrombocytopenia occurred in 8% of patients and there were two instances of grade 3 renal toxicity. No responses occurred in the pump arm. Eight of forty-six patients with non-small-cell lung cancer responded to treatment (response rate 17.3%) with two complete responses and six partial responses. Only one patient with mesothelioma responded to bolus therapy. There was no difference in survival for the subset of NSCLC patients. Survival for mesothelioma patients in the pump arm was superior but this was likely to be a result of early deaths in the bolus arm., Conclusions: The pump arm was well-tolerated but not active, whilst combination platinum-based therapy demonstrated activity but significantly more toxicity than the pump arm. Further studies of infusional carboplatin with this schedule are not warranted.

The distinction between borderline ovarian tumours (BOT) and ovarian carcinoma is made by histopathological assessment. Of 64 patients managed according to institutional BOT protocols, 27 (42%) had been referred with a diagnosis of ovarian carcinoma that was subsequently changed to BOT following histopathological review. The 70% 6-year event-free survival of the patients with a revised diagnosis was not significantly different from those who were referred with a diagnosis of BOT. This change in diagnosis is important as it avoids the need for chemotherapy for most patients and results in patients receiving appropriate information concerning prognosis. Interestingly, 24 patients (38.1%) reported a family history of epithelial cancer, a finding that has not been reported previously. Campaign

Basic fibroblast growth factor (bFGF) is dependent on heparan sulphate for its ability to activate the cell surface signal transducing receptor. We have investigated the FGF dual receptor mechanism in a novel model of the transformation from human colon adenoma to carcinoma in vitro. Reverse transcription-polymerase chain reaction showed that mRNA for FGF receptors 1 and 2 were expressed in both the adenoma and carcinoma cells whereas immunocytochemistry showed that the expression of the FGF R1 was reduced significantly in the carcinoma cells. We have reported previously that the composition and sequence of human colon adenoma and carcinoma heparan sulphate (HS) differ in a defined and specific manner. The functional significance of these changes was assessed by affinity co-electrophoresis, which showed that the affinity of adenoma HS for bFGF was 10-fold greater than that of the carcinoma HS (Kd 220 nM vs. 2493 nM, respectively). In addition, Northern studies of the expression of syndecan 1 and 4 mRNA showed that proteoglycan core protein expression was reduced significantly in the carcinoma cells. These findings were associated with a reduced biological response to bFGF in the carcinoma cells that could be partially reversed by the addition of exogenous heparin, suggesting that both the proteoglycan and signal transducing receptor control the cells' response to bFGF.

Ovarian cancer is the commonest cause of gynaecological cancer death in the UK, and guidelines for initial surgery and staging of this disease are widely available. We report a retrospective audit of the surgical management of patients with newly diagnosed ovarian cancer referred to the Christie Cancer Centre in Manchester in 1996. The aim was to assess compliance with surgical guidelines. The authors found that the majority of patients (92%) presented via an outpatient clinic and for these individuals surgery was therefore elective. This mode of presentation should allow management by a small number of dedicated gynaecologists at each hospital, but up to seven consultants in each hospital performed surgery on a relatively small number of patients. Furthermore, less than half the patients underwent the recommended surgical procedure. Although some patients may have 'inoperable' disease, these data suggest that a greater compliance with national and international guidelines are required to provide an optimal level of care.

We report a detailed analysis of heparan sulfate (HS) structure using a model of human colon carcinogenesis. Metabolically radiolabeled HS was isolated from adenoma and carcinoma cells. The chain length of HS was the same in both cell populations (Mr 20,000; 45-50 disaccharides), and the chains contained on average of two sulfated domains (S domains), identified by heparinase I scission. This enzyme produced fragments of approximate size 7 kDa, suggesting that the S domains were evenly spaced in the intact HS chain. The degree of polymer sulfation and the patterns of sulfation were strikingly different between the two HS species. When compared with adenoma HS, the iduronic acid 2-O-sulfate content of the carcinoma-derived material was reduced by 33%, and the overall level of N-sulfation was reduced by 20%. However, the level of 6-O-sulfation was increased by 24%, and this was almost entirely attributable to an enhanced level of N-sulfated glucosamine 6-O-sulfate, a species whose data implied was mainly located in the mixed sequences of alternating N-sulfated and N-acetylated disaccharides. The results indicate that in the transition to malignancy in human colon adenoma cells, the overall molecular organization of HS is preserved, but there are distinct modifications in both the S domains and their flanking mixed domains that may contribute to the aberrant behavior of the cancer cell.

Carcinomatous meningitis (CM) is an uncommon but devastating complication of malignancy. The management is controversial and clear recommendations cannot be made because: 1) Most series include patients with CM that has arisen from different primary malignancies which are associated with different median survival intervals. 2) There have been no prospective randomised investigations of treatment modalities in patients with CM from a particular tumour type. 3) The definition of response varies from one report to another so that some response rates refer to cytological changes in the CSF while others take clinical, cytological and biochemical parameters into account. 4) Reports include patients with and without parenchymal metastases and the natural history of carcinomatous meningitis in the two situations may differ. The median survival of solid tumour carcinomatous meningitis (excluding leukaemia and lymphoma) is approximately 2-3 months and patients with breast cancer have the longest survival (median 3 months). Currently patients are treated with radiotherapy to part or all of the neuraxis with either intrathecal or intravenous chemotherapy but the relative contribution of these modalities to survival or quality of life remains unknown. Approximately 50% of patients with carcinomatous meningitis die from other causes, including systemic disease. The two most important endpoints for the patient, neurological improvement and overall survival, are seldom used in isolation in the literature. Many reports have focused on surrogate markers of response, namely biochemical and cytological data points but the correlation between clinical status and these parameters is poor because of differences between lumbar and ventricular CSF and disturbances of CSF flow in CM. The current literature does not provide clear guidelines for the treatment of this condition. Multicentre, prospective, randomised trials should be conducted that address questions of most relevance to the patient, namely neurological status and overall survival.

Background: Carcinomatous meningitis is a rare and often devastating complication in patients with breast cancer, and the treatment is controversial., Methods: A retrospective analysis of 35 patients with carcinomatous meningitis from breast cancer was performed to define the biology of the disease and to guide treatment., Results: An aggressive variant of breast cancer was revealed: meningeal metastasis complicates less than 3.5% of cases of metastatic breast carcinoma. Sixty-seven percent of these patients had tumors that were lobular or combined lobular/ductal histology; the median intervals from primary treatment to disease recurrence and from recurrence to death were 10.9 and 15 months, respectively. The median survival after diagnosis of carcinomatous meningitis was 77 days. The most significant prognostic factor was the Karnofsky performance status (KP) at presentation of meningeal disease. Patients with a KP greater or equal to 70 survived a median of 313 days, whereas those with a KP of 60 or less survived for a median of 36 days (P = 0.0002). In addition, there was a trend suggesting that the response 2 weeks after treatment was initiated, correlated with survival., Conclusions: Carcinomatous meningitis from breast carcinoma is an aggressive metastatic complication with a poor prognosis. The authors suggest that patients with a poor KP (< 70) should be treated symptomatically and those with a good KP (> or = 70) should receive more aggressive treatment. The patients' survival in this study compared well with other reports, and yet, only one patient was treated with intraventricular chemotherapy. Therefore, these data question the superiority of intraventricular treatment versus other modalities.

Basic fibroblast growth factor (bFGF) is found in the extracellular matrix and around the endothelial and epithelial cells of some human colon carcinomas. It is believed to play a role in angiogenesis, but in addition, recent data suggest that it can directly stimulate mitogenesis in some colon carcinoma cell lines. To clarify the role of bFGF in human colon carcinoma, we developed a model of Caco-2 which grew in serum-free conditions so that the effect of bFGF on multiplication, migration, and differentiation could be studied in defined conditions. Through morphological and biochemical studies in serum-free conditions, we demonstrated that this subline of Caco-2 differentiated spontaneously on reaching confluence. Using this model, we found that bFGF did not affect differentiation but that multiplication and migration were increased. The implication of these findings is that bFGF, released from the extracellular matrix by invading cells or produced by neovascular endothelial cells, can increase the mitogenic rate and migratory potential of colon carcinoma cells. In addition, the dual role of bFGF in stimulating colon carcinoma cells directly and promoting angiogenesis suggests that anti-bFGF strategies could form the basis of a novel approach to the treatment of colon carcinoma.