Annals of Nuclear Medicine

, Volume 25, Issue 9, pp 625–633

18F-fluoromisonidazole positron emission tomography before treatment is a predictor of radiotherapy outcome and survival prognosis in patients with head and neck squamous cell carcinoma

Authors

    • Department of Otolaryngology, Head and Neck SurgeryKobe City Medical Center General Hospital
  • Tomohiko Yamane
    • Division of Molecular ImagingInstitute of Biomedical Research and Innovation
  • Shogo Shinohara
    • Department of Otolaryngology, Head and Neck SurgeryKobe City Medical Center General Hospital
  • Keizo Fujiwara
    • Department of Otolaryngology, Head and Neck SurgeryKobe City Medical Center General Hospital
  • Shin-ya Hori
    • Department of OtolaryngologyShizuoka General Hospital
  • Yosuke Tona
    • Department of OtolaryngologyShizuoka General Hospital
  • Hiroshi Yamazaki
    • Department of Otolaryngology, Head and Neck SurgeryKobe City Medical Center General Hospital
  • Yasushi Naito
    • Department of Otolaryngology, Head and Neck SurgeryKobe City Medical Center General Hospital
  • Michio Senda
    • Division of Molecular ImagingInstitute of Biomedical Research and Innovation
Original article

DOI: 10.1007/s12149-011-0508-9

Cite this article as:
Kikuchi, M., Yamane, T., Shinohara, S. et al. Ann Nucl Med (2011) 25: 625. doi:10.1007/s12149-011-0508-9

Abstract

Objective

To evaluate the usefulness of [18F]fluoromisonidazole ([18F]FMISO)-positron emission tomography (PET) prior to the treatment of head and neck squamous cell carcinoma.

Methods

Seventeen patients with untreated HNSCC underwent pretreatment [18F]FMISO PET. Six of them underwent definitive surgery and the remaining 11 definitive (chemo-)radiotherapy. We evaluated 30 lesions from the 17 patients. SUVmax and tumor-to-muscle ratios (TMR) were measured as hypoxia indicators. Tumors equal to or above the median value were defined as tumor with high uptake of [18F]FMISO and those below as tumor with low uptake of [18F]FMISO in both indicators. Local control rates with radiotherapy, event-free survival and disease-specific survival (DSS) rates with radiotherapy or operation were compared.

Result

[18F]FMISO-PET imaging of 30 lesions resulted in a SUVmax median value of 2.3 and a TMR median value of 1.3. Local control rates with radiotherapy (20-month median follow-up duration) were significantly lower in the tumor group with high uptake of [18F]FMISO compared to the tumor group with low uptake of [18F]FMISO using either SUVmax or TMR as the hypoxic indicator (P = 0.02 and 0.04, respectively). DSS rate with radiotherapy or operation (21-month median follow-up duration) was significantly lower in the patient group with high uptake of [18F]FMISO compared to the patient group with low uptake of [18F]FMISO defined by SUVmax (P = 0.04), but was not by TMR (P = 0.57).

Conclusions

Radiotherapy outcome and survival prognosis (radiotherapy or operation) in HNSCC may be predicted by carrying out [18F]FMISO PET before treatment.

Keywords

[18F]fluoromisonidazole ([18F]FMISO)Positron emission tomography (PET)HypoxiaHead and neck cancerPrediction of prognosis

Introduction

Unlike normal blood vessels, tumor blood vessels are chaotic, irregular, and leaky, which leads to uneven delivery of nutrients and oxygen [1]. Due to insufficient oxygen reaching cancerous cells can result in a hypoxic state. Tumors present in tissue where the partial pressure of oxygen is 8–10 mmHg or less are called hypoxic, and those at 3–5 mmHg or less are advanced hypoxic tumors [2]. Hypoxic tumors exhibit resistance to chemotherapy and radiotherapy, and this is a cause of tumor recurrence and metastasis [3].

In solid carcinomas such as head and neck carcinomas [4, 5], cervical carcinomas [6] and sarcomas [7], the prognosis has been reported to be poor if the tumor was in a hypoxic state prior to treatment. However, the method for evaluating the hypoxic status of these tumors required direct insertion of a polarographic needle electrode into the tumor to measure its oxygen tension (pO2) [8]. This method has not been widely performed since the electrode insertion is highly invasive and not always possible with deep tumors.

In 1979, Chapman et al. [9] indicated that nitroimidazole had an affinity for hypoxic cells and suggested that it may be useful for extracting images of hypoxic cells since it also exhibited a radiation sensitivity-increasing action. Subsequently, Jerabek [10] and Grierson [11] used fluorine-18 labeled fluoromisonidazole ([18F]-FMISO), a 18F-labeled nitroimidazole derivative, to extract images of a hypoxic region for the first time using positron emission tomography (PET). Since PET examinations are less invasive and suitable for systemic assessments, they are used widely in cancer clinics.

We previously demonstrated that [18F]FMISO uptake decreased after neoadjuvant chemotherapy (NAC) for 13 cases of locally advanced head and neck squamous cell carcinoma (HNSCC) and concluded that [18F]FMISO PET/CT may be a useful non-invasive tool for detecting hypoxia reduction after NAC [12].

Adding four new cases here in the present study, we report on an investigation into the outcome of definitive radiotherapy or operation, and the possibility of predicting radiotherapy outcome and survival prognosis using [18F]FMISO PET/CT prior to treatment of HNSCC.

Materials and methods

Patient population

Between January 2007 and April 2009, a total of 17 patients (14 men and three women; median age, 66 years; range, 44–76 years) with locally advanced HNSCC underwent [18F]FMISO PET/computed tomography (CT) and 2-deoxy-2-[18F] fluoro-d-glucose ([18F]FDG) PET/CT before initial therapy. Patient characteristics are summarized in Table 1.
Table 1

Patient characteristics

Patient no.

Age

Sex

Primary site

Clinical T status

Clinical N status

Clinical stage (UICC 6th)

Index lesion

Definitive therapy

Locoregional recurrence

Distant metastasis

Follow-up duration (month)

Outcome

1

61

M

Oropharynx

3

1

III

LN

RTa

P, LN

Lung

25

DOD

2

67

M

Oropharynx

4a

2c

IVa

P

OPa

29

NED

3

64

M

Hypopharynx

2

1

III

P

RTa

26

NED

4

72

F

Primary unknown

LN

OP

25

NED

5

74

F

Hypopharynx

4a

2b

IVa

P

OPa

P, LN

7

DOD

6

54

M

Oropharynx

4a

2b

IVa

P

OPa

Liver

21

DOD

7

44

M

Nasopharynx

2

3

IVb

LN

RTa

22

NED

8

67

M

Hypopharynx

4a

2b

IVa

P

RTa

22

NED

9

54

M

Hypopharynx

4a

2b

IVa

LN

RTa

21

NED

10

66

M

Hypopharynx

3

0

III

P

RTa

19

NED

11

73

M

Hypopharynx

2

3

IVb

LN

OPa (LN), RTa (P)

P, LN

Mediastinal LN

11

DOD

12

68

F

Cervical esophagus

2

1

III

P

RTa

P

14

NED

13

58

F

Oropharynx

2

2b

IVa

P

OPa

3

NED

14

55

M

Hypopharynx

4a

2b

IVa

P

OPa

Lung

14

AWD

15

58

M

Hypopharynx

4a

2b

IVa

P

RTa

P, LN

13

NED

16

73

M

Hypopharynx

4a

0

IVa

P

RTa

34

NED

17

76

M

Middle ear

P

RT

P, LN

15

DOD

P primary, LN lymph node, OP operation, RT radiation, NED no evidence of disease, AWD alive with disease, DOD dead of disease

aOne cycle of neoadjuvant chemotherapy was performed before definitive therapy

All but two patients were treated with one cycle of neoadjuvant chemotherapy using nedaplatin (CDGP) and TS-1 (S-1) [13]. After NAC was completed, surgical treatment or radiotherapy was carried out as the subsequent definitive therapy. The choice of definitive therapy was determined by the following criteria: surgical treatment was employed when tumors in the T1/T2 stage were resectable by a transoral approach or tumors were in the T3/T4 stage, while radiotherapy was considered when tumors in the T1/T2 stage were not resectable by a transoral approach, patients rejected surgical treatment, or tumors were inoperable. According to the criteria, six patients underwent definitive surgery and the remaining eleven underwent definitive radiation therapy after NAC.

Definitive radiotherapy was carried out using conventional fraction sizes of 2 Gy per fraction to a total dose of 60–66 Gy with weekly concomitant chemotherapy (docetaxel 10 mg/m2) or using hyperfractionation of 69.6–72 Gy in 58–60 fractions over 6 weeks without chemotherapy. Total radiation dose and indication for use of concomitant chemotherapy were not influenced by PET ([18F]FDG and [18F]FMISO) findings but modified by the general condition or tumor size of the patient.

Concomitant chemotherapy was given by intravenous infusion and the number of cycles varied from two to five depending on the mucosal reaction to the radiotherapy. Post-operative radiotherapy was carried out without chemotherapy in patients with high risks of locoregional recurrence such as positive multiple lymph nodes, extra-capsular invasion, large tumor size (in stage T3 or T4), positive or close resection margin and angiolymphatic invasion.

The primary lesions and metastatic cervical lymph nodes of all 17 patients were examined. For those with multiple metastatic lymph nodes, one with the largest diameter was taken for analysis as a representative node. We therefore registered 30 lesions from 17 patients, which included 16 primary lesions (one patient was primary unknown) and 14 representative lymph nodes (three patients were clinically node negative).

This protocol was approved both by the Ethics Committee of the Institute of Biomedical Research and Innovation (PET institute), and that of Kobe City Medical Center General Hospital, Japan (treatment and follow-up facility). All patients gave written informed consent before entering this study.

PET/CT protocol

PET investigations were performed with a PET/CT scanner (Discovery STEP, GE Healthcare, Milwaukee, WI). CT scanning was acquired using the following parameters: 140 kVp, 40 mA (maximum), a rotation time of 0.6 s, a table speed of 35 mm per gantry rotation, a pitch of 1.75:1, and a detector row configuration of 16 × 1.25 mm.

[18F]FMISO PET/CT scanning

[18F]FMISO was synthesized as previously described [14, 15]. Patients were required to fast for 6 h, then received 510–718 (median, 592) MBq of [18F]FMISO, and emission data were acquired at 150 min p.i. of [18F]FMISO. Images were acquired at a single bed position covering head and neck tumor and cervical lymph nodes in 2-dimensional mode for 20 min. Images were reconstructed using VUE point plus with five iterations and 15 subsets for iterative reconstruction. [18F]FMISO -uptake was quantified by calculating the maximum standardized uptake value (SUVmax) and tumor-to-muscle ratio (TMR) in the following manner.

Regions of interest (ROI) were placed over the primary tumor and metastatic lymph node, and SUVmax was measured. If the lesion was difficult to assess from [18F]FMISO-PET images, recent enhanced CT or magnetic resonance (MR) images were used to assist ROI placement. To measure TMR, one or several round ROIs of 10-mm diameters were drawn over the high uptake area within the primary tumor or lymph node. For reference, ROI on contralateral posterior cervical muscles were evaluated. TMR was then calculated by dividing the mean SUV of primary tumor or lymph node ROI by the mean SUV of the muscle ROI.

We estimated the hypoxic status of the 30 lesions using the [18F]FMISO uptake parameters of SUVmax and TMR. Because no large-scale [18F]FMISO studies for HNSCC have ever been reported, the optimal cut-off value of [18F]FMISO PET uptake to estimate hypoxic state or to predict therapy outcome has not been established yet. In addition, this study is also an exploratory one with a small sample size. Therefore, we used the median value of the 30 lesions as the cut-off value of [18F]FMISO PET uptake both for SUVmax and TMR in the following manner; a lesion was defined as tumor with high uptake of [18F]FMISO if its SUVmax or TMR was equal to or above that of the median, and as tumor with low uptake of [18F]FMISO if its SUV or TMR was less than the median. This methodological criteria, by which we aimed to explore the possibility of predicting HNSCC prognosis using [18F]FMISO PET, does not mean that half of the patient are likely to be hypoxic whereas the others are not.

Evaluation of radiation responses by lesion-by-lesion analysis

To evaluate radiation response, the local control rate of 18 lesions (11 primary lesions and seven lymph nodes) from 11 patients was estimated by the Kaplan–Meier method and the difference in the local control rate between the tumor group with high uptake of [18F]FMISO and the tumor group with low uptake of [18F]FMISO by [18F]FMISO PET/CT was estimated by the log rank test.

Evaluation of patient prognosis (radiation or operation) by patient-based analysis

The most hypoxic lesion of either a primary tumor or a lymph node for each patient was defined as an “index lesion” estimated by [18F]FMISO PET/CT.

By [18F]FMISO PET/CT, all the patients were divided into two groups in the two different indicators (SUV max and TMR): the patient group with high uptake of [18F]FMISO expressing their index lesions with high uptake of [18F]FMISO and the patient group with low uptake of [18F]FMISO expressing their index lesions with low uptake of [18F]FMISO. We evaluated the difference in event-free survival (EFS) rate and disease-specific survival (DSS) rate using the Kaplan–Meier and log rank tests between the two patient groups by [18F]FMISO PET/CT.

We also dichotomized the following clinical covariates: T stage (T0–T2 vs. T3–T4), N stage (N0–N1 vs. N2–N3), UICC clinical stage (stage III vs. stage IV) and definitive therapy choice (operation vs. radiotherapy), and then performed univariate analysis using the Kaplan–Meier method and log rank testing to examine the predictive value of the other clinical risk factors for EFS and DSS at 24 months. Because of the limited number of cases, we do not perform Cox multivariate regression analysis.

Statistical analysis

PASW Statistics software version 18 for Windows (Tokyo, Japan) was used, and for all tests the level of significance was set at a value of 0.05.

Results

Figure 1 shows a scatter diagram of the hypoxia indicator values from all 30 lesions. The median SUVmax value was 2.3 and the median TMR value was 1.3 by [18F]FMISO PET/CT.
https://static-content.springer.com/image/art%3A10.1007%2Fs12149-011-0508-9/MediaObjects/12149_2011_508_Fig1_HTML.gif
Fig. 1

Scatter diagram of hypoxia indicator values. SUVmax and TMR for [18F]FMISO PET are shown for 30 lesions from 17 cases (16 primary and 14 lymphatic). The median SUVmax value was 2.3 and the median TMR value was 1.3. Lesions with values greater than median values for each hypoxic indicator were defined as tumors with high uptake of [18F]FMISO, and those with values lower than median value were defined as tumors with low uptake of [18F]FMISO

Five of the patients who underwent surgery received post-operative radiotherapy. Patient no. 11 (hypopharyngeal cancer; T2N3M0) had radical neck dissection followed by chemoradiation with no surgery for the primary lesion.

Two patients did not receive NAC because one (patient no. 4) with primary unknown carcinoma underwent a tonsillectomy before therapy to detect the primary cancer. This patient consequently received a neck dissection with a tonsillectomy followed by radiotherapy. The other (patient no. 17) had decreased renal function and received chemoradiotherapy as a definitive therapy.

The range of follow-up duration was 3–3 months, with a median time of 21 months (Table 1).

Radiotherapy outcome

The local control rates with radiotherapy (20-month median follow-up duration) were significantly lower in the tumor group with high uptake of [18F]FMISO compared to the tumor group with low uptake of [18F]FMISO using either SUVmax or TMR as the hypoxic indicator (P = 0.02 and 0.04, respectively) (Fig. 2).
https://static-content.springer.com/image/art%3A10.1007%2Fs12149-011-0508-9/MediaObjects/12149_2011_508_Fig2_HTML.gif
Fig. 2

Hypoxic status and radiotherapy outcome. Eighteen tumors from 11 cases (11 primary and 7 lymphatic) undergoing radiotherapy as definitive treatment were divided into two groups: tumors with high uptake of [18F]FMISO and tumors with low uptake of [18F]FMISO. Comparative investigations of local control rate according to Kaplan–Meier method are shown. Local control rate (20-month median follow-up duration) was significantly lower for tumor group with high uptake of [18F]FMISO using either hypoxic indicator, SUVmax (left) or TMR (right) (P = 0.02 and 0.04, respectively)

Patient prognosis (radiotherapy or operation)

No significant difference was found between the patient groups with high or low uptake of [18F]FMISO for EFS (12-month median follow-up duration) using either SUVmax or TMR as the hypoxic indicator (P = 0.07 and 0.13, respectively). On the other hand, DSS (21-month median follow-up duration) was significantly lower in the patient group with high uptake of [18F]FMISO compared to the patient group with low uptake of [18F]FMISO defined by SUVmax (P = 0.04), but was not by TMR (P = 0.57) (Fig. 3).
https://static-content.springer.com/image/art%3A10.1007%2Fs12149-011-0508-9/MediaObjects/12149_2011_508_Fig3_HTML.gif
Fig. 3

Hypoxic status and patient prognosis (radiotherapy or operation). Lesion for which hypoxic indicator was greatest among primary and lymphatic lesions was defined as “index lesion”. Comparison of groups according to hypoxic status of index lesion for EFS rate (Kaplan–Meier method) and disease specific survival (DSS) rate (Kaplan–Meier method) in 17 cases are shown. Using both hypoxia indicators, SUVmax and TMR, EFS (12-month median follow-up duration) in group with high uptake of [18F]FMISO tended to be lower than group with low uptake of [18F]FMISO, but no statistically significant difference was found (P = 0.07 and 0.13, respectively). With DSS (21-month median follow-up duration), survival rate was significantly lower in group with high uptake of [18F]FMISO than in group with low uptake of [18F]FMISO only when SUVmax was used (P = 0.04). Upper figure EFS, lower figure DSS. Left SUVmax, right TMR

Log-rank test of PET parameters and other clinical risk factors for EFS and DSS at 24 months are summarized in Table 2.
Table 2

Univariate log-rank test of clinical risk factors for EFS and DSS at 2 years

Subgroup

No. of patients

EFS

DSS

Survival rates (%)

P value

Survival rates (%)

P value

FMISO-PET SUV max

 SUV max <2.3

6

83

0.07

100

0.04

 SUV max ≥2.3

11

30

 

27

 

FMISO-PET TMR

 TMR <1.3

5

80

0.13

75

0.57

 TMR ≥1.3

12

37

 

51

 

T stagea

 T0–T2

5

50

0.83

75

0.92

 T3–T4

10

50

 

50

 

N stagea

 N0–N1

5

60

0.38

67

0.36

 N2–N3

10

44

 

62

 

UICC stagea

 III

4

50

0.70

50

0.71

 IV

11

50

 

67

 

Definitive therapyb

 Operation

7

33

0.32

44

0.16

 Radiotherapy

10

60

 

58

 

aPatient no. 4 and no. 17 are excluded from T stage/N stage/UICC stage subgroups

bPatient no. 11 is included in operation-subgroup

Representative case

[18F]FMISO-PET/CT images from a representative patient (patient no. 1) are shown in Fig. 4.
https://static-content.springer.com/image/art%3A10.1007%2Fs12149-011-0508-9/MediaObjects/12149_2011_508_Fig4_HTML.gif
Fig. 4

Presentation of a typical case. Patient no. 1: 61-year-old male with oropharyngeal carcinoma (right lateral wall), findings from [18F]FMISO-PET/CT of cT3N1M0 case and [18F]FDG-PET/CT from imaging performed at same time. Upper image [18F]FDG, lower image [18F]FMISO. Left PET image, middle CT image, right PET-CT fusion. Arrowhead primary lesion. Arrow metastatic cervical lymph node. [18F]FDG-PET/CT: primary SUVmax 13.3, and cervical lymph node SUVmax 7.3. [18F]FMISO-PET/CT: primary SUVmax 2.2, TMR 1.7; cervical lymph node SUVmax 3.0, TMR 2.0 [18F]FMISO accumulation was found in both primary and cervical lymph nodes. Primary lesion had a greater diameter than cervical lymphatic lesion, and [18F]FDG SUVmax was also greater for primary than for cervical lymph node, but less [18F]FMISO accumulation was observed for primary than for cervical lymph node. Definitive chemoradiotherapy was carried out after NAC, but cervical lymphatic recurrence occurred 12 months after start of treatments, and primary lesion reappeared after 26 months, resulting in death from primary disease

Discussion

Eschmann et al. [16] previously examined 26 cases of head and neck carcinoma that had undergone radiotherapy and reported that prognosis can be predicted 4 h after administration of [18F]FMISO, by setting the TMR cutoff at 1.6, which is in line with the present results. In the present study, a significant difference was found in the local control rate with radiotherapy between the tumor groups with high or low uptake of [18F]FMISO, when either of the hypoxia indicators (SUVmax and TMR) was used by setting the cutoff value at the median for each (2.3 and 1.3, respectively) (Fig. 2).

DNA free radicals arising because of the ionizing effect in radiotherapy lead to cell death by the formation of hydroperoxide in the presence of oxygen. In a hypoxic state, the free radicals are reduced, and this recovery leads to resistance to treatment [3]. The sensitivity-increasing action of oxygen on radiotherapy is known as the oxygen effect, and in normal linear energy transfer (LET) radiotherapy, the effect is substantially constant at a partial pressure of oxygen of approximately above 20 mmHg, and the oxygen enhancement ratio (OER) is approximately 2.5–3 [17]. The OER drops rapidly if the partial pressure of oxygen is 3–10 mmHg or less [18, 19]. This threshold value substantially matches that for the partial pressure of oxygen where [18F]FMISO is trapped in the cells (approximately 2–3 mmHg) [2022]; therefore, it could be surmised that the radiotherapy outcome can be predicted by [18F]FMISO PET/CT.

In terms of survival prognosis, no significant difference was found in the EFS rate between the patient groups with high or low uptake of [18F]FMISO using either of the hypoxia indicators (SUVmax or TMR). However, the DSS rate was found to be significantly lower in the patient group with high uptake of [18F]FMISO only when SUVmax was used (Fig. 3). Previously, the prognosis was shown to be unfavorable for tumors with a hypoxic status (measured by a polarographic oxygen needle electrode) before treatment for head and neck carcinomas [4, 5] and Nordsmark et al. [23] indicated that tumor hypoxia (measured by a polarographic oxygen needle electrode) does not depend on tumor size, grade, or patient hemoglobin levels, and is therefore an independent predictor of outcome in head and neck carcinomas. The present results suggest that survival prognosis may be predicted with [18F]FMISO PET/CT uptake alone that was extracted from the most hypoxic region. In the present study, however, no other clinical indicators such as T stage, N stage, UICC clinical stage, and the choice of definitive therapy were able to predict radiation outcome or patient prognosis (Table 2). Thus, the present results have shown that evaluating hypoxic status by [18F]FMISO PET/CT before treatment may have more potential usefulness in predicting survival prognosis (whether radiation or operation is performed) rather than classifying TNM stages in patients with HNSCC.

Although [18F]FMISO PET is shown to be beneficial for predicting HNSCC prognosis, appropriate timing and effective hypoxia indicators for [18F]FMISO-PET imaging are still to be determined. Since [18F]FMISO is lipophilic, its clearance from hypoxic regions and blood plasma is poor, and if imaging is carried out too soon, the low signal to noise ratio reduces image clarity. The optimal imaging timing is 2–4 h after its injection [16, 24, 25]. Eschmann et al. [16] reported that imaging at 4 h provided superior contrast than that at 2 h and TMR at 4-h injection allowed better outcome prediction than at 2 h. By contrast, Krohn and his group [26] said that [18F]FMISO images can be produced 1 h after injection, although contrast is somewhat low and their laboratory typically images 90–150 min after injection. The later [18F]FMISO imaging produced after injection, the higher contrast imaging can be obtained, however, 4 h are thought to be too long for patients to wait for imaging. This is why, in the present study, PET scanning was performed 150 min after injection according to Krohn’s report [26] to lighten the patient’s time-dependent burden as much as possible. Since [18F]FMISO has a lower SUVmax and lower contrast with surrounding organs than [18F]FDG (Fig. 4), comparing [18F]FMISO uptake into the target tumor with that into the reference tissues is said to be more reasonable than using SUVmax only [27]. Thus, previous investigations have used blood count ratios [28, 29], muscle ratios [16] and kinetic analysis [30] for hypoxia indicators. In the present study, SUVmax and TMR were both useful hypoxic parameters to predict radiotherapy outcome, however, it is interesting to note that not TMR but SUVmax was only a statistically significant predictor of DSS rate. One plausible explanation for this bizarre result may be our early PET imaging timing of 150 min. Eschmann et al. [16] reported that the values of SUVmax at 2 h did not change compared to those at 4 h (1.85 ± 0.49 vs. 1.86 ± 0.56), however, the values of TMR increased from 1.54 ± 0.41 at 2 h to 1.72 ± 0.52 at 4 h. This finding may indicate that when we use SUVmax as a hypoxic indicator, imaging as early as 2 h is also useful as late as 4 h to estimate the hypoxic status of tumor, whereas when we use TMR, imaging as late as 4 h is more reliable than that as early as 2 h. It seems likely that TMR is a good hypoxic parameter, better than SUV max, if only PET imaging is performed such as late at 4 h [16].

In conclusion, the present study has shown that it may be possible to predict radiation outcomes and predict survival prognosis for HNSCC by evaluating SUVmax alone without measuring TMR at 150 min after [18F]FMISO injection. Evaluating SUVmax is simpler than TMR and has higher practicability for future clinical use. Since this is a study with a small sample size, further study with a larger sample size is necessary before determining the usefulness and appropriate parameters of [18F]FMISO-PET examination for evaluating HNSCC.

Conclusion

[18F]FMISO-PET imaging of 30 lesions from 17 cases of untreated HNSCC resulted in a SUVmax median value of 2.3 and a TMR median value of 1.3. Following lesion assessment, tumors with equal to or above the median value were defined as high uptake of [18F]FMISO and those below the median value were defined as low uptake of [18F]FMISO; the tumor group with high uptake of [18F]FMISO had a significantly lower local control rate with radiotherapy. When SUV max was used as the hypoxic indicator, the DSS rate was significantly lower in the patient group with high uptake of [18F]FMISO (operation or radiotherapy). The current results suggest that the radiotherapy outcome and survival prognosis (radiotherapy or operation) in HNSCC may be predicted by carrying out [18F]FMISO PET/CT before treatment.

Acknowledgments

This study was financially supported by Kasahara Fund for the promotion of cancer research.

Copyright information

© The Japanese Society of Nuclear Medicine 2011