 Hello everyone, I am Dr. Adarsh Anil and the paper that I will be presenting today is on the utility of advanced MRI techniques and radio genomics in glioma grading. So, as we all know, gliomas are the most common primary intracial brain tumors which are further graded from grades 1 to 4 among which grades 1 and 2 belong to the low grade category and grades 3 and 4 belong to the high grade category. Imaging plays a crucial role in the diagnosis of these brain tumors, among which the most important role is played by conventional and advanced MRI techniques. The various advanced MRI techniques that I used include diffusion weighted imaging, diffusion tension imaging, magnetical resonance spectroscopy, perfusion imaging and so on. Whatever is done, the goal standard for the diagnosis of these tumors is to pathological examination and the ideal are genetic studies. A midway in resource poor settings is that of immunohistochemistry. So, what happens when these genetic studies come together with radiology, what we get is radio genomics. So, the objectives of our study were basically three. The first objective was to differentiate between low grade and high grade gliomas on the basis of conventional and advanced MRI findings and correlate them with the histopathology findings. The second objective was to further investigate the utility of diffusion and perfusion metrics in differentiating these grades of gliomas. And third was to correlate the MRI features with genotypy expression of IHC markers like IDH1, ATRX, KI67 and P53. The study was conducted in the Department of Radio Diagnosis, AIMS-TRIPUD. It was an observation, analytical cross-sectional study conducted over a duration of 18 months after receiving IHC approval. It was conducted on a 3 Tesla MRI and patients with biopsy proven gliomas and age more than 7 years was included in our study. All patients with contraindications to MRI and previously operated patients were excluded. This is the standard image acquisition protocol that we used. The image analysis was conducted on standard post-processing software that is under review with ROIs placed in the tumor, the perituminal region and the contralateral normal white matter. Data analysis was conducted on Microsoft, Excel and IBM SDSS with appropriate statistical test of significance wherever considered appropriate. And surgical biopsy specimens were analyzed for histopathological examination and immunohistochemical analysis. So we included a total of 61 patients, among which 15 patients did not undergo surgery and of the remaining 37, 36 were inter-excel brain tumors, 7 were of non-glial lineage among this and 29 were of glial lineage. Among the 29 gliomas, 3 were oligodendrogliaomas, 3 were ependymomas and 23 were astrocytomas. Of these, 1 ependymoma and 13 astrocytomas we had the IHC data available. Among the 23 astrocytomas, 8 were of low grade, 15 were of high grade which were further used for conventional as well as advanced MRI characterization. So total of 29 patients we had 18 meals and 11 females, the mean age of which is shown here. Among the MRI parameters, the interrelational mean ADC was found to be significantly lower in high grade gliomas for which the cutoff of 997.55 was found with the 75% specificity and 75% sensitivity and 86.7% specificity. The interrelational mean RCBV and interrelational mean RCBF were found to be significantly higher for high grade gliomas compared to low grade gliomas and the cutoff values of 1.815 and 8.405 for the same are shown here. This is a box and whisker plot showing the interrelational mean ADC values and these are the ROC curves which we use for detecting the cutoffs. Among the DTI metrics, one parameter we found especially useful was the P by N FA ratio which is the FA value of the perillational region divided by the FA value of the contralateral normal white matter which we found to be significantly lower in high grade gliomas at a P value of 0.037 and a cutoff of 0.415 could differentiate between the two grades at 75% sensitivity and 86.5% specificity. The rest of the DTI metrics did not yield significant results in our study. This is a box and whisker plot of the same significant finding and the ROC curves used for detecting the cutoff. So among the amniotic IDC markers, we used ATRX, IDH1, P53 and KI67 and this is the take home message from this side is that in all grade 4 gliomas, we found the KI67 immuno reactivity more than 10% and in grade 2 gliomas in all it was less than 10%. So among the imaging features, we found that multi-lober involvement was found in all cases of IDH weld type which virus it was found only in 88.8% of IDH mutated cases. Heterogeneous enhancement with large non-enhancing areas and market perillational edema were found only in IDH mutated cases and the minimum ADC value and the was lower in IDH weld type cases and the maximum RCB value was higher in IDH mutated cases. Similar findings of multi-lober involvement heterogeneous enhancement market perillational edema in minimum ADC and maximum RCB were also used for ATRX loss and intact cases where we found market perillational edema was found only in ATRX loss cases and minimum ADC and maximum RCB value was lower and maximum RCB value was higher in case of ATRX lost cases. So coming to the discussion, the interrelation mean ADC as I already told it was lower for high grade gliomas which was in concordance with previous studies which can be explained on the basis of cellularity as the increasing cellularity it causes increased restriction of water molecules. The interrelation effect did not yield significant results in our study how in previous studies they have yielded significant results on which they have told like it may be because of it might be because of the loosely arranged cells in the fibrelia matrix that may be responsible for the lower FA values in low grade gliomas compared to the highly organized high grade gliomas having higher FA values. The P by N FA ratio which we found significantly lower in high grade gliomas has not been reported before and the advantages of this parameter is that the regional differences in FA values among different regions of the brain are negated. The interrelational RCB and RCB were for significantly higher in high grade gliomas which were in concordance with the previous studies and it can be explained by the increasing neurovascularity with increasing tumor grade. RCBF is a less reliable parameter because of the variation in tumor blood flow among different regions. So this is a prognostic grading in which we can see C, A, B and D are four groups based on the IHC markers in which A has the best prognosis which is an IDH mutant with ATRX intact status and D has the worst prognosis which is an IDH wild type with ATRX intact status. So these are the findings that I explained before which has concordance with previous studies except for the IDH mutation status and the RCBV in which we found the maximum RCB was higher in IDH mutated status whereas the Zhongfang et al. found it to be higher in IDH wild type. The reason for this difference may be because of the lesser number of IDH wild type cases we had in our study. The ATRX mutation status has not been correlated with imaging findings before. So the over expression of genes associated with hypoxia, angiogenesis and edema has been linked with the contrast material enhancement and necrosis as well as with hyperfusion and restricted water diffusion. So this is a representative case showing the conventional sequences, the advanced RCBV, RCBF, DTI and the MRS. The corresponding HPE and IHC panel of a glioblastoma which is of IDH1 mutant and ATRX intact status. So the basic take home message I would like to tell is that radio genomics holds great promise for aiding in the inexpensive non-invasive phenotyping of these gliomas and radiology should go beyond diagnosis and into prognosis to predict which glioma has a poor prognosis and which glioma has a better prognosis. Thank you. These are my references. Thank you.