Accurate DCE-MRI Measurement of Glioblastoma Using Point-of-care Portable Perfusion Phantom

The goal of this study is to test whether a new device developed at the University of Alabama at Birmingham (UAB) can decrease the error in calculating blood flow of a brain tumor, leading to better prognosis. UAB radiological research team has been studying a cutting-edge imaging technique named dynamic contrast enhanced (DCE) magnetic resonance imaging (MRI) , or DCE-MRI, over 10 years. This technique has been globally used to calculate blood flow of various tissues including tumors. Blood flow often serves as a critical indicator showing a disease status. For example, a brain tumor has typically high blood flow, so the magnitude of blood flow can be used as an indicator to identify the presence and aggressiveness of a brain tumor. In addition, an effective therapy can result in the alteration of the blood flow in a brain tumor. Therefore, the investigators may be able to determine whether the undergoing therapy is effective or not by measuring the blood flow in the brain tumor, and decide whether they need to continue the therapy or try a different one. However, unfortunately, the measurement of blood flow using DCE-MRI is often inaccurate. MRI scanners may use different hardware and software thus the measurement may be different across scanners. The measurement may also be different over time due to hardware instability. Therefore, the investigators propose to use an artificial tissue, named "phantom", together with a patient. The phantom has a constant blood flow thus it can serve as a standard. Errors, if it occurs, will affect the images of both the patient and the phantom. Therefore, the investigators will be able to correct the errors in the patient image using the phantom image. UAB radiological research team invented a new device for this purpose named point-of-care portable perfusion phantom, or shortly P4. The team recently demonstrated the utility of the P4 phantom for accurate measurement of blood flow in pancreatic cancer and prostate cancer. In this study, they will test whether the P4 phantom will improve the measurement accuracy in brain cancer.

Glioblastoma is the most common primary malignant type of brain tumor in adults. Surgical tumor resection followed by chemoradiation therapy is the standard of care for patients with glioblastoma, but its prognosis is still fairly dismal (median survival time = 15 months). One major concern that prevents effective treatment management is the difficulty of differentiating between pseudo-progression and true-progression. Pseudo-progression occurs in about 20-30% of glioblastoma patients typically within 3 months after chemoradiation therapy has been completed. Pseudo-progression is a local inflammatory reaction caused by irradiation and enhanced by concurrent chemotherapy, which leads to a transient increase of blood brain barrier (BBB) permeability. The BBB, however, is also disrupted by new cancer occurrence. Therefore, both pseudo- and true-progressions appear with an increased contrast enhancement in MRI, and there are currently no established techniques to differentiate between them. Pseudo-progression is typically known to be associated with better clinical outcomes, so pseudo-progression mistaken for true-progression results in the discontinuation of an effective therapy, while true-progression mistaken for pseudo-progression leads to the continuation of an ineffective therapy that may induce adverse side effects. DCE-MRI has potential to differentiate between pseudo- and true-progressions of glioblastoma. The enhancing lesions of pseudo-progression are due to inflammation, whereas those of true-progression are caused by cancer growing. Thus, true-progression typically presents higher perfusion than pseudo-progression does. DCE-MRI can quantitatively assess the tissue perfusion by monitoring the dynamic change of MRI contrast agent concentration. Several investigators have demonstrated the potential of quantitative DCE-MRI to differentiate between pseudo- and true-progressions. However, the variability in quantitative DCE-MRI measurement across different MRI scanners remains a major concern, as it hinders data comparison among institutes to retrieve a reliable threshold for accurate prognosis and subsequent treatment optimization. A point-of-care perfusion phantom may allow high reproducibility and accurate comparison of quantitative DCE-MRI data across MRI platforms. The UAB radiological research team recently developed the P4 phantom, which is small enough to be imaged concurrently with a patient for real-time quality assurance, but large enough not to suffer from the partial volume effect. The P4 phantom creates constant contrast enhancement curves with very robust repeatability, and thus the contrast agent concentration time-course in a tumor, which is a major source of error in quantitating DCE-MRI parameters, can be accurately calculated in reference to the values observed in the phantom. In our previous study, the variability in quantitating the volume transfer constant of various human tissues across two different MRI scanners was reduced fivefold after P4-based error correction. The investigators hypothesize that the variability in quantitative DCE-MRI measurement of glioblastoma across different scanners will be significantly reduced when the P4 is used for error correction, leading to better differentiation between pseudo- and true-progressions. The goal of this study is to test this hypothesis.

glioblastoma patients with newly or enlarged enhancing lesion within 3 months after completing 6 weeks of adjuvant chemoradiation therapy

P4 is a perfusion phantom developed by Dr. Harrison Kim that can significantly reduce variation in quantitating perfusion of human abdominal tissues across MRI scanners.

The goal is to measure the reproducibility of blood perfusion measurement in the glioblastoma using the two consecutive DCE-MRI scans with and without P4-based error correction. The pharmacokinetic (PK) parameter within the region of interest (ROI) will be averaged at each scan after P4-based error correction, and the mean values of two scans will be compared to calculate the reproducibility coefficient (%RDC) using the equation, %RDC=2.77wCV, where wCV is the within-subject coefficient of variation. The %RDC before P4-based error correction will also be calculated for comparison. Data reproducibility will be assessed using the intra-class correlation coefficient (ICC) as well. ICC = σ 2b / (σ 2b+ σ 2w), where σb is between-subject standard deviation and σw is within-subject standard deviation.

To determine whether the differentiation between the pseudo- and true-progressions of glioblastoma can be improved using qDCE-MRI after P4-based error correction.

The PK parameter (e.g., Ktrans) in the tumor with pseudoprogression will be statistically compared with that with true-progression before and after P4-based error correction to determine whether the differentiation between the pseudo- and true-progressions of glioblastoma can be improved using qDCE-MRI after P4-based error correction. Each tumor will be classified into pseudo- or true-progression based on RANO criteria.

Inclusion Criteria: Adult patients (age 18 years or older). Patients treated with surgery, followed by chemoradiation therapy, and currently under chemotherapy. Patients with a newly or enlarged enhancing lesion inside the radiation field at least three months after completion of radiation therapy. Patients with signed informed consent. Exclusion Criteria: Participants with safety contraindications to MRI examination (determined by standard clinical screening). Participants on hemodialysis or with acute renal failure. Participants who are pregnant, lactating or are planning to become pregnant during the study. Participants who are planning to farther a child during the study.