Oral Oncology 51 (2015) 536–540
Contents lists available at ScienceDirect
Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology
18FDG SUV in the primary tumor and lymph node metastases is not predictive for development of distant metastases in high risk head and neck cancer patients Jasmijn Meeuwis a, Otto S. Hoekstra b, Birgit I. Witte c, Ronald Boellaard b, C. René Leemans a, Remco de Bree a,d,⇑ a
Department of Otolaryngology-Head and Neck Surgery, VU University Medical Center, Amsterdam, The Netherlands Department of Radiology & Nuclear Medicine, VU University Medical Center, Amsterdam, The Netherlands c Department of Epidemiology and Biostatistics, VU University Medical Center, Amsterdam, The Netherlands d Department of Head and Neck Surgery, UMCU Utrecht Cancer Center, Utrecht, The Netherlands b
a r t i c l e
i n f o
Article history: Received 28 October 2014 Received in revised form 7 February 2015 Accepted 10 February 2015 Available online 27 February 2015 Keywords: Head and neck squamous cell carcinoma Distant metastases 18F-fluorodeoxyglucose Standardized uptake value
s u m m a r y Background: Pretreatment screening on distant metastases is particularly useful in head and neck squamous cell carcinoma (HNSCC) patients with high risk factors. Methods: In a retrospective study of 88 patients with previously identified clinical high risk factors the predictive value of standardized uptake value (SUV) of 18F-fluorodeoxyglucose (FDG) in the primary tumor and in the lymph node metastases for the development of distant metastases was examined. Different SUVs corrected for plasma glucose levels and body mass index were calculated and analysed in different patient groups: all patients (n = 88), patients with a follow up >6 months (n = 73), not previously treated patients (n = 51) and not previously treated patients with a follow up >6 months (n = 40). Results: Twenty-four of the 88 (27%) high risk HNSCC patients were diagnosed with distant metastases during screening and follow up. No correlation was found between different SUVs of the primary tumor and lymph nodes metastases and the development of distant metastases. Conclusion: SUVs of primary tumor and lymph node on FDG-PET are not predictive for distant metastases in a (selected) group of patients with already high risk factors for distant metastases. Ó 2015 Elsevier Ltd. All rights reserved.
Introduction The detection of distant metastases at the time of initial evaluation influences the prognosis and selection of treatment in head and neck squamous cell carcinoma (HNSCC) patients [1]. Patients with distant metastases are generally not considered curable and almost always receive only palliative treatment [2]. Detection of distant metastases may avoid futile extensive locoregional treatments with unnecessarily burden to the patient and use and costs of health resources. The incidence of distant metastasis from HNSCC at presentation is generally too low to warrant an extensive radiographic staging evaluation [3]. In the United States Surveillance Epidemiology
⇑ Corresponding author at: Department of Otolaryngology/Head and Neck Surgery, Vrije Universiteit Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, PO Box 7057, 1007 MB Amsterdam, The Netherlands. Tel.: +31 20 4443689; fax: +31 20 4443688. E-mail address: [email protected] (R. de Bree). http://dx.doi.org/10.1016/j.oraloncology.2015.02.002 1368-8375/Ó 2015 Elsevier Ltd. All rights reserved.
and End Results (SEER) database distant metastases were reported in 2066 out of 73,247 (2.82%) HNSCC patients [4]. This low number of distant metastases in unselected HNSCC patients emphasizes the need for selection based on risk factors. Jäckel and Rausch [5] found that screening is particularly useful in patients with advanced stage disease, local and/or regional recurrences and second primary tumors. Loh et al. [6] evaluated screening in HNSCC patients using chest CT and found T4 and/or N2 or N3 to be risk factors for the development of distant metastases. Leong et al. [7] found in their series of 102 patients who underwent screening by chest X-ray and CT that of the patients with positive screening, 86% had T3 or T4 disease and 71% had N2 or N3 disease. Haerle et al. [8] detected in 21% of patients with advanced stage disease (T3/4 and/or N2/3) distant metastases by PET–CT and follow-up. In previous studies we identified and validated clinically three or more lymph node metastases, bilateral lymph node metastases, lymph node metastases of 6 cm or larger, low jugular lymph node metastases, locoregional recurrence and second primary tumors as
537
J. Meeuwis et al. / Oral Oncology 51 (2015) 536–540
relevant high risk factors. Using these selection criteria, distant metastases were detected in 29–33% of the patients during initial screening or within 12 months follow-up after initial screening [9–11]. Several studies in HNSCC patients have found a correlation between standardized uptake value (SUV) of 18F-fluorodeoxyglucose (FDG) in primary tumor or lymph node metastases and the development of distant metastases [12–17]. However, Haerle et al. [8] could not find a correlation between SUV max of the primary tumor and the development of distant metastases. Several other factors thought to be associated with an increased risk for distant metastases have been reported in the literature [18]. Histopathologically determined risk factors obtained from the surgical specimen are generally not available for treatment planning. Pathologically and radiologically assessed extranodal spread has been associated with development of distant metastases [19]. Several studies have shown an association between the presence of extra nodal spread and increased SUV on FDGPET [14]. The aim of this study was to examine if SUV in the primary tumor or in the lymph node metastases is predictive for the development of distant metastases in HNSCC patients who underwent screening on distant metastases because of one or more of the aforementioned high risk factors. Material and methods A retrospective study was performed on consecutive head and neck cancer patients with high risk factors who underwent screening on distant metastases by PET–CT between April 2007 and February 2012 at VU University Medical Center Amsterdam. Initially 118 high risk head and neck cancer patients were included. Nine patients were incorrectly included, since they appeared to have no high risk factors after further locoregional examinations. In 21 patients the SUV could not be determined in the standardized way because the use of different scan protocol or unknown plasma glucose. All together 30 patients were excluded and 88 patients were evaluable (Fig. 1). Of these 88 patients, 69 patients were male and 19 were female. Risk factors were three or more lymph node metastasis (n = 17), bilateral lymph node metastasis (n = 33), lymph node metastasis >6 cm (n = 5), low jugular lymph node metastasis (n = 14), a second synchronous or metachronous primary tumor (n = 21), a locoregional recurrence (n = 21), assessed by physical examination, MRI, CT and biopsy. Some patients had more than one high risk factor. The mean age was 64 (range 36–88). The primary tumor sites were oral cavity (n = 25), oropharynx (n = 27), hypopharynx (n = 14), larynx (n = 22). Six patients had one or more synchronous primary tumors. The pathological primary tumor status were T1 (n = 10), T2 (n = 21), T3 (n = 25), T4 (n = 24), unknown primary
118 included 9 excluded no high risk factor 109 included 20 excluded 10 external revision scans 11 unknown glucose
88 evaluable 74 primary tumor SUV 56 lymph node SUV Fig. 1. Inclusion of patients.
tumor (n = 8). Some patients had synchronous second primary tumors. The pathological lymph node stage were N0 (n = 28), N1 (n = 10), N2a (n = 2), N2b (n = 20), N2c (n = 22), N3 (n = 6). The characteristics of the patients are summarized in Table 1.
FDG-PET–CT FDG-PET/CT scans were performed according the EANM procedure guidelines [20]. Patients were instructed to fast for 6 h, with the exception of water, coffee and tea both without sugar, sweets and milk, before the PET–CT-scan was performed. In the two hours before the scan the patient needed to drink approximately one litre of water or when indicated intravenous 0.9% NaCl was administered. Plasma glucose, weight and height were measured. FDG-PET–CT acquisition was acquired approximately 60 min after
Table 1 Patient’s characteristics. Characteristics
n (%)
Gender Male Female
69 (78.4) 19 (21.6)
Age (years) Mean Minimum Maximum
64 36 88
Pathologic T-status T1 T2 T3 T4 Unknown primary tumor
10 (11.4) 21 (23.9) 25 (28.4) 24 (27.3) 8 (9.1)
Pathologic N-status N0 N1 N2a N2b N2c N3
28 (31.8) 10 (11.4) 2 (2.3) 20 (22.7) 22 (25.0) 6 (6.8)
Site Oral cavity Oropharynx Hypopharynx Larynx Unknown primary tumor
25 (28.4) 27 (30.7) 14 (15.9) 22 (25.0) 8 (9.0)
Inclusion criteria >3 lymph node metastases Bilaterallymph node metastases Lymph node metastases >6 cm Low jugularlymph node metastases Second primary tumor Locoregionalrecurrence
17 (19.3) 33 (37.5) 5 (5.7) 14 (15.9) 21 (23.9) 21 (23.9)
Distant metastases Yes No
24 (27.3) 64 (72.7)
Distant metastases Lung Liver Skin Bone Adrenal
18 (75.0) 6 (25.0) 4 (16.7) 3 (12.5) 1 (4.2)
Treatment modality beforescreening Surgery Radiotherapy Surgery plusradiotherapy Chemotherapy plusradiotherapy
37 (42.0) 5 (13.5) 7 (18.9) 16 (43.2) 9 (24.3)
Follow up 6 months
15 (17.0) 73 (83.0)
538
J. Meeuwis et al. / Oral Oncology 51 (2015) 536–540
injection of 250–370 FDG, depending on body mass index. The FDGPET–CT scans (trajectory skull base to knee, 4 min per bed position) were performed with the Philips Gemini TF PET–CT scanner (Philips Medical Systems, Eindhoven, The Netherlands).Low-dose CT was performed with 120 kV and 50 mAs prior to emission scanning. PET–CT data were reconstructed using a time of flight row-action maximum likelihood algorithm, as implemented by the vendor. Final image matrix size equals 288 288 with a voxel size of 2 2 2 mm. Post-reconstruction image resolution was 5 mm full width at half maximum (FWHM).
Table 2 Summarised SUV score in patients with and without distant metastases. Mean
Median
Minimum
Maximum
Standard deviation
With distant metastasis Primarytumor SUV SUVA50 9.57 9.02 SUVmax 13.55 12.78
3.12 4.15
20.89 29.83
4.79 7.03
Lympnode SUV SUVA50 7.52 SUVmax 10.37
3.05 4.10
16.07 22.70
3.77 5.17
Primarytumor SUV 8.48 7.95 SUVA50 SUVmax 11.94 10.81
3.19 4.23
20.51 28.69
3.77 5.52
Lympnode SUV SUVA50 7.49 SUVmax 10.41
2.62 3.60
13.93 19.70
2.74 3.89
6.52 8.99
No distant metastasis
SUV calculation Volume of interest (VOI) of the primary tumor and most intense lymph nodes were drawn. For each VOI the highest value of one single pixel within the VOI (SUVmax) and the average value of 50% of the maximum pixel in the VOI corrected for the background absorption (SUVA50) were calculated through automated measurement using a VUmc research analysis tool developed with IDL software version 6.3 (Interactive Data Language, Research Systems Inc., Boulder, USA). When multiple lymph nodes were found only the lymph node with the highest SUV was used. All SUVs were corrected for plasma glucose levels and body mass index. The software provides the SUV automatically. Distant metastases Patients were considered to have distant metastases if distant metastases were detected during pretreatment screening (at the time of PET–CT) or follow-up. Mean follow up for all patients was 453 days with a range of 8 days (death due to postoperative complications) to 1353 days as measured from the PET–CT-scan. Distant metastases occurred in 24 patients: 11 pretreatment and 13 during follow-up. The most common site was the lung only (n = 18), liver (n = 6), skin (n = 4), bone (n = 3), and adrenal (n = 1) (Table 1).
7.30 10.40
(n = 51) and not previously treated patients with a follow up >6months (n = 40). As shown in Table 3 no statistically correlation between the different SUVs of the primary tumor and lymph nodes metastases and the development of distant metastases were found. For all patients the AUC for the primary tumor SUVA50 and the SUVmax are respectively 0.598 and 0.582 and 0.458 and 0.463 for the lymph node metastases. For patients with a follow up >6 months the AUC for the primary tumor SUVA50 and the SUVmax are respectively 0.628 and 0.609 and 0.474 and 0.485 for the lymph node metastases. For not previously treated patients the AUC for the primary tumor SUVA50 and the SUVmax are respectively 0.586 and 0.549 and 0.515 and 0.506 for the lymph node metastases. For patients with a follow up >6 months and not previously treated
Table 3 Results of logistic regression to determine the correlation of different SUVs and distant metastases in different (sub) groups. p-value
Statistical analysis Data were entered in Microsoft office Excel 2003. Statistical analyses were performed using SPSS 18.0 (SPSS Inc., Chicago, IL., USA). The logistic regression test was used to determine if there is a correlation between the presence of distant metastases and primary tumor SUVA50, primary tumor SUVmax, lymph node SUVA50 or lymph node SUVmax. After the logistic regression test with the continue variables the ROC analyses were performed to determine the best cut-off point in the statistical analysis. For the ROC analyses the same subgroups were used as in the logistic regression analysis. The area under the ROC curve (AUC) can range from 0.0 and 1.0 with lager indicative of better fit. Results In 74 patients SUV of the primary tumor and in 56 patients SUV of the lymph nodes metastases could be determined. In patients with distant metastases the mean (±SD) SUVA50 and the SUVmax were respectively 9.57 (±4.79) and 13.55 (±7.03) in the primary tumor and 7.52 (±3.77) and 13.55 (±5.17) in the lymph node metastases. In patients without distant metastases the mean (±SD) SUVA50 and the SUVmaxwere respectively 8.48 (±3.77) and 11.94 (±5.52) in the primary tumor and 7.49 (±2.74) and 10.41 (±3.89) in the lymph node metastases (Table 2). The different SUVs were used for logistic regression analysis in different patient groups: all patients (n = 88), patients with a follow up >6 months (n = 73), not previously treated patients
All patients Primary tumor SUVA50 SUVmax
0.309 0.307
Lymph node SUVA50 SUVmax
0.973 0.965
>6 months follow up Primary tumor SUVA50 SUVmax
0.271 0.287
Lymph node SUVA50 SUVmax
0.607 0.603
Not previously treated Primary tumor SUVA50 SUVmax
0.319 0.365
Lymph node SUVA50 SUVmax
0.517 0.638
>6 months follow up and not previously treated Primary tumor SUVA50 SUVmax
0.265 0.286
Lymph node SUVA50 SUVmax
0.331 0.338
J. Meeuwis et al. / Oral Oncology 51 (2015) 536–540
the AUC for the primary tumor SUVA50 and the SUVmax are respectively 0.630 and 0.583 and 0.560 and 0.546 for the lymph node metastases. These AUCs were too small to take advantage of a cut-off point. Therefore, it was not useful to determine the optimal cut-off value (Q point).
Discussion In the present study with high risk HNSCC patients 24 of the 88 (27%) patients were diagnosed with distant metastases during screening and follow up. In this selected group of patients SUV in the primary tumor and in the lymph node metastases were not associated with the development of distant metastases. Several authors reported a correlation between SUV and the clinical outcome [21–26], although not confirmed by others [27] Schwartz et al. [21] found by univariate analysis a significant correlation between SUV of the primary tumor and local recurrencefree and disease free survival. However, they did not examine the correlation with development of distant metastases. Moreover, they have not found a similar correlation for nodal SUV. Brun et al. [22] found low metabolic rate of FDG in the primary tumor was associated with complete remission, locoregional control and survival after therapy. No statistic significant differences between subgroups were found with respect to survival. Minn et al. [23] found an overall survival of patients with a SUV lower than or equal to the median value was clearly better in univariate analysis then that of patients with a higher then median, but in multivariate analysis this did not remain an independent predictor of survival. In multivariate analyses with 73 patients Halfpenny et al. [24] demonstrated that a SUV provides prognostic information independent of tumor stage and diameter. The result from Allal et al. [25] suggest that pretreatment tumor FDG uptake represents an independent prognostic factor in patients with head and neck cancer. Patients with higher SUV are at greater risk of failure and should be considered for more aggressive therapy. Kim et al. [26] identified also pretreatment FDG uptake as an independent prognostic factor. The association of high SUV with poor survival may be related to several factors. Most important are patients with advanced diseases presented with large tumors. FDG uptake may simply reflect to the large tumor burden of these patients [27]. There are several studies which examined the relation between SUV of FDG in primary tumor or lymph node metastases and distant metastases [8,12–18,28]. Suzuki et al. [12] showed that SUVmax was significantly associated with distant metastases-free survival in 24 oral squamous cell carcinoma patients Kubicek et al. [14] found in 152 head and neck cancer patients (all stages) that SUVmax of FDG in lymph node metastasis >10 was predictive for distant failure, but not for local and overall failure. SUVmax in the primary tumor was only predictive for overall survival and not for distant failure. Liao et al. [16] found in patients with oral cancer (all stages) that SUVmax of primary tumor (193 patients) >8.6 and SUVmax of lymph node(s) (57 patients) >5.7 were predictors for the development of distant metastases. Moreover, they developed a scoring system using three prognostic factors: SUVmax of primary tumor, SUVmax of lymph node(s) and the presence of pathological lymph node metastases. Patients with a score between 0 and 3 appeared to define distinct prognostic groups with significant different neck control, development of distant metastases and disease specific survival [16]. In another study (120 patients) Liao et al. [17] found a nodal SUVmax P 5.7 as an independent prognostic factor for head and neck cancer control, development of distant metastases, disease free survival, diseasespecific survival and overall survival in oral squamous cell carcinoma patients with pathologically positive lymph nodes. Chan et al. [13] examined the value of FDG-PET–CT and whole body MRI in
539
103 untreated oropharyngeal and hypopharyngeal carcinoma patients (all stages) and found that SUVmax of the regional lymph nodes above the median value of 8.7 g/ml was significantly associated with the occurrence of distant metastases. Yao et al. [15] examined the association of SUV in 203 HNSCC patients who underwent FDG-PET and found that SUVmax P 10.8 of lymph node(s) was associated with distant metastases. Inokuchi et al. [28] reported an association of SUVmax of lymph node(s) and distant metastases if 6.0 was used as cut-off point in 178 HNSCC patients with nodal metastases. On the contrary, Haerle et al. [8] examined the value of FDGPET–CT for the detection of high risk patients with head and neck squamous cell carcinomas and found no association between maximal SUV > 9 of the primary tumor and development of distant metastases. The present study also included only head and neck cancer patients with high risk factors for development of distant metastases and did not found an association between SUV of primary tumor and lymph nodes on FGD-PET and distant metastases. From the latter and the present study it can be concluded that SUV of primary tumor and lymph node on FDG-PET is not predictive for distant metastases in a (selected) group of patients with already high risk factors for distant metastases. In groups of patients who are not selected based on high risk factors for distant metastases, SUV of primary tumor and lymph node on FDG-PET have been shown to be associated with distant metastases. Therefore, it can be anticipated that SUV values are predictors for distant metastases, but are not independent on the clinical high risk factors previously used. SUV does not add to the risk of development of distant metastases in patients with clinically high risk factors. In many institutes FDG-PET–CT is not routinely used in all HNSCC patients. We use FDG-PET routinely for screening on distant metastases in high risk patients (with aforementioned high risk factors), lymph node metastases of unknown primary tumor (mostly with large lymph node metastases), response evaluation after chemoradiation and suspicion of recurrent tumor after radiotherapy. As a consequence almost all patients who undergo FDGPET–CT have high risk factors for distant metastases and we cannot retrospectively examine the predictive value of SUV in an unselected group of HNSCC patients. Moreover, in clinical practice SUV is difficult to use as an indication for screening on distant metastases. To assess an eventual indication for screening on distant metastases based on SUVFDG-PET, the currently best diagnostic technique for that [11,29], has yet to be made. In these aforementioned studies different cut-off values have been used to predict the development of distant metastases. Since different methods to assess the SUV (type, software, normalization (for BMI), glucose correction) are used, studies may be difficult to compare [30]. Also the reference standard varies in duration of follow-up between the different studies. Moreover, even within studies SUVs may not be compared reliable since scanners, uptake period of FDG (range 40–90 min) and scanning protocols (scanning traject and time per bed position and reconstruction) may vary within these studies. Previous treatment may have impact on vascularization and thus uptake of FDG. The SUVs determined after previous radiotherapy may not reflect the initial (original) metabolic activity of the tumor. Therefore, we made a subgroup analysis of patients without previous treatment in the head and neck area. Also in this subgroup no predictive value of SUV was found. SUVs were correlated with development of distant metastases. Although in most patients distant metastases are detected during pretreatment screening or the first 6 months follow up, they may become manifest after a longer period of time. In patients with locoregional control the development of distant metastases during follow-up must be due to distant metastases present but too small
540
J. Meeuwis et al. / Oral Oncology 51 (2015) 536–540
to detect during screening. Therefore, the reference standard will improve over time. However, to remain a sufficient number for statistical analysis (sub) group analysis was performed for all patients, for patients with a follow-up of at least 6 months, not previously treated patient and not previously treated patients with a followup of at least 6 months. In none of these subgroups an association of SUV and development of distant metastases was found. Conclusion In this retrospective study no statistically significant correlation between the primary tumor SUV or lymph node SUV and the development of distant metastases was found in head and neck cancer patients with high risk factors for distant metastases. This does not mean that SUV values cannot be a predictor for distant metastases but that it is at least not an independent prognostic factor in patients with previously assessed high risk factors. Therefore, we recommend to screen for distant metastases in patients with previously identified high risk factors and not to select patients based on SUV. Conflict of interest statement None declared. References [1] Takes RP, Rinaldo A, Silver CE, Haigentz Jr M, Woolgar JA, Triantafyllou A, et al. Distant metastases from head and neck squamous cell carcinoma. Part I. Basic aspects. Oral Oncol 2012;48(9):775–9. [2] Haigentz Jr M, Hartl DM, Silver CE, Langendijk JA, Strojan P, Paleri V, et al. Distant metastases from head and neck squamous cell carcinoma. Part III. Treatment. Oral Oncol 2012;48(9):787–93. [3] deBree R, Haigentz Jr M, Silver CE, Paccagnella D, Hamoir M, Hartl DM, et al. Distant metastases from head and neck squamous cell carcinoma. Part II. Diagnosis. Oral Oncol 2012;48(9):780–6. [4] Kuperman DI, Auethavekiat V, Adkins DR, Nussenbaum B, Collins S, Boonchalermvichian C, et al. Squamous cell cancer of the head and neck with distant metastasis at presentation. Head Neck 2011;33(5):714–8. [5] Jäckel MC, Rausch H. Fernmetastasierung von Plattenepithelkarzinom des oberenAerodigestivtrakts. Der Einfluss klinischer Tumorparameter und des Krankheitsverlauf. HNO 1999;47(1):38–44. [6] Loh KS, Brown DH, Baker JT, Gilbert RW, Gullane PJ, Irish JC. A rational approach to pulmonary screening in newly diagnosed head and neck cancer. Head Neck 2005;27(11):990–4. [7] Leong SC, Javed F, Elliot S, Mortimore S. Effectiveness of X-ray and computed tomography screening for assessing pulmonary involvement in patients with head and neck squamous cell carcinoma. J Laryngol Otol 2008;122(9):961–6. [8] Haerle SK, Schmid DT, Ahmad N, Hany TF, Stoeckli SJ. The value of (18)F-FDG PET/CT for the detection of distant metastases in high-risk patients with head and neck squamous cell carcinoma. Oral Oncol 2011;47(7):653–9. [9] de Bree R, Deurloo EE, Snow GB, Leemans CR. Screening for distant metastases in patients with head and neck cancer. Laryngoscope 2000;110(3):397–400. [10] Brouwer J, de Bree R, Hoekstra OS, Golding RP, Langendijk JA, Castelijns JA, et al. Screening for distant metastases in patients with head and neck cancer: is chest CT sufficient? Laryngoscope 2005;115(10):1808–12. [11] Senft A, de Bree R, Hoekstra OS, Kuik DJ, Golding RP, Oyen WJ, et al. Screening for distant metastases in head and neck cancer patients by chest CT or whole body FDG-PET: a prospective multitcenter trial study. Radiother Oncol 2008;87:221–9.
[12] Suzuki H, Hasegawa Y, Terada A, Hyodo I, Nakashima T, Nishio M, et al. FDGPET predicts survival and distant metastasis in oral squamous cell carcinoma. Oral Oncol 2009;45(7):569–73. [13] Chan SC, Wang HM, Yen TC, Lin CY, Chin SC, Liao CT, et al. 18F-FDG PET/CT and 3.0-T whole-body MRI for the detection of distantmetastasesand second primarytumours in patientswithuntreatedoropharyngeal/hypopharyngeal carcinoma: a comparativestudy. Eur J Nucl Med Mol Imaging 2011;38(9):1607–19. [14] Kubicek GJ, Champ C, Fogh S, Wang F, Reddy E, Intenzo C, et al. FDG-PET staging and importance of lymph node SUV in head and neck cancer. Head Neck Oncol 2010;2:19. [15] Yao M, Lu M, Savvides PS, Rezaee R, Zender CA, Lavertu P, et al. Distant metastases in head-and-neck squamous cell carcinoma treated with intensitymodulated radiotherapy. Int J Radiat Oncol Biol Phys 2012;83(2):684–9. [16] Liao CT, Chang JT, Wang HM, Ng SH, Hsueh C, Lee LY, et al. Preoperative [18F] fluorodeoxyglucose positron emission tomography standardized uptake value of neck lymph nodes predicts neck cancer control and survival rates in patients with oral cavity squamous cell carcinoma and pathologically positive lymph nodes. Int J Radiat Oncol Biol Phys 2009;74(4):1054–61. [17] Liao CT, Wang HM, Chang JT, Lin CY, Ng SH, Huang SF, et al. Influence of pathological nodal status and maximal standardized uptake value of the primary tumor and regional lymph nodes on treatment plans in patients with advanced oral cavity squamous cell carcinoma. Int J Radiat Oncol Biol Phys 2010;77(2):421–9. [18] Coca-Pelaz A, Rodrigo JP, Suárez C. Clinicopathologic analysis and predictive factors for distant metastases in patients with head and neck squamous cell carcinomas. Head Neck 2012;34(6):771–5. [19] Ljumanovic R, Langendijk JA, Hoekstra OS, Leemans CR, Castelijns JA. Distant metastases in head and neck carcinoma: identification of prognostic groups with MR imaging. Eur J Radiol 2006;60(1):58–66. [20] Boellaard R, O’Doherty MJ, Weber WA, Mottaghy FM, Lonsdale MN, Stroobants SG, et al. FDG PET and PET/CT: EANM procedure guidelines for tumour PET imaging: version 1.0. Eur J Nucl Med Mol Imaging 2010;37(1):181–200. [21] Schwartz DL, Rajendran J, Yueh B, Coltrera MD, Leblanc M, Eary J, et al. FDGPET prediction of head and neck squamous cell cancer outcomes. Arch Otolaryngol Head Neck Surg 2004;130(12):1361–7. [22] Brun E, Kjellén E, Tennvall J, Ohlsson T, Sandell A, Perfekt R, et al. FDG PET studies during treatment: prediction of therapy outcome in head and neck squamous cell carcinoma. Head Neck 2002;24(2):127–35. [23] Minn H, Lapela M, Klemi PJ, Grénman R, Leskinen S, Lindholm P, et al. Prediction of survival with fluorine-18-fluoro-deoxyglucose and PET in head and neck cancer. J Nucl Med 1997;38:1907–11. [24] Halfpenny W, Hain SF, Biassoni L, Maisey MN, Sherman JA, McGurk M. FDGPET. A possible prognostic factor in head and neck cancer. Br J Cancer 2002;86(4):512–6. [25] Allal AS, Slosman DO, Kebdani T, Allaoua M, Lehmann W, Dulguerov P. Prediction of outcome in head-and-neck cancer patients using the standardized uptake value of 2-[18F]fluoro-2-deoxy-D-glucose. Int J Radiat Oncol Biol Phys 2004;59(5):1295–300. [26] Kim SY, Roh JL, Kim MR, Kim JS, Choi SH, Nam SY, et al. Use of 18F-FDG PET for primary treatment strategy in patients with squamous cell carcinoma of the oropharynx. J Nucl Med 2007;48(5):752–7. [27] Dibble EH, Alvarez AC, Truong MT, Mercier G, Cook EF, Subramaniam RM. 18FFDG metabolic tumor volume and total glycolytic activity of oral cavity and oropharyngeal squamous cell cancer: adding value to clinical staging. J Nucl Med 2012;53(5):709–15. [28] Inokuchi H, Kodaira T, Tachibana H, Nakamura T, Tomita N, Nakahara R, et al. Clinical usefulness of [18F] fluoro-2-deoxy-D-glucose uptake in 178 head-andneck cancer patients with nodal metastasis treated with definitive chemoradiotherapy: consideration of its prognostic value and ability to provide guidance for optimal selection of patients for planned neck dissection. Int J Radiat Oncol Biol Phys 2011;79(3):747–55. [29] Ng SH, Chan SC, Liao CT, Chang JT, Ko SF, Wang HM, et al. Distant metastases and synchronous second primary tumors in patients with newly diagnosed oropharyngeal and hypopharyngeal carcinomas: evaluation of (18)F-FDG PET and extended-field multi-detector row CT. Neuroradiology 2008;50(11): 969–79. [30] Boellaard R. Need for standardization of 18F-FDG PET/CT for treatment response assessments. J Nucl Med 2011;52(Suppl 2):93S–100S.
Contents lists available at ScienceDirect
Oral Oncology journal homepage: www.elsevier.com/locate/oraloncology
18FDG SUV in the primary tumor and lymph node metastases is not predictive for development of distant metastases in high risk head and neck cancer patients Jasmijn Meeuwis a, Otto S. Hoekstra b, Birgit I. Witte c, Ronald Boellaard b, C. René Leemans a, Remco de Bree a,d,⇑ a
Department of Otolaryngology-Head and Neck Surgery, VU University Medical Center, Amsterdam, The Netherlands Department of Radiology & Nuclear Medicine, VU University Medical Center, Amsterdam, The Netherlands c Department of Epidemiology and Biostatistics, VU University Medical Center, Amsterdam, The Netherlands d Department of Head and Neck Surgery, UMCU Utrecht Cancer Center, Utrecht, The Netherlands b
a r t i c l e
i n f o
Article history: Received 28 October 2014 Received in revised form 7 February 2015 Accepted 10 February 2015 Available online 27 February 2015 Keywords: Head and neck squamous cell carcinoma Distant metastases 18F-fluorodeoxyglucose Standardized uptake value
s u m m a r y Background: Pretreatment screening on distant metastases is particularly useful in head and neck squamous cell carcinoma (HNSCC) patients with high risk factors. Methods: In a retrospective study of 88 patients with previously identified clinical high risk factors the predictive value of standardized uptake value (SUV) of 18F-fluorodeoxyglucose (FDG) in the primary tumor and in the lymph node metastases for the development of distant metastases was examined. Different SUVs corrected for plasma glucose levels and body mass index were calculated and analysed in different patient groups: all patients (n = 88), patients with a follow up >6 months (n = 73), not previously treated patients (n = 51) and not previously treated patients with a follow up >6 months (n = 40). Results: Twenty-four of the 88 (27%) high risk HNSCC patients were diagnosed with distant metastases during screening and follow up. No correlation was found between different SUVs of the primary tumor and lymph nodes metastases and the development of distant metastases. Conclusion: SUVs of primary tumor and lymph node on FDG-PET are not predictive for distant metastases in a (selected) group of patients with already high risk factors for distant metastases. Ó 2015 Elsevier Ltd. All rights reserved.
Introduction The detection of distant metastases at the time of initial evaluation influences the prognosis and selection of treatment in head and neck squamous cell carcinoma (HNSCC) patients [1]. Patients with distant metastases are generally not considered curable and almost always receive only palliative treatment [2]. Detection of distant metastases may avoid futile extensive locoregional treatments with unnecessarily burden to the patient and use and costs of health resources. The incidence of distant metastasis from HNSCC at presentation is generally too low to warrant an extensive radiographic staging evaluation [3]. In the United States Surveillance Epidemiology
⇑ Corresponding author at: Department of Otolaryngology/Head and Neck Surgery, Vrije Universiteit Medical Center, De Boelelaan 1117, 1081 HV Amsterdam, PO Box 7057, 1007 MB Amsterdam, The Netherlands. Tel.: +31 20 4443689; fax: +31 20 4443688. E-mail address: [email protected] (R. de Bree). http://dx.doi.org/10.1016/j.oraloncology.2015.02.002 1368-8375/Ó 2015 Elsevier Ltd. All rights reserved.
and End Results (SEER) database distant metastases were reported in 2066 out of 73,247 (2.82%) HNSCC patients [4]. This low number of distant metastases in unselected HNSCC patients emphasizes the need for selection based on risk factors. Jäckel and Rausch [5] found that screening is particularly useful in patients with advanced stage disease, local and/or regional recurrences and second primary tumors. Loh et al. [6] evaluated screening in HNSCC patients using chest CT and found T4 and/or N2 or N3 to be risk factors for the development of distant metastases. Leong et al. [7] found in their series of 102 patients who underwent screening by chest X-ray and CT that of the patients with positive screening, 86% had T3 or T4 disease and 71% had N2 or N3 disease. Haerle et al. [8] detected in 21% of patients with advanced stage disease (T3/4 and/or N2/3) distant metastases by PET–CT and follow-up. In previous studies we identified and validated clinically three or more lymph node metastases, bilateral lymph node metastases, lymph node metastases of 6 cm or larger, low jugular lymph node metastases, locoregional recurrence and second primary tumors as
537
J. Meeuwis et al. / Oral Oncology 51 (2015) 536–540
relevant high risk factors. Using these selection criteria, distant metastases were detected in 29–33% of the patients during initial screening or within 12 months follow-up after initial screening [9–11]. Several studies in HNSCC patients have found a correlation between standardized uptake value (SUV) of 18F-fluorodeoxyglucose (FDG) in primary tumor or lymph node metastases and the development of distant metastases [12–17]. However, Haerle et al. [8] could not find a correlation between SUV max of the primary tumor and the development of distant metastases. Several other factors thought to be associated with an increased risk for distant metastases have been reported in the literature [18]. Histopathologically determined risk factors obtained from the surgical specimen are generally not available for treatment planning. Pathologically and radiologically assessed extranodal spread has been associated with development of distant metastases [19]. Several studies have shown an association between the presence of extra nodal spread and increased SUV on FDGPET [14]. The aim of this study was to examine if SUV in the primary tumor or in the lymph node metastases is predictive for the development of distant metastases in HNSCC patients who underwent screening on distant metastases because of one or more of the aforementioned high risk factors. Material and methods A retrospective study was performed on consecutive head and neck cancer patients with high risk factors who underwent screening on distant metastases by PET–CT between April 2007 and February 2012 at VU University Medical Center Amsterdam. Initially 118 high risk head and neck cancer patients were included. Nine patients were incorrectly included, since they appeared to have no high risk factors after further locoregional examinations. In 21 patients the SUV could not be determined in the standardized way because the use of different scan protocol or unknown plasma glucose. All together 30 patients were excluded and 88 patients were evaluable (Fig. 1). Of these 88 patients, 69 patients were male and 19 were female. Risk factors were three or more lymph node metastasis (n = 17), bilateral lymph node metastasis (n = 33), lymph node metastasis >6 cm (n = 5), low jugular lymph node metastasis (n = 14), a second synchronous or metachronous primary tumor (n = 21), a locoregional recurrence (n = 21), assessed by physical examination, MRI, CT and biopsy. Some patients had more than one high risk factor. The mean age was 64 (range 36–88). The primary tumor sites were oral cavity (n = 25), oropharynx (n = 27), hypopharynx (n = 14), larynx (n = 22). Six patients had one or more synchronous primary tumors. The pathological primary tumor status were T1 (n = 10), T2 (n = 21), T3 (n = 25), T4 (n = 24), unknown primary
118 included 9 excluded no high risk factor 109 included 20 excluded 10 external revision scans 11 unknown glucose
88 evaluable 74 primary tumor SUV 56 lymph node SUV Fig. 1. Inclusion of patients.
tumor (n = 8). Some patients had synchronous second primary tumors. The pathological lymph node stage were N0 (n = 28), N1 (n = 10), N2a (n = 2), N2b (n = 20), N2c (n = 22), N3 (n = 6). The characteristics of the patients are summarized in Table 1.
FDG-PET–CT FDG-PET/CT scans were performed according the EANM procedure guidelines [20]. Patients were instructed to fast for 6 h, with the exception of water, coffee and tea both without sugar, sweets and milk, before the PET–CT-scan was performed. In the two hours before the scan the patient needed to drink approximately one litre of water or when indicated intravenous 0.9% NaCl was administered. Plasma glucose, weight and height were measured. FDG-PET–CT acquisition was acquired approximately 60 min after
Table 1 Patient’s characteristics. Characteristics
n (%)
Gender Male Female
69 (78.4) 19 (21.6)
Age (years) Mean Minimum Maximum
64 36 88
Pathologic T-status T1 T2 T3 T4 Unknown primary tumor
10 (11.4) 21 (23.9) 25 (28.4) 24 (27.3) 8 (9.1)
Pathologic N-status N0 N1 N2a N2b N2c N3
28 (31.8) 10 (11.4) 2 (2.3) 20 (22.7) 22 (25.0) 6 (6.8)
Site Oral cavity Oropharynx Hypopharynx Larynx Unknown primary tumor
25 (28.4) 27 (30.7) 14 (15.9) 22 (25.0) 8 (9.0)
Inclusion criteria >3 lymph node metastases Bilaterallymph node metastases Lymph node metastases >6 cm Low jugularlymph node metastases Second primary tumor Locoregionalrecurrence
17 (19.3) 33 (37.5) 5 (5.7) 14 (15.9) 21 (23.9) 21 (23.9)
Distant metastases Yes No
24 (27.3) 64 (72.7)
Distant metastases Lung Liver Skin Bone Adrenal
18 (75.0) 6 (25.0) 4 (16.7) 3 (12.5) 1 (4.2)
Treatment modality beforescreening Surgery Radiotherapy Surgery plusradiotherapy Chemotherapy plusradiotherapy
37 (42.0) 5 (13.5) 7 (18.9) 16 (43.2) 9 (24.3)
Follow up 6 months
15 (17.0) 73 (83.0)
538
J. Meeuwis et al. / Oral Oncology 51 (2015) 536–540
injection of 250–370 FDG, depending on body mass index. The FDGPET–CT scans (trajectory skull base to knee, 4 min per bed position) were performed with the Philips Gemini TF PET–CT scanner (Philips Medical Systems, Eindhoven, The Netherlands).Low-dose CT was performed with 120 kV and 50 mAs prior to emission scanning. PET–CT data were reconstructed using a time of flight row-action maximum likelihood algorithm, as implemented by the vendor. Final image matrix size equals 288 288 with a voxel size of 2 2 2 mm. Post-reconstruction image resolution was 5 mm full width at half maximum (FWHM).
Table 2 Summarised SUV score in patients with and without distant metastases. Mean
Median
Minimum
Maximum
Standard deviation
With distant metastasis Primarytumor SUV SUVA50 9.57 9.02 SUVmax 13.55 12.78
3.12 4.15
20.89 29.83
4.79 7.03
Lympnode SUV SUVA50 7.52 SUVmax 10.37
3.05 4.10
16.07 22.70
3.77 5.17
Primarytumor SUV 8.48 7.95 SUVA50 SUVmax 11.94 10.81
3.19 4.23
20.51 28.69
3.77 5.52
Lympnode SUV SUVA50 7.49 SUVmax 10.41
2.62 3.60
13.93 19.70
2.74 3.89
6.52 8.99
No distant metastasis
SUV calculation Volume of interest (VOI) of the primary tumor and most intense lymph nodes were drawn. For each VOI the highest value of one single pixel within the VOI (SUVmax) and the average value of 50% of the maximum pixel in the VOI corrected for the background absorption (SUVA50) were calculated through automated measurement using a VUmc research analysis tool developed with IDL software version 6.3 (Interactive Data Language, Research Systems Inc., Boulder, USA). When multiple lymph nodes were found only the lymph node with the highest SUV was used. All SUVs were corrected for plasma glucose levels and body mass index. The software provides the SUV automatically. Distant metastases Patients were considered to have distant metastases if distant metastases were detected during pretreatment screening (at the time of PET–CT) or follow-up. Mean follow up for all patients was 453 days with a range of 8 days (death due to postoperative complications) to 1353 days as measured from the PET–CT-scan. Distant metastases occurred in 24 patients: 11 pretreatment and 13 during follow-up. The most common site was the lung only (n = 18), liver (n = 6), skin (n = 4), bone (n = 3), and adrenal (n = 1) (Table 1).
7.30 10.40
(n = 51) and not previously treated patients with a follow up >6months (n = 40). As shown in Table 3 no statistically correlation between the different SUVs of the primary tumor and lymph nodes metastases and the development of distant metastases were found. For all patients the AUC for the primary tumor SUVA50 and the SUVmax are respectively 0.598 and 0.582 and 0.458 and 0.463 for the lymph node metastases. For patients with a follow up >6 months the AUC for the primary tumor SUVA50 and the SUVmax are respectively 0.628 and 0.609 and 0.474 and 0.485 for the lymph node metastases. For not previously treated patients the AUC for the primary tumor SUVA50 and the SUVmax are respectively 0.586 and 0.549 and 0.515 and 0.506 for the lymph node metastases. For patients with a follow up >6 months and not previously treated
Table 3 Results of logistic regression to determine the correlation of different SUVs and distant metastases in different (sub) groups. p-value
Statistical analysis Data were entered in Microsoft office Excel 2003. Statistical analyses were performed using SPSS 18.0 (SPSS Inc., Chicago, IL., USA). The logistic regression test was used to determine if there is a correlation between the presence of distant metastases and primary tumor SUVA50, primary tumor SUVmax, lymph node SUVA50 or lymph node SUVmax. After the logistic regression test with the continue variables the ROC analyses were performed to determine the best cut-off point in the statistical analysis. For the ROC analyses the same subgroups were used as in the logistic regression analysis. The area under the ROC curve (AUC) can range from 0.0 and 1.0 with lager indicative of better fit. Results In 74 patients SUV of the primary tumor and in 56 patients SUV of the lymph nodes metastases could be determined. In patients with distant metastases the mean (±SD) SUVA50 and the SUVmax were respectively 9.57 (±4.79) and 13.55 (±7.03) in the primary tumor and 7.52 (±3.77) and 13.55 (±5.17) in the lymph node metastases. In patients without distant metastases the mean (±SD) SUVA50 and the SUVmaxwere respectively 8.48 (±3.77) and 11.94 (±5.52) in the primary tumor and 7.49 (±2.74) and 10.41 (±3.89) in the lymph node metastases (Table 2). The different SUVs were used for logistic regression analysis in different patient groups: all patients (n = 88), patients with a follow up >6 months (n = 73), not previously treated patients
All patients Primary tumor SUVA50 SUVmax
0.309 0.307
Lymph node SUVA50 SUVmax
0.973 0.965
>6 months follow up Primary tumor SUVA50 SUVmax
0.271 0.287
Lymph node SUVA50 SUVmax
0.607 0.603
Not previously treated Primary tumor SUVA50 SUVmax
0.319 0.365
Lymph node SUVA50 SUVmax
0.517 0.638
>6 months follow up and not previously treated Primary tumor SUVA50 SUVmax
0.265 0.286
Lymph node SUVA50 SUVmax
0.331 0.338
J. Meeuwis et al. / Oral Oncology 51 (2015) 536–540
the AUC for the primary tumor SUVA50 and the SUVmax are respectively 0.630 and 0.583 and 0.560 and 0.546 for the lymph node metastases. These AUCs were too small to take advantage of a cut-off point. Therefore, it was not useful to determine the optimal cut-off value (Q point).
Discussion In the present study with high risk HNSCC patients 24 of the 88 (27%) patients were diagnosed with distant metastases during screening and follow up. In this selected group of patients SUV in the primary tumor and in the lymph node metastases were not associated with the development of distant metastases. Several authors reported a correlation between SUV and the clinical outcome [21–26], although not confirmed by others [27] Schwartz et al. [21] found by univariate analysis a significant correlation between SUV of the primary tumor and local recurrencefree and disease free survival. However, they did not examine the correlation with development of distant metastases. Moreover, they have not found a similar correlation for nodal SUV. Brun et al. [22] found low metabolic rate of FDG in the primary tumor was associated with complete remission, locoregional control and survival after therapy. No statistic significant differences between subgroups were found with respect to survival. Minn et al. [23] found an overall survival of patients with a SUV lower than or equal to the median value was clearly better in univariate analysis then that of patients with a higher then median, but in multivariate analysis this did not remain an independent predictor of survival. In multivariate analyses with 73 patients Halfpenny et al. [24] demonstrated that a SUV provides prognostic information independent of tumor stage and diameter. The result from Allal et al. [25] suggest that pretreatment tumor FDG uptake represents an independent prognostic factor in patients with head and neck cancer. Patients with higher SUV are at greater risk of failure and should be considered for more aggressive therapy. Kim et al. [26] identified also pretreatment FDG uptake as an independent prognostic factor. The association of high SUV with poor survival may be related to several factors. Most important are patients with advanced diseases presented with large tumors. FDG uptake may simply reflect to the large tumor burden of these patients [27]. There are several studies which examined the relation between SUV of FDG in primary tumor or lymph node metastases and distant metastases [8,12–18,28]. Suzuki et al. [12] showed that SUVmax was significantly associated with distant metastases-free survival in 24 oral squamous cell carcinoma patients Kubicek et al. [14] found in 152 head and neck cancer patients (all stages) that SUVmax of FDG in lymph node metastasis >10 was predictive for distant failure, but not for local and overall failure. SUVmax in the primary tumor was only predictive for overall survival and not for distant failure. Liao et al. [16] found in patients with oral cancer (all stages) that SUVmax of primary tumor (193 patients) >8.6 and SUVmax of lymph node(s) (57 patients) >5.7 were predictors for the development of distant metastases. Moreover, they developed a scoring system using three prognostic factors: SUVmax of primary tumor, SUVmax of lymph node(s) and the presence of pathological lymph node metastases. Patients with a score between 0 and 3 appeared to define distinct prognostic groups with significant different neck control, development of distant metastases and disease specific survival [16]. In another study (120 patients) Liao et al. [17] found a nodal SUVmax P 5.7 as an independent prognostic factor for head and neck cancer control, development of distant metastases, disease free survival, diseasespecific survival and overall survival in oral squamous cell carcinoma patients with pathologically positive lymph nodes. Chan et al. [13] examined the value of FDG-PET–CT and whole body MRI in
539
103 untreated oropharyngeal and hypopharyngeal carcinoma patients (all stages) and found that SUVmax of the regional lymph nodes above the median value of 8.7 g/ml was significantly associated with the occurrence of distant metastases. Yao et al. [15] examined the association of SUV in 203 HNSCC patients who underwent FDG-PET and found that SUVmax P 10.8 of lymph node(s) was associated with distant metastases. Inokuchi et al. [28] reported an association of SUVmax of lymph node(s) and distant metastases if 6.0 was used as cut-off point in 178 HNSCC patients with nodal metastases. On the contrary, Haerle et al. [8] examined the value of FDGPET–CT for the detection of high risk patients with head and neck squamous cell carcinomas and found no association between maximal SUV > 9 of the primary tumor and development of distant metastases. The present study also included only head and neck cancer patients with high risk factors for development of distant metastases and did not found an association between SUV of primary tumor and lymph nodes on FGD-PET and distant metastases. From the latter and the present study it can be concluded that SUV of primary tumor and lymph node on FDG-PET is not predictive for distant metastases in a (selected) group of patients with already high risk factors for distant metastases. In groups of patients who are not selected based on high risk factors for distant metastases, SUV of primary tumor and lymph node on FDG-PET have been shown to be associated with distant metastases. Therefore, it can be anticipated that SUV values are predictors for distant metastases, but are not independent on the clinical high risk factors previously used. SUV does not add to the risk of development of distant metastases in patients with clinically high risk factors. In many institutes FDG-PET–CT is not routinely used in all HNSCC patients. We use FDG-PET routinely for screening on distant metastases in high risk patients (with aforementioned high risk factors), lymph node metastases of unknown primary tumor (mostly with large lymph node metastases), response evaluation after chemoradiation and suspicion of recurrent tumor after radiotherapy. As a consequence almost all patients who undergo FDGPET–CT have high risk factors for distant metastases and we cannot retrospectively examine the predictive value of SUV in an unselected group of HNSCC patients. Moreover, in clinical practice SUV is difficult to use as an indication for screening on distant metastases. To assess an eventual indication for screening on distant metastases based on SUVFDG-PET, the currently best diagnostic technique for that [11,29], has yet to be made. In these aforementioned studies different cut-off values have been used to predict the development of distant metastases. Since different methods to assess the SUV (type, software, normalization (for BMI), glucose correction) are used, studies may be difficult to compare [30]. Also the reference standard varies in duration of follow-up between the different studies. Moreover, even within studies SUVs may not be compared reliable since scanners, uptake period of FDG (range 40–90 min) and scanning protocols (scanning traject and time per bed position and reconstruction) may vary within these studies. Previous treatment may have impact on vascularization and thus uptake of FDG. The SUVs determined after previous radiotherapy may not reflect the initial (original) metabolic activity of the tumor. Therefore, we made a subgroup analysis of patients without previous treatment in the head and neck area. Also in this subgroup no predictive value of SUV was found. SUVs were correlated with development of distant metastases. Although in most patients distant metastases are detected during pretreatment screening or the first 6 months follow up, they may become manifest after a longer period of time. In patients with locoregional control the development of distant metastases during follow-up must be due to distant metastases present but too small
540
J. Meeuwis et al. / Oral Oncology 51 (2015) 536–540
to detect during screening. Therefore, the reference standard will improve over time. However, to remain a sufficient number for statistical analysis (sub) group analysis was performed for all patients, for patients with a follow-up of at least 6 months, not previously treated patient and not previously treated patients with a followup of at least 6 months. In none of these subgroups an association of SUV and development of distant metastases was found. Conclusion In this retrospective study no statistically significant correlation between the primary tumor SUV or lymph node SUV and the development of distant metastases was found in head and neck cancer patients with high risk factors for distant metastases. This does not mean that SUV values cannot be a predictor for distant metastases but that it is at least not an independent prognostic factor in patients with previously assessed high risk factors. Therefore, we recommend to screen for distant metastases in patients with previously identified high risk factors and not to select patients based on SUV. Conflict of interest statement None declared. References [1] Takes RP, Rinaldo A, Silver CE, Haigentz Jr M, Woolgar JA, Triantafyllou A, et al. Distant metastases from head and neck squamous cell carcinoma. Part I. Basic aspects. Oral Oncol 2012;48(9):775–9. [2] Haigentz Jr M, Hartl DM, Silver CE, Langendijk JA, Strojan P, Paleri V, et al. Distant metastases from head and neck squamous cell carcinoma. Part III. Treatment. Oral Oncol 2012;48(9):787–93. [3] deBree R, Haigentz Jr M, Silver CE, Paccagnella D, Hamoir M, Hartl DM, et al. Distant metastases from head and neck squamous cell carcinoma. Part II. Diagnosis. Oral Oncol 2012;48(9):780–6. [4] Kuperman DI, Auethavekiat V, Adkins DR, Nussenbaum B, Collins S, Boonchalermvichian C, et al. Squamous cell cancer of the head and neck with distant metastasis at presentation. Head Neck 2011;33(5):714–8. [5] Jäckel MC, Rausch H. Fernmetastasierung von Plattenepithelkarzinom des oberenAerodigestivtrakts. Der Einfluss klinischer Tumorparameter und des Krankheitsverlauf. HNO 1999;47(1):38–44. [6] Loh KS, Brown DH, Baker JT, Gilbert RW, Gullane PJ, Irish JC. A rational approach to pulmonary screening in newly diagnosed head and neck cancer. Head Neck 2005;27(11):990–4. [7] Leong SC, Javed F, Elliot S, Mortimore S. Effectiveness of X-ray and computed tomography screening for assessing pulmonary involvement in patients with head and neck squamous cell carcinoma. J Laryngol Otol 2008;122(9):961–6. [8] Haerle SK, Schmid DT, Ahmad N, Hany TF, Stoeckli SJ. The value of (18)F-FDG PET/CT for the detection of distant metastases in high-risk patients with head and neck squamous cell carcinoma. Oral Oncol 2011;47(7):653–9. [9] de Bree R, Deurloo EE, Snow GB, Leemans CR. Screening for distant metastases in patients with head and neck cancer. Laryngoscope 2000;110(3):397–400. [10] Brouwer J, de Bree R, Hoekstra OS, Golding RP, Langendijk JA, Castelijns JA, et al. Screening for distant metastases in patients with head and neck cancer: is chest CT sufficient? Laryngoscope 2005;115(10):1808–12. [11] Senft A, de Bree R, Hoekstra OS, Kuik DJ, Golding RP, Oyen WJ, et al. Screening for distant metastases in head and neck cancer patients by chest CT or whole body FDG-PET: a prospective multitcenter trial study. Radiother Oncol 2008;87:221–9.
[12] Suzuki H, Hasegawa Y, Terada A, Hyodo I, Nakashima T, Nishio M, et al. FDGPET predicts survival and distant metastasis in oral squamous cell carcinoma. Oral Oncol 2009;45(7):569–73. [13] Chan SC, Wang HM, Yen TC, Lin CY, Chin SC, Liao CT, et al. 18F-FDG PET/CT and 3.0-T whole-body MRI for the detection of distantmetastasesand second primarytumours in patientswithuntreatedoropharyngeal/hypopharyngeal carcinoma: a comparativestudy. Eur J Nucl Med Mol Imaging 2011;38(9):1607–19. [14] Kubicek GJ, Champ C, Fogh S, Wang F, Reddy E, Intenzo C, et al. FDG-PET staging and importance of lymph node SUV in head and neck cancer. Head Neck Oncol 2010;2:19. [15] Yao M, Lu M, Savvides PS, Rezaee R, Zender CA, Lavertu P, et al. Distant metastases in head-and-neck squamous cell carcinoma treated with intensitymodulated radiotherapy. Int J Radiat Oncol Biol Phys 2012;83(2):684–9. [16] Liao CT, Chang JT, Wang HM, Ng SH, Hsueh C, Lee LY, et al. Preoperative [18F] fluorodeoxyglucose positron emission tomography standardized uptake value of neck lymph nodes predicts neck cancer control and survival rates in patients with oral cavity squamous cell carcinoma and pathologically positive lymph nodes. Int J Radiat Oncol Biol Phys 2009;74(4):1054–61. [17] Liao CT, Wang HM, Chang JT, Lin CY, Ng SH, Huang SF, et al. Influence of pathological nodal status and maximal standardized uptake value of the primary tumor and regional lymph nodes on treatment plans in patients with advanced oral cavity squamous cell carcinoma. Int J Radiat Oncol Biol Phys 2010;77(2):421–9. [18] Coca-Pelaz A, Rodrigo JP, Suárez C. Clinicopathologic analysis and predictive factors for distant metastases in patients with head and neck squamous cell carcinomas. Head Neck 2012;34(6):771–5. [19] Ljumanovic R, Langendijk JA, Hoekstra OS, Leemans CR, Castelijns JA. Distant metastases in head and neck carcinoma: identification of prognostic groups with MR imaging. Eur J Radiol 2006;60(1):58–66. [20] Boellaard R, O’Doherty MJ, Weber WA, Mottaghy FM, Lonsdale MN, Stroobants SG, et al. FDG PET and PET/CT: EANM procedure guidelines for tumour PET imaging: version 1.0. Eur J Nucl Med Mol Imaging 2010;37(1):181–200. [21] Schwartz DL, Rajendran J, Yueh B, Coltrera MD, Leblanc M, Eary J, et al. FDGPET prediction of head and neck squamous cell cancer outcomes. Arch Otolaryngol Head Neck Surg 2004;130(12):1361–7. [22] Brun E, Kjellén E, Tennvall J, Ohlsson T, Sandell A, Perfekt R, et al. FDG PET studies during treatment: prediction of therapy outcome in head and neck squamous cell carcinoma. Head Neck 2002;24(2):127–35. [23] Minn H, Lapela M, Klemi PJ, Grénman R, Leskinen S, Lindholm P, et al. Prediction of survival with fluorine-18-fluoro-deoxyglucose and PET in head and neck cancer. J Nucl Med 1997;38:1907–11. [24] Halfpenny W, Hain SF, Biassoni L, Maisey MN, Sherman JA, McGurk M. FDGPET. A possible prognostic factor in head and neck cancer. Br J Cancer 2002;86(4):512–6. [25] Allal AS, Slosman DO, Kebdani T, Allaoua M, Lehmann W, Dulguerov P. Prediction of outcome in head-and-neck cancer patients using the standardized uptake value of 2-[18F]fluoro-2-deoxy-D-glucose. Int J Radiat Oncol Biol Phys 2004;59(5):1295–300. [26] Kim SY, Roh JL, Kim MR, Kim JS, Choi SH, Nam SY, et al. Use of 18F-FDG PET for primary treatment strategy in patients with squamous cell carcinoma of the oropharynx. J Nucl Med 2007;48(5):752–7. [27] Dibble EH, Alvarez AC, Truong MT, Mercier G, Cook EF, Subramaniam RM. 18FFDG metabolic tumor volume and total glycolytic activity of oral cavity and oropharyngeal squamous cell cancer: adding value to clinical staging. J Nucl Med 2012;53(5):709–15. [28] Inokuchi H, Kodaira T, Tachibana H, Nakamura T, Tomita N, Nakahara R, et al. Clinical usefulness of [18F] fluoro-2-deoxy-D-glucose uptake in 178 head-andneck cancer patients with nodal metastasis treated with definitive chemoradiotherapy: consideration of its prognostic value and ability to provide guidance for optimal selection of patients for planned neck dissection. Int J Radiat Oncol Biol Phys 2011;79(3):747–55. [29] Ng SH, Chan SC, Liao CT, Chang JT, Ko SF, Wang HM, et al. Distant metastases and synchronous second primary tumors in patients with newly diagnosed oropharyngeal and hypopharyngeal carcinomas: evaluation of (18)F-FDG PET and extended-field multi-detector row CT. Neuroradiology 2008;50(11): 969–79. [30] Boellaard R. Need for standardization of 18F-FDG PET/CT for treatment response assessments. J Nucl Med 2011;52(Suppl 2):93S–100S.
