Search

BACKGROUND Thyroid nodules are extremely common and typically diagnosed with ultrasound whether benign or malignant. Imaging diagnosis assisted by Artificial Intelligence has attracted much attention in recent years. The aim of our study was to build an ensemble deep learning classification model to accurately differentiate benign and malignant thyroid nodules. MATERIAL AND METHODS Based on current advanced methods of image segmentation and classification algorithms, we proposed an ensemble deep learning classification model for thyroid nodules (EDLC-TN) after precise localization. We compared diagnostic performance with four other state-of-the-art deep learning algorithms and three ultrasound radiologists according to ACR TI-RADS criteria. Finally, we demonstrated the general applicability of EDLC-TN for diagnosing thyroid cancer using ultrasound images from multi medical centers. RESULTS The method proposed in this paper has been trained and tested on a thyroid ultrasound image dataset containing 26 541 images and the accuracy of this method could reach 98.51%. EDLC-TN demonstrated the highest value for area under the curve, sensitivity, specificity, and accuracy among five state-of-the-art algorithms. Combining EDLC-TN with models and radiologists could improve diagnostic accuracy. EDLC-TN achieved excellent diagnostic performance when applied to ultrasound images from another independent hospital. CONCLUSIONS Based on ensemble deep learning, the proposed approach in this paper is superior to other similar existing methods of thyroid classification, as well as ultrasound radiologists. Moreover, our network represents a generalized platform that potentially can be applied to medical images from multiple medical centers.

Context: - Proposed noninvasive follicular thyroid neoplasm with papillary-like nuclear features (NIFTPs), formerly noninvasive encapsulated papillary carcinoma, follicular variant (PTC-FV), is an indolent tumor with follicular growth and frequent RAS mutations., Objective: - To detect histologic and molecular differences separating NIFTP from follicular adenomas (FAs) and invasive carcinomas, particularly papillary carcinomas with extensive follicular growth (PTC-EFGs) and invasive encapsulated PTC-FV (IE-PTC-FV)., Design: - Sixty-one tumors were reviewed histologically and reclassified into 32 NIFTPs (52%), 4 IE-PTC-FVs (7%), 14 PTC-EFGs (23%), and 11 FAs (18%). Next-generation sequencing for mutations in 50 genes was performed. Clinical outcomes were recorded., Results: - The NIFTPs and FAs were well circumscribed and unencapsulated. The FAs had bland nuclei, whereas the NIFTPs showed at least 2 of 3 (67%; sufficient) nuclear features (enlargement, irregular contours, chromatin clearing). The IE-PTC-FVs had follicular growth, sufficient nuclear features, and extensive capsular invasion. The PTC-EFGs had a median of 5% papillae with intrathyroidal invasion (broad-based, sclerotic, or small follicle growth patterns); intranuclear pseudoinclusions were present only in PTC-EFGs (9 of 14; 64%). Mutations included RAS in 20 of the 32 NIFTPs (62%), 4 of the 11 FAs (36%), and 3 of the 4 IE-PTC-FVs (75%); BRAF K601E in 1 NIFTP (3%); BRAF V600E in 5 PTC-EFGs (36%). No NIFTPs or FAs recurred or metastasized. All 4 IE-PTC-FVs (100%) had hematogenous metastasis. Two PTC-EFGs (14%) had lymphatic metastasis., Conclusions: - The morphologic similarity and RAS mutations in FAs, NIFTPs, and IE-PTC-FVs supports the genetic similarity of those follicular neoplasms in contrast to the unique presence of BRAF V600E mutations in PTC-EFGs. Using strict diagnostic criteria supported by molecular testing, tumors with extensive follicular growth can be classified into follicular type or RAS-like (FA, NIFTP, IE-PTC-FV) versus papillary type or BRAF V600E-like (PTC-EFG).

Context: Papillary thyroid carcinoma (PTC) is the most common thyroid malignancy accounting for 80% of the thyroid cancers. Many histopathologic variants of PTC have been recognized, and few of these are of prognostic significance. The studies on clinicopathological features of PTC and its variants are so far seldom reported in India., Aim: The aim of the study was to study the percentage distribution of PTC among total thyroid specimens, the age and sex distribution of PTC, its histopathological features including frequency of nuclear findings, and various histological subtypes are also studied in detail. Methods: All cases of PTC diagnosed in our department from April 2003 to March 2013 formed the material for the study. The tissues were routinely processed and stained. On microscopic examination, tumors were classified according to 2004 WHO classification., Results: PTC formed the predominant type of malignancy accounting to 71% of the total cases. Of these, about 75% of patients were in the second to fifth decade. Male to female ratio was 1:5.4. Other than the usual classic variant and follicular variant, we also found rare types such as clear cell variant, tall cell type, oncocytic type, and macrofollicular variant. Microscopically, nuclear overcrowding and ground glass nuclei were seen in more than 90% of cases. Nodular goiter, Hashimoto's thyroiditis, and follicular adenoma were associated lesions in some cases., Conclusion: PTC is the most common thyroid malignancy, and it can affect any age group though it presents mostly in the third to fourth decade of life. Recognition of histological subtype is crucial in patient prognosis., Competing Interests: There are no conflicts of interest

Background: Thyroid nodules with atypia of undetermined significance (AUS) on fine-needle aspiration (FNA) have a low risk of malignancy that appears to vary based on specific features described in the AUS diagnosis. The Afirma gene expression classifier (GEC) is a molecular test designed to improve preoperative risk stratification of thyroid nodules, but its performance for different patterns of AUS has not been defined. The objective of this study was to assess GEC results and clinical outcomes in AUS nodules with architectural atypia (AUS-A), cytologic atypia (AUS-C) or both (AUS-C/A)., Methods: This was a retrospective review of all thyroid nodules with AUS cytopathology that underwent GEC testing at the authors' institution over a period of >4 years., Results: In 227 nodules that had AUS cytology results and Afirma GEC testing, the rate of benign GEC results was higher in AUS-A nodules (70 of 107; 65%) than in AUS-C/A nodules (25 of 65; 38%; P = .0008), and AUS-C nodules exhibited an intermediate rate of benign results (27 of 55 nodules; 59%). The risk of cancer among patients who had GEC-suspicious nodules, 86% of whom underwent resection, was 19% (6 of 25) for AUS-A nodules compared with 57% (21 of 37) for AUS-C/A nodules (P = .003) and 45% (10 of 22) for AUS-C nodules (P = .07). In nodules that had an indeterminate repeat cytology result, no difference was observed in the rate of benign GEC results or in the malignancy rate compared with nodules that had a single cytology result., Conclusions: The performance characteristics of Afirma GEC testing vary, depending on qualifiers of cytologic atypia. Recognition of these differences may enable clinicians to provide improved counseling and treatment recommendations to patients. Cancer Cytopathol 2017;125:313-322. © 2017 American Cancer Society., (© 2017 The Authors. Cancer Cytopathology published by Wiley Periodicals, Inc. on behalf of American Cancer Society.)

Thyroid tumors of follicular cell derivation are increasing in incidence. These lesions exhibit a spectrum of morphologic and behavioral features that provide the opportunity to understand malignant transformation and progression. Molecular data suggest that the thyroid undergoes a series of genetic alterations that account for the development of the various types of thyroid carcinoma. Our understanding of these tumors has progressed dramatically over the past 50 years and the classification has become complex and cumbersome. We provide a practical approach to clinical diagnosis and propose a simplified classification of these common neoplasms., (© 2013 Elsevier Inc. All rights reserved.)

The diagnosis of follicular epithelial neoplasms is an area of controversy. We provide our experience with common problems that practising pathologists face when confronted with follicular epithelial proliferations. One of the major issues is the recognition of the diagnostic nuclear features of papillary thyroid carcinoma and reactive cytologic atypia. We discuss the definitions of capsular invasion, vascular invasion, and extrathyroidal extension and their implications in cancer diagnosis and staging. We propose unified terminology for benign follicular epithelial proliferations in the setting of multinodular goiter. We also review challenges related to oncocytic change, malignant transformation in benign nodules, focal dedifferentiation, and the application of ancillary tools in thyroid pathology. We believe that this review contains comprehensive and up to date information that will be of value to pathologists who practice surgical pathology of thyroid.

The purpose of this study was to evaluate the immunohistochemical characteristics of thyroid tumors with a prominent hyalinizing trabecular pattern (TTHTP). Immunohistochemical stains for thyroid transcription factor 1 (TTF-1), thyroglobulin, CD56, cytokeratin (CK) 19, galectin-3, and Ki-67 were performed on 16 cases of TTHTP. All TTHTPs were positive for TTF-1 and thyroglobulin, and nine cases expressed CD56 and Ki-67. Of these nine cases, seven satisfied the strict criteria of Carney, and four cases were positive for CK19 or/and galectin-3 and negative for Ki-67 and CD56. The remaining three cases had no Ki-67, CD56, CK19 or galectin-3 expression. All TTHTPs meeting the criteria of Carney were hyalinizing trabecular tumors (HTT), and TTHTPs which did not satisfy the criteria of Carney were subclassified according to the immunohistochemical results as the HTT type [Ki-67 (+) and CD56 (+)], papillary thyroid carcinoma (PTC) type [CK19 or galectin-3 (+)/Ki-67 (-) and CD56 (-)], or null type [Ki-67 (-), CD56 (-), CK19 (-) and galectin-3 (-)]. TTHTPs with typical histologic features and/or with typical immunohistochemical manifestations can be categorized as HTT, and the remaining TTHTPs should be categorized as the PTC or null type., (© 2011 The Authors. APMIS © 2011 APMIS.)

Background/objective: Focal papillary thyroid carcinoma (PTC) arising within a follicular adenoma (PTCFA) represents a clinically significant, but rare, histopathologic subset of papillary carcinomas whose cytologic features have not been well described. This uncommon presentation of PTC may contribute to a subset of thyroid aspirates interpreted as 'atypia of undetermined significance/follicular lesion of undetermined significance' (AUS/FLUS)., Study Design: Seventeen fine-needle aspiration biopsy (FNAB) cases diagnosed as 'PTCFA' on corresponding surgical excision were identified from the archival records of 2 large academic medical centers. A control group of 40 FNAB comprised of 20 follicular adenomas (FA) and 20 PTC was identified (based on the corresponding surgical pathology diagnosis) for comparison. All 57 FNAB were reviewed in a masked fashion and scored for a series of 31 cytomorphologic features. The intraclass correlation between diagnostic categories and overall agreement between cytopathologists was statistically evaluated., Results: Aspirates of PTCFA were originally diagnosed as 'negative' (n = 3), 'AUS/FLUS' (n = 7), 'suspicious for a follicular neoplasm' (n = 3), 'suspicious for malignancy' (n = 3), and 'malignant' (n = 1). On masked review, the most common cytomorphologic features of PTCFA were a nonmacrofollicular cytoarchitectural pattern (71%), medium-large cell size (74%), and micronucleoli (79%). Intranuclear pseudoinclusions and a papillary architecture were absent in 85 and 88% of the cases, respectively. Relative to the 2 control groups, the PTCFA cases demonstrated overlapping features between FA and PTC for the majority of the 31 examined cytomorphologic features., Conclusion: PTCFA represent a rare subset of PTC that is difficult to recognize as PTC by FNAB. Most cases exhibit overlapping features between a benign thyroid nodule and conventional PTC, and they are often interpreted as 'AUS/FLUS'., (Copyright © 2011 S. Karger AG, Basel.)

Aims: To separate true oncocytic neoplasms from mitochondrion-rich non-oncocytic lesions based on the intracellular relationship between major cell organelles, and to establish the diagnostic and clinical relevance of this distinction., Methods and Results: Tissue samples from 276 follicular adenomas, 194 follicular carcinomas, 162 normal thyroids and 296 non-neoplastic lesions were classified as conventional, mitochondrion-rich or oncocytic based on the immunohistochemically assessed quantity and intracellular distribution of mitochondria and endoplasmic reticulum (ER) and nuclear position. Pathological and clinical features were compared among the groups. In oncocytes, densely packed mitochondria resulted in homogeneous immunolabelling of basal cytoplasmic regions, whereas ER and the nuclei were typically displaced to the apical position. This aberrant organelle distribution was not observed in non-oncocytes, which allowed reliable distinction between oncocytic and mitochondrion-rich lesions. Clinically, mitochondrial increase in non-oncocytic lesions was associated with neoplasia, malignancy and higher cancer recurrence rates. Similar correlation, albeit less pronounced, was observed within the oncocytic tumour group. By contrast, oncocytic change per se was not associated with neoplasia, malignancy or cancer aggressiveness., Conclusions: True oncocytic neoplasms can be distinguished from mitochondrion-rich non-oncocytic tumours based on aberrant distribution of all major cell organelles. This distinction has immediate clinical relevance and should be implemented in practice.

Hyalinizing trabecular tumors of the thyroid have been described on 4 occasions, by Carney and colleagues in 1987, by Ward and coworkers in 1982, by Pierre Masson in 1922, and by Rahel Zipkin in 1905. Zipkin credited her chief, Theodor Langhans (of Langhans giant cell fame), with identification of the cases she reported. Unaware of the 3 earlier descriptions, Carney and colleagues described 11 circumscribed or encapsulated thyroid tumors with elongated and polygonal cells arranged in trabeculae that contained a hyaline material resembling amyloid. The nuclei of the tumor cells had cytoplasmic invaginations and grooves similar to those of papillary carcinoma. Carney and colleagues labeled the neoplasms hyalinizing trabecular adenomas because of their microscopic appearance, absence of invasion, and benign natural history. Subsequently, the nuclear features of the tumor and the molecular genetic findings led to the introduction of equivocal designations for it, hyalinizing trabecular tumor and hyalinizing trabecular neoplasm, and later to its designation as a variant of papillary carcinoma. Experience has shown that most circumscribed or encapsulated follicular thyroid tumors with intratrabecular hyalin and nuclear features of papillary carcinoma behave as benign neoplasms. Hyalinizing trabecular carcinoma is a very rare tumor.

Objective: To analyze smears of 197 thyroid follicular tumors (adenoma and carcinoma)., Study Design: Several types of artificial neural networks (ANN) of various designs were used for diagnosis of thyroid follicular tumors. The typical complex of cytologic features, some nuclear morphometric parameters (area, perimeter, shape factor) and density features of chromatin texture (mean value and SD of gray levels) were defined for each tumor., Results: The ANN was trained by means of cytologic features characteristic for a thyroid follicular adenoma and a follicular carcinoma. At subsequent testing, the correct cytologic diagnosis was established in 93% (25 of 27) of cases. The morphometry increased the accuracy of diagnosis for follicular tumors in up to 97% (75 of 78) of cases. ANN correctly distinguished an adenoma or a carcinoma in 87% (73 of 84) of cases when using color microscopic images of tumors., Conclusion: The usage of ANN has raised sensitivity of cytologic diagnosis of follicular tumors to 90%, compared with a usual cytologic method (sensitivity of 56%). The automatic classification of thyroid follicular tumors by means of ANN is prospective.

Tumors of the thyroid characterized by a follicular growth pattern constitute the most common type of lesion of this organ encountered by pathologists. The vast majority of such lesions do not pose difficulties for histopathologic interpretation. A subset of these tumors, however, can represent a serious challenge for diagnosis. Thyroid tumors with a follicular growth pattern include a broad range of lesions that range from benign, hyperplastic nodules to follicular adenomas to follicular carcinomas. In addition, other types of tumors belonging in separate diagnostic categories can also present histologically with a follicular growth pattern, including the follicular variant of papillary thyroid carcinoma and medullary carcinoma. The histologic features and diagnostic criteria used for distinguishing among these conditions can often be subtle and subjective.

Background: Nodules of the thyroid gland are observed frequently in patients who undergo ultrasound studies. The majority of these nodules are benign, corresponding to goiters or adenomas, and only a small fraction corresponds to carcinomas. Among thyroid tumors, the diagnosis of follicular adenocarcinomas by preoperative fine-needle aspiration biopsy is a major challenge, because it requires inspection of the entire capsule to differentiate it from adenoma. Consequently, large numbers of patients undergo unnecessary thyroidectomy., Methods: Using data from gene expression analysis, the authors applied Fisher linear discriminant analysis and searched for expression signatures of individual samples of adenomas and follicular carcinomas that could be used as molecular classifiers for the precise classification of malignant and nonmalignant lesions., Results: Fourteen trios of genes were described that fulfilled the criteria for the correct classification of 100% of samples. The robustness of these trios was verified by using leave-1-out cross-validation and bootstrap analyses. The results demonstrated that, by combining trios, better classifiers could be generated that correctly classified >92% of samples., Conclusions: The strategy of classifiers based on individual signatures was a useful strategy for distinguishing between samples with very similar expression profiles.

A novel scheme for the classification of well-differentiated thyroid tumors composed of follicular cells and having a follicular pattern of growth is presented. In addition to the two conventional categories, i.e., adenoma and carcinoma, this scheme includes a group of tumors "of uncertain malignant potential."

Background: While the majority of minimally invasive follicular thyroid carcinoma (MIFTC) behave like follicular adenomas, some recur or metastasize. These studies were conducted to determine if molecular profiling can enhance our understanding of MIFTC and to perhaps offer a better classification schema., Methods: Microarray analysis was performed on 27 follicular neoplasms. Thirteen follicular adenomas (FAs) were compared with 7 widely invasive follicular thyroid carcinomas (WIFTCs) to generate a list of differentially expressed genes. Next, 7 MIFTCs were analyzed along with the other samples in a cluster analysis. The MIFTCs were then compared directly against both the adenoma and WIFTC groups to investigate genetic relatedness., Results: FAs and WIFTCs have distinct genetic profiles, with 401 differentially expressed genes. The 7 MIFTCs were added to the analysis. Six of 7 of the MIFTCs were grouped with the adenomas, 4 of which created their own subgroup. When analyzed directly, MIFTCs had 223 differently expressed genes, compared with FAs, and 365, compared with WIFTCs., Conclusions: Molecular profiling illustrates that the majority of MIFTCs comprise a subclass of follicular neoplasms, and, while most MIFTCs are genetically similar to adenomas, our patient data suggest that these tumors may deserve greater suspicion of malignant potential. Gene profiling can provide insight into the molecular pathogenesis of MIFTC.

Background: Thyroid nodules are common; fine-needle aspirations commonly are read as indeterminate, necessitating surgery to exclude carcinoma. We developed a 6-gene array-based predictor model to diagnose benign versus malignant thyroid lesions. In this study, we verified whether quantitative reverse transcription-polymerase chain reaction (qRT-PCR) using this model reliably can differentiate benign from malignant thyroid nodules., Methods: Molecular profiles of benign (follicular adenomas, hyperplastic nodules) and malignant tumors (papillary thyroid carcinomas, follicular variants of papillary thyroid carcinomas) were analyzed using qRT-PCR from our 6-gene model (kit, Hs.296031, Hs.24183, LSM7, SYNGR2, C21orf4). The gold standard was standard pathologic criteria. A diagnosis-predictor model was built by using the training samples and was then used to predict the class of 10 additional samples analyzed as unknowns., Results: Our predictor model using 47 training samples correctly predicted 9/10 unknowns. One sample diagnosed as benign by standard histologic criteria was diagnosed as malignant by our model (sensitivity 75%; specificity, 100%; positive predictive value, 100%; negative predictive value, 85.7%)., Conclusions: Molecular diagnosis with our 6-gene model can differentiate between benign and malignant thyroid tumors with high sensitivity and specificity. In combination, these genetic markers may be a reliable test to preoperatively diagnose the malignant potential of thyroid nodules.

The implications of global DNA hypomethylation were recently reported in several models of tumorigenesis. Little is known about this epigenetic event in thyroid neoplasia. The study aimed to evaluate the status of global DNA methylation in several types of thyroid tumors using a monoclonal antibody specific for 5-methylcytidine (5-mc) and to define the diagnosis potential of this marker. 5-mc immunostaining scores were calculated in 17 papillary thyroid carcinomas (PTC), 6 follicular thyroid carcinomas (FTC), 16 follicular adenomas (FA), 19 nodular goiters (NG) and ten Hürthle cells adenomas (HCA). The expression of galectin-3 was also evaluated. Computerized image analysis showed a significant lower level of 5-mc immunostaining in thyroid carcinoma when compared with benign tumors or adjacent normal thyroid parenchyma (P<0.0001). Overall, 5-mc accuracy to distinguish malign from benign thyroid tumors was similar to that of galectin-3 (89% versus 87%, P>0.05). The combination of 5-mc with galectin-3 led to an excellent accuracy level of 96%. Among follicular neoplasia 5-mc accuracy to differentiate malign tumors trends to be higher than galectin-3 one (90% versus 66%, P=0.06). These data stress the necessity of epigenetic events evaluation among thyroid nodules and propose global DNA methylation assessment as a potential diagnostic tool to combine with other valuable markers.

The precise diagnosis of follicular thyroid lesions is frequently debated because of the subjective nature of capsular invasion as well as both the histological and cytological characteristics. Furthermore, several different prognostic indices have been devised to examine prognosis associated with thyroid cancer. Herein, we describe how these confounding elements can affect the ability to accurately predict prognosis for patients with follicular thyroid lesions., ((c) 2005 Wiley-Liss, Inc.)

The authors review the group of thyroid tumors characterized by a follicular growth pattern; these include follicular adenoma, follicular carcinoma, and the follicular variant of papillary carcinoma. Most of these lesions can be diagnosed with ease, but a subgroup has generated recent controversy in the literature. The authors present their views based on their experience with the cytologic and histologic diagnosis of these tumors and propose a scheme to assist in their classification and appropriate clinical management.

In this study, we investigated the sensitivity, specificity, and predictive values of morphometric parameters of thyroideal follicular neoplasms based on concepts of fractal geometry. Thirty-seven follicular adenomas and 36 well-differentiated follicular carcinomas were assessed morphometrically. The nuclear area, nuclear area fraction, nuclear regularity factor, nuclear elongation factor, and slope setting (representing the ratio between the nuclear perimeter and nuclear regularity factor) were subjected to fractal dimensions analysis. By univariate analysis, the nuclear area, nuclear area fraction, nuclear regularity factor and slope values discriminate between adenomas and carcinomas. By multivariate analysis, the nuclear area, nuclear area fraction and slope values possess significant discriminatory powers in distinguishing between adenomas and carcinomas. Incorporating the nuclear area, nuclear area fraction, and slope values leads to a discriminatory power with 92% specificity and 83% sensitivity. The reciprocal relationships between the nuclear area, nuclear perimeter, and nuclear regularity factor of the cells of thyroideal adenomas and carcinomas may be expressed by fractal dimensions. Analysis limited to one parameter provides incomplete data. Expressing variations of the nuclear perimeter as a function of the nuclear regularity factor, the slope values constitute an independent attribute that significantly differentiates thyroideal adenomas from carcinomas.

Surgical (374 cases) and puncture (6148 cases) material from patients with multinodular goiter was studied. Definitions and a working classification characterizing morphological and cytological signs of adenomatous goiter were formulated. Prognostic significance of thyroid adenomatosis and approaches to its treatment in multinodular goiter are shown.

Background: In the works of various authors, performed with the help of morphometry, an assessment of cell changes was carried out, as a rule, by one sign of their atypia (e.g. area of cell nuclei). Moreover, the selective analysis of morphometric data for certain nosologic forms was performed. Such an approach allowed researchers to develop only additional diagnostic criteria of thyroid carcinoma. At the same time, the use of a set of quantitative signs of cell atypia as a criterion for tumor malignancy made it possible to improve oncologic diagnosis. In this study, which is a first step in the development of a quantitative approach to thyroid carcinoma diagnosis, a system of karyometric parameters was developed with the help of morphometry of cytologic material of patients with major nosologic forms of thyroid pathology (papillary and follicular carcinoma, follicular adenoma, autoimmune thyroiditis, and nodular colloid and diffuse toxic goiter). The significance of each parameter in thyroid carcinoma diagnosis was evaluated., Methods: Thyroid cells of autopsy, operative, and aspiration biopsy material were the object of the study. One hundred fifty-two preparations with histologically verified diagnosis were studied. Light morphometry was performed with the help of a semiautomatic computer analyzer of images. One hundred fifty thyroid nuclei were assessed in each preparation., Results: Among many karyometric parameters, six were selected: mean area of thyrocyte nuclei, ratio of frequencies, and number of classes on histograms of distribution of their area, coefficient and a free term of the regression equation to a scattered diagram of perimeter and nuclear area, and the discriminant index. Boundary values of these parameters between the malignant and benign pathology were united into an expert system for thyroid carcinoma diagnosis. Clinical trials of the system were performed. On the basis of the obtained results, the informativity (accuracy, sensitivity, specificity) of its karyometric parameters was assessed. The formula for calculation of the weighting coefficient for each parameter was derived, the diagnostic index being determined by the sum of their values., Conclusions: An expert system was created based on a set of karyometric parameters reflecting the regularities of pathologic changes in thyrocyte nuclei with malignant and benign pathology. The informativity and weighting coefficients included into the system of parameters have been assessed, which made it possible to perform the probabilistic diagnosis of thyroid carcinoma., (Copyright 2001 American Cancer Society.)

Objective: To investigate the capability of the learning vector quantizer (LVQ) in the discrimination of benign from malignant thyroid lesions., Study Design: The study was performed on May-Grünwald-Giemsa-stained smears taken by fine needle aspiration (FNA). Using a custom image analysis system, 25 features that describe the size, shape and texture of approximately 100 nuclei were measured from each case. Statistical features were extracted from each case, and a linear regression analysis was performed to detect the statistically significant features. The cases were distributed by category, as follows: 100 cases of goiter and follicular adenomas, 11 cases of follicular carcinoma, 35 cases of papillary carcinoma, 24 cases of oncocytic adenoma, 8 cases of oncocytic carcinoma and 20 cases of Hashimoto thyroiditis. About 30% of the cases from each class were used for training two LVQ classifiers. The remaining 139 cases, out of a total of 198, were used as the test set. A classifier was used to discriminate into four classes and a second into two classes., Results: The application of LVQ neural networks allows good discrimination between benign and malignant lesions (O.A. = 97.8). However, reliable discrimination of the cytologic types of the lesions was not obtained., Conclusion: These results indicate that the use of neural networks combined with image morphometry may offer useful information on the potential for malignancy of thyroid lesions and may improve the diagnostic accuracy of FNA of the thyroid gland, especially in cases of follicular neoplasms classified as suspicious for malignancy and in cases of oncocytic tumors.

Background: The current study was undertaken to evaluate the quantitative estimation of cytologic features on aspirated smears for the preoperative differential diagnosis of follicular lesions of the thyroid., Methods: The subjects were 60 patients with follicular lesions of the thyroid (including 20 follicular carcinomas, 15 follicular adenomas, and 25 adenomatous goiters) whose histopathologic explorations were conducted fully postoperatively. Using a microscope connected to a computerized video system, the mean nuclear area, the mean nuclear perimeter, the circular rate, the largest to the smallest dimension ratio (LS ratio) of the nuclei, and the coefficient of variation of the nuclear area (NACV) were measured and analyzed., Results: Among the quantitative morphometric parameters of nuclei, the circular rate was significantly higher in the group with adenomatous goiters than those with follicular carcinomas (P < 0.00001) and adenomas (P < 0.005). The group with follicular carcinomas had a higher LS ratio than the group with adenomatous goiters (P < 0.0005). The NACV value increased as the malignant potential of the lesion increased and showed significant differences between the groups. When a NACV of 21.5% was chosen as the cutoff point, the incidence of malignancy was significantly higher in patients with high NACV values than in those with low NACV values (P < 0.00001). Using this borderline value, it was possible to distinguish malignant from benign diseases with a sensitivity of 85.0%, a specificity of 82.5%, and an accuracy of 83.3%., Conclusions: Preoperative quantitative estimations of cytologic nuclear features are useful for the preoperative differential diagnosis of follicular lesions of the thyroid.

To determine the diagnostic implications of measuring the serum thyroglobulin level in patients with a solitary follicular thyroid tumor, a retrospective study was conducted on 122 consecutive patients with a solitary follicular thyroid nodule who underwent thyroidectomy. Data for eight variables were collected: the serum thyroglobulin (Tg) level (microgram/l), age, maximum diameter of the nodule, gender, histopathologic type, presence or absence of metastases, macroscopic characteristics of the cut surface of the resected tumor, and smoking habit. Multiple regression analyses were used to investigate the relationships between the serum Tg level and the seven other variables. The diagnostic value of serum Tg was examined by means of receiver operating characteristic (ROC) curves. There were significant correlations between the serum Tg level and the maximum diameter of the nodule, the macroscopic characteristics, and the smoking habit. The sensitivity and specificity of the serum Tg level with a cut-off value of 1,000 micrograms/l were 57% and 86%, respectively. The likelihood ratio favouring follicular carcinoma associated with the serum Tg > 1,000 micrograms/l was 4.41. Measuring the level of serum Tg may be useful in discriminating follicular carcinoma from follicular adenoma, but since there may be some biases in this retrospective study, the results are less definitive. Further research activities are mandatory to obtain valid evidence.

Clinical, biological, scintigraphic (99mTc, 123I, 201Tl) and histological data were reviewed in 90 patients operated for autonomous adenomas (hot nodules). Analysis of the data gave the following results: 74.4% of the nodules were solitary and 25.6% occurred in a multinodular goiter (MNG). Hyperthyroidism (FT3-, FT4-, TSH < 0.1) was present in 44% of solitary nodules and in 15% of MNG. 16.6% of patients were euthyroid, and 22.2% had a TSH < 0.1 and normal free hormones. The 201 Tl scintigraphy, performed instead of the administration of exogenous TSH, showed and uptake of Tl in the remainder of the thyroid. Four types of 201Tl uptake were observed in the pathological lobe: type 1: homogeneous uptake: type 2: heterogeneous uptake and hyperfixation in the nodule; type 3: faint uptake and hypofixation in the nodule; type 4: faint and very heterogeneous uptake. Histological analysis of the nodules showed 32.2% of follicular adenomas, 25.6% of MNG and 42.2% of colloid nodules. Among these lesions, 70% showed histological signs of hypermetabolism (resorption vesicles ...). Multivariate analysis of these data clearly demonstrates the need to redefine the clinical, biological, scintigraphic and histological concepts of autonomous adenomas. On the other hand, 91.6% of follicular adenomas show a type 2 Thallium uptake (22/24). On basis of these results, we prefer to operate these patients rather than to treat them by 131I or by percutaneous injection of ethanol.

This article reviews the pathology and classification of differentiated thyroid carcinomas. Diagnostic criteria are reviewed. Follicular and papillary carcinomas and their variants are described. New entities are detailed. Atypical adenomas and difficult follicular diseases are discussed. The need for a consensual terminology is assessed.

Treatment of thyroidal oxyphilic (Hurthle cell) tumors is controversial due to difficulties in grading tumor malignancy. We classify these tumors into 3 histological subtypes: adenomas, atypical adenomas, and carcinomas. The purpose of this study was to find out whether this classification is a useful criterion for determination of the extent of surgery. Our retrospective study included 5 cases of adenoma, 2 of atypical adenoma and 13 of carcinoma. All histological specimens were revised and were stained immunohistochemically for thyroglobulin. All specimens were positive for thyroglobulin; staining for factor VIII was performed in cases in which the tumor had penetrated into blood vessels. Hemithyroidectomy was used for typical adenomas, while atypical adenomas were treated as carcinomas and total thyroidectomy was performed. 5 patients were treated with radioactive iodine. Taking into account the clinical results, we conclude that histological subtype is a useful guide for proper surgical management. Hemithyroidectomy is proper therapy for adenomas. The management of atypical adenomas was not determined as all were treated as carcinomas.

Between February 1986 and January 1991, a total of 340 patients with thyroid diseases were seen; 115 (33.8%) had solitary nodules. Of the latter, 28 had toxic adenoma and 87 had euthyroid nodules (M:F = 0.14:7.21); the mean age for the non-toxic cases was 32.8 +/- 10.0 years and for the toxic cases, 40.2 +/- 15.2 years. Clinically, the toxic nodules were 0.5 to 6 cm in diameter, cystic in over 36% and firm in the rest; no regional lymph node enlargement was noted. Thyroid scan in 45 (52%) patients showed 38 'cold', 4 hypofunctioning and 3 hyperfunctioning nodules. Ultrasonography in 22 patients (25.3%) revealed 9 cystic and 13 solid nodules, consistent with the clinical finding. Fine needle aspiration biopsy (FNAB) in 61 patients (70.1%) showed 30 cystic, 15 benign, 2 follicular adenoma, 2 Hürtle cell adenoma, 1 papillary carcinoma and 1 thyroiditis. Ten patients were operated; histology confirmed the cytology diagnosis in all but 1 in whom the cytology was benign but histology showed papillary carcinoma. The fine needle aspiration was also useful in the treatment of cystic lesions; hence it should be adopted particularly in areas where there is bed shortage.

The differential diagnosis of thyroid neoplasms by routine cytology presents major difficulties. We therefore looked for measurable nuclear parameters that could be generated from Feulgen-stained smears obtained by fine needle aspiration biopsy (FNAB). These parameters would then be used to differentiate between benign and malignant lesions. Seventy-six patients whose cold thyroid nodule was surgically excised after FNAB and examined by a pathologist were used in this study: 56 benign, 18 malignant and 2 atypical adenomas. A set of 3,662 cells from the 33 benign cases was compared with the set of 1,712 cells from the 11 malignant nodules. Discrimination between the two populations was based on four nuclear features ranked according to their discriminating power. The first ranked was a textural parameter, followed by a densitometric and finally two other textural parameters. The rate of nuclei that could be regarded as benign was computed for each case using the ranked parameters. The average rate for the 33 benign cases was 85% (50.2 = 100%, set at the 95% confidence interval). In a prospective study of the 32 remaining patients, sensitivity and specificity were, respectively, 92% and 88%, which were higher than those obtained by conventional cytology.

The chemical and immunological properties of thyroglobulin (Tg) in tissue obtained from a patient co-existed with two types of thyroid tumors, i.e., functioning and non-functioning, and were compared with the properties of Tg that was isolated from adjacent peripheral tissue. In the present observations, the Tg content was markedly increased in the non-functioning thyroid tumor. On the other hand, the Tg content in the functioning tumor was at the normal level. The iodine content of Tg was significantly lower in the non-functioning tumor than in peripheral tissues. Affinity with Lectins differed among Tg preparations, suggesting that the carbohydrate chain in the Tg was different in each nodule in a single individual.

The study was done on cytologic material of 58 non-oxyphil follicular neoplasias of the thyroid, 32 of which were adenomas and 26 carcinomas. Three groups of nucleolar features were quantified using a routine microscope with an ocular micrometer: frequency-, size-, and margination-related features. Since value overlap was present between two categories for all the variables, stepwise discriminant analysis was applied. The following three features were selected by the computer for calculation of one canonical discriminant function: percentage of marginated nucleoli, percentage of nuclei with one nucleolus, mean major nucleolar diameter. The percentage of agreement between morphologic and computer classifications was 95%. Two follicular adenomas were allocated to the carcinoma category, whereas one follicular carcinoma was allocated to the adenoma category. Out of 58, 52 were diagnosed by the computer into one of the two diagnostic categories with a very high probability, i.e. P greater than 0.75, the remaining 6 being considered intermediate.

This study was conducted on fine-needle aspirates of well differentiated follicular neoplasms of the thyroid. A 'decision tree' classification based on the percentage of nucleolated nuclei, percentage of nuclei with two or more nucleoli and mean major nuclear diameter was adopted. We observed that the reproducibility and the validity of the follicular adenoma vs follicular carcinoma discrimination are greater than in the subjective evaluation. Moreover, similar classification results were obtained when measurements were performed either with a fully automated image analysis system or with semiautomatic or manual instrumentation. As for reproducibility of the inter-instrument comparisons, the k statistic values ranged from 0.85 to 1.00 (mean value 0.90, that is, an 'almost perfect' degree of agreement); in the subjective evaluations, the inter-observer comparisons showed values ranging from 0.20 to 0.56 (mean value 0.37, that is, 'fair'). In the decision tree classification, feature value thresholds were selected in order to have specificity of 100% and the predictive value of a positive result (carcinoma) of 100%; accuracy was 87% (range 86-89%), sensitivity 74% (71-79%), the predictive value of a negative result (adenoma) 79% (78-82%). In the subjective evaluation the values were as follows: accuracy 67% (64-71%), sensitivity 57% (50-64%), specificity 77% (71-79%), predictive value of a negative result (adenoma) 64% (61-69%), predictive value of a positive result (carcinoma) 71% (67-75%). The conclusion is that, by using a routine microscope equipped with a micrometer, the preoperative diagnosis of follicular carcinoma from smears can be formulated with a high degree of certainty.

Twenty-five Hürthle cell tumors of the thyroid gland, histopathologically classified into three groups--adenomas, carcinomas, and indeterminant--have been studied by DNA flow cytometry using archived, paraffin-embedded tissues. Tumor ploidy characteristics were correlated with patient follow-up and survival with the conclusions that (1) nuclear DNA ploidy alone does not distinguish benign from malignant Hürthle cell tumors; (2) diploid DNA Hürthle cell carcinomas behave far less aggressively than aneuploid Hürthle cell carcinomas; and (3) all patients with aneuploid carcinomas died of their disease or are alive with persistent carcinoma.

Thirty-one cases of medullary carcinoma of the thyroid have been studied over the past fifteen years at the University Hospital of the Canton of Vaud, Switzerland (CHUV). Twenty cases were of sporadic nature and eleven presented as part of the familial MEN II syndrome (multiple endocrine neoplasia), one of which showed the classical features of the rare MEN IIb type. It is important to distinguish between the familial and sporadic cases, because membership of the former group implies the investigation of associated endocrinopathies (pheochromocytoma, hyperparathyreoidism) and study of the family tree as the syndrome is autosomal dominant. Medullary carcinoma of the thyroid is a constant feature of the MEN II syndrome and is the cause of premature death in these patients. The familial type should be suspected if the carcinoma appears early in life, is located in the superior pole of the thyroid or is bilateral or multicentric, if the histology shows hyperplasia of the C cells and, of course, if there is a history of surgery for pheochromocytoma or hyperparathyroidism. Although total thyroidecomy is the rule for these familial cases, its role is debatable in sporadic medullary carcinoma of the thyroid. Postoperative follow-up of these patients is based on serum calcitonin determination, as this is an extremely sensitive marker. The ten year survival rate is 50%, with the worst prognosis in MEN IIb type.

During a 25-year period (1959 through 1983), 54 patients with Hürthle cell tumors were treated and monitored at the University of Chicago Medical Center. Thirty percent were men and 70% were women; mean age at diagnosis was 46.7 +/- 13.2 years (range: 19 to 69 years). Tumors were grouped into three categories at the time of initial diagnosis: group 1, grossly malignant (four patients, or 7.5%); group 2, intermediate (partial capsular and/or subcapsular vascular invasion) (10 patients, or 18.5%); and group 3, benign appearing (40 patients, or 74%). Twenty-one (39%) of the patients had a history of low-dose, external radiation to the head and neck in childhood (including three of four grossly malignant lesions). A separate non-Hürthle cell thyroid carcinoma was found within the thyroid gland in 22 (50%) of the patients--79% were papillary and 21% were follicular carcinomas. In half of these, there was a history of childhood irradiation. During a mean follow-up period of 8.4 years (range, 22 days to 35 years), three additional Hürthle cell tumors were recognized as malignant after metastases were discovered--two were originally classified as intermediate lesions and one was in the benign-appearing group. Thus, seven of 54 of our patients (13%) had Hürthle cell carcinomas. One of the seven patients died of widespread metastases after 35 years, and the other six are currently free of disease. We believe that therapy of these lesions should be individualized. Total thyroid ablation (surgical procedure followed by radioiodine therapy) is appropriate for frankly malignant Hürthle cell cancers, for all Hürthle cell tumors occurring in patients who received low-dose childhood irradiation, for associated papillary or follicular carcinomas, and in those patients who exhibit partial capsular or subcapsular vascular invasion. On the other hand, single, well-encapsulated, benign-appearing Hürthle cell tumors may be treated by lobectomy and careful follow-up, since the chance that they will later exhibit malignant behavior is low (2.5% in our series and 1.5% among patients described in the recent literature).

The validity of various scintigraphic and endocrinological methods was assessed in order to classify confirmed autonomous adenomas (AA) and to differentiate AA from euthyroid and hyperthyroid goitres. Methods used included: Measurement of 131J-uptake 6, 24 and 48 hours after 131J-application, before and after suppression by T3, thyroid scanning, count-rate measurement in the AA and the paranodular tissue, and determination of T4. All patients received 131J-therapy (40,000 rads). In 8 out of 30 scintigraphically compensated autonomous adenomas (CAA) there was no TSH response to TRH. 12 cases showed a subnormal TSH response and 10 patients had a normal TSH response. T4-values were increased in only 2 cases. Out of 73 decompensated autonomous adenomas (DAA), 66 cases (90%) had no TSH response and 7 cases showed only a subnormal response. In 48% of DAA T4-levels turned out to be in the hyperthyroid range. Both, scintigraphic and endocrinological tests were found to agree in most cases of DAA. However, the diagnosis of CAA could only be proved on the basis of a scintigraphically confirmed suppressive effect of T3 on the paranodular tissue. Thus, in the individual case, separation of CAA from DAA turned out to be impossible using the results of TRH-test alone.

The treatment of Hürthle-cell tumors of the thyroid is controversial because of their rarity and the inconsistent histopathologic criteria for their diagnosis. In order to obtain more objective criteria for the management of Hürthle-cell tumors, the nuclear DNA content of cells from 20 cases was measured with the MicroTICAS system and the correlation between the DNA distribution patterns and the clinical and histopathologic findings was evaluated. Three main DNA patterns were found: euploid, polyploid and aneuploid. The euploid or polyploid Hürthle-cell tumors came from patients who did not develop distant metastases or recurrence whereas the aneuploid variants came from patients who died of their disease and/or developed distant metastases and recurrence. Various correlation analyses were performed between DNA ploidy and age, sex, size of tumor, growth pattern, pleomorphism, invasion and metastases. Our data suggests that an aneuploid DNA pattern or one with a large percentage of aneuploid nuclei with DNA content exceeding 5N may predict eventual metastases or recurrence from Hürthle-cell tumor.