Background: This study characterises molecular effect of bevacizumab, and explores the relation of molecular and genetic markers with response to bevacizumab combined with chemoradiotherapy (CRT). Methods: From a subset of 59 patients of 84 rectal cancer patients included in a phase II study combining bevacizumab with CRT, tumour and blood samples were collected before and during treatment, offering the possibility to evaluate changes induced by one dose of bevacizumab. We performed cDNA microarrays, stains for CD31/CD34 combined with α -SMA and CA-IX, as well as enzyme-linked immunosorbent assay (ELISA) for circulating angiogenic proteins. Markers were related with the pathological response of patients. Results: One dose of bevacizumab changed the expression of 14 genes and led to a significant decrease in microvessel density and in the proportion of pericyte-covered blood vessels, and a small but nonsignificant increase in hypoxia. Alterations in angiogenic processes after bevacizumab delivery were only detected in responding tumours. Lower PDGFA expression and PDGF-BB levels, less pericyte-covered blood vessels and higher CA-IX expression were found after bevacizumab treatment only in patients with pathological complete response. Conclusions: We could not support the ‘normalization hypothesis’ and suggest a role for PDGFA , PDGF-BB, CA-IX and α -SMA. Validation in larger patient groups is needed.

Background: We examined the anti-tumor effect and radiosensitizing potential of a small molecule inhibitor of fibroblast growth factor receptor (FGFR) in colorectal cancer (CRC) in vitro and in vivo.Methods: Effects of in vitro drug treatment on cell survival, proliferation, FGFR signaling, cell cycle distribution, apoptosis and radiosensitivity were assessed using various CRC cell lines with FGFR wild type (Caco2 and HCA7) and FGFR2 amplification (HCT116, NCI-H716). In vivo tumor responses to FGFR inhibition with and without radiation therapy were evaluated by growth delay assays in two colorectal xenograft mouse models (NMRI nu/nu mice injected with NCI-H716 or CaCo2 cells). Mechanistic studies were conducted using Western blot analysis, immunohistochemistry and qPCR.Results: In the tested cell lines, the FGFR inhibitor (JNJ-42756493) was effective in vitro and in vivo in CRC tumors with highest expression of FGFR2 (NCI-H716). In vitro, cell proliferation in this line was decreased, associated with increased apoptotic death and decreased cell survival. In vivo, growth of NCI-H716 tumors was delayed by 5 days by drug treatment alone, although when drug delivery was stopped the relative tumor volume increased compared to control. The FGFR inhibitor did not radiosensitize NCI-H716 tumors either in vitro or in vivo.Conclusions: Among tested CRC cell lines, the growth inhibitory activity of this FGFR inhibitor was evident in cell lines with high constitutive FGFR2 expression, suggesting that FGFR addiction may provide a window for therapeutic intervention, though caution is advised. Preclinical study with NCI-H716 and Caco2 tumor demonstrated that continued presence of drug could be essential for tumor growth control, especially in cells with aberrant FGFR expression. In the tested set-up, the inhibitor showed no radiosensitizing effect.

Abstract Objective: We evaluated the potential of some recently proposed hypoxia markers, being monocarboxylic acid transporter 1 (MCT1), MCT4 and prolyl hydroxylase 2 (PHD2); and a more established hypoxia marker, glucose transporter-1 (GLUT-1), by testing the association with the exogenous marker pimonidazole. Materials and methods: Paraffin embedded tumour sections of 20 colorectal cancer patients were stained for blood vessels together with either pimonidazole or carbonic anhydrase-IX (CA-IX) and single stained for MCT1, MCT4, GLUT-1, and PHD2. Expression of all markers was compared with expression of pimonidazole and micro-vessel density (MVD) and with disease-free survival (DFS) and overall survival (OS). Results: No correlation was found between the different intrinsic hypoxia markers tested and pimonidazole. A trend for high MCT1 expression in biopsies with low CA-IX expression was found (R = −0.45, p = 0.06) and also the expression of MCT1 was higher in tumours with a high MVD (R = 0.49, p = 0.04). The more advanced tumours showed a higher expression of GLUT-1 (p = 0.03). A low CA-IX expression in the tumour correlated with better DFS (p = 0.03) and related to better OS (p = 0.07). Conclusion: Although none of the tested intrinsic hypoxia markers correlated with pimonidazole staining, we confirmed the important role of both GLUT-1 and CA-IX for a more advanced pTNM (pathological tumour-node-metastasis) stage and DFS respectively.

AIM:Recently, 18F-labeled 2-(5-fluoropentyl)-2-methylmalonic acid or ML10 has been proposed as a promising PET tracer for imaging of apoptosis. In this study we compared 18F-ML10, the 123I labeled 5-iodo derivative (123I-ML10) and a 68Ga-labeled Annexin A5 (AnxA5) and evaluated them as apoptosis tracers in several distinct models.METHODS:In vivo stability and biodistribution were studied in healthy mice. Apoptosis imaging was evaluated in anti-Fas treated mice and mice with muscular apoptosis. Furthermore, 18F-ML10 and 68Ga-Cys2-AnxA5 were evaluated in a rat model with reperfused liver infarct and a rat model with cerebral infarct as well as in Daudi tumor bearing mice, before and after treatment with cyclophosphamide and/or radiotherapy.RESULTS:18F-ML10 and 68Ga-Cys2-AnxA5 were both stable, while 123I-ML10 metabolized very quickly in vivo. All tracers showed a 3-4 times higher uptake in apoptotic muscular tissue in comparison to that in healthy muscular tissue. Animals with anti-Fas induced hepatic apoptosis showed an increased liver uptake which was most pronounced for 18F-ML10. The uptake of both 18F-ML10 and 68Ga-Cys2-AnxA5 increased in the apoptotic region surrounding the cerebral infarction and the reperfused liver infarction. Tumor uptake of 68Ga-Cys2-AnxA5, but not of 18F-ML10, was statistically significantly higher after therapy as measured with PET/MRI.CONCLUSION:All radiotracers were able to detect apoptosis in vitro and in vivo in each of the studied animal models of apoptosis. 68Ga-Cys2-AnxA5, but not 18F-ML10, allowed to visualize the effect of tumor therapy in a statistically significant way.

Aim. Recently, F-18-labeled 2-(5-fluoropentyl)-2-methylmalonic acid or ML10 has been proposed as a promising PET tracer for imaging of apoptosis. In this study we compared F-18-ML10, the I-123 labeled 5-iodo derivative (I-123-ML10) and a Ga-68-labeled Annexin A5 (AnxA5) and evaluated them as apoptosis tracers in several distinct models. Methods. In vivo stability and biodistribution were studied in healthy mice. Apoptosis imaging was evaluated in anti-Fas treated mice and mice with muscular apoptosis. Furthermore, F-18-ML10 and Ga-68-Cys2-AnxA5 were evaluated in a rat model with reperfused liver infarct and a rat model with cerebral infarct as well as in Daudi tumor bearing mice, before and after treatment with cyclophosphamide and/or radiotherapy. Results. F-18-ML10 and Ga-68-Cys2-Anx.A.5 were both stable, while I-123-ML10 metabolized very quickly in vivo. All tracers showed a 3-4 times higher uptake in apoptotic muscular tissue in comparison to that in healthy muscular tissue. Animals with anti-Fas induced hepatic apoptosis showed an increased liver uptake which was most pronounced for F-18-ML10. The uptake of both F-18-ML10 and Ga-68-Cys2-Anx.A.5 increased in the apoptotic region surrounding the cerebral infarction and the reperfused liver infarction. Tumor uptake of Ga-68-Cys2-AnxA5, but not of F-18-ML10, was statistically significantly higher after therapy as measured with PET/MRI. Conclusion. All radiotracers were able to detect apoptosis in vitro and in vivo in each of the studied animal models of apoptosis. Ga-68-Cys2-AnxA5, but not F-18-ML10, allowed to visualize the effect of tumor therapy in a statistically significant way.

INTRODUCTION:In vivo pH imaging has been a field of interest for molecular imaging for many years. This is especially important for determining tumor acidity, an important driving force of tumor invasion and metastasis formation, but also in the process of apoptosis. METHODS:2-(4-[(123)I]iodophenethyl)-2-methylmalonic acid (IPMM), 2-(4-[(123)I]iodophenethyl)-malonic acid (IPM), 2-(4-[(123)I]iodobenzyl)-malonic acid (IBMM) and 4-[(123)I]iodophthalic acid (IP) were radiolabeled via the Cu(+) isotopic nucleophilic exchange method. All tracers were tested in vitro in buffer systems to assess pH driven cell uptake. In vivo biodistribution of [(123)I]IPMM and [(123)I]IPM was determined in healthy mice and the pH targeting efficacy in vivo of [(123)I]IPM was evaluated in an anti-Fas monoclonal antibody (mAb) apoptosis model. In addition a mouse RIF-1 tumor model was explored in which tumor pH was decreased from 7.0 to 6.5 by means of induction of hyperglycemia in combination with administration of meta-iodobenzylguanidine. RESULTS:Radiosynthesis resulted in 15-20% for iodo-bromo exchange and 50-60% yield for iodo-iodo exchange while in vitro experiments showed a pH-sensitive uptake for all tracers. Shelf-life stability and in vivo stability was excellent for all tracers. [(123)I]IPMM and [(123)I]IPM showed a moderately fast predominantly biliary clearance while a high retention was observed in blood. The biodistribution profile of [(123)I]IPM was found to be most favorable in view of pH-specific imaging. [(123)I]IPM showed a clear pH-related uptake pattern in the RIF-1 tumor model. CONCLUSION:Iodine-123 labeled malonic acid derivates such as [(123)I]IPM show a clearly pH dependent uptake in tumor cells both in vitro and in vivo which allows to visualize regional acidosis. However, these compounds are not suitable for detection of apoptosis due to a poor acidosis effect.

Since the development and evaluation of novel anti-cancer therapies require molecular insight in the disease state, both FDG-PET and BLI imaging were evaluated in a Burkitt B-cell lymphoma xenograft model treated with cyclophosphamide or temsirolimus. Daudi xenograft mice were treated with either cyclophosphamide or temsirolimus and imaged with BLI and FDG-PET on d0 (before treatment), d2, d4, d7, d9 and d14 following the start of therapy. Besides tumor volume changes, therapy response was assessed with immunohistochemical analysis (apoptosis). BLI revealed a flare following both therapeutics that was significantly higher when compared to control tumors. FDG-PET decreased immediatelly, long before the tumor reduced in size. Late after therapy, BLI signal intensities decreased significantly compared to baseline subsequent to tumor size reduction while apoptosis was immediately induced following both treatment regimen. Unlike FDG, BLI was not able to reflect reduced levels of viable cells and was not able to predict tumor size response and apoptosis response.

PURPOSE:Evaluation and comparison of 3'-[(18)F]-fluoro-3'-deoxy-L-thymidine (FLT) and 2-[(18)F]-fluoro-2-deoxyglucose (FDG)-PET to monitor early response following both cyclophosphamide and temsirolimus treatment in a mouse model of Burkitt lymphoma. METHODS:Daudi xenograft mice were treated with either cyclophosphamide or temsirolimus and imaged with FLT-PET and FDG-PET on appropriate days post therapy inititiation. Immunohistochemical (IHC) studies (H&E, TUNEL, CD20, PCNA and ki-67) and DNA flow cytometry studies were performed. RESULTS:FDG tumor uptake decreased immediately after cyclophosphamide treatment while FLT-PET showed only a late and less pronounced decrease. A fast induction of apoptosis was observed together with an early accumulation of cells in the S-phase of the cell cycle, suggesting DNA repair. Temsirolimus treatment reduced both FDG and FLT tumor uptake immediately after therapy and resulted in a fast induction of apoptosis and G(0)-G(1) phase accumulation. CONCLUSION:FLT response was less distinct than FDG response and may be controlled by DNA repair early after cyclophosphamide. Nevertheless, FLT-PET was able to reflect decreased proliferation following temsirolimus.

In this study, we labeled N,N'-bis(diethylenetriamine pentaacetic acid)-pamoic acid bis-hydrazide (bis-DTPA-PA) with the generator produced PET radionuclide gallium-68 and evaluated 68Ga-bis-DTPA-PA as a potential tracer for in vivo visualization of necrosis by positron emission tomography (PET). Radiolabeling was achieved with a decay-corrected radiochemical yield of 63%. Biodistribution and in vivo stability studies in normal mice showed that 68Ga-bis-DTPA-PA is cleared faster from normal tissue than the previously reported 99mTc(CO)3 complex with bis-DTPA-PA which on the other hand is more stable in vivo. 68Ga-bis-DTPA-PA showed a 3.5-5 times higher binding to necrotic tissue than to viable tissue as shown by in vitro autoradiography while no statistically significant increased hepatic uptake was found in a biodistribution study in a mouse model of hepatic apoptosis. Specificity and avidity for necrosis was further evaluated in rats with a reperfused partial liver infarction and ethanol induced muscular necrosis. Dynamic microPET images showed a fast and prolonged uptake of 68Ga-bis-DTPA-PA in necrotic tissue with in vivo and ex vivo images correlating well with histochemical stainings. With necrotic to viable tissue activity ratios of 8-15 on ex vivo autoradiography, depending on the necrosis model, 68Ga-bis-DTPA-PA showed a faster and higher uptake in necrotic tissue than the 99mTc(CO)3 analog. These results show that 68Ga-bis-DTPA-PA specifically binds to necrotic tissue and is a promising tracer for in vivo visualization of necrosis using PET.

1185 Objectives Shortly after treatment the influx of inflammatory cells can interfere with [18F]FDG uptake while [18F]FLT is less affected by this phenomenon. In this study [18F]FDG and [18F]FLT uptake was monitored after cytotoxic or antiproliferative therapy and correlated to the number of viable cells assessed with bioluminescence imaging (BLI). Methods Daudi cells (Burkitt lymphoma) were transduced with a lentiviral vector (blasticidin selection marker and firefly luciferase reporter gene) and inoculated in SCID-mice (n=57). Mice were treated with Endoxan (125mg/kg, n=25) or Torisel (50mg/kg, n=25). [18F]FDG and [18F]FLT µPET and BLI were performed on days 0, 2, 4, 7, 9, 11 (no [18F]FLT) and 14. Histology was performed using H&E, TUNEL and Ki-67. Results Endoxan reduced [18F]FDG uptake from d2 without a significant reduction in [18F]FLT (only reduced from d7). BLI showed an increased signal on d2 and d4 while a reduction in the number of viable cells was measured from d7. Torisel treatment reduced [18F]FDG and [18F]FLT uptake from d2 after therapy. Reduction of [18F]FLT uptake was maximal on d4 versus d9 for [18F]FDG. BLI showed the same early signal increase as after Endoxan and a reduced number of viable cells was observed from d7 on. Proliferation (Ki-67) decreased only late after Endoxan treatment (d9-d14) while the decrease was faster (d2, d4) after Torisel treatment in line with the [18F]FLT response. In both treatment groups an increase in apoptotic and necrotic tumor fraction was measured. Further histological evaluation is underway to explore the rise in BLI signal, the role of inflammation and DNA repair early after therapy. Conclusions In conclusion, [18F]FLT-µPET was able to detect the reduced proliferation following therapy. After cytotoxic treatment [18F]FDG could predict response earlier than [18F]FLT

Purpose: Two variants of Annexin A5 (Cys2-AnxA5 and Cys 165-AnxA5) were labelled with Gallium-68 in order to evaluate their biological properties.Procedures: Biodistribution and pharmacokinetics of the radiotracers were studied with mu PET in healthy mice and in a mouse model of hepatic apoptosis. mu PET imaging after IV injection of the tracers in combination with mu MRI was performed in Daudi tumor bearing mice before and after treatment with a combination of chemotherapy and radiotherapy.Results: The biodistribution data indicated a fast urinary clearance with only minor hepatobilliary clearance, although a high retention in the kidneys was observed. Animals treated with anti-Fas showed a 3 to 8 times higher liver uptake as compared to healthy animals. Tumor uptake of Ga-68-Cys2-AnxA5 and Ga-68-Cys165-AnxA5 was low but significantly increased after therapy.Conclusion: Both Ga-68-Cys2-AnxA5 and Ga-65-Cys165-AnxA5 show a clear binding to apoptotic cells and are promising tracers for rapid evaluation of cancer therapy. (C) 2011 Elsevier Inc. All rights reserved.

1536 Objectives Radiolabelled AnnexinV (AnxA5) measures cell death both in vitro and in vivo. We radiolabelled AnxA5 with 68Ga using two variants of AnxA5 with a single cysteine residue at position 2 or 165, respectively Cys2-AnxA5 and Cys165-AnxA5, allowing site-specific coupling to 68Ga-Dota-maleimide. Methods In vitro cell binding was studied in control- and anti-Fas treated Jurkat cells. Biodistribution and pharmacokinetics were studied with µPET in healthy mice and in a hepatic apoptosis model (anti-Fas Ab treated) up to 60 min p.i. Daudi (Burkitt lymphoma) tumour bearing mice were scanned before and after treatment with combined chemotherapy (125 mg/g Endoxan) and radiotherapy (10 Gy/tumour) using µPET and µMRI. Tracer uptake was measured and imaged ex vivo using autoradiography and correlated to histological evidence of apoptosis (TUNEL). Results 68Ga-Dota-maleimide labelling yield was ≥ 98% and coupling yield of 68Ga-Dota-maleimide to Cys2-AnxA5 and Cys165-AnxA5 was ~70%. Labelling and purification took about 60 min, with a final radiochemical purity of ≥ 98%. In vitro binding of 68Ga-Cys2-AnxA5 and 68Ga-Cys165-AnxA5 to anti-Fas treated cells was 5 times higher compared to normal cells. Biodistribution data in normal mice indicated fast urinary clearance with only minor hepatobiliary clearance but high kidney retention. Anti-Fas treated animals showed a 3 to 8 times higher liver uptake (for respectively 68Ga-Cys2-AnxA5 and 68Ga-Cys165-AnxA5) as compared to healthy animals. Tumour uptake of 68Ga-Cys2-AnxA5 and 68Ga-Cys165-AnxA5 was respectively only 0.5 ± 0.1 % ID/g and 1.0 ± 0.3 % ID/g, but significantly increased to 1.5 ± 0.2 % ID/g and 1.6 ± 0.1 % ID/g after therapy. Conclusions Both 68Ga-Cys2-AnxA5 and 68Ga-Cys165-AnxA5 show a clear binding to apoptotic cells and are promising tracers for follow-up of cancer therapy. Research Support This study was financially supported by the European Union (IVA-VLANED-1.20)

PURPOSE:Early after therapy, 2-deoxy-2-[(18)F]fluoro-D-glucose ([(18)F]FDG) imaging is not always reliable due to the influx of inflammatory cells while apoptosis imaging offers a direct and early measurement of therapy effects. This study uses an improved apoptosis probe ((99m)Tc-hAnxA5) in combination with [(18)F]FDG imaging to evaluate therapy response.PROCEDURES:Daudi tumor tissue was implanted in the spleen of SCID mice. Treatment was performed with adriamycin and cyclophosphamide. Sequential [(18)F]FDG-positron emission tomography (PET) was acquired over 6 days and (99m)Tc-hAnxA5-SPECT was performed before and 1 day after therapy.RESULTS:On day 1, therapy induced apoptosis was visualized with (99m)Tc-hAnxA5 without a measurable change in [(18)F]FDG uptake. [(18)F]FDG uptake decreased significantly on day 3 and was even more pronounced on day 6.CONCLUSION:In this preclinical model, (99m)Tc-hAnxA5 imaging was able to detect apoptosis before metabolic changes were measured. These results confirm the value of apoptosis imaging for therapy response and give more insight in [(18)F]FDG imaging and its parameters to evaluate response.

PURPOSE:To determine the dependence of celecoxib on the tumour micro-environment in vitro and in vivo and to compare the use of (18)F-Fluorodeoxyglucose ((18)F-FDG) and (18)F- 3'-deoxy-3-fluorothymidine ((18)F-FLT) to measure tumour response.MATERIALS AND METHODS:In vitro, colony assays were performed on a cyclo-oxygenase 2 (COX-2) negative (HCT116) and a COX-2 positive cell line (HCA7). Xenograft models of these cell lines were treated with celecoxib and/or radiotherapy. Micro Positron Emission Tomography (microPET) scans with (18)F-FDG and (18)F-FLT were performed at different time-points.RESULTS:In vitro, no radiosensitising effect was seen in either of the cell lines. In vivo results showed a significant effect of celecoxib in the COX-2 negative tumours (HCT116) (enhancement ratio 1.5, p = 0.02) while no significant effect was observed in the COX-2 positive model (HCA7). A good correlation between (18)F-FDG and (18)F-FLT uptake was seen in both tumour models (r = 0.48, p = 0.002; r = 0.41, p = 0.005). After irradiation, a decrease in the uptake of both tracers was observed in both tumour models, which was more pronounced in the combination group, confirming the growth delay data.CONCLUSIONS:The contradicting in vitro and in vivo results suggest a major role of the tumour micro-environment. (18)F-FLT seems a good alternative for (18)F-FDG to follow tumour growth after radiation treatment.

375 Objectives: Evaluation of second cys2-annexin-A5 derivatives, radiolabelled with Tc99m or Ga68, as tracer agents for in vivo visualization of apoptosis. Methods: Cys2-annexin-A5 is a new generation annexin-A5, bearing a single cysteine residue for site-specific conjugation of suitable chelators via thiol-chemistry. HYNIC-cys2-annexin (1) and DTPA-cys2-annexin (2) were prepared using maleimide derivatives. 1 was labelled with Tc99m via a reported method and 2 was labelled with Tc99m(CO)3 using Isolink kit and with Ga68. Biodistribution of the radiolabelled complexes was studied in normal NMRI mice at 1 and 4 h p.i. In vivo specificity for apoptosis was tested in a Fas-mediated hepatic apoptosis model in mice for Tc99m-1 and Tc99m(CO)3-2 using SPECT-imaging, ex vivo biodistribution and autoradiography. H&E staining was performed to confirm apoptosis. Results: Radiolabelling yields were 85% for Tc99m-1 and 50-70% for Ga68-2 and Tc99m(CO)3-2. Biodistribution of the purified tracers in normal mice was similar to that of 1st-generation Tc99m-HYNIC-annexin-A5 for Tc99m-1, but different for Tc99m(CO)3-2 and especially Ga68-2, the latter showing persistent high blood retention. In the Fas-mediated hepatic apoptosis model uptake of Tc99m-1 and Tc99m(CO)3-2 in apoptotic tissue was significantly higher than in controls for both tracers, also visible on SPECT imaging at 1h p.i. Conclusions: Biodistribution of Tc99m-1 and Tc99m(CO)3-2 in normal mice was different depending on the way of labelling, while the specificity of the novel radiolabelled annexin derivatives for apoptosis was retained. Research Support: Euregional PACT project 4-BMG-II-2=70