If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
School of Dentistry, China Medical University, Taichung, TaiwanDepartment of Oral and Maxillofacial Surgery, Changhua Christian Hospital, Changhua, TaiwanDepartment of Oral and Maxillofacial Surgery, Show Chwan Memorial Hospital, Changhua, Taiwan
School of Dentistry, China Medical University, Taichung, TaiwanDepartment of Oral and Maxillofacial Surgery, China Medical University Hospital, Taichung, Taiwan
The aims of this paper were to evaluate the relation between the preoperative primary tumour maximum standardised uptake value (SUVmax, tSUVmax) and clinicopathological features, including depth of invasion (DOI), recurrence factors, and survival outcomes, and to compare the prognostic value of tSUVmax with that of other factors associated with the recurrence of early-stage oral squamous cell carcinoma (OSCC) of the tongue. We retrospectively analysed data from 155 patients. All patients were treated and regularly monitored at the Changhua Christian Hospital (CCH). Only those who had undergone 18F-fluorodeoxyglucose positron emission tomography (FDG-PET) in the 14 days before surgery were included. A tSUVmax of >5.2 was found to be the sole strong predictor of a DOI of >4 mm. A tSUVmax of >7.6 was strongly associated with pT2 SCC of the tongue, more aggressive DOI, and perineural invasion. DOI and tSUVmax could be used to predict disease-free survival (DFS) for early-stage SCC of the tongue, and they showed stronger predictive power than the traditional American Joint Committee on Cancer (AJCC) T stage. Therefore tSUVmax could be a prognostic tool for DFS in AJCC early-stage SCC of the tongue.
According to National Comprehensive Cancer Network (NCCN) guidelines, adjuvant therapy is indicated for advanced-stage OSCC or when resected specimens show evidence of certain recurrence factors.
For early-stage SCC of the tongue (pT1–pT2, pN0), however, the current T staging system by the American Joint Committee on Cancer (AJCC) is insufficient to gauge disease progression,
Early squamous cell carcinoma of the oral tongue with histologically benign lymph nodes: a model predicting local control and vetting of the eighth edition of the American Joint Committee on Cancer pathologic T stage.
Head-to-head comparison of chest X-ray/head and neck MRI, chest CT/head and neck MRI, and 18F-FDG PET/CT for detection of distant metastases and synchronous cancer in oral, pharyngeal, and laryngeal cancer.
The uptake of glucose by cancer cells is increased by metabolic alterations that support malignant properties, such as upregulation of the epithelial-to-mesenchymal (EMT) transition pathway.
Therefore, the SUVmax as a metabolic parameter may serve as a marker of the aggressiveness of OSCC.
In this study, we evaluated the relation of preoperative primary tumour SUVmax (tSUVmax) with clinicopathological features, including depth of invasion (DOI), recurrence factors, and survival outcomes, and compared the prognostic value of the tSUVmax with that of other factors associated with the recurrence of early-stage tongue SCC.
Material and methods
Experimental section
Patients
This retrospective cohort study was approved by the Institutional Study Review Board and Ethics Committee of the Changhua Christian Hospital (CCH), Taiwan on March 24, 2021(IRB number: 210210). All patient information was obtained from hospital charts and the hospital’s cancer registry centre. A total of 3221 OSCC patients who were treated with surgery were identified. They had had adjuvant therapy and had been monitored for their follow up at the CCH between 1 January 2008 and 31 December 2019. The follow-up period was determined as the period from the date of indexing to 31 December 2019. We enrolled patients with early-stage tongue SCC who had undergone FDG-PET in the two weeks before surgery. Patients who had not received treatment in line with NCCN treatment guidelines were excluded (n = 2). Those with positive margins (n = 2), those who failed to attend for follow up or who had incomplete records (n = 3), and those who had recurrent carcinoma or metastatic carcinoma from other sites (n = 10), were also excluded. A total of 155/172 patients (90%) were finally included.
FDG-PET/computed tomography (CT) scan protocols
All patients who underwent FDG-PET/CT (Gemini GXL 16 PET/CT system, Philips Healthcare) fasted for four to six hours before FDG injection. Blood glucose levels were checked before injection to ensure a glycaemic range of 126–150 mg/dL. FDG-PET was performed 60 minutes after intravenous injection of 185–370 MBq (5–10 mCi) of FDG according to the patient’s body weight. Low-dose CT without contrast medium was also conducted for attenuation correction and anatomical localisation (patients were scanned from the vertex of the skull to the mid-thigh, and images were reconstructed in coronal and sagittal planes). The tSUVmax was obtained automatically by a routinely used formula described elsewhere (the greatest activity response in the area of interest divided by the sum of the amount of injected FDG and body weight).
The study participants had wide excision of the tumour to achieve clear safe margins. They also concurrently underwent a neck dissection as required and determined by their stage of disease. Patients with superficial tumour invasion underwent tumour-side excision only. Those with clinical N0 disease had prophylactic neck dissection (levels I - III).
Clinical and pathological parameters
Patient variables included in the analysis were age at diagnosis, survival time, sex, TNM stage according to the AJCC (8th edition), recurrence, DOI, tumour stage, extranodal spread, perineural invasion, lymphovascular invasion, and distance of close margins. Additionally, smoking history and frequency, use of betel nuts, and consumption volume and frequency of alcohol, were also recorded. Mortality data were obtained from the CCH and from the Health Bureau of Changhua City, which is updated annually.
Statistical analysis
Continuous variables were summarised as mean (SD), and categorical variables were calculated as frequencies (percentages). Fisher’s exact test was performed to compare categorical variables. To assess the influence of clinicopathological factors on DFS, univariate and multivariate Cox proportional hazards models were employed. Hazard ratios and confidence intervals were also estimated. Kaplan–Meier analyses were used to estimate the rates of overall survival and DFS. Using the Youden index, we generated receiver operating characteristic (ROC) curves for the tSUVmax to determine the optimal cut-off level to predict DOI and DFS. The area covered by the ROC curve was used to compare the ability of the tSUVmax to predict DOI and DFS. We used the log-rank test to compare group survival functions according to DFS. The significance level was set at p values of <0.05 as indicative of statistical significance. SPSS for Windows version 16 (SPSS Inc) was used for statistical analysis.
Results
For the 155 patients with a diagnosis of early-stage SCC of the tongue who were enrolled in our retrospective study, ROC curve analyses were used to predict the optimal cut-off point for tSUVmax and DOI for DFS. According to the Youden index,
the optimal tSUVmax cut-off point for DFS was 7.6 (sensitivity: 46.0%; specificity: 72.4%), the optimal DOI cut-off point for DFS was 7 mm (sensitivity: 42.0%; specificity: 77.1%) (Supplementary Fig. 1). ROC curve analyses were also used to predict the optimal tSUVmax cut-off point for a DOI greater than or less than 4 mm. According to the Youden index, the optimal cut-off point for tSUVmax was 5.2 (sensitivity: 74.0%; specificity: 83.6%). Fig. 1 shows that the area under the curve was 0.801 for a tSUVmax that predicted a DOI of more or less than 4 mm (p<0.001). Fig. 2 shows that for a DOI of >4 mm, the mean tSUVmax was 7.89; for a DOI of 4, it was 4.41 (Δ=3.4875; p<0.001). A more detailed distributional relation between tSUVmax, DOI, and pathological T stage can be seen in Supplementary Figures 2 and 3.
Fig. 1Receiver operating characteristic (ROC) curve of primary tumour standardised uptake value (tSUVmax) for predicting a depth of invasion (DOI) of >4 or 4 mm. The optimal cut-off point, based on the Youden index, was 5.2; the area under the curve was 80.1% (95% confidence interval, 0.730 to 0.861; p < 0.001).
Fig. 2Mean primary tumour standardised uptake values (tSUVmax) for depth of invasion (DOI) in the >4 and 4 mm groups were 7.89 and 4.41, respectively (Δ = 3.4875; p < 0.001).
Table 1 shows the clinicopathological characteristics and results of the univariate analysis for DFS. Univariate analysis showed a significant association of poor DFS rate with a tSUVmax of >7.6 (p = 0.0082), DOI of >7 mm (p = 0.0043), and perineural invasion (p = 0.0384). As shown in Table 2, a tSUVmax of >7.6 showed a significant association with age greater than 60 years at diagnosis (p = 0.0392), pathological T2 stage (p < 0.0001), and DOI of >5 mm or >7 mm (both p < 0.0001).
Table 1Summary of clinicopathological features of patients and results of the univariate analysis.
No. (%) patients (n = 155)
No. (%) recurrence or death
Univariate analysis (crude)
Hazard ratio
95% CI
p value
Sex:
Female
16 (10.3)
5 (31.2)
Male
139 (89.7)
45 (32.4)
Age at diagnosis (years):
≦60
90 (58.1)
26 (28.9)
1
> 60
65 (41.9)
24 (33.7)
1.3616
0.7810 to 2.3740
0.2764
Pathological T stage:
1
66 (42.6)
18 (27.3)
1
2
89 (57.4)
32 (36)
1.4045
0.7882 to 2.5027
0.2491
Pathological N stage:
0
125 (80.6)
38 (30.4)
1
W/O ND
30 (19.4)
12 (40)
Primary tumour SUVmax:
≦ 7.6
103 (66.5)
27 (26.2)
1
> 7.6
52 (33.5)
23 (44.2)
2.1221
1.2151 to 3.7060
0.0082
Depth of tumour invasion:
≦ 5
75 (48.4)
19 (25.3)
1
> 5
80 (51.6)
31 (38.7)
1.7128
0.9673 to 3.0329
0.0649
≦ 7
93 (60)
22 (23.7)
1
> 7
62 (40)
28 (45.2)
2.2579
1.2907 to 3.9499
0.0043
Perineural invasion:
No
100 (71.2)
28 (28)
1
Yes
55 (28.8)
22 (40)
1.8139
1.0323 to 3.1873
0.0384
Lymphovascular invasion:
No
136 (87.7)
41 (30.1)
1
Yes
19 (12.3)
9 (47.4)
1.9209
0.9322 to 3.9583
0.0768
Grade:
Well
20 (13)
4 (20)
0.4399
0.1569 to 1.2336
0.1185
Moderate
118 (76).6
41 (34.7)
1
Poor
16 (10.4)
5 (31.2)
0.963
0.3803 to 2.4385
0.9366
Smoking:
No
44 (28.4)
16 (36.4)
1
Yes
111 (71.6)
34 (30.6)
0.8393
0.4631 to 1.5211
0.5636
Betel nut:
No
72 (46.5)
22 (30.6)
1
Yes
83 (53.5)
28 (33.7)
1.2631
0.7213 to 2.2120
0.4138
Alcohol:
No
73 (47.1)
24 (32.9)
1
Yes
82 (52.9)
26 (31.7)
0.9854
0.5652 to 1.7178
0.9586
W/O ND: without neck dissection; SUVmax: standardised uptake value
In multivariate Cox regression analysis for DFS (Table 3) we conducted two models to compare the survival parameter, including the pathological T stage, tSUVmax, perineural invasion, lymphovascular invasion, and DOI. Overall, a tSUVmax of >7.6 and a DOI of >7 mm were shown to have a significant prognostic ability for DFS.
Table 3Multivariate Cox proportional-hazards regression analysis of different models for disease-free survival.
Fig. 3 (Supplementary Figs. 4 and 5) depicts Kaplan–Meier curves for DFS disaggregated into two groups according to a tSUVmax cut-off level of 7.6, a DOI cut-off level of 7 mm, and pT1 versus pT2 stage. The curves show significant differences in DFS rates between the groups regarding DOI (p = 0.0034) and tSUVmax (p = 0.0068).
Fig. 3Kaplan–Meier curves for disease-free survival (DFS) in two groups of patients, classified according to a primary tumour standardised uptake value (tSUVmax) cut-off level of 7.6. These results show significant differences in DFS rates in the two groups regarding tSUVmax (p = 0.0068).
To our knowledge this study is the first to investigate the use of the tSUVmax for the prediction of DFS in patients with early-stage (pT1–pT2, pN0) tongue SCC. In this study, the ability of a preoperative tSUVmax to predict DOI and DFS was evaluated among patients with early-stage tongue SCC. A tSUVmax of >5.2 was found to be the sole strong predictor of a DOI of >4 mm (area under the curve, 0.801).
The maximum standardized uptake value increment calculated by dual-time-point 18F-fluorodeoxyglucose positron emission tomography predicts survival in patients with oral tongue squamous cell carcinoma.
and should be used to guide decision making for neck surgery. For tumours with a DOI of >4 mm, elective surgery should be strongly considered if postoperative radiotherapy is not planned.
A tSUVmax of >7.6 was strongly associated with pT2 tongue SCC, more aggressive DOI, and perineural invasion. DOI and tSUVmax could be used as a prediction tool for DFS in patients with early-stage tongue SCC, and showed stronger predictive power than the traditional AJCC T stage. Moreover, the tSUVmax can be calculated from preoperative FDG-PET scans, but DOI can be estimated only from resected specimens.
In several studies, investigators have used different tools to determine DOI for tongue SCC. Alsaffar et al reported that preoperative magnetic resonance imaging (MRI) could demonstrate whether the DOI was >5 or 5 mm; sensitivity and specificity were 80% and 97%, respectively.
However, their sample size was small, which could be the main limitation for interpreting the results. Xu et al showed that an MRI-determined DOI of 7.5 mm could be used as an indicator for prophylactic neck dissection, and an additional risk for recurrence and cancer-related mortality from stage cT1N0 tongue SCC.
Iida et al used intraoral ultrasonography to diagnose superficial invasion by tongue SCC in 56 patients, and to compare the values obtained with DOI measurements. They found that for a DOI of 5 or >5 mm, intraoral ultrasonography had a sensitivity and specificity of 92.3% and 70.6%, respectively.
To the best of our knowledge, our study is the first to use FDG-PET to interpret correlation of the tSUVmax with a DOI of >4 or 4 mm.
In a cohort study of 52 patients, Zheng et al proved that the tSUVmax strongly correlated with tumour dimension, tumour stage, and lymph node metastasis (cut-off point of tSUVmax: 6.57).
Relationship between the maximum standardized uptake value of fluoro-2-deoxyglucose-positron emission tomography/computed tomography and clinicopathological characteristics in tongue squamous cell carcinoma.
In comparison, we investigated the association of the tSUVmax with early-stage tongue SCC (pT1–pT2, pN0). Because of their small sample size, further investigation of early-stage tongue SCC in the study by Zheng et al would have been difficult, and survival was not clearly addressed.
Yonezawa et al reported the use of different metabolic parameters, including SUV early, SUV delayed, retention index, and change in SUVmax to predict survival among 52 patients with tongue SCC. In their multivariate analysis, change in the SUVmax appeared to be a better factor for predicting overall survival than the traditional pathological T stage, and their results regarding the change in the SUVmax were similar to ours for tSUVmax.
The maximum standardized uptake value increment calculated by dual-time-point 18F-fluorodeoxyglucose positron emission tomography predicts survival in patients with oral tongue squamous cell carcinoma.
Our sample size was larger, and we investigated cases of pathological N0 tongue SCC to exclude bias from neck lymph node metastasis because pN positivity could dramatically decrease the survival of affected patients.
In 2018 Lee et al studied 57 patients with cN0 stage tongue SCC26 and concluded from their univariate analysis that pN stage, overall AJCC stage, SUVmax, average SUV, metabolic tumour volume, and total lesion glycolysis were good prognostic factors for rate of survival. However, in multivariate analysis, pN stage, overall AJCC stage, and metabolic tumour volume were independent prognostic tools for the overall survival of patients with stage cN0 tongue SCC.
Prognostic value of volume-based (18)F-fluorodeoxyglucose PET/CT parameters in patients with clinically node-negative oral tongue squamous cell carcinoma.
Prognostic value of metabolic tumor volume measured by (18)F-FDG PET/CT in locally advanced head and neck squamous cell carcinomas treated by surgery. Nucl Med.
In these, however, the enrolled cases included all AJCC stages and all pathological sites, which could have contributed to major bias. By contrast, we enrolled only patients with early-stage tongue SCC and found that the tSUVmax could be an independent prognostic factor for DFS.
The SUVmax value is associated with the aggressiveness of malignant tumours and strongly correlates with the EMT pathway.
Kajiwara et al reported that oesophageal cancer patients with a SUVmax value greater than 10.26 had EMT gene alteration, including significantly different expression of Snail, E-cadherin, FN-1, integrin-α5, MMP-1, MMP-2, N-cadherin, TIMP-1, and IL‑8.
Maximum standardized uptake value in 18F-fluoro-2-deoxyglucose positron emission tomography is associated with advanced tumor factors in esophageal cancer.
Higashi et al also reported that a higher SUVmax value was associated with lower E-cadherin expression in patients with lung adenocarcinoma, and that it correlated with poor differentiation, aggressiveness, and postoperative recurrence.
Our study has several limitations that should be considered given the results. First, the dentate condition could contribute to false-positive results, and in very early-stage tumours, the tSUVmax may be difficult to capture by FDG-PET. Moreover, FDG-PET is liable to be affected by the inflammatory process - for example, periodontitis may cause false-positive results.
Next, the data were retrieved from only one medical centre, hence the results cannot be generalised. In addition, the results cannot not be generalised to other countries because of atypical cultural practice in Taiwan, wherein more people chew betel nuts than in other countries.
In future, other common metabolic parameters related to the shape and texture analysis of FDG-PET, such as metabolic tumour volume or total lesion glycolysis, which have been demonstrated in previous articles, may serve as powerful prognostic factors,
Prognostic value of volume-based (18)F-fluorodeoxyglucose PET/CT parameters in patients with clinically node-negative oral tongue squamous cell carcinoma.
In conclusion, the tSUVmax could play a better prognostic role than pathological T stage, and could be a prognostic tool for DFS in patients with AJCC early-stage SCC of the tongue.
Funding
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors
Ethics statement/confirmation of patients’ permission
The study was approved by the Institutional Study Review Board and Ethics Committee of CCH, Changhua, Taiwan on March 24, 2021 (IRB number: 210210). Patients’ permission was not required.
Declaration of competing interest
The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Acknowledgements
The authors would like to thank Enago (www.enago.com) for the English language review.
Appendix A. Supplementary material
The following are the Supplementary data to this article:
Supplementary Fig. 1According to the Youden index, the optimal tSUVmax cut-off point for DFS was 7.6 (sensitivity: 46.0%; specificity: 72.4%), the optimal DOI cut-off point for DFS was 7 mm (sensitivity: 42.0%; specificity: 77.1%).
Supplementary Fig. 4Kaplan–Meier curves for disease-free survival in two groups of patients, classified according to a depth of invasion (DOI) cut-off of 7 mm. These results show significant differences in DFS rates in the two groups regarding DOI (p = 0.0034).
Supplementary Fig. 5Kaplan–Meier curves for disease-free survival in two groups of patients, classified according to pT1 versus pT2. There were no significant differences in DFS rates between the two groups regarding stage pT1 versus pT2 (tSUVmax: tumour standardised uptake value; DOI: depth of tumour invasion; DFS: disease-free survival).
Early squamous cell carcinoma of the oral tongue with histologically benign lymph nodes: a model predicting local control and vetting of the eighth edition of the American Joint Committee on Cancer pathologic T stage.
Head-to-head comparison of chest X-ray/head and neck MRI, chest CT/head and neck MRI, and 18F-FDG PET/CT for detection of distant metastases and synchronous cancer in oral, pharyngeal, and laryngeal cancer.
The maximum standardized uptake value increment calculated by dual-time-point 18F-fluorodeoxyglucose positron emission tomography predicts survival in patients with oral tongue squamous cell carcinoma.
Relationship between the maximum standardized uptake value of fluoro-2-deoxyglucose-positron emission tomography/computed tomography and clinicopathological characteristics in tongue squamous cell carcinoma.
Prognostic value of volume-based (18)F-fluorodeoxyglucose PET/CT parameters in patients with clinically node-negative oral tongue squamous cell carcinoma.
Prognostic value of metabolic tumor volume measured by (18)F-FDG PET/CT in locally advanced head and neck squamous cell carcinomas treated by surgery. Nucl Med.
Maximum standardized uptake value in 18F-fluoro-2-deoxyglucose positron emission tomography is associated with advanced tumor factors in esophageal cancer.