Neuro-Oncology

Decision Making in Surveillance of High-Grade Gliomas Using Perfusion MRI as Adjunct to Conventional MRI and Artificial Intelligence.

Copyright © 2019 by American Society of Clinical Oncology
Journal of Clinical Oncology. 2019 May;37(15)_suppl doi: 10.1200/JCO.2019.37.15_suppl.2054

Abstract

BACKGROUND:
Surveillance of High-Grade Gliomas (HGGs) remains a major challenge in clinical neurooncology. Histopathological validation is not an option during the course of disease and imaging surveillance suffers from ambiguous features of both disease progression and treatment related changes. This study aimed to differentiate between Pseudoprogression (PsP) and Progressive Disease (PD) using an artificial intelligence (support vector machine – SVM) classification algorithm.
METHODS:
Two groups of patients with histologically proven HGGs were analysed, a group with a single time point DSC perfusion MRI (45 patients) and a group with multiple time point DSC perfusion MRI (19 patients). Both groups included conventional MRI studies prior and after each perfusion MRI. This study design aimed to replicate decision making in clinical practice including multiple previous studies for each patient. SVM training was performed with all available MRI studies for each group and classification was based on different feature datasets from a single or multiple (subtracted features) time points. Classification accuracy comparisons were performed by calculating prediction error rates for different feature datasets and different time point analyses.
RESULTS:
Our results indicate that the addition of multiple time point perfusion MRI combined with structural (conventional with gadolinium-enhanced sequences) MRI features results in optimal classification performance (median error rate: 0.016, lowest value dispersion). Subtracted feature datasets improved classification performance, more prominently when the final and first perfusion studies were included in the analysis. On the contrary, in the single time point group analysis, structural feature-based classification performed best (median error rate: 0.012).
CONCLUSIONS:
Validation of our results with a larger patient cohort may have significant clinical importance in optimising imaging surveillance and clinical decision making for patients with HGG.

Multi-Parametric MRI as Supplement to mRANO Criteria for Response Assessment to MDNA55 in Adults with Recurrent or Progressive Glioblastoma.

Copyright © 2019 by American Society of Clinical Oncology
Journal of Clinical Oncology. 2019 May;37(15)_suppl doi: 10.1200/JCO.2019.37.15_suppl.e13559

Abstract

BACKGROUND:
Modified response assessment in neuro-oncology (mRANO) criteria are widely used in GBM but seem insufficient to capture Pseudoprogression (PsP), which occurs due to extensive inflammatory infiltration, increased vascular permeability, tumor necrosis and edema. mRANO criteria recommend volumetric response evaluation using contrast-enhanced T1 subtraction maps for identifying PsP. Our approach incorporates multi-parametric MRI biomarkers to unravel the true PsP from recurrence or distinguish Pseudo Response (PsR) – following anti-VEGF agents – from delayed (immuno)response.
METHODS:
Multiple time-points MRI (18-24h after convection-enhance delivery of the anti-IL4-R agent MDNA55, then at 30-day intervals) was utilized to determine response. Multi-parametric MRI biomarkers analyzed included (1) 3D-FLAIR-T2-based tumor volume assessment reflecting edema, necrosis and tumor infiltration; (2) 3D-gadolinium-enhanced-based tumor volume estimation reflecting active tumor infiltration, neo-angiogenesis and disrupted blood brain barrier; (3) Dynamic susceptibility contrast-based relative cerebral blood volume (rCBV) measurements for estimation of the vascular tumour properties; and (4) Diffusion weighted imaging – Apparent diffusion coefficient measurements that assess interstitial edema, tumor cellularity and ischemic injury.
RESULTS:
We demonstrate similar imaging phenotypes on conventional FLAIR-T2- and enhanced T1- MR images among different disease states (PsP vs true progression, PsR vs and immuno-response) and describe the perfusion and diffusion MRI biomarkers that improve response staging including PsP masking true progression, PsP masking clinical response, early progression with delayed response, and differentiation between true and PsR. The results are compared with the mRANO-based assessments for concurrence.
CONCLUSIONS:
Incorporating multi-parametric MRI measurements to determine the complex underlying tissue processes enables a better assessment of PsP, PsR and delayed tumour response, and can supplement mRANO-based response assessments in GBM patients undergoing novel immunotherapies.

Multi-Parametric MRI As Supplement to mRANO Criteria for Response Assessment to MDNA55 in Adults with Recurrent or Progressive Glioblastoma.

Copyright © 2019 by American Society of Clinical Oncology
Journal of Clinical Oncology May 2019; 37(15)_suppl DOI: 10.1200/JCO.2019.37.15_suppl.e13559

Abstract

BACKGROUND:
Modified response assessment in neuro-oncology (mRANO) criteria are widely used in GBM but seem insufficient to capture Pseudoprogression (PsP), which occurs due to extensive inflammatory infiltration, increased vascular permeability, tumor necrosis and edema. mRANO criteria recommend volumetric response evaluation using contrast-enhanced T1 subtraction maps for identifying PsP. Our approach incorporates multi-parametric MRI biomarkers to unravel the true PsP from recurrence or distinguish Pseudo Response (PsR) – following anti-VEGF agents – from delayed (immuno)response.
METHODS:
Multiple time-points MRI (18-24h after convection-enhance delivery of the anti-IL4-R agent MDNA55, then at 30-day intervals) was utilized to determine response. Multi-parametric MRI biomarkers analyzed included (1) 3D-FLAIR-T2-based tumor volume assessment reflecting edema, necrosis and tumor infiltration; (2) 3D-gadolinium-enhanced-based tumor volume estimation reflecting active tumor infiltration, neo-angiogenesis and disrupted blood brain barrier; (3) Dynamic susceptibility contrast-based relative cerebral blood volume (rCBV) measurements for estimation of the vascular tumour properties; and (4) Diffusion weighted imaging – Apparent diffusion coefficient measurements that assess interstitial edema, tumor cellularity and ischemic injury.
RESULTS:
We demonstrate similar imaging phenotypes on conventional FLAIR-T2- and enhanced T1- MR images among different disease states (PsP vs true progression, PsR vs and immuno-response) and describe the perfusion and diffusion MRI biomarkers that improve response staging including PsP masking true progression, PsP masking clinical response, early progression with delayed response, and differentiation between true and PsR. The results are compared with the mRANO-based assessments for concurrence.
CONCLUSION:
Incorporating multi-parametric MRI measurements to determine the complex underlying tissue processes enables a better assessment of PsP, PsR and delayed tumour response, and can supplement mRANO-based response assessments in GBM patients undergoing novel immunotherapies.

Decision Making in Surveillance of High-Grade Gliomas Using Perfusion MRI as Adjunct to conventional MRI and Artificial Intelligence.

IAG & UCL poster for the 2019 ASCO Annual Meeting

Abstract

BACKGROUND:
Surveillance of High-Grade Gliomas (HGGs) remains a major challenge in clinical neurooncology. Histopathological validation is not an option during the course of disease and imaging surveillance suffers from ambiguous features of both disease progression and treatment related changes. This study aimed to differentiate between Pseudoprogression (PsP) and Progressive Disease (PD) using an artificial intelligence (support vector machine – SVM) classification algorithm.
METHODS:
Two groups of patients with histologically proven HGGs were analysed, a group with a single time point DSC perfusion MRI (45 patients) and a group with multiple time point DSC perfusion MRI (19 patients). Both groups included conventional MRI studies prior and after each perfusion MRI. This study design aimed to replicate decision making in clinical practice including multiple previous studies for each patient. SVM training was performed with all available MRI studies for each group and classification was based on different feature datasets from a single or multiple (subtracted features) time points. Classification accuracy comparisons were performed by calculating prediction error rates for different feature datasets and different time point analyses.
RESULTS:
Our results indicate that the addition of multiple time point perfusion MRI combined with structural (conventional with gadolinium-enhanced sequences) MRI features results in optimal classification performance (median error rate: 0.016, lowest value dispersion). Subtracted feature datasets improved classification performance, more prominently when the final and first perfusion studies were included in the analysis. On the contrary, in the single time point group analysis, structural feature-based classification performed best (median error rate: 0.012).
CONCLUSION:
Validation of our results with a larger patient cohort may have significant clinical importance in optimising imaging surveillance and clinical decision making for patients with HGG.

TargTex and IAG, Image Analysis Group Announce Partnership

TargTex and IAG, Image Analysis Group Announce Partnership

Lisbon, Portugal and London, UK, 8 February 2023

TargTex to partner with IAG to apply Artificial Intelligence and Advanced Imaging strategies to assess the effects of a clinical candidate in Recurrent Glioblastoma Patients

TargTex SA, a pre-clinical stage biopharmaceutical company focused on developing localized therapies for solid tumors, and IAG, Image Analysis Group, a leading global medical imaging company, are collaborating to apply Artificial Intelligence (AI) and Quantitative imaging to further the development of TargTex investigational product, a hydrogel dispersed drug nanosuspension used as adjunct-to-surgery in a single dose therapy, in patients with glioblastoma multiforme (GBM).

In this collaboration, the parties will utilize AI and advanced quantitative image analysis to identify early treatment changes in GBM patients and development of predictive response markers.

Advanced imaging techniques could play a critical role in response assessment in developing new and innovative cancer therapies.

Multiparametric magnetic resonance imaging (mpMRI) provides quantitative non-invasive imaging markers of early therapy-related changes.

TargTex foundation is also based on an artificial intelligence algorithm created by one of the founding members that can decipher relationships between biological targets and molecules of interest. They identified a new target for a daunting pathology – GBM. Considering GBM’s particularities, a hydrogel to be used as a neo-adjuvant in a single-dose therapy is being developed as a solution.

IAG’s advanced technology will be crucial for the detection of pseudo-progression, which is difficult to distinguish from true disease progression using routine clinical MRI assessment, avoid early patient withdrawal and save costs. The use of novel image analysis methodologies will allow the partners to address complex issues such as pseudo-progression and quantitatively measure the treatment response, thus deploying a truly precision medicine approach to monitor patient outcomes accurately.

IAG has deep expertise in blinded centralized reading and analysis of patient responses in real-time. IAG’s scientific and medical imaging expertise in GBM, coupled with IAG’s proprietary Artificial Intelligence-powered platform DYNAMIKA, will allow TargTex to review efficacy assessments and explore the drug effect in GBM patients thoroughly.

Dr. João Seixas, CEO of TargTex, said:

“In a complex indication such as GBM it is important for us to have IAG’s expertise in advanced imaging and to employ AI tools to differentiate our program and accelerate our development through the use of predictive response markers.”

Dr. Olga Kubassova, CEO of IAG commented,

“It is our pleasure to support TargTex team who have strong foundation in biology and AI. It is our joint long-term objective to bring TargTex clinical candidates for different oncological indications through clinical development to commercialization.”

About TargTex

 TargTex SA is a Portuguese biotech company that emerged from a blended academic research and medical environment. TargTex’s lead program on GBM is based on a machine learning algorithm that deciphered a relationship between an unexplored target for GBM – calcium channel – and a small molecule with known anticancer properties. From a long-term perspective, TargTex aims to develop clinical candidates for different oncological indications based on localized drug delivery, to prevent post-surgical recurrence and metastasis.

About IAG

IAG, Image Analysis Group is a unique partner to life sciences companies, leading AI-powered drug development and precision medicine. IAG leverages expertise in medical imaging and the power of Dynamika™ – our proprietary cloud-based platform, to de-risk clinical development and deliver lifesaving therapies into the hands of patients much sooner.  IAG provides early drug efficacy assessments, smart patient recruitment, and predictive analysis of advanced treatment manifestations, thus lowering investment risk and accelerating study outcomes. IAG bio-partnering takes a broader view of asset development, bringing R&D solutions, operational breadth, and radiological expertise via risk-sharing financing and partnering models.

Diagnostic yield of FDG PET/CT, MRI, and CSF cytology in nonbiopsiable Neurolymphomatosis as a heralding feature of Diffuse B-cell Lymphoma recurrence.

Neurolymphomatosis (NL) is a rare condition associated with lymphomas in which various structures of the nervous system are infiltrated by malignant lymphocytes. Rarely, it may be the presenting feature of recurrence of lymphoma otherwise deemed to be in remission. It is crucial, as is the case with all types of nodal or visceral involvement of lymphoma, to identify the disease early and initiate treatment with chemotherapy and/or radiation therapy. Positron emission tomography-computed tomography (PET-CT) has been shown to be a sensitive modality for staging, restaging, biopsy guidance, therapy response assessment, and surveillance for recurrence of lymphoma. Magnetic resonance imaging (MRI) is another useful imaging modality, which, along with PET/CT, compliment cerebrospinal spinal fluid (CSF) cytology and electromyography (EMG) in the diagnosis of NL. Performing nerve biopsies to confirm neurolymphomatosis can be challenging and with associated morbidity. The case presented herein illustrates the practical usefulness of these tests in detecting NL as a heralding feature of lymphoma recurrence, especially in the absence of histopathologic correlation.

Quantitative Imaging Analysis of FDG PET/CT Imaging for Detection of Central Neurolymphomatosis in a Case of Recurrent Diffuse B-Cell Lymphoma

Neurolymphomatosis (NL) is a rare disease characterized by malignant lymphocytes infiltrating various structures of the nervous system. It typically manifests as a neuropathy involving the peripheral nerves, nerve roots, plexuses, or cranial nerves. It often presents as a complication of lymphoma, but it can be the presenting feature of recurrent lymphoma. It is essential to identify and initiate treatment early with chemotherapy and/or radiation therapy in all cases of nodal or visceral (including neural) involvement with lymphoma. There are various diagnostic tests that can be used for its detection, such as cerebrospinal spinal fluid (CSF) cytology, electromyography (EMG), magnetic resonance imaging (MRI), and positron-emission tomography/computed tomography (PET/CT). FDG-PET/CT is the standard of care in lymphoma staging, restaging, and therapy response assessment, but has an inherent limitation in the detection of disease involvement in the central nervous system. While that is mostly true for visual assessment, there are quantitative methods to measure variation in the metabolic activity in the brain, which in turn helps detect the occurrence of neurolymphomatosis.