| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
Original Article |
Departments of General Surgery, Public Health Sciences, Nuclear Medicine, and Pathology, Wake Forest University, Medical Center Boulevard, Winston-Salem, North Carolina 27157
Correspondence: Address correspondence and reprint requests to: Nancy D. Perrier, MD, FACS, Department of Surgical Oncology, Unit 444, The University of Texas M.D. Anderson Cancer Center, P.O. Box 301402, Houston, TX 77230-1402, USA; E-mail: nperrier{at}mdanderson.org.
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Methods: The authors present a retrospective chart review on patients with parathyroid carcinoma from 1975 to 2004, identified by the tumor registry of a single tertiary-care center. Diagnoses were confirmed histologically, clinical and radiographical data were recorded, and statistical analyses were performed.
Results: Twenty-three cases were identified. The mean patient age was 54 years. The female:male ratio was 1.5:1. Follow-up ranged from 1 month to 23 years (median, 134 months). Mean preoperative calcium was 12.9 mg/dL. Median parathyroid hormone was 290 pg/mL. Two patients (9%) had an asymptomatic presentation, and five (22%) presented with a palpable neck mass. Only nine (39%) underwent initial comprehensive en-bloc resection. Median survival was 22 years. Five- and 10-year survival was 85.9% and 69.4%, respectively. Five- and 10-year survival with en-bloc resection was 90% and 67.5%, respectively. Local resection resulted in survival rates of 82.5% and 70.7%. Three of ten deaths were attributed to parathyroid carcinoma. In recurrent disease, computed tomography and scintigraphy had localization rates of 53% and 67%, respectively, with a concordance of 22%.
Conclusions: Long-term survival is possible with parathyroid carcinoma. Death associated with parathyroid carcinoma was uncommon. A staging may be warranted despite the rarity of this disease.
Key Words: Parathyroid • Carcinoma • Endocrine • Primary hyperparathyroidism • Survival
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
PATIENTS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
). Suggestive features including broad fibrous bands, capsular invasion, and diffuse sheetlike arrangement were noted as suspicious features. By using the institutional tumor registry, follow-up data concerning survival and cause of death were obtained when available. All patients had follow-up directly within the institution or through correspondence with primary physicians. Personal phone contact was made when necessary. Primary care physicians were contacted for further follow-up data when needed. The Social Security Death Index confirmed the dates of all deaths. Current records to within 3 years were available for all but eight patients. In these eight, the latest follow-up was not within 5 years, but most (seven of eight) had died before 1999. This left only one patient with no follow-up. Operative notes received intense review to determine whether characteristics suggestive of parathyroid carcinoma were noted, such as a hard mass, adherence or gross invasion into adjacent structures, or a white or gray color of the lesion. Clinical and biochemical parameters were recorded. Sestamibi and computed tomography (CT) scans were reviewed independently by a dedicated endocrine nuclear medicine physician. Median survival and 5- and 10-year survival rates were calculated using Kaplan-Meier methods. Log-rank tests were used to compare subsets of patients. Cox proportional hazards regression was used to assess factors that may predict survival. A thorough literature review was performed with the PubMed and MDConsult Web search engines.
|
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
Incidence
Data were not available for the total number of patients with primary HPT diagnosed from 1975 to 1995. However, data were available for the 8 years from 1996 to 2004. During this time, 583 patients at our institution were diagnosed with primary HPT, and 4 patients were diagnosed with parathyroid carcinoma. During these 8 years, parathyroid carcinoma therefore represented .68% of all patients with primary HPT. Nineteen (83%) patients were primarily diagnosed and treated at our institution; four (17%) patients were referred with a known diagnosis of parathyroid carcinoma. One patient had a family history of benign HPT. No patient had a family history of known parathyroid carcinoma. There was no evidence to suggest that any of these patients had or had family members with hyperparathyroid jaw tumor syndrome or multiple endocrine neoplasia type I or II. No patient had previous ethanol injection of parathyroid tissue.
Signs/Symptoms
The most common presenting signs and symptoms included a history of nephrolithiasis in nine patients (48%), hypertension in seven (30%), gastrointestinal entities such as gastroesophageal reflux disease, nausea, vomiting, or peptic ulcer disease in seven (30%), gout in three (13%), and arthritis in three (13%). Pathologic fracture, depression, and diabetes mellitus were all present in two patients (9%). Two patients (9%) had no objective symptoms at presentation. One presented with a serum calcium of 13.4 mg/dL, and the other, with 13.7 mg/dL. The former patient had a synchronous parathyroid hormone (PTH) of 126 pg/dL. Neither had a palpable mass or other signs to suggest to the surgeon the presence of parathyroid carcinoma. A palpable neck mass was documented before surgery in five patients (22%). In two (9%) patients, the surgeon suspected carcinoma before surgery on the basis of findings such as an extremely increased serum calcium level (15.4 mg/dL), palpable neck mass, or concurrent bone and renal disease.
Laboratory Data
Preoperative laboratory values revealed serum calcium levels ranging from 8.7 to 21.9 mg/dL, with a mean of 12.9 mg/dL. Postoperative calcium levels ranged from 7.6 to 11.8 mg/dL, with a mean value of 9.5 mg/dL. Preoperative PTH levels ranged from 96 to 1542 pg/dL, with a median value of 290 mg/dL for the 11 available values. The intraoperative PTH assay was used in one patient and demonstrated a 94% decline in PTH at 5 minutes.
Imaging
Only two patients underwent initial preoperative radiographic imaging. One had a chest radiograph, which revealed a large mediastinal mass that displaced the trachea. The second had scintigraphy, which localized a solitary hot spot assumed to be an adenoma. Most of these patients received care before the use of routine preoperative imaging (1980s and 1990s). Currently at our institution, routine use of sestamibi is used only for patients amenable to a minimally invasive surgical approach or reoperative surgery. Because of our excellent success rate, sestamibi imaging of solitary adenomas, CT, magnetic resonance imaging, positron emission topography, and ultrasound are not routinely used. Most imaging studies were for recurrent disease for which the primary surgical intervention had been performed elsewhere. Nine patients underwent a total of 32 studies in attempts to localize the source of recurrent disease. Several patients had multiple recurrences with multiple images. Scintigraphy scans were performed 15 times, and 17 CT scans were performed. The localization rate of scintigraphy was 67% (10 of 15 studies). CT had a localization rate of 53% (9 of 17 studies). In nine instances, patients underwent both CT and scintigraphy. The complementary studies were concordant in two (22%) of nine patients. Scintigraphy correctly identified sites of recurrence in five (71%) of the seven patients with discordant imaging studies. In the remaining two patients, CT properly identified the site of recurrent disease.
Treatment
Treatment included a total of 47 surgical procedures. Standard cervical exploration was performed in 21 patients (91%) at 38 different settings. Eleven patients (48%) underwent parathyroidectomy only. Nine patients (39%) underwent concurrent ipsilateral thyroid lobectomy (described as comprehensive en-bloc resection). Two patients (9%) underwent interval ipsilateral thyroid lobectomy. Three patients underwent concurrent thymectomy (13%). Five patients (22%) underwent interval mediastinal exploration for persistent or recurrent disease, and pulmonary metastasectomy was performed in one patient (4%). One other patient (4%) underwent minimally invasive parathyroidectomy. Two patients were treated with external beam radiation as adjuvant therapy. Both had recurrent parathyroid carcinoma, and both died as a result of this disease.
Diagnosis
In 19 patients (83%), the surgeon suspected carcinoma at the initial operation on the basis of preoperative or intraoperative data. In 12 (52%) of these 19 patients, intraoperative findings suggestive of carcinoma were noted in the operative dictation report. In 5 (22%) of these 19 patients, the patient had a preoperative diagnosis of cancer, and the surgeon was performing a resection for recurrence. In two other patients (9%), the surgeon suspected carcinoma on the basis of preoperative clinical data. Carcinoma was diagnosed after surgery in the final four patients (17%). In these patients, the diagnosis was based on histological findings.
Complications
Postoperative complications occurred in 4 (9%) of the 47 procedures. These included unilateral recurrent laryngeal nerve injury in a single patient, unilateral Horner’s syndrome in a single patient, and hypocalcemia in two patients.
Pathology
Pathologic data revealed well-differentiated carcinoma in 15 (70%) patients and well-differentiated cystic features in 2 patients (9%). Four cases (17%) were graded intermediate, and three (13%) were poorly differentiated. Tumor size ranged from 80 mg to 13 g, with a mean of 2710 mg. A single patient had an excisional biopsy of a large mediastinal mass; this confirmed parathyroid carcinoma. No patient had lymph node involvement. Fourteen patients had available material for re-review. Upon review of all available histological materials, two patients were thought to have borderline cases of parathyroid carcinoma. The first patient’s histology did not reveal metastases, definite soft tissue, or vascular invasion but did show lymphatic invasion. However, three of the suggestive features, including capsular invasion, broad fibrous bands, and a diffuse sheetlike arrangement, were also noted. It is interesting to note that the surgeon had a high suspicion of carcinoma at the time of surgery, stating "...the parathyroid was more firm and fixed than normal and gray-white in color." The second borderline patient underwent resection that showed two individual masses that seemed to have "collided." The first was believed to be a parathyroid adenoma; the second was believed to be a carcinoma. There was no local invasion, but capsular invasion, possible vascular invasion, increased mitotic figures, atypia, macronucleoli, and diffuse sheetlike cells were noted. This patient had no evidence of metastatic disease. Five of eight pathologists from three different institutions favored the diagnosis of collision tumor consisting of carcinoma and a concurrent adenoma.
Recurrence
Five patients (22%) developed a recurrence between 3 and 11 years from original diagnosis. Three of these patients had a local resection; two had comprehensive en-bloc resection (Table 1). Two of these patients died 48 and 97 months (4 and 6.3 years) after diagnosis, one with local failure and one with an unknown cause of death. These patients had not undergone subsequent attempts at surgical palliation. The remaining three have undergone a total of seven operations in attempts to palliate hypercalcemia, including repeated neck explorations, interval ipsilateral hemithyroidectomy, and pulmonary metastasectomy. All of the three surviving patients have biochemical evidence of residual parathyroid carcinoma (Table 2).
|
|
Follow-Up
Follow-up ranged from 1 to 279 months (23 years), with a median of 134 months. Of these 23 patients, only 3 died secondary to known complications of parathyroid cancer. The 5- and 10-year survival probabilities were 85.9% and 69.4%, respectively. The 5- and 10-year recurrence-free survival probability was 76.0% and 59.2%, respectively. When survival was analyzed for the performance of ipsilateral hemithyroidectomy at the initial resection, 5- and 10-year survivals were 90.0% and 67.5%, respectively. Patients who did not undergo this procedure as part of their initial resection had no significant change in survival at 5 and 10 years (82.5% and 70.7%, respectively; P = .61). The 5- and 10-year survival of patients who had preoperative or intraoperative findings suggestive of carcinoma at initial resection was 88.9% and 69.3%, respectively. Patients who received a diagnosis after surgery on pathologic criteria alone had a 75% 5- and 10-year survival (P = .50). A single patient developed pulmonary metastases nearly 20 years after initial diagnosis. She underwent pulmonary metastasectomy in an attempt at palliation. Cox proportional hazards regression failed to show that age, sex, size of tumor, preoperative calcium level, degree of differentiation by histological criteria, or type of operation were independent predictors of survival. Four patients had the diagnosis made only by histological criteria. Of these four patients, one expired because of presumed asphyxia in his home after 145 months’ (12 years’) survival. The remaining three are alive with survivals of 3, 11, and 19 years. The 5-year survival of this group was 89%, compared with 75% for patients with gross characteristics or biochemical suspicion of carcinoma (P = .50). Corresponding 10-year survivals were 75% and 68.3%.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
The histological diagnosis of parathyroid carcinoma can be difficult. The differences between atypical parathyroid adenoma and carcinoma are not always readily obvious. Prognostic markers such as cyclin D1 overexpression, DNA ploidy, Ki-67, and p27 all reveal overlapping results in separating atypical adenoma and carcinoma. The only definitive findings that may be used solely as diagnostic criteria include regional or distant metastases, local invasion into adjacent neck structures, and vascular or lymphatic invasion.14,15 Lloyd’s textbook of endocrine pathology15 suggests that in the absence of invasion or distant metastases, a diagnosis of parathyroid adenoma with suspicious features or borderline parathyroid tumor should be made.
Our series of 23 patients with carcinoma recorded by our institutional tumor registry included a small number of cases that met criteria to be classified as borderline parathyroid tumor or had atypical biochemical parameters. Two patients (9%) had multiple criteria suggestive of carcinoma, such as capsular invasion and broad fibrous bands, but had no definitive lymphatic, vascular, or soft tissue invasion. However, the surgeon’s operative dictation in both patients noted gross tissue findings of hard, firm masses consistent with well-known findings of parathyroid carcinoma. Both of these patients are alive with survivals of 8 and 18 years. Three patients had biochemical findings that did not support carcinoma, such as eucalcemia (8.7 mg/dL) in one and only minimally increased PTH in the other. One other patient had a tumor that weighed only 80 mg, a surprisingly low weight for most series. The eucalcemia patient was diagnosed by electron microscopy, an unusual means of making the diagnosis. The biochemical characteristics expected and noted in most large series differ in that the usual presentation includes extremely increased serum calcium and a PTH value 5 to 10 times above the reference range. Two other patients had the diagnosis made after a reoperative surgical procedure for original disease that was confirmed to be benign. The diagnosis of parathyroid carcinoma in the setting of recurrent benign HPT is extremely unusual. Review of the original pathology confirmed the benign nature of the original tumors. No features suggesting cystic nature of the adenoma were noted during surgery or histologically. Capsular fraction leading to local recurrence is possible and could have resulted in histological characteristics of scarring and fibrosis, which would mimic carcinoma. Capsular fraction leading to local recurrence is possible and could have resulted in histological characteristics of scarring and fibrosis, which would mimic carcinoma.
In practice, these seven patients (30%) pose the probability of atypical or borderline parathyroid tumor rather than carcinoma, because all findings do not support the diagnosis of malignancy. Further evaluation of these seven patients revealed that their survival was not significantly different from that of patients who met all criteria and had a classic presentation of symptoms and biochemical parameters (5-year survival, 86.7% vs. 83.3%; P = .57). One of these patients, however, did die secondary to complications of parathyroid cancer. This patient underwent excision of what was thought to be an adenoma at a referring institution and developed a local recurrence. All other patients died of other causes and had eucalcemia documented at the time of death. Small numbers prohibit quantitative analysis.
Previous literature suggests that a patient’s greatest hope for cure remains in an adequate surgical resection at the initial operation.11 Specifically, a comprehensive en-bloc resection should include resection of the parathyroid mass and the ipsilateral thyroid lobe and excision of adherent strap musculature, the paratracheal fibrolymphatic tissue, and, if it is involved, the thymus. Only twelve (43%) patients eventually underwent a comprehensive en-bloc resection with resection of adjacent organs. The reluctance of the primary surgeon to resect the adjacent thyroid lobe or thymus is not fully understood. One possible explanation may be that the surgeon believed that an adequate resection was performed with a less radical procedure. Despite this, these patients’ survival did not seem to vary from that of patients who underwent more radical procedures. Only one (14%) of the patients who received noncomprehensive resection died of recurrence (3 years after resection). All others remain disease free or had documented eucalcemia at the time of death.
The classic intraoperative diagnosis of parathyroid carcinoma is suggested by its gross characteristics as observed by the surgeon. A white, hard mass is more suggestive of carcinoma than the peanut-butter color of a soft adenoma.16 Involvement and adherence of structures such as the trachea, esophagus, carotid sheath, and recurrent laryngeal nerves with a mass also suggest carcinoma but are not purely diagnostic.16 Despite these classic characteristics, four patients in this series had a diagnosis made by postoperative pathologic criteria only. The fact that these patients had a phenotype of the disease that did not cause significant local invasion supports the finding that there is a broad spectrum of presentation and aggressiveness. Although the small number of patients precludes quantitative analyses, it should be realized that gross characteristics might be an important part of classification in a universal staging system.
These survival data are comparable to other published data.8 Overall survival in this series was 85.9% and 69.4% at 5 and 10 years. This wide range may represent a broad spectrum of disease from borderline atypical adenoma to aggressive carcinoma or may be due to our small sample size. Another explanation might be better medical management of hypercalcemia during the past two decades compared with older published series. Although this is a rare disease, a standardized classification might better serve clinicians and patients to predict outcomes, allow better communication and prognostication, and guide therapy or research.
|
CONCLUSIONS |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
|
ACKNOWLEDGMENTS |
|---|
Received for publication March 18, 2004. Accepted for publication October 21, 2004.
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONPATIENTS AND METHODSRESULTSDISCUSSIONCONCLUSIONSREFERENCES |
|---|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
