Ilwoo Park1, Rintaro Hashizume2, Joanna Phillips3,4, Sabine Mueller3,5, and Sarah Nelson1,6
1Radiology and Biomedical Imaging, University of California San Francisco, San Francisco, CA, United States, 2Neurological Surgery, Northwestern University, Chicago, IL, United States, 3Neurological Surgery, University of California San Francisco, San Francisco, CA, United States, 4Pathology, University of California San Francisco, San Francisco, CA, United States, 5Pediatrics, University of California San Francisco, San Francisco, CA, United States, 6Bioengineering and Therapeutic Sciences, University of California San Francisco, San Francisco, CA, United States
Synopsis
Diffuse intrinsic pontine gliomas (DIPGs) are one of the most difficult
pediatric cancers to treat. This study investigated the feasibility of 13C
magnetic resonance spectroscopic imaging (MRSI) of hyperpolarized (HP) [1-13C]pyruvate
for differentiating molecular characteristics of DIPGs. Differences in the lactate signal that were observed
in two distinct biopsy-originated orthotopic DIPG tumors were associated with changes in the levels of LHDA and HIF-1α activity. This suggests
that the non-invasive characterization of DIPGs using this new neuroimaging method
may be helpful for assessing treatment response and tumor progression. Introduction
Children with diffuse
intrinsic pontine glioma (DIPG) continue to have a dismal prognosis with a
median survival time of 9 month. Due to their brainstem location, surgical resection is impossible and
biopsies have rarely been performed. Recent studies that used a combination of postmortem
and biopsies samples have shed insight on genetic and molecular makeup of DIPGs
1,2
and propelled the identification of potential treatment targets
3,4. The
in vivo characterization of tumor molecular properties would be beneficial for
clinical management of the disease but there are currently no non-invasive
methods for assessing tumor molecular characteristics. This study investigates
the feasibility of using
13C magnetic resonance spectroscopic
imaging (MRSI) of hyperpolarized (HP) [1-
13C]pyruvate to
characterize DIPGs. The metabolism of two K27M-mutant, biopsy-originated DIPG xenografts,
one with features of glioblastoma (SF8628) and one with features of WHO grade
II infiltrative astrocytoma (SF7761), were investigated and compared with
molecular features from immunohistochemical analysis.
Methods
Primary human DIPG cells were implanted into the
brainstem of athymic rats (n=4 for SF8628 and n=4 for SF7761). All experiments
were performed using a GE clinical 3T scanner with a custom-designed
1H/
13C
coil. 35μL of [1-
13C]pyruvate
(with 1.5mM gadolinium) was polarized using a HyperSense
® DNP polarizer (Oxford
Instruments, Abingdon, UK)
5. Compressed sensing
13C 3D
MRSI data
6 were acquired at 20s from the start of the injection of
2.7mL hyperpolarized [1-
13C]pyruvate through tail vein. The injection
started ~10s after dissolution and lasted for 10s. The final dissolved solution
had a concentration of 100 mM pyruvate and pH of 7.5. The lactate and pyruvate
signals in the brainstem were normalized with respect to the relative signals
in normal appearing brain in the supratentorial region. In addition, the ratio
of lactate to pyruvate (Lac/Pyr) and lactate to total carbon (Lac/tC) were
calculated. T2-weighted fast spin-echo (FSE) images (TE/TR=60/4000ms) and
Gadolinium(Gd)-enhanced (0.2mmol/kg Gd-DTPA) T1-weighted spin-echo (SE) images
(TE/TR=10/700ms) were acquired in axial plane. The brains of rats were resected
and the following immunohistochemistry stains were performed: hematoxylin&eosin
(H&E), lactate dehydrogenase-A (LDHA), carbonic anhydrase-IX (CA-IX) and isocitrate
dehydrogenase-1 (IDH1).
Results and Discussion
HP 13C MRSI indicated that the SF8628
and SF7761 DIPG xenografts had distinct metabolic characteristics: The axial T2 FSE images showed
regions of T2 hyperintensity for both tumors (Fig1). SF8628 had well-defined
tumor margins (Fig1a), while SF7761 showed T2 lesion that spread throughout the
entire brainstem (Fig1e). There was no contrast enhancement on T1 post-Gd
images for either case (Fig1b, 1f). SF8628 produced elevated lactate signal
(Fig1c) and high Lac/Pyr (Fig1d) in the region of T2 hyperintensity. In
contrast, SF7761 did not show elevated levels of lactate (Fig1d, 1h). The mean
Lac/Pyr, Lac/tC and normalized lactate (nLac) in SF8628 tumors were
significantly higher than those in the contra-lateral brain tissue and in
SF7761 tumors (Fig2). The level of lactate signal in SF7761 tumors was similar
to that of contra-lateral normal brain tissue measured from rats with SF8628
tumors. The levels of normalized pyruvate (nPyr) were similar between SF8628
tumors, contra-lateral brain tissue and SF7761 tumors, highlighting that the
amount of pyruvate delivered and taken up by these tissues were comparable
between these different tissues.
Elevated lactate signal in SF8628 tumors was
associated with increased staining for LDHA and CA-IX: Immunohistochemical
analysis was used to further explain the findings from HP 13C MR
imaging. This indicated that in comparison to SF7761, the SF8628 tumors had
increased immunostaining for LDHA (Fig3), which is the enzyme that catalyzes pyruvate
to lactate conversion. The SF8628 tumors also produced a high level of immunostaining
for CA-IX (Fig3), which is an indicator of hypoxia-inducible factor 1α (HIF-1α)
activity that in turn transcriptionally activates LDH expression. The
differences in levels of LDHA and CA-IX staining are consistent with the levels
of hyperpolarized 13C lactate signal for these two orthotopic DIPG
tumors (Fig1, 2). A recent study reported that contrary to primary GBM tumors, IDH1
mutant oligoastrocytoma and oligodendroglioma did not produce significant
amount of lactate from hyperpolarized pyruvate7. In our study, there
was no difference in the level of IDH1 staining between the two types of tumor
(Fig3).
Conclusions
Hyperpolarized
13C
metabolic imaging is able to detect changes in metabolism that are associated
with distinct molecular characteristics in two types of orthotopic human DIPG xenografts.
The elevated lactate signal observed in DIPGs was associated with increased
levels of LHDA and HIF-1α activity. The results of the study indicate that this novel
neuroimaging method can be used to non-invasively characterize DIPGs and may be
applied to assist in stratifying patients to receive the most appropriate
treatments, and to assess treatment response and/or disease progression.
Acknowledgements
The
first author was supported by a NCI training grant in translational brain tumor
research (T32 CA151022) and Kure It Grant for Underfunded Cancer Research. The
support for the research studies came from NIH grants R01EB007588, P41EB13598,
R00 EB012064, R21CA170148 and R01CA154915.References
1. Wu G, Broniscer A,
McEachron TA et al. Somatic histone H3 alterations in pediatric diffuse
intrinsic pontine gliomas and non-brainstem glioblastomas. Nat Genet. Nat
Genet. 2012;44(3):251-3.
2. Schwartzentruber J,
Korshunov A, Liu XY, et al. Driver mutations in histone H3.3 and chromatin
remodelling genes in paediatric glioblastoma. Nature. Nature. 2012;482(7384):226-31.
3. Grasso CS, Tang Y, Truffaux N, et al. Functionally
defined therapeutic targets in diffuse intrinsic pontine glioma. Nat Med. 2015;21(7):827.
4. Mueller S, Hashizume R, Yang X, et al. Targeting Wee1 for the treatment
of pediatric high-grade gliomas. Nuero Oncol. 2014;16(3):352-60.
5. Ardenkjaer-Larsen JH, Fridlund B, Gram A, et al. Increase
in signal-to-noise ratio of > 10,000 times in liquid-state NMR. Proc Natl Acad Sci USA. 2003;100(18):10158-63.
6. Park I, Hu S, Bok R, et al. Evaluation
of heterogeneous metabolic profile in an orthotopic human glioblastoma
xenograft model using compressed sensing hyperpolarized 3D 13C magnetic
resonance spectroscopic imaging. Magn
Reson Med. 2012;70(1):33-9.
7. Chaumeil M, Radoul M, Eriksson P, et al. The tumor
exception that proves the rule: Hyperpolarized 13C MRS cannot be used to detect
the presence of mutant IDH1 glioma or their responses to Temozolomide therapy. Proc.
23rd Intl. Soc. Mag. Reson. Med. 2015;322.