Differentiating Tumor Progression Research Articles

The Impact of MRI-Based Advanced Neuroimaging on Neurooncologists' Clinical Decision-Making in Patients With Posttreatment High-Grade Glioma: A Prospective Survey-Based Study.

Background: Advanced MRI-based neuroimaging techniques, such as perfusion and spectroscopy, have been increasingly incorporated into routine follow-up protocols in patients treated for high-grade glioma (HGG), to help differentiate tumor progression from treatment effect. However, these techniques' influence on clinical management remains poorly understood. Objective: To evaluate the impact of MRI-based advanced neuroimaging on clinical decision-making in patients with HGG in the posttreatment setting. Methods: This prospective study, performed at a comprehensive cancer center from March 1, 2017, to October 31, 2020, included adult patients treated by chemoradiation for WHO grade 4 diffuse glioma who underwent MRIbased advanced neuroimaging (comprising multiple perfusion imaging sequences and spectroscopy) to further evaluate findings on conventional MRI equivocal for tumor progression versus treatment effect. The ordering neuro-oncologists completed surveys before and after each advanced neuroimaging session. The percent of care episodes with a change between the intended and actual management plan on the surveys conducted before and after advanced neuroimaging, respectively, was computed and compared with a previously published percent using the Wald test for independent samples proportions. Results: The study included 63 patients (mean age, 55±13 years; 36 women, 27 men) who underwent 70 advanced neuroimaging sessions. Ordering neuro-oncologists' intended and actual management plans on the surveys completed before and after advanced neuroimaging, respectively, differed in 44% (31/70, [95% CI: 33-56%]) of episodes, which differed from the previously published frequency of 8.5% (5/59) (p<.001). These management plan changes included selection of a different plan for 6/8 episodes with an intended plan to enroll patients in a clinical trial, 12/19 episodes with an intended plan to change chemotherapeutic agents, 4/8 episodes with an intended plan of surgical intervention, and 1/2 episodes with an intended plan of re-irradiation. The ordering neuro-oncologists found advanced neuroimaging to be helpful in 93% (95% CI: 87%-99%) (65/70) of episodes. Conclusion: Neuro-oncologists' management plans changed in a substantial fraction of adult patients with HGG who underwent advanced neuroimaging to further evaluate conventional MRI findings equivocal for tumor progression versus treatment effect. Clinical Impact: The findings support incorporation of advanced neuroimaging into HGG posttreatment monitoring protocols.

Fractional tumor burden maps increase the confidence of reading brain MR perfusion

Rationale and objectivesVarious methods exist to perform and post-process perfusion weighted MR imaging in the post-treatment imaging of glioma patients to differentiate tumor progression from tumor-related abnormalities. One of these post-processing methods produces ‘fractional tumor burden’ maps. This multi-reader study investigated the clinical feasibility of fractional tumor burden maps on real world data from radiological follow-up of high-grade astrocytoma patients. MethodsFive readers with background in radiology and varying levels of experience were tasked with assessing 30 astrocytoma and glioblastoma patients during one reader session. First, they were provided with a dataset of conventional MRI sequences, including perfusion MRI with regional cerebral blood volume maps. Then the dataset was expanded with a corresponding fractional tumor burden maps. Diagnostic accuracy, duration of post-processing, duration of the assessment of the fractional tumor burden maps, the diagnostic confidence reported by the readers and their diagnoses were recorded. Final diagnosis was determined by clinical and radiological follow-up and/or histopathological data used as gold standard. ResultsA mean sensitivity of 83.3 % and mean specificity of 55.1 % was obtained without the use of fractional tumor burden maps, whereas their additional of fractional tumor burden maps resulted in a mean sensitivity and specificity of 79.5 % and 54.2 %, respectively. Diagnostic accuracies with and without fractional tumor burden maps were not significantly different (Z = 0.76, p = 0.450). The median time spent post-processing was 313 s, while the median duration of the assessment of the FTB maps was 19 s. Interestingly, reader confidence increased significantly after adding the fractional tumor burden-maps to the assessment (Z = 454, p < 0.01). ConclusionsWhile the use of fractional tumor burden maps does not carry additional value in the radiological follow-up of post-operative high-grade astrocytoma and glioblastoma patients, it does give readers more confidence in their diagnosis.

Influence of arterial transit time delays on the differentiation between tumor progression and pseudoprogression in glioblastoma by arterial spin labeling magnetic resonance imaging.

Arterial spin labeling (ASL) and dynamic susceptibility contrast (DSC) magnetic resonance imaging (MRI) have shown potential for differentiating tumor progression from pseudoprogression. For pseudocontinuous ASL with a single postlabeling delay, the presence of delayed arterial transit times (ATTs) could affect the evaluation of ASL-MRI perfusion data. In this study, the influence of ATT artifacts on the perfusion assessment and differentiation between tumor progression and pseudoprogression were studied. This study comprised 66 adult patients (mean age 60 ± 13 years; 40 males) with a histologically confirmed glioblastoma who received postoperative radio (chemo)therapy. ASL-MRI and DSC-MRI scans were acquired at 3months postradiotherapy as part of the standard clinical routine. These scans were visually scored regarding (i) the severity of ATT artifacts (%) on the ASL-MRI scans only, scored by two neuroradiologists; (ii) perfusion of the enhancing tumor lesion; and (iii) radiological evaluation of tumor progression versus pseudoprogression by one neuroradiologist. The final outcome was based on combined clinical and radiological follow-up until 9months postradiotherapy. ATT artifacts were identified in all patients based on the mean scores of two raters. A significant difference between the radiological evaluation of ASL-MRI and DSC-MRI was observed only for ASL images with moderate ATT severity (30%-65%). The perfusion assessment showed ASL-MRI tending more towards hyperperfusion than DSC-MRI in the case of moderate ATT artifacts. In addition, there was a significant difference between the prediction of tumor progression with ASL-MRI and the final outcome in the case of severe ATT artifacts (McNemar test, p= 0.041). Despite using ASL imaging parameters close to the recommended settings, ATT artifacts frequently occur in patients with treated brain tumors. Those artifacts could hinder the radiological evaluation of ASL-MRI data and the detection of true disease progression, potentially affecting treatment decisions for patients with glioblastoma.

Prospective Randomized Phase 2 Trial of Hypofractionated Stereotactic Radiation Therapy of 25 Gy in 5 Fractions Compared With 35 Gy in 5 Fractions in the Reirradiation of Recurrent Glioblastoma

The aim of this work was to investigate whether reirradiation of recurrent glioblastoma with hypofractionated stereotactic radiation therapy (HSRT) consisting of 35 Gy in 5 fractions (35 Gy/5 fx) compared with 25 Gy in 5 fractions (25 Gy/5 fx) improves outcomes while maintaining acceptable toxicity. We conducted a prospective randomized phase 2 trial involving patients with recurrent glioblastoma (per the 2007 and 2016 World Health Organization classification). A minimum interval from first radiation therapy of 5 months and gross tumor volume of 150 cc were required. Patients were randomized 1:1 to receive HSRT alone in 25 Gy/5 fx or 35 Gy/5 fx. The primary endpoint was progression-free survival (PFS). We used a randomized phase 2 screening design with a 2-sided α of 0.15 for the primary endpoint. From 2011 to 2019, 40 patients were randomized and received HSRT, with 20 patients in each group. The median age was 50 years (range, 27-71); a new resection before HSRT was performed in 75% of patients. The median PFS was 4.9 months in the 25 Gy/5 fx group and 5.2 months in the 35 Gy/5 fx group (P = .23). Six-month PFS was similar at 40% (85% CI, 24%-55%) for both groups. The median overall survival (OS) was 9.2 months in the 25 Gy/5 fx group and 10 months in the 35 Gy/5 fx group (P = .201). Grade ≥3 necrosis was numerically higher in the 35 Gy/5 fx group (3 [16%] vs 1 [5%]), but the difference was not statistically significant (P = .267). In an exploratory analysis, median OS of patients who developed treatment-related necrosis was 14.1 months, and that of patients who did not was 8.7 months (P = .003). HSRT alone with 35 Gy/5 fx was not superior to 25 Gy/5 fx in terms of PFS or OS. Due to a potential increase in the rate of clinically meaningful treatment-related necrosis, we suggest 25 Gy/5 fx as the standard dose in HSRT alone. During follow-up, attention should be given to differentiating tumor progression from potentially manageable complications.

Low rCBV values in glioblastoma tumor progression under chemoradiotherapy.

After standard treatment for glioblastoma, perfusion MRI remains challenging for differentiating tumor progression from post-treatment changes. Our objectives were (1) to correlate rCBV values at diagnosis and at first tumor progression and (2) to analyze the relationship of rCBV values at tumor recurrence with enhancing volume, localization of tumor progression, and time elapsed since the end of radiotherapy in tumor recurrence. Inclusion criteria were (1) age > 18years, (2) histologically confirmed glioblastoma treated with STUPP regimen, and (3) tumor progression according to RANO criteria > 12weeks after radiotherapy. Co-registration of segmented enhancing tumor VOIs with dynamic susceptibility contrast perfusion MRI was performed using Olea Sphere software. For tumor recurrence, we correlated rCBV values with enhancing tumor volume, with recurrence localization, and with time elapsed from the end of radiotherapy to progression. Analyses were performed with SPSS software. Sixty-four patients with glioblastoma were included in the study. Changes in rCBV values between diagnosis and first tumor progression were significant (p < 0.001), with a mean and median decreases of 32% and 46%, respectively. Mean rCBV values were also different (p < 0.01) when tumors progressed distally (radiation field rCBV values of 1.679 versus 3.409 distally). However, changes and, therefore, low rCBV values after radiotherapy in tumor recurrence were independent of time. Chemoradiation alters tumor perfusion and rCBV values may be decreased in the setting of tumor progression. Changes in rCBV values with respect to diagnosis, with low rCBV in tumor progression, are independent of time but related to the site of recurrence.

IMMU-05. CD8 IMMUNOPET IMAGING IS ABLE TO MONITOR THE IMMUNE RESPONSE SENSITIVELY AND SPECIFICALLY IN GBM PATIENTS FOLLOWING NEOADJUVANT CHECKPOINT BLOCKADE IMMUNOTHERAPY

Abstract The sensitive and specific evaluation of treatment response in glioblastoma (GBM) patients following immunotherapy is a clinical dilemma. Traditional MRI assessments to visualize and quantify intracranial tumor burden following immunotherapy cannot reliably differentiate tumor progression and pseudo progression remains a complicating factor. In this study, we provide an early report on the use of an investigational new functional imaging PET tracer (CD8 ImmunoPET, 89Zr-crefmirlimab berdoxam, ImaginAb) prior to and immediately following neoadjuvant checkpoint blockade immunotherapy in five patients with recurrent GBM. Tumor samples were collected from each patient at surgery following a two-week neoadjuvant checkpoint blockade to assess gene expression changes and TIL density. We then evaluated how such tumor measurements correlated with PET metrics obtained at baseline and within 2 weeks after treatment initiation which included standardized uptake values (SUVmax, SUVpeak, and SUVmean), and two proposed new metrics indicating the Volume of Immunometabolism (VIM), and the Total Lesion or Lymph nodes’ T cells called TLTc (TLTc = VI x SUVmean) in the tumor bed and lymph node basins (neck, axillae, and inguinal regions). CD8 ImmunoPET demonstrated tracer uptake in the intracranial tumor at both time points, as well as specific physiologic uptake in lymph node basins, spleen, liver, and bone marrow. Neoadjuvant checkpoint blockade with PD-1 +/- CTLA-4 mAb (Nivolumab, Ipilimumab, BMS) at 2 weeks induced variable changes in SUV, but in four of the five initial patients studied in this clinical trial, an increase in VI and TLTc was observed within the tumor bed. Tracer uptake was associated with elevated T cell infiltration into the tumor bed and gene expression changes indicative of interferon signaling and immune cells in these same patients. Our early findings identify a noninvasive functional imaging platform to evaluate the immunometabolism secondary to T cell infiltration and/or activation changes following immunotherapy in GBM patients.

Radiomics: The New Promise for Differentiating Progression, Recurrence, Pseudoprogression, and Radionecrosis in Glioma and Glioblastoma Multiforme.

Glioma and glioblastoma multiform (GBM) remain among the most debilitating and life-threatening brain tumors. Despite advances in diagnosing approaches, patient follow-up after treatment (surgery and chemoradiation) is still challenging for differentiation between tumor progression/recurrence, pseudoprogression, and radionecrosis. Radiomics emerges as a promising tool in initial diagnosis, grading, and survival prediction in patients with glioma and can help differentiate these post-treatment scenarios. Preliminary published studies are promising about the role of radiomics in post-treatment glioma/GBM. However, this field faces significant challenges, including a lack of evidence-based solid data, scattering publication, heterogeneity of studies, and small sample sizes. The present review explores radiomics's capabilities in following patients with glioma/GBM status post-treatment and to differentiate tumor progression, recurrence, pseudoprogression, and radionecrosis.

Comparison of clinically available dynamic susceptibility contrast post processing software to differentiate progression from pseudoprogression in post-treatment high grade glioma

IntroductionThe purpose of this retrospective study was to compare two, widely available software packages for calculation of Dynamic Susceptibility Contrast (DSC) perfusion MRI normalized relative Cerebral Blood Volume (rCBV) values to differentiate tumor progression from pseudoprogression in treated high-grade glioma patients. Material and MethodsrCBV maps processed by Siemens Syngo.via (Siemens Healthineers) and Olea Sphere (Olea Medical) software packages were co-registered to contrast-enhanced T1 (T1-CE). Regions of interest based on T1-CE were transferred to the rCBV maps. rCBV was calculated using mean values and normalized using contralateral normal- appearing white matter. The Wilcoxon test was performed to assess for significant differences, and software-specific optimal rCBV cutoff values were determined using the Youden index. Interrater reliability was evaluated for two raters using the intraclass correlation coefficient. Results41 patients (18 females; median age = 59 years; range 21–77 years) with 49 new or size-increasing post-treatment contrast-enhancing lesions were included (tumor progression = 40 lesions; pseudoprogression = 9 lesions). Optimal rCBV cutoffs of 1.31 (Syngo.via) and 2.40 (Olea) were significantly different, with an AUC of 0.74 and 0.78, respectively. Interrater reliability was 0.85. DiscussionWe demonstrate that different clinically available MRI DSC-perfusion software packages generate significantly different rCBV cutoff values for the differentiation of tumor progression from pseudoprogression in standard-of-care treated high grade gliomas. Physicians may want to determine the unique value of their perfusion software packages on an institutional level in order to maximize diagnostic accuracy when faced with this clinical challenge. Furthermore, combined with implementation of current DSC-perfusion recommendations, multi-center comparability will be improved.

Liquid biopsy and glioblastoma.

Glioblastoma is the most common and malignant primary brain tumor. Despite a century of research efforts, the survival of patients has not significantly improved. Currently, diagnosis is based on neuroimaging techniques followed by histopathological and molecular analysis of resected or biopsied tissue. A recent paradigm shift in diagnostics ranks the molecular analysis of tissue samples as the new gold standard over classical histopathology, thus correlating better with the biological behavior of glioblastoma and clinical prediction, especially when a tumor lacks the typical hallmarks for glioblastoma. Liquid biopsy aims to detect and quantify tumor-derived content, such as nucleic acids (DNA/RNA), circulating tumor cells (CTCs), or extracellular vesicles (EVs) in biofluids, mainly blood, cerebrospinal fluid (CSF), or urine. Liquid biopsy has the potential to overcome the limitations of both neuroimaging and tissue-based methods to identify early recurrence and to differentiate tumor progression from pseudoprogression, without the risks of repeated surgical biopsies. This review highlights the origins and time-frame of liquid biopsy in glioblastoma and points to recent developments, limitations, and challenges of adding liquid biopsy to support the clinical management of glioblastoma patients.

Imaging of Brain Tumors.

This article focuses on neuroimaging as an essential tool for diagnosing brain tumors and monitoring response to treatment. Neuroimaging is useful at all stages of brain tumor care. Technologic advances have improved the clinical diagnostic capability of neuroimaging as a vital complement to history, examination, and pathologic assessment. Presurgical evaluations are enriched by novel imaging techniques, through improved differential diagnosis and better surgical planning using functional MRI (fMRI) and diffusion tensor imaging. The common clinical challenge of differentiating tumor progression from treatment-related inflammatory change is aided by novel uses of perfusion imaging, susceptibility-weighted imaging (SWI), spectroscopy, and new positron emission tomography (PET) tracers. Using the most up-to-date imaging techniques will facilitate high-quality clinical practice in the care of patients with brain tumors.

NIMG-17. A DEEP LEARNING MODEL FOR DISCRIMINATING TRUE PROGRESSION FROM PSEUDOPROGRESSION IN GLIOBLASTOMA PATIENTS

Abstract INTRODUCTION Glioblastomas (GBMs) are highly aggressive tumors. A common clinical challenge after standard of care treatment is differentiating tumor progression from treatment-related changes, also known as pseudoprogression (PsP). Usually, PsP resolves or stabilizes without further treatment or a course of steroids, whereas true progression (TP) requires more aggressive management. Differentiating PsP from TP will affect the patient's outcome. This study investigated using deep learning to distinguish PsP MRI features from progressive disease. METHOD We included newly diagnosed GBM patients who met the inclusion criteria, including a new or increasing enhancing lesion size within the original radiation field and who had clinical, medication, and any histopathology available. The interpretation of the MR images at the image inclusion time point had to be indeterminate. We labeled those who subsequently were stable or improved on imaging and clinically as PSP and those with clinical and imaging deterioration as TP. A subset of subjects underwent a second resection. We labeled these subjects as PSP or TP based on the histological diagnosis. We performed skull stripping, coregistered contrast-enhanced T1 MRIs with T2-weighted images for each patient and used them as input to a 3-D Densenet121 model. We used several augmentation techniques and five-fold cross-validation to achieve more robust predictions. RESULT We included 124 patients who met the criteria, and of those, 63 were PsP and 61 were TP. We trained a deep learning model that achieved 76.4% (ranged 70%-84%, SD 5.122) mean accuracy over the 5 folds, 0.7560(ranged 0.6553-0.8535, SD 0.069) mean AUROCC, 88.72% (SD 6.86) mean sensitivity, and 62.05%(SD 9.11) mean specificity. CONCLUSION We report the development of a deep learning model that distinguishes PsP from TP in GBM patients treated per the Stupp protocol. Further refinement and external validation are required prior to widespread adoption in clinical practice.

Diagnostic accuracy of MRI techniques for treatment response evaluation in patients with brain metastasis: A systematic review and meta-analysis

BackgroundTreatment response assessment in patients with brain metastasis uses contrast enhanced T1-weighted MRI. Advanced MRI techniques have been studied, but the diagnostic accuracy is not well known. Therefore, we performed a metaanalysis to assess the diagnostic accuracy of the currently available MRI techniques for treatment response. MethodsA systematic literature search was done. Study selection and data extraction were done by two authors independently. Meta-analysis was performed using a bivariate random effects model. An independent cohort was used for DSC perfusion external validation of diagnostic accuracy. ResultsAnatomical MRI (16 studies, 726 lesions) showed a pooled sensitivity of 79% and a specificity of 76%. DCE perfusion (4 studies, 114 lesions) showed a pooled sensitivity of 74% and a specificity of 92%. DSC perfusion (12 studies, 418 lesions) showed a pooled sensitivity was 83% with a specificity of 78%. Diffusion weighted imaging (7 studies, 288 lesions) showed a pooled sensitivity of 67% and a specificity of 79%. MRS (4 studies, 54 lesions) showed a pooled sensitivity of 80% and a specificity of 78%. Combined techniques (6 studies, 375 lesions) showed a pooled sensitivity of 84% and a specificity of 88%. External validation of DSC showed a lower sensitivity and a higher specificity for the reported cut-off values included in this metaanalysis. ConclusionA combination of techniques shows the highest diagnostic accuracy differentiating tumor progression from treatment induced abnormalities. External validation of imaging results is important to better define the reliability of imaging results with the different techniques.

Detecting monocyte trafficking in an animal model of glioblastoma using R2* and quantitative susceptibility mapping.

The role of tumor-associated macrophages (TAMs) in glioblastoma (GBM) disease progression has received increasing attention. Recent advances have shown that TAMs can be re-programmed to exert a pro-inflammatory, anti-tumor effect to control GBMs. However, imaging methods capable of differentiating tumor progression from immunotherapy treatment effects have been lacking, making timely assessment of treatment response difficult. We showed that tracking monocytes using iron oxide nanoparticle (USPIO) with MRI can be a sensitive imaging method to detect therapy response directed at the innate immune system. We implanted syngeneic mouse glioma stem cells into C57/BL6 mice and treated the animals with either niacin (a stimulator of innate immunity) or vehicle. Animals were imaged using an anatomical MRI sequence, R2* mapping, and quantitative susceptibility mapping (QSM) before and after USPIO injection. Compared to vehicles, niacin-treated animals showed significantly higher susceptibility and R2*, representing USPIO and monocyte infiltration into the tumor. We observed a significant reduction in tumor size in the niacin-treated group 7days later. We validated our MRI results with flow cytometry and immunofluoresence, which showed that niacin decreased pro-inflammatory Ly6C high monocytes in the blood but increased CD16/32 pro-inflammatory macrophages within the tumor, consistent with migration of these pro-inflammatory innate immune cells from the blood to the tumor. MRI with USPIO injection can detect therapeutic responses of innate immune stimulating agents before changes in tumor size have occurred, providing a potential complementary imaging technique to monitor cancer immunotherapies. We show that iron oxide nanoparticles (USPIOs) can be used to label innate immune cells and detect the trafficking of pro-inflammatory monocytes into the glioblastoma. This preceded changes in tumor size, making it a more sensitive imaging technique.

P15.06.B Influence of arterial transit time delays on the differentiation between tumor progression and pseudo-progression in glioblastoma by arterial spin labeling MRI

Abstract Background Perfusion MRI by Arterial Spin Labeling (ASL) and Dynamic Susceptibility Contrast (DSC) has shown its potential for differentiating tumor progression from pseudo-progression in glioblastoma patients. The ASL scans can be affected by arterial transit time (ATT) delays, which could be caused by treatment effects due to concomitant radiochemotherapy. A prolonged ATT is present as apparent signal increase in the large arteries due to labeled spins still residing within the vasculature, leading to underestimation of tissue perfusion and thus potentially affecting clinical decision-making. The research questions were: 1) Do delayed ATTs lead to a difference in the visual assessment of ASL perfusion (normal/increased) maps compared to DSC-MRI?; 2) Does the radiological evaluation (progression vs. pseudo-progression) of ASL and DSC perfusion maps differ when ATT artifacts are present?; 3) Do delayed ATTs affect the predictive value of ASL-MRI scans 3 months post-radiotherapy for detecting true disease progression? Material and Methods This retrospective, single-center study included 68 adult patients with histologically confirmed glioblastoma who received postoperative radio(chemo)therapy. ASL and DSC scans were acquired 3 months post-radiotherapy as part of routine clinical follow-up. The perfusion data were visually scored by a neuroradiologist who determined presence/absence of ATT artifacts and their severity (%), perfusion of the enhancing tumor lesion and the radiological evaluation of tumor progression versus pseudo-progression. Presence of true disease progression was determined by follow-up of clinical data until 9 months post-radiotherapy available for 49/68 patients. Logistic regression was performed with gender, age, treatment type and tumor genetic status as covariates to assess the predictive value of ASL. Results In 78% of the patients ATT artifacts were present. No statistically significant association between the agreement of the perfusion maps and presence of ATT artifacts was found, but presence of ATT artifacts lowered the agreement between the DSC and ASL radiological evaluation. The logistic regression analysis showed that the ASL-based radiological score could not predict true disease progression, whereas higher age and unmethylated MGMT gene were associated with progression. Presence of ATT artifacts was not associated with tumor progression. Conclusion The presence of delayed ATT in ASL data seems to impact the radiological evaluation of ASL data, steering interpretation towards tumor progression (as compared to the DSC evaluation), whereas in patients without ATT artifacts ASL and DSC provide more similar radiological scores. Therefore, it is highly recommended to consider these artifacts when interpreting ASL perfusion MRI to differentiate between tumor progression and pseudo-progression in glioblastoma patients.

A deep learning model for discriminating true progression from pseudoprogression in glioblastoma patients.

Glioblastomas (GBMs) are highly aggressive tumors. A common clinical challenge after standard of care treatment is differentiating tumor progression from treatment-related changes, also known as pseudoprogression (PsP). Usually, PsP resolves or stabilizes without further treatment or a course of steroids, whereas true progression (TP) requires more aggressive management. Differentiating PsP from TP will affect the patient's outcome. This study investigated using deep learning to distinguish PsP MRI features from progressive disease. We included GBM patients with a new or increasingly enhancing lesion within the original radiation field. We labeled those who subsequently were stable or improved on imaging and clinically as PsP and those with clinical and imaging deterioration as TP. A subset of subjects underwent a second resection. We labeled these subjects as PsP, or TP based on the histological diagnosis. We coregistered contrast-enhanced T1 MRIs with T2-weighted images for each patient and used them as input to a 3-D Densenet121 model and using five-fold cross-validation to predict TP vs PsP. We included 124 patients who met the criteria, and of those, 63 were PsP and 61 were TP. We trained a deep learning model that achieved 76.4% (range 70-84%, SD 5.122) mean accuracy over the 5 folds, 0.7560 (range 0.6553-0.8535, SD 0.069) mean AUROCC, 88.72% (SD 6.86) mean sensitivity, and 62.05% (SD 9.11) mean specificity. We report the development of a deep learning model that distinguishes PsP from TP in GBM patients treated per the Stupp protocol. Further refinement and external validation are required prior to widespread adoption in clinical practice.

DSC Perfusion MRI-Derived Fractional Tumor Burden and Relative CBV Differentiate Tumor Progression and Radiation Necrosis in Brain Metastases Treated with Stereotactic Radiosurgery.

Differentiation between tumor and radiation necrosis in patients with brain metastases treated with stereotactic radiosurgery is challenging. We hypothesized that MR perfusion and metabolic metrics can differentiate radiation necrosis from progressive tumor in this setting. We retrospectively evaluated MRIs comprising DSC, dynamic contrast-enhanced, and arterial spin-labeling perfusion imaging in subjects with brain metastases previously treated with stereotactic radiosurgery. For each lesion, we obtained the mean normalized and standardized relative CBV and fractional tumor burden, volume transfer constant, and normalized maximum CBF, as well as the maximum standardized uptake value in a subset of subjects who underwent FDG-PET. Relative CBV thresholds of 1 and 1.75 were used to define low and high fractional tumor burden. Thirty subjects with 37 lesions (20 radiation necrosis, 17 tumor) were included. Compared with radiation necrosis, tumor had increased mean normalized and standardized relative CBV (P = .002) and high fractional tumor burden (normalized, P = .005; standardized, P = .003) and decreased low fractional tumor burden (normalized, P = .03; standardized, P = .01). The area under the curve showed that relative CBV (normalized = 0.80; standardized = 0.79) and high fractional tumor burden (normalized = 0.77; standardized = 0.78) performed the best to discriminate tumor and radiation necrosis. For tumor prediction, the normalized relative CBV cutoff of ≥1.75 yielded a sensitivity of 76.5% and specificity of 70.0%, while the standardized cutoff of ≥1.75 yielded a sensitivity of 41.2% and specificity of 95.0%. No significance was found with the volume transfer constant, normalized CBF, and standardized uptake value. Increased relative CBV and high fractional tumor burden (defined by a threshold relative CBV of ≥1.75) best differentiated tumor from radiation necrosis in subjects with brain metastases treated with stereotactic radiosurgery. Performance of normalized and standardized approaches was similar.

A systematic review of amino acid PET in assessing treatment response to temozolomide in glioma.

The response assessment in neuro-oncology (RANO) criteria have been the gold standard for monitoring treatment response in glioblastoma (GBM) and differentiating tumor progression from pseudoprogression. While the RANO criteria have played a key role in detecting early tumor progression, their ability to identify pseudoprogression is limited by post-treatment damage to the blood-brain barrier (BBB), which often leads to contrast enhancement on MRI and correlates poorly to tumor status. Amino acid positron emission tomography (AA PET) is a rapidly growing imaging modality in neuro-oncology. While contrast-enhanced MRI relies on leaky vascularity or a compromised BBB for delivery of contrast agents, amino acid tracers can cross the BBB, making AA PET particularly well-suited for monitoring treatment response and diagnosing pseudoprogression. The authors performed a systematic review of PubMed, MEDLINE, and Embase through December 2021 with the search terms “temozolomide” OR “Temodar,” “glioma” OR “glioblastoma,” “PET,” and “amino acid.” There were 19 studies meeting inclusion criteria. Thirteen studies utilized [18F]FET, five utilized [11C]MET, and one utilized both. All studies used static AA PET parameters to evaluate TMZ treatment in glioma patients, with nine using dynamic tracer parameters in addition. Throughout these studies, AA PET demonstrated utility in TMZ treatment monitoring and predicting patient survival.

Sarcoid Like Reactıon Mimicking Disease Progression Associated With Nivolumab in A Case Wıth Malignant Mesothelıoma: Is There a Solution For This Dilemma

Sarcoid like reaction (SLR) is a well defined entity and SLR has been reported in cases treated by checkpoint inhibitors (CPIs) (1). So far the majority of the cases reported associated with CPIs are melanoma; here we reported a case with malignant mesothelioma (MM) tretaed by nivolumab and developing granulomatous reaction mimicking progressive disease. Case report 53 year old woman admitted with pleural MM epitheloid type. After surgery, pemetrexate-cisplatinum and bevacizumab had been given for 6 cycles and bevacizumab maintenance for 10 cycles. Vinorelbine was given for 4 cycles for progressive disease; there was partial response however at the end of 8 cycles PET/CT showed considerable progression and carboplatin-gemcitabine combination was given and at the end there was metabolic nodules at left lung, pleural thickening and soft tissue masses at PET/CT (Figure 1). Due to active tumor and severe pains nivolumab was given as 6 cycles, her general condition was excellent and she did not require analgesics. However PET/CT showed considerable progression (Figure 2). Her condition was very well and post-immunotherapy granulomatous inflammation was thought clinically. Biopsy was done from a pelvic lymph node and was reported as MM metastasis. She had no other treatment choice and salvage surgery was done: omentectomy, paraaortic and coeliac lymphadenectomy, right (6/11) and left pelvic lymphadenectomy, diaphragma resection, pulmonary parencyhmal resection, pleurectomy, pericardiectomy were performed. All of the tissue samples except 6 of 17 right pelvic lymph nodes and paraaortic lymph nodes were reported as granulomatous reaction. Figures 3 and 4 show granulomas Discussion: CPIs have various adverse events but SLR is relatively little known reaction in cases treated by CPIs. SLRs may mimic disease progression and may cause to discontinuation of the useful therapies (1, 2). It is well known that CPI related adverse effects are generally associated with favorable clinical outcome and lastly it has been suggested, at least in some reports, that SLR is associated with better outcome. Also sarcoid like reaction associated with CPIs have been found as better response to therapy (3). There is no specific recommendation for patients developing SLR during CPI treatment: asymptomatic patients do not reqire treatment and symptomatic cases respond to steroid. If there is no response to steroid, other immunosuppressives may be used (3, 4). In our case there was clear evidence of progression at PET/CT but her clinical condition was excellent. At this point granulomatous reaction was thought but biopsy showed MM and brutal surgery was done. However the majority of the surgical specimens showed granuolmatous reaction while only paraaortic and one third of right pelvic lymph nodes showed tumor. Clinicians should be aware of SLRs in cases treated by CPIs and tissues must be sampled and reviewed by an experienced pathologist to avoid misdiagnosis and drug stops while responding to tumor (1). In clinical practice FDG PET/CT is the most frequently used imaging to measure the response to CPIs but is not sensitive enough to predict atypical immune related adverse events including pseudoprogression and SLR. Newer imaging modalities including 18F-fluorothymidine (FLT) PET imaging may be useful. It is known that FLT is a substrate for thymidine kinase which is transported into the cell during DNA synthesis and trapped, but not incorporated into the DNA, it has been hypothesized that FLT PET is an important tool to show the proliferative activity of the tumor and has advantage to detect to differentiate tumor progression from pseudoprogression associated with tumor infiltrating immune cells which are low proliferative capacity than tumor cells (5). By analogy it can be proposed that FLT/PET may be used in cases with good clinical outcome in spite of progression detected by FDG/PET. The clinical outcome of this patient is very interesting: 1-A case with malignant mesothelioma responded CPI treatment at fourth line setting, 2-PET CT imaging showed progressive findings due to granulomatous reaction, 3-It can be said that SLR is not a rare event in cases treated with CPIs and awareness of these reactions will save the patients from unnecessary procedures and allows the patients to continue useful treatment. Newer PET/CT imaging modalities may be useful to differentiate SLR from progressive disease.

Dual-time point brain FDG PET to differentiate tumor progression from radionecrosis after stereotactic radiotherapy for brain metastases

Dual phase 18[F]-FDG brain PET (dual-PET) is useful to distinguish tumor recurrence (TR) from radionecrosis (RN) after stereotaxic radiosurgery (SRS) of brain metastases, when contrast-enhanced MRI is inconclusive. We aimed to compare six different visual and quantitative interpretation criteria to enhance diagnostic performances. In this retrospective french multicentric study (Clermont-Ferrand, Montpellier, Rennes), we evaluated 45 patients previously treated with SRS for BM, addressed for a dual-PET by the local neuro-oncological committees for an evolving lesion on MRI inconclusive between TR and RN, at least 3 months after the last SRS session. Dual-PET included both an “early” and a “delayed” acquisition, respectively 30 to 60 minutes and 4 to 5 hours after 18[F]-FDG injection. After measuring SUVmax values of both lesion (L) and mirror contralateral grey matter (GM) at early (1) and delayed (2) acquisitions, three quantitative metrics were calculated: ratios of L SUVmax to GM SUVmax at “early” (L1/GM1) and “delayed” (L2/GM2) acquisitions and the retention index which is the variation over time of the standardized SUVmax ratio (RI = [(Delayed-Early)/Early]). Visual analysis was also conducted using a subjective 6 points visual scale of the lesion uptake at both acquisitions. The five interpretation criteria were compared according to their area under the ROC curve (AUC), and to their diagnostic accuracy applying the best cut off value (maximizing the Youden's index) by the Cochran Q test. The final diagnosis was based on pathology, or by default radiological and clinical follow-up criteria after at least 6 months. Final diagnoses were TR for 24 patients and RN for 21 patients. There was no statistically significant difference regarding AUC between the different interpretation methods. AUC ranged from 0,78 [95 %CI 0,63; 0,89] with the L1/GM1 Ratio, to 0.85 [95 %CI 0.72; 0.94] with the L2/GM2 ratio. Visual analyses AUC were excellent at both early (0.835 [95 %CI 0.7; 0.93]) and delayed acquisitions (0.83 [95 %CI 0.69; 0.92]). There was no statistically significant difference between accuracies ( p = 0.87) ranging from 0.71 with the RI ratio to 0.80 with both the “delayed” visual and quantitative analyses. Dual-PET protocol distinguishes RN and TR after SRS with robust diagnostic performances in case of doubtful MR, with an accuracy up to 80 %.

Fully automated analysis combining [18F]-FET-PET and multiparametric MRI including DSC perfusion and APTw imaging: a promising tool for objective evaluation of glioma progression

PurposeTo evaluate diagnostic accuracy of fully automated analysis of multimodal imaging data using [18F]-FET-PET and MRI (including amide proton transfer-weighted (APTw) imaging and dynamic-susceptibility-contrast (DSC) perfusion) in differentiation of tumor progression from treatment-related changes in patients with glioma.Material and methodsAt suspected tumor progression, MRI and [18F]-FET-PET data as part of a retrospective analysis of an observational cohort of 66 patients/74 scans (51 glioblastoma and 23 lower-grade-glioma, 8 patients included at two different time points) were automatically segmented into necrosis, FLAIR-hyperintense, and contrast-enhancing areas using an ensemble of deep learning algorithms. In parallel, previous MR exam was processed in a similar way to subtract preexisting tumor areas and focus on progressive tumor only. Within these progressive areas, intensity statistics were automatically extracted from [18F]-FET-PET, APTw, and DSC-derived cerebral-blood-volume (CBV) maps and used to train a Random Forest classifier with threefold cross-validation. To evaluate contribution of the imaging modalities to the classifier’s performance, impurity-based importance measures were collected. Classifier performance was compared with radiology reports and interdisciplinary tumor board assessments.ResultsIn 57/74 cases (77%), tumor progression was confirmed histopathologically (39 cases) or via follow-up imaging (18 cases), while remaining 17 cases were diagnosed as treatment-related changes. The classification accuracy of the Random Forest classifier was 0.86, 95% CI 0.77–0.93 (sensitivity 0.91, 95% CI 0.81–0.97; specificity 0.71, 95% CI 0.44–0.9), significantly above the no-information rate of 0.77 (p = 0.03), and higher compared to an accuracy of 0.82 for MRI (95% CI 0.72–0.9), 0.81 for [18F]-FET-PET (95% CI 0.7–0.89), and 0.81 for expert consensus (95% CI 0.7–0.89), although these differences were not statistically significant (p > 0.1 for all comparisons, McNemar test). [18F]-FET-PET hot-spot volume was single-most important variable, with relevant contribution from all imaging modalities.ConclusionAutomated, joint image analysis of [18F]-FET-PET and advanced MR imaging techniques APTw and DSC perfusion is a promising tool for objective response assessment in gliomas.