Dose Response of Functionally Critical Brain Regions for Brain Radiotherapy

Normal tissue response is critical for brain radiotherapy, especially for dose escalation which carries with it an increased incidence of radiation-induced brain injury. Although radiation toxicity and limiting dose for anatomically critical structures of the brain have been well studied and documented, little is known for functionally critical brain regions and treatment of cognitive sequelae of cranial radiotherapy is limited. The objective of this clinical protocol is to accumulate preliminary data for future studies aiming to quantify dose response for functionally critical brain regions for brain radiotherapy. We plan to achieve this objective by correlating the radiation-induced complications and radiological changes with the radiation dose to the selected functionally critical brain regions for 25 patients. Each participating patient will receive brain fMRI to identify brain regions for processing visual, working memory and language functions. The image co-registration algorithm developed previously by our group will be used to co-register these regions on the CT scans for radiotherapy treatment planning for radiation dose calculation. Radiation-induced changes in cognitive functions will be evaluated using the modified mini mental status exam (3MS) and fMRI during the routine follow-up. The knowledge derived from this study might significantly improve the quality of life and allow safer dose escalation for patients receiving brain radiotherapy.

Research Question: Our goal is to answer the following three research questions in this study: 1. What is the dose response curve (degree of radiation complications/radiological changes vs. radiation dose) for brain regions involved in the processing of visual, language and working memory functions? 2. What is the limiting dose for these functionally critical brain regions that can be applied to radiotherapy treatment planning to minimize (e.g., less than 5%) the possibility of radiation-induced injury? 3. Can brain fMRI be used to guide the radiotherapy treatment planning to minimize the dose to functionally critical brain regions and evaluate treatment responses of these regions? Research Method: Imaging before radiotherapy: For each patient, active stimulation and resting fMRI examinations will be performed in addition to standard anatomic MRI for brain radiotherapy. Activation maps of regions involved in visual, language and working memory functions will be generated from the fMRI scans and the patient's responses on of the working memory task (i.e., the N-Back task) will be recorded during the fMRI studies. The MRI and fMRI will, combined, take approximately 90 minutes - 60 minutes for the standard MRI and 30 minutes for the fMRI. The activation maps will be overlaid onto the FLAIR MRI images, imported into the radiotherapy treatment planning system and co-registered with the simulation CT scan. Radiotherapy and routine follow-up: Functionally critical areas in the fMRI activation maps will be separately contoured as critical organs but will not be considered for plan optimization. Dose volume histograms (DVHs) of functionally critically brain regions will be calculated and documented. Each patient will receive fractionated brain radiotherapy according to the treatment plan and will be followed up with every three months after radiotherapy. Follow-up evaluations: The 3MS examination will be administered prior to initiation of treatment onset and every three months after treatment during routine follow-up to screen neuropsychological functions. Anatomic MRI (standard) and brain fMRI (research) will be acquired at 6 months after radiotherapy to determine the radiological and functional changes in regions involved in the processing of visual, language and working memory functions. Minimizing the interference of tumor recurrence or progression: Two actions will be taken to minimize the interference of tumor recurrence or progression with the evaluation of performance alterations caused by radiation. First, patient inclusion will be restricted to patients with slow growing tumors, e.g., meningioma, low grade glioma and anaplastic astrocytoma so that the chance for tumor recurrence or progression is small. Secondly, patients who develop disease progression at any of the post-radiotherapy follow-ups will be excluded from the study, as it would not be possible to distinguish the effects of disease progression from radiotherapy side effects on the fMRI or the 3MS examination, and the findings would not be interpretable. Dose Response Analysis: Logistic regression analysis will be used to assess the likelihood of complications to functionally critical brain regions as a function of radiation dose. The dependent variable will be the binary indicator that identifies patients' manifesting complications (decline of 3MS examination score and N-Back test, radiological/functional changes identified on MRI and fMRI) attributed to radiation exposure. The logistic regression model will include relevant subject level factors (e.g., age, gender, baseline Karnofsky performance status, tumor location, recent seizures, anti-epleptic medications) as covariates. The fitted logistic normal tissue complication probability (NTCP) model will permit estimation of the probability of significant complications associated with any radiation dose to the area, allowing the limiting dose to be calculated as the dose expected to induce complication with an acceptable probability (e.g., 5%). Incorporation of limiting dose for radiotherapy treatment planning: The limiting dose from the above "Dose Response Analysis" will be used for IMRT re-planning considering the functionally critical brain regions. The optimization goal is to minimize the dose to functionally critical brain regions while maintaining similar PTV (planning target volume) coverage and keeping dose to the critical structures within accepted tolerance. DVHs of the functionally critical brain regions, PTV and all other critical structures will be compared for both treatment plans with and without considering the functionally critical brain regions. A Wilcoxon matched pair signed rank test will be performed to compare the plans in terms of calculated NTCP of critical organs. The McNemar test will be used to compare plans with and without dose constraints of functional areas in terms of yield rate (i.e., the proportion of patients for whom the IMRT plan was successfully devised). The primary study endpoint of this study is normal tissue complication. The primary objective of this study is to determine the dose response curve and limiting dose for functionally critically brain regions for brain radiotherapy. The second objective of this study is to investigate the accuracy of active stimulation or resting state fMRI for avoidance of the functionally critical brain regions for radiotherapy treatment planning of brain tumors and for evaluating the treatment responses of these critical regions.