The central hypothesis of "radiomics" is that imaging features reflect tumor phenotype and genotype. Until now only correlative studies have been performed. The main objective of our study is to determine whether a causal relationship exists between genetic changes and image features. The secondary objective is to assess whether the combination with radiotherapy (RT) influences these image features. HCT116 doxycycline (dox) inducible GADD34 cells were grown as xenografts in the flanks of NMRI-nu mice. GADD34 overexpression decreases hypoxic fraction. Radiomics analyses were performed on computed tomography images obtained at 40kVp and again at 80kVp for validation, before radiotherapy at a volume of 200mm(3), 4days post RT (10Gy) and 500mm(3). To select reproducible features test-retest experiments were performed at baseline. Gene induction and/or irradiation translated into significant changes in radiomics features. Post irradiation, 17 features for 40kVp and 9 features for 80kVp differed significantly between dox+ and dox- combined with RT. 8 and 4 of these features remained consistent for 40 and 80kVp, respectively. Radiomics is able to identify early effects of changed gene expression combined with radiation treatment in tumors with similar volumes which are not visible to human eye.

Tumour hypoxia and its molecular responses have been shown to be associated with poor prognosis. Detection of hypoxia, preferably in a non-invasive manner, could therefore predict treatment outcome and serve as a tool to individualize treatment. This review gives an overview of recent literature on hypoxia imaging markers currently used in clinical trials. Furthermore, recent progress made in targeting hypoxia (hypoxia-activated prodrugs) or hypoxia response (carbonic anhydrase IX inhibitors) is summarized. Last, window-of-opportunity trials implementing non-invasive imaging are proposed as an important tool to prove anti-tumour efficacy of experimental drugs early during drug development.

Neo-adjuvant chemoradiotherapy followed by surgery is the standard treatment with curative intent for oesophageal cancer patients, with 5-year overall survival rates up to 50 %. However, patients’ quality of life is severely compromised by oesophagectomy, and eventually many patients die due to metastatic disease. Most solid tumours, including oesophageal cancer, contain hypoxic regions that are more resistant to chemoradiotherapy. The hypoxia-activated prodrug evofosfamide works as a DNA-alkylating agent under these hypoxic conditions, which directly kills hypoxic cancer cells and potentially minimizes resistance to conventional therapy. This drug has shown promising results in several clinical studies when combined with chemotherapy. Therefore, in this phase I study we investigate the safety of evofosfamide added to the chemoradiotherapy treatment of oesophageal cancer. A phase I, non-randomized, single-centre, open-label, 3 + 3 trial with repeated hypoxia PET imaging, will test the safety of evofosfamide in combination with neo-adjuvant chemoradiotherapy in potentially resectable oesophageal adenocarcinoma patients. Investigated dose levels range from 120 mg/m2 to 340 mg/m2. Evofosfamide will be administered one week before the start of chemoradiotherapy (CROSS-regimen) and repeated weekly up to a total of six doses. PET/CT acquisitions with hypoxia tracer 18F-HX4 will be made before and after the first administration of evofosfamide, allowing early assessment of changes in hypoxia, accompanied with blood sampling to measure hypoxia blood biomarkers. Oesophagectomy will be performed according to standard clinical practice. Higher grade and uncommon non-haematological, haematological, and post-operative toxicities are the primary endpoints according to the CTCAEv4.0 and Clavien-Dindo classifications. Secondary endpoints are reduction in hypoxic fraction based on 18F-HX4 imaging, pathological complete response, histopathological negative circumferential resection margin (R0) rate, local and distant recurrence rate, and progression free and overall survival. This is the first clinical trial testing evofosfamide in combination with chemoradiotherapy. The primary objective is to determine the dose limiting toxicity of this combined treatment and herewith to define the maximum tolerated dose and recommended phase 2 dose for future clinical studies. The addition of non-invasive repeated hypoxia imaging (‘window-of-opportunity’) enables us to identify the biologically effective dose. We believe this approach could also be used for other hypoxia targeted drugs. ClinicalTrials.gov Identifier: NCT02598687 .

Matrix metalloproteinase-2 (MMP2) is important in tumorigenesis, angiogenesis and tumor invasion. In this study, we investigated if the Cy5-tagged small immuno protein targeting the catalytic domain of human MMP2 (aMMP2-SIP) detects MMP2 in tumors non-invasively. For this purpose, we generated MMP2 expressing (empty vector EV) and knock-down (KD) HT1080, U373 and U87 cells, which were injected subcutaneously in the lateral flank of NMRI-nu mice. Optical imaging (Optix MX2) performed at 0.5, 2, 4, 8, 24 and 48 hour post injection (h.p.i.) of Cy5 tagged aMMP2-SIP, indicated significantly lower tumor to background ratios at both 24 (P = 0.0090) and 48 h.p.i. (P < 0.0001) for the U87 MMP2-KD compared to control tumors. No differences were found for HT1080 and U373 models. U87 MMP2-KD tumors had significantly lower MMP2 activity (P < 0.0001) than EV tumors as determined by gelatin zymography in tumor sections and lysates, while no differences were observed between EV and MMP2-KD in HT1080 and U373. In line with these data, only U87 MMP2-KD tumors had a reduced tumor growth compared to control tumors (P = 0.0053). aMMP2-SIP uptake correlates with MMP2 activity and might therefore be a potential non-invasive imaging biomarker for the evaluation of MMP2 activity in tumors.

Purpose: Current clinical measurements for tumor treatment efficiency rely often on changes in tumor volume measured as shrinkage by CT or MRI, which become apparent after multiple lines of treatment and pose a physical and psychological burden on the patient. Detection of therapy-induced cell death in the tumor can be a fast measure for treatment efficiency. However, there are no reliable clinical tools for detection of tumor necrosis. Previously, we studied the necrosis avidity of cyanine-based fluorescent dyes, which suffered long circulation times before tumor necrosis could be imaged due to low hydrophilicity. We now present the application of radiolabeled 800CW, a commercially available cyanine with high hydrophilicity, to image tumor necrosis in a mouse model. Procedures: We conjugated 800CW to DOTA via a PEG linker, for labeling with single-photon emission-computed tomography isotope indium-111, yielding [¹¹¹In]In-DOTA-PEG₄-800CW. We then investigated specific [¹¹¹In]In-DOTA-PEG₄-800CW uptake by dead cells in vitro, using both fluorescence and radioactivity as detection modalities. Finally, we investigated [¹¹¹In]In-DOTA-PEG₄-800CW uptake into necrotic tumor regions of a 4T1 breast tumor model in mice. Results: We successfully prepared a precursor and developed a reliable procedure for labeling 800CW with indium-111. We detected specific [¹¹¹In]In-DOTA-PEG₄-800CW uptake by dead cells, using both fluorescence and radioactivity. Albeit with a tumor uptake of only 0.37%ID/g at 6 h post injection, we were able to image tumor necrosis with a tumor to background ratio of 7:4. Fluorescence and radioactivity in cryosections from the dissected tumors were colocalized with tumor necrosis, confirmed by TUNEL staining. Conclusions: [¹¹¹In]In-DOTA-PEG₄-800CW can be used to image tumor necrosis in vitro and in vivo. Further research will elucidate the application of [¹¹¹In]In-DOTA-PEG₄-800CW or other radiolabeled hydrophilic cyanines for the detection of necrosis caused by chemotherapy or other anti-cancer therapies. This can provide valuable prognostic information in treatment of solid tumors.