Comparison of1H Blood Oxygen Level–Dependent
(BOLD) and19F MRI to Investigate Tumor Oxygenation
Dawen Zhao, Lan Jiang, Eric W. Hahn, and Ralph P. Mason*
Fluorine-19 [19F] MRI oximetry and1H blood oxygen level–de-
pendent (BOLD) MRI were used to investigate tumor oxygen-
ation in rat breast 13762NF carcinomas, and correlations be-
tween the techniques were examined. A range of tissue oxygen
partial pressure (pO2) values was found in the nine tumors while
the anesthetized rats breathed air, with individual tumor pO2
ranging from a mean of 1 to 36 torr and hypoxic fraction (HF10)
(<10 torr) ranging from 0% to 75%, indicating a large intra- and
intertumor heterogeneity. Breathing oxygen produced signifi-
cant increase in tumor pO2(mean ?pO2? 50 torr) and decrease
in HF10(P < 0.01).1H BOLD MRI observed using a spin echo-
planar imaging (EPI) sequence revealed a heterogeneous re-
sponse and significant increase in mean tumor signal intensity
(SI) (?SI ? 7%, P < 0.01). R*2measured by multigradient-echo
(MGRE) MRI decreased significantly in response to oxygen
(mean ?R*2? –4 s–1; P < 0.05). A significant correlation was
found between changes in mean tumor pO2 and mean EPI
BOLD ?SI accompanying oxygen breathing (r2> 0.7, P < 0.001).
Our results suggest that BOLD MRI provides information about
tumor oxygenation and may be useful to predict pO2changes
accompanying interventions. Significantly, the magnitude of the
BOLD response appears to be predictive for residual tumor
HFs.Magn Reson Med 62:357–364, 2009. © 2009 Wiley-Liss,
Key words: tumor oxygenation; BOLD;19F MRI; transverse re-
laxation rate R*2; oxygen; hexafluorobenzene
Tumor oxygenation has been widely recognized as a po-
tent factor influencing tumor response to various thera-
pies, especially radiotherapy, and hypoxia appears to pro-
mote malignant progression and metastasis of tumor (1).
Given the importance of tumor oxygenation, many mea-
surement techniques have been developed (1,2). While
each method has specific attributes, many are highly inva-
sive or cannot be applied to longitudinal studies of oxygen
Blood oxygen level–dependent (BOLD) MRI, exten-
sively used in studying brain function, is increasingly
being applied to noninvasively assess blood oxygenation
and vascular function in tumors (3–12). The underlying
rationale is that the paramagnetic deoxyhemoglobin cre-
ates microscopic field gradients, which enhance the trans-
verse relaxation rate, R*2, of water protons in blood and in
the tissue adjacent to blood vessels. Decrease in deoxyhe-
moglobin concentration leads to a decreased R*2, and thus,
to an increased signal intensity (SI) in T*2-weighted MRI
(13,14). Gradient-recalled echo (GRE), or spin echo-planar
imaging (EPI), is sensitive to changes in R*2. However,
BOLD contrast is also influenced by other factors such as
blood flow, blood volume, and vascular architecture
(3,15). Several recent studies have attempted to correlate
BOLD MRI with tissue oxygen partial pressure (pO2) mea-
sured by various techniques; notably, oxygen electrodes,
oxygen-sensitive fiber optic probes, electron spin reso-
nance (ESR), and19F MRI (4,11,16,17). Some of these stud-
ies have indicated a strong quantitative correlation with
tissue pO2 (11); some found a qualitative relationship
(16,17) while others suggested a lack of direct correlation
yet consistent temporal trends (4).
We have developed a method for measuring tumor oxy-
genation and dynamics based on19F NMR EPI following
direct intratumoral injection of the reporter molecule
hexafluorobenzene (HFB): fluorocarbon relaxometry using
EPI for dynamic oxygen mapping (FREDOM) (2). This
technique provides quantitative pO2 measurements at
multiple specific locations simultaneously within a tumor,
and reveals acute dynamic changes at individual locations
with respect to interventions, such as hyperoxic gas
breathing and vascular modifiers. The aim of this study
was to compare FREDOM with1H BOLD MRI in evaluating
tumor oxygenation in response to hyperoxic gas (100%
MATERIALS AND METHODS
Rat mammary carcinoma 13762NF (originally obtained
from the Division of Cancer Treatment, National Cancer
Institute [NCI], Bethesda, MD, USA) was implanted syn-
geneically in a skin pedicle surgically created on the fore-
back of Fisher 344 adult female rats (?150 g, N ? 9;
Harlan), as described in detail previously (18). Tumors
were allowed to grow and were investigated when tumor
volume was 0.2 to 2.1 cm3(mean volume ? 1 cm3). Inves-
tigations were approved by the Institutional Animal Care
and Use Committee.
MRI was performed using a 4.7T horizontal bore magnet
with a Varian Unity Inova system. Each rat was given
ketamine hydrochloride (120 ?l, 100 mg/ml; Aveco, Fort
Dodge, IA, USA) as a relaxant (intraperitoneally [i.p.]) and
maintained under general anesthesia (air and 1% isoflu-
Department of Radiology, University of Texas Southwestern Medical Center,
Dallas, Texas, USA.
*Correspondence to: Ralph P. Mason, Ph.D., Department of Radiology, UT
Southwestern Medical Center, 5323 Harry Hines Blvd., Dallas, TX 75390-
9058. E-mail: Ralph.Mason@UTSouthwestern.edu
Grant sponsor: U.S. Department of Defense (DOD) Breast Cancer Initiative
Awards IDEA DAMD; Grant number: 17-03-1-0363 and grant number 17-02-
1-0592; Grant sponsor: National Cancer Institute (NCI); Grant number: RO1
CA79515/EB002762; Grant sponsor: Southwestern Small Animal Imaging
Research Program (SW-SAIRP); Grant numbers: U24 CA12660801, P20 Pre-
Received 28 July 2008; revised 4 February 2009; accepted 18 February 2009.
Published online 12 June 2009 in Wiley InterScience (www.interscience.
Magnetic Resonance in Medicine 62:357–364 (2009)
© 2009 Wiley-Liss, Inc.
rane; Baxter International, Inc., Deerfield, IL, USA). The
oxygen reporter molecule HFB (50 ?l; Lancaster, Gaines-
ville, FL, USA) was injected directly into the tumor along
two or three tracks in a single central plane of the tumor,
coronal to the rat’s body using a Hamilton syringe (Reno,
NV, USA) with a custom-made fine sharp needle (32G), as
described in detail previously (2). A tunable (1H/19F) vol-
ume radio frequency (RF) coil was placed around the
tumor-bearing pedicle. Each animal was placed on its side
in the magnet with no change in position during the whole
study. A thermal blanket was used to maintain body tem-
perature. A single 2-mm slice coronal to the rat body
containing the strongest fluorine signal was chosen for
both19F pO2and1H BOLD studies.1H and19F MRI images
were acquired using a spin-echo sequence. Overlaying the
19F MRI image on the corresponding1H image revealed the
distribution of HFB.
A spin EPI sequence with pulse burst saturation recovery
(PBSR) signal preparation was applied, as described pre-
viously (8). The initial saturation was designed to mini-
mize in-flow effects. A series of 55 images, including five
at baseline with air breathing (images 1–5) and 50 with
oxygen breathing (images 6–55) was acquired on the 2-mm
coronal section at 5-s intervals using the following MR
parameters: ? ? 500 ms (?TR), TE ? 53.7 ms, field of view
(FOV) ? 40 mm ? 40 mm, matrix size ? 32 ? 32.
Multigradient-echo (MGRE) images with eight echoes were
acquired on the same 2-mm slice during air breathing
(baseline) and repeated immediately after the EPI BOLD
sequence (?5 min after start of O2breathing) for eight of
nine tumors. Acquisition parameters were as follows: rep-
etition time (TR) ?195 ms, initial echo time (TE) ? 7 ms,
echo spacing ? 6 ms, flip angle ? 45°, FOV ? 40 mm ?
40 mm, matrix size ? 128 ? 128, averages ? 2, acquisition
time ? 6 min 40 s.
19F Tumor Tissue Oximetry: FREDOM
Following a reequilibration period of air breathing
(?15 min), tumor oxygenation was estimated on the basis
previously (2). This approach provided pO2maps with
1.25-mm in-plane resolution and ?3-?l voxel size (FOV ?
40 mm ? 40 mm, matrix size ? 32 ? 32, thickness ?
2 mm) in 6.5 min. The spin-lattice relaxation rate [R1
(s–1) ? 1/T1] was estimated on a voxel-by-voxel basis using
a three-parameter monoexponential function, and pO2was
estimated using the relationship pO2 (torr) ? [(R1 –
0.0835)/0.001876] (2). Seven consecutive pO2 measure-
ments, including two baseline and five oxygen-breathing,
were acquired on the same 2-mm section as used for the1H
19F PBSR EPI relaxometry of the HFB, as described
Data were processed using IDL 5.3/5.4 (Research Systems,
Boulder, CO, USA). SI in the EPI BOLD study was assessed
on a voxel-by-voxel basis and averaged at every time point
for the whole tumor section. The SI change (?SI) of each
tumor was normalized to the mean baseline value ex-
pressed as a percentage change using the equation:
??SI? ? ?SIM? SIb?/SIb? 100%,
where SIMis the maximum mean SI and SIbis the mean
R*2 maps were generated using all eight images with
variable echo time by fitting an exponential model on a
voxel-by-voxel basis. Mean R*2of the whole section was
determined for baseline air and oxygen intervention. ?R*2
maps were obtained by subtracting the R*2oxygen map
from the R*2baseline map.
For the FREDOM data, typically 40 to 100 voxels pro-
vided an R1fit, and a potential pO2value. Since noise may
give an apparent relaxation curve (R1) fit, data were se-
lected by applying thresholds of T1error ? 2.5 s and ratio
T1error/T1? 50%. The two criteria are used since T1can
have a very large range from 1.5 to 12 s and thus the
absolute error is particularly important for long T1s and
the ratio for short T1s. While these criteria appear quite lax,
only those voxels that provided consistently reliable data
throughout the whole time course of seven measurements
were included for further analysis.
Statistical significance was assessed using an analysis of
variance (ANOVA) on the basis of Fisher’s Protected Least
Significant Difference (PLSD; Statview, SAS Institute, Inc.,
Cary, NC, USA) or paired Student’s t-tests. ANOVA was
applied for comparison of multiple repeat measurements
and the PLSD examines the importance of individual mea-
surements on the overall population. The assumption is
that inhaled gas at various time points is the independent
variable, while pO2, R*2, and BOLD signal changes are
dependant variables. Paired Student’s t-tests were used to
compare individual pairs of data such as pO2in a specific
tumor during air or oxygen breathing.
Overlay of19F on the corresponding1H MRI image con-
firmed that HFB was distributed in both peripheral and
central regions of a 2-mm-thick section, located in the
central plane of a representative tumor (Fig. 1). The fol-
lowing EPI BOLD, GRE R*2, and19F oximetry measure-
ments all interrogated this thin slice.
Normalized SI maps at different time points after switch-
ing to oxygen breathing showed heterogeneous response
(Fig. 2). Significant mean signal enhancement during oxy-
gen inhalation was observed in all nine tumors, with a
mean ?SI ? 6.7 ? 1.4% (range, 1% to 12%; Table 1).
However, individual voxel data showed some regions with
negative response (Fig. 2). The percentage of voxels with
negative response ranged from 12% to 38% in the nine
tumors. Baseline signal was quite stable, but there was a
rapid response within about 25 s of switching the inhaled
358 Zhao et al.
gas to oxygen. Some tumors showed a continual increase,
which was usually biphasic and approached a stable pla-
teau after about 3 min. In other cases there was a transient
maximum followed by a stable lower value, just margin-
ally above baseline.
Baseline maps revealed distinct heterogeneity with R*2
ranging from 6 to 450 s–1(T*2 ? 2–166 ms; Fig. 3). In
response to oxygen challenge, a small but significant de-
crease in R*2was seen, predominately in the tumor periph-
ery (P ? 0.001; Fig. 3). There was no correlation between
baseline R*2and ?R*2on a voxel-by-voxel basis (r2? 0.1;
Fig. 3b). For the group of eight tumors, a significant de-
crease in mean R*2(suggesting increased oxyhemoglobin
level) was found (mean ?R*2? –4 ? 2, P ? 0.05; Table 1).
One tumor (no. 5; Table 1) showed contrary behavior, with
increased mean R*2with oxygen breathing. A very close
correlation was observed between the mean R*2 values
during air vs. oxygen breathing (Fig. 3c). A general trend
was observed when the mean BOLD response for individ-
ual tumors was compared with ?R*2(Fig. 3d). In essence,
large ?R*2was associated with large ?SI, and an increase in
R*2coincided with a small signal change. However, most
tumors showed a ?R*2of –4 ms and this was associated
with a large range in BOLD signal change.
19F Tumor Tissue Oximetry: FREDOM
pO2maps showed a range of baseline pO2values and a
heterogeneous response to oxygen breathing (Fig. 4; Table
1). Oxygenation appeared clustered, with higher pO2re-
gions appearing close to the periphery when overlaid on
the anatomical1H MRI images of the corresponding tumor
slices. The time course of pO2dynamics showed differen-
tial response in both rate and magnitude (Fig. 4b). For the
group of nine tumors, baseline pO2varied from essentially
hypoxic (0.3 torr) to well-oxygenated (36 torr; Table 1).
With respect to oxygen challenge, mean pO2 increased
significantly in all the tumors (?pO2? 50 torr; Table 1) and
hypoxic fractions (?10 torr) decreased significantly, from
a mean tumor baseline of 31% to 8% (P ? 0.05). In most
tumors (7/9) the HF10was essentially eliminated (?5%
residual), but in two tumors a substantial HF remained,
albeit considerably diminished compared with baseline.
Histograms for the pooled individual voxels (N ? 265)
from the nine tumors in response to oxygen challenge
revealed significant increase in pO2to a mean of 75 ? 4
torr and a median of 63 torr (P ? 0.001; Fig. 5a and b). HFs
of HF5(?5 torr) and HF10(?10 torr) decreased signifi-
cantly from baseline values of 18% and 34%, to 8% and
10%, respectively (P ? 0.01). In common with previous
observations in this tumor line, baseline pO2tended to
decrease with tumor volume (r2? 0.52) and as a corollary
HF10increased with tumor volume (r2? 0.49) (19,20). A
strong inverse correlation was found between tumor mean
pO2and HF10(Fig. 5c). The change in pO2(?pO2) tended to
increase for tumors with higher baseline pO2(r2? 0.4), but
the relationship was much stronger when comparing pO2
during oxygen breathing vs. baseline pO2(r2? 0.6; Fig.
Correlation of pO2With1H BOLD
For the group of nine tumors, a significant linear correla-
tion was found between mean ?pO2and ?SI, detected by
19F oximetry and1H EPI BOLD (r2? 0.7, P ? 0.001; Fig. 6).
FIG. 2. Variations in T*2-weighted images with oxygen challenge
(BOLD). a: Normalized spin echo-planar images of two tumors
acquired at several time points after switching to oxygen breath-
ing. In response to oxygen breathing, the two tumors (upper: #2,
0.4 cm3; lower: #6, 1.4 cm3) showed heterogeneous signal en-
hancement. Regions of decreased SI were also observed (dark
regions). b: Variations of normalized SI change (?SI) vs. time for
these tumors with respect to oxygen challenge. Both tumors
showed an initial rapid response. For tumor #2 (E) this became
biphasic, reaching a plateau with 9% increase after about 200 s
of oxygen breathing. The second tumor (#6, ■) rapidly reached a
peak value of ?SI ? 1.4% after 35 s, then gradually decreased to
a plateau at about 0.5% increase. Data points are mean values ?
FIG. 1. Distribution of oxygen reporter probe. a:1H T1-weighted MR
image of 2-mm slice through tumor showing signal voids corre-
sponding to presence of HFB reporter molecule. b: Overlay of19F SI
on1H image showing distribution of oxygen reporter probe hexaflu-
orobenzene (HFB) in both central and peripheral tumor regions (see
#2 in Table 1). Bar ? 0.5 cm. [Color figure can be viewed in the
online issue, which is available at www.interscience.wiley.com.]
MRI Evaluation of Tumor Oxygenation359
However, there was a weak correlation between baseline
pO2and ?SI (r2? 0.3, data not shown). Likewise, compar-
ison of ?R*2with baseline pO2, maximum pO2during ox-
ygen breathing, or ?pO2indicated no correlations (r2?
0.01). Every tumor exhibiting a large BOLD response
(?SI ? 3%) had a negligible residual HF (HF10? 5%)
during oxygen breathing. In two of three tumors showing a
small BOLD response, a large HF10remained (Fig. 7).
Oxygenation in 13762NF rat breast tumors was found to
cover a considerable range, with both intra- and intertu-
mor heterogeneity, but a tendency toward lower pO2and
greater hypoxia in larger tumors, as we also reported pre-
viously (19,20). Likewise, tumor response to modulation
by hyperoxic gas breathing, which resulted in increased
pO2and reduced residual hypoxic fraction, was in line
with previous observations (19–21). Here, we have under-
taken consecutive1H MRI BOLD and19F MRI oximetry
investigations to examine potential correlations. A strong
correlation was found between mean BOLD signal re-
sponse (?SI) and change in mean pO2(Fig. 6), supporting
previous observations in various tumor types reported by
others (4,11,17). More significantly, a small BOLD re-
sponse usually indicates a substantial residual HF, while a
large BOLD response to breathing oxygen indicates that
the tumor is well-oxygenated or becomes well-oxygenated
Quantitative oximetry (absolute pO2values) has been
shown to predict local recurrence and disease-free sur-
vival in several human cancers (notably, cervical and head
and neck (22,23)). To date, electrodes have provided the
only quantitative clinical pO2measurements in tumors,
and electrodes are not only invasive, but the Eppendorf
Histograph cannot easily show changes in pO2with re-
spect to interventions. Furthermore, the Eppendorf Histo-
graph is no longer commercially available.19F MRI based
on perfluorocarbon (PFC) reporters has been used to mea-
sure pO2in the human eye (24), but it remains generally
FIG. 3. Response of R*2to oxygen breathing.
a: R*2maps showed heterogeneous baseline
R*2values for tumor #2. Top left: R*2while
breathing air; Center: R*2after 5 min breath-
ing oxygen; Right: difference map obtained
by subtracting the oxygen map from the air
map (mean ?R*2? –2.6 s–1). b: A voxel-to-
voxel comparison (N ? 2056) between
baseline R*2and ?R*2showed a poor corre-
lation (r2? 0.1). c: Comparison of mean
tumor R*2measured while rat breathed oxy-
gen vs. air showed a strong correlation (r2?
0.96). d: Comparison of mean signal change
in T*2-weighted image vs. change in mean R*2
in tumors accompanying change in gas
from air to oxygen breathing. No linear cor-
relation was observed, but greater change
in R*2was accompanied by changes in SI.
Comparison of pO2, BOLD, and R*2in Individual Tumors
36 ? 1
17 ? 2
13 ? 2
13 ? 1
12 ? 1
13 ? 1
24 ? 5
1 ? 2
35 ? 5
18 ? 4
31 ? 7
172 ? 7**
76 ? 4**
62 ? 8**
81 ? 6**
33 ? 4*
26 ? 5**
72 ? 12**
16 ? 9*
74 ? 9**
68 ? 15*
8 ? 4*
12.2 ? 0.6
8.7 ? 0.5
7.6 ? 0.6
12.4 ? 0.4
2.3 ? 0.6
1.6 ? 0.5
4.6 ? 0.8
2.4 ? 0.6
8.2 ? 0.6
6.7 ? 1.4
54 ? 1
68 ? 1
81 ? 1
116 ? 2
71 ? 1
60 ? 1
56 ? 1
90 ? 1
74 ? 7
51 ? 1*
65 ? 1**
77 ? 1**
102 ? 2**
73 ? 1
56 ? 1**
51 ? 1**
89 ? 1
70 ? 6*
1.0 ? 0.250 ? 13
?4 ? 2
aMean pO2over all voxels in the two repeated baseline measurements.
bMean highest pO2observed in all voxels during five oxygen breathing measurements.
*P ? 0.005.
**P ? 0.001 from baseline (21% O2).
NA ? not measured.
360Zhao et al.
restricted to preclinical animal studies (2) for two funda-
mental reasons: lack of access to clinical19F MRI capabil-
ities, and lack of U.S. Food and Drug Administration
(FDA) approval for human use of PFCs. ESR can also
measure pO2distributions in rodent tumors (16,25,26), but
while it has been used to measure oxygenation in humans
based on India ink tattoos, it is also handicapped by lack of
BOLD contrast MRI is an attractive surrogate for clinical
pO2measurements, since endogenous hemoglobin itself
serves as the reporter molecule. A few studies of human
tumor BOLD have been presented (5,27–29), but further
validation relative to other techniques is of the utmost
importance. Fundamental reports from Thulborn et al. (30)
and Wright (31), together with applications of functional
MRI (fMRI) in the brain, lay a strong foundation for vas-
cular oxygen measurements. Several groups (3,4,6,8,10–
14,16,17,26,32) have explored BOLD responses in diverse
tumor types with respect to varying oxygen concentrations
and carbogen. There is considerable evidence that signal
changes in T*2-weighted images reflect changes in pO2.
FIG. 4. Variation of pO2with oxygen challenge. a: pO2maps obtained
at successive times overlaid on T1-weighted1H images of two tumors
(#2 and #6). A range of pO2values was observed in both tumors under
baseline conditions. In response to breathing oxygen, all the individual
locations (34 voxels) in tumor #2 (upper row) responded significantly
and became well oxygenated. By contrast, some of the initially hypoxic
well oxygenated. b: Variation in mean pO2of each tumor. Tumor #2 (E;
mean baseline ? 17 torr) showed significantly increased pO2within
7 min of oxygen breathing, and continued to increase, reaching 76 torr
during the final measurement (49 min). Tumor #6 (■, mean baseline
pO2? 13 torr) reached a peak value (26 torr) after 14 min, but then
settled back to a lower level. *P ? 0.05; **P ? 0.001.
FIG. 5. Tumor oxygen tension distribution. a: Pooled pO2values for
individual regions (265) from the nine tumors showed a range of
baseline pO2values from hypoxia (?5 torr) to 55 torr, with a mean
(x) ? 15 ? 1 torr and median (m) ? 13.1 torr, while rats breathed air.
Binning is based on ranges; e.g., 10 refers to 5 ? pO2? 10 torr. b:
Oxygen breathing produced a significant increase in pO2, with mean
(75 ? 4 torr) and median (63 torr) (P ? 0.001). Hypoxic fractions HF5
(?5 torr) and HF10(?10 torr) decreased significantly from baseline
values of 18% and 34%, to 8% and 10%, respectively (P ? 0.01). c:
Correlation between mean baseline pO2 and hypoxic fraction
showed inverse relationship (r2? 0.87). d: Dependence of pO2
achieved with oxygen breathing on baseline pO2(r2? 0.6).
FIG. 6. Correlation of pO2and BOLD responses to oxygen chal-
lenge. A significant linear correlation was found between mean
increase in tissue pO2(?pO2) and mean spin echo planar BOLD
signal increase in the nine tumors with respect to oxygen interven-
tion (r2? 0.7; P ? 0.001).
FIG. 7. Assessment of residual hypoxic fraction. Comparison of the
final hypoxic fraction (HF10) with oxygen breathing as a function of
BOLD signal response (?SI) suggests strong predictive value. For
most tumors (6/9) a large BOLD response coincided with low resid-
ual HF10. A small BOLD response indicated a large residual hypoxic
fraction in two of three tumors.
MRI Evaluation of Tumor Oxygenation361
However, some investigations have shown a lack of direct
correlation; e.g., Baudelet and Gallez (4) reported that a
10% change in signal could correspond to a small (?25
torr) or large (approaching 100 torr) change in pO2 in
syngeneic fibrosarcoma (FSA) mouse tumors. Importantly,
both represent large changes in pO2 by radiobiological
standards. Elas et al. (16) showed correlation between
electron paramagnetic resonance imaging (EPRI) based on
vascular trityl spin probe and sequential BOLD measure-
ments in FSA tumors in mice. As with our study, their two
measurements were sequential rather than concomitant
and they had the added complexity of coregistering images
from separate modalities.
Fan et al. (17) previously compared19F MR oximetry
with BOLD response on a voxel-by-voxel basis in
R3230AC rat breast tumors. In common with many exper-
iments, they infused perfluorocarbon emulsion, which
progressively sequestered in the tumor tissue while clear-
ing from the vasculature. They reported that19F and1H
measurement agreed in 65% of pixels, viz. when19F MRI
showed increased pO2, then1H MRI linewidth decreased,
reflecting less deoxyhemoglobin, as assessed using high-
spatial/spectral resolution T*2-sensitive (HiSS) measure-
ments. Correlations were even stronger for subsets of pix-
els selected as showing no pO2change or a BOLD change.
Overall, they concluded that regions identified as hypoxic
tended to show a small BOLD response to carbogen inha-
lation in the R3230AC rat breast tumors and1H MRI gave
very few “false positives” (17). Our results are in accord
with these previous observations. Tumors exhibiting a
large BOLD response also showed a greater mean pO2
response (Fig. 6). Most significantly, tumors exhibiting a
large BOLD response (here, defined as ?4%) showed es-
sentially no residual HF (HF10? 5%; Fig. 7). We found no
false positives and only one false negative. While breath-
ing air, seven of nine tumors had HF10? 20% (Table 1). In
all but two cases this fell below 5% with oxygen breathing,
essentiality eliminating the HF. These remaining two tu-
mors showed a particularly small BOLD effect (?SI ?
2.5%). Meanwhile, six of seven responsive tumors exhib-
ited a large BOLD effect (?4%). We do note that the
13762NF tumors show relatively low HFs (HF10) compared
with many reports for tumors implanted in rodents. How-
ever, the values are closely in line with measurements
reported using the Eppendorf Histograph in breast tumors
in patients (33).
We should note differences between Karczmar’s team’s
approach (17) and ours. The use of systemically delivered
PFC emulsions as reporter molecules tends to provide
signal from well-perfused tumor regions only. Indeed, Fan
et al. (17) predominantly detected19F signal from the tu-
mor periphery, as also noted by others (34). While our
current approach allowed interrogation of central tumor
regions, as well as periphery, it required direct injection of
HFB into the tumor, which is invasive and samples limited
regions only. Thus, it is particularly reassuring that the
two approaches provide commensurate results.
BOLD changes reflect vascular oxygenation, whereas
FREDOM measures tissue pO2. As expected, the BOLD
changes occurred much more rapidly (in seconds; Fig. 2)
than the pO2changes (in minutes; Fig. 4). In the future it
will be interesting to examine1H T1-weighted tissue water
response, or so-called tissue oxygen level–dependant
(TOLD) response, as reported by Matsumoto et al. (35),
since then both vascular and tissue changes can be as-
sessed by1H MRI. Matsumoto et al. (35) examined re-
sponse to hyperbaric oxygen breathing, but others have
reported T1changes associated with hyperoxic gas in nor-
mal tissues (36) and tumors (3,10). The kinetic response of
the mean BOLD signal was consistent with previous global
near-infrared observations in this tumor type (20,37).
Changes in deoxyhemoglobin concentration generally fol-
lowed a biphasic time course, as also seen in the BOLD
response (Fig. 2b). This probably represents rapid arterio-
lar oxygenation, followed by more sluggish response in the
distant parts of the vascular tree. Meanwhile, pO2response
was more sluggish, with continued increase over 30 min
(Fig. 4b) for a responsive tumor. These observations are in
line with previous observations based on oxygen-sensitive
fiber optic probes and polarographic electrodes in this
tumor type (38).
BOLD MRI or susceptibility-weighted R*2measurement
based on the intrinsic paramagnetic properties of deoxy-
hemoglobin have been increasingly applied to assess tu-
mor vasculature (7,9). An increase in BOLD SI or a de-
crease in R*2may be related to decreased blood deoxyhe-
moglobin. However, BOLD SI change is also related to
several other factors; e.g., changes in tumor blood flow,
volume, hematocrit, and the ability of red blood cells to
traverse tumor capillaries (3). Here, we applied a presatu-
ration sequence across the whole tumor to minimize flow
effects. It has been shown that BOLD MRI is probably less
sensitive to changes in tumor oxygenation in regions con-
taining very sparse vasculature, and hence, little deoxyhe-
moglobin (9). There may be concern that poorly vascular-
ized tumors cannot show a measurable BOLD effect, due to
lack of deoxyhemoglobin (9,14). As a corollary, we pro-
pose that poorly vascularized tumors will also be hypoxic.
Thus, a small or absent BOLD effect will be indicative of
hypoxia. Indeed, Rodrigues et al. (12) showed that well-
vascularized rat pituitary tumor cell line (GH3) prolacti-
nomas tended to have a much higher R*2than sparsely
GH3 tumors showed a large ?R*2in response to carbogen
breathing, whereas the RIF-1 tumors showed essentially
no change. Breathing carbogen enhanced the response of
GH3 tumors to a single high dose of radiation (15 Gy),
whereas there was no effect on RIF-1 tumors. Our range of
R*2 values (54–116 s–1) is commensurate with previous
reports for animal tumors at 4.7T (12). In terms of potential
clinical applications, the ability to stratify patients based
on oxic or oxygenatable tumors vs. hypoxic (and resistant
to modulation) could be significant.
BOLD contrast is related to changes in local deoxyhe-
moglobin concentration. However, baseline R*2reflects not
only blood deoxyhemoglobin level, vascular blood flow,
and volume, but also local tissue architecture; i.e., cell
density, edema, or necrosis. Thus, the baseline R*2and
change in R*2 (?R*2) likely depends on tumor type. Re-
cently, Robinson et al. (9) showed heterogeneous intertu-
moral R*2among a variety of tumors, of which the GH3
prolactinoma had the highest mean R*2(89 s–1) and RIF-
1 had the lowest value (58 s–1). In response to carbogen
breathing, a significant decrease in R*2(–23 s–1) was found
362 Zhao et al.
in the GH3 prolactinoma, whereas the RIF-1 fibrosarcomas
showed a little increase in R*2(1 s–1). In the current study,
seven out of eight tumors had a modest decrease in R*2
(mean ? –4 s–1), while R*2increased in one tumor (#5) in
response to oxygen breathing. As expected, the largest ?SI
in response to oxygen breathing coincided with a large
decrease in R*2 and the one tumor with increased R*2
showed small ?SI. However, most tumors had similar ?R*2
(about 4 s–1) yet a large range of ?SIs (Fig. 3d). Generally,
there was no significant correlation between baseline ?R*2
values and R*2(Fig. 3b; Table 1). There was a strong corre-
lation between mean R*2measured during air or oxygen
breathing (Fig. 3c).
In the past, we had used a thick section (essentially
projection) for19F MR oximetry (FREDOM) studies (2). The
2-mm-thick slice used here allowed more satisfactory spa-
tial correlation between19F oximetry and1H BOLD con-
trast. The success of FREDOM depends largely on the
signal-to-noise ratio (SNR) of the acquired images. No ob-
vious decrease in SNR due to reduced slice thickness was
seen in the current study, which we attribute to greater
emphasis on injecting the HFB in a narrow plane and the
ability to image oblique angles in the upgraded Varian
Inova system. On average, 29 voxels (range, 10 to 43)
provided reliable pO2 readings that were traceable
throughout the oxygen challenge in each tumor.
We had hoped to correlate individual voxels in both19F
and1H EPI (FREDOM vs. BOLD). However, the directly
corresponding voxels did not allow meaningful correla-
tion. Signal voids were observed, which had only 10% to
25% as much1H signal as surrounding tumor (Fig. 1a).
Overlaying19F on1H images showed that the low1H SI
regions corresponded with19F signal (Fig. 1b). In addition
the R*2values were typically an order of magnitude smaller
(14 to 37 s–1, as opposed to 100 s–1) with smaller and
sometimes opposite changes in R*2 compared with sur-
rounding voxels in response to breathing oxygen. We have
previously observed such signal voids in
tumors with HFB (19) and with hexamethyldisiloxane, the
new alternative1H MRI pO2reporter under development
(39). While1H and19F voxels did not allow direct corre-
lation, we believe that the judicious placement of reporter
molecules in central and peripheral locations can provide
a representation of the whole tumor, and thus, correlation
with nonlabeled regions is reasonable. Importantly, pO2
values and dynamics observed using FREDOM are highly
consistent with other oximetry methods in rat breast and
prostate tumors, such as electrodes (38,40), fiber optic
probes (37,41), and immunohistochemistry (42).
Oxygen breathing produced a significant increase in19F-
measured pO2and EPI BOLD SI, and a decrease in GRE R*2
on the same section of tumor. Data indicated a strong
correlation between %?SI and ?pO2(Fig. 6). Although a
significant increase in global mean SI was found in all
tumors by BOLD in the current study, voxel-by-voxel data
analysis showed oxygen breathing induced signal loss in
many voxels for each tumor, averaged to 20% of the total
voxels. Similar findings have been observed in other tumor
types by others and us (6,8,10,14). Indeed, Fan et al. (17)
previously remarked that correlates between mean BOLD
and pO2were stronger when the whole tumor was consid-
ered, as opposed to individual regions. In a slightly differ-
1H images of
ent context Baudelet et al. (43) reported a better correlation
between mean tumor BOLD signal response and vascular
permeability (kep) than for individual voxels.
The ability to identify hypoxia could have far-reaching
implications for radiotherapy. We have previously shown
correlations between tumor pO2 measured by
(FREDOM) and response to single high-dose irradiation in
Dunning prostate R3327-HI and AT1 tumors (44,45). Fur-
thermore, we could correctly assess the ability to alter pO2
and modulate response to irradiation. Karczmar’s team,
i.e., Al-Hallaq et al. (46), also demonstrated the ability of
BOLD (viz. HiSS) to predict the relative efficacy of tumor
oxygenating agents.1H MRI BOLD assessment would be
far more practical in the clinic and we believe our results,
together with the previous report from Fan et al. (17),
provide strong impetus for translation.
We thank Ammar Adams and Dr. Matthew Merritt for
technical and collegial support. This work was supported
by U.S. Department of Defense (DOD) Breast Cancer Ini-
tiative Awards IDEA DAMD 17-03-1-0363 (to D.Z.) and a
predoctoral fellowship DAMD 17-02-1-0592 (to L.J.), in
conjunction with National Cancer Institute (NCI) RO1
CA79515/EB002762 and the Southwestern Small Animal
Imaging Research Program (SW-SAIRP) (which is sup-
ported by U24 CA12660801 and P20 Pre-ICMIC CA86354).
MRI experiments were performed in the Advanced Imag-
ing Research Center (formerly, Mary Nell and Ralph B.
Rogers MR Center) a National Institutes of Health (NIH)
Biotechnology Research Program P41-RR02584 facility.
1. Tatum JL, Kelloff GJ, Gillies RJ, Arbeit JM, Brown JM, Chao KS, Chap-
man JD, Eckelman WC, Fyles AW, Giaccia AJ, Hill RP, Koch CJ, Krishna
MC, Krohn KA, Lewis JS, Mason RP, Melillo G, Padhani AR, Powis G,
Rajendran JG, Reba R, Robinson SP, Semenza GL, Swartz HM, Vaupel
P, Yang D, Croft B, Hoffman J, Liu G, Stone H, Sullivan D. Hypoxia:
importance in tumor biology, noninvasive measurement by imaging,
and value of its measurement in the management of cancer therapy. Int
J Radiat Biol 2006;82:699–757.
2. Zhao D, Jiang L, Mason RP. Measuring changes in tumor oxygenation.
Methods Enzymol 2004;386:378–418.
3. Howe FA, Robinson SP, Rodrigues LM, Griffiths JR. Flow and oxygen-
ation dependent (FLOOD) contrast MR imaging to monitor the response
of rat tumors to carbogen breathing. Magn Reson Imaging 1999;17:
4. Baudelet C, Gallez B. How does blood oxygen level-dependent (BOLD)
contrast correlate with oxygen partial pressure (pO2) inside tumors?
Magn Reson Med 2002;48:980–986.
5. Padhani A, Krohn K, Lewis J, Alber M. Imaging oxygenation of human
tumours. Eur Radiol 2007;17:861–872.
6. Thomas CD, Chenu E, Walczak C, Plessis MJ, Perin F, Volk A. Morpho-
logical and carbogen-based functional MRI of a chemically induced
liver tumor model in mice. Magn Reson Med 2003;50:522–530.
7. Neeman M, Dafni H, Bukhari O, Braun RD, Dewhirst MW. In vivo
BOLD contrast MRI mapping of subcutaneous vascular function and
maturation: validation by intravital microscopy. Magn Reson Med
8. Jiang L, Zhao D, Constantinescu A, Mason RP. Comparison of BOLD
contrast and Gd-DTPA dynamic contrast enhanced imaging in rat pros-
tate tumor. Magn Reson Med 2004;51:953–960.
9. Robinson SP, Rijken PF, Howe FA, McSheehy PM, van der Sanden BP,
Heerschap A, Stubbs M, Van Der Kogel AJ, Griffiths JR. Tumor vascular
architecture and function evaluated by non-invasive susceptibility MRI
methods and immunohistochemistry. J Magn Reson Imaging 2003;17:
MRI Evaluation of Tumor Oxygenation363
10. Peller M, Weissfloch L, Stehling MK, Weber J, Bruening R, Sene-
kowitsch-Schmidtke R, Molls M, Reiser M. Oxygen-induced MR signal
changes in murine tumors. Magn Reson Imaging 1998;16:799–809.
11. Al-Hallaq HA, River JN, Zamora M, Oikawa H, Karczmar GS. Correla-
tion of magnetic resonance and oxygen microelectrode measurements
of carbogen-induced changes in tumor oxygenation. Int J Radiat Oncol
Biol Phys 1998;41:151–159.
12. Rodrigues LM, Howe FA, Griffiths JR, Robinson SP. Tumor R*2is a
prognostic indicator of acute radiotherapeutic response in rodent tu-
mors. J Magn Reson Imaging 2004;19:482–488.
13. Robinson SP, Howe FA, Rodrigues LM, Stubbs M, Griffiths JR. Mag-
netic resonance imaging techniques for monitoring changes in tumor
oxygenation and blood flow. Semin Radiat Oncol 1998;8:198–207.
14. Baudelet C, Gallez B. Current issues in the utility of blood oxygen level
dependent MRI for the assessment of modulations in tumor oxygen-
ation. Curr Med Imaging Rev 2005;1:229–243.
15. Duyn JH, Moonen CT, van Yperen GH, de Boer RW, Luyten PR. Inflow
versus deoxyhemoglobin effects in BOLD functional MRI using gradi-
ent echoes at 1.5 T. NMR Biomed 1994;7:83–88.
16. Elas M, Williams BB, Parasca A, Mailer C, Pelizzari CA, Lewis MA,
River JN, Karczmar GS, Barth ED, Halpern HJ. Quantitative tumor
oxymetric images from 4D electron paramagnetic resonance imaging
(EPRI): methodology and comparison with blood oxygen level-depen-
dent (BOLD) MRI. Magn Reson Med 2003;49:682–691.
17. Fan X, River JN, Zamora M, Al-Hallaq HA, Karczmar GS. Effect of
carbogen on tumor oxygenation: combined fluorine-19 and proton MRI
measurements. Int J Radiat Oncol Biol Phys 2002;54:1202–1209.
18. Hahn EW, Peschke P, Mason RP, Babcock EE, Antich PP. Isolated
tumor growth in a surgically formed skin pedicle in the rat: a new
tumor model for NMR studies. Magn Reson Imaging 1993;11:1007–
19. Song Y, Constantinescu A, Mason RP. Dynamic breast tumor oximetry:
the development of prognostic radiology. Technol Cancer Res Treat
20. Xia M, Kodibagkar V, Liu H, Mason RP. Tumour oxygen dynamics
measured simultaneously by near infrared spectroscopy and19F mag-
netic resonance imaging in rats. Phys Med Biol 2006;51:45–60.
21. Zhao D, Jiang L, Hahn EW, Mason RP. Tumor physiological response to
combretastatin A4 phosphate assessed by MRI. Int J Radiat Oncol Biol
22. Fyles A, Milosevic M, Hedley D, Pintilie M, Levin W, Manchul L, Hill
RP. Tumor hypoxia has independent predictor impact only in patients
with node-negative cervix cancer. J Clin Oncol 2002;20:680–687.
23. Brizel DM, Sibly GS, Prosmitz LR, Scher RL, Dewhirst MW. Tumor
hypoxia adversely affects the prognosis of carcinoma of the head and
neck. Int J Radiat Oncol Biol Phys 1997;38:285–289.
24. Wilson C, Berkowitz B, McCuen B, Charles C. Measurement of preret-
inal pO2in the vitrectomized human eye using19F NMR. Arch Oph-
25. Bratasz A, Pandian RP, Deng Y, Petryakov S, Grecula JC, Gupta N,
Kuppusamy P. In vivo imaging of changes in tumor oxygenation during
growth and after treatment. Magn Reson Med 2007;57:950–959.
26. Dunn JF, O’Hara JA, Zaim-Wadghiri Y, Lei H, Meyerand ME, Grinberg
OY, Hou H, Hoopes PJ, Demidenko E, Swartz HM. Changes in oxygen-
ation of intracranial tumors with carbogen: a BOLD MRI and EPR
oximetry study. J Magn Reson Imaging 2002;16:511–521.
27. Taylor NJ, Baddeley H, Goodchild KA, Powell ME, Thoumine M,
Culver LA, Stirling JJ, Saunders MI, Hoskin PJ, Phillips H, Padhani AR,
Griffiths JR. BOLD MRI of human tumor oxygenation during carbogen
breathing. J Magn Reson Imaging 2001;14:156–163.
28. Rijpkema M, Kaanders JH, Joosten FB, van der Kogel AJ, Heerschap A.
Effects of breathing a hyperoxic hypercapnic gas mixture on blood
oxygenation and vascularity of head-and-neck tumors as measured by
magnetic resonance imaging. Int J Radiat Oncol Biol Phys 2002;53:
29. Jiang L, McColl R, Weatherall P, Tripathy D, Mason RP. BOLD and
Gd-DTPA contrast enhanced MRI for early assessment of breast cancer
chemotherapy. In: Proceedings of the 13th Annual Meeting of ISMRM,
Miami Beach, FL, USA, 2005 (Abstract 408).
30. Thulborn KR, Waterton JC, Matthews PM, Radda GK. Oxygenation
dependence of the transverse relaxation time of water protons in whole
blood at high field. Biochim Biophys Acta 1982;714:265–270.
31. Wright GA, Hu BS, Macovski A. Estimating oxygen saturation of blood
in vivo with MR imaging at 1.5 T. J Magn Reson Imaging 1991;1:275–
32. Robinson SP, Collingridge DR, Howe FA, Rodrigues LM, Chaplin DJ,
Griffiths JR. Tumor response to hypercapnia and hyperoxia monitored
by FLOOD magnetic resonance imaging. NMR Biomed 1999;12:98–
33. Vaupel P, Hockel M, Mayer A. Detection and characterization of tumor
hypoxia using pO(2) histography. Antioxid Redox Signal 2007;9:1221–
34. Mason RP, Antich PP, Babcock EE, Constantinescu A, Peschke P, Hahn
EW. Non-invasive determination of tumor oxygen tension and local
variation with growth. Int J Radiat Oncol Biol Phys 1994;29:95–103.
35. Matsumoto K, Bernardo M, Subramanian S, Choyke P, Mitchell JB,
Krishna MC, Lizak MJ. MR assessment of changes of tumor in response
to hyperbaric oxygen treatment. Magn Reson Med 2006;56:240–246.
36. O’Connor JPB, Jackson A, Buonaccorsi GA, Buckley DL, Roberts C,
Watson Y, Cheung S, McGrath DM, Naish JH, Rose CJ, Dark PM, Jayson
GC, Parker GJM. Organ-specific effects of oxygen and carbogen gas
inhalation on tissue longitudinal relaxation times. Magn Reson Med
37. Gu Y, Bourke V, Kim JG, Constantinescu A, Mason RP, Liu H. Dynamic
response of breast tumor oxygenation to hyperoxic respiratory chal-
lenge monitored with three oxygen-sensitive parameters. Appl Opt
38. Kim JG, Zhao D, Constantinescu A, Mason RP, Liu H. Interplay of
tumor vascular oxygenation and tumor pO2observed using NIRS, ox-
ygen needle electrode, and19F MR pO2mapping. J Biomed Opt 2003;
39. Kodibagkar VD, Wang X, Pacheco-Torres J, Gulaka P, Mason RP. Proton
imaging of siloxanes to map tissue oxygenation levels (PISTOL): a tool
for quantitative tissue oximetry. NMR Biomed 2008;21:899–907.
40. Mason RP, Hunjan S, Constantinescu A, Song Y, Zhao D, Hahn EW,
Antich PP, Peschke P. Tumor oximetry: comparison of19F MR EPI and
electrodes. Adv Exp Med Biol 2003;530:19–27.
41. Zhao D, Constantinescu A, Hahn EW, Mason RP. Tumor oxygen dy-
namics with respect to growth and respiratory challenge: investigation
of the Dunning prostate R3327-HI tumor. Radiat Res 2001;156:510–
42. Zhao D, Ran S, Constantinescu A, Hahn EW, Mason RP. Tumor oxygen
dynamics: correlation of in vivo MRI with histological findings. Neo-
43. Baudelet C, Cron GO, Gallez B. Determination of the maturity and
functionality of tumor vasculature by MRI: correlation between BOLD-
MRI and DCE-MRI using P792 in experimental fibrosarcoma tumors.
Magn Reson Med 2006;56:1041–1049.
44. Bourke VA, Zhao D, Gilio J, Chang C-H, Jiang L, Hahn EW, Mason RP.
Correlation of radiation response with tumor oxygenation in the Dun-
ning prostate R3327-AT1 tumor. Int J Radiat Oncol Biol Phys 2007;67:
45. Zhao D, Constantinescu A, Chang C-H, Hahn EW, Mason RP. Correla-
tion of tumor oxygen dynamics with radiation response of the dunning
prostate R3327-HI tumor. Radiat Res 2003;159:621–631.
46. Al-Hallaq HA, Zamora M, Fish BL, Farrell A, Moulder JE, Karczmar GS.
MRI measurements correctly predict the relative effects of tumor oxy-
genating agents on hypoxic fraction in rodent BA1112 tumors. Int J
Radiat Oncol Biol Phys 2000;47:481–488.
364 Zhao et al.