EVALUATING ROLE OF MRI IN DIFFERENTIATING RECURRENT/RESIDUAL TUMOR FROM RADIATION INJURY IN A CASE SERIES OF RADIOTHERAPY TREATED CASES OF BRAIN TUMORS

Differentiating radiotherapy induced changes from residual/recurrent tumor is the mainstay challenge in neuro-oncoradiology. Contrast enhanced MRI is the current mainstay for evaluation of treatment response in brain tumors. However both shows similar clinical symptoms and post contrast enhancement.  Conventional MRI sequences provides only anatomical information, newer sequences also provides information regarding tumor microenvironment , metabolic and hemodynamic function.  The aim of our study was to evaluate the diagnostic effectiveness of MR spectroscopy and MR perfusion images to differentiate recurrent/residual tumor and radiation injury using parameters like Cho/NAA, Cho/Creat, NAA/Creat ratios and rCBV. In our hospital based observational study, Contrast MRI with advanced sequences eg. MR spectroscopy, DSC Perfusion MRI were performed using 3T MRI Machine. Total 40  cases of radiotherapy treated brain tumors were analyzed. In our study we found that mean value of Cho/NAA were significantly higher in residual/recurrent lesion (5.4), compared to radiation injury(0.33). Mean value of Cho/creat were significantly higher in residual/recurrent lesion(2.9), compared to radiation injury(1.01). Mean value of NAA/Creat were significantly lower in residual/recurrent lesion(0.45) compared to in radiation injury(4.2). Mean value of rCBV in a contrast enhancing lesion were significantly higher in residual/recurrent lesion(713), compared to in radiation injury(129). Mean value of rCBV ratio in lesion compared to contralateral normal white matter were significantly higher in residual/recurrent lesion(3.4), compared to in radiation injury(0.49). MR Spectroscopy and MR Perfusion may increase the accuracy of differentiating recurrent/residual tumor from radiation injury in radiotherapy treated cases of brain tumors and can help avoid unnecessary invasive procedures like biopsy.