Supplementary Figure 2 from Nelfinavir Down-regulates Hypoxia-Inducible Factor 1α and VEGF Expression and Increases Tumor Oxygenation: Implications for Radiotherapy

Supplementary Methods, Figures 1-3, Tables 1-2 from In Vivo Profiling of Hypoxic Gene Expression in Gliomas Using the Hypoxia Marker EF5 and Laser-capture Microdissection

Peritoneal carcinomatosis (PC) can occur as an advanced consequence of multiple primary malignancies. Surgical resection, radiation or systemic interventions alone have proven inadequate for this aggressive cancer presentation, since PC still has a poor survival profile. Photodynamic therapy (PDT), in which photosensitive drugs are exposed to light to generate cytotoxic reactive oxygen species, may be an ideal treatment for PC because of its ability to deliver treatment to a depth appropriate for peritoneal surface tumors. Additionally, epidermal growth factor receptor (EGFR) signaling plays a variety of roles in cancer progression and survival as well as PDT-mediated cytotoxicity, so EGFR inhibitors may be valuable in enhancing the therapeutic index of intraperitoneal PDT. This study examines escalating doses of benzoporphyrin derivative (BPD)-mediated intraperitoneal PDT combined with the EGFR-inhibitor cetuximab in a canine model. In the presence or absence of small bowel resection (SBR) and cetuximab, we observed a tolerable safety and toxicity profile related to the light dose received. Additionally, our findings that BPD levels are higher in the small bowel compared with other anatomical regions, and that the risk of anastomotic failure decreases at lower light doses will help to inform the design of similar PC treatments in humans.

The finding that most GBMs recur either near or within the primary site after radiotherapy has fueled great interest in the development of radiosensitizers to enhance local control. Unfortunately, decades of clinical trials testing a wide range of novel therapeutic approaches have failed to yield any clinically viable radiosensitizers. However, many of the previous radiosensitizing strategies were not based on clear pre-clinical evidence, and in many cases blood-barrier penetration was not considered. Furthermore, DNA repair inhibitors have only recenly arrived in the clinic, and likely represent potent agents for glioma radiosensitization. Here, we present recent progress in the use of small molecule DNA damage response inhibitors as GBM radiosensitizers. In addition, we discuss the latest progress in targeting hypoxia and oxidative stress for GBM radiosensitization.

The purpose of this pilot study was to determine whether blood-borne microvesicles from newly diagnosed glioblastoma patients could be used as biomarkers. We collected 2.8 mL blood from 16 post-operative patients at the time that they were being simulated for chemoradiation therapy (radiation with concurrent temozolomide). Two additional samples were collected during chemoradiation therapy and a final sample was collected at the end of chemoradiation therapy. Patients continued with the therapy suggested by their physicians, based on tumor conference consensus and were followed for recurrence and overall survival. Microvesicles were isolated using serial centrifugation and stained for surface markers (Annexin V for phosphotidyl serine, CD41 for platelets, anti-EGFR for tumor cells, and CD235 for red blood cells). Flow cytometry analysis was performed. Our findings provide initial evidence that increases in Annexin V positive microvesicle levels during chemoradiation therapy are associated with earlier recurrence and shorter overall survival in newly diagnosed glioblastoma patients. The effect is dramatic, with over a four-fold increase in the hazard ratio for an individual at the 75th versus the 25th percentile. Moreover the pattern of Annexin V positive microvesicles remain significant after adjustment for confounding clinical variables that have previously been shown to be prognostic for recurrence and survival. Inclusion of neutrophil levels at the start of chemoradiation therapy in the model yielded the largest attenuation of the observed association. Further studies will be needed to verify and further investigate the association between these two entities.

Diagnostic and prognostic indicators are key components to achieve the goal of personalized cancer therapy. Two distinct approaches to this goal include predicting response by genetic analysis and direct testing of possible therapies using cultures derived from biopsy specimens. Optimally, the latter method requires a rapid assessment, but growing xenograft tumors or developing patient-derived cell lines can involve a great deal of time and expense. Furthermore, tumor cells have much different responses when grown in 2D versus 3D tissue environments. Using a modification of existing methods, we show that it is possible to make tumor-fragment (TF) spheroids in only 2-3 days. TF spheroids appear to closely model characteristics of the original tumor and may be used to assess critical therapy-modulating features of the microenvironment such as hypoxia. A similar method allows the reproducible development of spheroids from mixed tumor cells and fibroblasts (mixed-cell spheroids). Prior literature reports have shown highly variable development and properties of mixed-cell spheroids and this has hampered the detailed study of how individual tumor-cell components interact. In this study, we illustrate this approach and describe similarities and differences using two tumor models (U87 glioma and SQ20B squamous-cell carcinoma) with supporting data from additional cell lines. We show that U87 and SQ20B spheroids predict a key microenvironmental factor in tumors (hypoxia) and that SQ20B cells and spheroids generate similar numbers of microvesicles. We also present pilot data for miRNA expression under conditions of cells, tumors, and TF spheroids.

The standard of care for glioblastoma (GB) is radiation therapy (RT) and temozolomide (TMZ) following optimal surgery. This regimen has been accompanied by an increase in the occurrence of equivocal imaging findings, e.g. tumor progression vs. treatment effect (TE), which includes pseudoprogression (PsP). Thus decisions regarding further treatment are difficult and often delayed. None of the current imaging methods for identifying TE/PsP have proven sensitive and specific. Therefore, we developed a method to isolate microvesicles (MV) from blood sample in patients with GB. MV are defined herein as lipid membrane-bound sacs with a diameter >300 nm. METHODS: 3 ml citrated blood was collected from GB patients at their RT simulation and at multiple times during and following treatment. MV were isolated during multiple centrifugations (300g, 2500g, 15,000g). The pellet from the final spin was analyzed using flow cytometry. Antibodies to phosphotidylserine were used to identify the MV. RESULTS: We analyzed 16 blood samples from 10 GB patients that met analysis criteria: the MV sample was obtained at or following the completion of CRT and, a definitive diagnosis (TP, TE or PSP) was reached within 60 days of the date of the sample. MV counts in the patients with stable disease or TE/PsP were significantly lower than patients who developed recurrence or died of their disease (p = 0.0385). Based on MRI and/or pathological assessment, 2 patients have died of their disease, 1 patient is alive with recurrence, 3 patients have stable disease and 4 patients are being followed for PsP vs. tumor progression (TP). SUMMARY/CONCLUSION: These preliminary data suggest that the analysis of blood (liquid biopsy) for MV may be useful to distinguish TE/PsP from TP in GB patients. MVs may be valuable in addition to standard imaging for decision making in patients with equivocal imaging findings.

Traditional anticancer chemotherapy often displays toxic side effects, poor bioavailability, and a low therapeutic index. Targeting and controlled release of a chemotherapeutic agent can increase drug bioavailability, mitigate undesirable side effects, and increase the therapeutic index. Here we report a polymersome-based system to deliver gemcitabine to Panc-1 cells in vitro. The polymersomes were self-assembled from a biocompatible and completely biodegradable polymer, poly(ethylene oxide)-poly(caprolactone), PEO-PCL. We showed that we can encapsulate gemcitabine within stable 200 nm vesicles with a 10% loading efficiency. These vesicles displayed a controlled release of gemcitabine with 60% release after 2 days at physiological pH. Upon treatment of Panc-1 cells in vitro, vesicles were internalized as verified with fluorescently labeled polymersomes. Clonogenic assays to determine cell survival were performed by treating Panc-1 cells with varying concentrations of unencapsulated gemcitabine (FreeGem) and polymersome-encapsulated gemcitabine (PolyGem) for 48 hours. 1 μM PolyGem was equivalent in tumor cell toxicity to 1 μM FreeGem, with a one log cell kill observed. These studies suggest that further investigation on polymersome-based drug formulations is warranted for chemotherapy of pancreatic cancer.

Classical descriptions of tumor physiology suggest two origins for tumor hypoxia; steady-state (diffusion-limited) hypoxia and cycling (perfusion-modulated) hypoxia. Both origins, primarily studied and characterized in murine models, predict relatively small, isolated foci or thin shells of hypoxic tissue interspersed with contrasting oxic tissue. These foci or shells would not be expected to scale with overall tumor size since the oxygen diffusion distance (determined by oxygen permeability and tissue oxygen consumption rate) is not known to vary dramatically from tumor to tumor. We have identified much larger (macroscopic) regions of hypoxia in rat gliosarcoma tumors and in larger human tumors (notably sarcomas and high-grade glial tumors), as indicated by biochemical binding of the hypoxia marker, EF5. Thus, we considered an alternative cause of tumor hypoxia related to a phenomenon first observed in window-chamber tumor models: namely longitudinal arteriole gradients. Although longitudinal arteriole gradients, as originally described, are also microscopic in nature, it is possible for them to scale with tumor size if tumor blood flow is organized in an appropriate manner. In this organization, inflowing blood would arise from relatively well-oxygenated sources and would branch and then coalesce to poorly-oxygenated outflowing blood over distances much larger than the length of conventional arterioles (multi-millimeter scale). This novel concept differs from the common characterization of tumor blood flow as disorganized and/or chaotic. The organization of blood flow to produce extended longitudinal gradients and macroscopic regional hypoxia has many important implications for the imaging, therapy and biological properties of tumors. Herein, we report the first experimental evidence for such blood flow, using rat 9L gliosarcoma tumors grown on the epigastric artery/vein pair.

UNLABELLED The primary goals of this study were to determine the biodistribution and excretion of (18)F-EF5 in oncologic patients, to estimate the radiation-absorbed dose and to determine the safety of this drug. METHODS Sixteen patients with histologically confirmed malignancy received a mean intravenous infusion of 217  MBq (range 107-364  MBq) of (18)F-EF5. Over a 4-6-hour period, four to five serial positron emission tomography (PET) or PET/computed tomography (CT) scans were obtained. To calculate the radiation dosimetry estimates, volumes of interest were drawn over the source organs for each PET scan or on the CT for each PET/CT scan. Serial blood samples were obtained to measure (18)F-EF5 blood clearance. Bladder-wall dose was calculated based on urine activity measurements. RESULTS The urinary bladder received the largest radiation-absorbed dose, 0.12 ± 0.034 mSv/MBq (mean ± SD). The average effective dose equivalent and the effective dose of (18)F-EF5 were 0.021 ± 0.003 mSv/MBq and 0.018 ± 0.002 mSv/MBq, respectively. (18)F-EF5 was well tolerated in all subjects. CONCLUSIONS (18)F-EF5 was demonstrated to be safe for patients, and the radiation exposure is clinically acceptable. As with any radiotracer with primary excretion in the urine, the bladder-wall dose can be minimized by active hydration and frequent voiding.

Abstract Hypoxia plays a key role in tumor aggressiveness and radiation resistance, yet little is known regarding hypoxic global gene regulation in vivo. Although the regulation of mRNA and miRNA expression by hypoxia has been investigated in isolation in vitro, it has not yet been analyzed directly in tumor samples. We have used the hypoxia marker EF5 coupled with laser capture microdissection (LCM) to isolate RNA from viable hypoxic and normoxic regions of 9L experimental rat gliomas and from human head and neck (H&N) tumor samples. This was followed by microarray analysis of mRNA expression and comparison of this signature with that obtained from treating 9L cells with hypoxia in vitro. Through this, we have identified several mRNAs (including the HIF targets Vegf, Glut-1 and Hsp27) with increased levels under hypoxia compared to normoxia both in vitro and in vivo. We also found striking differences between the global in vitro and in vivo hypoxic mRNA profiles. An intriguing finding was the hypoxic-downregulation of a number of immunomodulatory and DNA repair proteins including CXCL9, CD3D and RAD51 in vivo, consistent with a pro-tumorigenic role of hypoxia in solid tumors. Immunohistochemical staining verified increased HSP27 and decreased RAD51 protein levels in hypoxic vs. normoxic tumor regions. Moreover, CD8+ T cells which are recruited to tumors upon stimulation by CXCL9 and CXCL10, were largely excluded from viable hypoxic areas in vivo. Global microRNA expression changes have been reported to occur in response to hypoxia in vitro. Hence, our second objective was to identify the cluster of miRNAs differentially expressed in hypoxic vs. normoxic regions of the human tumors using TaqMan® Array MicroRNA Cards. We employed the same technique used with the 9L tumors to analyze miRNA expression patterns in hypoxic vs. normoxic samples from H&N cancer patients who have been administered EF5 prior to surgical removal of the tumor. Consistent with published in vitro studies, miR-210 emerged as a major miRNA robustly induced in the hypoxic regions. We also confirmed miR-210 induction by an individual real time-PCR assay and found it to be upregulated > 2-fold in all hypoxic RNA samples. In addition, we have also found several miRNAs whose levels were upregulated and downregulated in hypoxic vs. normoxic areas. This is the first study to analyze the influence of hypoxia on mRNA and miRNA levels in vivo and can be readily adapted to obtain a comprehensive picture of hypoxic regulation of gene expression and its influence on biological functions in solid tumors. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 5004. doi:10.1158/1538-7445.AM2011-5004

Glioblastoma multiforme (GBM) is the most aggressive of the astrocytic malignancies and the most common intracranial tumor in adults. Although the epidermal growth factor receptor (EGFR) is overexpressed and/or mutated in at least 50% of GBM cases and is required for tumor maintenance in animal models, EGFR inhibitors have thus far failed to deliver significant responses in GBM patients. One inherent resistance mechanism in GBM is the co-activation of multiple receptor tyrosine kinases (RTKs), which generates redundancy in activation of phosphoinositide-3'-kinase (PI3K) signaling. Herein we present a novel mechanism by which fibroblast growth factor receptors (FGFRs) and src family kinases (SFKs) impact PI3K signaling in GBM by phosphorylating the PTEN tumor suppressor at a conserved tyrosine residue, Y240. Phosphorylation of Y240 is associated with shortened overall survival and resistance to …

The primary purpose of this study was to assess the biodistribution and radiation dose resulting from administration of 18F-EF5, a lipophilic 2-nitroimidazole hypoxia marker in ten cancer patients. For three of these patients (with glioblastoma) unlabeled EF5 was additionally administered to allow the comparative assessment of 18F-EF5 tumor uptake with EF5 binding, the latter measured in tumor biopsies by fluorescent anti-EF5 monoclonal antibodies.

Primary malignant brain tumors are estimated to have caused more than 12,800 deaths in the United States in 2006, and nearly 19,000 new cases are diagnosed annually in this country. The glial-derived neoplasm, or glioma, is the most aggressive type of brain tumor and accounts for 44% of all primary brain tumors. Among gliomas, the glioblastoma is the most commonly seen glial brain tumor in the United States and the most lethal worldwide. Despite very active investigation of new therapeutic models, more aggressive surgical approaches, novel delivery modalities and protocols for radiation, types and dosage of chemotherapy, and use of new signaling modifiers, only small improvements in patient outcomes have been achieved in this disease. The purpose of this review is to summarize the important aspects of tumor hypoxia in human glial brain tumors. We will describe the techniques used to measure tumor hypoxia, highlight what is currently known of the pathophysiologic and genetic mechanisms linking hypoxia with tumor progression and illustrate the significance of the oxygen effect and its role in tumor resistance to radiation and chemotherapy. We will also review therapeutic modalities to combat hypoxic tumor cell resistance and discuss the future of diagnosing and combating brain tumor hypoxia, emphasizing noninvasive imaging and novel hypoxia-modifying agents. At the conclusion of this article, the reader should have an up-to-date understanding of a concept, tumor cell oxygenation, which has the potential to help differentiate which malignant glioma patients are most likely to have good outcomes and respond to antihypoxia therapy.

A recently published randomized study showed that combining an EGFR inhibitor with radiation increases survival and local control in patients with locally advanced head and neck cancer. We were interested in understanding the effects of EGFR inhibition on the tumor microenvironment in a head and neck cancer xenograft model, as this could provide insight into its effectiveness in combination with radiation and chemotherapy. We used SQ20B head and neck squamous cell carcinoma cells either in vitro or grown as subcutaneous xenografts in nude mice. Both in vitro(Western blotting, radiation cell survival assays) and in vivo studies (tumor regrowth assay, EF5 staining) were conducted. We saw a minimal effect of erlotinib on in vitro radiosensitization. However, mice with SQ20B xenografts treated with the drug had prolonged time to tumor regrowth after 6 Gy of radiation compared to mice given radiation only. Because the in vivo effects appeared greater than the in vitro effects, we investigated the effects of erlotinib on the tumor microenvironment. Erlotinib decreased both vascular endothelial growth factor (VEGF) and hypoxia-inducible factor (HIF)-1α expression in vitro and in tumor xenografts grown in vivo. We measured vascular permeability within the xenografts using Evan's blue dye and found a 75% reduction in permeability in the tumors of mice treated with the drug. Consistent with these results, tomato lectin staining of the tumor vasculature showed that erlotinib treatment led to a dramatic change in vessel morphology. Using Power Doppler we found that there was a greater than 3-fold increase in tumor perfusion within 48 hours of starting erlotinib treatment. This increased vascular perfusion was evident even at 96 hours, in spite of the fact that the tumors showed a 50% decrease in size by this time. Using the hypoxia marker EF5 we found that the pattern of hypoxia was heterogeneous in these tumors but that erlotinib treatment led to improved oxygenation. Erlotinib decreases HIF-1α/VEGF expression in SQ20B xenografts but paradoxically increases vascular perfusion. A potential explanation for this is the idea of "vascular normalization" as proposed by Jain (Science 2005;307:58). This could lead to increased delivery of chemotherapy to tumors in patients responding to erlotinib. Improved vascular perfusion may also lead to increased oxygenation that could play a role in increased in vivo radiosensitization.

Patients with glioblastomas (GBMs) survive an average of 12-14 months post-diagnosis. Both hypoxia and Ki67 expression increase with tumor grade but there is no general agreement that either has prognostic importance for GBM patients. We hypothesized that the pO2 of Ki67+ cells would be prognostically significant for outcome in GBM patients. Patients were given 21 mg/kg EF5 24 hours preceding surgery. Frozen tumor tissue sections were made and stained for nuclei (Hoechst 33342), hypoxia (EF5) and proliferation (Ki67); 2-4 sets of images were analyzed per patient. A binary nuclei mask was made to define all regions of viable tissue for further analyses. The EF5 images were calibrated by adjusting for variables previously reported by this group (camera variables, drug AUC, cube reference binding) and aligned to the nuclear images. The resulting overlays were merged to locate each KI67+ cell; the tissue oxygen levels (EF5 binding) in the immediate 50μ diameter area of each Ki67+ cell was determined and averaged. The overall EF5 values (histogram) for the entire image were also determined. Patients were followed to recurrence and/or death. Four predictors in addition to RPA (Recursive Partitioning Analysis) status were tested for association with outcome: overall EF5 binding (EF5_all), EF5 binding surrounding Ki67+ cells (EF5_Ki67), the ratio of the previous 2 predictors (Ratio, [EF5_Ki67/EF5_all], >50% vs. ≤50%) and proliferation index (PI). 16 GBM patients received definitive surgery and post-operative radiotherapy; a majority of patients received additional treatment when they recurred. At the time of analysis, 15 patients had recurred and/or died of their disease. Only the ratio was significant for outcome: for recurrence, p = 0.0074 and for survival p = 0.0196. Because RPA classification is an important prognostic factor for survival in glioma patients (p = 0.0026 in the current sample), we tested whether adding it to a model containing ratio was valuable; we found that the predictive ability of a model containing both RPA and Ratio was significantly improved (p = 0.038) compared to the model with RPA alone. All patients had regions of hypoxia in their tumors (8 mild, 8 moderate), yet the absolute pO2 of their tumors did not predict patient outcome. However, patients had a worse prognosis (recurrence and survival) if the pO2 of the proliferating cells was within 50% of the EF5 value of the overall pO2 (e.g. close to the overall level of hypoxia). Although mechanistic studies must be performed, we hypothesize that (1) hypoxic proliferating cells contribute to tumor recurrence and (2) the pO2 of hypoxic up-regulation of pro-invasion genes and cytokines is patient-specific, contributing to the complexity of understanding, diagnosing and treating gliomas.