Endocrine DOI 10.1007/s12020-015-0620-z
ORIGINAL ARTICLE
Predictive value of intratumoral heterogeneity of F-18 FDG uptake for characterization of thyroid nodules according to Bethesda categories of fine needle aspiration biopsy results Seong-Jang Kim1,2 • Samuel Chang3
Received: 20 February 2015 / Accepted: 27 April 2015 Ó Springer Science+Business Media New York 2015
Abstract The current study was aimed to investigate the clinical value of intratumoral heterogeneity of F-18 FDG uptake for characterization of thyroid nodule (TN) with inconclusive fine-needle aspiration biopsy (FNAB) results. The current study enrolled 200 patients who showed F-18 FDG incidentaloma and were performed FNAB. The intratumoral heterogeneity of F-18 FDG uptake was represented as the heterogeneity factor (HF), defined as the derivative (dV/dT) of a volume-threshold function for a primary tumor. The diagnostic and predictive values of HF and F-18 FDG PET/CT parameters were evaluated for characterization of inconclusive FNAB results. Among F-18 FDG PET/CT parameters, SUVmax, MTV, and TLG of malignant group were statistically higher than those of Bethesda category of suspicious malignant group. However, HF values were not statistically different between the groups of Bethesda categories (Kruskal–Wallis statistics, 9.924; p = 0.0774). In ROC analysis, when HF [ 2.751 was used as cut-off value, the sensitivity and specificity for prediction of malignant TN were 100 % (95 % CI 69.2–100 %) and 60 % (95 % CI 42.1–76.1 %), respectively. The AUC was 0.826 (95 % CI 0.684–0.922) and standard error was 0.0648 (p \ 0.0001). In conclusion, the intratumoral heterogeneity of F-18 FDG uptake represented
& Seong-Jang Kim [email protected]; [email protected] 1
Department of Nuclear Medicine, Pusan National University Hospital, 179 Gudeok-Ro, Seo-Gu, Busan 602-739, Republic of Korea
2
Biomedical Research Institute, Pusan National University Hospital, Busan 602-739, Republic of Korea
3
Department of Radiology, University of Colorado School of Medicine, 12401, 17th Ave, Aurora, CO 80045, USA
by HF could be a predictor for characterization of TN with inconclusive FNAB results. Additional large populationbased prospective studies are needed to validate the diagnostic utility of HF of F-18 FDG PET/CT. Keywords F-18 FDG PET/CT Thyroid nodule Inconclusive FNAB
Introduction Thyroid nodules (TN) are common, with up to 7 % of adults demonstrating clinically palpable and detectable up to 70 % of adults by ultrasonography [1, 2]. The majority of TN are benign, but approximately, 5–15 % of all palpable TN are malignant [3, 4]. Fine-needle aspiration biopsy (FNAB) is a safe, accurate, and cost-effective method and plays essential role in the evaluation of patients with TN [4–6]. Although FNAB is the most reliable and sensitive method to detect malignancy, 15–30 % of patients remain inconclusive or indeterminate findings on cytopathology [3, 7]. With use of Bethesda system for FNAB classification, indeterminate findings have been subdivided into three categories: follicular lesion of undetermined significance or atypia of undetermined significance, follicular neoplasm or lesion suspicious for follicular neoplasm, and lesion suspicious malignancy. Because this category is associated with 20 % of incidence of malignancy, most patients referred for surgery for diagnostic rather than therapeutic purposes [1, 8, 9]. Malignant cancer cells are composed of heterogenous components not only biologic constituents but also gene expression, metabolic, and behavioral characteristics [10–12]. Heterogeneity varies in the same cancer and has a wide spectrum even in the same stage because there are differences
123
Endocrine
in properties as the growth rate, vascularity, and necrosis within the same tumor cell population [13]. Recently, there has been increasing interest in assessment of intratumoral heterogeneity of F-18 FDG uptake demonstrating an association of prognosis of patients [14–16]. In the current study, we hypothesized that malignant TN would be more heterogenous and intratumoral heterogeneity could be used for characterization of TN. The current study was aimed to investigate the clinical value of intratumoral heterogeneity of F-18 FDG uptake for characterization of TN with inconclusive FNAB results.
Methods Patients From November 2012 to July 2014, among 23,462 subjects (15,015 men, 8447 women) who underwent F-18 FDG PET/CT for known or suspected cancer and health checkup in our institution were enrolled. Among these population, 493 patients (2.1 %) who showed focal thyroid F-18 FDG incidentaloma were referred for FNAB of TN. The current study included 200 patients who were confirmed cytological diagnosis by FNAB. Remaining 293 patients who did not perform FNAB were excluded from the current study. The reasons were as follows: transfer to other hospital (n = 172), follow-up loss (n = 53), refuse of FNAB (n = 36), and US follow-up (n = 32). Our institutional review board approved this retrospective study, and the requirement to obtain informed consent was waived. Ultrasonography-guided fine-needle aspiration biopsy FNAB was performed with a 21-gauge needle on a 10-ml syringe. Aspiration was done twice for each nodule. Specimens were smeared on slides, air dried, and stained using the Papanicolaou method. Typically, the procedure was performed by an experienced endocrinologist, and the cytological diagnosis was made by an experienced pathologist. In the current study, we defined inconclusive FNAB results as follows: Bethesda categories of non-diagnostic, AUS, follicular neoplasm, and suspicious malignancy.
FDG (5 MBq/kg of body weight). The emission scan time per bed position was 3 min, and six bed positions were acquired. PET data were obtained using a high-resolution whole-body scanner with an axial field of view of 21.6 cm. The average axial resolution varied between 2.0 mm at fullwidth at half-maximum in the center and 2.4 mm at 28 cm. The average total PET/CT examination time was 20 min. Measurement of volumetric parameters of F-18 FDG PET/CT The SUVmax was obtained using the following formula: SUVmax = maximum activity in the region of interest (MBq/gram)/(injected dose [MBq]/body weight [grams]). The metabolic tumor volume (MTV) was determined as the total number of voxels with threshold SUV of C40 % of the SUVmax in the volume of interest (VOI). The total lesion glycolysis (TLG) was calculated as the MTV multiplied by its SUVmean. Determination of intratumoral heterogeneity of F-18 FDG uptake The intratumoral metabolic heterogeneity was represented by the heterogeneity factor (HF) [17], which was determined for each patient as follows: a region of interest (ROI) was manually drawn to fully include the primary tumor and a surrounding region of normal tissue (normal background). The tumor volume was determined with a series of SUV thresholds (e.g., 40, 50, 60, 70, and 80 % of SUVmax) using a Syngo Multimodality Workspace (Siemens AG, Berlin, Germany). We excluded the values of \40 % from the heterogeneity analysis because a previous study reported that the minimal threshold that represents the actual tumor volume was 40 % and the values of\40 % included too much normal tissue background activity [18]. In addition, values of[80 % were also excluded because the volumes were small and the partial volume effect was pronounced [19]. A volumethreshold function of the tumor was acquired by plotting thresholds to volumes. Linear regression analysis was performed, and the HF was calculated by finding the derivative (dV/dT) of the volume-threshold function for each tumor. Next, the values of the HF were modified into absolute values so that all resulting values were positive; the more positive the factor, the more heterogeneous the tumor.
F-18 FDG PET/CT Statistical analysis F-18 FDG PET/CT images were obtained using a dedicated PET/CT scanner (Biograph 40; Siemens, Knoxville, Tennessee, USA). Standard patient preparation included a fasting at least 8 h and a serum glucose level lower than 120 mg/dl before F-18 FDG administration. F-18 FDG PET/ CT imaging was performed 60 min after injection of F-18
123
Statistical analyses were performed using commercially available software (MedCalc 14.12.0, Mariakerke, Belgium). Continuous data are expressed mean ± SD for normally distributed values and median (range) for nonparametric values (skewed data). An independent 2-sample
Endocrine
t test was used to compare the various parameters of F-18 FDG PET/CT between malignant and benign TN. To compare the F-18 FDG PET/CT parameters according to the Bethesda categories, Kruskal–Wallis test was used with appropriate Tukey post hoc test. Receiver operating characteristic (ROC) curves for each parameters for the prediction of malignant TN were derived and evaluated by comparing area under curve (AUC). The sensitivity and specificity of each parameter were determined at the optimal cut-off values. Statistical significance was defined as p \ 0.05.
Results Characteristics of patients Table 1 shows characteristics of patients included in the current study. The results of FNAB were 39 malignant (Category 6), 13 suspicious malignant (Category 5), 6 follicular neoplasm (Category 4), 12 atypia of undetermined significance (AUS, Category 3), 116 benign (Category 2), and 14 non-diagnostic (Category 1). The patients with non-diagnostic FNAB results underwent repeated FNAB. Repeated FNAB showed 2 malignant and 12 benign TN. The current study treated the first FNAB results in terms of inconclusive FNAB. All 72 patients with FNAB Table 1 Characteristics of patients Thyroid nodules Variables
Malignant (n = 49)
Benign (n = 151)
p value
Age
49.4 ± 10.1
51.8 ± 10.9
0.1654
Size
1.9 ± 1.1
2.1 ± 1.3
0.2144 0.0273
Sex Male
12
31
37
120
Category 1
2
12
Category 2
0
116
Category 3
1
11
Category 4
1
5
Category 5
6
7
Category 6
39
0
Female
F-18 FDG PET/CT parameters 5.96 ± 2.61
5.52 ± 3.18
Comparison of F-18 FDG PET/CT parameters according to Bethesda category Table 2 show the differences of SUVmax, SUVmean, MTV, and TLG according to the Bethesda category of FNAB results. Among the F-18 FDG PET/CT parameters, SUVmean was not statistically different between the groups of Bethesda categories. However, SUVmax, MTV, and TLG of Category 6 were statistically higher than those of Bethesda category 5. Comparison of HF according to Bethesda category Figure 1 shows differences of HF according to the Bethesda categories of FNAB results. The HF values were not statistically different between the groups of Bethesda categories (Kruskal–Wallis statistics, 9.924; p = 0.0774). Comparison of F-18 FDG PET/CT parameters Figure 2 shows the differences of parameters and HF of F-18 FDG PET/CT between benign, inconclusive, and malignant groups of FNAB results. All of these parameters were not statistically different between 3 groups (SUVmax; KW statistics, 3.555, p = 0.0624: SUVmean; KW statistics, 0.261, p = 0.7713: MTV; KW statistics, 1.455, p = 0.2358: TLG; KW statistics, 0.3873, p = 0.6794: HF; KW statistics, 2.634, p = 0.0743). Characterization of TN using F-18 FDG PET/CT parameters
Bethesda category
SUVmax
results of follicular neoplasm, AUS, malignant, and suspicious malignant underwent total thyroidectomy or lobectomy and final histologic diagnosis was surgically confirmed. The malignant pathologic results were 2 follicular carcinoma and 47 papillary thyroid cancers. Among the semiquantitative parameters, HF of malignant TN was higher than benign one.
0.3862
SUVmean
2.54 ± 0.8
2.53 ± 1.16
0.9548
MTV (ml)
5.76 ± 2.0
5 ± 2.13
0.5031
TLG
16.01 ± 6.9
15.27 ± 8.82
0.8655
HF
3.2019 ± 0.948
2.8065 ± 0.782
0.0040
FNAB fine-needle aspiration biopsy, AUS atypia of undetermined significance, FN follicular neoplasm
Table 3 shows the ROC analysis results for prediction of malignant TN. Among the various parameters, only HF could be used for differentiation of TN. When HF [ 3.027 was used as cut-off value, the sensitivity and specificity for prediction of malignant TN were 59.2 % (95 % CI 44.2–73 %) and 65.6 % (95 % CI 57.4–73.1 %), respectively. The AUC was 0.606 (95 % CI 0.535–0.675) and standard error was 0.0464 (p = 0.0217). Prediction of malignant TN with inconclusive FNAB Figure 3 demonstrates the ROC results of F-18 FDG PET/ CT parameters for prediction of malignant TN with
123
Tukey post hoc test showed higher value than that of Bethesda category of SM
Fig. 1 The differences of HF according to the Bethesda categories of FNAB results. The HF values were not statistically different between the groups of Bethesda categories (Kruskal–Wallis statistics, 9.924; p = 0.0774)
a
Data are expressed as median values with ranges
KW statistics Kruskal–Wallis statistics, SUVmax maximal standardized uptake value, SUVmean mean standardized uptake value, MTV metabolic tumor volume, TLG total lesion glycolysis, HF heterogeneity factor
0.0774
0.0121 14.62
3.095 (0.868–6.365) 3.354 (2.130–3.765) 2.738 (1.976–3.255) 2.553 (1.627–3.886) 2.825 (1.889–3.561) HF
2.876 (0.235–5.121)
11.16a (0.9–61.57) 3.36 (0.56–10.45) 12.21 (2.93–33.01) 3.77 (1.13–34.38) TLG
5.83 (0.6–146.23)
5.05 (0.97–211.74)
9.924
0.1849
0.0105 14.97 3.8a (0.4–30.5) 1.61 (0.33–4.8) 1.7 (0.47–6.47) MTV (ml)
2.25 (1.2–5.6) SUVmean
2.71 (0.3–33.85)
4.57 (1.54–7.86)
0.0092
2.3 (0.57–46.03)
7.517 2.5 (1.4–4.7) 2 (1.4–2.8) 2.65 (1.9–4.2) 2.1 (1.4–9.4)
15.28 5.1a (2.7–13.2) 4.2 (2.3–5.3) 5.2 (4.6–9.5) 4.45 (2.4–28) 4.35 (2.6–8.3) SUVmax
6 (n = 39) 5 (n = 13) 4 (n = 6) 3 (n = 12) 2 (n = 116) 1 (n = 14)
4.75 (2.3–8.3)
KW statistics Bethesda categories Variables
Table 2 Comparison of F-18 FDG PET/CT parameters according to Bethesda categories of FNAB results of thyroid nodules
123
2.2 (1.1–8.6)
p value
Endocrine
inconclusive FNAB results. Among the parameters, only the HF could predict malignant TN. When HF [ 2.751 was used as cut-off value, the sensitivity and specificity for prediction of malignant TN were 100 % (95 % CI 69.2–100 %) and 60 % (95 % CI 42.1–76.1 %), respectively. The AUC was 0.826 (95 % CI 0.684–0.922) and standard error was 0.0648 (p \ 0.0001).
Discussion The current study showed that HF could be used for characterization of TN. Also, in patients with inconclusive FNAB results, the HF could be used for prediction of malignant TN. Thyroid cancer is the most common endocrine malignancy. It represents approximately 1 % of all cancers, corresponding to up to 10.2 per 100,000 individuals in the United States with increasing incidence over the last decades [20]. Early detection and diagnosis is important in appropriate treatment of thyroid cancer, as delays in the diagnosis are associated with increased mortality [21]. The evaluation of TN includes a FNAB and measurement of serum thyroid-stimulating hormone. FNAB is the most important diagnostic test in the initial evaluation of a patient with TN and has high diagnostic accuracy. However, a number of reports have shown that 11–42 % of FNAB results are reported as indeterminate [22, 23]. These indeterminate TN present a clinical dilemma to the
Endocrine
Fig. 2 Comparison of parameters and HF between benign, inconclusive, and malignant groups of FNAB results. All of these parameters were not statistically different between 3 groups (SUVmax; KW
statistics, 3.555, p = 0.0624: SUVmean; KW statistics, 0.261, p = 0.7713: MTV; KW statistics, 1.455, p = 0.2358: TLG; KW statistics, 0.3873, p = 0.6794: HF; KW statistics, 2.634, p = 0.0743)
Table 3 ROC results for characterization of TN using F-18 FDG PET/CT parameters Variables
Cut-off
Sensitivity (%)
95 % CI
Specificity (%)
95 % CI
AUC
95 % CI
SE
p
SUVmax
[4.5
65.3
50.4–78.3 %
52.3
44–60.5 %
0.586
0.514–0.655
0.0463
0.0646
SUVmean
[2.5
42.9
28.8–57.8 %
69.5
61.5–76.8 %
0.549
0.478–0.620
0.0469
0.2915
MTV
[3.39
57.1
42.2–71.2 %
66.9
58.8–74.3 %
0.566
0.494–0.636
0.0516
0.2004
TLG
[7.98
57.1
42.2–71.2 %
63.6
55.4–71.2 %
0.562
0.490–0.632
0.0514
0.2286
HF
[3.027
59.2
44.2–73 %
65.6
57.4–73.1 %
0.606
0.535–0.675
0.0464
0.0217
95 % CI confidence interval, AUC area under curve, SE standard error
clinicians. Therefore, the next step in the current management of these patients with TN is usually surgical excision. At present, there is no alternative algorithm for a more conservative management of patients with TN of indeterminate results. Several methods had been used to characterize of such indeterminate FNAB results of TN. It has been reported that certain sonographic features of a TN are associated with an increased likelihood of malignancy. Recently, some studies have attempted to characterize the indeterminate TN using ultrasonography [24, 25]. Khoncarly et al. concluded that with the exception of nodule height greater than width, sonographic criteria were not helpful in deciding with patients with AFLUS should undergo thyroidectomy [24]. Batawil et al. also concluded that based on the limited accuracy or predictive value of ultrasonographic risk factors, surgery is the treatment of choice for indeterminate TN [25]. The use of molecular tests as an adjunct to diagnosis in indeterminate TN is standard care in the United States. A
167-gene expression classifier is used to minimize unnecessary diagnostic thyroid surgery, and another mutation marker panel is used to select patients for initial total thyroidectomy, thereby saving on the two-step surgery [26, 27]. Although molecular tests are useful in defining the natures of TN, these tests are not available at all centers, their cost is high, and greater experience is needed, particularly in the case of indeterminate TN. A recent costeffectiveness analytic study showed that full implementation of F-18 FDG PET/CT in the workup of adult patients with TN scheduled for surgery for FNAB-indeterminacy could lead to a decrease in costs and a moderate increase in health related quality of life compared with diagnostic surgery in all patients according to current European practice and in competitive to the current US standard of the gene expression classifier [28]. The use of F-18 FDG PET/CT in the management of thyroid disease has been limited primarily to the postoperative surveillance of patients with known differentiated thyroid carcinoma. Several studies have reported the role of
123
Endocrine
Fig. 3 ROC analysis of F-18 FDG PET/CT parameters for prediction of malignant TN with inconclusive FNAB results
F-18 FDG PET/CT in assessing cytological indeterminate TN with conflicting results [29–31]. Also, a recent pilot study showed that F-18 FDG PET/CT was not a predictable indicator of the presence of thyroid cancer [32]. To predict the correct diagnostic value of F-18 FDG PET/CT for characterization of cytological indeterminate TN, two meta-analytic studies were conducted [33, 34]. Wang et al. [33] conducted meta-analysis including 7 studies (267 patients). In their analysis, the pooled sensitivity of PET or PET/CT for the detection of thyroid cancer was 89.0 % (95 % CI 79.0–95.0 %). The pooled specificity was 55.0 % (95 CI 48.0 %–62.0 %). Vriens et al. [34] concluded that in patients with TN who have indeterminate FNAB results, a negative F-18 FDG PET scan improves diagnostic accuracy, particularly for patients who have lesions [15 mm. All false-negative F-18 FDG PET results were from lesions \15 mm. The current study investigated the diagnostic values of volumetric parameters and HF of F-18 FDG PET/CT for characterization of TN with inconclusive FNAB results. In the current study, only the HF could predict malignant TN in patients with inconclusive FNAB results. Also, in all TN, only the HF could differentiate the malignant and benign TN. Some previous studies have investigated the prognostic value of HF in patients with oral cavity cancer and breast cancer [16, 35]. Kwon SH et al. concluded that the intratumoral heterogeneity of F-18 FDG uptake may be a
123
significant prognostic factor for overall survival in addition to cervical lymph node metastasis in oral cavity cancer [35]. In breast cancer patients, intratumoral metabolic heterogeneity correlated closely with the MTV and significantly affected the OS in IDC patients. Therefore, they concluded the HF may act as a robust surrogate marker for the prediction of OS in IDC patients [16]. The current study adapted the HF for characterization of TN with inconclusive FNAB results and showed the possibility of clinical adaptation of HF in clinical situation of F-18 FDG incidentaloma. The current study has some limitations. First, it was a retrospective, single-center study. Second, the intratumoral metabolic heterogeneity on F-18 FDG PET scans can be represented by various methods. The parameter of intratumoral metabolic heterogeneity is not well standardized, although histologic studies have shown good spatial correlation between areas of F-18 FDG uptake and histologic findings. Furthermore, in nodular goiter, the heterogeneity might be notable because of necrosis, inflammation, and hemorrhage. A feasible and highly reproducible method for obtaining a heterogeneity parameter representing intratumoral metabolic heterogeneity is warranted. Third, the population of inconclusive FNAB results is relatively small. Finally, VOI selection is important for determining tumor MTV. Several different methods of VOI selection could be used: manually defined VOI, isocontour VOIs based on a fixed percentage of the SUVmax, and isocontour
Endocrine
VOIs based on a fixed percentage of SUVmax threshold, and generally 40 % of SUVmax is accepted for VOI selection.
Conclusion In conclusion, the intratumoral heterogeneity of F-18 FDG uptake represented by HF could be a predictor for characterization of TN with inconclusive FNAB results. Additional large population-based prospective studies are needed to validate the diagnostic utility of HF of F-18 FDG PET/CT.
Conflict of interest The authors declared that there is no conflict of interests.
References 1. C. Hambleton, E. Kandil, Appropriate and accurate diagnosis of thyroid nodule: a review of thyroid fine-needle aspiration. Int. J. Clin. Exp. Med. 6, 413–422 (2013) 2. M. Goldfarb, S. Gondek, C. Soloranzo, J.I. Lew, Surgeon-performed ultrasound can predict benignity in thyroid nodules. Surgery 50, 436–441 (2011) 3. D.S. Cooper, G.M. Doherty, B.R. Haugen, R.T. Kloos, S.L. Lee, S.J. Mandel, E.L. Mazzaferri, B. McIver, F. Pacini, M. Schlumberger, S.I. Sherman, D.L. Steward, R.M. Tuttle, Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 19, 1167–1214 (2009) 4. L. Hegedu¨s, Clinical practice. The thyroid nodule. N. Engl. J. Med. 351, 1764–1771 (2004) 5. J.R. Porterfield Jr, C.S. Grant, D.S. Dean, G.B. Thompson, D.R. Farley, M.L. Richards, C.C. Reading, J.W. Charboneau, B.K. Vollrath, T.J. Sebo, Reliability of benign fine needle aspiration cytology of large thyroid nodules. Surgery 144, 963–968 (2008) 6. J.B. Ogilvie, E.J. Piatigorsky, O.H. Clark, Current status of fine needle aspiration for thyroid nodules. Adv. Surg. 40, 223–238 (2006) 7. E.S. Cibas, S.Z. Ali, NCI Thyroid FNA State of the Science Conference, The Bethesda system for reporting thyroid cytopathology. Am. J. Clin. Pathol. 132, 658–665 (2009) 8. S. Rorive, N. D’Haene, C. Fossion, I. Delpierre, N. Abarguia, F. Avni, C. Decaestecker, I. Salmon, Ultrasound-guided fine-needle aspiration of thyroid nodules: stratification of malignancy risk using follicular proliferation grading, clinical and ultrasonographic features. Eur. J. Endocrinol. 162, 1107–1115 (2010) 9. L.J. Layfield, E.S. Cibas, H. Gharib, S.J. Mandel, Thyroid aspiration cytology: current status. CA Cancer J. Clin. 59, 99–110 (2009) 10. G.H. Heppner, Tumor heterogeneity. Cancer Res. 44, 2259–2265 (1984) 11. R.J. Gillies, P.A. Schornack, T.W. Secomb, N. Raghunand, Causes and effects of heterogeneous perfusion in tumors. Neoplasia 1, 197–207 (1999) 12. M.E. Lleonart, P. Martin-Duque, R. Sanchez-Prieto, A. Moreno, S. Cajal, Tumor heterogeneity: morphological, molecular and clinical implications. Histol. Histopathol. 15, 881–889 (2000)
13. B. Huang, T. Chan, D.L. Kwong, W.K. Chan, P.L. Khong, Nasopharyngeal carcinoma: investigation of intratumoral heterogeneity with FDG PET/CT. AJR Am. J. Roentgenol. 199, 169–174 (2012) 14. F. Tixier, C.C. Le Rest, M. Hatt, N. Albarghach, O. Pradier, Intratumor heterogeneity characterized by textural features on baseline 18F-FDG PET images predict response to concomitant radiochemotherapy in esophageal cancer. J. Nucl. Med. 52, 369–378 (2011) 15. G.J. Cook, C. Yip, M. Siddique, V. Goh, S. Chicklore, A. Roy, P. Marsden, S. Ahmad, D. Landau, Are pretreatment 18F-FDG PET tumor textural features in non-small cell lung cancer associated with response and survival after chemoradiotherapy? J. Nucl. Med. 54, 19–26 (2013) 16. S.H. Son, D.H. Kim, C.M. Hong, C.Y. Kim, S.Y. Jeong, S.W. Lee, J. Lee, B.C. Ahn, Prognostic implication of intratumoral metabolic heterogeneity in invasive ductal carcinoma of the breast. BMC Cancer 14, 585–595 (2014) 17. E.A. Kidd, P.W. Grigsby, Intratumoral metabolic heterogeneity of cervical cancer. Clin. Cancer Res. 14, 5236–5241 (2008) 18. T.R. Miller, P.W. Grigsby, Measurement of tumor volume by PET to evaluate prognosis in patients with advanced cervical cancer treated by radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 53, 353–359 (2002) 19. M. Soret, S.L. Bachrach, I. Buvat, Partial volume effect in PET tumor imaging. J. Nucl. Med. 48, 932–945 (2007) 20. R. Siegel, D. Naishadham, A. Jemal, Cancer statistics, 2012. CA Cancer J. Clin. 62, 10–29 (2012) 21. J.A. Sipos, E.L. Mazzaferri, Thyroid cancer epidemiology and prognostic variables. Clin. Oncol. (R. Coll. Radiol.) 22, 395–404 (2010) 22. B. Miller, S. Burkey, G. Lindberg, W.H. Snyder 3rd, F.E. Nwariaku, Prevalence of malignancy within cytologically indeterminate thyroid nodules. Am. J. Surg. 188, 459–462 (2004) 23. G.M. Scabas, G.A. Staerkel, S.E. Shapiro, Fine-needle aspiration of the thyroid and correlation with hispathology in a contemporary series of 240 patients. Am. J. Surg. 186, 702–710 (2003) 24. S.M. Khoncarly, S.W. Tamarkin, C.R. McHenry, Can ultrasound be used to predict malignancy in patients with a thyroid nodule and an indeterminate fine-needle aspiration biopsy ? Surgery 156, 967–971 (2014) 25. N. Batawil, T. Alkordy, Ultrasonographic features associated with malignancy in cytologically indeterminate thyroid nodules. EJSO 40, 182–186 (2014) 26. H. Li, K.A. Robinson, B. Anton, I.J. Saldanha, P.W. Ladenson, Cost-effectiveness of a novel molecular test for cytologically indeterminate thyroid nodules. J. Clin. Endocrinol. Metab. 96, E1719–E1726 (2011) 27. Y.E. Nikiforov, N.P. Ohori, S.P. Hodak, S.E. Carty, S.O. LeBeau, R.L. Ferris, L. Yip, R.R. Seethala, M.E. Tublin, M.T. Stang, C. Coyne, J.T. Johnson, A.F. Stewart, M.N. Nikiforova, Impact of mutational testing on the diagnosis and management of patients with cytologically indeterminate thyroid nodules: a prospective analysis of 1056 FNA samples. J. Clin. Endocrinol. Metab. 96, 3390–3397 (2011) 28. D. Vriens, E.M. Adang, R.T. Netea-Maier, J.W. Smit, J.H. de Wilt, W.J. Oyen, L.F. de Geus-Oei, Cost-effectiveness of FDGPET/CT for cytologically indeterminate thyroid nodules: a decision analytic approach. J. Clin. Endocrinol. Metab. 99, 3263–3274 (2014) 29. A.L. Traugott, F. Dehdashti, K. Trinkaus, M. Cohen, E. Fialkowski, F. Quayle, H. Hussain, R. Davila, L. Ylagan, J.F. Moley, Exclusion of malignancy in thyroid nodules with indeterminate fine-needle aspiration cytology after negative 18 F-fluorodeoxyglucose positron emission tomography: interim analysis. World J. Surg. 34, 1247–1253 (2010)
123
Endocrine 30. D. Deandreis, A. Al Ghuzlan, A. Auperin, P. Vielh, B. Caillou, L. Chami, J. Lumbroso, J.P. Travagli, D. Hartl, E. Baudin, M. Schlumberger, S. Leboulleux, Is (18)F-fluorodeoxyglucose-PET/ CT useful for the presurgical characterization of thyroid nodules with indeterminate fine needle aspiration cytology? Thyroid 22, 165–172 (2012) 31. J.M. Kim, J.S. Ryu, T.Y. Kim, W.B. Kim, G.Y. Kwon, G. Gong, D.H. Moon, S.C. Kim, S.J. Hong, Y.K. Shong, 18Ffluorodeoxyglucose positron emission tomography does not predict malignancy in thyroid nodules cytologically diagnosed as follicular neoplasm. J. Clin. Endocrinol. Metab. 92, 1630–1634 (2007) 32. N.W. Hales, G.A. Krempl, J.E. Medina, Is there a role for fluorodeoxyglucose positron emission tomography/computed tomography in cytologically indeterminate thyroid nodules? Am. J. Otolaryngol. 29, 113–118 (2008)
123
33. N. Wang, H. Zhai, Y. Lu, Is fluorine-18 fluorodeoxyglucose positron emission tomography useful for the thyroid nodules with indeterminate fine needle aspiration biopsy? A meta-analysis of the literature. J. Otolaryngol. Head Neck Surg. 42, 38 (2013). doi:10.1186/1916-0216-42-38 34. D. Vriens, J.H. de Wilt, G.J. van der Wilt, R.T. Netea-Maier, W.J. Oyen, L.F. de Geus-Oei, The role of [18F]-2-fluoro-2-deoxy-Dglucose-positron emission tomography in thyroid nodules with indeterminate fine-needle aspiration biopsy: systematic review and meta-analysis of the literature. Cancer 117, 4582–4594 (2011) 35. S.H. Kwon, J.K. Yoon, Y.S. An, Y.S. Shin, C.H. Kim, D.H. Lee, K.S. Jo, S.J. Lee, Prognostic significance of the intratumoral heterogeneity of (18) F-FDG uptake in oral cavity cancer. J. Surg. Oncol. 110, 702–706 (2014)
ORIGINAL ARTICLE
Predictive value of intratumoral heterogeneity of F-18 FDG uptake for characterization of thyroid nodules according to Bethesda categories of fine needle aspiration biopsy results Seong-Jang Kim1,2 • Samuel Chang3
Received: 20 February 2015 / Accepted: 27 April 2015 Ó Springer Science+Business Media New York 2015
Abstract The current study was aimed to investigate the clinical value of intratumoral heterogeneity of F-18 FDG uptake for characterization of thyroid nodule (TN) with inconclusive fine-needle aspiration biopsy (FNAB) results. The current study enrolled 200 patients who showed F-18 FDG incidentaloma and were performed FNAB. The intratumoral heterogeneity of F-18 FDG uptake was represented as the heterogeneity factor (HF), defined as the derivative (dV/dT) of a volume-threshold function for a primary tumor. The diagnostic and predictive values of HF and F-18 FDG PET/CT parameters were evaluated for characterization of inconclusive FNAB results. Among F-18 FDG PET/CT parameters, SUVmax, MTV, and TLG of malignant group were statistically higher than those of Bethesda category of suspicious malignant group. However, HF values were not statistically different between the groups of Bethesda categories (Kruskal–Wallis statistics, 9.924; p = 0.0774). In ROC analysis, when HF [ 2.751 was used as cut-off value, the sensitivity and specificity for prediction of malignant TN were 100 % (95 % CI 69.2–100 %) and 60 % (95 % CI 42.1–76.1 %), respectively. The AUC was 0.826 (95 % CI 0.684–0.922) and standard error was 0.0648 (p \ 0.0001). In conclusion, the intratumoral heterogeneity of F-18 FDG uptake represented
& Seong-Jang Kim [email protected]; [email protected] 1
Department of Nuclear Medicine, Pusan National University Hospital, 179 Gudeok-Ro, Seo-Gu, Busan 602-739, Republic of Korea
2
Biomedical Research Institute, Pusan National University Hospital, Busan 602-739, Republic of Korea
3
Department of Radiology, University of Colorado School of Medicine, 12401, 17th Ave, Aurora, CO 80045, USA
by HF could be a predictor for characterization of TN with inconclusive FNAB results. Additional large populationbased prospective studies are needed to validate the diagnostic utility of HF of F-18 FDG PET/CT. Keywords F-18 FDG PET/CT Thyroid nodule Inconclusive FNAB
Introduction Thyroid nodules (TN) are common, with up to 7 % of adults demonstrating clinically palpable and detectable up to 70 % of adults by ultrasonography [1, 2]. The majority of TN are benign, but approximately, 5–15 % of all palpable TN are malignant [3, 4]. Fine-needle aspiration biopsy (FNAB) is a safe, accurate, and cost-effective method and plays essential role in the evaluation of patients with TN [4–6]. Although FNAB is the most reliable and sensitive method to detect malignancy, 15–30 % of patients remain inconclusive or indeterminate findings on cytopathology [3, 7]. With use of Bethesda system for FNAB classification, indeterminate findings have been subdivided into three categories: follicular lesion of undetermined significance or atypia of undetermined significance, follicular neoplasm or lesion suspicious for follicular neoplasm, and lesion suspicious malignancy. Because this category is associated with 20 % of incidence of malignancy, most patients referred for surgery for diagnostic rather than therapeutic purposes [1, 8, 9]. Malignant cancer cells are composed of heterogenous components not only biologic constituents but also gene expression, metabolic, and behavioral characteristics [10–12]. Heterogeneity varies in the same cancer and has a wide spectrum even in the same stage because there are differences
123
Endocrine
in properties as the growth rate, vascularity, and necrosis within the same tumor cell population [13]. Recently, there has been increasing interest in assessment of intratumoral heterogeneity of F-18 FDG uptake demonstrating an association of prognosis of patients [14–16]. In the current study, we hypothesized that malignant TN would be more heterogenous and intratumoral heterogeneity could be used for characterization of TN. The current study was aimed to investigate the clinical value of intratumoral heterogeneity of F-18 FDG uptake for characterization of TN with inconclusive FNAB results.
Methods Patients From November 2012 to July 2014, among 23,462 subjects (15,015 men, 8447 women) who underwent F-18 FDG PET/CT for known or suspected cancer and health checkup in our institution were enrolled. Among these population, 493 patients (2.1 %) who showed focal thyroid F-18 FDG incidentaloma were referred for FNAB of TN. The current study included 200 patients who were confirmed cytological diagnosis by FNAB. Remaining 293 patients who did not perform FNAB were excluded from the current study. The reasons were as follows: transfer to other hospital (n = 172), follow-up loss (n = 53), refuse of FNAB (n = 36), and US follow-up (n = 32). Our institutional review board approved this retrospective study, and the requirement to obtain informed consent was waived. Ultrasonography-guided fine-needle aspiration biopsy FNAB was performed with a 21-gauge needle on a 10-ml syringe. Aspiration was done twice for each nodule. Specimens were smeared on slides, air dried, and stained using the Papanicolaou method. Typically, the procedure was performed by an experienced endocrinologist, and the cytological diagnosis was made by an experienced pathologist. In the current study, we defined inconclusive FNAB results as follows: Bethesda categories of non-diagnostic, AUS, follicular neoplasm, and suspicious malignancy.
FDG (5 MBq/kg of body weight). The emission scan time per bed position was 3 min, and six bed positions were acquired. PET data were obtained using a high-resolution whole-body scanner with an axial field of view of 21.6 cm. The average axial resolution varied between 2.0 mm at fullwidth at half-maximum in the center and 2.4 mm at 28 cm. The average total PET/CT examination time was 20 min. Measurement of volumetric parameters of F-18 FDG PET/CT The SUVmax was obtained using the following formula: SUVmax = maximum activity in the region of interest (MBq/gram)/(injected dose [MBq]/body weight [grams]). The metabolic tumor volume (MTV) was determined as the total number of voxels with threshold SUV of C40 % of the SUVmax in the volume of interest (VOI). The total lesion glycolysis (TLG) was calculated as the MTV multiplied by its SUVmean. Determination of intratumoral heterogeneity of F-18 FDG uptake The intratumoral metabolic heterogeneity was represented by the heterogeneity factor (HF) [17], which was determined for each patient as follows: a region of interest (ROI) was manually drawn to fully include the primary tumor and a surrounding region of normal tissue (normal background). The tumor volume was determined with a series of SUV thresholds (e.g., 40, 50, 60, 70, and 80 % of SUVmax) using a Syngo Multimodality Workspace (Siemens AG, Berlin, Germany). We excluded the values of \40 % from the heterogeneity analysis because a previous study reported that the minimal threshold that represents the actual tumor volume was 40 % and the values of\40 % included too much normal tissue background activity [18]. In addition, values of[80 % were also excluded because the volumes were small and the partial volume effect was pronounced [19]. A volumethreshold function of the tumor was acquired by plotting thresholds to volumes. Linear regression analysis was performed, and the HF was calculated by finding the derivative (dV/dT) of the volume-threshold function for each tumor. Next, the values of the HF were modified into absolute values so that all resulting values were positive; the more positive the factor, the more heterogeneous the tumor.
F-18 FDG PET/CT Statistical analysis F-18 FDG PET/CT images were obtained using a dedicated PET/CT scanner (Biograph 40; Siemens, Knoxville, Tennessee, USA). Standard patient preparation included a fasting at least 8 h and a serum glucose level lower than 120 mg/dl before F-18 FDG administration. F-18 FDG PET/ CT imaging was performed 60 min after injection of F-18
123
Statistical analyses were performed using commercially available software (MedCalc 14.12.0, Mariakerke, Belgium). Continuous data are expressed mean ± SD for normally distributed values and median (range) for nonparametric values (skewed data). An independent 2-sample
Endocrine
t test was used to compare the various parameters of F-18 FDG PET/CT between malignant and benign TN. To compare the F-18 FDG PET/CT parameters according to the Bethesda categories, Kruskal–Wallis test was used with appropriate Tukey post hoc test. Receiver operating characteristic (ROC) curves for each parameters for the prediction of malignant TN were derived and evaluated by comparing area under curve (AUC). The sensitivity and specificity of each parameter were determined at the optimal cut-off values. Statistical significance was defined as p \ 0.05.
Results Characteristics of patients Table 1 shows characteristics of patients included in the current study. The results of FNAB were 39 malignant (Category 6), 13 suspicious malignant (Category 5), 6 follicular neoplasm (Category 4), 12 atypia of undetermined significance (AUS, Category 3), 116 benign (Category 2), and 14 non-diagnostic (Category 1). The patients with non-diagnostic FNAB results underwent repeated FNAB. Repeated FNAB showed 2 malignant and 12 benign TN. The current study treated the first FNAB results in terms of inconclusive FNAB. All 72 patients with FNAB Table 1 Characteristics of patients Thyroid nodules Variables
Malignant (n = 49)
Benign (n = 151)
p value
Age
49.4 ± 10.1
51.8 ± 10.9
0.1654
Size
1.9 ± 1.1
2.1 ± 1.3
0.2144 0.0273
Sex Male
12
31
37
120
Category 1
2
12
Category 2
0
116
Category 3
1
11
Category 4
1
5
Category 5
6
7
Category 6
39
0
Female
F-18 FDG PET/CT parameters 5.96 ± 2.61
5.52 ± 3.18
Comparison of F-18 FDG PET/CT parameters according to Bethesda category Table 2 show the differences of SUVmax, SUVmean, MTV, and TLG according to the Bethesda category of FNAB results. Among the F-18 FDG PET/CT parameters, SUVmean was not statistically different between the groups of Bethesda categories. However, SUVmax, MTV, and TLG of Category 6 were statistically higher than those of Bethesda category 5. Comparison of HF according to Bethesda category Figure 1 shows differences of HF according to the Bethesda categories of FNAB results. The HF values were not statistically different between the groups of Bethesda categories (Kruskal–Wallis statistics, 9.924; p = 0.0774). Comparison of F-18 FDG PET/CT parameters Figure 2 shows the differences of parameters and HF of F-18 FDG PET/CT between benign, inconclusive, and malignant groups of FNAB results. All of these parameters were not statistically different between 3 groups (SUVmax; KW statistics, 3.555, p = 0.0624: SUVmean; KW statistics, 0.261, p = 0.7713: MTV; KW statistics, 1.455, p = 0.2358: TLG; KW statistics, 0.3873, p = 0.6794: HF; KW statistics, 2.634, p = 0.0743). Characterization of TN using F-18 FDG PET/CT parameters
Bethesda category
SUVmax
results of follicular neoplasm, AUS, malignant, and suspicious malignant underwent total thyroidectomy or lobectomy and final histologic diagnosis was surgically confirmed. The malignant pathologic results were 2 follicular carcinoma and 47 papillary thyroid cancers. Among the semiquantitative parameters, HF of malignant TN was higher than benign one.
0.3862
SUVmean
2.54 ± 0.8
2.53 ± 1.16
0.9548
MTV (ml)
5.76 ± 2.0
5 ± 2.13
0.5031
TLG
16.01 ± 6.9
15.27 ± 8.82
0.8655
HF
3.2019 ± 0.948
2.8065 ± 0.782
0.0040
FNAB fine-needle aspiration biopsy, AUS atypia of undetermined significance, FN follicular neoplasm
Table 3 shows the ROC analysis results for prediction of malignant TN. Among the various parameters, only HF could be used for differentiation of TN. When HF [ 3.027 was used as cut-off value, the sensitivity and specificity for prediction of malignant TN were 59.2 % (95 % CI 44.2–73 %) and 65.6 % (95 % CI 57.4–73.1 %), respectively. The AUC was 0.606 (95 % CI 0.535–0.675) and standard error was 0.0464 (p = 0.0217). Prediction of malignant TN with inconclusive FNAB Figure 3 demonstrates the ROC results of F-18 FDG PET/ CT parameters for prediction of malignant TN with
123
Tukey post hoc test showed higher value than that of Bethesda category of SM
Fig. 1 The differences of HF according to the Bethesda categories of FNAB results. The HF values were not statistically different between the groups of Bethesda categories (Kruskal–Wallis statistics, 9.924; p = 0.0774)
a
Data are expressed as median values with ranges
KW statistics Kruskal–Wallis statistics, SUVmax maximal standardized uptake value, SUVmean mean standardized uptake value, MTV metabolic tumor volume, TLG total lesion glycolysis, HF heterogeneity factor
0.0774
0.0121 14.62
3.095 (0.868–6.365) 3.354 (2.130–3.765) 2.738 (1.976–3.255) 2.553 (1.627–3.886) 2.825 (1.889–3.561) HF
2.876 (0.235–5.121)
11.16a (0.9–61.57) 3.36 (0.56–10.45) 12.21 (2.93–33.01) 3.77 (1.13–34.38) TLG
5.83 (0.6–146.23)
5.05 (0.97–211.74)
9.924
0.1849
0.0105 14.97 3.8a (0.4–30.5) 1.61 (0.33–4.8) 1.7 (0.47–6.47) MTV (ml)
2.25 (1.2–5.6) SUVmean
2.71 (0.3–33.85)
4.57 (1.54–7.86)
0.0092
2.3 (0.57–46.03)
7.517 2.5 (1.4–4.7) 2 (1.4–2.8) 2.65 (1.9–4.2) 2.1 (1.4–9.4)
15.28 5.1a (2.7–13.2) 4.2 (2.3–5.3) 5.2 (4.6–9.5) 4.45 (2.4–28) 4.35 (2.6–8.3) SUVmax
6 (n = 39) 5 (n = 13) 4 (n = 6) 3 (n = 12) 2 (n = 116) 1 (n = 14)
4.75 (2.3–8.3)
KW statistics Bethesda categories Variables
Table 2 Comparison of F-18 FDG PET/CT parameters according to Bethesda categories of FNAB results of thyroid nodules
123
2.2 (1.1–8.6)
p value
Endocrine
inconclusive FNAB results. Among the parameters, only the HF could predict malignant TN. When HF [ 2.751 was used as cut-off value, the sensitivity and specificity for prediction of malignant TN were 100 % (95 % CI 69.2–100 %) and 60 % (95 % CI 42.1–76.1 %), respectively. The AUC was 0.826 (95 % CI 0.684–0.922) and standard error was 0.0648 (p \ 0.0001).
Discussion The current study showed that HF could be used for characterization of TN. Also, in patients with inconclusive FNAB results, the HF could be used for prediction of malignant TN. Thyroid cancer is the most common endocrine malignancy. It represents approximately 1 % of all cancers, corresponding to up to 10.2 per 100,000 individuals in the United States with increasing incidence over the last decades [20]. Early detection and diagnosis is important in appropriate treatment of thyroid cancer, as delays in the diagnosis are associated with increased mortality [21]. The evaluation of TN includes a FNAB and measurement of serum thyroid-stimulating hormone. FNAB is the most important diagnostic test in the initial evaluation of a patient with TN and has high diagnostic accuracy. However, a number of reports have shown that 11–42 % of FNAB results are reported as indeterminate [22, 23]. These indeterminate TN present a clinical dilemma to the
Endocrine
Fig. 2 Comparison of parameters and HF between benign, inconclusive, and malignant groups of FNAB results. All of these parameters were not statistically different between 3 groups (SUVmax; KW
statistics, 3.555, p = 0.0624: SUVmean; KW statistics, 0.261, p = 0.7713: MTV; KW statistics, 1.455, p = 0.2358: TLG; KW statistics, 0.3873, p = 0.6794: HF; KW statistics, 2.634, p = 0.0743)
Table 3 ROC results for characterization of TN using F-18 FDG PET/CT parameters Variables
Cut-off
Sensitivity (%)
95 % CI
Specificity (%)
95 % CI
AUC
95 % CI
SE
p
SUVmax
[4.5
65.3
50.4–78.3 %
52.3
44–60.5 %
0.586
0.514–0.655
0.0463
0.0646
SUVmean
[2.5
42.9
28.8–57.8 %
69.5
61.5–76.8 %
0.549
0.478–0.620
0.0469
0.2915
MTV
[3.39
57.1
42.2–71.2 %
66.9
58.8–74.3 %
0.566
0.494–0.636
0.0516
0.2004
TLG
[7.98
57.1
42.2–71.2 %
63.6
55.4–71.2 %
0.562
0.490–0.632
0.0514
0.2286
HF
[3.027
59.2
44.2–73 %
65.6
57.4–73.1 %
0.606
0.535–0.675
0.0464
0.0217
95 % CI confidence interval, AUC area under curve, SE standard error
clinicians. Therefore, the next step in the current management of these patients with TN is usually surgical excision. At present, there is no alternative algorithm for a more conservative management of patients with TN of indeterminate results. Several methods had been used to characterize of such indeterminate FNAB results of TN. It has been reported that certain sonographic features of a TN are associated with an increased likelihood of malignancy. Recently, some studies have attempted to characterize the indeterminate TN using ultrasonography [24, 25]. Khoncarly et al. concluded that with the exception of nodule height greater than width, sonographic criteria were not helpful in deciding with patients with AFLUS should undergo thyroidectomy [24]. Batawil et al. also concluded that based on the limited accuracy or predictive value of ultrasonographic risk factors, surgery is the treatment of choice for indeterminate TN [25]. The use of molecular tests as an adjunct to diagnosis in indeterminate TN is standard care in the United States. A
167-gene expression classifier is used to minimize unnecessary diagnostic thyroid surgery, and another mutation marker panel is used to select patients for initial total thyroidectomy, thereby saving on the two-step surgery [26, 27]. Although molecular tests are useful in defining the natures of TN, these tests are not available at all centers, their cost is high, and greater experience is needed, particularly in the case of indeterminate TN. A recent costeffectiveness analytic study showed that full implementation of F-18 FDG PET/CT in the workup of adult patients with TN scheduled for surgery for FNAB-indeterminacy could lead to a decrease in costs and a moderate increase in health related quality of life compared with diagnostic surgery in all patients according to current European practice and in competitive to the current US standard of the gene expression classifier [28]. The use of F-18 FDG PET/CT in the management of thyroid disease has been limited primarily to the postoperative surveillance of patients with known differentiated thyroid carcinoma. Several studies have reported the role of
123
Endocrine
Fig. 3 ROC analysis of F-18 FDG PET/CT parameters for prediction of malignant TN with inconclusive FNAB results
F-18 FDG PET/CT in assessing cytological indeterminate TN with conflicting results [29–31]. Also, a recent pilot study showed that F-18 FDG PET/CT was not a predictable indicator of the presence of thyroid cancer [32]. To predict the correct diagnostic value of F-18 FDG PET/CT for characterization of cytological indeterminate TN, two meta-analytic studies were conducted [33, 34]. Wang et al. [33] conducted meta-analysis including 7 studies (267 patients). In their analysis, the pooled sensitivity of PET or PET/CT for the detection of thyroid cancer was 89.0 % (95 % CI 79.0–95.0 %). The pooled specificity was 55.0 % (95 CI 48.0 %–62.0 %). Vriens et al. [34] concluded that in patients with TN who have indeterminate FNAB results, a negative F-18 FDG PET scan improves diagnostic accuracy, particularly for patients who have lesions [15 mm. All false-negative F-18 FDG PET results were from lesions \15 mm. The current study investigated the diagnostic values of volumetric parameters and HF of F-18 FDG PET/CT for characterization of TN with inconclusive FNAB results. In the current study, only the HF could predict malignant TN in patients with inconclusive FNAB results. Also, in all TN, only the HF could differentiate the malignant and benign TN. Some previous studies have investigated the prognostic value of HF in patients with oral cavity cancer and breast cancer [16, 35]. Kwon SH et al. concluded that the intratumoral heterogeneity of F-18 FDG uptake may be a
123
significant prognostic factor for overall survival in addition to cervical lymph node metastasis in oral cavity cancer [35]. In breast cancer patients, intratumoral metabolic heterogeneity correlated closely with the MTV and significantly affected the OS in IDC patients. Therefore, they concluded the HF may act as a robust surrogate marker for the prediction of OS in IDC patients [16]. The current study adapted the HF for characterization of TN with inconclusive FNAB results and showed the possibility of clinical adaptation of HF in clinical situation of F-18 FDG incidentaloma. The current study has some limitations. First, it was a retrospective, single-center study. Second, the intratumoral metabolic heterogeneity on F-18 FDG PET scans can be represented by various methods. The parameter of intratumoral metabolic heterogeneity is not well standardized, although histologic studies have shown good spatial correlation between areas of F-18 FDG uptake and histologic findings. Furthermore, in nodular goiter, the heterogeneity might be notable because of necrosis, inflammation, and hemorrhage. A feasible and highly reproducible method for obtaining a heterogeneity parameter representing intratumoral metabolic heterogeneity is warranted. Third, the population of inconclusive FNAB results is relatively small. Finally, VOI selection is important for determining tumor MTV. Several different methods of VOI selection could be used: manually defined VOI, isocontour VOIs based on a fixed percentage of the SUVmax, and isocontour
Endocrine
VOIs based on a fixed percentage of SUVmax threshold, and generally 40 % of SUVmax is accepted for VOI selection.
Conclusion In conclusion, the intratumoral heterogeneity of F-18 FDG uptake represented by HF could be a predictor for characterization of TN with inconclusive FNAB results. Additional large population-based prospective studies are needed to validate the diagnostic utility of HF of F-18 FDG PET/CT.
Conflict of interest The authors declared that there is no conflict of interests.
References 1. C. Hambleton, E. Kandil, Appropriate and accurate diagnosis of thyroid nodule: a review of thyroid fine-needle aspiration. Int. J. Clin. Exp. Med. 6, 413–422 (2013) 2. M. Goldfarb, S. Gondek, C. Soloranzo, J.I. Lew, Surgeon-performed ultrasound can predict benignity in thyroid nodules. Surgery 50, 436–441 (2011) 3. D.S. Cooper, G.M. Doherty, B.R. Haugen, R.T. Kloos, S.L. Lee, S.J. Mandel, E.L. Mazzaferri, B. McIver, F. Pacini, M. Schlumberger, S.I. Sherman, D.L. Steward, R.M. Tuttle, Revised American Thyroid Association management guidelines for patients with thyroid nodules and differentiated thyroid cancer. Thyroid 19, 1167–1214 (2009) 4. L. Hegedu¨s, Clinical practice. The thyroid nodule. N. Engl. J. Med. 351, 1764–1771 (2004) 5. J.R. Porterfield Jr, C.S. Grant, D.S. Dean, G.B. Thompson, D.R. Farley, M.L. Richards, C.C. Reading, J.W. Charboneau, B.K. Vollrath, T.J. Sebo, Reliability of benign fine needle aspiration cytology of large thyroid nodules. Surgery 144, 963–968 (2008) 6. J.B. Ogilvie, E.J. Piatigorsky, O.H. Clark, Current status of fine needle aspiration for thyroid nodules. Adv. Surg. 40, 223–238 (2006) 7. E.S. Cibas, S.Z. Ali, NCI Thyroid FNA State of the Science Conference, The Bethesda system for reporting thyroid cytopathology. Am. J. Clin. Pathol. 132, 658–665 (2009) 8. S. Rorive, N. D’Haene, C. Fossion, I. Delpierre, N. Abarguia, F. Avni, C. Decaestecker, I. Salmon, Ultrasound-guided fine-needle aspiration of thyroid nodules: stratification of malignancy risk using follicular proliferation grading, clinical and ultrasonographic features. Eur. J. Endocrinol. 162, 1107–1115 (2010) 9. L.J. Layfield, E.S. Cibas, H. Gharib, S.J. Mandel, Thyroid aspiration cytology: current status. CA Cancer J. Clin. 59, 99–110 (2009) 10. G.H. Heppner, Tumor heterogeneity. Cancer Res. 44, 2259–2265 (1984) 11. R.J. Gillies, P.A. Schornack, T.W. Secomb, N. Raghunand, Causes and effects of heterogeneous perfusion in tumors. Neoplasia 1, 197–207 (1999) 12. M.E. Lleonart, P. Martin-Duque, R. Sanchez-Prieto, A. Moreno, S. Cajal, Tumor heterogeneity: morphological, molecular and clinical implications. Histol. Histopathol. 15, 881–889 (2000)
13. B. Huang, T. Chan, D.L. Kwong, W.K. Chan, P.L. Khong, Nasopharyngeal carcinoma: investigation of intratumoral heterogeneity with FDG PET/CT. AJR Am. J. Roentgenol. 199, 169–174 (2012) 14. F. Tixier, C.C. Le Rest, M. Hatt, N. Albarghach, O. Pradier, Intratumor heterogeneity characterized by textural features on baseline 18F-FDG PET images predict response to concomitant radiochemotherapy in esophageal cancer. J. Nucl. Med. 52, 369–378 (2011) 15. G.J. Cook, C. Yip, M. Siddique, V. Goh, S. Chicklore, A. Roy, P. Marsden, S. Ahmad, D. Landau, Are pretreatment 18F-FDG PET tumor textural features in non-small cell lung cancer associated with response and survival after chemoradiotherapy? J. Nucl. Med. 54, 19–26 (2013) 16. S.H. Son, D.H. Kim, C.M. Hong, C.Y. Kim, S.Y. Jeong, S.W. Lee, J. Lee, B.C. Ahn, Prognostic implication of intratumoral metabolic heterogeneity in invasive ductal carcinoma of the breast. BMC Cancer 14, 585–595 (2014) 17. E.A. Kidd, P.W. Grigsby, Intratumoral metabolic heterogeneity of cervical cancer. Clin. Cancer Res. 14, 5236–5241 (2008) 18. T.R. Miller, P.W. Grigsby, Measurement of tumor volume by PET to evaluate prognosis in patients with advanced cervical cancer treated by radiation therapy. Int. J. Radiat. Oncol. Biol. Phys. 53, 353–359 (2002) 19. M. Soret, S.L. Bachrach, I. Buvat, Partial volume effect in PET tumor imaging. J. Nucl. Med. 48, 932–945 (2007) 20. R. Siegel, D. Naishadham, A. Jemal, Cancer statistics, 2012. CA Cancer J. Clin. 62, 10–29 (2012) 21. J.A. Sipos, E.L. Mazzaferri, Thyroid cancer epidemiology and prognostic variables. Clin. Oncol. (R. Coll. Radiol.) 22, 395–404 (2010) 22. B. Miller, S. Burkey, G. Lindberg, W.H. Snyder 3rd, F.E. Nwariaku, Prevalence of malignancy within cytologically indeterminate thyroid nodules. Am. J. Surg. 188, 459–462 (2004) 23. G.M. Scabas, G.A. Staerkel, S.E. Shapiro, Fine-needle aspiration of the thyroid and correlation with hispathology in a contemporary series of 240 patients. Am. J. Surg. 186, 702–710 (2003) 24. S.M. Khoncarly, S.W. Tamarkin, C.R. McHenry, Can ultrasound be used to predict malignancy in patients with a thyroid nodule and an indeterminate fine-needle aspiration biopsy ? Surgery 156, 967–971 (2014) 25. N. Batawil, T. Alkordy, Ultrasonographic features associated with malignancy in cytologically indeterminate thyroid nodules. EJSO 40, 182–186 (2014) 26. H. Li, K.A. Robinson, B. Anton, I.J. Saldanha, P.W. Ladenson, Cost-effectiveness of a novel molecular test for cytologically indeterminate thyroid nodules. J. Clin. Endocrinol. Metab. 96, E1719–E1726 (2011) 27. Y.E. Nikiforov, N.P. Ohori, S.P. Hodak, S.E. Carty, S.O. LeBeau, R.L. Ferris, L. Yip, R.R. Seethala, M.E. Tublin, M.T. Stang, C. Coyne, J.T. Johnson, A.F. Stewart, M.N. Nikiforova, Impact of mutational testing on the diagnosis and management of patients with cytologically indeterminate thyroid nodules: a prospective analysis of 1056 FNA samples. J. Clin. Endocrinol. Metab. 96, 3390–3397 (2011) 28. D. Vriens, E.M. Adang, R.T. Netea-Maier, J.W. Smit, J.H. de Wilt, W.J. Oyen, L.F. de Geus-Oei, Cost-effectiveness of FDGPET/CT for cytologically indeterminate thyroid nodules: a decision analytic approach. J. Clin. Endocrinol. Metab. 99, 3263–3274 (2014) 29. A.L. Traugott, F. Dehdashti, K. Trinkaus, M. Cohen, E. Fialkowski, F. Quayle, H. Hussain, R. Davila, L. Ylagan, J.F. Moley, Exclusion of malignancy in thyroid nodules with indeterminate fine-needle aspiration cytology after negative 18 F-fluorodeoxyglucose positron emission tomography: interim analysis. World J. Surg. 34, 1247–1253 (2010)
123
Endocrine 30. D. Deandreis, A. Al Ghuzlan, A. Auperin, P. Vielh, B. Caillou, L. Chami, J. Lumbroso, J.P. Travagli, D. Hartl, E. Baudin, M. Schlumberger, S. Leboulleux, Is (18)F-fluorodeoxyglucose-PET/ CT useful for the presurgical characterization of thyroid nodules with indeterminate fine needle aspiration cytology? Thyroid 22, 165–172 (2012) 31. J.M. Kim, J.S. Ryu, T.Y. Kim, W.B. Kim, G.Y. Kwon, G. Gong, D.H. Moon, S.C. Kim, S.J. Hong, Y.K. Shong, 18Ffluorodeoxyglucose positron emission tomography does not predict malignancy in thyroid nodules cytologically diagnosed as follicular neoplasm. J. Clin. Endocrinol. Metab. 92, 1630–1634 (2007) 32. N.W. Hales, G.A. Krempl, J.E. Medina, Is there a role for fluorodeoxyglucose positron emission tomography/computed tomography in cytologically indeterminate thyroid nodules? Am. J. Otolaryngol. 29, 113–118 (2008)
123
33. N. Wang, H. Zhai, Y. Lu, Is fluorine-18 fluorodeoxyglucose positron emission tomography useful for the thyroid nodules with indeterminate fine needle aspiration biopsy? A meta-analysis of the literature. J. Otolaryngol. Head Neck Surg. 42, 38 (2013). doi:10.1186/1916-0216-42-38 34. D. Vriens, J.H. de Wilt, G.J. van der Wilt, R.T. Netea-Maier, W.J. Oyen, L.F. de Geus-Oei, The role of [18F]-2-fluoro-2-deoxy-Dglucose-positron emission tomography in thyroid nodules with indeterminate fine-needle aspiration biopsy: systematic review and meta-analysis of the literature. Cancer 117, 4582–4594 (2011) 35. S.H. Kwon, J.K. Yoon, Y.S. An, Y.S. Shin, C.H. Kim, D.H. Lee, K.S. Jo, S.J. Lee, Prognostic significance of the intratumoral heterogeneity of (18) F-FDG uptake in oral cavity cancer. J. Surg. Oncol. 110, 702–706 (2014)
