Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Head and neck cancers include malignancies originating in the oral cavity, nasopharynx, oropharynx, hypopharynx and larynx. In Belgium, more than 90 % of head and neck cancers are squamous cell carcinomas (HNSCC), and more than 2500 patients were diagnosed with HNSCC in the year 2016. (Belgium Cancer Registry) The use of alcohol and tobacco, with a substantial synergistic effect, induces most HNSCC except for oropharyngeal (OPC) and nasopharyngeal cancer. These HNSCC subsites often have a viral aetiology, Human Papillomavirus (HPV) and Epstein Barr virus, respectively. In the Western world, the incidence of HPV positive OPC is rapidly increasing.The treatment of HNSCC depends mainly on the tumour subsite, tumour size, lymph node involvement and the presence of distant metastases, summarised in the TNM classification. Most early-stage HNSCC can be cured by surgery or radiotherapy (RT), which have similar locoregional control rates. The selection of treatment modality in early disease depends on patient characteristics, the expected post-treatment organ functionality and cosmesis. Unfortunately, the majority of patients with HNSCC have locally advanced disease at diagnosis and should, therefore, be treated with chemoradiotherapy (CRT) or with radical surgery followed by adjuvant (C)RT. Thus, RT is one of the cornerstones of the treatment of HNSCC. Since RT damages both normal and neoplastic cells, the clinical challenge in radical RT is to attain a high therapeutic ratio, which is the highest probability of cure with the least toxicity to surrounding healthy structures.Technical evolutions have made it possible to design high precision three dimensional (3D) conformal and intensity-modulated radiotherapy (IMRT) in which the high dose region is better sculpted around the target volumes, which made it possible to spare the healthy organs at risk (OAR) better. IMRT has become the prefered RT technique in the treatment of HNSCC since the randomised controlled PARSPORT trial has shown the ability and advantage of sparing the parotid salivary glands with IMRT. Despite the improved RT dose distribution on the OAR, acute and late toxicity such as dysphagia, remains a major problem. The risk of dysphagia is correlated with the dose to the swallowing muscles. With the current standard RT dose and fractionation, it is difficult to lower the dose on the swallowing muscles because of the proximity of the target volumes (tumour, nodes and elective neck). Therefore, the current RT procedure and protocol should be optimised in order to reduce treatment-related toxicity.We hypothesised that dose-de-escalation to the elective nodal neck (PTVelect) could lower the RT related toxicity such as dysphagia and lead to better QoL without compromising the oncological outcome. Therefore, a multicentre randomised controlled trial was set up comparing a lower dose to the PTVelect, EQD2 40 Gy, against the standard dose, EQD2 50 Gy. In previous publications, the dosimetric comparison, the effect on acute and late toxicity and the two-year tumour outcome were published showing a significant reduction in dose on several swallowing structures resulting in a substantial reduction in grade ≥3 dysphagia three months after radiotherapy. At later time points, only a trend toward less dysphagia was found. Regarding survival and tumour control, no statistically significant differences were found at two years. We assessed the long term oncological outcome with a special interest in the regional recurrence rate in the PTVelect and investigated the differences in QoL between the 40 Gy and 50 Gy arm.Since in this prospective randomised trial reduction of the dose to PTVelect did not result in a clear benefit in terms of decrease in late dysphagia, other factors may contribute to the development of severe dysphagia. We examined the role of patient and treatment characteristics on dysphagia scored on swallowing videofluoroscopy (VFS) as well as patient and physician scored dysphagia. Next, we evaluated if an existing prediction model for dysphagia, the total dysphagia risk score (TDRS), was also valid in our patient cohort.Besides efforts to reduce the treatment-related toxicity, we should also strive to tailor the treatment to the individual risk profile of the patient. Presently, all patients treated with curative RT receive a standard high dose of 70 Gy (in 35 fractions of 2 Gy), even though some tumours are more radiosensitive than others. The difference in radiosensitivity and prognosis is particularly proven in OPC. In general, HPV positive OPC have a better prognosis than HPV negative OPC. Treatment intensification, e.g. by dose escalation or hypoxia modification, may improve the oncological outcome for patients with poor prognosis while for others with excellent prognosis who might be overtreated, dose reduction could prevent long term toxicity. Good prognostic models are needed to individualise the patients' treatment. At the moment, these models are mostly based on the TNM classification in combination with patient characteristics such as smoking. We examined the prognostic impact of a few models and the new TNM classification in our patient population. Although the new TNM classification (8th edition) provides better OS stratification than the 7th edition, we are convinced that in the future genetic and radiologic information must be integrated to better select patients for de-escalation or escalation of the treatment. First, we examined the prognostic value of a 15-gene hypoxia classifier in OPC. Next, the role of diffusion-weighted (DW) MRI in OPC was investigated. We compared the DW-MRI first-order histogram and radiomic features between HPV positive and negative OPC.Thus, the objectives of this research project are twofold. The first aim is to reduce the treatment-related toxicity in HNSCC with particular focus on dysphagia. The second aim is to predict the prognosis for patients with OPC to tailor the treatment to the individual risk profile.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Background: hepatocellular carcinoma (HCC) shows strong pathological and molecular heterogeneity. Prognostication and outcome prediction is currently mainly based on clinical patient and tumor characteristics. While histologically proven progenitor markers, including Keratin 19 (K19), have been shown to be independent predictors of survival, they require biopsy or tumor resection to be clinically effective. Although magnetic resonance imaging (MRI) is the primary imaging modality for diagnosing and locoregional staging of HCC, MRI features that predict potential biological aggressive behavior or adverse pathological markers or predict treatment outcome are lacking. Objectives: this study aims to identify MRI biomarkers for the identification of the adverse prognostic histopathological progenitor cell marker K19 and evaluate their utility to predict treatment outcome in radiofrequency ablation (RFA) of HCC defined by progression-free survival (PFS) and overall survival (OS). Methods: in a first retrospective and consecutive cohort of 74 patients with whole tumor resection or transplantation for HCC we correlated the MRI features, including dynamic enhancement pattern, nodule‐in‐nodule appearance, T2-weighted and T1-weighted signal intensity and qualitative diffusion-weighted imaging (DWI) appearance of 95 histopathologically proven HCC with K19 status using chi-square test with multiple variables. Subsequently, in a second retrospective and consecutive cohort of 68 patients treated by laparoscopic RFA with peroperative laparoscopic biopsy and RFA of 68 HCC-lesions, we correlated the presence or absence of identified MRI-based K19 features with PFS and OS using Kaplan-Meier and log-rank tests. Results: on the radiological-pathological correlation cohort, of the 95 HCCs, 29 expressed K19 and 66 were K19 negative. Mixed signal at T2-weighted images (P=0.001), rim enhancement at arterial, portal-venous and late venous contrast-phase (P<0.000001) and rim-like b600 hyperintensity at DWI (P=0,00001) were significantly correlated with K19 positivity. Dynamic enhancement showed the highest accuracy with a positive predictive value of 77% and negative predictive value of 92% for predicting K19 positivity. In the second cohort of patients treated with RFA, median PFS and OS were 12.5 and 24 months, respectively. Survival analyses showed significant effects of arterial (P=0.02), portal-venous (P<0.0001) and late venous phase rim enhancement (P<0.0001) on PFS and substantial effect by late venous phase rim enhancement on OS (P=0.11). At multivariable analysis, portal-venous and late venous phase rim enhancement were significant predictors of PFS (P<0.0001; Hazard Ratio=33.79) and OS (P<0.0001, HR=37.16). Conclusion: rim-like dynamic contrast enhancement, heterogeneous T2 signal and rim-like signal at DWI are predictive MRI biomarkers of K19 positivity in HCCs, with dynamic contrast-imaging achieving the highest accuracy. Rim-like contrast enhancement in either portal-venous or late-venous phase can be used as independent predictor of PFS and OS in HCCs treated by laparoscopic RFA.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Objective: To evaluate 3 Tesla (T) whole-body diffusion-weighted magnetic resonance imaging (WB DWI) for early treatment assessment in aggressive non-Hodgkin lymphoma (NHL). Methods: Fourteen patients with NHL treated with standard chemotherapy underwent 3-T WB DWI before and 2 and 4 weeks during treatment, using b-values of 01000 s/mm2 from which the apparent diffusion coefficient (ADC) was calculated. Patient follow-up (average 20.3 months, range 1523 months) was the reference standard. Volume and ADC changes between baseline and 2 weeks (Vratio2w, ADCratio2w) and 4 weeks (Vratio4w, ADCratio4w) of responding and non-responding lesions (lymph node and organ lesions) were compared using MannWhitney U tests. The per patient values of VratioN and ADCratioN to predict progression-free survival were determined with a log-rank test. Results: Eight patients showed complete remission and 6 showed tumour progression. The ADCratio2w and ADCratio4w differed significantly in lesions showing tumour progression versus complete remission (ADCratio2w ¼ 4 21% versus 119 68%; ADCratio4w ¼ 18 61% versus 155 78%; both P50.0001); the Vratio2w and Vratio4w did not (P40.05). Per body region, the ADCratio2w showed a negative predictive value of 100% and positive predictive value of 86%. Per patient, the ADCratio2w and ADCratio4w correlated significantly with progression-free survival (P50.05). Conclusion: 3-T WB DWI with ADC quantification may enable early treatment assessment of aggressive NHL.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Hepatic haemangiomas are congenital vascular malformations and are the most common benign solid tumours of the liver. If they exceed 4 centimeters in diameter they are called “giant haemangiomas”. Most cases are asymptomatic. Although they seldom spontaneous rupture, it is important to diagnose them as their global mortality rate is high. An accurate diagnosis of a ruptured giant haemangioma as the cause of a spontaneous haemoperitoneum would result in correct clinical decision making and treatment.

Choose an application

• Reference Manager
• EndNote
• RefWorks (Direct export to RefWorks)

Ovarian cancer is the second most lethal gynaecological cancer in women and is often referred to as ‘the silent killer’ because it often spreads throughout the abdomen without causing symptoms. As a consequence, most patients are diagnosed with extensive peritoneal disease and/or distant metastases. Primary debulking surgery (PDS) is the main treatment for ovarian cancer with complete macroscopic tumour resection (R0 resection) as the most important prognostic factor. If disease extent precludes upfront radical surgery, neoadjuvant chemotherapy (NACT) followed by interval debulking surgery (IDS) represents an alternative for patients with advanced-stage disease. But despite high complete resection rates after surgical debulking, most patients ultimately suffer from tumour recurrence with a 5-year survival of only 40%.The diagnosis and follow-up of patients with ovarian cancer partly relies on medical imaging but there is an obvious unmet clinical need related to the imaging of ovarian cancer. In ovarian cancer, imaging should aim to identify patients unfit for surgery, either at primary diagnosis, after NACT and in the recurrent setting. In addition, pretreatment prognostic imaging biomarkers should aim to individualize therapeutic decisions. Although 18F-fluorodeoxyglucose positron emission tomography/computed tomography (FDG-PET/CT) has an incremental benefit over CT for the detection of lymphadenopathies and distant metastases, both may underestimate peritoneal disease load. In particular, mesenteric and serosal bowel disease dissemination which is necessary for operability assessment, is often underestimated with CT and FDG-PET/CT. Therefore often surgical staging with diagnostic open laparoscopy is necessary for prediction of complete resection. Diffusion-weighted imaging provides image contrast based on water molecule movement allowing functional characterization of reflecting tissue microstructural properties on a cellular level. This high image contrast with millimetric spatial resolution of DWI allow visual depiction of small tumour deposits. In addition, DWI can be quantified by the apparent diffusion coefficient (ADC). The combination of diffusion-weighted imaging (DWI) with contrast-enhanced magnetic resonance imaging (MRI) has already been demonstrated to show a high accuracy of detection of PC and staging. Because at least 11% of ovarian cancer patients present with distant metastases at primary diagnosis and up to 22% in the recurrent setting, the development of whole body diffusion-weighted MRI (WB-DWI/MRI) is justified.In this thesis work, we explored the potential of whole body diffusion-weighted MRI (WB-DWI/MRI) to improve the workup of patients with ovarian cancer in seven experiments.In a first study, we assessed the clinical feasibility of WB-DWI/MRI in 32 patients suspected for ovarian cancer, in comparison with CT and FDG-PET/CT. We prospectively validated the interpretation criteria and diagnostic value of WB-DWI/MRI with pathology after surgery as reference standard. We demonstrated that the main gains of using WB-DWI/MRI were the detection of bowel serosal and mesenteric disease, important towards operability, with a higher accuracy of detecting peritoneal disease compared with CT and FDG-PET/CT. In addition, we showed that distant metastases were detected by WB-DWI/MRI with equal accuracy to FDG-PET/CT. The proposed imaging protocol was afterwards implemented in the other clinical studies in this thesis work.In a second study, the reproducibility of WB-DWI/MRI for the assessment of primary tumour origin, tumour characterization, staging and prediction of complete resection was validated. We observed a fair to perfect interobserver agreement for WB-DWI/MRI with a minimal training of a general radiologist without previous experience in WB-DWI/MRI. In contrast, the interobserver agreement for CT between two abdominal radiologists was somewhat lower. We therewith demonstrated the applicability of and reproducibility of WB-DWI/MRI.Third, a prospective superiority study was conducted in 166 patients with suspected ovarian cancer for prediction of complete resection. As the goals for preoperative staging of ovarian cancer are [1] the confirmation of a malignant mass, [2] the exclusion of non-ovarian primary tumours, [3] the assessment of disease extent and [4] the prediction of R0 resection, we compared the diagnostic accuracy of WB-DWI/MRI compared to CT for these 4 aims. We showed that WB-DWI/MRI showed higher accuracy for characterization of malignancy, identification of patients with primary non-ovarian cancer, staging and prediction of R0 resection, compared to CT. We therefore concluded that WB-DWI/MRI has an incremental value for individualized preoperative staging in suspected ovarian cancer and can therefore be an alternative for CT.As it generally believed that the prediction of optimal surgery is difficult based on imaging alone, we set up a fourth prospective non-inferiority study to compare the diagnostic accuracies of WB-DWI/MRI versus CT in combination visual assessment during diagnostic open laparoscopy for assessment of primary tumour origin, tumour characterization, staging and prediction of R0 resection. The main finding this study was the non-inferiority of WB-DWI/MRI for accurate prediction of R0 resection compared to DOL-CT. This was accompanied with significantly non-inferior accuracy for assessment of primary tumour origin, tumour characterization, identification of patients with non-ovarian primary cancer and assessment of disease extent according to FIGO staging and suggests that WB-DWI/MRI can be considered for operability assessment in patients suspected with ovarian cancer.In a fifth experiment, the prognostic value of first –and second order statistics was compared to the average ADC derived from pretreatment WB-DWI/MRI ADC histogram towards disease free (DFS) and overall survival (OS) in advanced-stage ovarian cancer. We showed that first –and second order statistical features from the 2D histogram derived from WB-DWI/MRI images were associated with survival in advanced-stage ovarian cancer and could provide prognostic non-invasively derived biomarkers to tailor personalized medicine.In experiment 6, we evaluated WB-DWI/MRI after neoadjuvant chemotherapy to predict R0 resection and to discriminate long –and short-term survival among patients with extensive advanced-stage ovarian cancer. From this study, we concluded that WB-DWI/MRI could accurately predict R0 resection after NACT.The last part of this thesis work focused on the prospective evaluation of WB-DWI/MRI compared to CT for detection, restaging and operability assessment in patients with suspected ovarian cancer recurrence. We showed that besides a significantly higher accuracy for tumour detection and improved sensitivity for per-site analysis of disease extent, WB-DWI/MRI showed better sensitivity for assessing disease extent critical towards operability, particularly for mesenteric root infiltration, small bowel and colon carcinomatosis, high volume peritoneal disease and irresectable distant metastases, compared to CT. As a consequence, WB-DWI/MRI was superior for prediction of complete cytoreduction as compared to CT.Although the results of this thesis contribute to establish the role of WB-DWI/MRI in ovarian cancer, future research and further standardization is warranted to gain acceptance for WB-DWI/MRI as a clinically valid and applicable imaging technique.