European Journal of Nuclear Medicine and Molecular Imaging

, Volume 34, Issue 5, pp 651–657

1-[11C]-acetate PET imaging in head and neck cancer—a comparison with 18F-FDG-PET: implications for staging and radiotherapy planning

Authors

    • Department of Radiation Sciences, Section of OncologyUmeå University Hospital
  • Jens Sörensen
    • Uppsala Imanet AB PET Center
  • Mikael Karlsson
    • Department of Radiation Sciences, Radiation PhysicsUmeå University Hospital
  • Ingela Turesson
    • Department of OncologyUppsala University Hospital
  • Bengt Langström
    • Uppsala Imanet AB PET Center
  • Per Nilsson
    • Department of Radiation PhysicsLund University Hospital
  • Lena Cederblad
    • Department of OncologyUppsala University Hospital
  • Jan Bertling
    • Hermes Medical Solution
  • Katrine Riklund
    • Department of Radiation Sciences, Diagnostic RadiologyUmeå University Hospital
  • Silvia Johansson
    • Department of Radiation Sciences, Section of OncologyUmeå University Hospital
Original Article

DOI: 10.1007/s00259-006-0298-9

Cite this article as:
Sun, A., Sörensen, J., Karlsson, M. et al. Eur J Nucl Med Mol Imaging (2007) 34: 651. doi:10.1007/s00259-006-0298-9
  • 118 Views

Abstract

Purpose

The aim of this study was to evaluate the feasibility of using 1-[11C]-acetate positron emission tomography (ACE-PET) to detect and delineate the gross tumour volume of head and neck cancer before radiotherapy, and to compare the results with those obtained using 18F-fluoro-2-deoxy-D-glucose (FDG) PET.

Methods

Ten patients with histologically verified squamous cell carcinoma were investigated by FDG-PET and dynamic ACE-PET prior to radiotherapy. The two scans were performed on the same day or on consecutive days, except in one patient in whom they were done 5 days apart. Diagnostic CT or MRI was performed in all patients. The image data sets were analysed both visually and semi-quantitatively. All primary tumours and metastases were delineated automatically by using the 50% threshold of maximum radioactivity corrected for background. The mean standardised uptake value (SUV) and the tumour volumes were evaluated and compared.

Results

All ten primary tumours were detected by ACE-PET, while nine primaries were detected by FDG-PET and CT and/or MRI. The ACE SUV tended to be lower than the FDG SUV (5.3±2.7 vs 9.6±7.0, p=0.07). The tumour volumes delineated with ACE were on average 51% larger than the FDG volumes (p<0.05). ACE-PET identified 20/21 lymph node metastases, while only 13/21 lesions were detected by FDG-PET and 16/21 lesions by CT or MRI.

Conclusion

ACE-PET appears promising for the staging of head and neck cancer. The biological information provided by both FDG and ACE must be carefully validated before it can be used in clinical routine for radiation treatment planning. More studies are needed to evaluate the differences in volumes and to confirm the clinical potential of both FDG and ACE-PET, especially in radiotherapy.

Keywords

11C-acetate18F-FDGPETHead and neck cancerSUV

Introduction

Head and neck squamous cell carcinoma is curable when diagnosed at an early stage [1]. Both accurate diagnosis and accurate staging of tumours are important for prognosis and determination of treatment strategy. Conventional anatomical imaging techniques, such as computed tomography (CT), magnetic resonance imaging (MRI) and ultrasonography, are routinely used for evaluation of tumour size and local tumour extent. However, there are inherent limitations associated with all these techniques [2].

Positron emission tomography (PET) may improve the ability to non-invasively detect the biological characteristics of tumours. 18F-fluoro-2-deoxy-D-glucose (FDG) PET has been widely applied for tumour staging, distinguishing tumour recurrence from other conditions such as fibrosis and predicting treatment response in head and neck cancer [25]. Use of PET to delineate gross tumour volume is also increasing [6].

FDG is an analogue of glucose with high uptake in malignant cells owing to the increased energy requirement [7]. However, FDG is not a specific tumour marker. It accumulates in inflammatory tissues and also has limitations in detecting well-differentiated tumours [8, 9]. Development of new tracers to improve the efficiency of PET imaging in head and neck cancer is therefore warranted.

Several recent studies have demonstrated that 1-[11C]-acetate (ACE) may be a useful tracer for various cancer types, such as lung cancer, hepatocellular carcinoma, renal cancer, prostate cancer and astrocytomas [1014]. Ho et al. [11] reported that well-differentiated hepatocellular carcinoma displayed increased ACE uptake and minimal FDG uptake. These findings indicated that ACE has a high sensitivity and specificity as a radiotracer complementary to FDG in the PET imaging of hepatocellular carcinoma. Sandblom et al. [15] reported PET with ACE to be a promising method for the early detection and localisation of prostate cancer recurrence.

While these results are promising, currently only sparse data are available on the use of ACE-PET in head and neck cancer. One report showed the clearance rate of ACE in nasopharyngeal carcinoma to be slower than in normal nasopharyngeal tissues [16]. Experimental and clinical studies have shown that acetate is incorporated into the lipid pool with a high lipid synthesis rate [17] and low oxidative metabolism [16]. This implies that ACE may be a valuable PET tracer in head and neck cancer.

FDG-PET has already been accepted as a complementary tool in treatment planning at many radiotherapy departments [1820]. It has been shown to enhance the accuracy of tumour definition and it might permit and improve individualised radiotherapy treatment. However, to our knowledge, there are no reports on tumour delineation with ACE. The aim of this study was to evaluate the feasibility of using ACE-PET to detect and delineate head and neck cancer before radiotherapy, and to compare the results with those obtained with FDG-PET.

Materials and methods

Patients

Ten consecutive patients (eight males and two females; median age 56, range 18–77 years) with histologically confirmed squamous cell carcinoma of the head and neck were included in the study. None of the patients suffered from diabetes. The patients had been treated with neither radiotherapy nor chemotherapy prior to inclusion and were candidates for radiotherapy. The clinical characteristics, including the stage and location of the primary tumour, are shown in Table 1. Conventional staging of the tumours was performed by CT (n = 9) or MRI (n = 1), histopathology and clinical examination. Histological confirmation was obtained by guided biopsies in all the primary tumours and at most metastatic sites. The metastases not verified with biopsies (n = 5) were deemed to be malignant based on the combination of all the available information, including 3-month follow-up. All patients participating in the study provided informed consent. The study was accepted by the ethical committee of the participating hospital.
Table 1

Clinical characteristics of the patients

Patient

Sex

Age (years)

Stage

Location

Histology (diff)

1

M

77

T4N2cM0

Larynx

Low

2

F

57

T2N0M0

Nose

Moderate

3

M

59

T2N0M0

Nose

High

4

M

53

T3N0M0

Nose/sinus

Low

5

F

67

T4N3M1

Tonsil

Low

6

M

59

T3N1M0

Tonsil

Low

7

M

47

T4N3M0

Epipharynx

Low

8

M

64

T2N2aM0

Tonsil

Low

9

M

18

T3N3M0

Epipharynx

Low

10

M

45

T2N2bM0

Tongue base

High

M male, F female, diff cell differentiation

PET imaging

In all patients, both FDG-PET and ACE-PET were performed before radiotherapy. ACE- and FDG-PET scans were performed on the same day or on consecutive days, except in one patient in whom they were done 5 days apart. PET studies were carried out with a dedicated PET scanner (Siemens ECAT HR+, Knoxville, TN, USA) or with a PET/CT scanner (GE Discovery ST, Milwaukee, WI, USA). All patients were normoglycaemic and had fasted for at least 6 h before tracer injection.

ACE-PET imaging

Six patients were studied with dedicated PET and four patients were investigated with PET/CT. A 32-min dynamic emission scan was performed immediately after intravenous injection of 10 MBq/kg body weight ACE. The scan time was 12 × 5 s, 6 × 10 s, 4 × 30 s, 4 × 60 s, 2 × 120 s and 4 × 300 s. Frame 30 (17–22 min after injection) generally provided the best image quality, with the highest tumour to background ratio, and was therefore chosen for subsequent data analysis.

FDG-PET imaging

Whole-body scanning was performed 1 h after intravenous injection of 5 MBq/kg body weight FDG. Scanning was performed in two steps: First, a scan was done from the thorax to the pelvis with the arms resting above the head. Secondly, the scan was performed from the eyebrows to the upper mediastinum with the arms alongside the body. The images were reconstructed with a zoom factor of 2 to improve the resolution. Six patients were examined by PET/CT (4 min per bed position) and four patients were studied by PET alone (5 min per bed position). The patients were instructed to remain recumbent and to avoid using the voice or performing other activities involving the neck muscles during the uptake period.

Data analysis

All PET images were co-registered with the diagnostic contrast-enhanced CT or MRI and dose-planning CT images by a normalised mutual information procedure supported by manual correction using Hermes Multimodality software (Nuclear Diagnostics, Stockholm, Sweden). The CT component of the PET/CT was only used for attenuation correction. FDG-PET and ACE-PET images were analysed both qualitatively and quantitatively, using Hermes Volume Display version V2β. In qualitative analysis, PET images were interpreted visually by two nuclear medicine physicians and any disagreement was resolved by consensus. The tumour uptake of FDG and ACE was graded as negligible, mild, moderate or intense in comparison with the contralateral or surrounding tissues. Abnormal uptake equal to or exceeding mild was considered positive. In quantitative analysis, the mean standardised uptake value (SUV) and tumour volumes delineated by ACE- and FDG-PET were evaluated. SUV was calculated as mean radioactivity concentration in the tumour volume (Bq/cc) divided by injected activity (Bq) per gram body weight. For lesions with negligible uptake, similar tumour volumes were drawn manually using visually correlated fusion images.

Each tumour volume on FDG-PET and ACE-PET was delineated automatically by tracing an isoactivity pixel value set to the 50% threshold of the maximum radioactivity corrected for background. The background was measured from a separately drawn region of interest (ROI) adjacent to but at a safe distance from the tumour. The isoactivity pixel value of each volume was calculated as:
$${\text{Isoactivity}}\,\,{\text{pixel}}\,\,{\text{value}} = {\left( {{\text{MPV}}_{{{\text{tumour}}}} + {\text{APV}}_{{{\text{background}}}} } \right)} \times 50\% $$
where MPV is the maximum pixel value and APV is the average pixel value of the background ROI. This approach takes into account the variable background activity, effectively cancels the effect of varying background uptake on tumour volume measurements and was found to be highly reproducible. In those cases in which the tumour location was near to the salivary glands with normally high physiological uptake of ACE, the tumour volumes were adjusted manually based on the combined information of CT and PET. Only one primary tumour volume and five metastases needed manual adjustments for this reason.

Statistical analysis

The relationship between FDG SUV and ACE SUV was determined by Pearson’s correlation coefficient. ANOVA was used to compare the tracer uptake with histological cell differentiation. The differences between the FDG and ACE SUVs and volumes were analysed by non-parametric Wilcoxon signed rank test. Volumes of metastases were presented by median ± interquartile, since it did not show a normal distribution. A p value < 0.05 was considered statistically significant. Calculations were performed by SPSS version 11.5.

Results

Primary tumours

Results of the qualitative and semi-quantitative comparisons between ACE-PET and FDG-PET with regard to the detection of primary tumours are shown in Table 2. All of the primary tumours (10/10) were detected by ACE-PET, while nine were detected by FDG-PET and CT or MRI. PET and CT images are shown in Fig. 1 for one of the patients with cancer of the tonsil. The primary tumour of patient no. 10 in the left base of the tongue could not be detected by either FDG-PET (SUV 1.9) or CT. ACE-PET, however, clearly visualised the tumour with high uptake (SUV 3.7) (Fig. 2). One of the contralateral lymph node metastases was also visualised in this patient.
Table 2

Qualitative and semi-quantitative comparisons between ACE- and FDG-PET with regard to the detection of primary tumours

Patient

ACE-PET

FDG-PET

SUV

Visual

Volumes (cc)

SUV

Visual

Volumes (cc)

1

9.2

+++

9.5

13.5

+++

3.1

2

2.5

+

3.9

3.8

++

3.4

3

6.0

+++

8.6

24.5

+++

6.0

4

2.7

+

24.6

4.4

++

13.0

5

5.7

+++

17.5

12.9

+++

10.1

6

3.8

++

15.0

7.9

+++

10.5

7

10.6

+++

4.7

6.7

+++

5.3

8

4.9

++

1.8

4.7

++

1.5

9

3.9

++

15.1

15.4

+++

13.5

10

3.7

++

1.8

1.9

NM

Mean±SD

5.3  ± 2.7

 

11.2 ± 7.4

9.6 ± 7.0

 

7.4 ± 4.5

SUV standardised uptake value; − negligible uptake, + mild uptake, ++ moderate uptake, +++ intense uptake, SD standard deviation, NM not measurable

https://static-content.springer.com/image/art%3A10.1007%2Fs00259-006-0298-9/MediaObjects/259_2006_298_Fig1_HTML.jpg
Fig. 1

Patient no. 6 with squamous cell carcinoma in the left tonsil. a CT, b FDG-PET, c fused FDG-PET, d ACE-PET, e fused ACE-PET. The tumour exhibited increased uptake of FDG (SUV 7.9) and ACE (SUV 3.8)

https://static-content.springer.com/image/art%3A10.1007%2Fs00259-006-0298-9/MediaObjects/259_2006_298_Fig2_HTML.jpg
Fig. 2

Patient no. 10 with squamous cell carcinoma in the left base of the tongue and metastases at the right side of the neck. a CT, b FDG-PET, c fused FDG-PET, d ACE-PET, e fused ACE-PET. ACE-PET clearly exhibited high uptake in the primary tumour, with a SUV of 3.7. However, FDG-PET failed to show significantly increased uptake, with a SUV of only 1.9, and missed the primary tumour. CT has also yielded a false negative result. All of the images showed the contralateral lymph node metastasis

The range (mean±SD) of ACE SUV and FDG SUV was 2.5–10.6 (5.3 ± 2.7) and 1.9–24.5 (9.6 ± 7.0) respectively. FDG SUV tended to be higher than ACE SUV, although the difference was not statistically significant (p = 0.07). No positive relation was found between ACE SUV and FDG SUV (r = 0.296, p = 0.41). Furthermore, neither FDG SUV nor ACE SUV correlated with the histological grade of cell differentiation (p = 0.44 and p = 0.81, respectively).

Metastases

A total 21 metastatic lesions were detected in seven patients (Table 3). Twelve of 21 lesions (12/21) were visualised by all used techniques. Almost all lymph node metastases (20/21) were detected by ACE-PET. The only false negative lesion on ACE-PET had a volume of 0.8 cc. This small lesion displayed increased uptake on FDG-PET (SUV 3.8) and was also visualised by CT. Thirteen of 21 lesions were true positive on FDG-PET, whereas eight lymph node metastases in three patients were false negative; 16/21 metastases were true positive on CT or MRI, while five lesions in two patients were false negative (data not shown). In patient no. 10, four lesions were false negative on both FDG-PET and CT, but all of them were true positive on ACE-PET (Fig. 3). The range of ACE SUV was 2.4–6.2 (4.0 ± 1.3) and that of FDG SUV, 0.9–10.0 (4.47 ± 3.3). No significant difference (p = 0.52) or correlation (r = 0.383, p = 0.09) was found between FDG SUV and ACE SUV.
Table 3

Qualitative and semi-quantitative comparisons between ACE- and FDG-PET with regard to the detection of metastases

Patient

No. of metastases

ACE-PET

FDG-PET

SUV

Visual

Volume (cc)

SUV

Visual

Volume (cc)

1

1

5.9

+++

1.2

1.5

NM

2

5.9

+++

2.2

5.2

+++

1.4

3

5.9

+++

1.8

5.4

+++

0.8

4

4.7

++

1.9

4.9

++

1.7

5

1

5.1

+++

101.0

13

+++

82.4

2

3.1

++

1.4

3.8

++

1.7

6

1

2.9

++

1.4

3.6

++

1.2

7

1

4.5

++

1.5

1.7

NM

2

3.9

++

1.5

1.3

NM

3

4.6

++

1.7

1.5

NM

8

1

5.8

+++

4.6

7.6

+++

3.9

9

1

3.0

NM

3.8

++

0.8

2

2.8

+

1.8

6.4

+++

1.5

3

2.6

+

3.5

6.2

+++

2.9

4

3.7

++

14.5

8.1

+++

13.2

5

3.0

++

4.8

5.7

+++

3.7

10

1

6.2

+++

14.5

10.0

+++

10.5

2

2.4

+

1.1

1.1

NM

3

2.6

+

1.6

1.2

NM

4

2.7

+

0.8

0.9

NM

5

3.5

++

1.8

1.0

NM

The median volume of the metastases using ACE was 2.9 ± 10.3 cc, compared with 2.3 ± 7.4 cc when FDG was used

SUV standardised uptake value, − negligible uptake, + mild uptake, ++ moderate uptake, +++ intense uptake, NM not measurable

https://static-content.springer.com/image/art%3A10.1007%2Fs00259-006-0298-9/MediaObjects/259_2006_298_Fig3_HTML.jpg
Fig. 3

A lymph node metastasis at the right side of the neck in patient no 10. a CT, b FDG-PET, c fused FDG-PET, d ACE-PET, e fused ACE-PET. The metastasis displayed increased uptake of ACE (SUV 3.5), but FDG-PET was false negative with no increased uptake (SUV 1.0). CT also missed this lymph node metastasis

High physiological uptake of ACE was found in the salivary glands and tonsils, and the images displayed a lower ratio of uptake in tumour to background compared with FDG-PET images.

Volumes

The calculated volumes of the primary tumours and metastases delineated by ACE-PET and FDG-PET are shown in Tables 2 and 3, respectively. The mean primary tumour volumes derived from ACE-PET were 11.2 ± 7.4 cc (range 1.8–24.6 cc, n = 9) compared with 7.4 ± 4.5 cc (range 1.5–13.5 cc, n = 9) for FDG-PET. The mean ACE-PET volumes were thus 51% larger than the volumes delineated by FDG-PET (p = 0.02). In particular, in patient no. 1, the ACE volume of the primary was three times larger than the corresponding FDG volume, while for patient no. 4 it was larger by a factor of almost 2. Only in patient no. 7 was the ACE-delineated tumour volume smaller than the FDG-delineated volume. The ratio of ACE volume to FDG volume hence exceeded unity in eight of the nine patients (Fig. 4).
https://static-content.springer.com/image/art%3A10.1007%2Fs00259-006-0298-9/MediaObjects/259_2006_298_Fig4_HTML.gif
Fig. 4

The ratio of the ACE and FDG volumes of the primary tumours. Eight of the nine volume ratios between ACE and FDG exceeded unity

Volumes of metastases delineated by ACE-PET were also larger (23%, p = 0.005) than those delineated by FDG-PET. The median volumes of lymph node metastases on ACE-PET were 2.9 ± 10.3 cc, compared with slightly lower values when FDG was used (2.3 ± 7.4 cc).

Discussion

The results of this study in ten patients indicate that ACE-PET may be more sensitive than FDG-PET for the detection of primary tumours and metastases in patients with head and neck squamous cell carcinoma.

It is well known that FDG reflects the tumour glycolysis and may have diagnostic limits in the identification of well-differentiated tumours [9]. In patient no. 10, the primary lesion with high cell differentiation and four lymph node metastases showed no increased uptake of FDG. This was probably due to the low glycolysis in the tumour. All the lesions displayed increased uptake of ACE, which provided information of importance for treatment strategies and prognostic evaluation. Only one small metastasis was missed by ACE-PET, as it was obscured by the high physiological uptake in the adjacent salivary gland. The other false negative lymph node metastases using FDG-PET may have been caused by the individual tumour characteristics.

ACE was initially used as a tracer of myocardial oxidative metabolism. It is converted to acetyl coenzyme A in the mitochondria, followed by quick clearance as 11C-labelled carbon dioxide through citric acid cycle in myocardial myocytes. The exact mechanism of acetate accumulation in tumour cells is not yet known. Yoshimoto et al. [17] reported that high uptake in tumour cells was caused by enhanced lipid synthesis, which correlated with tumour growth activity by increased membrane synthesis. This suggests that ACE may be an important probe of the anabolic pathway of metabolism in cancer tissues. On the dynamic scan, ACE uptake by the tumour reached a peak value in the initial minutes after injection, followed by a slow clearance compared with the normal tissues. This might indicate that the ACE signal reflects both tumour blood flow and mixed catabolic and anabolic pathways. Further pharmacokinetic studies of ACE-PET are needed to clarify the mechanism of tumour uptake in head and neck cancer.

The short half-life of 11C-labelled acetate allows patients to undergo repeated PET scan on the same day, but also limits its use without an on-site cyclotron. The use of 18F-labelled acetate should be explored, potentially expanding the use of ACE for research and clinical oncological applications.

FDG-PET and especially FDG-PET/CT is currently gaining acceptance as a complementary tool for guidance in delineating the gross target volume in radiotherapy. This imaging modality may provide additional biological information on the tumour, optimise tumour volume delineation and hence spare the normal tissues from an unnecessary radiation dose [2123]. However, the methodology to delineate tumour volumes with PET is not straightforward. Different methods, leading to different results, have been presented in the literature [18]. The problem is too complex for setting a fixed standard to be used for contouring of all tumour volumes by PET in a clinical setting, especially in the head and neck region, with its complicated anatomical structures. Low reproducibility was shown in studies where the tumour volumes were drawn manually [4, 24] . Some studies [25, 26] traced the tumour volumes based on a certain intensity level without considering the background activity. Tumour delineation in the present study was traced in three dimensions based on an isoactivity value halfway between the maximum SUV and the adjacent background. The intensity level representing the actual activity distribution may depend on a number of specific factors. However, for these types of comparative study the choice of intensity level (50% in our case) was assumed to be less critical. The only operator-dependent input needed with this approach was definition of the background activity, while repetitive and blinded measurements indicated that this introduced a variability in the final volumes of only a few percent. However, this tool might be confounded in tumours with obvious heterogeneous uptake.

One finding of this study of possible interest and importance for the clinician is the large difference between tumour volumes delineated by ACE and by FDG. This finding raises several questions: Does FDG-PET accurately define the tumour volume? Should a different dose regimen be sought for the excess 51% volume delineated by ACE-PET compared with FDG-PET? In addition to confirmation of our findings in a larger patient group, a deeper understanding of the differential metabolic pathways for these tracers may be needed to answer these questions. As previously reported, ACE and FDG may be taken up by different parts of certain tumours [11]. As suggested by the current study, FDG may be a relatively poor tracer for some head and neck cancers in which metabolic needs are sustained by routes other than glycolysis.

Conclusion

Increased acetate uptake was a prominent feature of the primary tumours and lymph node metastases of head and neck squamous cell carcinomas in this study. ACE-PET provided diagnostic images of good quality and may be a more sensitive tool than FDG-PET for staging of head and neck tumours. The use of ACE-PET for tumour volume delineation resulted in 51% larger volumes than were obtained with FDG-PET. Both ACE and FDG must be carefully validated before they are routinely used in a clinical setting. More studies are needed to evaluate the difference in volumes and to confirm the clinical potential of using PET/CT with different tracers in radiotherapy treatment planning.

Acknowledgements

The authors wish to thank the staff at Uppsala Imanet PET Center for assistance in performing the studies. This work was supported by grants from Uppsala University Amersham PET Research Fund, Laryngfonden and Umeå University Lion’s Cancer Research Foundation, Sweden. None of the authors have any financial conflict of interest.

Copyright information

© Springer-Verlag 2006