Relationship of invivo MR parameters to molecular characteristics of non-enhancing lower-grade gliomas
Tracy L Luks1, Tracy Richmond McKnight1, Aurelia Williams1, Evan Neill1, Khadjia Lobo1, Anders Persson2, Arie Perry3, Joanna Phillips3, Annette Molinaro4, Susan Chang4, and Sarah J Nelson1

1Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, 2Neurology, University of California San Francisco, San Francisco, CA, United States, 3Pathology, University of California San Francisco, San Francisco, CA, United States, 4Neurosurgery, University of California San Francisco, San Francisco, CA, United States

Synopsis

Invivo MR anatomy, diffusion, perfusion, and spectroscopy profiles from non-enhancing grade 2 and grade 3 gliomas were examined by histologic and molecular characteristics associated with clinical outcome. Patients underwent a pre-surgical 3T MR exam including IRSPGR, FSE, FLAIR, DWI, MRSI and DSC. For surgical biopsies, histological sub-type, grade, cleaved caspase-3, MIB-1, Ki67, IDH1R132H, ATRX, p53, and co-deletion of 1p19q were determined. Overall, molecular characteristics associated with worse clinical outcome were associated with higher ADC and lower FA, lower nCBV and nPH, and higher Recov, and higher nLAC.

Purpose

In vivo MR anatomy, diffusion, perfusion, and spectroscopy profiles of 88 biopsy samples from 45 non-enhancing grade 2 and grade 3 gliomas were examined by histologic and molecular characteristics associated with clinical outcome. In lower-grade gliomas, higher grade, IDH wild-type, 1p19q intact, p53 mutated and ATRX mutated tumors often have worse clinical outcomes1. The goal of this study was to identify MR parameters associated with these molecular characteristics that are predictive of tumor aggressiveness or therapeutic response.

Methods

Patients underwent a pre-surgical 3T GE MR exam with an 8 channel head coil. The MRI examination included T1-weighted IRSPGR, T2-weighted 3D FSE and/or XETA T2 FLAIR, 6 directional axial Diffusion Weighted Imaging with b=1000s/mm2; lactate-edited 3D MRSI with PRESS volume localization; and dynamic susceptibility contrast enhanced Perfusion Weighted Imaging with a 3- 5ml/s injection of 0.1mmol/kg body weight Gd-DTPA.

During surgery, the locations of biopsy samples were recorded relative to the FSE image using a surgical nativigation system. Tissue samples were sectioned and fixed in 4% formalin, dehydrated by graded ethanols, and embedded in wax for histopathology analysis. Histological sub-type (AS, OA, or OD), grade, cleaved caspase-3 (CC-3), MIB-1, Ki67, IDH1R132H, ATRX, p53, and co-deletion of 1p19q were determined by standard pathology and immunohistochemistry and FISH assessment of tumor tissue samples. In order to compare imaging with histopathological parameters, 5mm spherical ROIs at biopsy locations were used to calculate the median imaging value around each biopsy location. The following MRI parameters were calculated: Diffusion: nADC, eigenvalues (nEVaxial, nEVradial), nFA, Perfusion: nCBV, recov, and nPeak Height (nph) MRS: nNAA, nCho, nCre, nLac, nLip, Cho/Naa, Cho/Cre. The effects of histological and molecular tumor characteristics on MRI parameters were assessed with repeated-measures analyses of variance. Correlations between imaging and continuous histopathological parameters were evaluated using Kendall Tau tests.

Diffusion Results

Grade 3 biopsies had higher nADC than grade 2. IDH1 mutant biopsies had higher nADC, nEVaxial, nEVradial. p53 mutant biopsies had higher nADC. In grade 3, p53 mutant biopsies had higher nADC and lower nFA. ATRX mutant biopsies had higher nADC, nFA, and nEVaxial. Within Grade 3, ATRX mutant biopsies had higher nADC, nFA, nEVaxial, nEVradial. Overall, with the exception of IDH mutation which showed the opposite trend, molecular characteristics associated with worse clinical outcome were associated with higher ADC and lower FA.

Perfusion Results

Grade 2 biopsies had higher nCBV and nph. Higher MIB-1 biopsies had higher Recov. Higher CC-3 biopsies had lower Recov. 1p19q co-deleted biopsies had higher nCBV and nPH. In Grade 3 biopsies and AS biopsies, 1p19q co-deleted biopsies had lower Recov and those that were p53 mutant had higher Recov. Overall, molecular characteristics associated with worse clinical outcome were associated with lower nCBV and nPH, and higher Recov.

MRS Results

1p19q co-deleted OD biopsies had lower nLAC. IDH-1 mutant biopsies had lower nLAC. In Grade 3 IDH-1 mutant biopsies, nLAC was lower. P53 mutant biopsies had higher nLIP and nLAC. Overall, molecular characteristics associated with worse clinical outcome were associated with higher nLAC.

Conclusion

In our sample of 88 non-enhancing, newly-diagnosed lower grade glioma biopsies, molecular characteristics associated with poorer clinical outcome were associated with higher ADC and lower FA. These findings are in contrast to previous reports, primarily in GBMs, that lower ADC is associated with worse clinical outcomes 2,3,4. ADC is generally considered to be a measure of tumor cellularity, because as proliferating tumor crowds out normal tissue architecture, the diffusion of extracellular water is abnormally restricted. However, in the current data, the diffusion of water in the tumor region increased in many of the tumors with molecular characteristics of aggressiveness and poorer prognosis. Part of this may be due to our study including exclusively newly-diagnosed, non-enhancing lower grade gliomas, that still have relatively low tumor cellularity 2. In these tumors ADC may instead reflect the disruption of neuronal integrity and axonal and glia microstructure. Similarly, in our sample, molecular characteristics associated with poorer clinical outcome were associated with lower nCBV. nCBV generally increases with tumor aggressiveness, as tumor angiogenesis increases in GBMs. Again, the perfusion results in these exclusively newly-diagnosed, non-enhancing lower grade gliomas suggests that angiogensis is still relatively low, and the low perfusion values are reflective of the disruption of normal microvascular architecture.

Acknowledgements

This work was supported by NIH R01 CA159869 and NIH P50 CA097257.

References

1. Claus, EB, Walsh, KM, Wiencke, JK et al. Survival and low-grade glioma: the emergence of genetic information. Neurosurgical Focus 38:1-10, 2015.

2.Khayal, IS, Vandenberg, SR, Smith, KJ et al. MRI apparent diffusion coefficient reflects histopathologic subtype, axonal disruption, and tumor fraction in diffuse-type grade II gliomas. Neuro-Oncology 13:1192-1201, 2011.

3. Wen Q, Jalilian L, Lupo JM et al. Comparison of ADC metrics and their association with outcome for patients with newly diagnosed glioblastoma being treated with radiation therapy, temozolomide, erlotinib and bevacizumab.Journal of Neuro-Oncology 121:331-339, 2015.

4. Togoa, O, Hiwatashi, A, Yamashita, K et al. Differentiation of high-grade and low-grade diffuse gliomas by intravoxel incoherent motion MR imaging. Neuro-Oncology 2015: doi:10.1093/neuronc/nov147

Figures

Figure 1. Example FLAIR, ADC and CBV images for Grade 2 and Grade 3 gliomas with biopsy locations

Figure 2. Molecular characteristics of lower-grade non-enhancing Glioma biopsies associated with worse clinical outcome were associated with higher ADC.

Figure 3. Molecular characteristics of lower-grade non-enhancing Glioma biopsies associated with worse clinical outcome were associated with lower nCBV .

Figure 4. Molecular characteristics of lower-grade non-enhancing Glioma biopsies associated with worse clinical outcome were associated with higher nLAC.

Figure 5. Example ADC colormaps for Grade 2 and Grade 3 gliomas



Proc. Intl. Soc. Mag. Reson. Med. 24 (2016)
0293