Follicular neoplasms of the thyroid: importance of clinical and cytological correlation

Transcription

1 Cir Cir 2010;78: Follicular neoplasms of the thyroid: importance of clinical and cytological correlation Martín Granados-García,* Ana Olivia Cortés-Flores,* Imelda del Carmen González-Ramírez,** Ana María Cano-Valdez,*** Lorena Flores-Hernández**** and José Luis Aguilar-Ponce* Abstract Background: Thyroid cancer presents as nodules. Thyroid nodules are frequent, but only 5-30% are malignant. Fine needle aspiration biopsy (FNAB) is useful for initial evaluation; nevertheless, malignancy is uncertain when follicular neoplasm is reported. Some factors can be associated with malignancy. Therefore, we analyzed our follicular neoplasms in order to identify those factors associated with a higher risk of malignancy. Methods: We analyzed the clinical files of consecutive patients with cytological diagnoses of follicular neoplasm. Results: From 1,005 cases of thyroid nodules, 121 were follicular neoplasms according to cytology. Of these, 75 were surgically treated. Definitive report showed 45 benign (60%) and 30 malignant (40%) cases. Benign cases included 29 goiters, 11 follicular adenomas, and 5 cases of thyroiditis. Malignant cases were comprised of 12 papillary carcinomas, 4 follicular carcinomas, 3 papillary carcinomas follicular variant, 1 lymphoma, 1 teratoma, 5 medullary carcinomas, 2 insular carcinomas, 1 anaplastic carcinoma and 1 metastatic breast carcinoma. Tumor size of benign lesions was 3.43 ± 2.04 cm and 4.67 ± 2.78 (p = 0.049) for malignant lesions. Age was ± years for benign lesions and ± for malignant lesions (p = 0.66). Fifty percent of males showed malignancy vs. 37.7% of females (p <0.005). Conclusions: Our results suggest that size and gender, but not age, are associated with cytological pattern. Ultrasonographic characteristics may be useful discriminating patients with a higher risk of malignancy. FNAB is a useful tool for initial evaluation of thyroid nodules, but clinical evaluation can enhance predictive value. Key words: follicular neoplasm, fine needle aspiration biopsy, follicular carcinoma, thyroid nodule. Introduction Thyroid cancer accounts for <1% of the malignant neoplasms and its incidence rate is 0.004% per year. 1,2 In Mexico, in 2002 there were 1,937 thyroid malignancies reported among 108,064 cancer cases. Of these, 1,580 occurred in * Departamento de Tumores de Cabeza y Cuello, ** Departamento de Atención a la Salud, Universidad Autónoma Metropolitana, México, D.F., Mexico, *** Departamento de Patología, **** Servicio de Citología, Instituto Nacional de Cancerología, Secretaría de Salud, México, D.F., Mexico Correspondence: Martín Granados-García Departamento de Tumores de Cabeza y Cuello Instituto Nacional de Cancerología San Fernando 22, Col. Sección XVI, Del. Tlalpan México, D.F., Mexico Tel: (55) martingranadosmx@yahoo.com.mx Received for publication: Accepted for publication: females and 357 in males, leading to a female/male ratio of 4.4:1. In this same year, 460 deaths were attributed to thyroid cancer, for a mortality rate of 0.4/100,000 inhabitants. 3 Despite being a rare neoplasm, the presence of thyroid nodules is a common clinical problem: its frequency is 4-10% by palpation and up to 67% when using ultrasound in the population > 40 years of age. 4 Most of the nodules of this nature are benign, but 5-30% are malignant 5 and require surgical treatment. Fine needle aspiration biopsy (FNAB) is the most useful test for the initial evaluation of thyroid nodules, but a proportion of cases are recorded as indeterminate nodules and the majority of them are follicular neoplasms, where by definition the malignancy is uncertain. These cases represent 10-25% of the reports and require a precise assessment in order to avoid unnecessary surgical intervention. 6 In selected cases of follicular neoplasm, observation is recommended because thyroid surgery is not innocuous. The most feared complication of a total thyroidectomy is permanent hypoparathyroidism, occurring in 2-11%, 7,8 which is difficult to justify in benign lesions although the observation is dangerous because of the risk of underlying malignancy. Routine total thyroidectomy, without definitive diagnosis, is an overtreatment in a high proportion Volume 78, No. 6, November-December

2 Granados-García M et al. of patients because only 15-20% of these patients require surgery for malignancy. To distinguish benign from malignant nodules, FNAB has been proposed to be used together with ultrasound and nuclear medicine imaging, although the usefulness of the association has not been determined. FNAB FNAB is recognized for its cost-effectiveness in diagnosing thyroid diseases. 1,2 Due to the use of FNAB, the number of patients referred for surgery has declined by 25% 4 and the proportion of thyroidectomy with underlying malignancy has increased by two to three times. FNAB is commonly reported as benign, malignant, inadequate or indeterminate, and the latter case represents a management problem, 9 particularly the so-called follicular neoplasms, which may be hyperplastic nodules, nodular goiters and follicular adenomas, but also follicular carcinomas (FC), papillary carcinomas (PC) and papillary carcinomas-follicular variant (FVPCA). 1,10,11 The problem is that it is impossible to distinguish between adenoma and FC based on cytological characteristics. Vascular and capsular invasion need to be demonstrated, which is not possible to document with cytology. 9,12 However, lesions with a microfollicular pattern, solid and trabecular lesions may be FCs (23%), whereas virtually no macro-pattern lesion results in FC. A lesion is considered macrofollicular when >70% of the area is occupied by macrofollicles and microfollicular when >70% is occupied by microfollicles, solid mantles or trabeculae of follicular cells (Figures 1 and 2). An adequate specimen contains five to six fragments of follicular epithelium of adequate size in at least two slides. Other authors require eight to ten well-established and stained cell groups on two slides, and each group should contain <10 cells. 13 A sample that does not meet these conditions or is inadequate is an indication for a repeat of the FNAB. Thyroid scan has been used to distinguish benign from malignant lesions, but its low predictive value limits its usefulness. The addition of an ultrasound has also been used to enhance the diagnostic power to identify suspicious areas to biopsy. Suggestive findings include microcalcifications, solid nodules, hypoechogenic lesions, irregular margins, nonencapsulated lesions, and spherical and hypervascularitic intranodular lesions, but its predictive value is far from perfect. 14 FC represents 3-10% of cases of thyroid malignancy. These are usually solitary or associated with multinodular goiter. They sometimes affect individuals with a history of radiation. Metastasis to lungs and bones is common, and >33% have subclinical metastases at the time of diagnosis. 4,15 In contrast, follicular cells forming monolayers and follicular groups on a colloidal background characterize the FVPCA cellular specimens. In addition, the cells may be oncocytic, with eosinophilic cytoplasm and elongated nucleus, focally superimposed on clear chromatin, pseudoinclusions and nuclear grooves. 11,16,17 To define the diagnosis in patients with indeterminate aspirates and to avoid total thyroidectomy or re-operations that have morbid outcomes, the study of intraoperative frozen sections has been used, but its usefulness is not clear. Figure 1. Follicular cells with a microfollicular pattern (Diff-Quick stain, original magnification x40). Figure 2. Discohesive cells that form microfollicles, some with small nucleoli (Papanicolaou stain, original magnification x40). 474 Cirugía y Cirujanos

3 Follicular neoplasm of thyroid gland Basolo et al. 1 studied 1,472 cases of slides and frozen sections and found a diagnostic accuracy of 88.8% and 95.7% when compared with the definite histological diagnosis. Kingston et al. 18 analyzed 395 patients with follicular neoplasms, of whom 198 had frozen sections performed during their surgery. Frozen sections were accurate in 79% of the cases differentiating adenomas from carcinomas, yielding a sensitivity of 52% and specificity of 100%. The positive predictive value of frozen sections showing carcinoma was 100%, but the negative predictive value was 73%. An incorrect diagnosis was made in 21% of the patients in whom the final diagnosis was carcinoma. Consequently, the evidence points to the existence of an inadequate cost-effective relation for frozen sections that systematically halts their use. Due to the lack of frozen sections to resolve the problem, we have attempted to introduce clinical factors to enhance the predictive value. Some authors have found clinical criteria that indicate malignancy when the cytology is reported as indeterminate or as follicular neoplasm. These include male gender, tumor size >4 cm, and age between 40 and 45 years. Baloch et al. 19 studied 184 cases of follicular neoplasm in 1024 cases of FNAB and report a statistically significant difference when tumors were larger or smaller than 3 cm (55 vs. 23%, p = ) when age was >40 years in relation to when it was <40 years (20 vs. 10%, p = ). In addition, male patients were associated with an increased risk (47 vs. 29%, p = ). We believe there must be clinical factors associated with an increased likelihood of malignancy when reporting follicular neoplasms, but these may differ between the populations. To identify the clinical factors associated with underlying malignancy, we analyzed our cases of follicular neoplasm and compared benign and malignant according to the definitive diagnosis Materials and Methods We reviewed records of patients treated at the Departamento de Tumores de Cabeza y Cuello del Instituto Nacional de Cancerología (INCan) with the diagnosis of thyroid nodules. The review took place between January 1, 2000 and January 1, We analyzed those files with cytological diagnosis of follicular neoplasm, with correlation between the cytological and definitive histological diagnosis. The study was not associated with any additional risk; however, we obtained approval from the Research Committee of our Institution. We eliminated from the analysis those patients with incomplete medical records. FNAB FNAB was performed by the cytologist with a 20-cm syringe and a 23-g or 25-g needle. In lesions <1 cm and in those not easily palpable, suspicious areas were examined by ultrasound-assisted biopsy. Four to eight slides were made and half of these were air-dried and stained with Diff- Quick to assess the quality and quantity of the cytological material. If insufficient material was obtained, the puncture was then repeated. The other slides were fixed in 95% ethanol and Papanicolaou stain was used. Cytological diagnoses were made by two dedicated cytopathologists and their consensus. We performed the analysis with the original cytological diagnoses to avoid bias, but all histological material was reviewed by an expert pathologist who was unaware of the previous results. Statistical Analysis Comparisons of the clinical variables between benign and malignant follicular neoplasms after the definitive diagnosis were made using SPSS program v.12. We used χ 2 test for comparisons between dichotomous variables and Student t-test for comparisons of continuous variables. Statistical significance was accepted when p <0.05. Results We analyzed 1,005 thyroid nodules collected during a 5-year period of five (January 1, 2000 to January 1, 2005) in the INCan. All patients underwent FNAB. One hundred and twenty-one cases were reported as follicular neoplasm (12%), but 75 cases with significant suspicion of malignancy were referred for surgery with the surgical component for cytohistological correlation. Fifty cases were initially reported as insufficient material, but biopsies were repeated and were later classified as follicular neoplasm. Our institutional policy, in patients who were not referred for surgery, was a yearly follow-up to avoid leaving a false negative untreated. After an average of >2 years, we detected only one case of a malignancy initially omitted. No patients had other manifestations suggestive of malignancy such as metastatic cervical lymph nodes or manifestations of a locally invasive disease. Usually the nodules were solitary: 14 clinical cases involved both lobes, 21 cases involved the left lobe and 20 cases involved the right lobe. Data were missing in all other cases. In the analyzed group there were 61 females (81.3%) and 14 males (18.7%). Median age was ± years (range: years) and median nodule size was 3.98 ± 2.46 cm (range: cm). Volume 78, No. 6, November-December

4 Granados-García M et al. Cytological diagnosis of follicular neoplasm was made in 75 cases, whereas the specific diagnoses were follicular neoplasm in 60 cases (79%), follicular lesion in 9 cases (12%), possible follicular adenoma in 3 cases (4%), follicular neoplasm with oncocytic changes in 1 case (1.3%), follicular neoplasm with oxyphilic cells in one patient (1.3%), and Hurtle cell neoplasm in one case (1.3%). Definitive histological diagnosis was benign in 45 patients (60%) and malignant in 30 cases (40%). Among those nodules reported as benign, 64% were nodular goiters (29 cases), 24% were follicular adenomas (11 cases) and 11% were Hashimoto s thyroiditis (5 cases). Malignant tumors were PC in 12 cases (40%), FC in 4 cases (13%), FVPCA in 3 cases (10%); there were 11 cases (37%) of various neoplasms: five medullary carcinomas, two insular carcinomas, one anaplastic carcinoma, one metastatic breast cancer, one lymphoma and one teratoma (Table 1). Among other factors analyzed (age, gender, size, laterality, associated diffuse growth, and consistency), the following relevant results were as follows: mean age for benign neoplasms was ± years and ± years for malignant neoplasms (not significant, p = 0.66). The median size for benign neoplasms was 3.43 ± 2.04 cm vs ± 2.78 cm for malignant neoplasms (p = 0.049). In the case of gender, 50% of males had a malignant disease, whereas benign neoplasms predominated in females with only 37.7% reporting malignancy (p <0.005). Discussion Twelve percent of our cytological diagnoses were reported as follicular neoplasms, a low proportion according to Table 1. Benign and malignant lesions reported in cytological study Cases % Benign pathology Multinodular goiters Follicular adenoma Hashimoto s thyroiditis 5 11 Malignant pathology Papillary carcinoma Follicular carcinoma 4 13 Papillary carcinoma-follicular variant 3 10 Others Medullary carcinoma Insular carcinoma Anaplastic carcinoma Metastatic carcinoma Lymphoma Teratoma reports from other authors. 6 In addition, previous reports showed that an indeterminate diagnosis was associated with malignancy in 20-40% of the cases. 11 In this regard, Bakhos 2 and Baloch 19 reported malignancy with indeterminate cytology in 15-20% of the cases. In contrast, our study showed a low proportion of follicular neoplasm and a high risk of malignancy (40%), probably because of the strict diagnostic criteria or because our data come from a reference cancer center where patients with high probability of malignancy are referred. It has been assumed that follicular neoplasm is an unresolved problem because it is impossible to distinguish between follicular adenoma and carcinoma but, interestingly, in our study, as shown by other authors (Table 2), a significant proportion of follicular neoplasms were not adenomas or FC. We believe that, in practice, diagnosis of follicular neoplasm is a waste basket for suspected aspirated material or with unrepresentative samples due to sampling errors. We believe that clinicians should be aware of this phenomenon in order to avoid errors. Our evaluation identified some clinical parameters that may increase the predictive value of malignancy when the patient is diagnosed with follicular neoplasm. These include a tumor size >4 cm and male gender. In contrast with other reports, age was not a predictive factor; therefore, in our population the possibility of a malignant diagnosis should not be ruled out in younger subjects. There is no doubt regarding the usefulness of FNAB. But there is a problem because of its low predictive value, especially when a multinodular goiter, follicular adenoma, FC, or FVPCA are present because they are cytologically indistinguishable. 2,20 But other considerations of cytological characteristics may be useful such as micro- or macrofollicular pattern. Microfollicular pattern is associated with a significant frequency of malignancy, whereas macrofollicular points to a benign disease such as goiter. Similarly, ultrasonographic findings such as microcalcifications, solid nodules, irregular margins, noncapsular lesions and intranodular hypervascularity suggest malignancy. Furthermore, use of calcitonin in the initial evaluation of thyroid nodules may be of value and a cost-effective measure to preoperatively diagnose some medullary carcinomas. 20 Finally, certain features such as male gender and nodular size >4 cm may offer guidance to the clinician for arriving at the most optimal surgical decision. Unfortunately, the predictive values remain low and, therefore, we are unable to avoid some unnecessary surgery. Consequently, evaluation of certain molecular markers for preoperative diagnosis has been suggested. 12,21 The suggested markers include cytokeratin-19, galectin-3, HBME-1, Leu-M1 thyroid peroxidase, lactoferrin, thyroglobulin, di- 476 Cirugía y Cirujanos

5 Follicular neoplasm of thyroid gland Table 2. Analysis of definitive diagnosis after diagnosis of follicular neoplasm using FNAB First author/year Goiter Adenoma FVPCA FC Others Total Lin (1997) * 71.2 NE NE Logani (2000) Bakhos (2000) NE Baloch (2002) Wu (2006) 13 NE NE Basolo (2007) 1 213* 79.2 NE NE Kapur (2007) This series FNAB, fine needle aspiration biopsy; FVPCA, follicular variant papillary carcinoma; FC, follicular carcinoma; NS, not specified. *Only reported as benign. Includes papillary carcinoma without specifying if follicular variant is included. ß Includes benign pathologies distinct from follicular adenoma. Includes all nodules diagnosed using FNAB. peptidyl peptidase IV (DAPIV) and topoisomerase II alpha. These may help distinguish between benign and malignant follicular neoplasms. Another approach is to perform a thyroid scan with 99m- TC-MIBI. Some authors have reported that the absence of uptake ensures the absence of malignancy. On the other hand, 18F-FDG-PET could accomplish this task, but at a high cost and with problems because of its limited availability. In addition, to date, the experience is limited and results must be corroborated. 22,23 Despite its limitations, FNAB remains to be a useful tool in the initial study of patients with thyroid nodules, but we believe that clinical assessment is essential in the comprehensive assessment of patients, especially with regard to surgical decisions. What should we do when a report of follicular neoplasm by frozen section fails to define the diagnosis? We believe that total thyroidectomy is justified if the described parameters (cytological pattern, gender, nodule size, sonographic characteristics, serum calcitonin) have been analyzed and suggest malignancy, provided that total thyroidectomy can be done with minimal morbidity. This avoids second operation associated with significant morbidity. In conclusion, in addition to cytological and ultrasonographic evaluations along with evaluations of serum calcitonin, clinical parameters may be helpful. Male gender is a strong predictor of malignancy. In our population, age showed no statistical difference; therefore, the possibility of malignancy cannot be ruled out in younger patients. The increasing size of the nodule was associated with increased risk (p = 0.049), supporting the notion that patients with nodules >4 cm are associated with an increased risk of malignancy. References 1. Basolo F, Ugolini C, Proietti A, Iacconi P, Berti P, Miccoli P. Role of frozen section associated with intraoperative cytology in comparison to FNA and FS alone in the management of thyroid nodules. Eur J Surg Oncol 2007;33: Bakhos R, Selvaggi S, DeJong S. Fine-needle aspiration of the thyroid: rate and causes of cytohistopathologic discordance. Diagn Cytopathol 2000;23: Centro Nacional de Vigilancia Epidemiológica y Control de Enfermedades. Dirección General de Epidemiología. Compendio de cáncer Registro Histopatológico de las Neoplasias Malignas en México México: SSA; pp. 3, St Louis JD, Leight GS, Tyler DS. Follicular neoplasms: the role of observation, fine needle aspiration biopsy, thyroid suppression, and surgery. Semin Surg Oncol 1999;16: Kimoto T, Suemitsu K, Eda I, Shimizu T, Ohtani M, Nabika T. The efficiency of performing ultrasound-guided fine-needle aspiration biopsy following mass screening for thyroid tumors to avoid unnecessary surgery. Surg Today 1999;29: Rosen JE, Stone MD. Contemporary diagnostic approach to the thyroid nodule. J Surg Oncol 2006;94: Prim MP, Diego JI, Hardisson D, Madrid R, Gavilan J. Factors related to nerve injury and hypocalcemia in thyroid gland surgery. Otolaryngol Head Neck Surg 2001;124: Thomusch O, Machens A, Sekulla C, Ukkat J, Brauckhoff M, Dralle H. The impact of surgical technique on postoperative hypoparathyroidism in bilateral thyroid surgery: a multivariate analysis of 5846 consecutive patients. Surgery 2003;133: Galera-Davidson H. Diagnostic problems in thyroid FNAs. Diagn Cytopathol 1997;17: Berner A, Sigstad E, Pradhan M, Groholt KK, Davidson B. Fineneedle aspiration cytology of the thyroid gland: comparative analysis of experience at three hospitals. Diagn Cytopathol 2006;34: Logani S, Gupta PK, LiVolsi VA, Mandel S, Baloch ZW. Thyroid nodules with FNA cytology suspicious for follicular variant of papillary thyroid carcinoma: follow-up and management. Diagn Cytopathol 2000;23: Volume 78, No. 6, November-December

6 Granados-García M et al. 12. Henry JF, Denizot A, Porcelli A. Thyroperoxidase immunodetection for the diagnosis of malignancy on fine-needle aspiration of the thyroid nodules. World J Surg 1994;18: Wu HH, Jones JN, Osman J. Fine-needle aspiration cytology of the thyroid: ten years experience in a community teaching hospital. Diagn Cytopathol 2006;34: Kovacevic DO, Skurla MS. Sonographic diagnosis of thyroid nodules: correlation with the results of sonographically guided fine-needle aspiration biopsy. J Clin Ultrasound 2007;35: Gardner HA, Ducatman BS, Wang HH. Predictive value of fine-needle aspiration of the thyroid in the classification of follicular lesions. Cancer 1993;71: Lin JD, Huang BY, Weng HF, Jeng LB, Hsueh C. Thyroid ultrasonography with fine-needle aspiration cytology for the diagnosis of thyroid cancer. J Clin Ultrasound 1997;25: Gharib H. Fine-needle aspiration biopsy of thyroid nodules: advantages, limitations and effect. Mayo Clin Proc 1994;69: Kingston GW, Bugis SP, Davis N. Role of frozen section and clinical parameters in distinguishing benign from malignant follicular neoplasm of the thyroid. Am J Surg 1992;164: Baloch ZW, Fleisher S, LiVolsi V, Gupta PK. Diagnosis of follicular neoplasms : a gray zone in thyroid fine-needle aspiration cytology. Diagn Cytopathol 2002;26: Elisei R, Bottici V, Luchetti F, Di Coscio G, Romei C, Grasso L, et al. Impact of routine measurement of serum calcitonin on the diagnosis and outcome of medullary thyroid cancer: experience in patients with nodular thyroid disorders. J Clin Endocrinol Metab 2004;89: Saggiorato E, De Pompa R, Volante M. Characterization of thyroid follicular neoplasms in fine-needle aspiration cytological specimens using a panel of immunohistochemical markers: a proposal for clinical application. Endocr Relat Cancer 2005;12: Hurtado-López LM, Arellano-Montaño S, Torres-Acosta EM. Combined use of fine-needle aspiration biopsy, MIBI scans and frozen section biopsy offers the best diagnostic accuracy in the assessment of the hypofunctioning solitary thyroid nodule. Eur J Nucl Med Mol Imaging 2004;31: De Geus-Oei LF, Pieters GF, Bonekamp JJ. 18F-FDG PET reduces unnecessary hemithyroidectomy for thyroid nodules with inconclusive cytologic results. J Nucl Med 2006;47: Cirugía y Cirujanos