Somme et al.

Cancer Imaging (2020) 20:70

https://doi.org/10.1186/s40644-020-00348-5

REVIEW Open Access

Usefulness of 18F-FDOPA PET for the

management of primary brain tumors: a
systematic review of the literature
François Somme1* , Laura Bender2, Izzie Jacques Namer1,3, Georges Noël4,5 and Caroline Bund1

Abstract
Contrast-enhanced magnetic resonance imaging is currently the standard of care in the management of primary
brain tumors, although certain limitations remain. Metabolic imaging has proven useful for an increasing number of
indications in oncology over the past few years, most particularly 18F-FDG PET/CT. In neuro-oncology, 18F-FDG was
insufficient to clearly evaluate brain tumors. Amino-acid radiotracers such as 18F-FDOPA were then evaluated in the
management of brain diseases, notably tumoral diseases. Even though European guidelines on the use of amino-
acid PET in gliomas have been published, it is crucial that future studies standardize acquisition and interpretation
parameters. The aim of this article was to systematically review the potential effect of this metabolic imaging
technique in numerous steps of the disease: primary and recurrence diagnosis, grading, local and systemic
treatment assessment, and prognosis. A total of 41 articles were included and analyzed in this review. It appears
that 18F-FDOPA PET holds promise as an effective additional tool in the management of gliomas. More consistent
prospective studies are still needed.
Keywords: F-DOPA, Glioma, Primary brain tumor, Systematic review

Background Differentiating low-grade and high-grade features of a

Management of primary brain tumors is based on primary brain tumor as well as identifying patients with
contrast-enhanced magnetic resonance imaging relapse and those with pseudo-progression are crucial in
(MRI). Despite progress in MRI as a perfusion and choosing the best treatment. Positron emission tomog-
diffusion technique, it has a number of limitations raphy with computed tomography (PET/CT) could be a
(mainly due to the disruption of the blood–brain tool to help reach these goals given the large number of
barrier), most particularly in differentiating recur- radioisotopes that can be used in various clinical
rence from post-therapeutic effects [1, 2]. Additional situations.
18
magnetic resonance spectroscopy was developed to Fluor-fluorodeoxyglucose (18F-FDG) PET/CT is
improve sensitivity and specificity. However, overlap widely used in oncology and can provide relevant infor-
between low-grade tumor values and those obtained mation, even in the management of primary brain tu-
for high-grade tumors has made this technique dis- mors [6, 7]. Nevertheless, a high rate of glucose
putable [3, 4]. Its use to differentiate relapse and metabolism in normal brain tissue (generating a poor
pseudo-progression remains under study [5]. signal-to-noise ratio with the tumor), tumors with low
glucose metabolism (such as low-grade gliomas) and
18
* Correspondence: [email protected]
F-FDG’s lack of specificity remain limitations [8]. For
1
Nuclear medicine Department, Hautepierre University Hospital, 1, rue these reasons, alternative PET radiotracers without these
Molière, F-67000 Strasbourg, France
Full list of author information is available at the end of the article

© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,
which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give
appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if
changes were made. The images or other third party material in this article are included in the article's Creative Commons
licence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commons
licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain
permission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.
The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the
data made available in this article, unless otherwise stated in a credit line to the data.
Somme et al. Cancer Imaging (2020) 20:70 Page 2 of 13

limitations have been evaluated quite recently in the Results

management of brain tumors. Study design
Amino-acid radiotracers have low uptake in normal A total of 222 articles were identified using the search
brain tissue. Therefore, they more easily reveal the strategy mentioned above. One hundred sixty-five arti-
brain’s progressive processes, such as neoplastic disease. cles were excluded for the following reasons: review arti-
There are several radiolabeled amino acids currently in cles, case reports, duplicates, irrelevant, abstract only,
use. The first of them was 11C-methionine (MET); how- animal or pediatric studies, and articles in a language
ever, its use is limited because carbon 11’s half-life is 20 other than English. A total of 41 articles were finally in-
min and it is reserved for PET centers with an on-site cluded in the review (see Fig. 1).
cyclotron unit. Other amino-acid radiotracers labeled The methodological quality of the studies included
with fluor 18 have been synthetized, such as [18F]-fluor- was evaluated using the Quality Assessment of Diagnos-
oethyl-tyrosine (18F-FET) and [18F]-L-dihydroxypheny- tic Accuracy Studies tool (QUADAS-2) [18] by two in-
lalanine (18F-FDOPA), which are easier to use in clinical dependent reviewers (LB and FS). The results are
routine because of their longer half-life (110 min). Stud- summarized in Fig. 2 and Fig. 3. They show that in most
ies have shown that 18F-FDOPA PET/CT is more accur- studies (n = 28, 68%) patient selection can potentially
ate than 18F-FDG PET/CT and is more sensitive in introduce bias. This was mainly due to the small sample
detecting primary or recurrent gliomas [9–11]. Several size of the population and the heterogeneity of the tu-
studies have shown that these different amino-acid ra- mors included. Histological findings were almost always
diotracers (18F-FET, 11C-MET and 18F-FDOPA) per- the reference standard defined in the studies included.
formed equally well in the visual assessment of primary However, only 13 studies (32%) were able to strictly cor-
brain tumors [12–14]. relate their results with histology. This could introduce
Moreover, the cellular transport mechanism of bias in those articles as well.
amino-acid radiotracers is active, and their uptake
seems to be correlated with the glioma grade. This is Disease detection
predominantly due to the increased transport of Several studies examined the sensitivity of 18F-FDOPA.
amino acids into tumor cells via an overexpression of Beuthien-Baumann et al. investigated 19 patients with
the amino-acid transport system L (LAT 1 and LAT suspected primary brain tumors. They obtained an over-
2). Youland et al. showed a statistically significant all sensitivity of 94% in correlation with histological or
positive correlation between 18F-FDOPA SUVmedian clinical and MRI follow-up [19]. The main limitation of
and LAT1 expression (p = 0.04) [15]. this study was the lack of correlation with histological
Amino-acid radiotracers such as 18F-FDOPA have an results: histological data were obtained for only eight pa-
impact on the management of patients with primary tients (42%).
brain tumors [16]. Its use is now recommended by joint For the detection of primary brain tumor, other radio-
EANM, EANO and RANO guidelines in several isotopes were compared with 18F-FDOPA. Chen et al.
indications [17]. However, it remains crucial that future included 81 patients with newly diagnosed or recurrent
studies standardize acquisition and interpretation brain tumors. All patients underwent both 18F-FDOPA
parameters. This systematic review of the literature and 18F-FDG imaging. Sensitivity, specificity, positive
attempts to summarize the data on the usefulness of predictive value (PPV), negative predictive value (NPV)
18
F-FDOPA PET for the diagnosis and management of and overall accuracy were calculated. For 18F-FDOPA,
primary brain tumors. the results were 98, 86, 95, 95 and 95%, respectively, and
for 18F-FDG, they were 61, 43, 78, 25 and 57%, respect-
ively [9]. Interestingly, with the criterion that any uptake
Methods above the background should be considered abnormal,
A systematic review was conducted with reference to the using only visual inspection provided a higher sensitivity
PRISMA statement and the current methodological lit- with 18F-FDOPA, but the specificity was as poor as that
erature. Because of the heterogeneity of studies, per- with 18F-FDG. Tripathi et al. prospectively included 15
forming a meta-analysis was not relevant. patients suffering from low-grade gliomas. Among them,
The PubMed, ScienceDirect and Cochrane databases 13 patients (87%) with primary and recurrent low-grade
were searched for relevant English-language articles gliomas had a visual uptake of 18F-FDOPA, whereas vis-
until July 2020. The search terms used were “DOPA” ual uptakes of 18F-FDG and 18F-FLT were observed in
AND “PET” AND (“gliomas” OR “brain tumor”). The only 54 and 31%, respectively. Moreover, with a better
literature was systematically reviewed by two inde- tumor/normal brain uptake ratio (SUVmax T/N, defined
pendent reviewers (LB and FS) using the Covidence by SUVmax of tumor divided by SUVmax of normal brain
tool. tissue), visual tumor delineation was easier with 18F-
Somme et al. Cancer Imaging (2020) 20:70 Page 3 of 13

Fig. 1 Flow chart of systematic review of the literature

FDOPA [11]. However, due to the small number of pa- provided comparable results [21]. The authors split the
tients included, the authors could not calculate sensitiv- patients into two groups: group 1 (n = 21), those patients
ity and specificity. Jacob et al. compared 18F-FDOPA, who underwent 18F-FDOPA and MRI followed by tumor
13
N-ammonia and 18F-FDG. The authors concluded that resection; group 2 (n = 70), those who lacked patho-
the sensitivity with 18F-FDOPA was substantially higher logical confirmation. For group 1, the sensitivity of
compared to the other radiotracers, but the figures and disease detection for PET was 95.2% compared to 90.5%
p-value were not provided. Furthermore, with only nine for MRI. Interestingly, in one case, increased 18F-
patients included, the number of cases was too small to FDOPA activity was clearly detected in an area of none-
draw statistical conclusions [10]. nhancing tumor, a finding that may have been missed if
Some authors have demonstrated that 18F-FDOPA MRI had been used alone. After a 7-month follow-up,
PET/CT can be compared favorably with contrast- imaging demonstrated significant tumor growth in this
enhanced MRI. Pafundi et al. published a prospective area (which also developed contrast enhancement) [21].
study including 10 patients and correlated 23 biopsy One problem reported in this study was the diagnosis of
samples with 18F-FDOPA PET uptake and contrast en- residual tumor using 18F-FDOPA after surgery. High
hancement on MRI images. The sensitivity and NPV of levels of amino-acid transport into cells are also de-
18
F-FDOPA PET and T1-CE MRI were 72.7% versus scribed for macrophages, which are activated after sur-
27.3 and 14.3% versus 5.9%, respectively (p-value not gery. This might explain the mild 18F-FDOPA tracer
available (NA)). Both modalities had 100% specificity uptake that can be present around resection cavities and
and PPV. Overall accuracy was 73.9% for PET and 30.4% should be interpreted with caution.
18
for T1 MRI (p-value NA) [20]. Ledezma et al. published F-FDOPA PET/CT could be helpful in the initial
a retrospective study, which included 91 patients and diagnosis of primary brain tumors, in addition to MRI.
Somme et al. Cancer Imaging (2020) 20:70 Page 4 of 13

Fig. 2 Summary of the risk of bias and applicability concerns according to the QUADAS-2 tool
Somme et al. Cancer Imaging (2020) 20:70 Page 5 of 13

Fig. 3 Graphic presentation of the risk of bias and applicability concerns according to the QUADAS-2 tool

These two imaging modalities demonstrated that they prospective study with 10 patients. A significant correl-
could contribute complementary information. Moreover, ation was found between SUVmax and tumor grade
in some cases where MRI remains uncertain (e.g., none- across biopsy samples (p = 0.0005). A significant differ-
nhancing gliomas), the relatively new contribution of ence was found between grade II and grade IV disease
PET/MRI fused images could be a tool to obtain a non- (p = 0.008) and between grade III and grade IV (p =
invasive diagnosis [22]. Table 1 provides sensitivity, 0.024) but not between grade II and grade III (p = 0.174)
specificity, PPV, NPV and accuracy of 18F-FDOPA in the [20]. Similar results were also found in several other
detection of primary brain tumors in each study in- studies. Youland et al. correlated MRI and 18F-FDOPA
cluded in this review. PET/CT with tumor grade in 13 patients. Regions of
MRI contrast enhancement correlated with the presence
Grading and correlation with histopathological features of high-grade recurrent tumor (p = 0.03), with 63%
In addition to tumor grade, the last World Health sensitivity and 80% specificity. A SUVmax T/N ratio
Organization (WHO) classification of gliomas dating greater than 2.0 (p = 0.0004) and SUVmax above 2.0 (p =
from 2016 emphasizes integrating molecular parameters. 0.002) correlated with high-grade recurrent tumor with
In clinical practice, gliomas are split into two main sensitivity at 85 and 80%, and specificity at 93 and 60%,
groups with different prognoses: low-grade tumors respectively [24]. The inclusion of low- and high-grade
(grade II) and higher-grade tumors (grades III and IV). gliomas in this series may influence the results, as the
Still, noninvasive categorization of tumor grade with radiographic appearance of recurrent disease may be in-
image findings remains a challenge. 18F-FDOPA PET fluenced by histology. Moreover, the low frequency of
could be helpful in this matter. negative biopsies and the fact that the spatial distribu-
Chiaravalloti et al. conducted a prospective study on tion of MRI contrast enhancement is not clearly com-
97 patients examining 18F-FDOPA uptake after surgery pared with 18F-FDOPA avidity can limit the robustness
and radiotherapy. SUVmax and SUVmean were signifi- of this work. Janvier et al. found that all SUV-derived in-
cantly correlated with tumor grade (p < 0.05) [23]. As a dices (SUVmax, SUVmean, T/N ratios and tumor/striatum
limit, most of the patients underwent PET/CT a long ratios (T/S, defined by tumoral uptake divided by stri-
time after surgery (mean, 41.48 months), when abnormal atum uptake)) were significantly correlated with tumor
18
F-FDOPA uptake related to post-therapeutic inflam- grade in 31 patients. The two best-correlated indices
mation had clearly decreased. Pafundi et al. published a were SUVmeanT/N (p = 0.001) and SUVmeanT/S (p =

Table 1 Performance of 18F-FDOPA in disease detection

Authors Year Patients (#) Sensitivity (%) Specificity (%) PPV (%) NPV (%) Accuracy (%) Optimal cut-off
Pafundi et al. 2013 10 72.7 100 100 14.3 73.9 NA (visual analysis)
Ledezma et al. 2009 21 95.2 NA NA NA NA NA (visual analysis)
Tripathi et al. 2009 15 100 NA NA NA NA NA (visual analysis)
Chen et al. 2006 81 98 86 95 95 95 T/S > 0.75
Beuthien-Baumann et al. 2003 19 94 NA NA NA NA NA (visual analysis)
NA Not available, NPV Negative predictive value, PPV Positive predictive value, T/S Tumoral uptake divided by striatum uptake
Somme et al. Cancer Imaging (2020) 20:70 Page 6 of 13

0.003). SUVmeanT/N had a sensitivity of 71%, a specifi- as mitotic activity (Ki-67 index) and the IDH mutation.
city of 100%, and an area under curve (AUC) of 0.85 for Pafundi et al. found a significant correlation between
an optimal threshold of 1.33 [25]. These results were SUVmean and cellularity (p = 0.01) and an approaching
significant considering only newly diagnosed gliomas significance with the Ki-67 index (p = 0.053). They used
since the authors failed to perform a significant sub- SUVmean because it may be more representative of the
group analysis (newly diagnosed opposed to recurrent entire cellular area used in calculations of both cellular-
gliomas), mainly because they had only six recurrences ity and Ki-67 than SUVmax [20]. Verger et al. included
in their study. Patel et al. retrospectively included 45 43 patients with grades II and III gliomas before surgery.
patients and correlated the uptake with numerous Surprisingly, patients with the IDH mutation showed
pathological data. They demonstrated that the SUVmax higher 18F-FDOPA T/N (1.6 vs. 1.2; p = 0.046) and T/S
T/N was significantly higher in high-grade versus lower- ratios (0.9 vs. 0.6; p = 0.024) than patients without the
grade glioma (p = 0.0002). The authors did not find a IDH mutation. The authors discussed hypotheses to ex-
significant correlation between 18F-FDOPA uptake and plain this unexpected result. Changes in metabolic path-
the IDH mutation (p = 0.022) or the MGMT methylation ways including increases in free amino acids, which lead
(p = 0.66). However, the use of SUVmax could limit these to an increased activity of the amino-acid transporter
results since it does not take into account the heterogen- and a better differentiation of gliomas were the main
eity of the tumor [26]. reasons suggested. Moreover, there was a significant
Schiepers et al. evaluated static and dynamic 18F- positive correlation between 18F-FDOPA uptake and Ki-
FDOPA PET in 37 patients with brain tumors (33 pri- 67 expression (p = 0.02) but not with presence/absence
mary brain tumors). Statistically significant differences of 1p/19q co-deletion or ATRX loss of expression [30].
(p < 0.01) for volume distribution of the radiotracer were However, the retrospective design, the exclusion of
found between newly diagnosed high-grade tumors and glioblastoma and PET methodology (the authors used
low-grade tumors and between newly diagnosed high- two-dimensional regions of interest to obtain the differ-
grade tumors and tumors with post-treatment changes ent ratios) were some of the limitations. In 29 patients
[27]. The main limitation of this study was the hetero- with recurrent high-grade gliomas, Karavaeva et al.
geneity of the tumors included (13 grade II, 10 grade III found a significant correlation between 18F-FDOPA PET
and 10 grade IV). The contribution of dynamic PET SUVmean and mitotic activity (p = 0.0362) [31]. It is im-
studies remains uncertain. A more recent study on 33 portant to note that biopsy samples were taken from
patients who underwent both static and dynamic 18F- MRI contrast-enhanced areas and were afterwards corre-
FDOPA PET/CT showed that static PET/CT was able to lated with PET uptake. Another study reported similar
determine the disease grade with 94% sensitivity and results using additional dynamic PET acquisitions. Ginet
66% specificity for a SUVmean threshold of 2.5. Interest- et al. analyzed the correlation between the IDH mutation
ingly, all PET/CT indices were significant to distinguish and the 1p/19q co-deletion with numerous static (SUV-
between low-grade glioma and high-grade glioma in max, SUVmean, T/S, T/N and MTV) and dynamic (time
newly diagnosed tumors. Sensitivity and specificity were to peak and slope) PET indices [32]. They found no cor-
90 and 70%, respectively. Considering recurrent tumors, relation for static parameters but a significant correl-
only the SUVmean index was significant (p < 0.001) to ation for both dynamic parameters. The best index was
distinguish grade with 100% sensitivity and specificity. time to peak, both for the IDH mutation (p < 0.001,
However, dynamic imaging did not significantly improve AUC = 0.789) and for 1p/19q co-deletion (p = 0.034,
the diagnosis compared to static parameters [28]. AUC = 0.679). Still, three main applicability concerns re-
Even when MRI evaluation is limited, notably in cases main due to the use of carbidopa for some patients, the
of nonenhancing tumors, 18F-FDOPA PET/CT contrib- combination with MRI in cases of no radiotracer uptake
uted useful information. Bund et al. analyzed 33 patients (17% in this study) and a selection bias (patients were in-
with nonenhancing primary brain tumors. An optimal cluded only if both molecular data were known and a
threshold of SUVmax T/N = 2.16 (AUC = 0.87) discrimi- dynamic PET acquisition had been made). Moreover, it
nated low-grade from high-grade gliomas with 60% sen- is worth noting that a recent study reported by Girard
sitivity, 100% specificity, 100% PPV, and 83.3% NPV et al. sought to improve the time frame binning for dy-
(p < 0.01) [29]. Moreover, the authors reported that 18F- namic 18F-FDOPA PET imaging [33]. The authors used
FDOPA was also useful in the subgroup of low-grade a three-compartment model, which is the most com-
gliomas. Indeed, it was able to discriminate between dys- monly used for full kinetic analysis of PET, and com-
embryoplastic neuroepithelial tumor and grade II oligo- pared five different time samplings in 14 patients. They
dendroglioma (p < 0.01). reported that the average K1 value obtained by the 8 ×
A number of studies have evaluated the correlation of 15 s–2 × 30 s–2 × 60 s–3 × 300 s time sampling was the
18
F-FDOPA PET with precise molecular features, such closest to the target K1 value. Yet, variations in
Somme et al. Cancer Imaging (2020) 20:70 Page 7 of 13

methodological factors such as 18F-FDOPA dose, a non- gliomas. The metabolic tumor volume (MTV) and T/N
TOF PET system, image reconstruction or post-filtering failed to differentiate IDH-wildtype from IDH-mutant
could bias K1 estimates. (p > 0.5 for both) when considering the whole popula-
However, four studies highlighted various discrepan- tion. Nevertheless, when considering only diffuse gli-
cies. Considering the radiotracer’s uptake and the grade, omas (with the exclusion of glioblastomas), MTV was
three authors reported no correlation between those two higher in IDH-mutant gliomas (p = 0.002) and a trend
parameters. Fueger et al. found a significant correlation was observed for the T/N ratio (p = 0.1) [37]. Cicone
between only newly diagnosed tumors and not in recur- et al. replied that MTV could not be considered as an
rent ones. Fifty-nine patients were analyzed. There was a uptake parameter and highlighted the trend found for T/
significant correlation between uptake and grade and be- N in the previous study [38].
tween uptake and the proliferation rate (Ki-67 index) Considering only newly diagnosed primary brain tu-
only for newly diagnosed tumors. Uptake was signifi- mors, prior to any treatment, the data suggest that 18F-
cantly higher in high-grade than in low-grade tumors for FDOPA PET is able to discriminate between low- and
newly diagnosed tumors (p = 0.005) but not for recurrent high-grade gliomas. Table 2 presents 18F-FDOPA PET
tumors (p = 0.22). Similarly, uptake correlated signifi- sensitivity and specificity data to discriminate between
cantly with the Ki-67 index in newly diagnosed tumors low-grade gliomas and high-grade gliomas in each study
(p = 0.001) but not in recurrent ones (p = 0.41) [34]. The included in this review.
main explanation reported by the authors was the wide
range of blood–brain barrier breakdown indicators, de-
pending on previous treatments, when considering re- Target volume delineation and radiation treatment
current tumors. This could be why correlation in these monitoring
cases is relatively more uncertain. Chen et al. did not Several studies showed that 18F-FDOPA PET could be a
find a significant difference between uptake levels in 48 useful examination in the diagnosis of primary tumor
high-grade tumors and 18 low-grade tumors (p = 0.40) (paragraph 3.2). The objective of using 18F-FDOPA PET/
in a study of their 81 patients [9]. In a study evaluating CT was then consistent with trying to include it in the
the value of 18F-FDOPA PET in cases of nonenhancing management of local treatment (biopsy planning,
MRI primary brain tumors, Todeschi et al. prospectively radiotherapy).
included 20 patients [35]. They reported an average Despite the lack of recommendations defining the bio-
SUVmax of 2.18 for high-grade and 2.025 for low-grade logical target volume [17], several studies compared the
tumors, with no significant difference (p = 0.64). delineation of target volumes obtained with MRI and
18
Regarding the correlation between uptake and molecu- F-FDOPA PET/CT. Pafundi et al. prospectively in-
lar data, Cicone et al. pointed out a number of contra- cluded 10 patients and correlated volume definition with
dictions. The authors studied 33 patients with gliomas. histological findings. Each patient underwent one to
They did not find significant correlations between PET three biopsies in concordant and discordant areas,
uptake parameters and the IDH mutational or the 1p/19 uptaking 18F-FDOPA and contrast-enhancing MRI. For
co-deletion status, neither for SUVmax (p = 0.56 and p = the six patients with T1-contrast enhancement, the per-
0.29 respectively) nor SUVmean T/N (p = 0.32 and p = centage of 18F-FDOPA PET volume with a T/N > 2.0
0.82, respectively). The main limitation of this study was outside the MRI contoured volume was on average
that almost all of the PET/CT examinations (94%) were 47.3% (range, 15.1–81.0%). The T2/FLAIR volume out-
made after the surgical procedure. This could have re- side the high-grade threshold 18F-FDOPA PET uptake
moved a portion of disease with uptake characteristics volume was on average 87.3% (range, 70.6–99.9%).
different from those of the remaining disease [36]. Zara- These results suggest that there could be an impact of
gori et al. discussed this controversial result. The authors delivering a higher radiation dose into the volume delin-
conducted a new analysis based on another cohort of 58 eated with 18F-FDOPA PET [20].

Table 2 Optimal indices and cut-off to discriminate between low- and high-grade gliomas
Authors Year Patients (#) Sensitivity (%) Specificity (%) Optimal ratio used
Patel et al. 2018 45 70 78 SUVmax T/N > 1.7
Youland et al. 2018 13 85 93 SUVmax T/N > 2
Bund et al. 2017 33 60 100 SUVmax T/N > 2.16
Janvier et al. 2015 31 71 100 SUVmean T/N > 1.33
Nioche et al. 2013 33 94 66 SUVmean > 2.5
T/N Tumor uptake divided by normal brain uptake.
Somme et al. Cancer Imaging (2020) 20:70 Page 8 of 13

Other retrospective studies highlighted the low correl- coefficient (ADC) within the 18F-FDOPA-defined
ation between target volumes defined with MRI and tumor volume. PET/CT volumes were significantly
PET/CT. Kosztyla et al. analyzed the interobserver vari- larger than ADC volumes (p = 0.0009). More import-
ability and volume localizations, especially in cases of re- antly, considering the overlap between these two
current gliomas. Five observers contoured gross tumor volumes, most patients presented with no or only
volumes (GTVs) using MRI and PET/CT, and interob- modest overlap [41]. The authors suggested that re-
server variability were quantified by the percentage of gions of minimum ADC may primarily be associated
volume overlap. The mean interobserver volume over- with tumor ischemia, but there was no correlation
laps for PET GTVs and MRI GTVs were not signifi- with histological findings. This proposal needs to be
cantly different, 42% versus 41%, respectively (p = 0.67). evaluated with specific ischemia radiotracers (such as
18
The mean consensus volume was significantly larger for F-fluoromisonidazole). Approximately the same cor-
PET GTVs (58.6 cm3) than for MRI GTVs (30.8 cm3) relation was found with perfusion-weighted MRI.
(p = 0.003). Moreover, the percentage of the recurrence Cicone et al. defined tumor volume semiautomatically
volume that extended beyond the PET GTV (52%) was on 18F-FDOPA PET (threshold value, 1.6 over back-
significantly less than the percentage that extended be- ground) and was compared with the relative cerebral
yond the MRI GTV (62%), (p = 0.04) [39]. Kazda et al. blood volume (rCBV) defined by perfusion-weighted
retrospectively included eight patients. The aim was to MRI in 44 patients. 18F-FDOPA volume greatly
compare treatment planning with and without the in- exceeded rCBV volume (p < 0.00001). A median over-
corporation of 18F-FDOPA PET imaging. For patients lapping volume of 0.28 mL resulted in a 1.38% overall
with contrast enhancement on T1-MRI (n = 5), bio- median spatial congruence. Interestingly, high-grade
logical target volumes (BTV60Gy) were less than 4.4 gliomas had a significantly larger 18F-FDOPA volume
times as large as GTV60Gy; the planning target volume than low-grade gliomas (p = 0.023), which was not
(PTV60Gy) including MRI + PET ranged from being the significant with rCBV volume (p = 0.071) [42]. No tar-
same to 1.8 times larger than PTV60Gy using MRI only. geted biopsies were undertaken to confirm the results
For non-contrast-enhanced patients (n = 3), BTV60Gy presented. Moreover, one might argue that the object-
ranged from 48 to 202 times smaller than the GTV60Gy ive of perfusion-weighted MRI is not to define the
(composed of the FLAIR MRI volume), while the result- precise tumor extent but instead to identify subre-
ing PTV60Gy ranged from 3.2 to 72 times smaller. Inter- gions of high-grade disease.
estingly, after inclusion of 18F-FDOPA PET biologic A recent paper evaluated the feasibility of dose-
imaging, the average 60-Gy isodose volumes for the five painted radiation therapy using 18F-FDOPA PET/CT.
patients with contrast enhancement increased 1.3-fold The authors included 10 patients with a high-grade
and decreased 2.5-fold in the three patients without con- glioma and analyzed the irradiation of the PTV with
trast enhancement. All priority dose volume constraints dose-painting (using MTV delineated with different
for PTV60Gy (V100% ≥ 95% and V110% < 0.5 cc) were met thresholds). They demonstrated that the median vol-
in both treatment plans for all patients, and all plans ume of PTV receiving at least 95% of the prescribed
met critical organs at risk constraints (according to the dose was 99.6% with and 99.5% without dose painting
Radiation Therapy Oncology Group) [40]. In the two (p = 0.5). There was no significant difference when
preceding studies listed, the authors reported problems considering the organs at risks as well. As limitations,
with the PET volume delineation. The physiological up- this study included a small number of patients and
take of 18F-FDOPA in the basal ganglia may well inter- the authors only used cell density to calculate the
fere in the clear delineation of gliomas located near dose; they did not include the partial volume effect or
these structures. Moreover, postsurgical changes around the tumor hypoxia [43].
the resection cavity can also exhibit radiotracer uptake Several studies highlighted great differences between
and may have modified volume delineation. This may volumes defined with MRI and with 18F-FDOPA PET.
have added uncertainty to the study contours. Yet, study has demonstrated better outcome using
Certain authors also compared the target volume amino-acid PET volume delineation. One prospective,
definition of 18F-FDOPA PET/CT with perfusion- multicenter, randomized phase II trial is currently in
weighted or diffusion-weighted MRI, with poor results progress in an attempt to give an objective answer to
in terms of volume overlap. Rose et al. prospectively this matter (NOA 10/ARO 2013–1) [44]. It is de-
analyzed 15 patients with newly diagnosed, confirmed signed to test whether radiotherapy target volume
high-grade gliomas. Two volumes were defined: re- delineation based on FET-PET improves progression-
gions of maximum 18F-FDOPA uptake within the free survival (PFS) in patients with recurrent glioblast-
tumor volume (voxels with the 20% highest SUV T/N oma treated with re-irradiation, compared to target
ratio) and regions of minimum apparent diffusion volume delineation based on MRI (NCT01579253).
Somme et al. Cancer Imaging (2020) 20:70 Page 9 of 13

Chemotherapy and targeted therapy assessment issues remain [49]. 18F-FDOPA PET may help differenti-
Considering systemic treatment monitoring in high- ate progression from post-treatment changes.
grade gliomas, there is currently a lack of data due to Many studies have attempted to evaluate the potential
the limited effectiveness of using chemotherapy or tar- value of 18F-DOPA PET in this matter. A recent article
geted drugs in primary brain tumors. Multiple factors by Humbert et al. evaluated the impact of PET/CT
explain this: the impermeability of the blood–brain bar- through a multidisciplinary brain tumor board. A first
rier is one of the most frequently suggested. However, decision was made with clinical and MRI data, without
there have been promising studies, especially in the knowing the results of 18F-DOPA PET/CT; then a sec-
evaluation of bevacizumab therapy. Bevacizumab is a re- ond decision was made with the inclusion of PET/CT.
combinant humanized monoclonal antibody that blocks The authors demonstrated that this technique was able
angiogenesis by inhibiting vascular endothelial growth to modify the diagnosis or the therapeutic strategy in up
factor A (VEGF-A). Schwarzenberg et al. showed that to 33.3% of cases when considering patients with glio-
18
F-FDOPA PET/CT could identify early responders blastomas. The main limitation was the lack of correl-
after 2 weeks of treatment with bevacizumab. Thirty pa- ation with pathological data (9.4% of patients) [50].
tients were prospectively included. In multivariate ana- Youland et al. correlated 37 stereotactic biopsies and
lysis, the 18F-FDOPA MTV (defined by all voxels that histological results from 13 patients according to areas
fall within an SUV threshold determined by the mean of increased 18F-FDOPA uptake and areas of MRI con-
SUV of the contralateral striatum) at 2 weeks (p < 0.05) trast enhancement. To distinguish between radionecrosis
and MTV changes at 6 weeks (p < 0.05) were the most and recurrence, MRI sensitivity and specificity were 52
significant predictors of overall survival (OS). 18F- and 50%, respectively, 18F-FDOPA PET sensitivity and
FDOPA MTV change at 2 weeks (p < 0.01) was also the specificity were 82 and 50%, respectively [24]. Herrmann
most significant predictor of progression-free survival et al. analyzed 110 patients retrospectively. Images were
[45]. However, the patients included had varying num- correlated with histological data in 41 (37.3%) cases and
bers of recurrences (median, 1.77). Since the number of clinical and imaging follow-up in 69 (62.7%) cases. The
recurrences is also predictive of survival, this could authors did not separate the two groups in their results.
interfere with the results. Wardak et al. also identified a Overall, visual analysis resulted in sensitivity, specificity,
significant correlation between 18F-FDOPA PET/CT and accuracy, and positive and negative predictive values of
overall survival. In their study, information from kinetic 85.2, 72.4, 81.8, 89.6 and 63.4%, respectively, considering
parameters (either from 18F-FLT alone, 18F-FDOPA 18
F-FDOPA PET/CT. All PET indices (SUVmax, SUV-
alone, or both together; best adjusted R2 = 0.83) showed mean, T/N ratios and T/S ratios) were significantly higher
better predictive results than standardized uptake values in progressive than in nonprogressive patients. Interest-
(best adjusted R2 = 0.25) [46]. Consequently, the authors ingly, semiquantitative image analysis did not improve
highlighted the need for dynamic PET/CT studies. This accuracy over visual PET/CT image analysis, with the
requirement leads to modifications in the acquisition AUC ranging from 0.77 to 0.82 versus 0.82 for visual
procedures: starting the image acquisition simultan- analysis [51]. Nevertheless, several limitations remain in
eously with radiotracer injection and a slightly longer ac- this study: the retrospective design, the fact that different
quisition time (about 35 min for dynamic images versus PET systems were used, which might have affected SUV
20 min for the static acquisition). measurements (even if checks were made with phan-
PET indices such as metabolic tumor volume could be toms) and a potential selection bias since patients were
factors that predict early responders to systemic drugs, included based on a positive MRI diagnosis of recurrent
but larger prospective studies are required to confirm disease. Karunanithi et al. prospectively included 35 pa-
this assumption. tients comparing 18F-FDOPA PET/CT and MRI to de-
tect recurrence. Sensitivity, specificity and overall
accuracy were, for 18F-FDOPA PET/CT, 100, 88.9 and
Diagnosis between recurrence and post-therapeutic 91.1%, respectively, compared to, for contrast-enhanced
changes MRI, 92.3, 44.4 and 80%, respectively [52].
18
The ability to distinguish between progression and post- F-FDOPA PET/CT was also compared with other ra-
treatment changes (mostly pseudo-progression within diotracers developed for SPECT/CT and PET/CT. The
the first 12 weeks after completion of chemoradiotherapy most frequently used was obviously 18F-FDG. Karuna-
or radionecrosis) is a major issue in the management of nithi et al. prospectively compared 28 patients who
primary brain tumor. Indeed, it is recognized that MRI underwent PET/CT with both 18F-FDOPA and 18F-FDG.
has limitations in diagnosing early recurrence [47]. How- The sensitivity, specificity and accuracy of 18F-FDG
ever, RANO (response assessment in neuro-oncology) PET/CT were 47.6, 100 and 60.7%, respectively, and
criteria are always based on MRI [48], even though those for 18F-FDOPA PET/CT were 100, 85.7 and 96.4%,
Somme et al. Cancer Imaging (2020) 20:70 Page 10 of 13

respectively. The difference in the findings between 18F- all PET parameters, except time to peak, were correlated
FDG PET/CT and 18F-FDOPA PET/CT was significant with progression-free survival. Thus, the authors showed
(p = 0.0005) [53]. 18F-FDOPA also showed better per- that none of the dynamic parameters provided any add-
formance than radiotracers developed for SPECT/CT, itional diagnostic information. As limitations, Zaragori
such as 99mTC-GH. Karunanithi et al. prospectively et al. were able to obtain a pathological confirmation of
compared these two tracers in 30 patients. Sensitivity, recurrence for only four patients and the authors mixed
specificity, and accuracy were 86.4, 62.5 and 80% for low- and high-grade gliomas that may have benefited
99m
Tc-GH SPECT/CT and 100, 87.5 and 96% for 18F- from different therapeutic strategies, which could have
FDOPA PET/CT, respectively [54]. All results reported influenced survival [56].
18
by Karunanithi et al. in their three studies may present F-FDOPA PET should be considered as a comple-
certain limitations: the sample size was relatively small, mentary tool to assess real progression when MRI re-
especially considering low-grade gliomas, and a confirm- mains uncertain. Table 3 collects the sensitivity,
ation of recurrence by histological findings was obtained specificity and accuracy of 18F-FDOPA PET in the diag-
for a small number of cases. nosis of recurrence and post-therapeutic changes in each
More recently, Fraioli et al. took an interest in the in- study included in this review.
novative PET/MRI technique [55]. They prospectively
compared 40 residual tumor volumes (obtained from Prognostic value
PET and MRI) and several PET (SUVmax, T/N and T/S) The ability to sort patients into subgroups of different
and MRI (relative cerebral blood volume and relative prognoses is important. It can help the clinician adapt or
cerebral blood flow) parameters. PET volumes were sig- even change treatment. In this indication, 18F-FDOPA
nificantly larger than those obtained from MRI, both for PET could have a place to claim. Many authors have ex-
low-grade and high-grade tumors (p = 0.02 and p = amined this issue, resulting in several PET indices posi-
0.0002, respectively). Both modalities were concordant tively correlated with survival. In this section, we will
in 37 patients (93%). A single-modality analysis of PET summarize the most significant indices objectified in
imaging demonstrated an AUC of 0.94. A combined studies.
multiparameter PET/MRI approach resulted in an AUC Villani et al. prospectively included 50 grade II gli-
of 0.99, showing that MRI and F-DOPA are complemen- omas. After a median follow-up of 16 months, on multi-
tary modalities for assessment of tumor burden. How- variate analysis, a maximum standardized uptake value
ever, there was no pathological confirmation of residual greater than 1.75 (p = 0.005) was an independent pre-
tumor and a dichotomous evaluation of presence or ab- dictor of disease progression [57]. Correlation with over-
sence of active disease was made (without taking into all survival was not calculated because patient follow-up
consideration other important relevant information). was too short. On multivariate analysis, another study
Lastly, Zaragori et al. assessed dynamic acquisition in considering 12 patients with low-grade gliomas demon-
this issue by evaluating the predictive value of static strated a significant correlation between follow-up status
(SUVmax, SUVmean, T/S and T/N) and dynamic param- (stable versus disease progression at 1 year) and T/N
eters (time to peak and slope) in terms of recurrence with a cut-off > 1.7 (p = 0.05 [58];. In addition to the
and survival. Except time to peak, all the PET parame- small population examined in this study, there was a
ters studied were significant univariate predictors of gli- majority of oligodendroglioma cases (eight patients,
oma recurrence (p < 0.001) and the T/S ratio was the 67%). As already shown in the literature, this histological
sole significant parameter in the multivariate analysis. subtype may have a specific presentation pattern. Indeed,
No indices were predictive of overall survival, whereas it may show increased amino-acid uptake and high rCBV

Table 3 Performance of 18F-FDOPA in discriminating recurrence from post-therapeutic effects

Authors Year Patients (#) Sensitivity (%) Specificity (%) PPV (%) NPV (%) Accuracy (%) Cut-off
Zaragori et al. 2020 51 97.1 94.1 NA NA 96 T/S > 1
Youland et al. 2018 13 82 50 NA NA NA T/N > 2.0
Karunanithi et al.a 2014 30 100 87.5 NA NA 96 T/S > 0.6
a
Karunanithi et al. 2013 35 100 88.9 NA NA 97.1 NA (visual)
Herrmann et al. 2013 110 85.2 72.4 89.6 63.4 81.8 T/N > 1.81
Karunanithi et al.a 2013 28 100 85.7 NA NA 96.4 T/N > 1.3
NA Not available, NPV Negative predictive value, PPV Positive predictive value, T/N Tumor uptake divided by normal brain uptake, T/S Tumor uptake divided by
striatum uptake
a
: the results of these studies are based on the same patient population
Somme et al. Cancer Imaging (2020) 20:70 Page 11 of 13

Table 4 Optimal indices and cut-off that correlated with prognosis

Authors Year Patients (#) Population Optimal index and cut-off P-value
Chiaravalloti et al. 2019 133 II = 68, III = 34, IV = 31 SUVr > 1.37 0.01
Isal et al. 2018 20 II = 13, III = 7 SUVmax T/N > 1.8 0.04
Rossi Espagnet et al. 2016 12 II = 12 SUVmax T/N > 1.7 0.05
Villani et al. 2015 50 II = 50 SUVmax > 1.75 0.005
Dowson et al. 2014 9 IV = 9 ΔSUVmax > 4.74 0.002
Herrmann et al. 2014 110 III = 33, IV = 77 SUVmean T/S > 1.06 < 0.001
Karunanithi et al. 2014 33 I = 2, II = 9, III = 6, IV = 16 SUVmax T/N > 1.51 0.005
SUVr Tumor uptake divided by contralateral occipital uptake, T/N Tumor uptake divided by normal brain uptake, T/S Tumor uptake divided by striatum uptake.

values that are not related to tumor grade but more con- limitations remain: a high proportion of glioma without
sistently related to 1p/19q co-deletion [59]. Another radiotracer uptake (n = 41) and a large interval between
retrospective study was conducted in 27 low-grade gli- the surgical intervention and the PET/CT (low-grade gli-
omas [60]. The authors analyzed the rates of change in oma could have switched to a more malignant grade).
FLAIR volume and in 18F-FDOPA SUVmax normalized Several authors tried to correlate 18F-FDOPA PET/
to the basal ganglia (nSUVmax). General linear models CT with MRI findings and survival. Isal et al. com-
were used to integrate clinical information (age and pared T/N with velocity of diameter expansion
treatment) with MRI and PET measurements to predict (VDE), calculated on MRI, a known prognostic factor
malignant transformation. A model using age, treatment, [63]. A ratio higher than 1.8 was significantly more
rate of change in FLAIR volume and in radiotracer up- frequent in patients with a VDE < 4 mm compared to
take predicted a malignant transformation within 6 those with a VDE ≥ 4 mm (45% vs 0%, p = 0.04) [64].
months (p = 0.0248). Moreover, only the rate of change The tumor growth rate was chosen as a surrogate for
in radiotracer uptake was correlated with overall survival clinical course in this study, since the overall survival
in the multivariate analysis (p = 0.0033). The limited could not be obtained owing to the long clinical
number of patients, the retrospective nature of the study course of low-grade glioma.
and the fact that PET and MRI scans were not made at When considering only recurrent gliomas, Karunanithi
a specific controlled interval could limit the applicability et al. prospectively included 33 patients. After a median
of this study. follow-up of 20.2 months, on multivariate analysis, only
The prognostic impact of 18F-FDOPA PET was also size of the recurrent tumor on MRI (p = 0.002) and the
highlighted when considering all grades of gliomas. Patel T/N ratio of 18F-FDOPA PET (p = 0.005) were found to
et al. demonstrated that age (p = 0.001) and the meta- be independent predictors of survival [65]. Another large
bolic tumor volume on PET (p = 0.016, using a SUVmax retrospective study analyzed 110 patients with a median
T/N threshold) were correlated with the 2-year overall follow-up of 34.9 months. All PET indices were signifi-
survival time, in multivariate analysis [26]. Dowson et al. cant predictors of progression-free survival, with the
studied radiotracer uptake in nine patients, at baseline, mean lesion-to-T/N ratio providing the best discrimin-
immediately before tumor resection and 12 weeks after ation (p < 0.001). Conversely, none of the parameters in-
resection. The results demonstrated that a decrease in vestigated were predictive of overall survival [51].
18
F-FDOPA uptake (ΔSUVmax) is a predictor of extended When examining all the studies included in this re-
survival (p = 0.002) [61]. The population size was very view, it appears that 18F-FDOPA PET can help stratify
small in this study and could limit its statistical power. patients into subgroups of different prognoses. Several
Nevertheless, a recent study reported results that are PET indices were exploited; the most frequently used is
more restrained. Chiaravalloti et al. retrospectively in- SUVmax T/N. However, there is still a huge diversity of
cluded 133 primary brain tumors [62]. They correlated PET indices used, which limits comparison between
the OS and the PFS with the SUVmax and the SUVr (de- studies. Nevertheless, these results need to be confirmed
fined by the SUVmax of the tumor divided by the SUVmax by larger prospective studies. We have summarized the
of the contralateral occipital region). In the whole co- best indices and cut-offs found in each study in Table 4.
hort, the uptake was significantly correlated with OS
(p = 0.01) but not with PFS. These two indices were cor- Conclusion
related with OS and PFS (p = 0.03 and p = 0.007, respect- Due to the poor prognosis of gliomas, especially consid-
ively) for grade II gliomas and no correlations were ering high-grade tumors, their management remains a
found for grade III and grade IV gliomas. Several huge challenge. It is widely accepted that 18F-FDOPA
Somme et al. Cancer Imaging (2020) 20:70 Page 12 of 13

PET can provide useful information in terms of initial 5. Zhang H, Ma L, Wang Q, Zheng X, Wu C, Xu B-N. Role of magnetic
diagnosis and the extent of gliomas in the context of re- resonance spectroscopy for the differentiation of recurrent glioma from
radiation necrosis: a systematic review and meta-analysis. Eur J Radiol. 2014;
current tumors. This systematic review gathers the 83:2181–9.
current data reported in the literature regarding the use 6. Yoon JH, Kim J, Kang WJ, Sohn C-H, Choi SH, Yun TJ, et al. Grading of
of 18F-FDOPA in the other steps of primary brain tumor cerebral glioma with multiparametric MR imaging and 18F-FDG-PET:
concordance and accuracy. Eur Radiol. 2014;24:380–9.
management. Nevertheless, standardization of acquisi- 7. Santra A, Kumar R, Sharma P, Bal C, Kumar A, Julka PK, et al. F-18 FDG PET-
tion and interpretation parameters (as addressed by CT in patients with recurrent glioma: comparison with contrast enhanced
EANM, SNNMI and EANO) remains essential. MRI. Eur J Radiol. 2012;81:508–13.
8. Dankbaar JW, Snijders TJ, Robe PA, Seute T, Eppinga W, Hendrikse J, et al.
Abbreviations The use of (18)F-FDG PET to differentiate progressive disease from
MRI: Magnetic resonance imaging; PET/CT: Positron emission tomography treatment induced necrosis in high grade glioma. J Neurooncol. 2015;125:
with computed tomography; SUV: Standardized uptake value; WHO: World 167–75.
Health Organization; AUC: Area under the curve; GTV: Gross tumor volume; 9. Chen W, Silverman DHS, Delaloye S, Czernin J, Kamdar N, Pope W, et al.
PTV: Planning target volume; ADC: Apparent diffusion coefficient; 18F-FDOPA PET imaging of brain tumors: comparison study with 18F-FDG
rCBV: Relative cerebral blood volume; VEGF-A: Vascular endothelial growth PET and evaluation of diagnostic accuracy. J Nucl Med Off Publ Soc Nucl
factor A; MTV: Metabolic tumor volume; RANO: Response assessment in Med. 2006;47:904–11.
neuro-oncology; VDE: Velocity of diameter expansion 10. Jacob MJ, Pandit AG, Jora C, Mudalsha R, Sharma A, Pathak HC.
Comparative study of (18)F-DOPA, (13)N-ammonia and F18-FDG PET/CT in
Acknowledgements primary brain tumors. Indian J Nucl Med IJNM Off J Soc Nucl Med India.
Not applicable. 2011;26:139–43.
11. Tripathi M, Sharma R, D’Souza M, Jaimini A, Panwar P, Varshney R, et al.
Authors’ contributions Comparative evaluation of F-18 FDOPA, F-18 FDG, and F-18 FLT-PET/CT for
FS carried out the systematic review of literature and wrote the manuscript, metabolic imaging of low grade gliomas. Clin Nucl Med. 2009;34:878–83.
LB was the second independent reviewer for the inclusion and the quality 12. Lapa C, Linsenmann T, Monoranu CM, Samnick S, Buck AK, Bluemel C, et al.
assessment of articles, CB reviewed the manuscript initially, IJN and GN Comparison of the amino acid tracers 18F-FET and 18F-DOPA in high-grade
reviewed and made major corrections to the manuscript. The authors read Glioma patients. J Nucl Med. 2014;55:1611–6.
and approved the final manuscript. 13. Kratochwil C, Combs SE, Leotta K, Afshar-Oromieh A, Rieken S, Debus J,
et al. Intra-individual comparison of 18F-FET and 18F-DOPA in PET imaging
Funding of recurrent brain tumors. Neuro-Oncol. 2014;16:434–40.
Authors declare there was no source of funding. 14. Becherer A, Karanikas G, Szabó M, Zettinig G, Asenbaum S, Marosi C, et al.
Brain tumour imaging with PET: a comparison between [18F]fluorodopa
Availability of data and materials and [11C]methionine. Eur J Nucl Med Mol Imaging. 2003;30:1561–7.
Not applicable. 15. Youland RS, Kitange GJ, Peterson TE, Pafundi DH, Ramiscal JA, Pokorny JL,
et al. The role of LAT1 in (18)F-DOPA uptake in malignant gliomas. J
Ethics approval and consent to participate Neurooncol. 2013;111:11–8.
Not applicable. 16. Walter F, Cloughesy T, Walter MA, Lai A, Nghiemphu P, Wagle N, et al.
Impact of 3,4-dihydroxy-6-18F-fluoro-L-phenylalanine PET/CT on managing
Consent for publication patients with brain tumors: the referring physician’s perspective. J Nucl Med
Not applicable. Off Publ Soc Nucl Med. 2012;53:393–8.
17. Law I, Albert NL, Arbizu J, Boellaard R, Drzezga A, Galldiks N, et al. Joint
Competing interests EANM/EANO/RANO practice guidelines/SNMMI procedure standards for
Authors declare no competing interests. imaging of gliomas using PET with radiolabelled amino acids and [18F]FDG:
version 1.0. Eur J Nucl Med Mol Imaging. 2019;46:540–57.
Author details 18. Whiting PF, Rutjes AWS, Westwood ME, Mallett S, Deeks JJ, Reitsma JB, et al.
1
Nuclear medicine Department, Hautepierre University Hospital, 1, rue QUADAS-2: a revised tool for the quality assessment of diagnostic accuracy
Molière, F-67000 Strasbourg, France. 2Oncology Department, Hautepierre studies. Ann Intern Med. 2011;155:529–36.
University Hospital, 1, rue Molière, F-67000 Strasbourg, France. 3Strasbourg 19. Beuthien-Baumann B, Bredow J, Burchert W, Füchtner F, Bergmann R, Alheit
University, Unistra/CNRS UMR 7237, Strasbourg, France. 4Radiotherapy H-D, et al. 3-O-methyl-6-[18F]fluoro-L-DOPA and its evaluation in brain
Department, Paul Strauss Comprehensive Cancer Center, 3, rue de la porte tumour imaging. Eur J Nucl Med Mol Imaging. 2003;30:1004–8.
de l’hôpital, F-67065 Strasbourg, France. 5Strasbourg University, CNRS, IPHC 20. Pafundi DH, Laack NN, Youland RS, Parney IF, Lowe VJ, Giannini C, et al.
UMR 7178, Centre Paul Strauss, UNICANCER, F-67000 Strasbourg, France. Biopsy validation of 18F-DOPA PET and biodistribution in gliomas for
neurosurgical planning and radiotherapy target delineation: results of a
Received: 29 October 2019 Accepted: 21 September 2020 prospective pilot study. Neuro-Oncol. 2013;15:1058–67.
21. Ledezma CJ, Chen W, Sai V, Freitas B, Cloughesy T, Czernin J, et al. 18F-
FDOPA PET/MRI fusion in patients with primary/recurrent gliomas: initial
References experience. Eur J Radiol. 2009;71:242–8.
1. Stockham AL, Tievsky AL, Koyfman SA, Reddy CA, Suh JH, Vogelbaum MA, 22. Morana G, Piccardo A, Milanaccio C, Puntoni M, Nozza P, Cama A, et al.
et al. Conventional MRI does not reliably distinguish radiation necrosis from Value of 18F-3,4-dihydroxyphenylalanine PET/MR image fusion in pediatric
tumor recurrence after stereotactic radiosurgery. J Neurooncol. 2012;109: supratentorial infiltrative astrocytomas: a prospective pilot study. J Nucl Med
149–58. Off Publ Soc Nucl Med. 2014;55:718–23.
2. Young RJ, Gupta A, Shah AD, Graber JJ, Zhang Z, Shi W, et al. Potential 23. Chiaravalloti A, Fiorentini A, Villani V, Carapella C, Pace A, Di Pietro B, et al.
utility of conventional MRI signs in diagnosing pseudoprogression in Factors affecting 18F FDOPA standardized uptake value in patients with
glioblastoma. Neurology. 2011;76:1918–24. primary brain tumors after treatment. Nucl Med Biol. 2015;42:355–9.
3. Zou Q-G, Xu H-B, Liu F, Guo W, Kong X-C, Wu Y. In the assessment of 24. Youland RS, Pafundi DH, Brinkmann DH, Lowe VJ, Morris JM, Kemp BJ, et al.
supratentorial glioma grade: the combined role of multivoxel proton MR Prospective trial evaluating the sensitivity and specificity of 3,4-dihydroxy-6-
spectroscopy and diffusion tensor imaging. Clin Radiol. 2011;66:953–60. [18F]-fluoro-L-phenylalanine (18F-DOPA) PET and MRI in patients with
4. Bulik M, Jancalek R, Vanicek J, Skoch A, Mechl M. Potential of MR recurrent gliomas. J Neurooncol. 2018;137:583–91.
spectroscopy for assessment of glioma grading. Clin Neurol Neurosurg. 25. Janvier L, Olivier P, Blonski M, Morel O, Vignaud J-M, Karcher G, et al.
2013;115:146–53. Correlation of SUV-derived indices with Tumoral aggressiveness of Gliomas
Somme et al. Cancer Imaging (2020) 20:70 Page 13 of 13

in static 18F-FDOPA PET: use in clinical practice. Clin Nucl Med. 2015;40: with recurrent malignant glioma on bevacizumab therapy. Clin Cancer Res
e429–35. Off J Am Assoc Cancer Res. 2014;20:3550–9.
26. Patel CB, Fazzari E, Chakhoyan A, Yao J, Raymond C, Nguyen H, et al. 18F- 46. Wardak M, Schiepers C, Cloughesy TF, Dahlbom M, Phelps ME, Huang S-C.
18
FDOPA PET and MRI characteristics correlate with degree of malignancy F-FLT and 18F-FDOPA PET kinetics in recurrent brain tumors. Eur J Nucl
and predict survival in treatment-naïve gliomas: a cross-sectional study. J Med Mol Imaging. 2014;41:1199–209.
Neurooncol. 2018;139:399–409. 47. Ellingson BM, Chung C, Pope WB, Boxerman JL, Kaufmann TJ.
27. Schiepers C, Chen W, Cloughesy T, Dahlbom M, Huang S-C. 18F-FDOPA Pseudoprogression, radionecrosis, inflammation or true tumor progression?
kinetics in brain tumors. J Nucl Med Off Publ Soc Nucl Med. 2007;48: Challenges associated with glioblastoma response assessment in an
1651–61. evolving therapeutic landscape. J Neurooncol. 2017;134:495–504.
28. Nioche C, Soret M, Gontier E, Lahutte M, Dutertre G, Dulou R, et al. 48. Linhares P, Carvalho B, Figueiredo R, Reis RM, Vaz R. Early
Evaluation of quantitative criteria for glioma grading with static and Pseudoprogression following Chemoradiotherapy in Glioblastoma patients:
dynamic 18F-FDopa PET/CT. Clin Nucl Med. 2013;38:81–7. the value of RANO evaluation. J Oncol. 2013;2013:690585.
29. Bund C, Heimburger C, Imperiale A, Lhermitte B, Chenard M-P, Lefebvre 49. Nasseri M, Gahramanov S, Netto JP, Fu R, Muldoon LL, Varallyay C, et al.
F, et al. FDOPA PET-CT of nonenhancing brain tumors. Clin Nucl Med. Evaluation of pseudoprogression in patients with glioblastoma multiforme
2017;42:250–7. using dynamic magnetic resonance imaging with ferumoxytol calls RANO
30. Verger A, Metellus P, Sala Q, Colin C, Bialecki E, Taieb D, et al. IDH criteria into question. Neuro-Oncol. 2014;16:1146–54.
mutation is paradoxically associated with higher 18F-FDOPA PET uptake 50. Humbert O, Bourg V, Mondot L, Gal J, Bondiau P-Y, Fontaine D, et al. 18F-
in diffuse grade II and grade III gliomas. Eur J Nucl Med Mol Imaging. DOPA PET/CT in brain tumors: impact on multidisciplinary brain tumor
2017;44:1306–11. board decisions. Eur J Nucl Med Mol Imaging. 2019;46:558–68.
31. Karavaeva E, Harris RJ, Leu K, Shabihkhani M, Yong WH, Pope WB, et al. 51. Herrmann K, Czernin J, Cloughesy T, Lai A, Pomykala KL, Benz MR, et al.
Relationship between [18F]FDOPA PET uptake, apparent diffusion coefficient Comparison of visual and semiquantitative analysis of 18F-FDOPA-PET/CT for
(ADC), and proliferation rate in recurrent malignant Gliomas. Mol Imaging recurrence detection in glioblastoma patients. Neuro-Oncol. 2014;16:603–9.
Biol MIB Off Publ Acad Mol Imaging. 2015;17:434–42. 52. Karunanithi S, Sharma P, Kumar A, Khangembam BC, Bandopadhyaya GP,
32. Ginet M, Zaragori T, Marie P-Y, Roch V, Gauchotte G, Rech F, et al. Kumar R, et al. Comparative diagnostic accuracy of contrast-enhanced MRI
Integration of dynamic parameters in the analysis of 18F-FDopa PET and 18F-FDOPA PET-CT in recurrent glioma. Eur Radiol. 2013;23:2628–35.
imaging improves the prediction of molecular features of gliomas. Eur J 53. Karunanithi S, Sharma P, Kumar A, Khangembam BC, Bandopadhyaya GP,
Nucl Med Mol Imaging. 2020;47:1381–90. Kumar R, et al. 18F-FDOPA PET/CT for detection of recurrence in patients
33. Girard A, Saint-Jalmes H, Chaboub N, Le Reste P-J, Metais A, Devillers A, with glioma: prospective comparison with 18F-FDG PET/CT. Eur J Nucl Med
et al. Optimization of time frame binning for FDOPA uptake quantification Mol Imaging. 2013;40:1025–35.
in glioma. PLoS One. 2020;15:e0232141. 54. Karunanithi S, Bandopadhyaya GP, Sharma P, Kumar A, Singla S, Malhotra A, et al.
34. Fueger BJ, Czernin J, Cloughesy T, Silverman DH, Geist CL, Walter MA, et al. Prospective comparison of (99m)Tc-GH SPECT/CT and (18)F-FDOPA PET/CT for
Correlation of 6-18F-Fluoro-L-Dopa PET uptake with proliferation and tumor detection of recurrent glioma: a pilot study. Clin Nucl Med. 2014;39:e121–8.
grade in newly diagnosed and recurrent Gliomas. J Nucl Med. 2010;51:1532–8. 55. Fraioli F, Shankar A, Hyare H, Ferrazzoli V, Militano V, Samandouras G, et al.
35. Todeschi J, Bund C, Cebula H, Chibbaro S, Lhermitte B, Pin Y, et al. The use of multiparametric 18F-fluoro-L-3,4-dihydroxy-phenylalanine PET/
Diagnostic value of fusion of metabolic and structural images for MRI in post-therapy assessment of patients with gliomas. Nucl Med
stereotactic biopsy of brain tumors without enhancement after contrast Commun. 2020;41:517–25.
medium injection. Neurochirurgie. 2019;65:357–64. 56. Zaragori T, Ginet M, Marie P-Y, Roch V, Grignon R, Gauchotte G, et al. Use of
36. Cicone F, Carideo L, Scaringi C, Arcella A, Giangaspero F, Scopinaro F, et al. static and dynamic [18F]-F-DOPA PET parameters for detecting patients
18F-DOPA uptake does not correlate with IDH mutation status and 1p/19q with glioma recurrence or progression. EJNMMI Res. 2020;10:56.
co-deletion in glioma. Ann Nucl Med. 2019;33:295–302. 57. Villani V, Carapella CM, Chiaravalloti A, Terrenato I, Piludu F, Vidiri A, et al.
37. Zaragori T, Guedj E, Verger A. Is IDH mutation status associated with 18F- The role of PET [18F]FDOPA in evaluating low-grade Glioma. Anticancer Res.
FDopa PET uptake? Ann Nucl Med. 2020;34:228–9. 2015;35:5117–22.
58. Rossi Espagnet MC, Romano A, Mancuso V, Cicone F, Napolitano A, Scaringi
38. Cicone F, Carideo L, Scaringi C, Minniti G. Reply to Zaragori et al.: “is IDH
C, et al. Multiparametric evaluation of low grade gliomas at follow-up:
mutation status associated with 18F-FDopa PET uptake”. Ann Nucl Med.
comparison between diffusion and perfusion MR with (18)F-FDOPA PET. Br
2020;34:230–1.
J Radiol. 2016;89:20160476.
39. Kosztyla R, Chan EK, Hsu F, Wilson D, Ma R, Cheung A, et al. High-grade Glioma
59. Smits M. Imaging of oligodendroglioma. Br J Radiol. 2016;89:20150857.
radiation therapy target volumes and patterns of failure obtained from
60. Oughourlian TC, Yao J, Schlossman J, Raymond C, Ji M, Tatekawa H, et al.
magnetic resonance imaging and 18F-FDOPA positron emission tomography
Rate of change in maximum 18F-FDOPA PET uptake and non-enhancing
delineations from multiple observers. Int J Radiat Oncol. 2013;87:1100–6.
tumor volume predict malignant transformation and overall survival in low-
40. Kazda T, Pafundi DH, Kraling A, Bradley T, Lowe VJ, Brinkmann DH, et al.
grade gliomas. J Neurooncol. 2020;147:135–45.
Dosimetric impact of amino acid positron emission tomography imaging
61. Dowson N, Thomas P, Fay M, Jeffree RL, Gal Y, Bourgeat P, et al. Early
for target delineation in radiation treatment planning for high-grade
prediction of treatment response in advanced gliomas with (18)F-dopa
gliomas. Phys Imaging Radiat Oncol. 2018;6:94–100.
positron-emission tomography. Curr Oncol Tor Ont. 2014;21:e172–8.
41. Rose S, Fay M, Thomas P, Bourgeat P, Dowson N, Salvado O, et al.
62. Chiaravalloti A, Esposito V, Ursini F, Di Giorgio E, Zinzi M, Calabria F, et al.
Correlation of MRI-derived apparent diffusion coefficients in newly
Overall survival and progression-free survival in patients with primary brain
diagnosed gliomas with [18F]-fluoro-L-dopa PET: what are we really
tumors after treatment: is the outcome of [18F] FDOPA PET a prognostic
measuring with minimum ADC? AJNR Am J Neuroradiol. 2013;34:758–64.
factor in these patients? Ann Nucl Med. 2019;33:471–80.
42. Cicone F, Filss CP, Minniti G, Rossi-Espagnet C, Papa A, Scaringi C, et al.
63. Pallud J, Blonski M, Mandonnet E, Audureau E, Fontaine D, Sanai N, et al.
Volumetric assessment of recurrent or progressive gliomas: comparison
Velocity of tumor spontaneous expansion predicts long-term outcomes for
between F-DOPA PET and perfusion-weighted MRI. Eur J Nucl Med Mol
diffuse low-grade gliomas. Neuro-Oncol. 2013;15:595–606.
Imaging. 2015;42:905–15.
64. Isal S, Gauchotte G, Rech F, Blonski M, Planel S, Chawki MB, et al. A high
43. Kosztyla R, Raman S, Moiseenko V, Reinsberg SA, Toyota B, Nichol A. Dose- 18F-FDOPA uptake is associated with a slow growth rate in diffuse grade II-
painted volumetric modulated arc therapy of high-grade glioma using 3,4- III gliomas. Br J Radiol. 2018;91:20170803.
dihydroxy-6-[ 18 F]fluoro-L-phenylalanine positron emission tomography. Br 65. Karunanithi S, Sharma P, Kumar A, Gupta DK, Khangembam BC, Ballal S,
J Radiol. 2019;92:20180901. et al. Can (18)F-FDOPA PET/CT predict survival in patients with suspected
44. Oehlke O, Mix M, Graf E, Schimek-Jasch T, Nestle U, Götz I, et al. Amino-acid recurrent glioma? A prospective study. Eur J Radiol. 2014;83:219–25.
PET versus MRI guided re-irradiation in patients with recurrent glioblastoma
multiforme (GLIAA) - protocol of a randomized phase II trial (NOA 10/ARO
2013-1). BMC Cancer. 2016;16:769. Publisher’s Note
45. Schwarzenberg J, Czernin J, Cloughesy TF, Ellingson BM, Pope WB, Grogan Springer Nature remains neutral with regard to jurisdictional claims in
T, et al. Treatment response evaluation using 18F-FDOPA PET in patients published maps and institutional affiliations.