Tracy L Luks1, Tracy Richmond McKnight1, Evan Neill1, Llewellyn Lynn Jalbert1, Arie Perry1, Soonmee Cha1, Joanna Phillips1, Annette Molinaro1, Susan Chang1, and Sarah J. Nelson1
1University of California San Francisco, San Francisco, CA, United States
Synopsis
The relationship of diffusion imaging parameters with prognostic histological
and molecular factors for patients with grade II and III gliomas is unclear,
particularly for tumors that are non-enhancing on post-Gadolinium images. We investigated the relationship of ADC and FA values with histological tumor score, tumor grade, and molecular
characteristics for non-enhancing (NE) vs contrast-enhancing (CE) and
newly-diagnosed (ND) vs recurrent (REC) disease. In NE patients, histopathological and
molecular characteristics associated with poorer clinical outcome were found to have higher ADC and lower FA. In CE patients, some
characteristics associated with poorer outcome hadlower ADC
and higher FA.
Purpose
Treatment planning for patients with grade II and grade III glioma is
challenging, and there is no consensus on the timing of radiation and
chemotherapy. Decisions are based upon histopathological and molecular
characteristics of resected tissue, in combination with MRI findings. Obtaining
tissue samples from regions that reflect the most malignant tumor
characteristics is essential for optimal treatment planning. Diffusion imaging
can be used to assist in identifying target locations for tissue sampling,
especially in non-enhancing gliomas. Estimates of the Apparent Diffusion
Coefficient (ADC) and Fractional Anisotropy (FA) have been associated with the degree
of abnormal tissue microstructure. In WHO
grade IV glioma, regions of low ADC and high FA are reliably indicative of malignant
behavior1-3. But in grade II and III gliomas, the relationship
between diffusion imaging parameters and prognostic histological and molecular factors is unclear, particularly in non-enhancing
tumors. In the current study, we
investigated the relationship between diffusion imaging and histological tumor
score, tumor grade, and molecular characteristics for non-enhancing (NE) vs
contrast-enhancing (CE) and newly-diagnosed (ND) vs recurrent (REC) disease. Methods
285 patients underwent a pre-surgical 3T GE
MR exam using an 8 channel head coil. The MRI examination included T1-weighted
IRSPGR, T2-weighted 3D FSE and/or XETA T2 FLAIR, and 6 directional axial
Diffusion Weighted Imaging with b=1000s/mm2. During surgery, the locations of tissue
samples were recorded relative to the FSE image with a surgical navigation
system. WHO 2016 diagnosis, grade, MIB-1/ki-67, tumor score (based on
contribution of tumor cellularity to total cellularity), IDH mutation status, and co-deletion of
1p19q were determined by standard pathology, immunohistochemistry and FISH. 5mm
spherical ROIs centered at the tissue sample locations were used to calculate
the median ADC and FA values around each location, normalized to normal brain
tissue. The nADC (normalized
ADC) and nFA (normalized FA) values were also calculated for the entire T2
Flair lesion, and in CE gliomas, the contrast-enhancing T1 lesion. The effects
of histological and molecular tumor characteristics on diffusion parameters were
evaluated according to enhancement and recurrence status using repeated-measure
analysis of variance. Results
In ND-NE patients, 185 tissue samples were obtained from 78 patients. Tissue samples with higher nADC and lower nFA values had
worse tumor scores, higher MIB scores, and came from tumors with higher WHO grade,
astrocytoma diagnosis, intact 1p19, and IDH wildtype. In the entire T2 flair lesion, IDH wildtype patients
had higher nADC than IDH mutant patients, and 1p19q intact patients had higher nADC
than co-deleted patients. No tissue samples were obtained from
contrast-enhancing locations in ND patients with contrast enhancement.
In REC-NE patients, 72 tissue samples
were obtained from 38 patients. Tissue samples with lower nFA values had better
tumor scores, and tissue samples with higher nADC values had lower MIB scores, and
were from tumors with astrocytoma diagnosis, intact 1p19q, and IDH wildtype. In
the entire T2 flair lesion, IDH wildtype patients had higher nADC than IDH mutant
patients. Across all patients, nFA was lower for ND-NE than REC-NE patients.
In REC-CE patients, 23 tissue samples in contrast enhancing regions were
obtained from 16 patients. There were no
significant diffusion MRI differences associated with histopathological or
molecular characteristics. nADC was lower and nFA was higher for REC-CE than
REC-NE tissue samples. In the entire contrast-enhancing lesion, grade III patients
had lower nADC than grade II patients, and 1p19q co-deleted patients had higher
nFA than 1p19q intact patients. In the entire T2 flair lesion, grade III
patients had lower nADC than grade III patients. Across all patients, nADC was
lower and nFA was higher for REC-CE than REC-NE T2 flair lesions.
Discussion
In NE lower grade glioma patients, especially ND, histopathological and molecular
characteristics that are associated with poorer clinical outcome4
were found to have higher nADC and lower nFA. In CE patients, some
histopathological and molecular characteristics associated with poorer clinical
outcome were associated with lower nADC and higher nFA.
In ND-NE patients, higher nADC and lower nFA may be driven by increased
disruption of normal tissue architecture, whereas in CE patients, increased
tumor cell density may drive lower nADC. These results suggest that when
considering nADC and nFA values in tissue sample targeting and in evaluations
of tumor malignancy and treatment response in grade II and III gliomas, the
interpretation of these values depends upon the patient’s enhancement and recurrence
status. Acknowledgements
This work was supported by NIH R01 CA159869 and NIH P50 CA097257. References
1.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.
2. 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.
3. Togao, O., Hiwatashi,
A., Yamashita, K. et al. Differentiation of high-grade and low-grade diffuse
gliomas by intravoxel incoherent motion MR imaging. Neuro-Oncology 18:132-41, 2015.
4. Claus, EB, Walsh, KM,
Wiencke, JK et al. Survival and low-grade glioma: the emergence of genetic
information. Neurosurgical Focus 38:1-10, 2015.