Sage Journals: Discover world-class research

Abstract

Background:

Following changes in primary tumor (T) and lymph node (N) staging for nasopharyngeal carcinoma (NPC) in the Eighth Edition AJCC Cancer Staging Manual, simplification of T staging has been proposed. However, a limited range of 2-deoxy-2-[fluorine-18] fluoro-D-glucose positron emission tomography–computed tomography (¹⁸F-FDG PET–CT) metabolic parameters has been investigated. Therefore, we aimed to evaluate the primary tumor invasiveness and the lymph node metastasis (LNM) of NPC from a metabolic perspective.

Methods:

A total of 435 NPC patients underwent ¹⁸F-FDG PET/CT before treatment were retrospectively examined. The primary endpoint was differences in standard uptake value (SUV), lean body mass-normalized SUV (SUL), body surface area-normalized SUV (SUS), glucose-normalized SUV (GN), metabolic tumor volume (MTV), total lesion glycolysis (TLG), and glucose-normalized total lesion glycolysis (GNTLG) of primary tumors and LNM between different T and N stages. The metabolic parameters associated with T and N staging were identified.

Results:

There were significant differences between all parameters relative to the primary tumor but no significant differences in any parameter relative to the LNM and T stages. Higher mean values of TGN_max, TGN_mean, TSUV_peak, and TSUS_max were associated with advanced T stages. Higher mean values of all the LNM parameters were associated with more advanced N stages. Only primary tumor metabolic tumor volume (TMTV), TSUV_peak, TSUL_max, and TSUS_max showed a significant positive association with T staging, while lymph node metabolic tumor volume (LNMTV) and TSUS_max were significantly positive in N staging.

Conclusions:

Our findings suggest that metabolic parameters are useful indicators of tumor invasiveness and LNM based on the Eighth Edition manual. Compared with volume-dependent parameters, TGN_max, TGN_mean, TSUV_peak, and TSUS_max may be better indicators of local tumor aggressiveness. SUS_max of the primary tumor was associated with LNM. In addition to SUV_max, other metabolic parameters (eg, SUL_max, SUS_max, GN_max, and GN_mean) could evaluate tumor aggressiveness and LNM better.

Keywords

Nasopharyngeal carcinoma metabolic parameters primary tumor aggressiveness lymph node metastasis

Introduction

Nasopharyngeal carcinoma (NPC) exhibits clear geographical distribution patterns, especially in East and Southeast Asia.¹ The geographical global distribution of NPC is extremely unbalanced with over 70% of new cases diagnosed in eastern and southeast Asia corresponding to a 5-year prevalence (all ages) of 12.91 cases per 100 000 in China.² With the advancement of radiotherapy technology, induction and concurrent chemotherapy, and the development of an accurate staging system, the treatment of NPC has continuously improved.¹

In 2017, the American Joint Committee on Cancer/Union for International Cancer Control (AJCC/UICC) revised the staging system for NPC³ in the Eighth Edition AJCC Cancer Staging Manual. The Eighth Edition AJCC Cancer Staging Manual is currently the most common prognostic tool for NPC and remains the most influential resource for guiding treatment, assessing the treatment response, comparing outcomes between different institutions, and academic research. The main changes to T staging in the Eighth Edition are as follows: medial pterygoid, lateral pterygoid, or prevertebral muscle involvement was included in T2, cervical vertebral invasion was included in T3, and the ambiguous terms infratemporal fossa/masticator space invasion was removed and replaced with extensive soft tissue invasion (soft tissue outside the lateral pterygoid muscle and parotid gland) in T4.⁴ Regarding N staging changes, stages N3a and N3b in the Seventh Edition were merged into a single stage, N3, and the N3 criterion was changed to “below the caudal border of cricoid cartilage.”⁴ However, some studies have expressed support for these changes,^5-9 while others have not.^10-14 Many studies have shown that N staging is reasonable^5-9 while T staging needs further improvement.^10-14 A comparative assessment was conducted by Li et al¹¹ on the 5-year overall survival (OS) and progression-free survival (PFS) of individuals diagnosed with NPC who presented severe skull base invasion (SBI) versus those with slight SBI. Their results suggested that patients with slight SBI experienced improved outcomes and that the OS and PFS of patients with slight SBI is not significantly improved by additional induction chemotherapy therapy. They advised to downgrade the stage of NPC patients with mild SBI from T3 to T2 stage. This suggestion could potentially address the issue of conflicting prognoses for T2 and T3 stages and indications for using induction chemotherapy. Pan et al¹⁴ have simplified T stage as the use of 17 anatomical structures in the Eighth Edition is too complicated, its popularization and application are greatly limited. The differences in local control and survival between T stage have narrowed down due to the advances in diagnostics and treatment. Therefore, some studies have proposed the simplification of T staging.^14,15

Magnetic resonance imaging (MRI) is the preferred imaging technique for assessment, staging, evaluating treatment effectiveness, and monitoring NPC progression due to its advantages of high soft tissue resolution, multiparametric imaging, and non-ionizing radiation.¹⁶ However, 2-deoxy-2-[fluorine-18] fluoro-D-glucose positron emission tomography–computed tomography (¹⁸F-FDG PET–CT) surpasses MRI in accuracy in the identification of cervical nodal metastases, and should therefore be the preferred reference for assessing the metastases of the neck. ¹⁸F-FDG PET–CT demonstrates proficient diagnostic accuracy and a low false positive rate in identifying distant metastases, rendering it a viable substitute for conventional approaches. The National Comprehensive Cancer Network and the Chinese Society of Clinical Oncology currently recommend ¹⁸F-FDG PET–CT as a proven imaging strategy in NPC management.^16,17

¹⁸F-FDG PET–CT parameters offer metabolic information that can be used to evaluate tumor aggressiveness,^18-20 predict lymph node metastasis (LNM)^21-23 and may be correlated with patient survival.²⁴ The maximum standard uptake value (SUV_max) is the most commonly used metabolic parameter. SUV_max is an averaged value of the tracer uptake in sufficient numbers of cancer cells and other cells in the most metabolically aggressive part of the potentially heterogeneous tumor. The limitation of SUV_max is that it only represents the maximum uptake within the volume of interest (VOI) and not the entire mass. The use of ¹⁸F-FDG-PET–CT parameters reflecting tumor size and metabolic information, such as metabolic tumor volume (MTV) and total lesion glycolysis (TLG), has been proposed to overcome the limitation. MTV and TLG have been acknowledged as useful indicators for tumor aggressiveness, LNM, and patient survival for various tumors. MTV is defined as the total number of voxels exceeding the predetermined SUV threshold within the VOI, while TLG is calculated by multiplying the MTV by the mean of SUV (SUV_mean).²⁵ Other metabolic parameters, such as the maximum lean body mass (LBM)-normalized SUV (SUL_max),²⁶ maximum body surface area (BSA)-normalized SUV (SUS_max),²⁷ maximum glucose-normalized SUV (GN_max), GN_mean, and glucose-normalized total lesion glycolysis (GNTLG),²⁸ have rarely been utilized in previous studies.

Given the changes to T and N staging, and the few¹⁸ F-FDG PET–CT metabolic parameters investigated in previous studies, the main objectives of this study were to evaluate the primary tumor invasiveness and the LNM in the new staging system by comparing metabolic parameters of primary tumors and LNM of different T and N stages. We also aimed to determine the relationship between metabolic parameters and T and N stages to provide reference values for future research.

Patients and Methods

Patients

The clinical records of all patients with NPC with undifferentiated carcinoma diagnosed between June 2016 and January 2020 at Jiangsu Cancer Hospital, Nanjing, China, were investigated. Carcinoma staging was based on the Eighth Edition AJCC Cancer Staging Manual, and patients who were not staged according to the latest staging system were re-staged by clinicians according to the Eighth Edition manual. Only patients who had undergone MRI of the neck were eligible. All patients provided written informed consent to participate before the initiation of the study. The Ethics Committee of Jiangsu Cancer Hospital approved this study (protocol code 2022k-k026 and date of approval: April 22, 2022).

PET–CT imaging

All patients underwent examination with a PET–CT scanner (Discovery 710; GE Medical Systems, Waukesha, Wisconsin, USA). Intravenous administration of ¹⁸F-FDG (0.1–0.2 mCi/kg) was performed in patients who fasted for a minimum period of 6 hours. Before injection, blood glucose concentrations were measured to ensure that they were less than 11 mmol/L. During the radiotracer distribution, patients rested in the waiting room and orally ingested approximately 1 000 mL of water. The patients were instructed to urinate immediately before the examination. PET–CT image acquisition started 50–70 minutes after ¹⁸F-FDG injection. CT and PET scan parameters were as follows: vertex to mid-thigh scanning range, 140 kV tube voltage, auto mA (noise index, 28.5) tube current, 0.8 second rotation time, 3.75 mm slice thickness, 2 minutes/bed position emission scan time, and six to seven bed positions scanning range. The acquired PET images were reconstructed with an iterative reconstruction algorithm (VUE point FX +SharplR: iteration = 2; subset = 24) with CT-based attenuation correction. Images were analyzed on the workstation (Advanced Workstation AW 4.6, GE HealthCare, Chicago, IL, USA).

PET–CT image analysis

An experienced nuclear medicine physician analyzed the images. VOIs were placed around the primary tumor and LNM. Methods for the placement of VOI have been described previously.²⁹ Within the selected VOI, SUV_max, SUL_max, SUS_max, SUV_mean, GN_max, GN_mean, MTV, TLG, and GNTLG of the primary tumor and LNM with the highest SUV_max values were evaluated using the fixed percentage of SUV_max threshold algorithm (42% of SUV_max).³⁰ The SUV measures the uptake in a tumor normalized according to the distribution volume. SUV_max and SUV_mean were defined as the maximum and average values of SUVs, respectively. Glucose-normalized SUV_max and SUV_mean were defined as GN_max and GN_mean, respectively.²⁸ SUV_peak is the average SUV computed within a fixed VOI, most often containing (and not necessarily centered on) the hottest pixel value.³¹ GNTLG is calculated by multiplying MTV by GN_mean.²⁸ The LBM-normalized SUV was defined as SUL, while the BSA-normalized SUV was defined as SUS.²⁷

Statistical analysis

Multigroup comparisons were performed using analysis of variance (ANOVA) to compare the differences in metabolic parameters of the primary tumors and LNM of stages T1–T4 and N0–N3. The Spearman rank correlation coefficient was used to characterize the relationship between the PET parameters and NPC T and N staging. Ordered logistic regression analysis was used to identify the PET parameters that are significantly associated with T and N staging. Two-tailed P-values of < .05 were considered statistically significant. Statistical Package for Social Sciences software (SPSS, version 22.0, IBM, Armonk, NY, USA) was used for all statistical analyses.

Results

A total of 435 patients with histologically confirmed undifferentiated carcinoma who underwent ¹⁸F-FDG-PET–CT imaging before treatment were included in this study.

Clinical characteristics

The clinical characteristics of patients with NPC included in this study are listed in Table 1.

Table 1.

Clinical characteristics of patients with stages I–IV nasopharyngeal carcinoma.

	I	II	III	IVA	IVB
Number, %	15 (3.45%)	71 (16.32%)	164 (37.70%)	157 (36.09%)	28 (6.44%)
Sex (M/F)	(11/4)	(53/18)	(121/43)	(112/45)	(24/4)
Age (years)	49.07 ± 9.49	50.73 ± 11.50	50.59 ± 11.74	49.48 ± 12.18	52.57 ± 11.86
Height (cm)	169.87 ± 7.33	168.28 ± 7.21	167.81 ± 7.50	166.57 ± 7.11	166.04 ± 9.07
Weight (kg)	72.93 ± 12.12	70.09 ± 11.95	67.45 ± 11.67	63.81 ± 11.07	63.05 ± 11.32
¹⁸F-FDG dose (mCi/kg)	0.15 ± 0.03	0.15 ± 0.03	0.15 ± 0.03	0.16 ± 0.04	0.16 ± 0.04
Blood glucose (mmol/L)	5.25 ± 0.60	5.70 ± 0.88	5.86 ± 1.05	5.79 ± 0.88	5.71 ± 0.77

Abbreviation: ¹⁸F-FDG, 2-deoxy-2-[fluorine-18] fluoro-D-glucose.

IV includes IVA and IVB.

Comparison of PET parameters between stage T1–T4

Significant differences in TMTV, TSUV_max, TSUV_mean, TTLG, TGN_max, TGN_mean, TGNTLG, TSUV_peak, TSUL_max, and TSUS_max values between stages T1 and T4 tumors were found (P < .05).

We did not observe any significant differences in the lymph node metabolic tumor volume (LNMTV), lymph node standard uptake value (LNSUV)_max, LNSUV_mean, LNTLG, lymph node glucose-normalized standard uptake value (LNGN)_max, LNGN_mean, lymph node glucose-normalized total lesion glycolysis (LNGNTLG), LNSUV_peak, lymph node lean body mass-normalized SUV (LNSUL)_max, or lymph node body surface area-normalized SUV (LNSUS)_max values between stages T1 and T4 carcinomas (P > .05).

The pairwise comparison of the TGN_max, TGN_mean, TSUV_peak, and TSUS_max values between stages T1 and T4 showed significant differences (P < .05). Further analysis suggested that higher TGN_max, TGN_mean, TSUV_peak, and TSUS_max mean values were associated with more advanced T staging.

We observed no significant differences in TMTV, TTLG, and TGNTLG values between stages T1 and T2 (P > .05) (Table 2, Figure 1).

Figure 1.

¹⁸F-FDG PET–CT imaging of primary tumors (yellow arrows) in patients with nasopharyngeal carcinoma. A, B, C, and D depict stages T1, T2, T3, and T4 carcinomas, respectively. ¹⁸F-FDG PET–CT indicates 2-deoxy-2-[fluorine-18] fluoro-D-glucose positron emission tomography–computed tomography.

Table 2.

Comparison of PET parameters of primary tumors and lymph node metastasis between stages T1 and T4.

	T1 (n = 56)	T2 (n = 91)	T3 (n = 147)	T4 (n = 141)	F	P
TMTV	4.13 ± 2.39^a	4.90 ± 3.16^b	7.96 ± 6.23	20.76 ± 15.75	71.855	.000
TSUV_max	8.65 ± 3.82	12.01 ± 4.58^c	13.15 ± 5.19^a	15.45 ± 6.06	24.441	.000
TSUV_mean	5.18 ± 2.32	7.38 ± 3.03^c	8.08 ± 3.36^a	9.37 ± 3.77	22.557	.000
TTLG	19.77 ± 11.86^a	34.10 ± 22.31^b	65.41 ± 72.88	193.88 ± 167.35	65.400	.000
TGN_max	8.50 ± 4.34	11.28 ± 5.36	13.38 ± 5.35	15.60 ± 6.47	25.012	.000
TGN_mean	5.07 ± 2.65	6.93 ± 3.46	8.22 ± 3.43	9.45 ± 4.01	23.488	.000
TGNTLG	19.55 ± 13.24^a	31.60 ± 22.93^b	65.88 ± 69.18	195.93 ± 171.32	66.318	.000
T_peak	6.05 ± 2.53	9.02 ± 3.75	10.25 ± 4.21	12.72 ± 5.13	36.207	.000
TSUL_max	6.60 ± 2.88	9.40 ± 3.65^c	10.40 ± 4.07^a	12.37 ± 4.83	28.228	.000
TSUS_max	2.30 ± 1.19	3.10 ± 1.23	3.48 ± 1.38	4.19 ± 1.59	27.489	.000
LNMTV	4.72 ± 5.19	4.89 ± 5.42	4.41 ± 5.37	3.81 ± 3.29	1.097	.35
LNSUV_max	9.48 ± 7.47	8.06 ± 6.56	9.49 ± 5.95	9.93 ± 6.00	1.705	.165
LNSUV_mean	6.03 ± 4.87	5.11 ± 4.26	5.99 ± 3.89	6.17 ± 3.87	1.366	.252
LNTLG	44.33 ± 68.98	39.48 ± 66.48	35.84 ± 57.36	27.83 ± 36.46	1.523	.208
LNGN_max	8.96 ± 7.27	8.02 ± 6.74	9.56 ± 6.09	10.02 ± 6.29	1.910	.127
LNGN_mean	5.70 ± 4.72	5.07 ± 4.36	6.04 ± 3.98	6.28 ± 4.05	1.661	.175
LNGNTLG	39.68 ± 59.86	39.41 ± 65.73	35.83 ± 57.25	28.47 ± 37.92	1.028	.380
LN_peak	7.39 ± 6.13	6.27 ± 5.39	7.27 ± 4.98	7.23 ± 4.75	0.917	.433
LNSUL_max	7.33 ± 5.72	6.31 ± 5.01	7.50 ± 4.72	7.92 ± 4.73	2.024	.110
LNSUS_max	2.41 ± 1.84	2.21 ± 1.93	2.52 ± 1.59	2.69 ± 1.60	1.505	.213

Abbreviations: LN, lymph node; LNGN, lymph node glucose-normalized standard uptake value; LNGNTLG, lymph node glucose-normalized total lesion glycolysis; LNMTV, lymph node metabolic tumor volume; LNSUL, lymph node lean body mass-normalized SUV; LNSUS, lymph node body surface area-normalized SUV; LNSUV, lymph node standard uptake value; LNTLG, lymph node total lesion glycolysis; PET, positron emission tomography; TGN, primary tumor glucose-normalized standard uptake value; TGNTLG, primary tumor glucose-normalized total lesion glycolysis; TMTV, primary tumor metabolic tumor volume; TSUL, primary tumor lean body mass-normalized SUV; TSUS, primary tumor body surface area-normalized SUV; TSUV, primary tumor standard uptake value; TTLG, primary tumor total lesion glycolysis.

Not significantly different from T2.

Not significantly different from T1.

Not significantly different from T3.

Comparison of PET parameters between stages N0 and N3

Significant differences were found between TMTV, TSUV_max, TSUV_mean, TTLG, TGN_max, TGN_mean, TGNTLG, TSUV_peak, TSUL_max, TSUS_max, LNMTV, LNSUV_max, LNSUV_mean, LNTLG, LNGN_max, LNGN_mean, LNGNTLG, LNSUV_peak, LNSUL_max, and LNSUS_max values of stages N0–N3 (P < .05).

The pairwise comparison of LNMTV, LNSUV_max, LNSUV_mean, LNTLG, LNGN_max, LNGN_mean, LNGNTLG, LNSUV_peak, LNSUL_max, and LNSUS_max values between stages N0 and N3 showed significant differences (P < .05). Further analysis suggested that higher mean LNMTV, LNSUV_max, LNSUV_mean, LNTLG, LNGN_max, LNGN_mean, LNGNTLG, LNSUV_peak, LNSUL_max, and LNSUS_max values were associated with more advanced N stages (Table 3).

Table 3.

Comparison of PET parameters of primary tumors and lymph node metastasis between stages N0 and N3.

	N0 (n = 58)	N1 (n = 165)	N2 (n = 167)	N3 (n = 45)	F	P
TMTV	7.13 ± 5.36^a	10.80 ± 13.09^a,b	13.01 ± 12.49^c	9.01 ± 10.77^c,d	4.083	0.007
TSUV_max	10.96 ± 4.60	12.81 ± 5.64^b	13.65 ± 5.44^a,c	14.67 ± 6.72^b	4.776	0.003
TSUV_mean	6.73 ± 3.00	7.85 ± 3.58^a,b	8.29 ± 3.46^a,c	8.92 ± 4.12^{b, c}	4.007	0.008
TTLG	46.43 ± 43.12^a	89.71 ± 131.73^a	117.52 ± 135.42^a	89.83 ± 124.97^{b, c,d}	4.887	0.002
TGN_max	10.33 ± 5.27	12.62 ± 6.24^b	13.84 ± 5.55^a,c	15.03 ± 7.13^b	6.915	0.000
TGN_mean	6.31 ± 3.36	7.73 ± 3.94^b	8.40 ± 3.52^a,c	9.12 ± 4.33^b	6.117	0.000
TGNTLG	45.33 ± 45.73^a	89.09 ± 134.58^a	118.93 ± 136.52^a	90.94 ± 124.50^{b, c,d}	5.131	0.002
T_peak	8.57 ± 3.83	10.01 ± 4.76^a,b	10.84 ± 4.77^a,c	11.16 ± 5.38^{b, c}	4.020	0.008
TSUL_max	8.56 ± 3.70	10.15 ± 4.50^b	10.80 ± 4.37^a,c	11.63 ± 5.26^b	5.072	0.002
TSUS_max	2.88 ± 1.37	3.39 ± 1.46^b	3.64 ± 1.49^a,c	3.97 ± 1.84^b	5.499	0.001
LNMTV	0	4.13 ± 3.78	5.11 ± 4.02	8.01 ± 8.39	31.945	0.000
LNSUV_max	0	9.22 ± 5.56	11.26 ± 5.11	14.62 ± 4.78	95.761	0.000
LNSUV_mean	0	5.80 ± 3.65	7.10 ± 3.34	9.14 ± 3.46	85.870	0.000
LNTLG	0	29.08 ± 47.12	40.25 ± 48.52	83.31 ± 95.17	23.219	0.000
LNGN_max	0	8.99 ± 5.70	11.40 ± 5.21	14.74 ± 5.36	91.946	0.000
LNGN_mean	0	5.64 ± 3.73	7.18 ± 3.40	9.38 ± 3.67	85.409	0.000
LNGNTLG	0	27.18 ± 42.26	40.62 ± 48.39	84.92 ± 94.71	26.504	0.000
LN_peak	0	6.79 ± 4.46	8.60 ± 4.26	11.48 ± 4.95	81.590	0.000
LNSUL_max	0	7.23 ± 4.28	8.90 ± 4.06	11.53 ± 3.58	97.197	0.000
LNSUS_max	0	2.47 ± 1.53	2.99 ± 1.37	3.94 ± 1.29	92.760	0.000

Not significantly different from N3.

Not significantly different from N2.

Not significantly different from N1.

Not significantly different from N0.

Correlation between PET parameters, and T and N staging

The parameters TMTV, TSUV_max, TSUV_mean, TTLG, TGN_max, TGN_mean, TGNTLG, TSUV_peak, TSUL_max, and TSUS_max were correlated with T staging (P < .05), and the correlation coefficients were 0.662, 0.391, 0.384, 0.731, 0.393, 0.384, 0.732, 0.477, 0.415, and 0.421, respectively.

The parameters LNMTV, LNSUV_max, LNSUV_mean, LNTLG, LNGN_max, LNGN_mean, LNGNTLG, LNSUV_peak, LNSUL_max, and LNSUS_max were not correlated with T staging (P > .05).

All primary tumor and LN parameters, except TMTV (P = 0.120), were correlated with N staging (P < 0.05), and the correlation coefficients were 0.165, 0.155, 0.117, 0.202, 0.192, 0.143, 0.150, 0.168, 0.175, 0.506, 0.568, 0.559, 0.560, 0.574, 0.573, 0.574, 0.569, 0.567, and 0.568, respectively.

Relationship between PET parameters, and T and N staging

TMTV, TSUV_peak, TSUL_max, and TSUS_max had significant positive associations with T staging, and the odds ratio (OR) and P-values were 1.357 and 0.000, 1.709 and 0.000, 1.436 and 0.020, and 1.643 and 0.033, respectively. LNMTV and TSUS_max had significant positive associations with N staging, and the OR and P-values were 1.404 and 0.000, and 1.913 and 0.024 (Table 4, Figure 2).

Table 4.

Summary of logistic regression analysis results of PET parameters, and T and N staging.

Variables	T stage		N stage
Variables	OR (95% CI)	P	OR (95% CI)	P
TMTV	1.357 (1.253–1.470)	0.000	/	/
TSUV_max	1.200 (0.184–7.813)	0.849	4.286 (0.605–30.382)	.145
TSUV_mean	0.276 (0.013–5.981)	0.412	0.072 (0.003–1.703)	.103
TTLG	0.975 (0.947–1.004)	0.093	1.010 (0.989–1.032)	.335
TGN_max	1.151 (0.174–7.622)	0.884	0.248 (0.035–1.747)	.162
TGN_mean	0.827 (0.038–18.226)	0.904	10.501 (0.450–244.928)	.143
TGNTLG	1.012 (0.983–1.041)	0.428	0.991 (0.971–1.012)	.411
T_peak	1.709 (1.347–2.166)	0.000	0.989 (0.801–1.221)	.918
TSUL_max	1.436 (1.090–1.891)	0.010	0.942 (0.654–1.359)	.750
TSUS_max	1.643 (1.015–2.661)	0.044	1.913 (1.088–3.366)	.024
LNMTV	/	/	1.404 (1.270–1.553)	.000
LNSUV_max	/	/	2.115 (0.515–8.682)	.298
LNSUV_mean	/	/	0.268 (0.021–3.382)	.309
LNTLG	/	/	0.987 (0.947–1.029)	.545
LNGN_max	/	/	0.731 (0.171–3.120)	.672
LNGN_mean	/	/	2.239 (0.174–28.815)	.536
LNGNTLG	/	/	0.987 (0.946–1.029)	.545
LN_peak	/	/	1.103 (0.848–1.433)	.466
LNSUL_max	/	/	1.048 (0.690–1.592)	.826
LNSUS_max	/	/	1.173 (0.689–1.996)	.556
McFadden R²	0.289		0.239

Abbreviations: CI, confidence interval; LN, lymph node; LNGN, lymph node glucose-normalized standard uptake value; LNGNTLG, lymph node glucose-normalized total lesion glycolysis; LNMTV, lymph node metabolic tumor volume; LNSUL, lymph node lean body mass-normalized SUV; LNSUS, lymph node body surface area-normalized SUV; LNSUV, lymph node standard uptake value; LNTLG, lymph node total lesion glycolysis; OR, odds ratio; PET, positron emission tomography; TGN, primary tumor glucose-normalized standard uptake value; TGNTLG, primary tumor glucose-normalized total lesion glycolysis; TMTV, primary tumor metabolic tumor volume; TSUL, primary tumor lean body mass-normalized SUV; TSUS, primary tumor body surface area-normalized SUV; TSUV, primary tumor standard uptake value; TTLG, primary tumor total lesion glycolysis.

Figure 2.

¹⁸F-FDG PET–CT imaging of lymph node metastasis (yellow arrows) in patients with nasopharyngeal carcinoma. A, B, C, and D depict stages N0, N1, N2, and N3 lymph node metastases, respectively. ¹⁸F-FDG PET–CT indicates 2-deoxy-2-[fluorine-18] fluoro-D-glucose positron emission tomography–computed tomography.

Discussion

The unique ¹⁸F-FDG probe is utilized for molecular imaging, exploiting the ability of cancer cells to take up the tracer in proportion to their glucose utilization rate. The Warburg effect refers to the enhanced utilization of glucose through aerobic metabolism in cancer cells, which becomes more pronounced as the cells become less differentiated and more aggressive. Therefore, as a tumor becomes increasingly undifferentiated and aggressive, it results in elevated glucose utilization and increased uptake of the radioactive tracer ¹⁸F-FDG.^19,20 Black patients with breast cancer have poorer prognosis and higher mortality. Abubakar et al¹⁹ observed differences in ¹⁸F-FDG PET–CT metabolic parameters of locally advanced invasive ductal carcinoma (IDC) among patients of different racial groups and molecular subtypes. A notable increase in the SUV_max, MTV, and TLG values of the primary tumor was observed in black patients when compared with patients of other ethnicities. Moreover, the luminal subtype showed a significant increase in SUV_max, while both SUV_max and TLG values were significantly elevated in the basal subtype of the primary tumor. Overall, Black patients with IDC exhibited markedly elevated PET parameters, implying a more aggressive disease phenotype for this racial group, particularly with luminal and basal carcinoma subtypes.¹⁹

Some studies have proposed to simplify the T staging.^14,15 Our study aimed to compare metabolic parameters in patients of different T stages and to evaluate the differences in primary tumor aggressiveness by identifying the metabolic parameters associated with T staging. We observed significant differences in the values of primary tumor MTV, SUV_max, SUV_mean, TLG, GN_max, GN_mean, GNTLG, SUV_peak, SUL_max, and SUS_max between stages T1–T4. SUV_max, SUS_max, and GN_max only represent the maximum uptake within the VOI, while MTV, TLG, and GNTLG reflect not only the tumor size but also the metabolic information. In our study, we observed no significant differences in TMTV, TTLG, and TGNTLG values between the stages T1 and T2. Conversely, the pairwise comparison of the primary tumor GN_max, GN_mean, SUV_peak, and SUS_max between the four groups showed significant differences. Further analysis suggested that higher mean values of TGN_max, TGN_mean, TSUV_peak, and TSUS_max were associated with advanced T stages. The T staging of NPC is based entirely on the anatomical tumor extent rather than the size of the malignant tumors.¹⁶ The above results suggest that GN_max, GN_mean, SUV_peak, and SUS_max can be used to evaluate the local aggressiveness of the primary tumor and may be more effective indicators compared with volume-dependent parameters, such as MTV and TLG. Our results showed that the primary tumor metabolic parameter values were significantly different in patients with tumors of different T stages. In other words, our results suggested that T staging based on the Eighth Edition AJCC Cancer Staging Manual is credible from a metabolic perspective.

Our study further aimed to elucidate the relationship between PET-CT parameters and LNM. In our study, all parameters of the primary tumor and LNM, except TMTV, were correlated with N staging. We observed significant differences in all parameters of the primary tumor and LNM between stages N0 and N3. The pairwise comparison of all LNM parameters between the four stages showed significant differences. Furthermore, our analysis showed that higher mean values of LNMTV, LNSUV_max, LNSUV_mean, LNTLG, LNGN_max, LNGN_mean, LNGNTLG, LNSUV_peak, LNSUL_max, and LNSUS_max were associated with advanced N staging. Our findings suggest that metabolic information is a useful indicator in N staging of LNM based on the Eighth Edition AJCC Cancer Staging Manual. Two metabolic parameters (LNMTV and TSUS_max) exhibited a significant positive association with N staging. The primary tumor SUS_max was associated with LNM. Previous studies have also described the association between PET–CT parameters and LNM. For example, a study by Yilmaz et al²² showed that higher primary tumor SUV_max was associated with a higher probability of LNM in patients with cervical cancer. A study by Crivellaro et al²³ illustrated that primary tumor MTV and TLG were significantly related to the presence of LNM in early-stage cervical cancer. Li et al³² also proposed that in early-stage cervical cancer, primary tumor TLG may predict LNM. Husby et al²¹ demonstrated that SUV_max, SUV_mean, MTV, and TLG of the primary tumor tend to indicate LNM in endometrial carcinoma. In incidentally detected thyroid cancer, high MTV and TLG are associated with LNM.³³ A study unveiled an association between the retention index (calculated as the difference between SUV_max delayed point and SUV_max early point, divided by SUV_max early point) of the primary tumor and a heightened likelihood of LN metastasis in patients afflicted by non-small cell lung cancer (NSCLC).³⁴ Another study illustrated that the SUV_max of the mediastinal LNM and primary tumor SUV_max, SUV_peak, SUV_mean, MTV, and TLG were significantly associated with the presence of mediastinal LNM in NSCLC. In particular, the study found a strong correlation between the mediastinal LNM SUV_max, the primary tumor SUV_peak, and the occurrence of mediastinal LNM in patients with NSCLC.³⁵ PET–CT metabolic parameters have also been used to predict LNM in other tumors, such as vulvar cancer,³⁶ esophageal squamous cell carcinoma,³⁷ breast cancers,³⁸ rectal cancer,³⁹ and gastric cancer.⁴⁰ In addition to some similarities, differences between the findings of our study and those of previous studies were noted. We investigated not only the presence of LNM but also the staging of LNM. In addition, we investigated more metabolic parameters than previous studies. SUL_max, SUS_max, GN_max, and GN_mean have rarely been investigated. SUV is typically normalized to total body mass. However, because FDG accumulation in white adipose tissue (WAT) is minimal in the fasting state, SUV in tissues other than WAT tends to be higher in patients with obesity. SUV values of the target tissue in patients with higher body mass will be higher than those in leaner patients. Therefore, some studies have proposed that SUV should be normalized to LBM instead of total body mass.²⁶ Normalization to BSA also has been proposed.²⁷ Our study suggested that TSUL_max and TSUS_max were positively associated with T staging, while TSUS_max instead of SUV_max was positively associated with N staging. Few studies have investigated GN. Hence, evidence supporting that GN improves the treatment response monitoring or the prediction of outcomes compared with uncorrected SUVs is lacking.²⁸ Our study showed that higher mean values of TGN_max and TGN_mean rather than SUV_max and SUV_mean were associated with advanced T staging. Our study also suggested that glucose-normalized SUV_max and glucose-normalized SUV_mean reflect the metabolic parameter differences between T stages. Our study provides reference values for follow-up studies that investigate the utility of other metabolic parameters in addition to SUV_max (eg, SUL_max, SUS_max, GN_max and GN_mean) to evaluate tumor aggressiveness and LNM.

Currently, the treatment of NPC mainly involves radiotherapy combined with chemotherapy, and most patients do not need surgical resection of the nasopharyngeal primary tumor and LNM. Due to the lack of gross surgical specimens, the depth of tumor invasion and LNM could not be accurately judged from the pathological level. In fact, comprehensive imaging is generally used to stage NPC in clinical practice, but when the disease (primary or metastatic) is in a critical state, clinical staging based on imaging involves certain difficulties. A study by Feng et al⁴¹ showed that a model based on PET and MRI features (1 T2WI feature and 11 PET features) and metabolic parameters (primary tumor SUV_max and TLG) exhibited good diagnostic performance for predicting NPC staging in the testing set (AUC, 0.90). Our study retrospectively analyzed the association between various PET–CT metabolic parameters and T and N staging, which may provide novel directions for clinical work and prove helpful for clinicians.

Several limitations of this study should be acknowledged. First, some patients underwent a biopsy of the primary lesion before the PET examination, which may have affected the accuracy of the metabolic parameters measurements. Second, our measurement method for MTV and TLG is only one of the many measurement methods, and we only used a 42% SUV_max threshold. The relationship between metabolic parameters determined via more measurement methods and thresholds and the primary tumor and LNM requires further study. Third, outcome analysis was not performed. Follow-up studies with larger, homogeneous patient cohorts are planned to investigate this aspect.

Conclusion

We observed significant differences in certain primary tumor PET–CT metabolic parameters between different T-stage tumors. Further analysis showed that higher mean values of TGN_max, TGN_mean, TSUV_peak, and TSUS_max were associated with advanced T staging. Compared with volume-dependent parameters, such as MTV and TLG, TGN_max, TGN_mean, TSUV_peak, and TSUS_max may be more useful indicators of the local aggressiveness of tumors. Furthermore, our analysis suggests that higher mean values of all the LN metabolic parameters were associated with advanced N staging. Our results indicate that metabolic parameters are useful for T and N staging based on the Eighth Edition AJCC Cancer Staging Manual. LNMTV and TSUS_max were positively associated with N staging. The SUS_max of the primary tumor was associated with LNM. In addition to SUV_max, other metabolic parameters (eg, SUL_max, SUS_max, GN_max, and GN_mean) were indicated to be useful in the evaluation of tumor aggressiveness and LNM. With further improvement and validation, PET–CT metabolic parameters may become useful predictors of local tumor aggressiveness and LNM of NPC.

Footnotes

Acknowledgements

none

Declarations

ORCID iDs

Yun Zhang

Rong Huang

References

Chen

Y-P

Chan

ATC

Q-T

Blanchard

Sun

Nasopharyngeal carcinoma. Lancet. 2019;394:64-80. doi:10.1016/S0140-6736 (19)30956-0

Ferlay

Ervik

Lam

, et al. Global cancer observatory: cancer today. International Agency for Research on Cancer. Published December 2020. Accessed December 28, 2020. https://gco.iarc.fr/today

Amin

Edge

Greene

, et al. Eighth Edition AJCC Cancer Staging Manual. 8th ed. Springer; 2017.

Chiang

Guo

, et al. Prognostic factors for overall survival in nasopharyngeal cancer and implication for TNM staging by UICC: a systematic review of the literature. Front Oncol. 2021;11:703995. doi:10.3389/fonc.2021.703995

Luo

Yang

Qiu

, et al. A new T classification based on masticator space involvement in nasopharyngeal carcinoma: a study of 742 cases with magnetic resonance imaging. BMC Cancer. 2014;14:653. doi:10.1186/1471-2407-14-653

Pan

Zong

, et al. Proposal for the 8th edition of the AJCC/UICC staging system for nasopharyngeal cancer in the era of intensity-modulated radiotherapy. Cancer. 2016;122:546-558. doi:10.1002/cncr.29795

Huang

Chen

Zhu

Pan

XB.

Maximal lymph nodal diameter on N stage of nasopharyngeal carcinoma. Medicine (Baltimore). 2021;100:e26543. doi:10.1097/MD.0000000000026543

Yan

Chen

, et al. Comparing the 7th and 8th editions of UICC/AJCC staging system for nasopharyngeal carcinoma in the IMRT era. BMC Cancer. 2021;21:327. doi:10.1186/s12885-021-08036-8

Pan

Chen

Liang

Zhu

XD.

Validation of stage N3 of the eighth edition AJCC staging system for nasopharyngeal carcinoma. Laryngoscope. 2021;131:535-540. doi:10.1002/lary.28825

10.

Huang

Quan

Zhao

, et al. MRI of nasopharyngeal carcinoma: parapharyngeal subspace involvement has prognostic value and influences T-staging in the IMRT era. Eur Radiol. 2022;32:262-271. doi:10.1007/s00330-021-08113-3

11.

Luo

Huang

, et al. Value of skull base invasion subclassification in nasopharyngeal carcinoma: implication for prognostic stratification and use of induction chemotherapy. Eur Radiol. 2022;32:7767-7777. doi:10.1007/s00330-022-08864-7

12.

Wang

Zhu

, et al. Extent of paranasal sinus involvement and its prognostic value in nasopharyngeal carcinoma: proposed modification in the current UICC/AJCC staging system. Radiother Oncol. 2021;160:221-227. doi:10.1016/j.radonc.2021.04.024

13.

Zhao

Zhou

Shen

Paranasal sinus invasion should be classified as T4 disease in advanced nasopharyngeal carcinoma patients receiving radiotherapy. Front Oncol. 2020;10:01465. doi:10.3389/fonc.2020.01465

14.

Pan

Tong

Chen

, et al. A simplified T classification based on the 8th edition of the UICC/AJCC staging system for nasopharyngeal carcinoma. Cancer Manag Res. 2019;11:3163-3169. doi:10.2147/CMAR.S185860

15.

Tang

Liang

Huang

, et al. The development and external validation of simplified T category classification for nasopharyngeal carcinoma to improve the prognostic value in the intensity-modulated radiotherapy era. Cancer Med. 2019;8:2213-2222. doi:10.1002/cam4.2131

16.

Tang

Chen

, et al. The Chinese Society of Clinical Oncology (CSCO) clinical guidelines for the diagnosis and treatment of nasopharyngeal carcinoma. Cancer Commun (Lond). 2021;41:1195-1227. doi:10.1002/cac2.12218

17.

Colevas

Yom

Pfister

, et al. NCCN guidelines insights: head and neck cancers, version 1.2018. J Natl Compr Canc Netw. 2018;16:479-490. doi:10.6004/jnccn.2018.0026

18.

Oriuchi

Sugawara

Shiga

Positron emission tomography for response evaluation in microenvironment-targeted anti-cancer therapy. Biomedicines. 2020;8:371. doi:10.3390/biomedicines8090371

19.

Abubakar

Tag

Olabinjo

Isah

Lawal

Differences in tumour aggressiveness based on molecular subtype and race measured by [¹⁸F]FDG PET metabolic metrics in patients with invasive carcinoma of the breast. Diagnostics (Basel). 2023;13:2059. doi:10.3390/diagnostics13122059

20.

Bezzi

Bergamini

Mathoux

, et al. Role of machine learning (ML)-based classification using conventional 18F-FDG PET parameters in predicting postsurgical features of endometrial cancer aggressiveness. Cancers (Basel). 2023;15:325. doi:10.3390/cancers15010325

21.

Husby

Reitan

Biermann

, et al. Metabolic tumor volume on 18F-FDG PET/CT improves preoperative identification of high-risk endometrial carcinoma patients. J Nucl Med. 2015;56:1191-1198. doi:10.2967/jnumed.115.159913

22.

Yilmaz

Adli

Celen

Zincirkeser

Dirier

FDG PET-CT in cervical cancer: relationship between primary tumor FDG uptake and metastatic potential. Nucl Med Commun. 2010;31:526-531. doi:10.1097/MNM.0b013e32833800e7

23.

Crivellaro

Signorelli

Guerra

, et al. 18F-FDG PET/CT can predict nodal metastases but not recurrence in early stage uterine cervical cancer. Gynecol Oncol. 2012;127:131-135. doi:10.1016/j.ygyno.2012.06.041

24.

Qiu

, et al. Reflecting on the utility of standardized uptake values on (18)F-FDG PET in nasopharyngeal carcinoma. BMC Cancer. 2022;22:495. doi:10.1186/s12885-022-09626-w

25.

Larson

Erdi

Akhurst

, et al. Tumor treatment response based on visual and quantitative changes in global tumor glycolysis using PET-FDG imaging. Clin Positron Imaging. 1999;2:159-171. doi:10.1016/s1095-0397(99)00016-3

26.

Chen

Cypess

Laughlin

, et al. Brown adipose reporting criteria in imaging studies (BARCIST 1.0): recommendations for standardized FDG-PET/CT experiments in humans. Cell Metab. 2016;24:210-222. doi:10.1016/j.cmet.2016.07.014

27.

Wahl

Jacene

Kasamon

Lodge

MA.

From RECIST to PERCIST: evolving considerations for PET response criteria in solid tumors. J Nucl Med. 2009;50:122S-150S. doi:10.2967/jnumed.108.057307

28.

Boellaard

Delgado-Bolton

Oyen

, et al. FDG PET/CT: EANM procedure guidelines for tumour imaging: version 2.0. Eur J Nucl Med Mol Imaging. 2015;42:328-354. doi:10.1007/s00259-014-2961-x

29.

Burger

Vargas

Beattie

, et al. How to assess background activity: introducing a histogram-based analysis as a first step for accurate one-step PET quantification. Nucl Med Commun. 2014;35:316-324. doi:10.1097/MNM.0000000000000045

30.

Wang

Zhao

Liu

Yang

Zhu

Comparison of different automated lesion delineation methods for metabolic tumor volume of 18F-FDG PET/CT in patients with stage I lung adenocarcinoma. Medicine (Baltimore). 2017;96:e9365. doi:10.1097/MD.0000000000009365

31.

Sher

Lacoeuille

Fosse

, et al. For avid glucose tumors, the SUV peak is the most reliable parameter for [(18)F]FDG-PET/CT quantification, regardless of acquisition time. EJNMMI Res. 2016;6:21. doi:10.1186/s13550-016-0177-8

32.

Sun

Guo

Combinative evaluation of primary tumor and lymph nodes in predicting pelvic lymphatic metastasis in early-stage cervical cancer: a multiparametric PET-CT study. Eur J Radiol. 2019;113:153-157. doi:10.1016/j.ejrad.2019.02.013

33.

Kim

, et al. High metabolic tumor volume and total lesion glycolysis are associated with lateral lymph node metastasis in patients with incidentally detected thyroid carcinoma. Ann Nucl Med. 2015;29:721-729. doi:10.1007/s12149-015-0994-2

34.

Ema

Kojima

Mizuno

, et al. Retention index of FDG-PET/CT SUVmax of the primary tumor in non-small cell lung cancer as a predictor of lymph node metastasis: a retrospective study. Eur J Hybrid Imaging. 2022;6:21. doi:10.1186/s41824-022-00141-6

35.

Liao

Liu

, et al. The value on SUV-derived parameters assessed on 18F-FDG PET/CT for predicting mediastinal lymph node metastasis in non-small cell lung cancer. BMC Med Imaging. 2023;23:49. doi:10.1186/s12880-023-01004-7

36.

Rufini

Garganese

Ieria

, et al. Diagnostic performance of preoperative [18F]FDG-PET/CT for lymph node staging in vulvar cancer: a large single-centre study. Eur J Nucl Med Mol Imaging. 2021;48:3303-3314. doi:10.1007/s00259-021-05257-8

37.

Liao

Wei

Zhang

, et al. Modified method to improve the diagnostic efficiency of 18F-FDG PET/CT in regional lymph node metastasis of esophageal squamous cell carcinoma. Ann Transl Med. 2021;9:1549-1549. doi:10.21037/atm-21-4926

38.

Zhang

Shi

Xiao

, et al. Early SUVmax is the best predictor of axillary lymph node metastasis in stage III breast cancers. Quant Imaging Med Surg. 2021;11:1680-1691. doi:10.21037/qims-20-423

39.

Yukimoto

Uemura

Tsuboyama

, et al. Efficacy of positron emission tomography in diagnosis of lateral lymph node metastases in patients with rectal cancer: a retrospective study. BMC Cancer. 2021;21:520. doi:10.1186/s12885-021-08278-6

40.

Zhang

Shao

PET/CT for predicting occult lymph node metastasis in gastric cancer. Curr Oncol. 2022;29:6523-6539. doi:10.3390/curroncol29090513

41.

Feng

Liang

Wang

Ding

A diagnosis model in nasopharyngeal carcinoma based on PET/MRI radiomics and semiquantitative parameters. BMC Med Imaging. 2022;22:150. doi:10.1186/s12880-022-00883-6

The Utility of 18 F-FDG-PET–CT Metabolic Parameters in Evaluating the Primary Tumor Aggressiveness and Lymph Node Metastasis of Nasopharyngeal Carcinoma

Abstract

Background:

Methods:

Results:

Conclusions:

Keywords

Introduction

Patients and Methods

Patients

PET–CT imaging

PET–CT image analysis

Statistical analysis

Results

Clinical characteristics

Comparison of PET parameters between stage T1–T4

Comparison of PET parameters between stages N0 and N3

Correlation between PET parameters, and T and N staging

Relationship between PET parameters, and T and N staging

Discussion

Conclusion

Footnotes

Acknowledgements

Declarations

ORCID iDs

References