Naar een behandeling op maat voor patiënten met lokaal geavanceerde rectale kanker

Rectal tumors account for about 30% of all colorectal cancers (CRC) which are the second leading cause of cancer-related death in the Western world. Despite major advances achieved in the treatment of rectal cancer, treatment response remains disappointing, especially for a subgroup of patients with locally advanced rectal cancer (LARC), for which the current standard treatment is chemoradiation (CRT) followed by surgery. Unfortunately, with this treatment still about 30% of the patients will die within 5 years of completion of treatment due to the development of local and distant recurrences. For this subgroup of patients, an intensification of the preoperative treatment is needed. Therefore movement towards an individualized medicine could be the future in which we should be able to stratify patients and adapt therapy according to the individual's needs. To address those needs, three research questions were examined. 1/ Can we find endogenous hypoxia markers selecting for tumors with a less favorable response? 2/ Can we understand the mechanism of action of bevacizumab alone or in combination with CRT and can we find prognostic markers? 3/ Is inhibition of the fibroblast growth factor receptor (FGFR) pathway an option to intensify the preoperative treatment?1/ One of the major factors responsible for the heterogeneous response and resistance to chemo- and radiotherapy is hypoxia. Measuring the oxygenation status of tumors could therefore identify patients with a bad prognosis and bad response to standard treatment, as well as offer an ideal target for treatment intensification for those patients. However, measuring intra-tumoral hypoxia is still a big challenge. Further research is needed before this factor can be taken into account in routine clinical practice to decide on an individual's treatment. One of the less invasive methods to detect hypoxia could be the use of immunohistochemical (IHC) stains for endogenous hypoxia markers, which are products of genes that are activated upon low oxygen levels. Various hypoxia-inducible proteins are proposed, but none of them has been validated so far. In a first study we investigated the potential of different possible endogenous hypoxia markers by performing IHC stainings on tumor material from 20 colorectal cancer patients which underwent surgery as primary therapy. Expression of those candidates (MCT1, MCT4, PHD2, GLUT-1, CA-IX) was compared with the expression of pimonidazole, as gold standard to measure hypoxia, and with clinical outcomes. IHC expression of none of the tested hypoxia markers correlated with pimonidazole staining. Therefore we can conclude that none of them is able to detect hypoxia properly in CRC patients. However we found a prognostic value for CA-IX, with a low tumoral expression correlating with better survival. Also a high GLUT-1 expression was shown in more advanced tumors. We can conclude that both GLUT-1 and CA-IX, although not entirely hypoxic-specific, might be involved in other molecular mechanisms that are of importance for response to treatment. 2/ Hypoxia in tumors is caused by an increase in diffusion distances with tumor expansion or by chaotic vasculature within tumors. Anti-angiogenic treatments have been proposed to transiently normalize the tumor vasculature resulting in a transient increase in oxygenation and a window of opportunity for enhanced radioresponse and drug delivery. Such combination of an anti-angiogenic agent and CRT could consequently be superior to individual treatment alone. The first angiogenesis inhibitor to be approved for cancer treatment was bevacizumab (Avastin®). Many phase I/II trials reported promising response rates and safety results for adding bevacizumab prior and concurrent to CRT in the preoperative treatment of LARC. However, the exact effect of bevacizumab monotherapy and the molecular interactions between the angiogenesis inhibitor and radiotherapy still need to be further unraveled to be able to optimize this combination treatment. Moreover at present there are no standardized or validated biomarkers of response, toxicity or resistance for anti-angiogenic therapy available. Extensive biomarker research needs to be conducted to allow us to use anti-angiogenic agents in the best way for the right patients. In the context of those problems, translational research was performed in a second project on patient material obtained as part of a phase II trial in which patients with advanced rectal cancer have been treated with bevacizumab in combination with CRT. Blood and tissue samples of 59 patients were collected before treatment and after one dose of bevacizumab but before CRT. The molecular effects of bevacizumab on tumoral gene expression, tumor cells and blood vessels and circulating angiogenic proteins were investigated by performing cDNA microarrays, IHC stainings and ELISA's. Possible prognostic biomarkers were evaluated by correlating expression levels with the pathological response. Our data could not show a normalization of the tumor vasculature after bevacizumab treatment, but rather suggest for bevacizumab a role as inhibitor of new vessel formation shown by a decrease in micro-vessel density and mature blood vessels, and a slight increase in hypoxia after one dose of bevacizumab treatment. Our exploratory data also demonstrated that tumors with an angiogenic expression profile correlated with a higher response rate and points towards a possible prognostic role for PDGFA, PDGF-BB, CA-IX and mature blood vessels. Based on these findings we hypothesize that a blocking of pericyte recruitment, with a parallel decrease of PDGF-BB, and a disruption of the vasculature, sensitizes the blood vessels for further anti-angiogenic therapy, resulting in a better treatment response. Overall, the effects of anti-angiogenic agents are context dependent calling for further mechanistic and biomarker studies before it can be used in clinical practice to intensify the preoperative treatment for LARC patients.3/ Another major hurdle to overcome is the development of resistance or tumor escape to anti-angiogenic agents. Evidence is growing that the FGFR signaling may be part of the mechanism of resistance to anti-VEGF therapy. Appropriate targeting of the FGFR pathway may be a promising strategy to overcome resistance to current anti-angiogenic agents. Furthermore FGFs and their receptors are recognized oncogenes associated with a variety of cancers. All four FGF receptors and their ligands are expressed in CRC and have been shown to play roles in CRC progression and tumor angiogenesis. Most clinically advanced drugs against FGFR are not selective for FGFR alone, which often result in increased side effects and the limitation to achieve the required doses for effective FGFR inhibition. Therefore it will be necessary to develop FGFR-selective agents. Moreover, targeting dysregulated pathways is a promising approach but generally will have to be combined with conventional cytotoxic therapies like radiation therapy. The effectiveness of FGFR-targeting therapy for CRC has been demonstrated in vitro and in vivo in basic research studies, but to our knowledge FGFR inhibitors have not been tested particularly in rectal cancer in combination with (chemo) radiotherapy. Accordingly in a third research project, we assessed the antitumor effect and radiosensitizing potential of a small molecule inhibitor of FGFR in CRC in vitro and in vivo. The expression profile of FGFRs was evaluated by qPCR on different human colorectal cell with and without drug incubation. In vitro, the effect of the inhibitor on survival, proliferation, cell cycle distribution, apoptosis and radiosensitivity was evaluated by respectively SRB, BrdU, FACS using propidium iodide, BrdU and Annexin V and colony assays. Western blotting was performed to assess the modulation of FGFR phosphorylation, apoptosis and downstream signalling by the inhibitor. The effect of the inhibitor with and without radiation was evaluated in vivo on 2 mouse models (Caco2 and NCI-H716). In vivo mechanistic studies were performed on isolated tumors with IHC staining, western blots and qPCR. The tested inhibitor showed potent and selective antitumor activity in vitro and in vivo. The effect of the inhibitor was restricted to the NCI-H716 cells, which show an FGFR2 overexpression. No effects were observed on the other cell lines without deregulated FGFR activity. In vitro proliferation was abolished consistent with blockade of the ERK pathway, followed by a dramatic increase in apoptosis and a significant decrease in cell survival. These events were reflected at the molecular level by an increase in the apoptotic marker cleaved PARP and inhibition of the pro-survival AKT pathway. FGFR inhibition in the NCI-H716 xenograft model resulted initially in a growth delay of 5 days with decreased tumor cell proliferation with related ERK inhibition. However, the continued presence of drug seems essential with accelerated in vivo regrowth of treated tumors after treatment cessation associated with an increase in tumor cell proliferation. In the tested set-up, the inhibitor does not augment the cytotoxicity ofradiotherapy. In summary based on our data, combining this inhibitor concurrent with radiotherapy does not seem beneficial. The drug on its own slows down tumor growth, but attention needs to be paid when stopping drug delivery. Moreover, only tumors with deregulated FGFR activity respond to the drug, providing an option for selective intervention but also stressing the importance of patient selection based on molecular aberrations. In conclusion, these projects illustrate that a lot of hurdles still need to be taken before we can provide a personalized treatment to LARC patients. First, although hypoxia is known for decades as a major factor of therapy resistance, still trustable ways to image this factor are needed before we can implement this factor in treatment decision. Secondly, further insights in the effect of bevacizumab and validated biomarkers are urgently needed before giving bevacizumab to LARC patients outside a clinical trial setting. Last, we have to keep searching for new molecular targets for radiosensitization in order to intensify the preoperative treatment. In parallel, biomarker research needs to be enclosed in those preclinical studies to support patient selection strategies in further clinical trials.Dankwoord List of abbreviations Table of contents CHAPTER I 11 1. General introduction 13 1.1. Rectal cancer 13 1.2. Evolution to a multimodal treatment in rectal cancer 13 1.3. Heterogeneity in response 14 1.4. Tumor hypoxia as major factor of resistance against chemoradiotherapy 15 1.4.1. Development of hypoxia 15 1.4.2. Response to hypoxia and its clinical relevance 16 1.4.3. Hypoxia measurement and modulation 18 1.5. Angiogenesis: key players and their role in cancer 18 1.6. Targeting VEGF as therapeutic approach in rectal cancer 21 1.6.1. Rationale for combining anti-angiogenic drugs with chemoradiation 21 1.6.2. VEGF inhibitors 25 1.6.3. Clinical trials: bevacizumab for locally advanced rectal cancer 27 1.6.4. Obstacles to conquer before using anti-VEGF agents in the clinic 30 1.7. FGFR as molecular target in rectal cancer 32 1.7.1. Rationale to target FGFR in rectal cancer treatment 32 1.7.2. FGFR inhibitors 37 CHAPTER II 39 2. Objectives 41 2.1. General aim 41 2.2. Specific aims 42 2.3. Strengths and limits of the methodology 43 CHAPTER III 45 3. Investigation of possible endogenous hypoxia markers in colorectal cancer 47 3.1. Introduction 48 3.2. Material and methods 50 3.2.1. Patients 50 3.2.2. Immunohistochemical staining 50 3.2.3. Quantitative analysis 50 3.2.4. Statistical analysis 51 3.3. Results 52 3.3.1. Clinical characteristics 52 3.3.2. Immunohistochemistry 52 3.3.3. Associations between intrinsic hypoxia markers, pimonidazole and MVD 53 3.3.4. Associations between intrinsic hypoxia markers, pimonidazole, MVD and tumor characteristics 54 3.3.5. Survival analysis 56 3.4. Discussion 57 CHAPTER IV 61 4. Combining bevacizumab and chemoradiation in rectal cancer. Translational results of the AXEBeam trial 63 4.1. Introduction 64 4.2. Materials and methods 66 4.2.1. Patient selection, treatment and outcome 66 4.2.2. Sample collections and storage 66 4.2.3. ELISA assays 67 4.2.4. Immunohistochemistry and scoring 68 4.2.5. RNA extraction and cDNA synthesis 68 4.2.6. Microarray and data processing 69 4.2.7. Pathway analysis 69 4.2.8. Statistical analysis 69 4.3. Results 71 4.3.1. Impact of bevacizumab on gene expression, blood markers and tissue markers 71 4.3.2. Exploration of prognostic biomarkers 75 4.4. Discussion 78 4.4.1. Changes induced by bevacizumab in AxeBeam trial versus Willett et al. 78 4.4.2. Angiogenic gene profiling as biomarker 80 4.4.3. Role of tissue biomarkers 81 4.4.4. Angiogenic circulating markers as biomarkers 82 4.4.5. Hypothesis 83 4.4.6. Conclusion 84 CHAPTER V 105 5. In vitro and in vivo evaluation of FGFR as a molecular target for rectal cancer radiosensitization 107 5.1. Introduction 108 5.2. Material and methods 109 5.2.1. Cells and cell culture 109 5.2.2. FGFR inhibitor 109 5.2.3. Cell viability 109 5.2.4. Flow cytometry 109 5.2.5. Clonogenic assays 110 5.2.6. Mice 110 5.2.7. Irradiation 111 5.2.8. Gene expression 111 5.2.9. Western blotting 111 5.2.10. Immunohistochemical staining and analysis 111 5.2.11. Statistical analysis 112 5.3. Results 113 5.3.1. Antitumor activity in vitro 113 5.3.2. In vitro effects on cell proliferation, apoptosis and radiosensitivity 114 5.3.3. Xenograft growth delay 116 5.3.4. Mechanism of action of FGFR inhibitor in vivo 117 5.3.5. Combination treatment in vivo 118 5.4. Discussion 120 CHAPTER VI 129 6. Conclusions, general discussion and future perspectives 131 6.1. Challenges to measure hypoxia: problems and opportunities 131 6.1.1. The tested endogenous markers do not reflect intra-tumoral hypoxia 131 6.1.2. Challenges in the search for endogenous hypoxia markers 133 6.1.3. Opportunities for endogenous markers 134 6.2. Is there a future for bevacizumab in combination with radiotherapy for rectal cancer? 136 6.2.1. Bevacizumab as disrupter of tumor vasculature 137 6.2.2. Search for biomarkers 138 6.2.3. Future use of anti-angiogenic agents 142 6.3. Targeting FGFR signaling in colorectal cancer 144 6.3.1. Affecting survival but not radiosensitivity 144 6.3.2. Considerations for further development of the inhibitor 145 6.3.3. Ongoing search for molecular targets 146 6.4. Overall conclusion 148 6.5. Future perspectives 149 CHAPTER VII 165 Summary 167 Samenvatting 171 Curriculum vitae 175 Bibliography 177nrpages: 179status: publishe