1139 Medullary Thyroid Carcinoma Clinical Characteristics, Treatment, Prognostic Factors, and a Comparison of Staging Systems Electron Kebebew, M.D.1 Philip H. G. Ituarte, M.P.H., Allan E. Siperstein, M.D.1 Quan-Yang Duh, M.D.1,2 Orlo H. Clark, M.D.1 1 Ph.D. 1 University of California, San Francisco School of Medicine, and Department of Surgery, University of California, San Francisco/Mount Zion Medical Center, San Francisco, California. 2 Veterans Affairs Medical Center, Surgical Services, San Francisco, California. Presented in part at the 80th Annual Meeting of The Endocrine Society, June 1998, New Orleans, Louisiana. Supported in part by the Mt. Zion Health Systems and Friends of Endocrine Surgery Fund. Dr Kebebew is a fellow in Surgical Oncology Basic Science, with an NIH T32 training grant, and a resident in surgery at the University of California, San Francisco. Address for reprints: Orlo H. Clark, M.D., Department of Surgery, C342, University of California, San Francisco /Mt. Zion Medical Center, 1600 Divisadero Street, San Francisco, CA 94143-1674. Received June 17, 1999; revision received October 15, 1999; accepted November 19, 1999. © 2000 American Cancer Society BACKGROUND. The clinical courses of patients with medullary thyroid carcinoma (MTC) vary, and a number of prognostic factors have been studied, but the significance of some of these factors remains controversial. METHODS. The study group consisted of 104 patients with MTC or C-cell hyperplasia managed at the hospitals of the University of California, San Francisco, between January 1960 and December 1998. Patients were classified as having sporadic MTC, familial non-multiple endocrine neoplasia (MEN) MTC, MEN 2A, or MEN 2B. The TNM, European Organization for Research and Treatment of Cancer (EORTC), National Thyroid Cancer Treatment Cooperative Study (NTCTCS), and Surveillance, Epidemiology, and End Results (SEER) extent-of-disease stages were determined for each patient. The predictive values of these staging or prognostic scoring systems were compared by calculating the proportion of variance explained (PVE) for each system. RESULTS. Fifty-six percent of the patients had sporadic MTC, 22% had familial MTC, 15% had MEN 2A, and 7% had MEN 2B. The overall average age at diagnosis was 38 years, and patients with sporadic MTC presented at an older age (P ⬍ 0.05). Thirty-two percent of the patients with hereditary MTC were diagnosed by screening (genetic and/or biochemical). These patients had a lower incidence of cervical lymph node metastasis (P ⬍ 0.05) and 94.7% were cured at last follow-up (P ⬍ 0.0001) compared with patients not screened. Patients with sporadic MTC who had systemic symptoms (diarrhea, bone pain, or flushing) had widely metastatic MTC and 33.3% of those patients died within 5 years. Overall, 49.4% of the patients were cured, 12.3% had recurrent MTC, and 38.3% had persistent MTC. The mean follow-up time was 8.6 years (median, 5.0 years) with 10.7% (n⫽11) and 13.5% (n⫽14) cause specific mortality at 5 and 10 years, respectively. Patients with persistent or recurrent MTC who died of MTC lived for an average of 3.6 years (ranging from 1 month to 23.7 years). Patients who had total or subtotal thyroidectomy were less likely to have persistent or recurrent MTC (P ⬍ 0.05), and patients who had total thyroidectomy with cervical lymph node clearance required fewer reoperations for persistent or recurrent MTC (P ⬍ 0.05) than patients who underwent lesser procedures. In univariate analysis, age, gender, clinical presentation, TNM stage, sporadic/hereditary MTC, distant metastasis, and extent of thyroidectomy were significant prognostic factors. Only age and stage, however, remained independent prognostic factors in multivariate analysis. The TNM, EORTC, NTCTCS, and SEER staging systems were all accurate predictors of survival, but the EORTC prognostic scoring system had the highest PVE in this cohort. CONCLUSIONS. Screening for MTC and early treatment (total thyroidectomy with central neck lymph node clearance) had nearly a 100% cure rate. Patients with postoperative hypercalcitoninemia without clinical or radiologic evidence of residual tumor after apparently curative surgery may enjoy long term survival but have occult MTC. Only patient age at presentation and TNM stage were independent predictors of survival. The EORTC criteria, which included the greatest number of significant prognostic factors in our cohort, had the highest predictive value. Cancer 2000;88:1139 – 48. © 2000 American Cancer Society. KEYWORDS: medullary thyroid carcinoma, stage, prognosis, hereditary, multiple endocrine neoplasia 2, RET, treatment and prognostic factors. 1140 M CANCER March 1, 2000 / Volume 88 / Number 5 edullary thyroid carcinoma (MTC) is a rare malignancy. It accounts for 3–10% of all thyroid carcinomas, but it is responsible for up to 13.4% of all deaths related to this disease.1–3 MTCs arise from the parafollicular cells or C cells of the thyroid gland. Familial MTC (non–multiple endocrine neoplasia [MEN]) and MEN 2 (A and B) comprise the hereditary forms of MTC.4,5 Hereditary MTC is inherited in an autosomal dominant pattern with virtually a 100% penetrance.6 A germline mutation in the RET protooncogene, which encodes for a transmembrane tyrosine kinase receptor, predisposes individuals to develop MTC. Screening for RET germline mutations has allowed for early and accurate diagnosis of patients at risk of developing MTC.7,8 Although significant advances have been made in screening patients at risk for hereditary MTC and diagnosing MTC by fineneedle aspiration cytology, some patients (especially those with sporadic MTC) still present late.3 Generally, the overall survival of patients with MTC is intermediate to that of patients with differentiated thyroid carcinoma (papillary and follicular) and anaplastic thyroid carcinoma.9 However, the clinical courses of patients with MTC vary, with long term survival reported for patients with persistent hypercalcitoninemia.10 A number of clinical (age, stage, gender, hereditary type of MTC, and symptoms), biochemical (serum calcitonin pre- and postoperative, serum carcinoembryonic antigen [CEA]), and molecular (calcitonin gene expression, calcitonin-related gene peptide, amyloid staining, and specific RET mutations) prognostic factors in patients with MTC have been evaluated.9 –20 The significance of some of these factors, however, remains controversial. Furthermore, relatively few contemporary studies have evaluated the clinical characteristics, treatment, and prognostic factors simultaneously in patients with MTC. The appropriate staging, risk stratification, and management of patients with MTC still remain of great interest to physicians (internists, endocrinologists, pediatricians, oncologists, and surgeons alike) who care for these patients. The prognostic factors that predict the courses of patients with papillary and follicular thyroid carcinoma have been addressed in the literature, with multiple prognostic scoring systems proposed over the last 4 decades.9 In patients with differentiated thyroid carcinoma, prognostic scoring systems have been helpful in categorizing patients into “high” or “low” risk groups.9 Some investigators have also compared the various prognostic staging systems to assess the “best” system for predicting the outcomes of patients with differentiated thyroid carcinoma.21–25 The utility of the different prognostic scoring systems in evaluating patients with MTC who have a higher overall mortality rate than those with differentiated thyroid carcinoma has not been addressed. In this study, we evaluated the courses of patients with MTC or C-cell hyperplasia who were managed at our institution over 4 decades. The goal of this investigation was to determine the important clinical factors that predict patient outcome in univariate and multivariate analysis. We also applied and compared the different prognostic staging systems (previously used to stage patients with MTC) in our cohort to determine the most accurate predictors of survival. PATIENTS AND METHODS Patient Selection and Classification The study group consisted of patients evaluated and/or treated at the hospitals of the University of California, San Francisco (UCSF), for MTC or C-cell hyperplasia between January 1960 and December 1998. The diagnoses of MTC or C-cell hyperplasia were all confirmed either on histology of resected specimens or by the presence of a germline RET mutation in those patients with only C-cell hyperplasia. A total of 108 patients were identified. Two patients were diagnosed at autopsy and two patients were lost to follow-up: These patients were excluded from the study group analysis. Data on the remaining 104 patients were used for survival analysis. We classified patients with MTC or C-cell hyperplasia into 4 groups: 1) sporadic, 2) familial non-MEN, 3) MEN 2A, and 4) MEN 2B.5 Patients without any first-degree relative with MEN 2, MTC, or a germline RET protooncogene mutation were considered to have sporadic MTC. Patients in the familial MTC group had a firstdegree relative with MTC but no evidence of MEN 2 (A and B) in themselves or family members. In the MEN 2A group, the patients had a first-degree relative with MEN 2A or had MTC or C-cell hyperplasia with hyperparathyroidism and/or pheochromocytoma. The patients in the MEN 2B group had MTC or C-cell hyperplasia with typical phenotypic features (e.g., marfanoid body habitus, mucosal neuroma and intestinal ganglioneuromatosis) with or without pheochromocytoma. Early in our experience, basal and pentagastrin or calcium stimulated calcitonin tests were used to screen patients for hereditary MTC. Since 1994, we have used DNA analysis for the germline RET protooncogene mutation in exons 10, 11, 13, and 14. The genetic testing was done commercially. In general, calcitonin stimulation testing for screening and follow-up consisted of pentagastrin injected intravenously with plasma calcitonin levels measured at 0, 1, 2, 5, and 10 minutes after injection. Follow-up data were obtained through the UCSF Outcomes of Patients with MTC/Kebebew et al. cancer registry and from review of patients’ clinical records. Prognostic variables were evaluated from the time of diagnosis to the last follow-up date. The study endpoints were death from thyroid carcinoma (cause specific mortality) and tumor status (free of MTC, recurrent MTC, or persistent MTC). Patients with short follow-up time or patients who died from unrelated causes were censored in the survival analysis. Tumor recurrence was defined as new evidence of MTC (locoregional or distant metastasis) after successful tumor resection in which basal or stimulated serum calcitonin levels were normal for at least 6 months. Patients with elevated (basal or stimulated) postoperative serum calcitonin levels were classified as having persistent MTC. Data Analysis Time-independent continuous variables were evaluated using the Student t test or by analysis of variance for multiple group comparison, and the chi-square test was used for categorical data. Survival analysis was done with the SAS 6.12 and StatView 4.51 statistical software packages (SAS Inc., Cary, NC). For univariate analysis of time-dependent variables, the Kaplan–Meier method and the log rank test were used to determine significant prognostic factors. In multivariate analysis, a Cox proportional hazards stepwise regression model was used for all factors found to be significant in univariate analysis. The observed differences were assumed to be statistically significant if the probability of chance occurrence was less than 0.05. The TNM,26 European Organization for Research and Treatment of Cancer (EORTC),27 National Thyroid Cancer Treatment Cooperative Study (NTCTCS),22 and Surveillance, Epidemiology, and End Results program (SEER extent of disease)1 stages were determined for each patient in the study cohort (Table 1). These prognostic classification systems have been used previously to stage patients with MTC. To compare the predictive values of the different staging systems, the proportion of variance explained (PVE) was determined for each system.22,23,28,29 The Cox proportional hazards model was used to calculate the PVE as follows: PVE ⫽ 关1 ⫺ 共Lr/Lu兲 2/N 兴, where Lr and Lu were the restricted and unrestricted, respectively, maximum likelihood ratios from the Cox proportional hazards model and N was the total number of patients. 1141 TABLE 1 The TNM, EORTC, NTCTCS, and SEER Stage Classifications for Patients with MTC Stage classification TNMa Stage I Stage II Stage III Stage IV EORTC Group 1: ⬍50 Group 2: 50–65 Group 3: 66–83 Group 4: 84–108 Group 5: ⱖ109 Definition T1, N0, M0 T2–4, N0, M0 Any T, N1, M0 Any T, any N, M1 Scoring system ⫹12 if male ⫹10 if medullary or follicular less differentiated ⫹45 if anaplastic thyroid carcinoma ⫹10 if extrathyroidal tumor invasion ⫹15 if there is 1 distant metastasis ⫹15 (in addition to above ⫹15) for multiple distant metastasis ⫽ Total score NTCTCS Stage I C-cell hyperplasia Stage II Tumor size ⬍1 cm Stage III Tumor size ⱖ1 cm or positive cervical lymph nodes Stage IV Extrathyroidal invasion or extracervical metastates SEER extent of disease Local Tumor confined to the thyroid gland Regional Extrathyroid extension or regional lymph node metastasis Metastasis Metastasis to extracervical lymph nodes or organ MTC: medullary thyroid carcinoma; EORTC: European Organization for Research and Treatment of Cancer; NTCTCS: National Thyroid Cancer Treatment Cooperative Study; SEER: Surveillance, Epidemiology, and End Results. a Tumor size: T1 ⱕ1 cm; T2 ⫽ tumor ⬎1 cm but ⬍4 cm; T3 ⬎4 cm; T4 ⫽ tumor of any size extending beyond thyroid capsule. Lymph node: N0 ⫽ no regional lymph node metastasis; N1 ⫽ positive regional lymph node metastasis. Distant metastasis: M0 ⫽ no distant metastasis; M1 ⫽ positive distant metastasis. RESULTS Clinical Characteristics Fifty-eight patients (55.8%) had sporadic MTC, 23 patients (22.1%) had familial MTC, 16 patients (15.4%) had MEN 2A, and 7 patients (6.7%) had MEN 2B (Table 2). The overall mean age at diagnosis for the study group was 38 years. Patients with sporadic MTC presented at an older age (mean ⫽ 47.3 years) than patients with hereditary MTC (P ⬍ 0.05 even when patients diagnosed by screening were excluded). The overall male-to-female ratio was 1.0:1.1, with no difference between the sporadic and hereditary MTC groups. In the sporadic group, 74.2% of patients presented with a thyroid mass, 15.5% had local symptoms (dysphagia, dyspnea, or hoarseness), and 10.3% had systemic symptoms (bone pain, flushing, and/or diarrhea) attributable to their tumor. All the patients with sporadic MTC who presented with systemic symptoms had distant metastases at the time of presentation. It is noteworthy that 4 patients with sporadic MTC had coexisting occult papillary thyroid carci- 1142 CANCER March 1, 2000 / Volume 88 / Number 5 TABLE 2 Clinical Characteristics of Patients with Sporadic and Hereditary MTC Features Sporadic MEN 2A MEN 2B Familial MTC Total No. of patients Age (mean ⫾ SD, yrs) Gender (male/female) Presenting symptom or signs Thyroid nodule Locala Systemicb Family historyc Stage (TNM) I II III IV Unknown 58 47.3 ⫾ 16.9 27/31 16 25.2 ⫾ 15.2 7/9 7 19.9 ⫾ 17.0 5/2 23 29.1 ⫾ 14.6 10/13 104 38.0 ⫾ 16.1 49/55 43 9 6 0 4 0 2 10 6 1 0 0 9 1 1 12 62 11 9 22 9 12 22 13 2 10 3 1 2 — 1 0 5 1 — 10 5 7 1 — 30 20 35 17 2 a “Local” refers to those patients who presented with dysphagia, hoarseness, dyspnea, or vocal cord paralysis. Systemic symptoms and signs that prompted workup and the eventual diagnosis of MTC were bone pain, flushing, and diarrhea in the patients with sporadic MTC. Those patients in the hereditary group had manifestation of pheochromocytoma (2 patients) or bone pain from MTC metastasis that prompted a workup for MTC. c “Family history” refers to those patients who were at risk for MTC and were screened for MTC with biochemical and/or genetic testing. b noma discovered on permanent histology. In patients with hereditary MTC, 41.3% presented initially with a thyroid mass, 4.4% with local symptoms, and 6.5% with systemic symptoms; 47.8% were diagnosed by biochemical and/or genetic screening and were asymptomatic. Of the three patients who presented with systemic symptoms in the hereditary MTC groups, one patient had metastasis to the bone and reported bone pain at presentation. One patient presented in hypertensive crisis secondary to a pheochromocytoma and one patient presented with a retroperitoneal hemorrhage from a ruptured pheochromocytoma. The latter two patients were then subsequently found to have a germline RET mutation by genetic testing and had a prophylactic total thyroidectomy with central neck lymph node clearance. The patients with MEN 2A presented with significantly fewer cervical lymph node metastases than patients with sporadic MTC, MEN 2B, or familial MTC (P ⬍ 0.05). Eighty-five percent of the patients with hereditary MTC had bilateral tumors, whereas 66.7% of the patients with sporadic MTC had bilateral tumors. There was no difference in the overall stage of MTC at presentation between patients with hereditary and sporadic MTC. Almost half (47.8%) of the patients with hereditary MTC were diagnosed by screening (biochemical and/or genetic testing). Of note, 3 of these patients who underwent prophylactic total thyroidectomy with central neck lymph node clearance for positive germline RET mutations (2 patients with familial MTC and 1 patient with MEN 2A) had no evidence of MTC or C-cell hyperplasia on histologic examination, even after calcitonin immunostaining.30 In the remaining 19 patients, 2 had only C-cell hyperplasia and 17 had MTC. Patients with hereditary MTC diagnosed by screening had a lower frequency of cervical lymph node metastases than patients with hereditary MTC who were not screened (P ⬍ 0.05). Furthermore, 94.7% of patients diagnosed by screening were free of MTC, compared with 32.5% of patients with hereditary MTC who were not screened (P ⬍ 0.0001 with a mean follow-up time of 6.3 years and 8.8 years, respectively). Treatment and Outcome Two patients with MEN 2A await surgery at this time. In the remaining 102 patients, 76.6% had total or neartotal thyroidectomy, 9.8% had subtotal thyroidectomy, 9.8% had thyroid lobectomy, and 3.8% had incomplete resection of their tumor. Those patients who had lessthan-total or near-total thyroidectomy had their initial surgery at other medical centers or were treated before 1974. Fifty-five percent of patients had cervical lymph node dissection: 85.3% had central neck lymph node dissection and 14.7% had both central neck and modified radical neck dissection. One patient had a planned resection of the recurrent laryngeal nerve because of tumor invasion. There was one recurrent laryngeal nerve, phrenic nerve, sympathetic nerve, and spinal accessory nerve injury, each in different patients, during operative resection of locally invasive MTC, at our institution. On reoperation, one patient Outcomes of Patients with MTC/Kebebew et al. TABLE 3 Evaluation of Clinicopathologic Variables as Risk Factors for Recurrent or Persistent MTC 1143 TABLE 4 Clinical Prognostic Variables for Patients with MTC in Univariate and Multivariate Analysis Clinicopathologic variables Variable significancea P value Clinicopathologic variables Univariate P value Multivariate P value Age (yrs) Gender Type of MTC FMTC vs. others Sporadic vs. others MEN 2A vs. others MEN 2B vs. others TNM stage I vs. II/III/IV I/II vs. III/IV I/II/III vs. IV Cervical lymph node metastasis Multicentricity Extent of thyroidectomy Total/near total vs. lesser procedures Total and subtotal vs. lesser procedures Incomplete resection Cervical lymph node dissection S NS S S S NS NS S S S S S NS S NS S S NS 0.048 0.187 0.022 0.005 0.017 1.00 0.671 0.002 0.002 0.002 0.006 0.047 0.166 Agea Gender Ethnicity Presenting symptoms and signsb M vs. A M vs. L M vs. S Type of MTC TNM stage I vs. II I vs. III I vs. IV Tumor size Cervical lymph node metastasis Tumor multicentricity Treatment Extent of thyroidectomy Lymph node dissection External beam radiation Chemotherapy S S NS 0.0149 0.0535 0.3638 S NS — 0.0272 — — S NS S S NS S NS NS S NS 0.0033 0.3178 0.0006 ⬍0.0001 0.2323 0.0001 0.1170 0.0758 0.0006 0.084 NS — — S — 0.0013 — — NS NS 0.5238 0.2015 — — — — S NS NS NS ⬍0.0001 0.5778 0.7365 0.8855 NS — — — — — — 0.058 0.029 0.006 0.257 MTC: medullary thyroid carcinoma; FMTC: familial MTC; MEN: multiple endocrine neoplasia; S: significant; NS: not significant. a Significance was determined by the chi-square analysis, except for age (t test). had an injury to the thoracic duct with postoperative chyle leak, which resolved after evacuation and ligation of the duct. At last follow-up, 49.4% of the patients with MTC were free of disease, 38.3% had persistent MTC, and 12.3% had recurrent MTC. In the patients who were cured, 77.5% had total thyroidectomy with central neck lymph node clearance and 57.5% had Stage I or II MTC. Overall, patients who were younger, had familial MTC or MEN 2A, had lower stage MTC, had no diarrhea or bone pain, and had total or subtotal thyroidectomy were less likely to have recurrent or persistent MTC (Table 3). All of the patients diagnosed by screening were cured except for a woman age 62 years who had an asymptomatic MTC metastasis to her spine at the time of screening. Furthermore, patients who had total thyroidectomy with central neck lymph node clearance on average (0.8 vs. 2.8) required fewer reoperations than patients who had lesser procedures (P ⬍ 0.031). Eight patients received therapeutic external beam radiation (one patient hyperfractionated) to the cervical region and four patients received chemotherapy regimens (two patients received systemic chemotherapy, one patient had chemoembolization for liver metastases, and one patient was treated with anti-CEA radiolabeled antibody). No appreciable tumor regression was observed in any of these patients. Two patients, however, were treated by radiofre- MTC: medullary thyroid carcinoma; S: significant; NS: not significant. a Age at presentation was a significant prognostic factor for patients aged ⱕ45 years. b M: thyroid mass; A: asymptomatic; L: local; S: systemic (see Table 2 for a description of these categories). quency ablation for liver metastases with tumor regression. Prognostic Factors and Staging Systems The mean follow-up time was 8.6 years (median, 5.0 years). The overall cause specific mortality was 10.7% at 5 years and 13.5% at 10 years. Age, gender, clinical presentation, TNM stage of MTC, distant metastasis, and extent of thyroidectomy were significant predictors of survival by univariate analysis (Table 4). Only age and stage of disease remained independent predictors of survival in multivariate analysis (Table 4 and Fig. 1). The 5-year survival rates by stage were 100% (I), 90% (II), 86.5% (III), and 55.5% (IV). Patients who were age ⬍45 years, female, and with MTC confined to the thyroid had the best overall prognosis (100% survival at 10 years). All the prognostic staging systems used to evaluate patients with MTC were accurate predictors of survival (Table 5). The EORTC risk assessment model had the highest PVE value. DISCUSSION The differences in prognostic factors reported for MTC may be due in part to the low frequency of occurrence 1144 CANCER March 1, 2000 / Volume 88 / Number 5 TABLE 5 Comparison of Staging Systems for Patients with MTC Staging system EORTC group 1 2 3 4 5 TNM stage I II III IV NTCTCS stage I II III IV SEER criteria Local Regional Distant FIGURE 1. Cancer specific cumulative survival by Kaplan–Meier analysis is shown for all patients with medullary thyroid carcinoma. (A) Survival for patients ages ⬍ 45 and ⱖ 45 years is compared with (B) survival by TNM stage of disease at presentation. of MTC and its heterogeneous clinical behavior. Although there have been large cohort multicenter or cancer registry– based studies, many of them lack complete clinical data and suffer from nonuniform diagnostic criteria. A single-institution study, although of a smaller cohort, may circumvent such problems. A long period of follow-up is usually required for such studies at a single institution, especially to study multiple prognostic factors. We believe our study period and follow-up interval represent an adequate length of time to evaluate the important clinical factors that determine the outcomes of patients with MTC.11–13,15,17,18 Furthermore, we have evaluated the influence of a comprehensive group of clinical factors on the outcomes of patients with MTC. The mean age at diagnosis in our cohort is similar to that in other studies.12,15,17–19,23 Although in most series sporadic MTC accounts for 75–95% of all MTCs, we found that 54.3% of the patients had sporadic MTC during the study period.3,12,16,18 –20 The number of hereditary MTC cases in our study is comparable to that of Modigliani et al.17 The higher proportion of hereditary MTC in our cohort most likely reflects a referral bias. In addition, a higher frequency of hereditary MTC may be observed in contemporary series because Cause specific mortality (%) by stage or group P value PVE (%) Rank order based on PVE 0.0001 15.147 1 0.0017 13.118 2 0.0012 12.715 3 0.0007 12.131 4 0 8.7 33.3 40.0 100.0 0 10.0 14.5 44.5 0 5.3 10.5 38.9 10.8 10.8 29.6 MTC: medullary thyroid carcinoma; PVE: proportion of variance explained; EORTC: European Organization for Research and Treatment of Cancer; NTCTCS: National Thyroid Cancer Treatment Cooperative Study; SEER: Surveillance, Epidemiology, and End Results. of the increased awareness in screening patients for MTC that is now possible with commercially available with genetic testing. Unlike other series,15,17,18,20 which have reported a higher frequency of MEN 2A in the hereditary types of MTC, we found that 50% of our patients had familial MTC. Patients with MEN 2A and MEN 2B accounted for 33.3% and 16.7%, respectively, of the hereditary MTC cases. Because of the incomplete penetrance of hyperparathyroidism and pheochromocytoma in patients with MEN 2A, the distinction between familial MTC and MEN 2A can be reliably made only after long term follow-up. This would ensure that those patients with familial MTC do not eventually develop hyperparathyroidism and/or pheochromocytomas.31 The majority of MEN 2B patients presented in the second decade of life, and familial non-MEN and MEN 2A patients presented in the third decade of life. The ratio of women to men (1:1.1) is consistent with the observations of others11–13,15–20 and is unlike the ratios for other thyroid tumors, which have a female preponderance.1,2 The majority of patients with sporadic MTC (70%) initially presented with a thyroid nodule or mass, which led to the diagnosis. Patients who presented with local symptoms were more likely to have either persistent or recurrent disease compared with pa- Outcomes of Patients with MTC/Kebebew et al. tients who presented with a thyroid nodule (P ⬍ 0.05). All patients with sporadic MTC who presented with systemic symptoms had Stage IV MTC, and all were dead at follow-up. One of 11 patients with seemingly sporadic MTC was diagnosed as having familial MTC because of a positive germline RET mutation. None of the patients diagnosed by screening had cervical lymph node or distant metastases, with the exception of the asymptomatic woman age 62 years who was mentioned earlier. Furthermore, 3 patients who underwent prophylactic total thyroidectomy with central neck lymph node clearance had no MTC or C-cell hyperplasia. These observations emphasize that early detection in patients at risk for MTC allowed for nearly a 100% rate of cure, compared with the overall 49.4% cure rate in our cohort. We believe that patients diagnosed by screening warrant a prophylactic total thyroidectomy at about age 6 years. The positive RET germline mutation should be confirmed at least once prior to thyroidectomy. Patients with MEN 2B need to have a total thyroidectomy when they are initially diagnosed. In children, we recommend central cervical lymph node clearance at the time of total thyroidectomy if 1) the basal or stimulated calcitonin is elevated, 2) there is a thyroid lesion identified on ultrasound or at the time of surgery, or 3) there are any abnormal lymph nodes. Although C-cell hyperplasia and bilateral MTC are less common in sporadic MTC, our patients with sporadic MTC had a high rate of bilateral MTC. In all such cases familial disease must be considered. However, these patients with sporadic MTC had no family history and 17% had no germline RET mutations in exon 10, 11, 13, and 14. Nonetheless, we recommend that all patients with MTC be screened for the RET germline proto-oncogene mutations, as up to 10% of patients with sporadic MTC may represent an index case of hereditary MTC.3,31 Fine-needle cytology is normally accurate for diagnosing MTC. In some patients the type of thyroid carcinoma is difficult to determine and Hürthle cell neoplasms can occasionally be confused with MTC cytologically and on frozen section. Staining for amyloid, calcitonin, CEA, and thyroglobulin can help confirm the diagnosis. All patients with MTC must also be screened for pheochromocytoma prior to any operation. It is intriguing that the extent of thyroidectomy was not an independent predictor of mortality, given the higher likelihood of persistent or recurrent MTC in those patients who had lesser procedures than total or subtotal thyroidectomy. We speculate that the long survival time of some patients with persistent postoperative hypercalcitoninemia may account for this. Fur- 1145 thermore, the majority of the patients with MTC had total thyroidectomy with central neck lymph node clearance. Patients who had lesser procedures, however, did require more reoperations than patients who had total thyroidectomy with central neck lymph node clearance, as we have previously reported.5 Most of the patients treated at our institution had a central neck dissection, removing all fibrofatty tissue from just above the innominate vein between the carotid sheath and trachea to the level of the thyroid cartilage (central neck dissection), and a functional neck dissection, preserving the motor and sensory nerves but removing all fibrofatty and lymphatic tissue from the carotid sheath to the trapezius and from the clavicle to just above the hyoid bone. In retrospective studies, chemotherapy and radiotherapy have been reported to be either helpful or detrimental.3,19,32,33 The differing results probably depend on patient selection. In our experience, those patients with MTC who had chemotherapy and/or external radiotherapy had shorter survival times (not statistically significant), but these patients also had metastatic MTC. We recommend external radiation treatment when macroscopic tumor is left or is unresectable. When evaluating patients with persistent hypercalcitoninemia, we first determine whether the patient has had appropriate initial surgical treatment. We believe this includes 1) at least a total or near-total thyroidectomy, 2) central neck dissection, and 3) modified functional neck and superior mediastinal dissection.34 Prior to this treatment, we generally obtain a neck ultrasound or magnetic resonance imaging scan of the neck and mediastinum. A computed tomography scan of the liver and lung is also helpful for identifying liver and lung metastasis, which, however, are usually miliary. In our experience, virtually all patients who have markedly elevated calcitonin levels after appropriate initial treatment have distant metastasis. Bone scans are useful to screen for bony metastasis. Our study revealed that 38.3% of the patients had persistent MTC, but these patients lived for 1 month to 23.7 years and 65.9% were alive at last follow-up. This emphasizes that the subset of patients with postoperative persistent hypercalcitoninemia and no clinical or radiographic evidence of residual tumor could be followed with apparent long-term survival. As reported by Tissel et al.,35 Dralle et al.,36 and Moley et al.,37 we recommend that patients with persistent hypercalcitoninemia have a meticulous total thyroidectomy and central neck dissection, if not already done; this is a functional lateral neck dissection in patients who have what appears to be curable disease. We also recom- 1146 CANCER March 1, 2000 / Volume 88 / Number 5 mend palliative resection of locoregional disease even in patients with known distant metastasis.38 The overall cause specific mortality observed in our cohort is slightly better than that reported in other studies (13.3–32.6% and 21.6 –38.6% at 5 years and 10 years, respectively).15–20 Patient age and stage of MTC have been consistently observed to be the most important prognostic factors in patients with MTC.1,12,13,15,17,18 Indeed, in our study, these were the only two independent prognostic factors in patients with MTC. Patient gender approached statistical significance in univariate analysis and probably represents a minor risk factor. Some,12,13,16,18 but not all,11,15,19,20 studies have also reported gender to be a significant prognostic factor. The type of MTC (familial MTC, MEN 2A, sporadic MTC, and MEN 2B, in increasing order of mortality rates) has been reported to be an important prognostic factor in some studies.12,17,18 However, when controlling for the stage of disease or multivariate analysis was done, the type of MTC was not an independent prognostic factor.11,13,17,18,20,32 This is consistent with our finding. Although patients with cervical lymph node metastasis were more likely to have persistent or recurrent MTC, the presence of cervical lymph node metastasis was not associated with significantly higher mortality. Although this is in conflict with some studies,11,15,16 others have reported that cervical lymph node metastasis was not an independent prognostic factor.12,19,33,39 Previous studies of patients with differentiated thyroid carcinoma have demonstrated the importance of the prognostic staging systems.9,22–25,27 To our knowledge, however, the utility and predictive value of these staging systems have not been addressed regarding patients with MTC. The rationale for comparing these staging systems was 1) to determine the most accurate staging system that best predicts survival, and 2) to evaluate the influence of the individual prognostic factors on survival in these prognostic staging models. We recognize that some of these retrospective risk models involving patients with MTC have been originally derived from a single population and might not necessarily apply to another population.40 Their utility, however, has been validated in follow-up studies or by other investigators who have evaluated such staging systems in different cohorts, or in prospective studies.21,22,23,41 It is noteworthy that the EORTC staging system had the highest PVE value and that age, gender, and distant metastasis are included in this system. This is consistent with our observation that age and stage (mainly the presence of distant metastasis) were the main determinants of survival in our multivariate analysis. The apparent higher PVE of the EORTC compared with the TNM system may also be due to the absence of patient age in the TNM system for MTC, which has been observed to be an important prognostic factor in most studies, including ours.1,11,12,13,16,17,39 This suggests that a similar TNM staging system should be used for MTC as is used for papillary and follicular thyroid carcinoma, in which patient age is included in the TNM staging system. It is interesting that the PVE value was lower for the NTCTCS staging system than for the TNM system because both are based on the same pathologic data, with distinction between the two systems only in the stage distribution (Table 1). The inclusion of otherthan-distant metastasis (i.e., extrathyroidal invasive tumors) into the Stage IV group of the NTCTCS system may account for this apparent difference, which did not portend a poor prognosis in our cohort (Table 1). Furthermore, the higher PVE and mortality rates by stage (for Stage ⱖII) observed in the TNM staging system compared with the NTCTCS system suggest a more meaningful stage classification for the TNM system, at least in our cohort. Because the PVE is sensitive to the number of categories in a particular staging system, it is not surprising, then, that the SEER extentof-disease staging system (local, regional, and distant) had the least predictive value for patient survival. In summary, the initial clinical presentation of patients with MTC was useful in determining which patients were at risk for poor outcome or for persistent or recurrent MTC. Patients with familial MTC and MEN 2A who were diagnosed at an early age and had total thyroidectomy with central neck lymph node clearance had an excellent prognosis. Patient age and stage of disease were independent prognostic factors in all patients with MTC. The model that included the greatest number of significant prognostic factors yielded a higher predictive value for survival. Although the EORTC scoring system had the highest PVE value in our cohort, the more commonly used TNM staging system should probably be used to report treatment outcome, for uniform comparison of treatment and outcomes of patients with MTC. Furthermore, our findings suggest that patients with persistent hypercalcitoninemia and no clinical or radiographic evidence of residual tumor may have long survival time. These findings should be reassuring to those patients with persistently elevated serum calcitonin levels after apparent curative surgical resection and who are without clinical evidence of MTC. REFERENCES 1. Gilliland FD, Hunt WC, Morris DM, Key CR. Prognostic factors for thyroid carcinoma: a population-based study of 15,698 cases from the Surveillance, Epidemiology and End Results (SEER) program, 1973–1991. Cancer 1997;79:564 –73. Outcomes of Patients with MTC/Kebebew et al. 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. Hundahl SA, Fleming ID, Fremgen AM, Menck HR. A national cancer database report on 53,856 cases of thyroid carcinoma treated in the U.S., 1985–1995. Cancer 1998;83: 2638-48. Marsh DJ, Learoyd DL, Robinson BG. Medullary thyroid carcinoma: recent advances and managment update. Thyroid 1995;5:407–20. Block MA, Jackson CE, Greenwald KA, Yott JB, Tashjian A. Clinical characteristics distinguishing hereditary from sporadic medullary thyroid carcinoma: treatment implications. Arch Surg 1980;115:142– 8. Duh QY, Sancho JJ, Greenspan FS, Hunt TK, Galante M, deLorimier AA, et al. Medullary thyroid carcinoma: the need for early diagnosis and total thyroidectomy. Arch Surg 1989; 124:1206 –10. Ponder BA, Ponder MA, Coffey R, Pembrey ME, Gagel RF, Telenius-Berg M, et al. Risk estimation and screening in families of patients with medullary thyroid carcinoma. Lancet 1988;1:397– 401. Mulligan LM, Kwok JB, Healey CS, Elsdon MJ, Eng C, Gardner E, et al. Germ-line mutations of the RET proto-oncogene in multiple endocrine neoplasia type 2A. Nature (1993;363: 458 – 60. Donis-Keller H, Dou S, Chi D, Carlson KM, Toshima K, Lairmore TC, et al. Mutations in the RET proto-oncogene are associated with MEN 2A and FMTC. Hum Mol Genet 1993;2:851– 6. Treseler PA, Clark OH. Prognostic factors in thyroid carcinoma. Surg Oncol Clin North Am 1997;6:555–98. van Heerden JA, Grant CS, Gharib H, Hay ID, Ilstrup DM. Long-term course of patients with persistent hypercalcitoninemia after apparent curative primary surgery for medullary thyroid carcinoma. Ann Surg 1990;212:395– 400. Chong GC, Beahrs OH, Sizemore GW, Woolner LH. Medullary carcinoma of the thyroid gland. Cancer 1975;35:695– 704. Saad MF, Ordonez NG, Rashid RK, Guido JJ, Hill CS, Hickey RC, et al. Medullary carcinoma of the thyroid: a study of the clinical features and prognostic factors in 161 patients. Medicine 1984;63:319 – 42. Schroder S, Bocker W, Baisch H, Burk CG, Arps H, Meiners I, et al. Prognostic factors in medullary thyroid carcinomas: survival in relation to age, sex, stage, histology, immunocytochemistry and DNA content. Cancer 1988;61: 806 –16. Moley JF. Medullary thyroid cancer. Surg Clin North Am 1995;75:405–20. Dottorini ME, Assi A, Sironi M, Sangalli G, Spreafico G, Colombo L. Multivariate analysis of patients with medullary thyroid carcinoma: prognostic significance and impact on treatment of clinical and pathologic variables. Cancer 1996; 77:1556 – 65. Bergholm U, Adami HO, Bergstrom R, Johansson H, Lundell G, Telenius-Berg M, et al. Clinical characteristics in sporadic and familial medullary thyroid carcinoma: a nationwide study of 249 patients in Sweden from 1959 through 1981. Cancer 1989;63:1196 –204. Modigliani E, Cohen R, Campos JM, Conte-Devolx B, Maes B, Boneu A, et al. Prognostic factors for survival and for biochemical cure in medullary thyroid carcinoma: results in 899 patients. The GETC Study Group. Clin Endocrinol 1998; 48:265–73. 1147 18. Raue F, Kotzerke J, Reinwein D, Schroder S, Roher HD, Deckart H, et al. Prognostic factors in medullary thyroid carcinoma: evaluation of 741 patients from the German Medullary Thyroid Carcinoma Register. Clin Invest 1993;71: 7–12. 19. Brierley JD, Tsang R, Simpson WJ, Gospodarowicz M, Sutcliffe S, Panzarella T. Medullary thyroid cancer: analyses of survival and prognostic factors and the role of radiation therapy in local control. Thyroid 1996;6:305– 10. 20. Gharib H, McConahey WM, Tiegs RD, Bergstralh EJ, Goellner JR, Grant CS, et al. Medullary thyroid carcinoma: clinicopathologic features and long-term follow-up of 65 patients treated during 1946 through 1970. Mayo Clin Proc 1992;67:934 – 40. 21. Davis NL, Bugis SP, McGregor GI, Germann E. An evaluation of prognostic scoring systems in patients with follicular thyroid cancer. Am J Surg 1995;170:476 – 80. 22. Brierley JD, Panzarella T, Tsang RW, Gospodarowicz MK, O’Sullivan B. A comparison of different staging systems predicting patient outcome: thyroid carcinoma as an example. Cancer 1997;79:2414 –23. 23. Sherman SI, Brierley JD, Sperling M, Ain KB, Bigos ST, Cooper DS, et al. Prospective multicenter study of thyroid carcinoma treatment: initial analysis of staging and outcome. Cancer 1998;83:1012–21. 24. Sanders LE, Cady B. Differentiated thyroid cancer: reexamination of risk groups and outcome of treatment. Arch Surg 1994;133:419 –25. 25. Hay ID, Bergstralh EJ, Goellner JR, Ebersold JR, Grant CS. Predicting outcome in papillary thyroid carcinoma: development of a reliable prognostic scoring system in a cohort of 1779 patients surgically treated at one institution during 1940 through 1989. Surgery 1993;114: 1050 – 8. 26. Thyroid gland. In Fleming ID, Cooper JS, Henson DE, et al., editors. AJCC cancer staging manual. 5th edition. Philadelphia: Lippincott-Raven, 1997:59 – 64. 27. Byar DP, Green SB, Dor P, Williams D, Colon J, van Gilse HA, et al. A prognostic index for thyroid carcinoma: a study of the E.O.R.T.C. thyroid cancer cooperative group. Eur J Cancer 1979;15:1033– 41. 28. Schemper M. The relative importance of prognostic factors in studies of survival. Stat Med 1993;12:2377– 82. 29. Schemper M, Stare J. Explained variation in survival analysis. Stat Med 1996;15:1999 –2012. 30. Kebebew E, Treseler PA, Siperstein AE, Duh Q-Y, Clark OH. Normal thyroid pathology in patients undergoing thyroidectomy for finding a RET gene germline mutation: a report of three cases and review of the literature. Thyroid 1999;9: 127–31. 31. Decker RA, Peacock ML. Update on the profile of multiple endocrine neoplasia type 2a RET mutations. Cancer 1997; 80:557– 68. 32. Samaan GA, Schultz PN, Hickey RC. Medullary thyroid carcinoma: prognosis of familial versus sporadic disease and the role of radiotherapy. J Clin Endocrinol Metab 1988;67: 801–5. 33. Rougier P, Parmentier C, Laplanche A, Lefevre M, Travagli JP, Cailou B, et al. Medullary thyroid carcinoma: prognostic factors and treatment. Int J Radiat Oncol Biol Phys 1983;9: 161–9. 1148 CANCER March 1, 2000 / Volume 88 / Number 5 34. Moley JF, DeBenedetti MK. Patterns of nodal metastases in palpable medullary thyroid carcinoma: recommendations for extent of node dissection. Ann Surg 1999;229: 880 –7. 35. Tissel LE, Hansson G, Jansson S, Salander H. Reoperation in the treatment of asymptomatic metastasizing medullary thyroid carcinoma. Surgery 1986;99:60 – 6. 36. Dralle H, Damm I, Scheumann GF, Kotzerke J, Kupsch E, Geerlings H, et al. Compartment-oriented microdissection of regional lymph nodes in medullary thyroid carcinoma. Surg Today 1994;24:112–21. 37. Moley JF, Dilley WG, DeBenedetti MK. Improved results of cervical reoperation for medullary thyroid carcinoma. Ann Surg 1997;225:734 – 43. 38. Chen H, Roberts JR, Ball DW, Eisele DW, Baylin SB, Udels- man R, et al. Effective long-term palliation of symptomatic, incurable metastatic medullary thyroid cancer by operative resection. Ann Surg 1997;59:279 – 87. 39. Rossi RL, Cady B, Meissner WA, Wool MS, Sedgewick CE, Werber J. Nonfamilial medullary thyroid carcinoma. Am J Surg 1980;139:554 –9. 40. Hannequin P, Liehn JC, DeLisle MJ. Multifactorial analysis of survival in thyroid cancer: pitfalls of applying the results of published studies to another population. Cancer 1986;58: 1749 –55. 41. Tennvall J, Birkland A, Moller T, Ranstam J, Akerman M. Is the E.O.R.T.C. prognostic index of thyroid cancer valid in differentiated thyroid carcinoma? Retrospective multivariate analysis of differentiated thyroid carcinoma with long follow-up. Cancer 1986;57:1405–14.