Abstract
Oligodendroglial components (OC) and loss of heterozygosity on chromosomes 1p and 19q (LOH 1p/19q) are associated with better outcome in patients with glioma. We aimed to assess the fitness of [18F]fluoroethyltyrosine positron-emission-tomography (FET-PET) for noninvasively identifying these important prognostic/predictive factors. One hundred forty-four patients with MRI-suspected WHO grade II and III glioma underwent FET-PET scans prior to histological diagnosis. FET-PET analyses included maximal tumoral uptake (SUVmax/BG), biological tumor volume (BTV), mean tumoral uptake (SUVmean/BG), total tumoral uptake (SUVtotal/BG), and kinetic analysis. Suspicion of OC was based on static and dynamic FET-uptake parameters. PET results were correlated with histology and 1p/19q status. OC tumors exhibited significantly higher uptake values, compared with astrocytomas (AC) (SUVmax/BG 3.1 vs 2.3, BTV 15.5 mL vs 7.2 mL, SUVtotal/BG 38.5 vs 17.4, P < .01 each; SUVmean/BG 2.2 vs 2.1, P < .05). These differences were more pronounced in WHO grade II gliomas. Comparable results were found with respect to 1p/19q status. Kinetic analysis misclassified 18 of 34 low-grade OC tumors as high-grade glioma but misclassified only 5 of 45 of the low-grade ACs. FET-based suspicion of OC resulted in concordance rates of both 76% for the prediction of OC and LOH 1p/19q. FET-uptake was significantly higher in gliomas with OC, compared with AC, and likewise in 1p/19q codeleted, compared with noncodeleted tumors. However, FET-PET analysis did not reliably predict the presence of OC/LOH 1p/19q in the individual patient, mostly because of an overlap in PET characteristics of OC tumors and high-grade AC. Histological examination is still required for an accurate diagnosis.
Keywords: FET-PET, FET uptake, glioma, kinetic analysis, LOH1p/19q, oligodendroglial tumor components
Diffusely infiltrating astrocytoma (AC), mixed oligoastrocytoma, and oligodendroglioma represent the most frequent WHO grade II and III gliomas;1 those with oligodendroglial components (OC) usually exhibit better outcome measurements.2–4 The histological criteria in particular for mixed oligoastrocytoma, however, are somewhat subjective and suffer from considerable inter-observer variability; consequently estimates of their incidence widely varies.
Recent advances in molecular genetics have contributed considerably to a better understanding of glioma genesis, diagnostic classification, and prognostic evaluation: during glioma genesis, the acquisition of an unbalanced chromosomal translocation t(1;19)(q10;p10) (LOH 1p/19q) has been suggested to determine an oligodendroglial tumor cell differentiation.5–7 Indeed, a 1p/19q codeletion can be detected in the majority of oligodendrogliomas/oligoastrocytomas and has been shown to be homogeneously distributed throughout the tumor volume, indicating a monoclonal origin of tumor cells even in oligoastrocytoma.8,9 Recently, it has been suggested that the determination of the 1p/19q status might help to overcome the ambiguity inherent to the conventional histological diagnosis of oligoastrocytoma.6
From a clinical perspective, screening methods with noninvasive identification of gliomas harboring OC and/or LOH 1p/19q are of utmost interest and might influence treatment strategies, particularly for those patients with nonresectable tumors located in highly eloquent areas of the brain that might benefit from neoadjuvant chemotherapy regimes. MRI-based methods have been repeatedly evaluated but have failed to reliably predict a more detailed histological, molecular-genetic, and prognostic differentiation of grade II and III gliomas.10–17
Recently, dynamic [18F]fluoroethyltyrosine positron emission tomography (FET-PET) evaluation has been shown to be valuable for noninvasive tumor grading of AC.18 In addition, preliminary data suggest that FET-uptake might be more pronounced in OC tumors (as compared with AC tumors); the relevance of these FET-PET findings with respect to noninvasive discrimination between OC and AC tumors and tumor grading of OC tumors has not yet been systematically evaluated.
Because of this background, we obtained in the present study FET-PET recordings in a series of consecutively evaluated patients with an MRI-based suspicion of a WHO grade II and III glioma; we aimed to correlate both static and dynamic FET-uptake parameters with WHO grading and the presence of OC and 1p/19q codeletion.
Patients and Methods
Patients
For the current single-institutional study (Departments of Neurosurgery and Nuclear Medicine, Ludwig-Maximilians-University [LMU], Munich, Germany), adult patients with an MRI-based suspicion of a de novo or recurrent supratentorial glioma WHO grade II and III (consecutively evaluated and/or treated from 2005 through 2010) were considered to be eligible. In case of tumor recurrence, only patients who were previously exclusively surgically treated were included. Patients with a history of radiotherapy and/or chemotherapy were excluded. The study protocol was reviewed and approved by the LMU institutional review board. Informed consent was obtained from all patients.
MRI Acquisition and Assessment
Standard MRI images were obtained with either 1.5 T (Magnetom Symphony; Siemens Erlangen) or 3.0 T (Signa HDx; GE Healthcare) magnets at the LMU Department of Neuroradiology. MRI-based criteria for the suspicion of WHO II or III gliomas were the presence of hypointense lesions in T1-weighted images (with or without sparse contrast enhancement) and hyperintense lesions in T2/fluid attenuated inversion recovery (FLAIR)–weighted images.
FET-PET Image Acquisition and Assessment
All patients underwent FET-PET examination prior to histological diagnosis according to standard protocols.18
Static evaluation included (1) the maximal tumoral FET-uptake corrected for the mean background activity in the contralateral hemisphere (SUVmax/BG), (2) an estimated biological tumor volume (BTV) as defined by assessment of increased FET-uptake in semi-automatic threshold-based (SUVma/BG ≥1.8) delineation of a volume of interest (VOI), (3) the mean tracer uptake in the BTV (SUVmean/BG), and (4) the total tumoral uptake volume (SUVtotal/BG = SUVmean/BG × BTV) to accommodate the differing sizes of the lesions.
Evaluation of dynamic FET-PET data has been described previously.18 In brief, a region of interest (ROI) analysis was performed to extract time activity curves (TAC) for each individual slice with suspicious FET-uptake. For this purpose, 90% iso-contour threshold ROIs were defined on summation images for the interval 10–30 min after tracer injection and, afterwards, applied to the dynamic PET data.
FET-PET Criteria for OC Tumors
On the basis of preliminary observations, we hypothesized that glioma with OC should manifest with the following combined static and dynamic tracer uptake characteristics: visually intense focal uptake pattern in conjunction with SUVmax/BG ≥2.4 and temporally increasing TAC for low-grade OC and SUVmax/BG ≥3.0 and temporally decreasing TAC for high-grade OC. Final FET-PET evaluation and discrimination in pure AC vs OC were formed by the consensus of 2 experienced investigators (C.l.F. and N.J.) who were blinded to histological and molecular-genetic data. In a second analysis, receiver operating characteristics (ROC) calculations were performed to obtain optimal cutoff values for the discrimination of OC and AC.
Tissue Sampling
Tumor specimens for histological and molecular-genetic evaluation were harvested from either serial stereotactic biopsy procedures or open tumor surgery. Molecular stereotactic biopsy procedures were performed as published elsewhere.8,19 In brief, besides CT (2-mm contrast-enhanced axial images) and MRI (2-mm T1-weighted gadolinium-enhanced and 2-mm T2-weighted axial images), PET data were routinely co-registrated and integrated in multiplanar trajectory planning (i-plan stereotaxy/BrainLab). Both static and dynamic PET data were used as a guide for stepwise (1-mm steps) histopathological evaluation throughout the tumor space. Serial biopsy with a mean tissue volume of approximately 1 mm3 per sample was taken along the predefined trajectory including the area of highest FET uptake. The mean number of tumor samples obtained per patient was 5. PET-guided microsurgical resections were performed by experienced neurosurgeons using modern neurosurgical equipment, including neuronavigation with MRI and PET image fusion (Brainlab, Germany). Tumor specimens used for histopathological evaluation always included samples from FET-PET pre-delineated tumor parts.
Histopathology, WHO Grading, and Molecular-Genetic Markers
Histological classification, tumor grading, and molecular-genetic analyses were performed by experienced neuropathologists (R.E., H.A.K., S.E.) who were blinded to FET-PET data. Tumors were classified according to the current WHO guidelines.1 In case of discordant findings, a conference for reevaluation was initiated to achieve a consensus concerning the diagnosis. Determination of LOH 1p/19q was performed according to standard protocols.8 The following microsatellite markers were used: D1S548, D1S1184, D1S1608, D1S1592, D1S1161, D19S601, D19S559, D1S433, D19S718, and D19S431.
Statistical Analysis
SPSS for Windows (version 17.0, SPSS) was used for statistical calculations. To evaluate the concordance rate of PET-based classification (OC tumors vs AC tumors) with histology, the ratio of all concordant (true positive and true negative) cases was calculated. Furthermore, the sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were determined. In a second model, similar analyses were performed to assess concordance/discordance rates between the PET-based classification and the 1p/19q status. Continuously scaled data were analyzed using the Mann–Whitney U-test and categorical data with the χ2 statistic. All continuously scaled parameters were reported as mean ± standard error of mean. Statistical significance was defined as P values ≤.05.
Results
Study Population
The study comprised 144 patients (84 men, 60 women), including 21 patients (14.6%) with an MRI-based suspicion of a recurrent tumor. The mean age at the time of study inclusion was 45.3 ± 12.2 years (range, 22–81 years). FET-PET scans and histopathological evaluations were available for all patients. Tumor tissue samples were obtained using stereotactic biopsy procedure in 124 patients (86.1%) and open tumor resection in 20 patients (13.9%). The median time from FET-PET evaluation to tissue sample obtainment (in newly diagnosed and recurrent tumors) was 5 days.
Histopathology, WHO Grading, and 1p/19q Status
Detailed results of the histological and molecular-genetic analyses are shown in Table 1. In total, 45 diffuse astrocytomas WHO II, 44 anaplastic astrocytomas WHO III, 11 oligodendrogliomas WHO II, 23 mixed oligoastrocytomas WHO II, and 21 anaplastic oligoastrocytomas WHO III were found. No case of anaplastic oligodendroglioma WHO III or glioblastoma multiforme WHO grade IV was detected.
Table 1.
Study population: histopathology and molecular-genetic markers
| Variable | All tumors (N = 144) | DA (N = 45) | AA (N = 44) | OD (N = 11) | OA (N = 23) | AOA (N = 21) |
|---|---|---|---|---|---|---|
| LOH 1p/19q | 47 (32.6%) | 7 (15.6%) | 1 (2.3%) | 10 (90.9%) | 18 (78.3%) | 11 (52.4%) |
| LOH 1p or 19q | 16 (11.1%) | 3 (6.7%) | 5 (11.4%) | — | 2 (8.7%) | 6 (28.6%) |
| No LOH 1p/19q | 72 (50.0%) | 33 (73.3%) | 32 (72.7%) | 1 (9.1%) | 2 (8.7%) | 4 (19.0%) |
| 1p/19q status not available | 9 (6.3%) | 2 (4.4%) | 6 (13.6%) | — | 1 (4.3%) | — |
Abbreviations: DA, diffuse astrocytoma; AA, anaplastic astrocytoma; OD, oligodendroglioma; OA, oligoastrocytoma; AOA, anaplastic oligoastrocytoma.
The 1p/19q status was available for 135 patients: 47 patients exhibited a 1p/19q codeletion, including 10 of 11 oligodendrogliomas, 29 of 43 mixed oligoastrocytomas, and 8 of 81 astrocytomas; a codeletion was seen in 72% of the analyzed OC and in 9% of the AC tumors (P < .001).
Analysis of FET-PET Parameters
FET-PET findings are summarized in Table 2. The distribution of the PET data did not differ between biopsied and surgically treated patients or between de novo and recurrent tumors (P > .05).
Table 2.
Detailed FET-PET parameters (mean value ± standard error) according to histology and WHO grading
| Parameter | FET-uptake values |
|||
|---|---|---|---|---|
| AC tumors |
OC tumors |
|||
| SUVmax/BG | 2.3 ± 0.1** | 3.1 ± 0.1** | ||
| BTV [mL] | 7.2 ± 1.8** | 15.5 ± 3.4** | ||
| SUVmean/BG | 2.14 ± 0.04* | 2.24 ± 0.04* | ||
| SUVtotal/BG | 17.4 ± 4.3** | 38.5 ± 10.2** | ||
| WHO II | WHO III | WHO II | WHO III | |
| SUVmax/BG | 2.1 ± 0.2** | 2.6 ± 0.2# | 3.1 ± 0.2** | 3.2 ± 0.2# |
| BTV [mL] | 7.1 ± 2.9** | 7.3 ± 2.0## | 11.3 ± 2.4** | 22.3 ± 7.8## |
| SUVmean/BG | 2.15 ± 0.07 | 2.14 ± 0.04 | 2.24 ± 0.06 | 2.23 ± 0.05 |
| SUVtotal/BG | 17.1 ± 7.1** | 17.7 ± 4.9## | 27.0 ± 6.5** | 57.0 ± 24.5## |
Group differences are significant at *or #P < .05 and **or ##P < .01.
WHO III VS WHO II Glioma
WHO grade III gliomas tended to exhibit higher values for SUVmax/BG (2.8 vs 2.5; P = .09), BTV (12.2 mL vs 8.9 mL; P = .08), and SUVtotal/BG (30.4 vs 21.4; P = .06), compared with their grade II counterparts; no such trend was found for SUVmean/BG (2.2 vs 2.2, P = .87). Most of the detected differences referred to the AC subpopulation. In particular, significantly higher SUVmax/BG values were found (2.6 vs 2.1; P < .01) for grade III AC (compared with grade II AC). OC tumors WHO grade II and III, however, did not show significantly different values for SUVmax/BG, BTV, SUVmean/BG, and SUVtotal/BG (Fig. 1; Table 2).
Fig. 1.
Comparison of SUVmax/BG in AC gliomas and OC gliomas. OC tumors presented with significantly higher uptake values compared with AC tumors, independently of the WHO grading. Subgroup comparison in AC tumors revealed significantly higher values in WHO grade III, whereas no significant differences were seen between WHO grade II and III OC tumors. *P < .05; **P < .01; n.s., not significant.
Dynamic analysis of FET uptake revealed decreasing TACs (suggestive for high-grade glioma) in 72 tumors, which correlated strongly with WHO grade III histology: The corresponding sensitivity, specificity, and PPV values were 88%, 63%, and 69%, respectively. The diagnostic power for tumor grading was higher in AC tumors (sensitivity, 84%; specificity, 81%; PPV, 86%), compared with the OC subgroup (sensitivity, 95%; specificity, 50%; PPV, 53%). Of note, 8 of 11 patients with grade II oligodendrogliomas had tumors misclassified as grade III.
OC VS AC Tumors
FET uptake was above background level in 98% of OC tumors and 75% of AC tumors (P < .001). OC tumors significantly differed from AC tumors with respect to SUVmax/BG (3.1 vs 2.3; P < .01), BTV (15.5 mL vs 7.2 mL; P < .01), SUVmean/BG (2.24 vs 2.14; P < .05), and SUVtotal/BG (38.5 vs 17.4; P < .01) (Table 2). No difference was found between oligoastrocytomas and oligodendrogliomas. In the subgroup of WHO grade II gliomas, low FET uptake (SUVmax/BG <2.4) was also seen in 9 (26%) of 34 OC tumors, whereas 14 (31%) of 45 AC tumors had SUVmax/BG values ≥2.4. In the subpopulation of WHO grade III gliomas, 11 (52%) of 21 OC tumors had a SUVmax/BG <3.0, whereas 10 (23%) of 44 AC tumors exhibited an SUVmax/BG ≥3.0.
According to our prospectively defined combined static and dynamic FET-uptake criteria, an OC tumor was suspected in 47 patients (Figs 2 and 3) and a pure AC tumor in 97 patients. Histological evaluation confirmed an OC tumor in 34 (72%) of 47. Of note, 2 of the 13 misclassified tumors had a 1p/19q codeletion despite a pure astroglial histology. Vice versa, a pure astroglial differentiation was histologically confirmed in 76 (78%) of 97 patients with FET-PET suspicion of an AC tumor. However, 9 of 21 tumors that were erroneously suspected to belong to the AC group were histologically classified as oligodendroglial tumors but without a 1p/19q codeletion. The resulting concordance rate of FET-PET predictions regarding the presence/absence of OC histology was 76% (Table 3); the implementation of the ROC-based cutoff values did not considerably increase this rate (detailed results are shown in Table 4).
Fig. 2.
FET-PET analysis of an MRI-suspected low-grade glioma. (A) Axial T1-weighted, contrast-enhanced MRI, showing a non-contrast enhancing, cortical, and subcortical lesion in the frontal lobe. (B) Coregistered axial FET-PET image reveals an intensive tracer uptake with an SUVmax/BG of 4.6. (C) Dynamic analysis of FET-uptake within the lesion reveals a constantly increasing time activity curve, suggestive of low-grade tumor tissue. According to our combined static and dynamic FET-PET criteria, a low-grade oligodendroglial tumor was expected. Histopathological and molecular-genetic evaluation revealed a WHO grade II oligoastrocytoma with LOH 1p/19q.
Fig. 3.
FET-PET analysis of an MRI-suspected anaplastic glioma. (A) Axial T1-weighted, contrast-enhanced MRI, showing a contrast-enhancing, cortical, and subcortical left-sided insular lesion. (B) Coregistered axial FET-PET image reveals an intensive tracer uptake with an SUVmax/BG of 5.0. (C) Dynamic analysis of FET-uptake within the lesion presents with a decreasing time activity curve, suggesting high-grade tumor tissue. According to our combined static and dynamic FET-PET criteria, an anaplastic oligodendroglial tumor was suspected. Histopathological and molecular-genetic evaluation revealed a WHO grade III astrocytoma without LOH 1p/19q.
Table 3.
Concordance rates of FET-PET–based predictions for presence of oligodendroglial tumor components and LOH 1p/19q
| Variable | PET/Histology (%) | PET/LOH 1p/19q (%) |
|---|---|---|
| Concordance Rate | 76 | 76 |
| Sensitivity | 62 | 62 |
| Specificity | 85 | 83 |
| PPV | 72 | 66 |
| NPV | 78 | 80 |
Table 4.
Cut-off values of the ROC analyses for FET-PET parameters with respective sensitivity and specificity
| Parameter | Cut-off value | Sensitivity (%) | Specificity (%) |
|---|---|---|---|
| SUVmax/BG | 2.6 | 70 | 72 |
| BTV [mL] | 4.0 | 71 | 69 |
| SUVmean/BG | 2.1 | 61 | 59 |
| SUVtotal/BG | 6.9 | 75 | 66 |
1p/19q Codeleted VS Noncodeleted Tumors
Significant differences were found for static FET-PET parameters of 1p/19q codeleted tumors and those lacking these chromosomal abnormalities (LOH-negative tumors; ie, SUVmax/BG: 3.2 vs 2.4 [P < .01]; BTV: 17.7 mL vs 6.7 mL [P < .01]; SUVmean/BG: 2.26 vs 2.13 [P < .05]; and SUVtotal/BG: 43.3 vs 16.3 [P < .01]). A comparative analysis of tumors without 1p/19q codeletion and those with partial deletions (either on 1p or 19q) revealed no difference of static PET parameters (Table 5).
Table 5.
Detailed FET-PET parameters (mean value ± standard error) according to 1p/19q status
| Parameter | No LOH 1p/19q | LOH 1p or 19q | LOH 1p/19q |
|---|---|---|---|
| SUVmax/BG | 2.4 ± 0.1** | 2.2 ± 0.2## | 3.2 ± 0.2**## |
| BTV [mL] | 6.7 ± 1.5** | 6.4 ± 2.0 | 17.7 ± 4.4** |
| SUVmean/BG | 2.14 ± 0.04* | 2.07 ± 0.05# | 2.26 ± 0.05*# |
| SUVtotal/BG | 16.7 ± 4.0** | 14.1 ± 4.9 | 43.3 ± 12.8** |
| SUVmax/BG | 2.4 ± 0.1** | 3.2 ± 0.2** | |
| BTV [mL] | 6.7 ± 1.3** | 17.7 ± 4.4** | |
| SUVmean/BG | 2.13 ± 0.04* | 2.26 ± 0.05* | |
| SUVtotal/BG | 16.3 ± 3.4** | 43.3 ± 12.8** |
Group differences are significant at *or #P < .05 and **or ##P < .01.
According to our PET criteria, 73 (83%) of 88 non-codeleted gliomas were assigned to the AC group (including 14 of 16 with partial deletions). Of note, 6 of the remaining 15 noncodeleted tumors (that were allocated to the OC group by FET-PET criteria) indeed exhibited an oligodendroglial histology. Conversely, 29 (62%) of 47 codeleted gliomas were assigned to the OC group (based on our PET criteria). Again, 6 of the remaining 18 codeleted tumors that were suspected to belong to the AC group were histologically confirmed as pure astrocytomas. The overall concordance rate of the PET-based predictions regarding the presence/absence of a 1p/19q codeletion was 76% and did not differ from the aforementioned concordance rate of PET predictions regarding the presence/absence of OC histology.
Discussion
The current study was conducted to define the role of FET-PET as a noninvasive screening method for the identification of OC tumors; 89 AC tumors (grade II/grade III) and 55 OC tumors (grade II/grade III) were analyzed. We correlated static and dynamic uptake parameters of FET with WHO grading, the presence/absence of an OC, and the presence/absence of a 1p/19q codeletion. The latter correlation analysis was considered to be important, because LOH 1p/19q, which represents a key marker of OC tumors, is suspected to be a more objective measurement of the presence/absence of an OC. Consequently, this marker might help to overcome the ambiguity inherent to the conventional histological diagnosis, especially of an oligoastrocytoma. FET-PET criteria for the identification of OC tumors were based on our previously published preliminary data,18 which were intended to be validated by the current analysis.
We showed that histologically verified OC tumors generally exhibit significantly higher uptake values than do AC tumors. Because the apparent magnitude of SUV is affected by partial volume effects, resulting in an underestimation of SUV in smaller lesions, this effect could be a limitation of our study; our AC tumors presented with half the size of our OC tumors. However, our data suggest that the size relation of SUV may only have a minor effect on our results; the uptake differences between AC and OC tumors were more pronounced in low-grade tumors (SUVmax/BG 3.1 vs 2.1) than in high-grade tumors (SUVmax/BG 3.2 vs 2.6), although the discrepancy in BTV was more evident in the latter (Table 2). Similar conclusions can be made when comparing low- and high-grade OC tumors, in which larger tumor volumes did not correlate with higher FET uptake. Therefore, we hold that our findings truly reflect actual higher FET uptake in OC tumors.
Of note, although high FET uptake was independent of the tumor grade in OC tumors, a positive correlation between FET uptake and tumor grade was seen in the AC tumors. Accordingly, a marked overlap of the distributions of the respective SUVmax/BG values of grade III AC and grade II and III OC tumors was seen. Remarkably, no uptake differences were seen between oligodendrogliomas and oligoastrocytomas; instead, both tumor entities exhibited a similar metabolic pattern in this series.
Any possible selection bias resulting from nonrepresentative biopsy procedures and consecutive missing of a considerable number of mixed cell tumors was ruled out by the 1p/19q status-based correlation model. Chromosomal abnormalities on 1p and 19q have been shown to be homogeneously distributed throughout the tumor space and are regarded as early events in glioma genesis.6,8 Assessment of concordance/discordance rates of the PET-based classification and the 1p/19q status revealed similar results, although a different patient population was addressed. Forty-seven patients harbouring a 1p/19q codeletion (including 8 patients with AC tumors) were compared with 88 patients lacking combined chromosomal abnormalities (including 15 patients with OC tumors); 1p/19q codeleted tumors exhibited significantly higher uptake values and larger BTVs.
It was surprising that, despite these distinct differences of static tracer uptake parameters, our prospectively applied classification criteria did not generate prediction rates as high as one might have expected. Relatively high specificity values (>80%) were accompanied by sensitivity values in the range of 63%; these estimates were not better in grade II tumors than in grade III tumors. Moreover, even the implementation of the ROC-based cutoff values did not considerably increase the concordance rate of FET-PET and OC/LOH 1p/19q presence.
One of the major findings of this study was that dynamic PET analyses were less accurate for OC tumors than for AC tumors. Specificity values of dynamic FET-PET were 81% in AC tumors but only 50% in OC tumors. Considering that WHO criteria of tumor grading are more subjective in OC tumors (compared with AC tumors),1 one might speculate that those low-grade OC tumors with decreasing TACs were at risk of histological misclassification. However, we do not expect any significant bias, because almost all low-grade OC tumors with decreasing TACs (except for one) clearly presented with low proliferation indices and no histological sign of anaplasia. It will be interesting to see how the clinical courses of those patients with low-grade OC tumors and decreasing TACs compare with those with low-grade oligodendroglial histology but increasing TACs and with patients with anaplastic OC tumors, respectively. Thus far, it remains speculative whether differences in dynamic FET-PET findings, especially in OC tumors, are of prognostic relevance independent of WHO grading.
In summary, the PET-based prediction of presence/absence of OC was hampered by the large overlap of FET-uptake values in OC tumors and WHO grade III AC tumors and by the high rate of WHO grade II OC tumors with decreasing TACs, thus mimicking high-grade tumor tissue.
To elucidate further the FET-PET characteristics of OC gliomas and their similarity to those of high-grade AC tumors, a better understanding of tracer uptake mechanisms is needed. In particular, the pathophysiological mechanisms resulting in enhanced FET uptake and the factors responsible for the different kinetic behaviors of FET in low- and high-grade gliomas remain to be clarified. Different speculative explanations may account for this effect: (1) OC and AC tumors may differentially express the L-system amino acid transporter (LAT), which was previously suggested to be the rate-limiting factor for FET uptake in gliomas;20 (2) increased FET uptake partially depends on higher cell density,21 which was shown to be higher in low-grade OC tumors than in low-grade AC tumors;22 and (3) tracer uptake pattern in gliomas positively correlates with higher vascular density21 and is influenced by the regional cerebral blood flow, which is likely subserved by tumor angiogenesis.23 Of interest, microvessel density and luminal surface area both tend to be higher in OC tumors.24,25 Accordingly, a significantly higher cerebral blood flow has been reported in OC gliomas than in AC gliomas, by means of perfusion-weighted MR studies.26–28 The higher cerebral blood flow in OC tumors might also influence the washout of the untrapped and unmetabolized FET29,30 that diffuses back into the extracellular milieu through the bidirectional facilitated diffusion transporter LAT2.31 As a consequence, increased blood flow in OC tumors, while subserving higher tracer delivery (and thus higher SUVmax/BG), may by the same token result in rapid washout kinetics, accounting for the high false-positive rate of kinetic analysis in the low-grade OC glioma uptake.
From a clinical perspective (taking into account the aforementioned limitations), our findings support combined static and dynamic FET-PET analyses (as introduced here) as a valuable screening method for OC/LOH 1p/19q in suspected grade II/III gliomas. It will be of utmost interest to prospectively evaluate the effect of our results on clinical prognosis and treatment decisions for these patients in future studies. However, we think that, especially in patients with non-resectable tumors located in highly eloquent areas of the brain highly eloquent tumors, any suspicion of OC/LOH 1p/19q by noninvasive FET-PET may additionally justify a risk-adapted management concept consisting of minimal-invasive biopsy procedures first, followed by neoadjuvant chemotherapy if favorable prognostic/predictive histological and molecular-genetic profiles suggest a chemosensitive tumor.
Conclusion
OC tumors presented with significantly higher FET-uptake than did pure AC tumors, especially within grade II gliomas. However, FET-PET analysis does not reliably predict the presence of OC or LOH 1p/19q in individual patients. Moreover, tumor grading by dynamic analysis is less accurate in OC tumors. The reasons for different FET tracer dynamics in gliomas with an OC require further clarification. Our findings indicate that FET-uptake analysis cannot substitute histological diagnosis and molecular genetics for the detection of OC. Finally, the individual prognostic relevance of metabolic imaging findings, in particular the highly variable static and dynamic tracer uptake characteristics in our patient cohort and their correlation with clinical outcome, remain to be evaluated.
Funding
The study was supported by the German Glioma Network through German Cancer Aid (Deutsche Krebshilfe 70-3163-Wi 3).
Acknowledgments
We thank Katrin Richter for the superb technical support. Authors N. L. J., C. S., C. l. F., and N. T. contributed equally to this work.
Conflict of interest statement. None declared.
References
- 1.Louis D, Ohgaki H, Wiestler O, Cavenee W, editors. WHO Classification of Tumours of the Central Nervous System. Lyon: IARC; 2007. pp. 1–306. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Cairncross G, Berkey B, Shaw E, et al. Phase III trial of chemotherapy plus radiotherapy compared with radiotherapy alone for pure and mixed anaplastic oligodendroglioma: Intergroup Radiation Therapy Oncology Group Trial 9402. J Clin Oncol. 2006;24:2707–2714. doi: 10.1200/JCO.2005.04.3414. doi:10.1200/JCO.2005.04.3414. [DOI] [PubMed] [Google Scholar]
- 3.Glass J, Hochberg FH, Gruber ML, Louis DN, Smith D, Rattner B. The treatment of oligodendrogliomas and mixed oligodendroglioma-astrocytomas with PCV chemotherapy. J Neurosurg. 1992;76:741–745. doi: 10.3171/jns.1992.76.5.0741. doi:10.3171/jns.1992.76.5.0741. [DOI] [PubMed] [Google Scholar]
- 4.Okamoto Y, Di Patre PL, Burkhard C, et al. Population-based study on incidence, survival rates, and genetic alterations of low-grade diffuse astrocytomas and oligodendrogliomas. Acta Neuropathol. 2004;108:49–56. doi: 10.1007/s00401-004-0861-z. doi:10.1007/s00401-004-0861-z. [DOI] [PubMed] [Google Scholar]
- 5.Jenkins RB, Blair H, Ballman KV, et al. A t(1;19)(q10;p10) mediates the combined deletions of 1p and 19q and predicts a better prognosis of patients with oligodendroglioma. Cancer Res. 2006;66:9852–9861. doi: 10.1158/0008-5472.CAN-06-1796. doi:10.1158/0008-5472.CAN-06-1796. [DOI] [PubMed] [Google Scholar]
- 6.Kim YH, Nobusawa S, Mittelbronn M, et al. Molecular classification of low-grade diffuse gliomas. Am J Pathol. 2010;177:2708–2714. doi: 10.2353/ajpath.2010.100680. doi:10.2353/ajpath.2010.100680. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Watanabe T, Nobusawa S, Kleihues P, Ohgaki H. IDH1 mutations are early events in the development of astrocytomas and oligodendrogliomas. Am J Pathol. 2009;174:1149–1153. doi: 10.2353/ajpath.2009.080958. doi:10.2353/ajpath.2009.080958. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Thon N, Eigenbrod S, Grasbon-Frodl EM, et al. Novel molecular stereotactic biopsy procedures reveal intratumoral homogeneity of loss of heterozygosity of 1p/19q and TP53 mutations in World Health Organization grade II gliomas. J Neuropathol Exp Neurol. 2009;68:1219–1228. doi: 10.1097/NEN.0b013e3181bee1f1. doi:10.1097/NEN.0b013e3181bee1f1. [DOI] [PubMed] [Google Scholar]
- 9.Reifenberger G, Louis DN. Oligodendroglioma: toward molecular definitions in diagnostic neuro-oncology. J Neuropathol Exp Neurol. 2003;62:111–126. doi: 10.1093/jnen/62.2.111. [DOI] [PubMed] [Google Scholar]
- 10.Barker FG, Chang SM, Huhn SL, et al. Age and the risk of anaplasia in magnetic resonance-nonenhancing supratentorial cerebral tumors. Cancer. 1997;80:936–941. doi:10.1002/(SICI)1097-0142(19970901)80:5<936::AID-CNCR15>3.0.CO;2-X. [PubMed] [Google Scholar]
- 11.Chaichana KL, Kosztowski T, Niranjan A, et al. Prognostic significance of contrast-enhancing anaplastic astrocytomas in adults. J Neurosurg. 2010;113:286–292. doi: 10.3171/2010.2.JNS091010. doi:10.3171/2010.2.JNS091010. [DOI] [PubMed] [Google Scholar]
- 12.Emblem KE, Zoellner FG, Tennoe B, et al. Predictive modeling in glioma grading from MR perfusion images using support vector machines. Magn Reson Med. 2008;60:945–952. doi: 10.1002/mrm.21736. doi:10.1002/mrm.21736. [DOI] [PubMed] [Google Scholar]
- 13.Walker C, du Plessis DG, Fildes D, et al. Correlation of molecular genetics with molecular and morphological imaging in gliomas with an oligodendroglial component. Clin Cancer Res. 2004;10:7182–7191. doi: 10.1158/1078-0432.CCR-04-0681. doi:10.1158/1078-0432.CCR-04-0681. [DOI] [PubMed] [Google Scholar]
- 14.Cha S, Tihan T, Crawford F, et al. Differentiation of low-grade oligodendrogliomas from low-grade astrocytomas by using quantitative blood-volume measurements derived from dynamic susceptibility contrast-enhanced MR imaging. AJNR Am J Neuroradiol. 2005;26:266–273. [PMC free article] [PubMed] [Google Scholar]
- 15.Spampinato MV, Smith JK, Kwock L, et al. Cerebral blood volume measurements and proton MR spectroscopy in grading of oligodendroglial tumors. AJR Am J Roentgenol. 2007;188:204–212. doi: 10.2214/AJR.05.1177. doi:10.2214/AJR.05.1177. [DOI] [PubMed] [Google Scholar]
- 16.White ML, Zhang Y, Smoker WR, et al. Fluid-attenuated inversion-recovery MR imaging in assessment of intracranial oligodendrogliomas. Comput Med Imaging Graph. 2005;29:279–285. doi: 10.1016/j.compmedimag.2004.10.005. doi:10.1016/j.compmedimag.2004.10.005. [DOI] [PubMed] [Google Scholar]
- 17.White ML, Zhang Y, Kirby P, Ryken TC. Can tumor contrast enhancement be used as a criterion for differentiating tumor grades of oligodendrogliomas? AJNR Am J Neuroradiol. 2005;26:784–790. [PMC free article] [PubMed] [Google Scholar]
- 18.Popperl G, Kreth FW, Mehrkens JH, et al. FET PET for the evaluation of untreated gliomas: correlation of FET uptake and uptake kinetics with tumour grading. Eur J Nucl Med Mol Imaging. 2007;34:1933–1942. doi: 10.1007/s00259-007-0534-y. doi:10.1007/s00259-007-0534-y. [DOI] [PubMed] [Google Scholar]
- 19.Kunz M, Thon N, Eigenbrod S, et al. Hot spots in dynamic (18)FET-PET delineate malignant tumor parts within suspected WHO grade II gliomas. Neuro Oncol. 2011;13:307–316. doi: 10.1093/neuonc/noq196. doi:10.1093/neuonc/noq196. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Langen KJ, Hamacher K, Weckesser M, et al. O-(2-[18F]fluoroethyl)-L-tyrosine: uptake mechanisms and clinical applications. Nucl Med Biol. 2006;33:287–294. doi: 10.1016/j.nucmedbio.2006.01.002. doi:10.1016/j.nucmedbio.2006.01.002. [DOI] [PubMed] [Google Scholar]
- 21.Stockhammer F, Plotkin M, Amthauer H, van Landeghem FK, Woiciechowsky C. Correlation of F-18-fluoro-ethyl-tyrosin uptake with vascular and cell density in non-contrast-enhancing gliomas. J Neurooncol. 2008;88:205–210. doi: 10.1007/s11060-008-9551-3. doi:10.1007/s11060-008-9551-3. [DOI] [PubMed] [Google Scholar]
- 22.Derlon JM, Chapon F, Noel MH, et al. Non-invasive grading of oligodendrogliomas: correlation between in vivo metabolic pattern and histopathology. Eur J Nucl Med. 2000;27:778–787. doi: 10.1007/s002590000260. doi:10.1007/s002590000260. [DOI] [PubMed] [Google Scholar]
- 23.Wyss MT, Hofer S, Hefti M, et al. Spatial heterogeneity of low-grade gliomas at the capillary level: a PET study on tumor blood flow and amino acid uptake. J Nucl Med. 2007;48:1047–1052. doi: 10.2967/jnumed.106.038489. doi:10.2967/jnumed.106.038489. [DOI] [PubMed] [Google Scholar]
- 24.Nojiri T, Nariai T, Aoyagi M, et al. Contributions of biological tumor parameters to the incorporation rate of L: -[methyl-(11)C] methionine into astrocytomas and oligodendrogliomas. J Neurooncol. 2009;93:233–241. doi: 10.1007/s11060-008-9767-2. doi:10.1007/s11060-008-9767-2. [DOI] [PubMed] [Google Scholar]
- 25.Kracht LW, Friese M, Herholz K, et al. Methyl-[11C]- l-methionine uptake as measured by positron emission tomography correlates to microvessel density in patients with glioma. Eur J Nucl Med Mol Imaging. 2003;30:868–873. doi: 10.1007/s00259-003-1148-7. doi:10.1007/s00259-003-1148-7. [DOI] [PubMed] [Google Scholar]
- 26.Bisdas S, Kirkpatrick M, Giglio P, Welsh C, Spampinato MV, Rumboldt Z. Cerebral blood volume measurements by perfusion-weighted MR imaging in gliomas: ready for prime time in predicting short-term outcome and recurrent disease? AJNR Am J Neuroradiol. 2009;30:681–688. doi: 10.3174/ajnr.A1465. doi:10.3174/ajnr.A1465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Lev MH, Ozsunar Y, Henson JW, et al. Glial tumor grading and outcome prediction using dynamic spin-echo MR susceptibility mapping compared with conventional contrast-enhanced MR: confounding effect of elevated rCBV of oligodendrogliomas [corrected] AJNR Am J Neuroradiol. 2004;25:214–221. [PMC free article] [PubMed] [Google Scholar]
- 28.Covarrubias DJ, Rosen BR, Lev MH. Dynamic magnetic resonance perfusion imaging of brain tumors. Oncologist. 2004;9:528–537. doi: 10.1634/theoncologist.9-5-528. doi:10.1634/theoncologist.9-5-528. [DOI] [PubMed] [Google Scholar]
- 29.Laverman P, Boerman OC, Corstens FH, Oyen WJ. Fluorinated amino acids for tumour imaging with positron emission tomography. Eur J Nucl Med Mol Imaging. 2002;29:681–690. doi: 10.1007/s00259-001-0716-y. doi:10.1007/s00259-001-0716-y. [DOI] [PubMed] [Google Scholar]
- 30.Wester HJ, Herz M, Weber W, et al. Synthesis and radiopharmacology of O-(2-[18F]fluoroethyl)-L-tyrosine for tumor imaging. J Nucl Med. 1999;40:205–212. [PubMed] [Google Scholar]
- 31.Meier C, Ristic Z, Klauser S, Verrey F. Activation of system L heterodimeric amino acid exchangers by intracellular substrates. EMBO J. 2002;21:580–589. doi: 10.1093/emboj/21.4.580. doi:10.1093/emboj/21.4.580. [DOI] [PMC free article] [PubMed] [Google Scholar]