DefinePK

Abstract

Abstract


High grade brain tumours are treated with surgery, chemotherapy and radiation therapy and despite such aggressive treatment, can recur in a short span of time. MRI scan has been the conventional diagnostic modality to diagnose recurrence, although at times it becomes difficult for the neuroradiologists to differentiate between tumour recurrence and radiation necrosis. Herein lies the emergent need to explore the efficacy of functional imaging to assist in this diagnostic challenge. Recent studies have sought to do so with promising implications, which we have attempted to summarize in this review.

Keywords: Functional imaging, neuro-oncology; tumor recurrence; treatment necrosis; perfusion.

Introduction

Despite aggressive treatmenrt regimes for high grade brain tumours, recurrence remains inevitable.1 For a considerable amount of time, conventional contrast enhanced MR imaging has been the mainstay of assessing post treatment tumour recurrence, especially in CNS neoplasms.The usual indicators of a recurrent tumour on follow up MRI include progressive enlargement of lesion, causing mass effect and infiltration of corpus callosum; whereas the enhancement pattern following a Swiss cheese or spreading wavefront pattern is more indicative of radiation necrosis. 2 Over the years considerable overlap between these two types has been observed which has prompted new research efforts into investigating advanced non-invasive imaging methods measuring physiological tumour properties. 3



Novel modern day research centers on bringing forth the capabilities of MR perfusion in exploiting the functional differences at the cellular level between recurrent or progressive tumour growth from treatment-induced necrosis after radiation therapy.These include increased cell proliferation with neo-angiogenesis in case of tumour; and liquefactive necrosis, vascular hyalinization and endothelial damage in case of radiation induced changes. 4 The observed clinical symptoms vary from none to significant neurological deficit predominantly affecting the white matter. Vermaet al., 5 reported an incidence of 3–24% for radiation necrosis, showing a direct correlation with the dose of radiation, duration and volume of targeted brain parenchyma.


Technique

In perfusion scanning, successive images are obtained during the first pass of contrast. There are two methods of obtaining perfusion sequences namely dynamic susceptibility-weighted contrast-enhanced (DSC) and dynamic contrast-enhanced (DCE) imaging. In DSC-MRI, the susceptibility effect of contrast causes a T2 signal drop in DSC-MR thus allowing measurement of haemodynamic parameters: relative cerebral blood volume (rCBV), relative peak height (rPH), and percentage of signal-intensity recovery (PSR). Higher rCBV indicates highly permeable blood vessels as in the case of tumour neo-angiogenesis whereas lower values stipulate treatment necrosis reducing blood flow. 5 CBV data is compared with the contralateral side for normalization hence the term relative. 6 In DCE-MRI, rapid sequence T2 imaging is used to measure signal intensities of contrast bolus which reflects lesion perfusion, permeability and extracellular volume. However very few precedents are found in literature that show compelling results in application of DCE-MRI to the question of tumour recurrence versus radiation necrosis. 5


Review of Evidence
 

Prager et al. 7, conducted perfusion analysis in post treatment enhancing lesions for patients with primary high grade gliomas. They reported lower rCBV values in treatment related changes, rCBV lesion (P = 0.003) and rCBVROI (region of interest) (P = 0.011). An optimized rCBV lesion threshold of ≥1.27 had 86.5% sensitivity and 83.3% specificity with AUC (area under curve) of 0.863 for the diagnosis of recurrence.Barajas et al. 8, performed a retrospective review of 27 patients that underwent gamma knife radiosurgery for metastatic lesions of the brain.Upon follow up patients that presented with enhancing lesions on conventional imaging were selected and ROI were drawn around the entire contrast enhancing region.Their observations stipulated lower rCBV (P