极T代谢磁共振全球科研集锦

295

where T1 was assumed to be equal for all 13C metabolites.

The reverse conversion rate kLP was assumed to be 0 in the

model because it is much lower compared to the forward

reaction in physiological conditions, and such assumption

improves the stability of fitting computations.19,20

3 | RESULTS

3.1 | Preclinical studies

The acquisition was delayed by 15 s from the beginning of

injection. Immediately following the arrival of HP-pyruvate

in TRAMP tumor, lactate dehydrogenase (LDH) rapidly catalyzed the reduction reaction to HP-lactate. This reflected the

upregulated LDH gene expression/activity in cancer as found

previously.1 Pyruvate signal displayed its maximum near the

beginning of the acquisition and then decreased because of

the metabolic conversion, the RF excitation pulses, and the

T1 relaxation. The lactate increased at the beginning of the

sequence because of the rapid pyruvate-to-lactate conversion,

reaching maximum at approximately t 5 14 6 4 s. The

decreasing lobe at the latter half of the lactate curve indicated

the timing where the combined loss from progressively

increasing flip angle and T1 relaxation exceeded the contribution from pyruvate conversion. Alanine was converted to a

much lower degree from pyruvate as a key step in gluconeogenesis pathway, which is governed by the alanine transaminase. The alanine time curve approached maximum at

t 5 20 6 6 s. The amount of alanine production was only a

fraction of lactate in the TRAMP tumor, whereas higher alanine level can be seen in the liver of both cancerous and

healthy animals.13

FIGURE 6 Prostate cancer patient 3D dynamic CS-EPSI data with volumetric coverage from base to apex of HP pyruvate and its conversion to lactate (signal summed through time is shown in the overlays). Spatial resolution 5 0.5 cm3

, temporal 5 2 s, 18 time points, starting 5 s after injection of HP

(37%) [1-13C]pyruvate. Region of high lactate conversion correlated with the bilateral biopsy-confirmed cancer.

FIGURE 7 (A) 18 timepoints for HP 13C-pyruvate from a single slice with bilateral biopsy-confirmed prostate cancer. The acquisition began !5 s

after injection. HP-13C pyruvate appears in the prostate at !10 s into the dynamic 3D CS-EPSI acquisition. This data demonstrates the feasibility of acquiring dynamically in three dimensions that covered the entire prostate with 2 s temporal resolution. (B) Temporal dynamics of 13C-lactate from the same data

and slice as in (A). Conversion to lactate in the bilateral cancer regions was observed at !20 s.

6 | Magnetic Resonance in Medicine CHEN ET AL.

极T代谢磁共振全球科研集锦

296

To investigate the feasibility of imaging larger FOVs in

vivo, we applied the 3D CS-EPSI sequence on 3 healthy

rats. Rapid pyruvate perfusion/uptake into the kidney was

observed since the beginning of the sequence, namely at

around !t 5 3 s post-injection. Appreciable amount of lactate exchange was detected as well (Figure 3). Lactate and

alanine production was also found in rat liver (data not

shown). These outcomes were highly consistent with the

previous observations on HP-13C rat studies.21,22

From TRAMP studies (Figure 4), the mean SNR,

summed across time, of total carbon was 69.2 6 28.4 for

tumor, 115.5 6 45.8 for vena cava, and 135.5 6 56.2 for kidneys with !25–35% polarization on dissolution. In the rat

scans (Figure 5), both FOV and voxel size were doubled

from mice. High SNR was found in both rat kidney (172.5 6

100.3) and liver (85.7 6 50.1). Such SNR was adequate for

both direct data visualization and dynamic modeling of metabolic interconversion. The mean size of TRAMP tumors,

2.2 cm3

, was approximately equivalent to 37 voxels (voxel

size 5 0.059 cm3

).

3.2 | Clinical phantom studies

Phantom studies were conducted using clinical setup,

sequence, and coils. The signal pattern on the built-in urea

phantom was found to be of higher homogeneity compared

FIGURE 8 The biopsy-proven Gleason 4 1 3 tumor in the patient’s right lateral midgland (red arrow) exhibited high lactate conversion following

HP pyruvate injection. (A) T2-FSE image showing the tumor voxel selected for the dynamic spectral plot in (B). Also shown is the ADC map where the

tumor region has substantially reduced ADC. (C) Dynamic curves (corrected for variable flip angle) are shown with far higher conversion to lactate in cancer compared to normal appearing regions. (D) Representative spectra for these regions at t 5 36 s. (E) Pyruvate-to-lactate conversion rate kPL parameter

map overlays showed high kPL on the opposite side (yellow arrows) as well, which was also confirmed as Gleason 4 1 3 prostate cancer by post-surgical

histopathology.

CHEN ET AL.

Magnetic Resonance in Medicine | 7

极T代谢磁共振全球科研集锦

297

to the ethylene glycol phantoms (Figure 3). Such signal profile probably resulted from the reduced sensitivity in regions

further away from the endorectal coil. The phantom data was

acquired with our 3D CS-EPSI sequence in tandem with the

new spectral–spatial RF pulses with reduced peak power and

duration. On both pyruvate and lactate bands, the phantom

dynamics showed good fitting agreement with the simulated

signal profile (Figure 2B). The apparent SNR of ethylene

glycol phantom was 16.1, and of urea phantom was 354 at

the final time point, where the 13C compounds were individually excited by the lactate excitation band with a mean flip

angle !40 8.

3.3 | Patient data

The subject of the human study testing the feasibility of the

new 3D CS-EPSI methods was a 66-year-old male patient

with biopsy-confirmed prostate cancer of stage T2c, with

PSA level of 6 ng/mL and PI-RADS score of 4/5. At radical

prostatectomy (RP), bilateral Gleason 4 1 3 was found at the

midgland of the prostate. Figure 6 depicts the pyruvate and

lactate area under curve (AUC) overlays in the prostate

region overlaid on T2-weighted reference scan. It demonstrated full gland coverage of this new sequence, from apex

to base, with a spatial resolution of 8 3 8 3 8 mm isotropic

(volumetric 5 0.5 cm3

). Whereas pyruvate intensities encompass the prostate gland as well as some surrounding vasculatures, regions of elevated conversion to lactate correlated

with cancer in the bilateral pathology data of this patient.

Figure 7 illustrates the temporal dynamics in a single

slice containing bilateral prostate cancer confirmed at RP.

HP-13C pyruvate is seen to perfuse into the vasculature surrounding the prostate, and the bolus entered prostate !t 5 10

s into the dynamic 3D acquisition. Rapid conversion to lactate in cancerous regions was observed to occur at !t 5 20 s.

The biopsy-proven Gleason 4 1 3 tumor in the right

peripheral zone exhibited more than 4-fold higher pyruvateto-lactate conversion (kPL) compared to normal-appearing

regions, reflecting high LDH enzymatic activity. The tumor

size was !1.5 cm3

, or 3 voxels (voxel size 5 0.5 cm3).

Figures 8C and 8D showed the HP-13C spectra and dynamics

in a representative voxel of this tumor versus a normalappearing region voxel. The dynamic curves were corrected

for progressive flip angles to show the estimated HP magnetization for kinetic modeling visualization purposes. The raw

signal of both pyruvate and lactate appears to monotonically

increase until the end of acquisition (Figure 8B) because of

the progressively increasing flip angles, whereas the corrected signal (i.e. the HP magnetization) shows the pyruvate

maximizing near !t 5 20 s post-injection and that of lactate

!30 s and decreasing toward the end (Figure 8C). The tumor

region also corresponds to darker region in T2-weighted FSE

image (Figure 8A, top, as encircled by the red box), and high

intensity in high b-value ADC maps (Figure 8A, bottom),

both of which exhibited good consistency with biopsy and

HP 13C findings. The bilateral midgland Gleason 4 1 3 cancer found at radical prostatectomy was also consistent with

the kPL map in Figure 8E, where the tumor in the right was

larger than the one in the left.

In the right midgland cancer, the apparent SNR calculated for pyruvate was 104, and for lactate, it was 10.7 at the

last time point. The mean SNR over the acquisition for pyruvate was 45.2 and 6.1 for lactate. The mean SNR over all

time points for total carbon was 51.3. In normal-appearing

regions, the mean SNR of total carbon in this patient was

48.2, comparable to the tumor region.

4 | DISCUSSION

Because HP 13C MR encodes chemical as well as spatial

information, this new molecular imaging technique allows

the simultaneous detection of multiple biologic compounds

and metabolic products with sensitivity enhancements of

>10,000-fold.23 This technique therefore presents the fields

of oncology and medical imaging with an opportunity to

improve our ability to investigate human disease and to ultimately translate these techniques into the clinic for more

individualized patient care.

The translation from animal to clinical HP-13C imaging

faces the challenges of larger imaging volume, reduced peak

RF power, and higher B1 inhomogeneity. To address these

challenges, specialized sequence modifications were developed including a low-power spectral spatial RF excitation,

“FID” acquisition mode and associated reconstruction methods, 3D coverage of the entire prostate with 0.5 cm3 spatial

and 2-s temporal resolution for prostate cancer patient imaging. This study was designed to determine and test the

sequence properties and signal behavior transitioning from

mouse studies to rats and, then to a human subject, with the

intent to optimize the performance and robustness of this

new 3D dynamic acquisition approach and then determine its

feasibility for imaging patients.

One major sequence modification made for the translation to human imaging was in the RF pulse design. The new

spectral-spatial RF pulse provide 67% savings in peak B1 by

means of relaxing the constraints on urea flip angle, which

accounts for the reduced peak RF power from clamshell

transmit coils in clinical setup compared to preclinical settings. The designed peak B1 was chosen to be !60% the

nominal maximum allowable power for the transmit coil to

provide sufficient headroom in transmit power allow for

varying coil loading when scanning different patients. The

1 ppm (!30 Hz) passband for each metabolite in this spectral–spatial pulse was reasonably wide to account for offresonance, which is typically <0.2 ppm (!6 Hz) for TRAMP

8 | Magnetic Resonance in Medicine CHEN ET AL.

极T代谢磁共振全球科研集锦

298

mice and <0.5 ppm (!15 Hz) for human prostate. The timebandwidth of 5 provides good compromise between pulse

duration, peak power, and transition profile sharpness (Figure

2A). The reduction in RF pulse width (8.9 to 6.3 ms) shortened the echo time, which may slightly improve SNR given

limited T2 and T"

2. The phantom signal curve agrees well

with the simulated signal profile, indicating that the excitation pulses can be confidently generated with the clinical coil

configuration. Removing the DSE refocusing pulses also

mitigated the issue of limited peak RF power. The mean

SNR were comparable in TRAMP prostate tumor in the

DSE-enabled sequence from prior studies and the new

sequence in this study (N 5 8, SNRDSE 5 61.6 6 43.6,

SNRFID 5 69.2 6 28.4, P > 0.3).

Imaging a larger subject using the 3D dynamic CS-EPSI

acquisition protocol provided a way to investigate the

sequence parameters and signal behavior because of increasing FOV. The rat scans in this study use a dual-tuned rat coil

that was !5 times the volume of mouse coil. The FOV and

voxel size were doubled both in the phase encode and EPSI

readout direction (in-plane resolution: mouse 5 3.3 mm,

rat 5 6.7 mm; axial resolution: mouse 5 5.4 mm, rat 5

10.8 mm), giving 4 3 voxel volume. The total injection dose

also increased by !6-fold. However, rat received lower

HP-13C dose per unit weight (!60% that of TRAMP). In

addition, the larger coil volumes inherently lead to decreased

sensitivity compared to mouse setup. The other sequence

parameters (e.g., flip angles, temporal resolution, and undersampling ratio) remained the same. Substantial pyruvate and

lactate were observed in rat kidneys just like in TRAMP

mice tumors. The SNR did not deteriorate in rat relative to

mouse (Figures 4 and 5). As such, transitioning to rat scans

revealed the key elements to scaling up the sequence, and

this protocol showed robustness in signal and image quality

with larger imaging subjects and coils.

A 2-s temporal resolution was chosen for human protocol

identical to that in TRAMP studies. In TRAMP scans, this

consistently provided >30 (apparent) SNR for both pyruvate

and lactate (!25–35% polarization), which translated into

<3% error in quantitative metabolism models for kPL evaluation based on simulations. For the clinical study, the pyruvate

SNR was similar to TRAMP (!30), while lactate SNR was

relatively lower (!6). Two primary sources of uncertainty

contribute to the apparent SNR of this 3D CS-EPSI acquisition—the data noise and CS reconstruction errors. Referring

to Figure 3 of the paper by Larson et al.,15 the reconstruction

error was <0.001 for SNR of 45 and <0.007 for SNR of 6.

Therefore, it can be concluded that the data noise was dominant source of error under both clinical and mouse scheme.

SNR improvement is theoretically possible through a longer

temporal resolution, because it results in effective signal

averaging and decrease of undersampling ratio. Nevertheless,

a longer temporal resolution can create temporal blurring and

ambiguity on timing of dynamic curve, which can negatively

impact quantitative modeling.

In the patient research, the acquisition began at t 5 5 s

post-injection, compared to the t 5 0 s in mouse studies,

whereas both shared the same 2-s temporal resolution. The

acquisition time window covered both bolus dynamics and

pyruvate-to-lactate conversion in human prostate cancer,

whereas the TRAMP scan focused more on the latter half of

the pharmacokinetics that mainly reflected pyruvate metabolism. A main reason for the 5-s delay in human studies was

to prevent hyperpolarized magnetization in the intravenous

tubing and arm being excited by the clamshell transmit coils.

Because at the end of the injection, the bolus could still be

traversing through the antecubital vein, which is typically

located inside the “hot” zones of the transmitter, the delay

allows hyperpolarized bolus to perfuse into tumor region

before RF excitation.

In clinical hyperpolarized 13C imaging, because of the

absence of arterial coverage or perfusion markers, it is more

challenging to account for pharmacokinetic parameters such

as circulation and AIF. In contrast, because the bolus delivery was more rapid in small animals (e.g., TRAMP mice),

and a reference AIF was relatively easy to estimate using the

HP 13C urea through arterial voxels according to a preclinical

co-polarized imaging study of TRAMP tumor,1 the sequence

can be configured to put more focus on net metabolism by

acquiring at a longer (15 s) delayed window.1,24 Additionally, the circulation and bolus delivery is generally slower in

human versus small animals. Therefore, carefully selecting

an acquisition window that both accounts for pyruvate infusion and pyruvate–lactate conversion benefits the clinical

quantitation of prostate tumor metabolism.

Prostate cancer is a major health concern in the United

States with >160,000 new cases per year and >26,000

deaths.25 Because of increased screening using serum prostate specific antigen (PSA) and extended-template transrectal

ultrasound (TRUS) guided biopsies, patients with prostate

cancer are being identified at an earlier and potentially more

treatable stage. Unfortunately, the aggressiveness of individual tumors cannot be predicted with great confidence in individual patients using currently available clinical and imaging

prognostic data.26–30 Preliminary data strongly indicate that

hyperpolarized 13C-pyruvate MRI using DNP has the potential to dramatically improve prostate cancer clinical management. In transgenic prostate cancer mouse models, this

method demonstrated the unprecedented ability to separate

early stage (low-grade) tumors from late stage (high-grade)

cancer based on this metabolic parameter (conversion

through the LDH-catalyzed pyruvate metabolism).2 Higher

grade prostate cancers, both in transgenic models and human

biopsies, have demonstrated several fold increases in LDH

expression.2,31 No other imaging method has demonstrated

this ability to differentiate low grade, clinically insignificant

CHEN ET AL.

Magnetic Resonance in Medicine | 9

极T代谢磁共振全球科研集锦

299

prostate cancers that are the majority of cases from highgrade disease that kills >26,000 Americans per year. The

Phase 1 clinical trial of HP 13C-pyruvate in prostate cancer

patients demonstrated feasibility and safety through this first

human study of hyperpolarized MR metabolic imaging12

using single-slice 1D dynamic, 2D dynamic, and 3D single

time point acquisitions. This clinical trial indicated the potential to characterize the extent and aggressiveness of prostate

cancer in individual subjects to ultimately benefit clinical

treatment decisions and to monitor treatment response. However, the acquisition methods used in that trial did not provide simultaneously the required spatial and dynamic

temporal resolution necessary for optimal HP 13C MR clinical research studies. Development and translation of 3D

dynamic HP 13C MRSI offers a new method to quantitatively

analyze metabolism in human prostate cancer throughout the

gland. In this study, the human 3D MRSI acquisition demonstrated the ability to obtain dynamic information on the conversion of [1-13C]pyruvate to [1-13C]lactate that is catalyzed

by lactate dehydrogenase (LDH) that is upregulated in prostate cancer. Adequate SNR and temporal resolution enabled

the calculation of kPL maps with a spatial resolution of

0.5 cm3

. This supports the use of this 3D acquisition

approach in future studies to investigate in sufficiently large

patient populations prostate cancer aggressiveness and

response to therapy. Importantly, the capability to image biochemical processes and visualize diseases with high spatiotemporal resolution opens the door to many potential HP 13C

translational and research applications. For instance, high

lactate/pyruvate ratios were detected in various cancer types

such as xenografts of human brain tumor,32 renal carcinoma

cells,33 and breast tumor xenografts.7 Modulation of lactate

production was found in tumors subjected to chemo34 and

targeted therapy.35 Besides cancer, HP 13C imaging has been

used to study kidney urea transporters,36 diabetes and gluconeogenesis,37 cardiac diseases,38 and neurodevelopment.39

5 | CONCLUSIONS

This new 3D dynamic MRSI acquisition method incorporating new spectral–spatial RF pulses, “FID” readout, and

modified CS reconstruction addressed the challenges of

larger imaging volumes and reduced available peak RF

power required for human studies. Scalability in acquisition,

reconstruction, and quantitation methods was demonstrated

by the satisfactory image quality, SNR, and apparent kinetic

rate constants between cancer–normal during the transition

from mice to human patient studies. The results demonstrate

the feasibility to characterize prostate cancer metabolism in

the clinical setting using this new 3D dynamic HP MR technique to quantify and image the kinetic rate constant, kPL, of

the conversion of [1-13C]pyruvate to [1-13C]lactate that has

been shown to be increased in prostate cancer.1,2,40,41

ACKNOWLEDGMENTS

We would like to thank Dr. Renuka Sriram for the helpful

discussions.

ORCID

Peder E.Z. Larson http://orcid.org/0000-0003-4183-3634

REFERENCES

[1] Chen HY, Larson PEZ, Bok RA, et al. Assessing prostate cancer

aggressiveness with hyperpolarized dual-agent 3D dynamic

imaging of metabolism and perfusion. Cancer Res. 2017;77:

3207-3216.

[2] Albers MJ, Bok R, Chen AP, et al. Hyperpolarized 13C lactate,

pyruvate, and alanine: noninvasive biomarkers for prostate cancer detection and grading. Cancer Res. 2008;68:8607-8615.

[3] Brindle K. New approaches for imaging tumour responses to

treatment. Nat Rev Cancer. 2008;8:94-107.

[4] Day SE, Kettunen MI, Gallagher FA, et al. Detecting tumor

response to treatment using hyperpolarized C-13 magnetic resonance imaging and spectroscopy. Nat Med. 2007;13:1382-1387.

[5] Golman K, in ’t Zandt R, Thaning M. Real-time metabolic imaging. Proc Natl Acad Sci U S A. 2006;103:11270-11275.

[6] Golman K, Olsson LE, Axelsson O, Mansson S, Karlsson M,

Petersson J. Molecular imaging using hyperpolarized 13C. Br J

Radiol. 2003;76:S118-S127.

[7] Golman K, Zandt RI, Lerche M, Pehrson R, Ardenkjaer-Larsen

JH. Metabolic imaging by hyperpolarized 13C magnetic resonance imaging for in vivo tumor diagnosis. Cancer Res. 2006;

66:108551-108560.

[8] Ishii M, Emami K, Kadlecek S, et al. Hyperpolarized 13C MRI

of the pulmonary vasculature and parenchyma. Magn Reson

Med. 2007;57:459-463.

[9] Kurhanewicz J, Vigneron DB, Brindle K, et al. Analysis of cancer metabolism by imaging hyperpolarized nuclei: prospects for

translation to clinical research. Neoplasia. 2011;13:81-97.

[10] Merritt ME, Harrison C, Storey C, Jeffrey FM, Sherry AD, Malloy CR. Hyperpolarized C-13 allows a direct measure of flux

through a single enzyme-catalyzed step by NMR. Proc Natl

Acad Sci U S A. 2007;104:19773-19777.

[11] Nelson SJ, Vigneron D, Kurhanewicz J, Chen A, Bok R, Hurd

R. DNP-hyperpolarized C-13 magnetic resonance metabolic

imaging for cancer applications. Appl Magn Reson. 2008;34:

533-544.

[12] Nelson SJ, Kurhanewicz J, Vigneron DB, et al. Metabolic imaging of patients with prostate cancer using hyperpolarized 1-C-13]

pyruvate. Sci Transl Med. 2013;5:198ra108.

[13] Chen AP, Albers MJ, Cunningham CH, et al. Hyperpolarized c13 spectroscopic imaging of the TRAMP mouse at 3T - initial

experience. Magn Reson Med. 2007;58:1099-1106.

[14] Hu S, Lustig M, Balakrishnan A, et al. 3D Compressed sensing

for highly accelerated hyperpolarized C-13 MRSI with in vivo

10 | Magnetic Resonance in Medicine CHEN ET AL.

极T代谢磁共振全球科研集锦

300

applications to transgenic mouse models of cancer. Magn Reson

Med. 2010;63:312-321.

[15] Larson PEZ, Hu S, Lustig M, et al. Fast dynamic 3D MR spectroscopic imaging with compressed sensing and multiband excitation pulses for hyperpolarized C-13 studies. Magn Reson Med.

2011;65:610-619.

[16] Xing Y, Reed GD, Pauly JM, Kerr AB, Larson PEZ. Optimal

variable flip angle schemes for dynamic acquisition of exchanging hyperpolarized substrates. J Magn Reson. 2013;234:75-81.

[17] Larson PEZ, Bok R, Kerr AB, et al. Investigation of tumor

hyperpolarized [1-C-13]-pyruvate dynamics using time-resolved

multiband RF excitation echo-planar MRSI. Magn Reson Med.

2010;63:582-591.

[18] Crane JC, Olson MP, Nelson SJ. SIVIC: open-source, standardsbased software for DICOM MR spectroscopy workflows. Int J

Biomed Imaging. 2013;2013:169526.

[19] Zierhut ML, Yen YF, Chen AP, et al. Kinetic modeling of

hyperpolarized 13C1-pyruvate metabolism in normal rats and

TRAMP mice. J Magn Reson. 2010;202:85-92.

[20] Bahrami N, Swisher CL, Von Morze C, Vigneron DB, Larson

PE. Kinetic and perfusion modeling of hyperpolarized (13)C

pyruvate and urea in cancer with arbitrary RF flip angles. Quant

Imaging Med Surg. 2014;4:24-32.

[21] Hu S, Larson PEZ, Vancriekinge M, et al. Rapid sequential

injections of hyperpolarized [1-C-13]pyruvate in vivo using a

sub-kelvin, multi-sample DNP polarizer. Magn Reson Imaging.

2013;31:490-496.

[22] Kohler SJ, Yen Y, Wolber J, et al. In vivo (13)carbon metabolic

imaging at 3T with hyperpolarized C-13-1-pyruvate. Magn

Reson Med. 2007;58:65-69.

[23] 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 U S A. 2003;100:10158-10163.

[24] von Morze C, Larson PEZ, Hu S, et al. Imaging of blood flow

using hyperpolarized [C-13] urea in preclinical cancer models.

J Magn Reson Imaging. 2011;33:692-697.

[25] American Cancer Society. Cancer facts & figures 2017. Atlanta:

American Cancer Society; 2017.

[26] Bill-Axelson A, Holmberg L, Ruutu M, et al. Radical prostatectomy versus watchful waiting in early prostate cancer. New Engl

J Med. 2005;352:1977-1984.

[27] Carroll PR. Early stage prostate cancer - do we have a problem

with over-detection, overtreatment or both? J Urol. 2005;173:

1061-1062.

[28] Draisma G, Boer R, Otto SJ, et al. Lead times and overdetection

due to prostate-specific antigen screening: estimates from the

European randomized study of screening for prostate cancer.

J Natl Cancer Inst. 2003;95:868-878.

[29] Etzioni R, Penson DF, Legler JM, et al. Overdiagnosis due to

prostate-specific antigen screening: lessons from US prostate

cancer incidence trends. J Natl Cancer Inst. 2002;94:981-990.

[30] Johansson JE, Andren O, Andersson SO, et al. Natural history

of early, localized prostate cancer. JAMA. 2004;291:2713-2719.

[31] Tessem MB, Swanson MG, Keshari KR, et al. Evaluation of lactate and alanine as metabolic biomarkers of prostate cancer using

H-1 HR-MAS spectroscopy of biopsy tissues. Magn Reson Med.

2008;60:510-516.

[32] Park I, Larson PEZ, Zierhut ML, et al. Hyperpolarized C-13

magnetic resonance metabolic imaging: application to brain

tumors. Neuro Oncol. 2010;12:133-144.

[33] Keshari KR, Sriram R, Van Criekinge M, et al. Metabolic

reprogramming and validation of hyperpolarized 13C lactate as a

prostate cancer biomarker using a human prostate tissue slice

culture bioreactor. Prostate. 2013;73:1171-1181.

[34] Park I, Bok R, Ozawa T, et al. Detection of early response to

temozolomide treatment in brain tumors using hyperpolarized

13C MR metabolic imaging. J Magn Reson Imaging. 2011;33:

1284-1290.

[35] Lodi A, Woods SM, Ronen SM. Treatment with the MEK inhibitor U0126 induces decreased hyperpolarized pyruvate to lactate

conversion in breast, but not prostate, cancer cells. NMR

Biomed. 2013;26:299-306.

[36] von Morze C, Larson PEZ, Hu S, et al. Imaging of blood flow

using hyperpolarized [13C]Urea in preclinical cancer models.

J Magn Reson Imaging. 2011;33:692-697.

[37] Hu S, Chen AP, Zierhut ML, et al. In vivo Carbon-13 dynamic

MRS and MRSI of normal and fasted rat liver with hyperpolarized C-13-pyruvate. Mol Imaging Biol. 2009;11:399-407.

[38] Schroeder MA, Lau AZ, Chen AP, et al. Hyperpolarized (13)C

magnetic resonance reveals early- and late-onset changes to in

vivo pyruvate metabolism in the failing heart. Eur J Heart Fail.

2013;15:130-140.

[39] Chen YR, Kim H, Bok R, et al. Pyruvate to lactate metabolic

changes during neurodevelopment measured dynamically using

hyperpolarized C-13 imaging in juvenile murine brain. Dev Neurosci. 2016;38:34-40.

[40] Keshari KR, Sriram R, Van Criekinge M, et al. Metabolic

reprogramming and validation of hyperpolarized C-13 lactate as

a prostate cancer biomarker using a human prostate tissue slice

culture bioreactor. Prostate. 2013;73:1171-1181.

[41] Wilson DM, Kurhanewicz J. Hyperpolarized 13C MR for molecular imaging of prostate cancer. J Nucl Med. 2014;55:1567-1572.

SUPPORTING INFORMATION

Additional Supporting Information may be found in the

online version of this article.

How to cite this article: Chen H-Y, Larson PEZ, Gordon JW, et al. Technique development of 3D dynamic

CS-EPSI for hyperpolarized 13C pyruvate MR molecular imaging of human prostate cancer. Magn Reson

Med. 2018;00:1–11. https://doi.org/10.1002/mrm.

27179

CHEN ET AL.

Magnetic Resonance in Medicine | 11

极T代谢磁共振全球科研集锦

301

14

8.1 Supplementary Figures and Tables

Label # Region Label # Region

21 L superior frontal gyrus 65 L inferior occipital gyrus

22 R superior frontal gyrus 66 R inferior occipital gyrus

23 L middle frontal gyrus 67 L cuneus

24 R middle frontal gyrus 68 R cuneus

25 L inferior frontal gyrus 81 L superior temporal gyrus

26 R inferior frontal gyrus 82 R superior temporal gyrus

27 L precentral gyrus 83 L middle temporal gyrus

28 R precentral gyrus 84 R middle temporal gyrus

29 L middle orbitofrontal gyrus 85 L inferior temporal gyrus

30 R middle orbitofrontal gyrus 86 R inferior temporal gyrus

31 L lateral orbitofrontal gyrus 87 L parahippocampal gyrus

32 R lateral orbitofrontal gyrus 88 R parahippocampal gyrus

33 L gyrus rectus 89 L lingual gyrus

34 R gyrus rectus 90 R lingual gyrus

41 L postcentral gyrus 91 L fusiform gyrus

42 R postcentral gyrus 92 R fusiform gyrus

43 L superior parietal gyrus 101 L insular cortex

44 R superior parietal gyrus 102 R insular cortex

45 L supramarginal gyrus 121 L cingulate gyrus

46 R supramarginal gyrus 122 R cingulate gyrus

47 L angular gyrus 161 L caudate

48 R angular gyrus 162 R caudate

49 L precuneus 163 L putamen

50 R precuneus 164 R putamen

61 L superior occipital gyrus 165 L hippocampus

62 R superior occipital gyrus 166 R hippocampus

63 L middle occipital gyrus 181 Cerebellum

64 R middle occipital gyrus 182 Brainstem

Table 1: (Supplementary) The numerical labels used in the LPBA40 atlas and the corresponding brain regions.

极T代谢磁共振全球科研集锦

302

15

A

B

Figure 5: (Supplementary) (A) Average (± SD) of segmented LPBA40 regional volumes from

14 subjects. Red line indicates isotropic 1.5 cm voxel volume. Regions that were smaller

than the size of a single voxel were: L/R caudate, L/R putamen, L/R gyrus rectus, L/R

hippocampus. (B) Volume z -scores plotted vs. the LPBA40 atlas region labels, each colour

showing a different subject.

极T代谢磁共振全球科研集锦

303

16

A

B

C

-4 -3 -2 -1 0123

-2

-1

0

1

2

3

4

5

6

Pyruvate z-score

Lactate z-score

-4 -3 -2 -1 0123

-3

-2

-1

0

1

2

3

4

Bicarbonate z-score

Lactate z-score

D

Figure 6: (Supplementary) Metabolite signal ratios vs. LPBA40 atlas region number

(N=14). (a) lactate-to-pyruvate and (b) lactate-to-bicarbonate ratios, calculated by computing the area-under-the-curve from the timecourse for each metabolite and taking the

ratio between these areas. Each colour represents a different subject. (c) For reference, the

pyruvate z -scores are plotted vs. the LPBA 40 regions, showing a consistent pattern across

subjects which is different from the lactate and bicarbonate patterns. (d) Scatter plots for

regional pyruvate z-scores vs. lactate z-scores (left) and bicarbonate z-scores vs. lactate

z-scores for all subjects.

极T代谢磁共振全球科研集锦

304

17

0 50 0

10

20

30

40

50

60

time [s]

signa [a.u.]

Subject 4

AUCR =0.37

0 50 0

10

20

30

40

50

60

time [s]

Subject 6

AUCR =0.22

0 50 0

10

20

30

40

50

60

70

80

90

time [s]

Subject 7

AUCR =0.26

0 50 0

5

10

15

20

25

time [s]

Subject 8

AUCR =0.30

0 50 100 0

20

40

60

80

100

120

140

160

time [s]

Subject 9

AUCR =0.18

lactate

pyruvate

a

b

Figure 7: (Supplementary) Venous drainage signal measured in the right jugular. (a) Representative region-of-interest (ROI) drawn in the right jugular of subject #7, with 13C-lactate

signal shown in the colour overlay. The 13C-pyruvate and 13C-lactate signal vs. time in the

right jugular ROI for a subset of the subjects (N=5), with lactate-to-pyruvate area-underthe-curve ratios (AUCR) for each subject.

极T代谢磁共振全球科研集锦

305

18

7 Author Contributions

CYL, BJG, APC, CHC developed the data acquisition methods, CYL and RE operated the

MRI scanner, CYL and BJG did the image reconstruction and image analysis, HS and KAC

assessed the subjects and performed the 13C-pyruvate injections, RE loaded the injector,

WJP performed compounding and pharmacy release of the 13C-pyruvate doses, APC and

CHC built the 13C head coil, CH and SEB interpreted the images. All authors critically

reviewed the manuscript.

8 Acknowledgements

Funding support from the Canadian Cancer Society grant 705246 and Canadian Institutes

of Health Research grant PJT-152928.

极T代谢磁共振全球科研集锦

306

19

REFERENCES

1. Luc Pellerin and Pierre J Magistretti. Glutamate uptake into astrocytes stimulates aerobic glycolysis: a mechanism coupling neuronal activity to glucose utilization. Proceedings

of the National Academy of Sciences, 91(22):10625–10629, 1994.

2. Luc Pellerin, Giovanni Pellegri, Philippe G Bittar, Yves Charnay, Constantin Bouras,

Jean-Luc Martin, Nephi Stella, and Pierre J Magistretti. Evidence supporting the existence of an activity-dependent astrocyte-neuron lactate shuttle. Developmental neuroscience, 20(4-5):291–299, 1998.

3. Mireille B´elanger, Igor Allaman, and Pierre J Magistretti. Brain energy metabolism:

focus on astrocyte-neuron metabolic cooperation. Cell metabolism, 14(6):724–738, 2011.

4. Pierre J Magistretti and Igor Allaman. Lactate in the brain: from metabolic end-product

to signalling molecule. Nature Reviews Neuroscience, 19(4):235, 2018.

5. Akinobu Suzuki, Sarah A Stern, Ozlem Bozdagi, George W Huntley, Ruth H Walker,

Pierre J Magistretti, and Cristina M Alberini. Astrocyte-neuron lactate transport is

required for long-term memory formation. Cell, 144(5):810–823, 2011.

6. Mauro DiNuzzo and Maiken Nedergaard. Brain energetics during the sleep–wake cycle.

Current opinion in neurobiology, 47:65–72, 2017.

7. Gerald A Dienel. Lack of appropriate stoichiometry: Strong evidence against an energetically important astrocyte–neuron lactate shuttle in brain. Journal of neuroscience

research, 95(11):2103–2125, 2017.

8. Carlos Manlio D´ıaz-Garc´ıa, Rebecca Mongeon, Carolina Lahmann, Dorothy Koveal,

Hannah Zucker, and Gary Yellen. Neuronal stimulation triggers neuronal glycolysis and

not lactate uptake. Cell metabolism, 26(2):361–374, 2017.

9. S Neil Vaishnavi, Andrei G Vlassenko, Melissa M Rundle, Abraham Z Snyder, Mark A

Mintun, and Marcus E Raichle. Regional aerobic glycolysis in the human brain. Proceedings of the National Academy of Sciences, 107(41):17757–17762, 2010.

10. Manu S Goyal, Michael Hawrylycz, Jeremy A Miller, Abraham Z Snyder, and Marcus E

Raichle. Aerobic glycolysis in the human brain is associated with development and

neotenous gene expression. Cell metabolism, 19(1):49–57, 2014.

11. Manu S Goyal, Andrei G Vlassenko, Tyler M Blazey, Yi Su, Lars E Couture, Tony J

Durbin, Randall J Bateman, Tammie L-S Benzinger, John C Morris, and Marcus E

Raichle. Loss of brain aerobic glycolysis in normal human aging. Cell metabolism,

26(2):353–360, 2017.

12. Tyler Blazey, Abraham Z Snyder, Yi Su, Manu S Goyal, John J Lee, Andrei G Vlassenko,

Ana Maria Arbel´aez, and Marcus E Raichle. Quantitative positron emission tomography

reveals regional differences in aerobic glycolysis within the human brain. Journal of

Cerebral Blood Flow & Metabolism, page 0271678X18767005, 2018.

极T代谢磁共振全球科研集锦

307

20

13. J. H. Ardenkjaer-Larson, B. Fridlund, A. Gram, G. Hansson, L. Hansson, M. H. Lerche,

R. Servin, M. Thaning, and K. Golman. Increase in signal-to-noise ration of > 10,000

times in liquid-state NMR. Proc Natl Acad Sci USA, 100:10158–10163, 2003.

14. Sarah J Nelson, John Kurhanewicz, Daniel B Vigneron, Peder EZ Larson, Andrea L

Harzstark, Marcus Ferrone, Mark van Criekinge, Jose W Chang, Robert Bok, Ilwoo

Park, et al. Metabolic imaging of patients with prostate cancer using hyperpolarized

[1-13c] pyruvate. Science translational medicine, 5(198):198ra108–198ra108, 2013.

15. Charles H Cunningham, Justin Y Lau, Albert P Chen, Benjamin J Geraghty, William J

Perks, Idan Roifman, Graham A Wright, and Kim A Connelly. Hyperpolarized 13c

metabolic mri of the human heart: Initial experience. Circulation Research, pages

CIRCRESAHA–116, 2016.

16. Vesselin Z Miloushev, Kristin L Granlund, Rostislav Boltyanskiy, Serge K Lyashchenko,

Lisa M DeAngelis, Ingo K Mellinghoff, Cameron W Brennan, Vivian Tabar, T Jonathan

Yang, Andrei I Holodny, et al. Metabolic imaging of the human brain with hyperpolarized

13c pyruvate demonstrates 13c lactate production in brain tumor patients. Cancer

research, 78(14):3755–3760, 2018.

17. C.H. Cunningham, A.P. Chen, M. Lustig, B.A. Hargreaves, J. Lupo, D. Xu,

J. Kurhanewicz, R.E. Hurd, S.J. Pauly, J.M. an d Nelson, and D.B. Vigneron. Pulse sequence for dynamic volumetric imaging of hyperpolarized m etabolic products. J. Magn.

Reson., 193:139–146, Jul 2008.

18. Benjamin J Geraghty, Justin YC Lau, Albert P Chen, and Charles H Cunningham. Dualecho epi sequence for integrated distortion correction in 3d time-resolved hyperpolarized

13c mri. Magnetic resonance in medicine, 79(2):643–653, 2018.

19. David E Rex, Jeffrey Q Ma, and Arthur W Toga. The loni pipeline processing environment. Neuroimage, 19(3):1033–1048, 2003.

20. David W Shattuck, Mubeena Mirza, Vitria Adisetiyo, Cornelius Hojatkashani, Georges

Salamon, Katherine L Narr, Russell A Poldrack, Robert M Bilder, and Arthur W Toga.

Construction of a 3d probabilistic atlas of human cortical structures. Neuroimage,

39(3):1064–1080, 2008.

21. James T Grist, Mary A McLean, Frank Riemer, Rolf F Schulte, Surrin S Deen, Fulvio

Zaccagna, Ramona Woitek, Charlie J Daniels, Joshua D Kaggie, Tomasz Matyz, et al.

Quantifying normal human brain metabolism using hyperpolarized [1–13c] pyruvate and

magnetic resonance imaging. NeuroImage, 2019.

22. Jeremy W Gordon, Hsin-Yu Chen, Adam Autry, Ilwoo Park, Mark Van Criekinge,

Daniele Mammoli, Eugene Milshteyn, Robert Bok, Duan Xu, Yan Li, et al. Translation of carbon-13 epi for hyperpolarized mr molecular imaging of prostate and brain

cancer patients. Magnetic resonance in medicine, 81(4):2702–2709, 2019.

极T代谢磁共振全球科研集锦

308

21

23. James Prichard, Douglas Rothman, Edward Novotny, Ognen Petroff, Takeo Kuwabara,

Malcolm Avison, Alistair Howseman, Christopher Hanstock, and Robert Shulman. Lactate rise detected by 1h nmr in human visual cortex during physiologic stimulation.

Proceedings of the National Academy of Sciences of the United States of America,

88(13):5829, 1991.

24. Klaus-Dietmar Merboldt, Harald Bruhn, Wolfgang Hanicke, Thomas Michaelis, and

Jens Frahm. Decrease of glucose in the human visual cortex during photic stimulation.

Magnetic resonance in medicine, 25(1):187–194, 1992.

25. Silvia Mangia, Ivan Tk´aˇc, Rolf Gruetter, Pierre-Francois Van De Moortele, Federico

Giove, Bruno Maraviglia, and Kˆamil U˘gurbil. Sensitivity of single-voxel 1h-mrs in investigating the metabolism of the activated human visual cortex at 7 t. Magnetic resonance

imaging, 24(4):343–348, 2006.

26. Yury Koush, Robin A de Graaf, Lihong Jiang, Douglas L Rothman, and Fahmeed Hyder.

Functional mrs with j-edited lactate in human motor cortex at 4 t. NeuroImage, 184:101–

108, 2019.

27. Fahmeed Hyder, Peter Herman, Christopher J Bailey, Arne Møller, Ronen Globinsky,

Robert K Fulbright, Douglas L Rothman, and Albert Gjedde. Uniform distributions

of glucose oxidation and oxygen extraction in gray matter of normal human brain: no

evidence of regional differences of aerobic glycolysis. Journal of Cerebral Blood Flow &

Metabolism, 36(5):903–916, 2016.

28. MN Subhash, SK Shankar, and BSS RAMA RAO. Regional distribution of lactate

dehydrogenase isoenzymes in human brain. Current Science, pages 770–772, 1986.

29. Jocelyn D Laughton, Philippe Bittar, Yves Charnay, Luc Pellerin, Enik¨o Kovari, Pierre J

Magistretti, and Constantin Bouras. Metabolic compartmentalization in the human

cortex and hippocampus: evidence for a cell-and region-specific localization of lactate

dehydrogenase 5 and pyruvate dehydrogenase. BMC neuroscience, 8(1):35, 2007.

30. Xia Liang, Qihong Zou, Yong He, and Yihong Yang. Coupling of functional connectivity

and regional cerebral blood flow reveals a physiological basis for network hubs of the

human brain. Proceedings of the National Academy of Sciences, 110(5):1929–1934, 2013.

31. Patrizia Mecocci, Usha MacGarvey, Allan E Kaufman, Deborah Koontz, John M

Shoffner, Douglas C Wallace, and M Flint Beal. Oxidative damage to mitochondrial

dna shows marked age-dependent increases in human brain. Annals of Neurology: Official Journal of the American Neurological Association and the Child Neurology Society,

34(4):609–616, 1993.

32. Zuyao Y Shan, Andrew J Leiker, Arzu Onar-Thomas, Yimei Li, Tianshu Feng, Wilburn E

Reddick, David C Reutens, and Barry L Shulkin. Cerebral glucose metabolism on

positron emission tomography of children. Human brain mapping, 35(5):2297–2309,

2014.

极T代谢磁共振全球科研集锦

309

22

33. Sam E Day, Mikko I Kettunen, Ferdia A Gallagher, De-En Hu, Mathilde Lerche, Jan

Wolber, Klaes Golman, Jan Henrik Ardenkjaer-Larsen, and Kevin M Brindle. Detecting

tumor response to treatment using hyperpolarized 13 c magnetic resonance imaging and

spectroscopy. Nature medicine, 13(11):1382, 2007.

极T代谢磁共振全球科研集锦

310

ࢫႀሞፌ

ג极ࣅ磁共振ׯၟरຍ๟ႜᄽڦ݀ቛඤۅLj全球௅౎݀՗ڦ࿔ቤࡗגଇӥೊLjপኹణമ

݀՗࿔ቤᅙࡗג 2000 ೊLjഄዐփ݄ NatureĂScience ۥڪप೺਽ݴߛڦ࿔ቤăᆯᇀೊޗᆶ၌Lj

Ԩກৈৈ๭୤କג极ࣅ代谢ׯၟڦ 20 ܠೊඤۅ࿔ቤăٗ࿔၅ڦለ܁ዐLj࿢்փৈକ঴କᆶ

ג࠲极ࣅरຍ全球科研ڦ৊ቛLjཞ้ᄺሺेକ࿢்ܔᇀ࿄ઠᆖၟ݀ቛڦ႑႐ăג极ࣅ磁共振

रຍ๟ܔ࿄ઠ代谢ፇბႎڦາୟഔڱă

极 T 代谢磁共振ਸഔକํ้代谢ᆖၟڦႎೊቤLjӝ࿢்၄ሞፌമᄂڦ代谢ፇბ研৯ᇑణ

മፌံ৊ڦ磁共振ׯၟरຍ঳ࢇഐઠLjඁ༑൱዗ୀLj႐సदթLj代૙ૌदթڪዘۇٴदթڦ৛

ጚᅅଐڦ঴ਦݛӄ , ඗ܸ极 T 代谢磁共振ڦᆌᆩৈৈߑߑഐօLj࿄ઠڦႎݛڦၠࢅ݀ቛ࣏ᆶ

ځᇀ࿄ઠ߳࿋ጆॆڦ共ཞ౮૰ࢅኸڞă

ቛ݀ბ科ڦ代谢ፇბܸक़Lj้ڦ౎କৎӥࡗLjঢ়؜༵ڦ౒߁ፇბڟဆݴ࿿代谢ڦཥدٗ

ඐኻᆶփڟܾๆ౎ڦ้क़ăᅅბ代谢ፇბڦბ科݀ቛৎ౎ઠᄺᅃ኱ྷජᅅბڦ݀ቛփ܏മ৊Lj

ࢅጚඓႠڦ݆ݛമᄂLjܠ৊ံܠरຍᆶ࠶փ඗ܸጚᅅბă৛ڟᅅბLjምࣅገڟბิ࿿ဣཥٗ

੗ዘްႠᆦᇺ๟ഄጽၠଣڦضਦۨႠᅺ໎ăᅺُႴᄲ࿢்ᅃഐᇑ全球研৯ኁ்փ܏ڦศࢇࣅ

ፕLjፔڟ੗໓ႠࢅՔጚࣅLj研৯঳ࡕ੗ዘްLjג极ࣅरຍ֍ీሞ৛ጚᅅბڦଶᇘۇٴᆶፕྺă

ፌࢫ GE ᅅଐ Ljࣷփ܏ڦ৆ໜ้代ڦօ݃Ljᅃ኱ዂ૰ᇀ঴ਦᅅბଶᇘڦవ༶Ljགցႜᄽ

रຍ੣ӣăփৈॽፌႎፌമᄂڦरຍڟټዐࡔLj࣏ॽփ܏߀৊ࢅਸྊLjྺዐڦࡔ磁共振ڦ݀

ቛࠋ၅ᅃݻ૰ଉă

GE ᅅଐ磁共振ׂ೗๚ᄽև

© 2022 General Electric Company保留所有权利。

GE, GE Monogram是通用电气公司（General Electric Company）的商标。

PromoMats No. : JB00587XC

* Internal Use Only