Interobserver Agreement in Ultrasound Risk Stratification Systems for Thyroid Nodules on Static Images Versus Cine-Loop Video Sequences.

PURPOSE: To evaluate the impact of video sequences (cine-loops) on the interobserver agreement (IOA) using risk stratification systems (RSSs) for thyroid nodules (TNs). METHODS: Twenty TNs were randomly selected from a large database and evaluated by twelve experienced observers using five different RSSs (Kwak-, ACR-, EU-, Korean-TIRADS, ATA Guidelines). In the first step, the evaluation was conducted based on static ultrasound (US) images in two planes ("static"). Six months later, these cases were reevaluated by the same observers using video sequences in two planes ("cine-loops"). Fleiss' kappa (κ) was calculated for the IOA analyses. RESULTS: IOA on static was moderate with κ values of 0.46, 0.42, 0.40, 0.45, and 0.38 for the Kwak-, ACR-, EU-, Korean-TIRADS, and ATA Guidelines, respectively, while the IOA on cine-loops was fair with κ values of 0.41, 0.38, 0.37, 0.36, and 0.34 for the Kwak-, ACR-, EU-, Korean-TIRADS, and ATA Guidelines, respectively. The overall IOA was superior in static images versus cine-loops (p = 0.024). Among other findings, the subgroup analyses (related to age, gender, US certificates, number of thyroid US per week, and RSSs experience) particularly showed that the experience of the observers in using RSSs had a significant influence on the IOA. CONCLUSIONS: The overall IOA (all twelve observers and all five RSSs) was superior on static US images in comparison to cine-loops. Furthermore, the overall IOA of the five US features revealed superior κ values of the static images over cine-loops. However, this impact was significantly lower when the observers were highly experienced in the use of US RSSs of TNs.

Thyroid scintigraphy: establishing a clinically useful normal range for 99mTc pertechnetate uptake.

PURPOSE: This study aims to establish a normal range for the thyroid uptake derived from 99mTc pertechnetate scans. In particular, variations of uptake with TSH stimulation and other factors such as urinary iodine concentration are taken into account and compared with the calculation of a raw uptake value. METHODS: Clinical multicentric (center A, B and C) prospective study on 125 consecutive healthy patients undergoing thyroid scans for thyroid nodules. Normal functional thyroid status was assured by normal TSH, normal thyroid size, no thyroid antibodies and no symptoms of thyroid functional disorders. Calculations of raw Tc-uptake (uptake) and modified uptake values regarding current TSH value (uptakeTSH1), urinary iodine concentration (uptakeTSH1&uic), gland volume, age, smoking status, weight and tissue thickness ventral to the thyroid were performed. RESULTS: There is a positive correlation of thyroid uptake with TSH allowing for the calculation of a normalized uptake value (uptakeTSH1). The normal range for uptakeTSH1 compares favourable to that for raw uptake in that it yields a clear distinction from thyroid functional disorders. The additional normalization for urinary iodine concentration (uptakeTSH1&uic) may even improve the distinctive power whereas further normalizations such as for gland volume, age and others are not warranted by this study. The 95% CI of uptakeTSH1 for sites A, A&B, and A&B&C were 0.21%-2.06%, 0.22%-2.38% and 0.24%-2.40%. CONCLUSION: A normal range for the thyroid uptake can be established with respect to the current TSH stimulation. This normalization (uptakeTSH1) overcomes the drawback of raw uptake by yielding a clinically useful parameter with obviously high distinctive power against functional thyroid disorders.

Thyroid Imaging Reporting and Data Systems: Applicability of the "Taller than Wide" Criterium in Primary/Secondary Care Units and the Role of Thyroid Scintigraphy.

BACKGROUND: To examine the applicability of the "taller than wide" (ttw) criterium for risk assessment of thyroid nodules (TNs) in primary/secondary care units and the role of thyroid scintigraphy therein. METHODS: German bicenter study performed in a setting of primary/secondary care. Patient recruitment and analysis in center A was conducted in a prospective manner. In center B, patient data were retrieved from a database that was originally generated by prospective data collection. TNs were assessed by ultrasound and thyroid scans, mostly fine needle biopsy and occasionally surgery and others. In center A, only patients who presented for the first time were included. The inclusion criterion was any TN ≥ 10 mm that had at least the following two sonographic risk features: solidity and a ttw shape. In center B, consecutive patients who had at least ttw and hypofunctioning nodules ≥ 10 mm were retrieved from the above-mentioned database. The risk of malignancy was determined according to a mixed reference standard and compared with literature data. RESULTS: In center A, 223 patients with 259 TNs were included into the study. For further analysis, 200 nodules with a reference standard were available. The overall malignancy rate was 2.5% (upper limit of the 95% CI: 5.1%). After the exclusion of scintigraphically hyperfunctioning nodules, the malignancy rate increased slightly to 2.8% (upper limit of the 95% CI: 5.7%). Malignant nodules exhibited sonographic risk features additional to solidity and ttw shape more often than benign ones. In addition to the exclusion of hyperfunctioning nodules, when considering only nodules without additional US risk features, i.e., exclusively solid and ttw-nodules, the malignancy rate decreased to 0.9% (upper limit 95% CI: 3.7%). In center B, from 58 patients, 58 ttw and hypofunctioning TNs on thyroid scans with a reference standard were available. Malignant nodules from center B were always solid and hypoechoic. The overall malignancy rate of hypofunctioning and ttw nodules was 21%, with the lower limit of the 95% CI (one-sided) being 12%. CONCLUSIONS: In primary/secondary care units, the lowest TIRADS categories for indicating FNB, e.g., applying one out of five sonographic risk features, may not be appropriate owing to the much lower a priori malignancy risk in TNs compared to tertiary/quaternary care units. Even the combination of two sonographic risk features, "solidity" and "ttw", may only be appropriate in a limited fashion. In contrast, the preselection of TNs according to hypofunctioning findings on thyroid scans clearly warranted FNB, even when applying only one sonographic risk criterion ("ttw"). For this reason, thyroid scans in TNs may not only be indicated to rule out hyperfunctioning nodules from FNB but also to rule in hypofunctioning ones.

Correct and Incorrect Recommendations for or against Fine Needle Biopsies of Hypofunctioning Thyroid Nodules: Performance of Different Ultrasound-based Risk Stratification Systems.

PURPOSE: To evaluate the recommendations for or against fine needle biopsy (FNB) of hypofunctioning thyroid nodules (TNs) using of five different Ultrasound (US) -based risk stratification systems (RSSs). METHODS: German multicenter study with 563 TNs (≥ 10 mm) in 534 patients who underwent thyroid US and surgery. All TNs were evaluated with ACR TI-RADS, EU-TIRADS, ATA, K-TIRADS 2016 and modified K-TIRADS 2021. A correct recommendation was defined as: malignant TN with recommendation for FNB (appropriate) or benign TN without recommendation for FNB (avoided). An incorrect recommendation was defined as: malignant TN without recommendation for FNB (missed) or benign TN with recommendation for FNB (unnecessary). RESULTS: ACR TI-RADS demonstrated the highest rate of correct (42.3 %) and lowest rate of incorrect recommendations (57.7 %). The other RRSs showed similar results for correct (26.5 %-35.7 %) and incorrect (64.3 %-73.5 %) recommendations. ACR TI-RADS demonstrated the lowest rate of unnecessary (73.4 %) and the highest rate of appropriate (26.6 %) FNB recommendation. For other RSSs, the rates of unnecessary and appropriate FNB were between 75.2 %-77.1 % and 22.9 %-24.8 %. The lowest rate of missed FNB (14.7 %) and the highest rate of avoided FNB (85.3 %) was found for ACR TI-RADS. For the other RSSs, the rates of missed and avoided FNB were between 17.8 %-26.9 % and 73.1 %-82.2 %. When the size cutoff was disregarded, an increase of correct recommendations and a decrease of incorrect recommendations was observed for all RSSs. CONCLUSION: The RSSs vary in their ability to correctly recommend for or against FNB. An understanding of the impact of nodule size cutoffs seems necessary for the future of TIRADS.

Introducing a Pole Concept for Nodule Growth in the Thyroid Gland: Taller-than-Wide Shape, Frequency, Location and Risk of Malignancy of Thyroid Nodules in an Area with Iodine Deficiency.

Purpose: (i) To examine the criterion taller-than-wide (TTW) for the sonographic assessment of thyroid nodules in areas of iodine deficiency in terms of frequency, anatomical distribution within the thyroid gland and risk of malignancy. (ii) To develop a model for nodule growth in the thyroid gland. Methods: German multicenter study consisting of two parts. In the prospective part, thyroid nodules were sonographically measured in all three dimensions, location within the thyroid gland and contact to a protrusion-like formation (horn) in the dorsal position of thyroid gland was noted. In addition, further sonographic features such as the composition, echogenity, margins and calcifications were investigated. All nodules from the prospective part were assessed for malignancy as part of clinical routine at the decision of the treating physician adhering to institutionally based algorithms. In the retrospective part, only nodules with fine needle aspiration and/or histology were included. The risk of malignancy in TTW nodules was determined by correlating them with cyotological and histological results. Results: Prospective part: out of 441 consecutively evaluated thyroid nodules, 6 were found to be malignant (1.4%, 95% CI 0.6-2.7%). Among the 74 TTW nodules (17%), 1 was malignant (1%, 95% CI 0-4%). TTW nodules were more often located in the dorsal half of the thyroid than non-TTW nodules (factor 2.3, p = 0.01, 95% CI 2.1-2.5) and more often located in close proximity to a horn than non-TTW nodules (factor 3.0, p = 0.01, 95% CI 2.4-3.8). Retrospective part: out of 1315 histologically and/or cytologically confirmed thyroid nodules, 163 TTW nodules were retrieved and retrospectively analyzed. A TTW nodule was 1.7 times more often benign when it was dorsal (95% CI 1.1-2.5) and 2.5 times more often benign when it was associated with a horn (95% CI 1.2-5.3). The overall probability of malignancy for TTW nodules was 38% (95% CI 30-46%) in this highly preselected patient group. Conclusion: TTW nodules are common in iodine deficient areas. They are often located in the dorsal half of the thyroid gland and are frequently associated with a dorsal protrusion-like formation (horn) of the thyroid. Obviously, the shape of benign nodules follows distinct anatomical preconditions within the thyroid gland. The frequency of TTW nodules and their predominant benignity can be explained by a pole concept of goiter growth. The difference between the low malignancy risk of TTW nodules found on a prospective basis and the high risk found retrospectively may be the result of a positive preselection in the latter.

FDG-PET/CT for the early prediction of histopathological complete response to neoadjuvant chemotherapy in breast cancer patients: initial results.

BACKGROUND: Up to about one-quarter of patients treated with neoadjuvant chemotherapy do not adequately respond to the given treatment. By a differentiation between responders and non-responders ineffective toxic therapies can be prevented. PURPOSE: To retrospectively test if FDG-PET/CT is able to early differentiate between breast cancer lesions with pathological complete response (pCR) and lesions without pathological complete response (npCR) after two cycles of neoadjuvant chemotherapy (NACT). MATERIAL AND METHODS: In this retrospective study 26 breast cancer patients (mean age, 46.9 years ± 9.9 years) underwent a pre-therapeutic FDG-PET/CT scan and a subsequent FDG-PET/CT after the second cycle of NACT. Histopathology of resected specimen served as the reference standard. Maximum standardized uptake values (SUVmax) of cancer lesions before and after the second cycle of NACT were measured. Two evaluation algorithms were used: (a) pCR: Sinn Score 3 and 4, npCR: Sinn Score 0-2; (b) pCR: Sinn Score 4, npCR: Sinn Score 0-3. The absolute and relative decline of the SUVmax (ΔSUVmax, ΔSUVmax(%))was calculated. Differences of the SUVmax as well as of the SUVmax decline between pCR lesions and npCR lesions were tested for statistical significance P < 0.05. To identify the optimal cut-off value of ΔSUVmax(%) to differentiate between pCR lesions and npCR lesions a receiver-operating curve (ROC) analysis was performed. RESULTS: Using evaluation algorithm A the ΔSUVmax was 13.5 (pCR group) and 3.9 (npCR group) (P = 0.006); the ΔSUVmax(%) was 79% and 47%, respectively (P = 0.001). On ROC analysis an optimal cut-off ΔSUVmax(%) of 66% was found. Using evaluation algorithm B the ΔSUVmax was 17.5 (pCR group) and 4.9 (npCR group) (P = 0.013); the ΔSUVmax(%) was 89% and 51%, respectively (P = 0.003). On ROC analysis an optimal cut-off ΔSUVmax(%) of 88% was found. CONCLUSION: FDG-PET/CT may be able to early differentiate between pCR and npCR of primary breast cancer lesions after two cycles of NACT.

Pre-therapeutic blood dosimetry in patients with differentiated thyroid carcinoma using 124-iodine: predicted blood doses correlate with changes in blood cell counts after radioiodine therapy and depend on modes of TSH stimulation and number of preceding radioiodine therapies.

OBJECTIVE: Pre-therapeutic blood dosimetry prior to a high-dose radioiodine therapy (RAIT) is recommended and a blood dose of 2 Gy is considered to be safe. In this study, changes in the blood cell count after radioiodine therapy of high risk differentiated thyroid carcinoma (DTC) were analyzed and compared with the results of the pre-therapeutic blood dosimetry using 124I. Moreover, the influence of different modes of TSH stimulation and the number of preceding radioiodine therapies on the blood dose were assessed. METHODS: 198 patients with locally advanced or metastasized DTC received a pre-therapeutic blood dosimetry using 124I. To analyze the influence of the modes of TSH stimulation and the number of preceding RAITs on blood dose subgroups were built as follows: patients with endogenous TSH stimulation versus patients with exogenous TSH stimulation and patients with no preceding RAIT versus patients with at least one preceding RAIT. In 124/198 patients subsequent RAIT was performed. In 73/124 patients, hemograms were performed from day 2 to 12 month after RAIT. RESULTS: There was no high-grade bone marrow toxicity (i.e. ≥ grade 3) in patients receiving less than 2 Gy blood dose-independent of the therapeutic history. Within the first month after radioiodine therapy, there was an overall decrease in the white blood cell and platelet counts. The erythrocyte count was essentially stable. There was a correlation between cell count decrease and predicted blood doses (Spearman's correlation coefficient >-0.6 each) for the white cell line and the platelets. With regard to the subgroups, the blood dose per administered 131I activity (BDpA) was significantly higher in patients with endogenous TSH stimulation (median 0.08 Gy/GBq) than in patients with exogenous TSH stimulation (0.06 Gy/GBq) and in patients with no previous RAIT (0.08 Gy/GBq) compared to patients who had previously undergone at least one RAIT (0.07 Gy/GBq). CONCLUSIONS: The range of BDpA among DTC patients is rather wide. Our results suggest that lower blood doses can be expected when using exogenous TSH stimulation and blood doses are generally higher at first RAIT compared to subsequent RAITs. Thus, we advise to make blood dosimetry standard praxis prior to a high-activity RAIT.

Time course of tumor SUV in 18F-FDG PET of breast cancer: presentation of a simple model using a single reference point for time corrections of tumor SUVs.

UNLABELLED: Tumor standardized uptake values (SUVs) vary with the interval between 18F-FDG injection and image acquisition. This paper presents a simple method using a single reference point to make appropriate time corrections for tumor SUVs. METHODS: The reference point method was algebraically deduced from observations made by Beaulieu et al., who found that tumor SUVs behaved linearly over time (∼30 to 75 min after 18F-FDG injection). Eighteen patients with breast cancer were dynamically examined with PET/CT (∼60 and 80 min after 18F-FDG injection). Maximum SUV was calculated by applying 2 different iterative reconstruction methods (high-definition reconstruction and attenuation-weighted ordered-subsets expectation maximization). Reference points for time corrections were given, and errors for corrections obtained with the reference point method were calculated. RESULTS: Variations in the reconstruction algorithm strongly influenced the coordinates of the reference point. Time corrections using the reference point method were more accurate at higher tumor SUVs (>8 at high-definition reconstruction and>6 at attenuation-weighted ordered-subsets expectation maximization) than at lower ones. CONCLUSION: A common origin of tumor SUVs over time exists in breast cancer. In combination with the linear behavior of tumor SUVs between approximately 30 and 80 min, such a reference point allows for straightforward time corrections of tumor SUVs. Parameters for image reconstruction must be considered because they influence the coordinates of the reference point.

The XbaI G>T polymorphism of the glucose transporter 1 gene modulates 18F-FDG uptake and tumor aggressiveness in breast cancer.

UNLABELLED: We investigated the relevance of single-nucleotide polymorphisms (SNPs) in the glucose transporter 1 (GLUT1) gene to the uptake of (18)F-FDG and tumor aggressiveness in breast cancer. METHODS: In 52 individuals with breast cancer, a diagnostic PET/CT scan was obtained, and the standardized uptake value was determined as a measure of (18)F-FDG uptake using a region-of-interest technique. Three GLUT1 SNPs (XbaI G>T, HpyCH4V A>T, and HaeIII T>C) were investigated in genomic DNA that was isolated from the paraffin-embedded specimens of all patients. Tumors were typed and graded according to the World Health Organization classifications. RESULTS: The GG genotype of the XbaI G>T SNP was associated with increased tumor uptake of (18)F-FDG, with a mean standardized uptake value of 11.7 (TT/GT genotypes, 5.9; P = 0.03). Furthermore, the GG genotype was positively related to enhanced tumor proliferation (mitotic count, P = 0.01). In line with this finding, the GG genotype was absent in grade 1 carcinomas and increasingly prevalent in tumors with higher malignancy (grade 2, 28.0%; grade 3, 50%; P = 0.04). CONCLUSION: This study found that the XbaI G>T SNP of the GLUT1 gene is associated with an increased (18)F-FDG uptake and a more advanced tumor grade or growth in breast cancer. Thus, this genetic variant might favor aggressive phenotypes by modulating the efficiency of cancer cells to recruit glucose and escalate growth rate, suggesting the XbaI G>T SNP as a proliferation-related prognostic factor.

Incidental head and neck (18)F-FDG uptake on PET/CT without corresponding morphological lesion: early predictor of cancer development?

PURPOSE: To retrospectively determine whether increased/asymmetric FDG uptake on PET without a correlating morphological lesion on fully diagnostic CT indicates the development of a head and neck malignancy. METHODS: In 590 patients (mean age 55.4 +/- 13.3 years) without a head and neck malignancy/inflammation FDG uptake was measured at (a) Waldeyer's ring, (b) the oral floor, (c) the larynx, and (d) the thyroid gland, and rated as absent (group A), present (group B), symmetric (group B1) or asymmetric (group B2). Differences between groups A and B and between B1 and B2 were tested for significance with the U-test (p < 0.05). An average follow-up of about 2.5 years (mean 29.5 +/- 13.9 months) served as the reference period to determine whether patients developed a head and neck malignancy. RESULTS: Of the 590 patients, 235 (40%) showed no evidence of enhanced FDG uptake in any investigated site, and 355 (60%) showed qualitatively elevated FDG uptake in at least one site. FDG uptake values (SUV(max), mean+/-SD) for Waldeyer's ring were 3.0 +/- 0.89 in group A (n = 326), 4.5 +/- 2.18 in group B (n = 264; p < 0.01), 5.4 +/- 3.35 in group B1 (n = 177), and 4.1 +/- 1.7 in group B2 (n = 87; p < 0.01). Values for the oral floor were 2.8 +/- 0.74 in group A (n = 362), 4.7 +/- 2.55 in group B (n = 228; p < 0.01), 4.4 +/- 3.39 in group B1 (n = 130), and 5.1 +/- 2.69 in group B2 (n = 98, p = 0.01). Values for the larynx were 2.8 +/- 0.76 in group A (n = 353), 4.2 +/- 2.05 in group B (n = 237; p < 0.01), 4.0 +/- 2.02 in group B1 (n = 165), and 4.6 +/- 2.8 in group B2 (n = 72; p = 0.027). Values for the thyroid were 2.4 +/- 0.63 in group A (n = 404), 3.0 +/- 1.01 in group B (n = 186; p < 0.01), 2.6 +/- 0.39 in group B1 (n = 130), and 4.0 +/- 1.24 in group B2 (n = 56; p < 0.01). One patient developed a palatine tonsil carcinoma (group B1, SUV(max) 3.2), and one patient developed an oral floor carcinoma (group B1, SUV(max) 3.7). CONCLUSION: Elevated/asymmetric head and neck FDG accumulation without a correlating morphological lesion can frequently be found and does not predict cancer development. In populations in which goitre is endemic, FDG uptake by the thyroid is common and not associated with thyroid cancer.

A novel view on dosimetry-related radionuclide therapy: presentation of a calculatory model and its implementation for radioiodine therapy of metastasized differentiated thyroid carcinoma.

PURPOSE: Targeted radionuclide therapy is being increasingly used in malignancies. According to regulatory requirements patient-specific dosimetry must be performed in the context of any radiotherapeutic procedure. A calculatory model is presented demonstrating how to individualize radionuclide therapy through dosimetric data by weighing the success of therapy against risk. The model is exemplarily implemented for radioiodine therapy of differentiated thyroid carcinoma (DTC). METHODS: For DTC dose-response relationships were retrieved from the literature. From these data a three-variable model was developed which consists of a target variable weighing response against risk, a measured variable representing the lesion dose per activity [LDpA, in units gray per gigabecquerel (Gy/GBq)] and a manipulated variable constituting the therapeutic activity. RESULTS: Dosimetry-related radioiodine therapy along the three-variable model increases response probability in individual patients by up to > 50% (e.g. from 18 to 72% at a LDpA of 6 Gy/GBq) compared to "standard" therapy with 7 GBq. On a patient population scale, by escalating and de-escalating activity along the model, the overall response rate can be enhanced by 8% (62 vs 70%) while saving on average 0.9 GBq per patient (7 vs 6.1 GBq). CONCLUSION: Redistribution of therapeutic activities along the model, i.e. taking into account success and risk, may enhance response while on average saving activity as exemplarily shown by a virtual comparison with standard approaches using literature data from DTC for implementation. The model may thus provide a guideline for the prescription of therapeutic activities; the results underline the potential impact of individual dosimetry in radionuclide therapy.

PET/CT with Gluc-Lys-([(18)F]FP)-TOCA: correlation between uptake, size and arterial perfusion in somatostatin receptor positive lesions.

PURPOSE: Somatostatin receptor (sstr) positive tumours vary widely in uptake of radiolabelled somatostatin (sst) analogues. This study determinates variability in lesion uptake of the glycosylated sst analogon N(alpha)-(1-deoxy-D-fructosyl)-N(epsilon)-(2-[(18)F]fluoropropionyl)-Lys(0)-Tyr(3)-octreotate (Gluc-Lys([(18)F]FP)-TOCA) and correlates it with lesion size and arterial perfusion as measured on computed tomography (CT). METHODS: Ten patients with metastasized neuroendocrine carcinomas were investigated with positron emission tomography PET/CT (Biograph 16, Siemens, Germany). Lesion standardized uptake values (SUVs) were determined at approximately 50 min post tracer injection according to a 60% isocontour volume of interest around each lesion. Lesion size and enhancement in the arterial phase (hounsfield units, HUs) were derived from CT. RESULTS: 114 lesions in the upper abdomen had a correlate on both, PET and CT. Variability in lesion SUVs was high (SUV(mean) 22 +/- 13). Intraindividually, there was a sigmoid positive correlation between lesion SUV and lesion diameter indicating partial volume effects. Residual variability in lesions > or =3 cm (> or =2.5 cm) ranged down to about half (third) of the maximum lesion uptake and remained unexplained by partial volume effects. No correlation with measured HU in the arterial phase was found, neither intraindividually nor interindividually. CONCLUSION: Partial volume effects were a major source of intraindividual variability in tumour tracer uptake. Lesions below 2.5 to 3 cm should thus be used with caution when performing dose calculations. In larger lesions residual variability in uptake must be considered; it may be due to variable sstr2 expression on the tumours' cell surfaces.