|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Hürthle Cell Carcinoma: A Critical Histopathologic AppraisalFrom the Departments of Surgery and Pathology, and Laboratory of Epithelial Cancer Biology, Memorial Sloan-Kettering Cancer Center, New York, NY. Address reprint requests to Bhuvanesh Singh, MD, Head and Neck Service, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021; email: singhb{at}mskcc.org
PURPOSE: Controversy exists over the ability of morphology to predict the biologic behavior of Hürthle cell carcinoma. The aim of this study was to conduct a critical histopathologic review of Hürthle cell carcinoma and to correlate morphologic parameters with clinical outcome. PATIENTS AND METHODS: Patients with histologically confirmed Hürthle cell carcinoma treated between 1940 and 2000 form the basis of this study. Adenomas were excluded. Tumors of unknown malignant behavior ([UMB] n = 17) had solid growth pattern, incomplete capsular invasion (Ci), or both but no vascular invasion (Vi). Minimally invasive carcinomas ([MIC] n = 23) had one focus of intra- or extracapsular Vi, one focus of complete Ci, or both. Widely invasive carcinomas ([WIC] n = 33) demonstrated more than one focus of Vi, more than one focus of Ci, or both. The primary end points were relapse-free survival (RFS) and disease-specific survival (DSS). Rates of recurrence/death were estimated by Kaplan-Meier method. The univariate influence of prognostic factors on end points was analyzed by log-rank test, and multivariate analysis was performed by Cox regression. RESULTS: Median follow-up was 8 years. No patients with UMB or MIC relapsed or died of disease. Of WIC, 73% relapsed and 55% died of disease. Age, size, and extent of resection did not influence outcome. Adverse predictors of RFS and DSS among WIC were extrathyroidal extension, nodal metastasis, positive margin, and solid growth pattern (P < .05). Both Ci and Vi were associated with worse DSS (P < .05). On multivariate analysis, extrathyroidal extension and nodal metastases were independent predictors of outcome (P < .05). CONCLUSION: Patients with Hürthle cell carcinoma have a prognosis that is predicted by well-defined histomorphologic characteristics. Unlike differentiated thyroid cancer, nodal metastases predict a worse outcome in widely invasive Hürthle cell carcinoma, as does extrathyroidal extension.
HURTHLE CELL or oncocytic tumors are rare thyroid neoplasms of follicular cell origin. Askanazy is credited with the first description of the oxyphil or oncocytic cell, characterized by its polygonal shape, acidophilic granular cytoplasm, hyperchromatic nuclei, and abundant mitochondria.1 Hürthle cell carcinoma represents fewer than 5% of all differentiated carcinomas of the thyroid gland. Fewer than 400 cases of Hürthle cell carcinoma have been reported in the English literature since 1935.2 Although previously considered as variants of follicular carcinomas and classified as such by the World Health Organization, Hürthle cell carcinomas have different oncogenic expression than follicular carcinomas and are now recognized as a distinct clinicopathologic entity.3 Until pathologic criteria became better defined and more uniformly applied, distinguishing benign from malignant Hürthle cell tumors remained difficult. The diagnosis of malignancy is reliably based on histologic evidence of capsular invasion (Ci) or vascular invasion (Vi), extrathyroidal local tissue invasion, and nodal or distant metastasis. Tumors with minimal Ci, although considered carcinomas, have been found to have relatively benign biologic behavior.4 Some authors have considered this group to be intermediate in the adenoma to carcinoma sequence on the basis of incomplete Ci or subcapsular/intracapsular Vi.5 It appears that the natural history of Hürthle cell tumors spans a continuum that includes benign oncocytic adenomas; Hürthle cell tumors of unknown malignant behavior (UMB); minimally invasive, nonthreatening malignancy; and aggressive, carcinomas demonstrating widespread invasion. Considerable controversy exists over the ability of morphology to predict biologic behavior for Hürthle cell tumors. Thus far, progress in clearly defining histopathologic prognostic factors for Hürthle cell carcinoma has been limited by the rarity of the disease. Moreover, the biologic behavior and treatment approach for this malignancy remains controversial. The purpose of this study was to conduct a critical histopathologic review of a large patient cohort with Hürthle cell tumors and to determine to what extent morphologic parameters correlate with clinical outcome.
Patient Search Patients with Hürthle cell tumors were identified from a search of the Departments of Surgery, Head and Neck and Pathology databases, and the hospital tumor registry. This is a retrospective review of patients diagnosed with Hürthle cell tumors who were consecutively treated and followed at Memorial Sloan-Kettering Carcinoma Center from 1940 to 2000. Pathology reports were reviewed for all patients to confirm the diagnosis of Hürthle cell tumor or Hürthle cell carcinoma.
Pathologic Review and Inclusion Criteria
Histopathologic Definitions
As stated by Rosai, a tumor ". . . must penetrate the entire thickness of the capsule to be regarded as having unequivocal Ci. Irregularities of contour along the inner border or clusters of cells embedded within the capsule are not sufficient."6 The latter is interpreted by some pathologists as incomplete Ci. Vi is evident when a vessel located within or immediately external to the capsule contains oncocytes within it, and these cells are found to adhere to the vascular wall. The tumor cells "should project into the vessel lumen in a thrombus-like fashion (Fig 1C), in such a way that they partly or completely obliterate it."6 Tumors with solid/trabecular or follicular growth pattern with a single focus of Vi, a single focus of complete Ci, or both (Fig 1D) were classified as minimally invasive Hürthle cell carcinomas (n = 23, Table 1). Hürthle cell lesions with either solid/trabecular or follicular pattern of growth, more than one focus of intracapsular or extracapsular Vi, or more than one focus of complete Ci were classified as widely invasive Hürthle cell carcinomas (n = 33, Table 1). The majority (88%) of widely invasive Hürthle cell carcinomas in this study demonstrated both extracapsular angioinvasion and complete capsular penetration. Major Vi was defined by extracapsular Vi by tumor. Major Ci was defined by complete capsular penetration by tumor. Tumors that had > 75% follicular cell features with Hürthle cell characteristics and that had separate microscopic foci of papillary carcinoma were included in the study. Thus, widely invasive Hürthle cell carcinomas were defined as those tumors with more than one focus of intracapsular or extracapsular angioinvasion, more than one focus of complete Ci, or both (Fig 1E). These definitions of Hürthle cell UMB, minimally invasive carcinoma (MIC), and widely invasive carcinoma (WIC) were established before slide review.
Histopathologically Confirmed Hürthle Cell UMB and Carcinoma
Clinicopathologic Categories
Study Cohort
Statistics Descriptive statistics were used to summarize study data. Associations between categoric variables were evaluated using the Fisher’s exact or  2 test where appropriate. Kruskal-Wallis analysis of variance was used to compare numerical data between the three groups. Primary study end points included relapse-free survival (RFS) and disease-specific survival (DSS). The following histopathologic factors were correlated with study end points: tumor histology, size, growth pattern, extent and type of invasion, number of foci of Ci or Vi, focality, synchronous foci of microscopic papillary thyroid carcinoma, extrathyroidal extension, regional nodal involvement, presence of distant metastases, and microscopic resection margin.
Outcome was classified according to sites of first disease recurrence. Time to recurrence, RFS, and DSS were all calculated from the date of primary surgery. Deaths resulting from disease were treated as an end point for DSS. Those who died of other causes were considered to have been alive without evidence of disease and were censored. RFS was calculated from the time of surgery to any local, regional nodal, or distant disease recurrence. The rate of recurrence or death was estimated using the Kaplan-Meier product-limit methods. Univariate influence of prognostic factors on study end points was analyzed using the log-rank test. Multivariate analysis, based on Cox’s regression model, was used to associate covariates to time-dependent end points. Only those factors identified to be potentially significant in the univariate analysis were evaluated in the multivariate analysis to assess the independent prognostic effect of these variables. Statistical analysis was performed using JMP software (JMP, Cary, NC). P
Tumor and Treatment Characteristics This study consisted of 73 patients with UMB/MIC or WIC Hürthle cell carcinomas. Tumor and treatment characteristics of Hürthle cell UMB and carcinomas are listed in Table 2. The median size of widely invasive Hürthle cell carcinomas was significantly larger than that of UMBs or MICs (P < .001). Approximately 80% of UMBs and MICs were  4 cm (T1 or T2). Primary WICs exceeded 4 cm in size (T3 or T4) in 76% of cases; 19 (76%) of these demonstrated extrathyroidal extension. Twelve (16%) patients had synchronous microscopic foci of papillary thyroid carcinoma, six each with UMB/MIC and WIC Hürthle cell carcinoma. Only WICs manifested primary regional nodal metastases (seven of 33, 21%, P < .001). The majority of patients with UMB (71%) and MIC (78%) were treated with thyroid lobectomy. Of patients with widely invasive Hürthle cell carcinomas, 45%, 18%, and 36% were treated with lobectomy, subtotal (lobectomy plus contralateral partial lobectomy) thyroidectomy, and total thyroid resection, respectively.
Histopathologic Prognostic Factors
Survival Analysis Median follow-up for the entire study group was 8 years. Median RFS and DSS were 63 months (range, 5 to 381 months) and 91 months (range, 6 to 466 months), respectively. At the time of last follow-up, 45 patients (62%) were alive and relapse free, four (5%) were alive with disease, and 18 (25%) had died of disease. Six (8%) patients have died of other causes without evidence of Hürthle cell carcinoma. The 10-year RFSs and DSSs for the entire cohort were 61% and 73%, respectively. No patient with UMB or minimally invasive Hürthle cell carcinoma developed recurrence or metastases nor died of disease ( Figs 2 and 3). For patients with WICs, 73% relapsed, 64% developed distant metastases, and 55% died of disease at the time of last follow-up. There was no significant difference (P = .28) in local recurrence among patients with WIC treated with less than total (10 of 21, 48%) or total thyroidectomy (three of 12, 25%). The rate of distant metastases was not significantly different (P = .27) among patients treated with less than total (15 of 21, 71%) or total thyroidectomy (six of 12, 50%).
Univariate comparisons were made between histopathologic variables for RFS and DSS. Because no patient with UMB or MIC relapsed or died of disease, the univariate analysis was limited to patients with WIC. Statistically significant factors predicting disease recurrence among patients with WICs included extrathyroidal disease extension (P = .002), solid/trabecular pattern of growth (P = .01), regional nodal metastases (P = .02), and positive microscopic resection margin (P = .02). Neither advanced age (> 45 years old v 45 years old, P = .60) nor large primary tumor size (> 4 cm v 4 cm, P = .31) was predictive of disease recurrence. The 10-year RFS for these patients was 19% (Fig 2). Median RFS was 28 months (range, 7 to 381 months). Median DSS for patients with WICs was 86 months (range, 6 to 466 months). Univariate analysis identified several pathologic factors associated with significantly worse DSS. These included extrathyroidal extension of disease (P = .01), solid growth pattern (P = .01), nodal metastases (P = .02), both major Ci and Vi (P = .02), and positive microscopic resection margin (P = .03). Neither advanced age (> 45 years old) nor large primary tumor size (> 4 cm) was predictive of tumor-related mortality. The 10-year DSS for widely invasive Hürthle cell carcinomas was 48% (Fig 3). On multivariate analysis, extrathyroidal extension of disease (P = .04; relative risk [RR], 2.0; confidence interval [CI], 1.1 to 4.6) and regional nodal metastases (P = .05; RR, 0.6; CI, 0.3 to 1.0) remained independent predictors of disease recurrence following primary resection. The 10-year RFS for patients with and without extrathyroidal disease extension was 75% and 30% ( Fig 4). The 10-year RFS for lymph node–negative and lymph node–positive carcinomas was 31% and 0%. When both extent of disease and nodal status were combined in the univariate analysis, a significant difference in RFS was identified between those patients with both favorable and those with both adverse factors; 10-year DFS was 67% versus 14%, P = .002. Of the five factors associated with tumor-related mortality by univariate analysis, only one emerged as an independent predictor of survival: absence of primary regional nodal disease (P = .02; RR, 0.4; CI, 0.2 to 0.8). The 10-year DSS for patients with and without nodal disease at primary presentation was 18% versus 88% ( Fig 5).
The natural history and treatment of Hürthle cell tumors remains incompletely defined. Hürthle cell carcinomas have been long considered variants of follicular carcinoma and have been classified as a subgroup of follicular cancers by the World Health Organization since 1973. The finding of differential oncogenetic expression between these two types of carcinoma suggests that they are distinct disease entities.3 Distinguishing benign from malignant Hürthle cell tumors remains a topic of debate. As architectural distortion and cellular atypia are found in benign Hürthle cell neoplasms, these diagnostic criteria of malignancy may not be applied to this disease. In the absence of extrathyroidal extension or nodal or distant metastases, the criteria for the diagnosis of malignancy remain Ci or Vi. In most cases, Hürthle cell adenomas can be distinguished from carcinomas by permanent section microscopy on the basis of Ci or Vi.6,7 However, controversy exists over the pathologic diagnostic criteria that differentiate those tumors characterized by a benign course and those that behave aggressively. In the present study, we have conducted a critical histopathologic review on a relatively large, well-characterized patient cohort treated and followed at a single institution with the aim of defining histologic parameters that could be used to predict clinical outcome. Our experience with Hürthle cell tumors included 161 patients for which histologic slides were available for review in 102 patients. Critical review of histopathology resulted in diagnostic revisions in 28% of cases. The majority of revisions in diagnosis consisted of Hürthle cell tumor to adenoma and Hürthle cell carcinoma to tall-cell variant papillary carcinoma. The tall-cell papillary carcinoma cells resemble oncocytes in that they have abundant pink cytoplasm on hematoxylin-and-eosin–prepared slides; however, they are differentiated from oncocytes by having the classic nuclear features of papillary thyroid cancer (ie, ground glass [clear] nuclei, nuclear grooves, and pseudoinclusions). Most diagnostic modifications involved cases initially treated before 1990. This review underscores the fact that our understanding of these oncocytic neoplasms has certainly evolved over this period of time. Accordingly, the diagnosis of Hürthle cell carcinoma can be challenging, and diligent scrutiny of multiple permanent sections is required to define the nature and extent of Ci and Vi. The absence of complete capsular penetration and angioinvasion reliably predicted benignity in our patients. Our data support the finding that the presence of a solitary focus of complete Ci or Vi poses minimal risk of recurrence or carcinoma-related death. A Hürthle cell carcinoma characterized by more than one focus of Vi and/or complete Ci, particularly when both are present, should be regarded as potentially lethal. Nearly 90% of WICs in this study demonstrated both major Ci and Vi, of which 80% of the patients experienced a recurrence and 60% of the patients died of disease. Twenty patients (61%) with WICs had more than four foci of major Vi, and 61% demonstrated more than four foci of complete Ci. We identified two patients with WICs that recurred with more than four foci of Ci and no or only two foci of Vi associated with the primary tumor. Although uncommon, recurrent disease may develop in the setting of few identifiable foci of major Ci or Vi.8 The distinction between MICs and WICs appears clear; within the realm of histologically malignant Hürthle cell tumors, microscopic morphology appears to correlate closely with the biology of individual lesions. No patient with UMB or minimally invasive Hürthle cell carcinoma as defined in this study either relapsed or died of disease. However, 73% of patients with WICs have experienced a recurrence, 64% have developed distant metastases, and 55% have died of disease. Our findings confirm observations from the Lahey Clinic that minimally invasive Hürthle cell carcinomas do not demonstrate malignant biology.4 These data suggest that the clinical behavior of Hürthle cell tumors may be predicted by well-defined histopathologic criteria. These observations have important therapeutic implications. Thompson et al9 and Gundry et al10 from the University of Michigan in 1974 and 1983 regarded all Hürthle cell neoplasms, irrespective of size, as malignant and advocated an aggressive surgical approach to their treatment. The recommendation for total thyroidectomy was based on tumor-related mortality in 12% (three of 36) of benign and 31% of malignant Hürthle cell tumors over a study period of 31 years. Gosain and Clark, however, identified only four cases (5%) of carcinoma among 75 patients with Hürthle cell tumors (71 adenomas) of which one recurred after primary treatment.11 A later comprehensive review of the literature from 1954 to 1988 identified six of 642 (0.9%) reported cases of Hürthle cell carcinomas inappropriately diagnosed as adenomas.7 As all UMBs and MICs in our study had unifocal disease and exhibited a benign clinical course over a follow-up period of 8 years, thyroid lobectomy and isthmusectomy alone may be considered sufficient treatment. In the case of multiple bilobar benign Hürthle cell tumors, UMBs, or MICs, total thyroidectomy is indicated. On the basis of our experience, little justification can be made for routine total thyroidectomy for all Hürthle cell tumors, and an individualized treatment approach is appropriate based on the risk of recurrence. We cannot conclusively state that total thyroidectomy clearly improves outcome when compared with less extensive operations given the modest number of patients studied, the retrospective nature of the study, and the fact that patients with locally advanced disease were treated more aggressively than those with favorable characteristics. Although radioiodine has demonstrated efficacy in the setting of recurrent, well-differentiated thyroid carcinoma, there has been no proven role of radioactive iodine therapy for Hürthle cell carcinoma.4 This is largely based on the finding that metastatic Hürthle cell carcinomas do not concentrate radioiodine.12-14 Moreover, adjuvant chemotherapy has limited efficacy in the treatment of this disease.8,15 Given the highly malignant nature of widely invasive Hürthle cell carcinomas and the paucity of effective adjuvant therapy, total thyroidectomy is recommended in an effort to optimize local control and cure of this aggressive variant of Hürthle cell carcinoma. In the univariate analysis of widely invasive Hürthle cell carcinomas, extrathyroidal extension, regional nodal metastases, positive microscopic resection margin, and solid/trabecular growth pattern emerged as adverse prognostic factors for RFS and DSS. In addition, the category of complete Ci and Vi was a significant adverse predictor of tumor-related mortality. Although patient age and primary tumor size have been shown to be significant predictors of survival in all risk-stratifying systems for papillary and follicular carcinoma, we have found neither to be predictive of outcome in this population of Hürthle cell carcinoma.4,16-20 Multivariate analysis identified two independent predictors of disease recurrence: extrathyroidal disease extension and regional nodal metastases at time of primary surgery. Nodal metastases emerged as the only independent predictor of DSS. All patients with primary nodal metastases had evidence of extrathyroidal disease extension. The prognostic significance of regional nodal metastases in the setting of differentiated thyroid carcinoma remains controversial.16-20 The finding that nodal metastasis is an independent adverse predictor of outcome for widely invasive Hürthle cell carcinoma must be interpreted cautiously within the confines of the relatively small patient subset analyzed. From a clinical standpoint, differentiating between MICs and WICs on the basis of gross morphology can be challenging, and we have found frozen section analysis to be inaccurate in assessing the degree of Ci or Vi. In our experience, the diagnosis of adenoma by frozen section analysis is changed by permanent section evaluation to carcinoma (in most cases minimally invasive) in 20% to 30% of patients.16 In the absence of gross evidence of extrathyroidal extension of disease, regional nodal metastases, or frozen section microscopic evidence of WIC at primary operation, we advocate thyroid lobectomy and isthmusectomy. If widely invasive Hürthle cell carcinoma is identified on histologic review of multiple permanent tumor sections, completion thyroidectomy is recommended. The latter scenario is distinctly uncommon in our experience, because frozen section analysis has been reliable in identifying WICs. In this study, widely invasive Hürthle cell carcinomas had extrathyroidal extension of disease in nearly 60% of patients at the time of primary operation; 37% of these patients had cervical node metastases. Our practice has been to perform modified neck dissections (sparing the internal jugular vein, sternocleidomastoid muscle, and spinal accessory nerve) for palpable cervical node metastases. When local control was analyzed in patients with WIC on the basis of the extent of resection, no statistically significant difference could be demonstrated. Thus, in this small patient subset, we were unable to statistically show an improved outcome with more extensive operative approaches. Aggressive locoregional surgical treatment must be balanced with the risk of operative morbidity at a given institution. Although most patients relapse within the first 2 years following surgery, relapse-free intervals on the order of decades have been observed with this disease, underscoring the need for careful, lifelong clinical surveillance of these patients. This study represents a critical histopathologic appraisal of a relatively large, well-characterized cohort with Hürthle cell tumors. Although our findings suggest that the clinical behavior of Hürthle cell tumors may be predicted on the basis of well-defined histopathologic criteria, a number of shortcomings must be addressed. The reliability of the findings may be limited by the retrospective methodology used in this study. As Hürthle cell carcinoma is a rare disease, prospective studies to evaluate natural history and efficacy of treatment are not feasible. Thus, we are reliant on retrospective analyses of prognostic factors for outcome in an effort to define the natural history of the disease and to make appropriate treatment and follow-up recommendations. In terms of quantifying Ci and Vi the retrospective review of permanent histopathology, sections cannot account for areas of invasion that were not sampled at the time of initial specimen processing. We have identified 28% of cases for which the diagnosis was revised; thus we cannot account for cases of Hürthle cell carcinoma misdiagnosed as the morphologically similar tall-cell variant of papillary carcinoma. The interpretation of extrathyroidal extension of disease and regional nodal metastasis as independent outcome predictors must take into consideration that the multivariate analysis was conducted on a small subset of patients with WICs, as they were the only patients that recurred and suffered tumor-related deaths. Morphology appears to be a powerful predictor of clinical outcome for Hürthle cell tumors. No patient with UMB or MIC relapsed or died of disease. Widely invasive Hürthle cell carcinomas recurred in 73% of cases, and 55% of patients died of disease. Extranodal extension and regional nodal metastases appear to be independent predictors of outcome for WICs. An aggressive, locoregional treatment approach to WICs is recommended. Given the considerable variation in time to recurrence, lifelong follow-up is indicated in patients with widely invasive Hürthle cell carcinoma.
We thank Denis Hy Leung, PhD, for his assistance with the statistical analysis.
A.S. and R.A.G. both contributed equally to this work.
1. Askanazy M: Pathologisch-anatomische Beiträge zur Kenntniss des morbus basedowii, insbesondere über die dabei auftretende. Muskelerkrankkung Dtsch Arch Klin Med 61: 118-186, 1898 2. Grossman RF, Clark OH: Hürthle cell carcinoma. Carcinoma Control 4: 13-17, 1997 3. Masood S, Auguste LJ, Westerbank A, et al: Differential oncogenic expression in thyroid follicular and Hürthle cell carcinomas. Am J Surg 1666: 366-368, 1993 4. Sanders LE, Silverman M: Follicular and Hürthle cell carcinoma: Predicting outcome and directing therapy. Surgery 124: 967-974, 1998[Medline] 5. Arganini M, Behar R, Wu TC, et al: Hürthle cell tumors: A twenty-five-year experience. Surgery 100: 1108-1115, 1986[Medline] 6. Rosai J, Caracangui M, DeLellis R: Tumors of the thyroid gland, in Rosai J, Sobin L (eds): Atlas of Tumor Pathology (series 3) Fascicle. Washington DC, Armed Forces Institute of Pathology, 1992 7. Grant CS, Barr D, Goellner JR, et al: Benign Hürthle cell tumors of the thyroid: A diagnosis to be trusted? World J Surg 12: 488-495, 1988[Medline] 8. Goldstein NS, Czako P, Neill JS: Metastatic minimally invasive (encapsulated) follicular and hurthle cell thyroid carcinoma: A study of 34 patients. Mod Pathol 13: 123-130, 2000[Medline] 9. Thompson NW, Dunn EL, Batsakis JG, et al: Hürthle cell lesions of the thyroid gland. Surg Gynecol Obstet 139: 555-560, 1974[Medline] 10. Gundry SR, Burney RE, Thompson NW, et al: Total thyroidectomy for Hürthle cell neoplasm of the thyroid. Arch Surg 118: 529-532, 1983[Abstract] 11. Gosain AK, Clark OH: Hürthle cell neoplasms: Malignant potential. Arch Surg 119: 515-519, 1984[Abstract] 12. Shaha AR, Shah JP, Loree TR: Patterns of nodal and distant metastasis based on histologic varieties in differentiated carcinoma of the thyroid. Am J Surg 172: 692-694, 1996[Medline] 13. Schoem SR: Oxyphil cell neoplasms of the thyroid gland. South Med J 86: 919-923, 1993[Medline] 14. Nishiyama RH: Overview of surgical pathology of the thyroid gland. World J Surg 24: 898-906, 2000[Medline] 15. Har-El G, Hadar T, Segal K, et al: Hürthle cell carcinoma of the thyroid gland. Carcinoma 57: 1613-1617, 1986 16. Shaha AR, Loree TR, Shah JP: Prognostic factors and risk group analysis in follicular carcinoma of the thyroid. Surgery 118: 1131-1136, 1995[Medline] 17. Hay ID, Grant CS, Taylor WF, et al: Ipsilateral lobectomy versus bilateral lobar resection in papillary thyroid carcinoma: A retrospective analysis of surgical outcome using a novel prognostic scoring system. Surgery 102: 1088-1095, 1987[Medline] 18. Cady B, Ross R: An expanded view of risk group defini-tion in differentiated thyroid carcinoma. Surgery 104: 947-953, 1988[Medline] 19. Shah JP, Loree TR, Dharker D, et al: Prognostic factors in differentiated carcinomas of the thyroid gland. Am J Surg 164: 658-661, 1992[Medline] 20. Shaha AR, Loree TR, Shah JP: Intermediate risk group for differentiated carcinoma of thyroid. Surgery 116: 1036-1041, 1994[Medline] Submitted December 11, 2000; accepted February 14, 2001. This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
