DCE-CT of Thoracic Tumors as an Early Biomarker for Treatment Monitoring in Comparison With Morphologic Criteria

DCE-CT of Thoracic Tumors as an Early Biomarker for Treatment Monitoring in Comparison With Morphologic Criteria

DCE-CT of Thoracic Tumors as an Early Biomarker for Treatment Monitoring in Comparison With Morphologic Criteria

DCE-CT of thoracic tumors as an early biomarker for treatment monitoring in comparison with morphologic criteria. Rationale of the clinical investigation For the evaluation of response to anti-tumoral therapy in thoracic tumors, merely morphologic information is often not sufficient for early response evaluation as dimensions of the oncologic lesions are not changing during the first weeks of treatment. To be able to measure functional changes, dynamic contrast-enhanced CT (DCE-CT) seems promising as a biomarker for early therapy monitoring. Having an early biomarker for treatment monitoring will allow to increase patients' prognosis if a non-responder is earlier detected, will optimize the use of expensive treatments, is expected to shorten hospitalization and shorten absence at work, and to decrease side-effects of (adjuvant) medication. Objective of the study 2.1.Primary objectives The primary objective is to investigate the potential of functional imaging (i.e. DCE-CT), as analyzed by the Hyperfusion analytic software, as an early biomarker for the evaluation of therapy response in primary thoracic malignancy. 2.2.Secondary objectives There are two secondary objectives: To define internal system parameters and perfusion parameter thresholds that maximize the accuracy of the outcomes and to define the correct category (PD, SD, PR, CR); and To compare the predicted categorization to the assessed RECIST1.1 categorization. Endpoints 3.1.Primary Endpoint The primary endpoint is to directly compare the biomarker of the HF analysis software at week 3 (+- 1 week) and week 8 (+- 3 weeks) with the eventually reported Progression-Free Survival (PFS) intervals and Overall Survival (OS) in this study. PFS intervals are determined by the clinician and are based on RECIST1.1 and additional clinical and biochemical progression markers. The focus will be on evaluating the accuracy of the prediction as well as how early the prediction was correct. 3.2.Secondary Endpoints There are two secondary endpoints corresponding to the two secondary objectives. The internal parameters for the HF biomarker, e.g. magnitude of the Ktrans decrease, and the change in volume of unhealthy tissue, need to be determined to define the classification (PD, SD, PR and CR) by the HF analysis software. These parameters are optimized to optimally predict the classification according to PFS and OS. This will be done by splitting the data into a train and test set to ensure generalization. The classification of the HF analysis software will be compared to the purely morphological classification by RECIST1.1 to identify correlation. Furthermore, some cases will be investigated where the HF analysis performs noticeably better or worse than RECIST1.1 in predicting PFS and OS. Finally, the difference in time to the first correct prediction is compared between HF and RECIST1.1. 4.Study Design This prospective study is part of the clinical β-phase. We aim to test pre-release versions of the Hyperfusion.ai software under real-world working conditions in a hospital (clinical) setting. It is important to note, though, that the results of the software analysis will not be used by interpreting physicians to alter clinical judgement during the course of the clinical trial. A prospective study including 100 inoperable patients in UZ Gent suffering from primary thoracic malignancy (≥15mm diameter) will be conducted. For this study, in total 3 CT scan examinations of the thorax will be performed (a venous CT examination of the thorax in combination with a DCE-CT scan of the tumoral region). All patients will be recruited from the pulmonology department. Oncologic patients are clinically referred with certain intervals for a clinically indicated CT scan (being part of standard care). In the study, two clinical CT examinations that are performed standard of care (baseline CT examination and CT examination at week 8 (+- 3 weeks) after start of systemic therapy) will be executed by also adding a DCE-image of the lung adenocarcinoma to this examination. This DCE-image is performed during the waiting time before the venous/morphologic phase. Consequently, from a clinical point-of-view, the time to scan remains exactly the same. With regard to the contrast agent, an identical amount is injected as is the case in standard of care, but the contrast bolus is split in two parts - see also addendum with DCE protocol. In this study there is one additional CT-examination (DCE-scan of the thoracic malignancy in combination with venous CT scan of the thorax) at week 3 (± 1 week).

Introduction Background information Below, you will find an overview the publications on DCE-CT, in the context of the evaluation of treatment effect in cancer patients: Strauch L et al (Diagnostics 2016: 21;6(3) - doi: 10.3390/diagnostics6030028) provide an overview of the literature available on DCE-CT as a tool to evaluate treatment response in patients with lung cancer. In studies where patients were treated with systemic chemotherapy with or without anti-angiogenic drugs, four out of the seven studies found a significant decrease in permeability after treatment. Four out of five studies that measured blood flow post anti-angiogenic treatments found that blood flow was significantly decreased. This review concluded that DCE-CT may be a useful tool in assessing treatment response in patients with lung cancer. It seemed that particularly permeability and blood flow are important perfusion values for predicting treatment outcome. However, the heterogeneity in scan protocols, scan parameters, and time between scans makes it difficult to compare the reviewed papers. The research group of van Elmpt W and Lambin P (Radiother Oncol 2017: 125(3):379-384) identified tumor subregions with characteristic phenotypes based on pre-treatment multi-parametric functional imaging and correlated these subregions to treatment outcome. The subregions were created using imaging of metabolic activity (FDG- Positron Emission Tomography (PET)/CT), hypoxia (HX4-PET/CT) and tumor vasculature (DCE-CT). Thirty-six non-small cell lung cancer (NSCLC) patients underwent functional imaging prior to radical radiotherapy. Kinetic analysis was performed on DCE-CT scans to acquire blood flow (BF) and volume (BV) maps. HX4-PET/CT and DCE-CT scans were non-rigidly co-registered to the planning FDG-PET/CT. Two clustering steps were performed on multi-parametric images: first to segment each tumor into homogeneous subregions (i.e. supervoxels) and second to group the supervoxels of all tumors into phenotypic clusters. Patients were split based on the absolute or relative volume of supervoxels in each cluster; overall survival was compared using a log-rank test. Unsupervised clustering of supervoxels yielded four independent clusters. One cluster (high hypoxia, high FDG, intermediate BF/BV) related to a high-risk tumor type: patients assigned to this cluster had significantly worse survival compared to patients not in this cluster (p = 0.035). It was concluded that subregional analysis for multi-parametric imaging in NSCLC has the potential as a biomarker for prognosis. This methodology allows for a comprehensive data-driven analysis of multi-parametric functional images. Qiao P et al (Clin Transl Oncol 2016: 18(1):47-57) have studied the feasibility and clinical value of DCE-CT for early evaluation of targeted therapy efficacy in non-small cell lung cancer (NSCLC). They measured tumor diameter, peak height (PH), time to peak (TP), tumor mass-aortic peak height ratio (M/A), and blood perfusion (BP) in 20 patients with advanced NSCLC using DCE-CT before and 7 days after treatment. Therapy efficacy was assessed with conventional CT 4-6 weeks post-treatment. Patients were grouped into those with partial response (PR), stable disease (SD), and progressive disease (PD) according to the therapy efficacy assessment at 4-6 weeks post-treatment. The PR group primary tumor diameter (P = 0.0007) and BP (P = 0.0225) were reduced at 7 days post-treatment; the SD group DCE-CT value changes were not significant. The PD group M/A (P = 0.0443) and BP (P = 0.0268) were increased 7 days post-treatment. The BP decrease group had significantly longer progression-free survival than the BP increase group (median, 54 vs. 6 weeks). This study concluded that DCE-CT can evaluate targeted therapy efficacy at 7 days post-treatment. Decreased primary tumor diameter and BP indicate tumor sensitivity to therapy; increased BP with unchanged tumor diameter suggests the tumor is not sensitive to therapy. Reduced BP suggests treatment effectiveness. Hwang et al (Eur Radiol 2013: 23(6):1573-81) compared tumor enhancement patterns measured using DCE-CT with tumor metabolism measured using PET-CT in patients with non-small cell lung cancer (NSCLC) and stable disease after chemotherapy or chemoradiotherapy. After treatment, 75 NSCLC tumors in 65 patients who had stable disease on DCE-CT according to Response Evaluation Criteria in Solid Tumor (RECIST) were evaluated using PET-CT. On DCE-CT, relative enhancement ratios (RER) of tumor at 30, 60, 90, 120 s and 5 min after injection of contrast material were measured. Metabolic responses of tumors were classified into two groups according to the maximum standardized uptake value (SUVmax) by PET-CT: complete metabolic response (CR) with an SUVmax of less than 2.5, and non-complete metabolic response (NR) with an SUVmax of at least 2.5. Using the optimal RER₆₀ cutoff value of 43.7 % to predict NR of the tumor gave 95.7 % sensitivity, 64.2 % specificity, and 82.1 % positive and 95.0 % negative predictive values. After adjusting for tumor size, the odds ratio for NR in the tumor with an RER60 of at least 43.7 % was 70.85 (95 % CI = 7.95-630.91; P < 0.05). Even when disease was stable according to RECIST, DCE-CT predicted hypermetabolic status of residual tumor in patients with NSCLC after treatment. Rationale of the clinical investigation For the evaluation of response to anti-tumoral therapy in thoracic tumors, merely morphologic information is often not sufficient for early response evaluation as dimensions of the oncologic lesions are not changing during the first weeks of treatment. To be able to measure functional changes, dynamic contrast-enhanced CT (DCE-CT) seems promising as a biomarker for early therapy monitoring. Having an early biomarker for treatment monitoring will allow to increase patients' prognosis if a non-responder is earlier detected, will optimize the use of expensive treatments, is expected to shorten hospitalization and shorten absence at work, and to decrease side-effects of (adjuvant) medication. Objective of the study Primary objectives The primary objective is to investigate the potential of functional imaging (i.e. DCE-CT), as analyzed by the Hyperfusion analytic software, as an early biomarker for the evaluation of therapy response in primary thoracic malignancy. Secondary objectives There are two secondary objectives: To define internal system parameters and perfusion parameter thresholds that maximize the accuracy of the outcomes and to define the correct category (PD, SD, PR, CR); and To compare the predicted categorization to the assessed RECIST1.1 categorization. Endpoints Primary Endpoint The primary endpoint is to directly compare the biomarker of the HF analysis software at week 3 (+- 1 week) and week 8 (+- 3 weeks) with the eventually reported Progression-Free Survival (PFS) intervals and Overall Survival (OS) in this study. The prediction of the four classes (PD, SD, PR and CR) is based on "significant changes" (see 7.2) in HF biomarker. PFS intervals are determined by the clinician and are based on RECIST1.1 and additional clinical and biochemical progression markers. The focus will be on evaluating the accuracy of the prediction as well as how early the prediction was correct. Secondary Endpoints There are two secondary endpoints corresponding to the two secondary objectives. The internal parameters for the HF biomarker, e.g. magnitude of the Ktrans decrease, and the change in volume of unhealthy tissue, need to be determined to define the classification (PD, SD, PR and CR) by the HF analysis software. These parameters are optimized to optimally predict the classification according to PFS and OS. This will be done by splitting the data into a train and test set to ensure generalization. The classification of the HF analysis software will be compared to the purely morphological classification by RECIST1.1 to identify correlation. Furthermore, some cases will be investigated where the HF analysis performs noticeably better or worse than RECIST1.1 in predicting PFS and OS. Finally, the difference in time to the first correct prediction is compared between HF and RECIST1.1. Study Design This prospective, monocentric (currently, however we do foresee to collaborate with other centres to include patients with colorectal liver metastases; also other types of solid tumors could be included; see ) study is part of the clinical β-phase. We aim to test pre-release versions of the Hyperfusion.ai software under real-world working conditions in a hospital (clinical) setting. It is important to note, though, that the results of the software analysis will not be used by interpreting physicians to alter clinical judgement during the course of the clinical trial. A prospective study including 100 inoperable patients in UZ Gent suffering from primary thoracic malignancy (≥15mm diameter) will be conducted. For this study, in total 3 CT scan examinations of the thorax will be performed (a venous CT examination of the thorax in combination with a DCE-CT scan of the tumoral region). All patients will be recruited from the pulmonology department. Oncologic patients are clinically referred with certain intervals for a clinically indicated CT scan (being part of standard care). In the study, two clinical CT examinations that are performed standard of care (baseline CT examination and CT examination at week 8 (+- 3 weeks) after start of systemic therapy) will be executed by also adding a DCE-image of the primary thoracic malignancy to this examination. This DCE-image is performed during the waiting time before the venous/morphologic phase. Consequently, from a clinical point-of-view, the time to scan remains exactly the same. With regard to the contrast agent, an identical amount is injected as is the case in standard of care, but the contrast bolus is split in two parts - see also addendum with DCE protocol. In this study there is one additional CT-examination (DCE-scan of the thoracic malignancy in combination with venous CT scan of the thorax) at week 3 (± 1 week). Population Number of subjects At least one hundred patients will be included in this study. Inclusion criteria Inoperable patients suffering from primary thoracic malignancy or second line patients having had a therapy pause of at least 6 weeks; at least one tumoral lesion/component should be at least 15mm in diameter. Exclusion criteria All patients less than 18-years-old. Documented allergy for iodine. Neutropenia (absolute White Blood Cell count ≤ 1.5 × 109/l). Thrombopenia (absolute platelet count ≤ 100 × 109/l). Renal insufficiency: serum creatinine ≥ 1.5× the upper limit of normal (ULN); 24-hours creatinine clearance ≤ 50ml/min). Serum bilirubine ≥ 1,5 x ULN, AST ≥ 2,5 x ULN, ALT ≥ 2,5x ULN. Brain metastases. Withdrawal and replacement of subjects Criteria for withdrawal Subjects may prematurely discontinue from the clinical investigation at any time. Premature discontinuation from the study is to be understood when the subject did not undergo end of study examination. Subjects can be withdrawn under the following circumstances: at their own request; if the investigator feels it would not be in the best interest of the subject to continue; if the subject violates conditions laid out in the informed consent form or disregards instructions by the clinical investigation personal; in case of significant study intervention non-compliance; In all cases, the reason why subjects are withdrawn will be recorded in detail in the CRF and in the subject's medical records. A subject will be considered lost to follow-up if he or she fails to return for the scheduled visits and is unable to be contacted by the study site staff. The following actions must be taken if a subject fails to return to the clinic for a required study visit: The site will attempt to contact the subject and reschedule the missed visit. Before a subject is deemed lost to follow-up, the investigator or designee will make every effort to regain contact with the subject (where possible, 3 telephone calls and, if necessary, a certified letter to the subject's last known mailing address or local equivalent methods). These contact attempts should be documented in the subject's medical record or study file. Should the subject continue to be unreachable, he or she will be considered to have withdrawn from the study with a primary reason of being lost to follow-up. Should the clinical investigation be discontinued prematurely, all clinical investigation materials will be retained and the sponsor will notify the relevant regulatory authorities and ethical committees. Replacement policy Drop-outs will be replaced. Restrictions and prohibitions for the subjects All patients must refrain from any food and drinks at least 4 hours before the CT-examination (standard of care). Possible advantages and risks for the subjects At this stage there are no advantages for the patients studied. As a result of this study, we anticipate that DCE-CT and the concurring Hyperfusion.ai analysis will function as an early biomarker for treatment monitoring, thereby optimizing tumor treatments. So, for future patients it is hoped that this study will prolong the PFS (progression-free survival). There are no other risks than those associated with an intravenous injection of iodine-based contrast-agent (fully comparable with standard of care IV injections of iodine). In this patient population, there is no risk associated with the increased x-ray dose applied. Methodology Medical device description Intended use and instructions for use Intended use Hyperfusion.ai's software as a medical device is intended to be used by physicians in order to perform a post-processing analysis of DCE-imaging, with the intention of accurately calculating perfusion parameters. Hyperfusion.ai's software is a radiological computer-assisted therapy monitoring software device intended to be used concurrently by interpreting physicians while analysing DCE-images that stem from medical imaging types and modalities, such as compatible computed tomography scanners (CT), magnetic resonance imaging scanners, and others. The system computes perfusion parameters within one scanning session and calculates the relative difference between different scanning sessions. These findings assist interpreting physicians in evaluating tumor response rate that may be confirmed or dismissed by the interpreting physician. Consequently, this software is perceived as Class IIb, according to (Rule 11) of the MDR 2017/745, in the idea that Hyperfusion's software is 'intended to provide information which is used to take decisions with diagnosis or therapeutic purposes.' Delivery of the medical device The medical device software will not be accessible at the site during the course of the clinical trial. Hyperfusion's software analysis is cloud-based and can currently only be accessible by personnel of the manufacturer. During the course of the clinical trial, the site (Department of Radiology) will be asked to send the DICOM-data to Hyperfusion.ai via a highly-secure DNS-server that will be provided by Hyperfusion.ai. The processing result will be made available to the physician for download at an agreed cloud location. Storage of the medical device The output of Hyperfusion.ai software analysis, will be stored at the highly-secure cloud storage servers Hyperfusion uses. Packaging and labeling of the medical device The results will be provided to the physician in a digital form including the perfusion parameter maps and the generated PDF-report. There will be a supporting document (Instructions for use) to aid physicians in accurately interpreting the report. The following information will be mentioned as 'label' on the PDF report: The Medical Device Regulation 2017/745 requires for each investigational device to be specifically labelled. As this concerns a software analysis (without user interface) which generates a PDF-report for the user, the first page of this (unique) report contains the following information: Identification of the investigational device Hyperfusion.ai Kleemburg 1, 9050 Gentbrugge Belgium info@hyperfusion.ai The information in this report can only be used in the context of clinical investigation. If interpreting physicians intend to use this report to alter their clinical judgement, Hyperfusion.ai will not be held accountable. UDI: 'HF-61fcda38 Report identificator: 'HF-8901 Known reactions/side effects of the medical device Since the medical device functions in the form of a pipeline of software-algorithms, no side effects can occur. Study Specific Procedures Screening visit This step is standard of care. At time of the first referral consultation within the framework of the patient's oncologic rehabilitation, the pneumonologist will determine whether the patient is eligible for systemic treatment, and fits the inclusion and exclusion criteria. Informed consent will be explained to the patient at first by the study coordinator. If the patient is willing to participate more detailed information is communicated by the physician. Final informed consent is obtained by the physician and 3 CT examinations (baseline, week 3 and week 8) will be planned. Baseline CT examination A baseline CT-examination is standard of care. Maximally 28 days before the start of systemic treatment the baseline CT examination is performed. This includes a standard of care CT in combination with a study specific DCE-CT scan. Detailed information on how to perform the study specific DCE-CT scan can be found in addendum with DCE-CT protocol. The patient receives an identical dose of intravenous iodine based contrast agent as in standard of care (venous phase) CT examination (contrast agent bolus is split in 2 parts - see explanation earlier/above). Follow-up visit 1 The follow-up visit 1 is standard of care. 3 weeks (+/- 1 week) after the start of the systemic treatment, a study specific DCE-CT scan will be scheduled at this visit in combination with a venous phase CT. Detailed information on how to perform the study specific DCE scan can be found in addendum with DCE-CT protocol. The patient receives an identical amount of contrast agent during this CT examination when compared to standard of care (venous) CT. Follow-up visit 2 The follow-up visit 2 is standard of care. 8 weeks (+/- 3 weeks) after the start of the systemic treatment follow-up visit 2 is planned. This visit is identical as follow-up visit 1. Follow-up until PFS. After follow-up visit 2, no study specific assessments will be performed anymore. Data on the morphologically (standard of care) CT examinations based RECIST 1.1 will be collected until PFS (and potentially Overall Survival). All patients will be in follow-up at least until PFS (even after closing the clinical trial when the last patient has been included). A maximum term for follow-up is however set at 1 year after the initiation of therapy in all patients. All patients are followed-up in standard care until PFS. For this clinical trial we aim at following-up the patient until PFS but a maximum term for follow-up is set at 1 year after the initiation of therapy for the patient. Data on adverse events will not be collected anymore during follow-up until PFS. Only 'Death' will be reported to the sponsor. During the clinical trial, data on tumor-type, applied systemic treatment, medical history, gender and age will be collected. Flowchart Screening Baseline (T0 -28d) start systemic treatment (T0) FU visit 1 (T0 + 3 weeks (+/- 1 week)) FU visit 2 (T0 +8 weeks (+/- 3 weeks)) FU until PFS (max 1 year) Informed consent x Inclusion/exclusion criteria check x medical history x demographic data x morphologic scan, including RECIST x x x x x documentation of applied treatment x DCE- CT scan x x x Adverse event check x x x x reporting of death x x x x x FU = Follow-up End of study The end of study intervention is reached when the last subject (inclusion of patient 100), has completed follow-up visit 2. Overall, the end of the study is reached when PFS is reached in all included patients or 1 year after the last patient has started treatment. Between the end of study intervention and PFS, data of standard of care CT examinations based on RECIST 1.1 will be collected. Blinding Blinding of patients or study personnel is not applicable in this trial. Study analysis Sample size calculation One hundred patients will be used as a sample size. A preliminary power analysis for determining the sample size was performed. Preliminary results have shown that a Ktrans reduction of 40% would be indicative for differentiating between treatment responders and non-responders. A preliminary power analysis under these assumptions shows that for a statistically significant difference in mean value of 0.6 results in a power of 0.8439 for 100 patients. Statistical analysis The statistical analysis would be performed in-house by Ir. Verhack, PhD (ruben@hyperfusion.ai) and Ir. Van den Abeele, PhD (floris@hyperfusion.ai). In the following, two analyses will be proposed that support the primary endpoint and the second secondary endpoint. Both analyses will be performed twice, once using the HF prediction at week 3 (+- 1 week) and once using the prediction of week 8 (+- 3 weeks). Two interim analyses will be performed over the course of the clinical study. The first will be after 10 patients have been followed up for 6 months, and the second will be when 50 patients have been followed up for a year. The final analysis will be performed at the end of the year 2 follow-up. Analysis of correlation between HF biomarker and PFS and OS First, the goal is to investigate if the HF biomarker based on Ktrans is indeed a viable early biomarker to indicate treatment response or not. Therefore, a to-be-defined descriptive statistic of the evolution of Ktrans (i.e. HF biomarker) will be compared to the binary dependent variable responder/non-responder. The dependent variable is determined from PFS and, in parallel, from OS. If there has been no event (progression or death) before the last follow-up, then the patient is considered to be a "responder". Two hypotheses will be tested. The first hypothesis states that the HF biomarker does not significantly decrease for the responders group. The second hypothesis states that for the non-responder group, there is no significant decrease in the HF biomarker. In order to assess statistical significance for testing the null hypotheses, a paired t-test is used with a p-value of 0.05 is considered. Note that the above sample size calculation is based on this analysis. Second, an analysis will be performed in which the survival curves of each prediction class (CR, PR, SD, PD) is compared. As such, the duration until progression is taken into account as well. Alternatively, these categories can be grouped into responders (CR, PR) and non-responders (SD, PD). The survival curves are estimated using Kaplan-Meier estimates [Friedman2010], keeping in mind the common pitfalls in working with censored data using PFS [Korn2013]. Analysis of the correlation between the HF classification and RECIST1.1 Both prediction strategies result in the same classification which involves ordinal data with four choices (CR, PR, SD, PD). The statistical analysis will investigate the correlation between the two ordinal variables. The null hypothesis is that there is no linear relation between the two variables. A p-value of 0.05 will be used to determine statistical significance. The two ordinal variables will be compared using cross tabulation, Pearson and Spearman's correlation coefficient. Alternatively, if there are issues with this statistical analysis, e.g. it might be difficult for the HF software to distinguish between SD and PD, then these two categories will be merged accompanied by an explanation. [Korn2013] Korn, R. L., & Crowley, J. J. (2013). Overview: Progression-Free Survival as an Endpoint in Clinical Trials with Solid Tumors. Clinical Cancer Research, 19(10), 2607-2612. https://doi.org/10.1158/1078-0432.CCR-12-2934 [Friedman2010] Friedman, L. M., Furberg, C. D., & DeMets, D. L. (2010). Fundamentals of Clinical Trials. https://doi.org/10.1007/978-1-4419-1586-3 Indemnity insurance During their participation in the clinical investigation the patients will be insured as defined by legal requirements. An insurance with no fault responsibility has been foreseen by the sponsor in accordance with the Belgian law concerning experiments on humans, 7 May 2004. Final Report Within one year after the final completion of the study, a full final report will be written by the sponsor and submitted to the central ethical committee and competent authority. The Sponsor will pass this report to all local Principal Investigators for submission to their local EC if defined by their institution's procedure. Publication policy This study will be registered at Clinicaltrials.gov prior to inclusion of the first subject. Results information from this study will be submitted to Clinicaltrials.gov. Furthermore, the sponsor & cooperating departments aim to publish in the following journals: The New England Journal of Medicine British Medical Journal The Lancet Oncology Journal of Clinical Oncology Data Handling Case Report Form (CRF) The source documents are to be completed at the time of the subject's visit. The CRFs are to be completed within reasonable time after the subject's visit. For this study a digital CRF, Smart-Trial software will be used. The following data might be collected in the CRFs over the course of the clinical trial: Demographic info Age Gender Medical history Other relevant demographic parameters Diagnostic info: Clinical diagnosis files Medication Specific to tumoral pathology Chemotherapy: Platinum-based doublets, carboplatin/nab-P regimen, … Targeted therapy (biologicals: Pembrolizumab, anti-PD-(L1) inhibitors, Nivolumab plus ipilimumab,...) Immunotherapy General medication Blood analysis Complet (red blood cells, white blood cells, platelets) Liver enzymes (transaminases, bilirubin, gGT) Renal function (creatinine, ureum, Glomerular Filtration Rate) Bone biochemistry (calcium, calcitonin) Tumor markers: CEA PET/CT-examinations: DICOM-images of other examinations that will be performed over the course of this clinical trial. For each subject enrolled the CRF will be signed by the principal investigator or co-investigator. This also applies to those subjects who fail to complete the study. If a subject withdraws from the study, the reason must be noted on the CRF. CRF entries and corrections will only be performed by study site staff, authorized by the investigator and in accordance with ISO 14155. Errors will be logged by the software and corrections will be initiated and dated by the study member that made the correction. Entries will be checked by trained personnel (Monitor) and any errors or inconsistencies will be changed immediately. The original completed and signed CRFs will be collected at the end of the study by the sponsor. The Principal investigator must verify that all data entries in the CRFs are accurate and correct. If certain information is Not Done, Not Available or Not Applicable, "N.D." or "N.AV." or "N.AP", should be entered in the appropriate space. Data directly collected in the CRF (no source available) The DCE-images (in DICOM-format) can be regarded as sources-files and will, consequently, be collected in the CRF directly. They will be pseudonymized before being sent to the sponsor. Pseudonymized data will be sent to the cloud where it can be downloaded by Hyperfusion.ai. Hyperfusion.ai analysis will be performed in a pseudonymised way. A PDF-report will be shared after each CT examination to the study coordinator. Direct access to source data / documents The investigator will permit trial-related monitoring, audits, IRB/IEC review, and regulatory inspection(s), providing direct access to source data/documents. Archiving The investigator and sponsor specific essential documents will be retained for at least 20 years. At that moment, it will be judged whether it is necessary to retain them for a longer period, according to applicable regulatory or other requirement(s). Quality assurance and periodic monitoring The investigator will maintain adequate and accurate records to enable the conduct of the study to be fully documented and the study data to be subsequently verified. These documents will be classified into two different categories: investigator's file, and subject clinical source documents. The investigator's file will contain the documents as per EUROPEAN Standard of EN ISO 14155 (incl. GCP) and local regulations. Regular monitoring will be performed by Hiruz CTU according to ICH GCP and ISO 14155. Data will be evaluated for compliance with the protocol and accuracy in relation to source documents. Following written standard operating procedures, the monitors will verify that the clinical trial is conducted and data are generated, documented and reported in compliance with the protocol, ISO 14155 and the applicable regulatory requirements. To be ISO 14155 compliant at least 3 monitoring visits are scheduled. An initiation visit, one routine visit and a final visit after the last patient had finished the study. The monitor will be working according to SOPs and will provide a monitoring report after each visit for the sponsor and a follow-up letter to the investigator. Depending on the quality of the data, additional monitoring visits will be necessary according to the sponsor's discretion. More detailed information regarding the monitoring can be found in the monitoring plan which will be the responsibility of the HIRUZ. 14 Designation I certify that I will conduct the study in compliance with the protocol, any amendments, GCP/ISO 14155, the declaration of Helsinki, and all applicable regulatory requirements. Principal Investigator: Name: Professor Veerle Surmont Title: Prof. dr. Date: 18/07/2020

Cancer, Lung, Perfusion Computed Tomography Target Lesion, Cancer Liver, Metastatic Lung Cancer, Metastatic Liver Cancer

Patients suffering from primary malignant thoracic tumoral pathology or second line patients having had a therapy pause of at least 6 weeks.

DCE-CT of thoracic tumors as an early biomarker for treatment monitoring in comparison with morphologic criteria.

The primary endpoint is to directly correlate the biomarker of the HF analysis software at week 3 (+- 1 week) with the eventually reported Progression-Free Survival (PFS) intervals and Overall Survival (OS) in this study.

The HF biomarker is calculated from DCE perfusion and permeability metrics such as arterial blood flow fraction (alpha), total blood plasma flow (F_p), volume transfer coefficient (K-trans), extracellular volume ratio reflecting vascular permeability (v_e) and plasma volume ratio (v_p). Additionally, semi-quantitative DCE signal metrics, such as signal enhancement and time until contrast agent arrival, may also be taken into account. PFS intervals are determined by the clinician and are based on RECIST1.1 and additional clinical and biochemical progression markers. The focus will be on evaluating the accuracy of the prediction as well as how early the prediction was correct.

The primary endpoint is to directly correlate the biomarker of the HF analysis software at week 8 (+- 3 weeks) with the eventually reported Progression-Free Survival (PFS) intervals and Overall Survival (OS) in this study.

The HF biomarker is calculated from DCE perfusion and permeability metrics such as arterial blood flow fraction (alpha), total blood plasma flow (F_p), volume transfer coefficient (K-trans), extracellular volume ratio reflecting vascular permeability (v_e) and plasma volume ratio (v_p). Additionally, semi-quantitative DCE signal metrics, such as signal enhancement and time until contrast agent arrival, may also be taken into account. PFS intervals are determined by the clinician and are based on RECIST1.1 and additional clinical and biochemical progression markers. The focus will be on evaluating the accuracy of the prediction as well as how early the prediction was correct.

The secondary endpoint is to find an optimal classification system based on changes in DCE perfusion and permeability parameters to classify a treatment response as (PD, SD, PR and CR).

The classification system will be optimized to optimally predict the classification according to PFS and OS. This will be done by splitting the data into a train and test set to ensure generalization.

The classification of the HF analysis software will be compared to the purely morphological classification by RECIST1.1 to identify correlation.

Inclusion Criteria: Patients suffering from primary malignant thoracic tumoral pathology or second line patients having had a therapy pause of at least 6 weeks; at least one tumoral lesion/component should have ≥15mm in diameter. All patients willing to participate and to sign the informed consent. Exclusion Criteria: All patients younger than 18-years-old. Documented allergy for iodine. Neutropenia (absolute White Blood Cell count ≤ 1.5 × 109/l). Thrombopenia (absolute platelet count ≤ 100 × 109/l). Renal insufficiency: serum creatinine ≥ 1.5× the upper limit of normal (ULN); 24-hours creatinine clearance ≤ 50ml/min). Serum bilirubine ≥ 1,5 x ULN, AST ≥ 2,5 x ULN, ALT ≥ 2,5x ULN. Brain metastases

Hwang SH, Yoo MR, Park CH, Jeon TJ, Kim SJ, Kim TH. Dynamic contrast-enhanced CT to assess metabolic response in patients with advanced non-small cell lung cancer and stable disease after chemotherapy or chemoradiotherapy. Eur Radiol. 2013 Jun;23(6):1573-81. doi: 10.1007/s00330-012-2755-0. Epub 2013 Jan 9.

Korn RL, Crowley JJ. Overview: progression-free survival as an endpoint in clinical trials with solid tumors. Clin Cancer Res. 2013 May 15;19(10):2607-12. doi: 10.1158/1078-0432.CCR-12-2934.

Strauch LS, Eriksen RO, Sandgaard M, Kristensen TS, Nielsen MB, Lauridsen CA. Assessing Tumor Response to Treatment in Patients with Lung Cancer Using Dynamic Contrast-Enhanced CT. Diagnostics (Basel). 2016 Jul 21;6(3):28. doi: 10.3390/diagnostics6030028.

Even AJG, Reymen B, La Fontaine MD, Das M, Mottaghy FM, Belderbos JSA, De Ruysscher D, Lambin P, van Elmpt W. Clustering of multi-parametric functional imaging to identify high-risk subvolumes in non-small cell lung cancer. Radiother Oncol. 2017 Dec;125(3):379-384. doi: 10.1016/j.radonc.2017.09.041. Epub 2017 Nov 6.

Qiao PG, Zhang HT, Zhou J, Li M, Ma JL, Tian N, Xing XD, Li GJ. Early evaluation of targeted therapy effectiveness in non-small cell lung cancer by dynamic contrast-enhanced CT. Clin Transl Oncol. 2016 Jan;18(1):47-57. doi: 10.1007/s12094-015-1335-6. Epub 2015 Aug 5.