Langenbecks Arch Surg (2005) 390: 385–390 DOI 10.1007/s00423-005-0555-6

Maurizio Iacobone Cesare Ruffolo Franco Lumachi Gennaro Favia

Received: 6 November 2004 Accepted: 3 March 2005 Published online: 3 June 2005 # Springer-Verlag 2005

This paper was presented at the First Congress of the European Society of Endocrine Surgeons (ESES), Pisa, Italy, 13–15 May 2004 M. Iacobone (*) . C. Ruffolo . F. Lumachi . G. Favia Endocrine Surgery Unit, Department of Surgical and Gastroenterological Sciences, University of Padua, Via Giustiniani, 2, 35128 Padua, Italy e-mail: [email protected] Tel.: +39-049-8212388 Fax: +39-049-8211378

ORIGINA L ARTI CLE

Results of iterative surgery for persistent and recurrent parathyroid carcinoma

Abstract Background and aims: Parathyroid carcinoma (PC) is a rare cause of primary hyperparathyroidism. Surgery is the only effective treatment; re-operations are often required, because recurrences occur in most of the cases. The aim of this retrospective study was to analyse the rate of biochemical cure, clinical relief, sensitivity of localizing studies and morbidity after re-operations. Patients and methods: From January 1980 to December 2000, 19 patients underwent surgery for PC. PC persisted or recurred in all cases. Fourteen re-operations were performed in six patients. Results: Twelve reoperations at loco-regional site and two pulmonary metastasectomies were performed. Iterative surgery

Introduction Parathyroid carcinoma (PC) is an extremely rare neoplasm, accounting for 1–5% of cases of primary hyperparathyroidism (HPT) [1–4]. PC is usually an indolent tumour with tendency to local invasion and, lately, to distant metastases, more often to lungs (40%), liver and bone [5–7]. In most cases, PC is a functioning neoplasm; thus, hypercalcemia and subsequent metabolic complications, rather than tumour invasiveness, are the main causes of death [7–10]. Surgery is the only effective treatment for PC, but, in spite of apparent radical resection of the primary lesion at the time of initial operation, recurrence may occur in most of cases [1, 4, 10–12]. Radiation therapy and chemotherapy are ineffective [6]; subsequently, re-operations are often required. Studies on the results of re-operations for persistent and recurrent PC are scarce, because of its rarity

achieved a symptomatic relief in 86% of cases and a transient biochemical remission only in one patient, but significantly reduced parathormone and calcemia. The sensitivity of scintigraphy, CT and ultrasonography was 86, 79 and 100%, respectively. Conclusion: When recurrences occur, complete cure of PC is unlikely, despite re-operations. Iterative surgery is associated with some morbidity and never achieves a definitive cure, but provides significant clinical and biochemical palliation. Localizing studies are mandatory but do not detect all recurrences. Keywords Recurrent and persistent hyperparathyroidism . Parathyroid carcinoma . Re-operations

[1, 3, 13]. Usually, these data derive from scattered case reports and wider collective reviews; unfortunately, in these cases diagnostic criteria are not uniform; clinical and follow-up data are incomplete. The aim of this retrospective study was the analysis of the long-term results of iterative surgery for persistent and recurrent PC in a single university referral centre, in order to assess the rate of biochemical cure, clinical relief, accuracy of localizing studies, survival and morbidity.

Materials and methods From January 1980 to December 2000, PC was diagnosed in 19 patients (12 men and seven women, median age 65 years, range 30–78 years). It accounted for 4.7% of observed cases of primary HPT.

386

The diagnosis of PC, suggested by intraoperative findings such as invasion of the surrounding structures, was assessed according to the Schantz and Castleman criteria (i.e. the presence of a fibrous capsule or dense fibrous trabeculae, uniform trabecular or rosette-like cellular architecture, vascular or capsular invasion, and presence of mitotic figures within tumour cells) [14] and further confirmed by the following clinical course (presence of local recurrences and distant metastases). Medical records of in- and out-patients were reviewed for data extraction and for gathering information about clinical follow-up. Prevalence of signs and symptoms before each operation and after 3 months was scheduled. The following clinical features were assessed: renal symptoms (nephrolitiasis, nephrocalcinosis, impaired glomerular filtration, renal colic), skeletal symptoms (bone pain, osteoporosis, fractures, radiographic signs), cardiovascular symptoms (hypertension, arrhythmias) and general symptoms (fatigue, weakness, lethargy, depression, nausea, vomiting, dyspepsia). Pre- and post-operative serum calcium and parathormone (PTH) levels were scheduled. The relative efficacy of each surgical procedure in reducing serum calcium and PTH levels was also calculated as rate of decrease between pre- and post-operative levels. Vocal cord motility was assessed before and after surgery by indirect laringoscopy. Before re-operations, ultrasonography (US), computerized tomography (CT) and/or magnetic resonance (MRI), and SestaMIBI or Tl/Tc scintigraphy (SC) were used to localize the site of recurrence. Sensitivity of localizing techniques (defined as True Positive/True Positive+False Negative) and positive predictive value (PPV) (defined as True Positive/True Positive+False Positive) were determined by comparing the results of the study with histopathological findings of the resected tissues. Specificity and negative predictive value were not calculated because there were no true negative results since all the patients had recurrent or persistent PC. The reported data are expressed as mean±standard deviation (SD). Statistical analysis was performed by the mean of Student’s t test, chi-square test, Fisher’s exact test, Mann–Whitney U test and McNemar test, according to the type of variables. Pearson’s correlation coefficient was used to evaluate the linear relationship between pairs of variables. Survival was calculated according to the method of Kaplan and Meier, and significance was tested by the log-rank test. A value of p<0.05 was considered to be statistically significant.

Results PC was histopathologically confirmed in 19 cases by the presence of vascular or capsular invasion and by the following clinical course. All patients presented local recurrences; distant metastases occurred in 16 cases.

Thirteen patients underwent a single surgical procedure, while six cases (five men and one woman; mean age 58±11 years) underwent repeated surgery. Tumour size in patients who underwent a single operation was 30.8±13 mm vs 30±11 mm in patients who underwent iterative surgery (p=0.81). Fourteen re-operations were performed: five re-operations in one case, three in another one, two in two patients and one re-operation in two patients. In re-operated patients, initial surgery consisted of “en bloc” parathyroidectomy, ipsilateral hemithyroidectomy and thymectomy in four cases, while a wide local excision was performed in two patients with a mediastinal PC. Iterative surgery consisted of excision of the tumour and possibly of the surrounding tissues. It was performed at a loco-regional site for recurrences in 12 cases (ten at neck and two at mediastinum), while two patients underwent multiple pulmonary metastasectomies. Local cervical recurrences in soft tissues were found in all patients who underwent re-operations. Mediastinal recurrences were found in the upper mediastinum and were excised by the mean of mid-sternothomic approach; nodal metastases were found in one case, at re-operation, deeply located at mediastinal level, inferiorly to the anonymous vein. There were no intra-operative deaths. None of the patients was treated by chemotherapy. One patient underwent radiation therapy, but he required a re-operation for local recurrence 4 months later. Although apparent capsular rupture of the tumour never occurred at initial surgery, PC persisted or recurred in all cases. Repeated surgery never achieved a definitive cure; a temporary biochemical remission occurred only in one case, with a new recurrence 5 months later. Table 1 reports the results of serum calcium and PTH levels before and after each surgical procedure in the six patients who underwent repeated surgery. The first and second re-operations significantly reduced serum calcium and PTH levels in all patients (p<0.05). The reduction of calcemia and PTH levels became not statistically significant after the second re-operation. Concerning the efficacy of each re-operation in reducing calcium and PTH levels, the first re-operation was less effective in reducing the serum calcium levels than the initial surgery (serum calcium reduction: 22.4±6 vs 36± 5.9%; p=0.005), whereas no differences were found for PTH levels (80.5±6 vs 79.7±13.5%, respectively; p=0.91). The second re-operations were progressively less effective in reducing both calcemia (13.7±3%; p=0.03) and PTH levels (69±7%; p=0.04), whereas no difference was found between further operations. All patients had symptomatic HPT before re-operations. General, skeletal, renal and cardiovascular symptoms were present in 100% (n=14), 93% (n=13), 86% (n=12) and 86% (n=12) of cases, respectively, before 14 re-operations. Postoperatively, as assessed 3 months after each re-operation, the predominance of general, skeletal, renal and cardio-

387

Table 1 Pre- and post-operative serum calcium and parathormone (PTH) levels (mean±SD) in six patients undergoing repeated surgery for parathyroid carcinoma

Post-operative values are referred to immediate post-operative period Because of the various methods used for PTH measurements during the last 20 years, values have been converted considering 65 ng/l the upper limit of normal a PTH available in five patients

Preoperative Postoperative P value Preoperative Postoperative P value calcium levels calcium levels PTH levels PTH levels (ng/l) (ng/l) (mmol/l) (mmol/l) Initial surgery (n=6) First reoperation (n=6) Second reoperation (n=4) Third reoperation (n=2) Fourth reoperation (n=1) Fifth reoperation (n=1)

3.82±21

2.39±0.15

0.0001

408±257a

48±7a

0.047

3.47±0.34

2.68±0.18

0.011

492±96

95±30

0.005

3.6±0.42

3.1±0.36

0.006

620±155

189±35

0.033

3.4

2.95±0.07

0.07

602±493

199±144

0.33

3.4

3

–

350

120

–

3.6

3.2

–

423

310

–

vascular symptoms was 14% (n=2), 86% (n=12), 79% (n=11) and 79% (n=11), respectively. The clinical improvement was significant only for general symptoms (especially fatigue, weakness, nausea and dyspepsia) (p<0.001). Only one patient showed no symptomatic relief, after the fourth and fifth surgical re-explorations. All the patients underwent at least two localization procedures before iterative surgery (Table 2); no statistically significant differences were found for sensitivity and PPV between the different techniques. The concordance rates between US and SC (n=11), US and CT or MRI (n=11), SC and CT or MRI (n=14) were 100, 73 and 64%, respectively. No false negative results for neck recurrences occurred for US, but four false positive results were recorded (enlarged lymph nodes without metastases), since US showed more than one suspect localization. CT and MRI failed to localize three cases of cervical recurrences (correctly diagnosed by both US and SC), but there were not false negative results for mediastinal and pulmonary localizations. SC failed to correctly detect in the same patient a primitive mediastinal localization, local recurrences and pulmonary metastases. According to the time of execution of the imaging techniques, all the false negative results were observed when

localizing studies were performed early after biochemical recurrence (less than 3 months); thus, the sensitivity was 100% when imaging was performed later, and the number of false positive results was not increased. In the group of patients treated by repeated surgery, the overall mortality rate was 100% (median survival 48 months, range 21–146); in the group treated by a single surgical procedure, the median overall survival was 24 months (range 20–153) and four patients were still alive. No correlations were found between the number of reoperations and survival (r=0.61; p=0.19). The disease-free interval represented the main prognostic factor, since it was significantly correlated with the overall survival (r=0.768, p=0.039 in patients after repeated surgery; r=0.805, p= 0.001 after a single operation). In the group of re-operated patients, the mean biochemical disease-free interval after initial surgery was 15 months (range 2–74); it was less than 6 months after initial surgery in five cases, while only in one case a recurrence occurred later, after 72 months. In patients who underwent a single operation, the mean disease-free interval was 18 months; persistent disease after initial surgery occurred in six cases, while in the remaining seven patients the mean disease-free interval was 33 months (range 12–74 months). Although

Table 2 Results of the localizations studies before 14 re-operations for persistent and recurrent parathyroid carcinoma Overall sensitivity (%) Overall PPV (%) Sensitivity, early (%) Sensitivity, late (%) PPV, early (%) PPV, late (%) Ultrasonography (n=11) CT and MRI (n=14) Scintigraphy (n=14)

100.0

73.3

100.0a

100b

60a

80b

78.6 85.7

68.7 100.0

50.0c 66.6c

100b 100b

50c 100c

80b 100b

Sensitivity and positive predictive value (PPV) are also reported according to the time of execution of the study after the biochemical relapse early: Study performed less than 3 months after biochemical relapse; late: study performed more than 3 months after biochemical relapse a Referred to three cases b Referred to eight cases c Referred to six cases

388

the 5-year survival rate was 33% for re-operated patients and 8% in patients who underwent a single operation, the difference for overall survival between the two groups was not significant (p=0.109). On the contrary, when we compared only the patients of the two groups with persistent and early recurrent HPT after initial surgery (i.e. patients with a biochemical disease-free interval less than 6 months), repeated surgery significantly prolonged the survival after PC relapse (41 vs 24 months; p<0.05). The overall morbidity consisted of three unilateral definitive vocal cord paralyses at initial surgery following the intentional resection of the nerve. After iterative surgery, complications consisted of five transient, six permanent recurrential palsies (one bilateral) and two cases of bleeding, requiring re-exploration (p<0.01).

Discussion PC is usually considered a slow-growing neoplasm. Prognosis is quite variable because of the various biological aggressiveness of the tumour and the wide range of subsequent hypercalcemia. Various histological and biological markers have been proposed as diagnostic and prognostic factors; recently, HPRT2 mutations have been identified as relevant contributor factors to PC tumorogenesis. HPRT2 is a suppressor gene encoding a novel nuclear protein named parafibromin. Mutations may influence the phenotype of PC, including its metastatic potential and, subsequently, prognostic outcome [15]. Primary extensive surgery of parathyroid carcinoma is the treatment of choice, because radical initial surgery is the main prognostic factor [7]. Unfortunately, instead of apparent radical treatment, recurrences may occur in most of cases, since malignancy is confirmed [6]. Histology alone is unable to confirm the diagnosis of cancer [2]. Some authors have suggested that only recurrences and metastases may assess the definitive diagnosis of malignancy [9, 11, 16]. In our experience, PC persisted or recurred in all cases, although apparent capsular ruptures of the tumour never occurred at initial surgery. Metastases occurred in 84% of cases. The main cause of death in these patients is usually represented by the metabolic complications of hypercalcemia rather than local tumour invasion or distant metastases, because of the rather indolent biology of the tumour [7, 8, 10]. For these reasons, repeated surgery is often required to keep hypercalcemia under control, and possibly improve the prognosis [6, 7, 17–19]. In the present study, repeated surgery never provided a definitive cure, although distant metastases were not evaluable by imaging studies before each re-operation, except in patients who underwent multiple pulmonary metasta-

sectomies. A biochemical transient remission was achieved only in one patient, followed by early recurrence, 5 months later. At any rate, re-operations significantly reduced serum calcium and PTH levels. Furthermore, the best results in terms of biochemical palliation were achieved by the first two re-operations. Further re-explorations were progressively less effective, and the decrease in serum calcium levels and PTH became not statistically significant, possibly because of the presence of occult distant metastases. According to literature [13, 20], this study confirmed that iterative surgery also provides significant clinical palliation, although symptoms are difficult to assess in retrospective studies. Before surgery, all patients had two or more major clinical features; general and skeletal symptoms occurred most often. The symptomatic relief was not evident for renal and osteo-articular involvement, possibly because clinical records were assessed early (3 months after each re-operation), while a significant regression of general symptoms was observed in most of cases, possibly resulting in an objective improvement of the quality of life [21]. Before re-operations, localization studies should be performed, although recurrences most often occur at the original site [6, 13]. It is difficult to evaluate the efficacy of localization studies because they have greatly changed over the last 20 years. US can be considered excellent for localizing cervical PC, while CT and MRI are more sensitive for mediastinal sites and for pulmonary metastases [13]. In this series, US never showed false negative results for neck recurrences, and CT never failed to identify mediastinal and lung metastases (sensitivity 100%). Unfortunately, when we considered the overall sensitivity of CT and MRI, the rate dropped to 78.6%, because three cervical recurrences were not localized. The specificity of US, CT and MRI is limited, because they are not able to differentiate between recurrence and non-neoplastic masses (lymph nodes, scar tissue); in the present series the PPV for US and CT was 73.3 and 68.7%, respectively. Scintigraphy is considered very specific for parathyroid tissue and sensitive for ectopic localizations and distant metastases [13, 20]. In our series it had a PPVof 100% since no false positive results were recorded, but the overall sensitivity was 85.7%. For these reasons, at least two localization studies should be performed before embarking on re-operative surgery, in order to avoid negative re-explorations. The sensitivity of localizing techniques is also related to the time of execution of the study. Studies performed too early after biochemical relapse could fail to detect all the sites of recurrences, because even very small tumour deposits may produce significant PTH levels to cause hypercalcemia. This was confirmed in our experience; when imaging was performed less than 3 months since biochemical HPT recurred, the sensitivity of CT and scintigraphy was lower. The sensitivity raised to 100% (i.e. no false negative results) when localization studies and subsequent surgical

389

exploration were performed later, without increased number of false positive results. Recently, positron emission tomography (PET) scan was proposed for the localization of the site of recurrence. The technique seems to be more sensitive than other imaging studies; however, only few cases have been reported, because of the rarity of the disease [22]. The present experience also confirmed that iterative surgery for recurrent HPT is associated with increased morbidity [13], and the rate of complications should be kept in mind before planning repeated and extensive resections of loco-regional recurrences. However, surgery is the sole effective therapy, because chemotherapy and radiation therapy are ineffective [6], although rare and scattered cases of apparent cure are reported [10, 23]. We never used chemotherapy; one patient underwent radiation therapy that turned out to be futile since an early recurrence was observed. In recent years, novel promising medical treatments have been developed in adjunct or alternative to iterative surgery. Bisphosphonates, a group of drugs inhibiting the osteoclast-mediated bone resorption, have been proposed to lower serum calcium levels [6], but the effects are exclusively transitory. Calcimimetic agents are low molecular weight compounds acting as potent allosteric activators of the calcium receptor. They can be effectively used to lower PTH production and serum calcium concentrations in patients with parathyroid carcinoma [24], although systemic side effects are often observed. Immunization against PTH has also been investigated. It consists of administration of human and bovine PTH peptides causing production of blocking antibodies against native PTH, followed by a decrease in the serum calcium levels [25]. At the moment, only few cases have provided significant results; thus, further studies are needed. However, despite the promising results of these complementary medical treatments, the role of repeated surgery in biochemical and symptomatic palliation is undisputed, while only a limited amount of data is available on the specific effects on long-term survival [1, 7, 16, 17, 26]. In

literature, the overall 5-year relative survival rates vary from 40 to 86% [6, 7, 27]. The wide range is possibly related to the inclusion of some “histologically equivocal” cases, with better survival, possibly not true carcinoma, because different criteria of inclusion are used for diagnosis of PC [2, 3, 9, 11, 14]. In the present series, at a mean follow-up of 5 years, all the patients were dead. Iterative surgery did not prolong the overall survival in all patients, but it should be considered that most of our cases (83%) underwent re-operations because of persistent and early recurrent HPT (i.e. diseasefree interval after initial surgery shorter than 6 months). As previously reported [5, 28] and confirmed in the present series, persistence and early recurrences are predictive of worsening prognosis; in fact, disease-free interval is an important prognostic factor, since it is correlated with survival. Thus, although the small number of cases limits the value of our report, iterative surgery could prolong the survival after HPT relapse in patients with persistent and early recurrent PC, who usually have the worst prognosis and, subsequently, should be selected for iterative surgery.

Conclusion PC is a disease with a high rate of recurrence. Although repeated surgery never led to a definitive cure, it achieved significant biochemical and clinical palliation—although the effects of re-operations became progressively less effective. In order to avoid negative re-explorations, localization studies should be performed. US was sensitive for cervical recurrences, while CT and MRI were more sensitive for mediastinal sites and for pulmonary metastases. Only scintigraphy gave no false positive results. Thus the execution of more than one localization method should be suggested. After recurrence, too early localizing investigations should be avoided, because they could fail to localize all the sites of recurrence. Iterative surgery was associated with some morbidity, but might prolong survival in patients with persistent and early recurrent PC.

References 1. Obara T, Okamoto T, Ho Y, Yamashita T, Kawano M, Nishi T, Tani M, Sato K, Demura H, Fujimoto Y (1993) Surgical and medical management of patients with pulmonary metastasis from parathyroid carcinoma. Surgery 114:1040– 1048 2. Fujimoto Y, Obara T (1987) How to recognize and treat parathyroid carcinoma. Surg Clin North Am 67:343–357 3. Obara T, Fujimoto Y (1991) Diagnosis and treatment of patients with parathyroid carcinoma: an update and review. World J Surg 15:738–744

4. Obara T, Okamoto T, Kanbe M, Iihara M (1997) Functioning parathyroid carcinoma: clinicopathologic features and rational treatment. Sem Surg Oncol 13:134–141 5. Akerstrom G, Juhlin C, Johansson H (1994) Parathyroid carcinoma. In: Akerstrom G (ed) Current controversies in parathyroid operations and reoperations. Landes Company, Austin TX, pp 201–218 6. Shane E (2001) Parathyroid carcinoma. J Clin Endocrinol Metab 86:485–493

7. Sandelin K, Auer G, Bondeson L, Grimelius L, Farnebo LO (1992) Prognostic factors in parathyroid cancer: a review of 95 cases. World J Surg 16:724–731 8. Fraker DL (2000) Update on the management of parathyroid tumors. Curr Opin Oncol 12:41–44 9. Vetto JT, Brennan MF, Woodruff J, Burt M (1993) Parathyroid carcinoma: diagnosis and clinical history. Surgery 114:882–892

390

10. Wynne AG, van Heerden J, Carney JA, Fitzpatrick LA (1992) Parathyroid carcinoma: clinical and pathologic features in 43 patients. Medicine 71:197–205 11. Levin K, Galante M, Clark OH (1987) Parathyroid carcinoma versus parathyroid adenoma in patients with profound hypercalcemia. Surgery 101:647–660 12. Wang C, Gaz RD (1985) Natural history of parathyroid carcinoma. Am J Surg 149:522–527 13. Kebebew E, Arici C, Duh QY, Clark OH (2001) Localization and reoperation results for persistent and recurrent parathyroid carcinoma. Arch Surg 136:878–885 14. Schantz A, Castleman B (1973) Parathyroid carcinoma: a study of 70 cases. Cancer 31:600–605 15. Shattuck TM, Valimaki S, Obara T, Gaz RD, Clark OH, Shoback D, Wierman ME, Tojo K, Robbins CM, Carpten JD, Farnebo LO, Larsson C, Arnold A (2003) Somatic and germ-line mutations of the HRPT2 gene in sporadic parathyroid carcinoma. N Engl J Med 349:1722–1729

16. Koea JB, Shaw JH (1999) Parathyroid cancer: biology and management. Surg Oncol 8:155–165 17. Dotzenrath C, Goretzki PE, Sarbia M, Cupisti K, Feldkamp J, Roher HD (2001) Parathyroid carcinoma: problems in diagnosis and the need for radical surgery even in recurrent disease. Eur J Surg Oncol 27:383–389 18. Shortell C, Andrus CH, Phillips CE, Schwatz S (1991) Carcinoma of the parathyroid glands: a 30 year experience. Surgery 110:704–708 19. Flye MW, Brennan MF (1981) Surgical resection of metastatic parathyroid carcinoma. Ann Surg 193:425–435 20. Fujimoto Y, Obara T, Ito Y, Kodama T, Nobori M, Ebihara S (1986) Localization and surgical resection of metastatic parathyroid carcinoma. World J Surg 10:539–547 21. Quiros RM, Alef MJ, Wilhelm SM, Djuricin G, Loviscek K, Prinz RA (2003) Health-related quality of life in hyperparathyroidism measurably improves after parathyroidectomy. Surgery 134:675–681 22. Arslan N, Rydzewski B (2002) Detection of a recurrent parathyroid carcinoma with FDG positron emission tomography. Clin Nucl Med 27:221– 222 23. Rasmuson T, Kristoffersson A, Boquist L (2000) Positive effect of radiotherapy and surgery on hormonally active pulmonary metastases of primary parathyroid carcinoma. Eur J Endocrinol 143:749–754

24. Collins MT, Skarulis MC, Bilezikian JP, Silverberg SJ, Spiegel AM, Marx SJ (1998) Treatment of hypercalcemia secondary to parathyroid carcinoma with a novel calcimimetic agent. J Clin Endocrinol Metab 83:1083–1088 25. Bradwell AR, Harvey TC (1999) Control of hypercalcaemia of parathyroid carcinoma by immunization. Lancet 353:370–373 26. Sandelin K, Thompson NW, Bondeson L (1991) Metastatic parathyroid carcinoma: dilemmas in management. Surgery 110:978–988 27. Hundahl SA, Fleming ID, Fremgen AM, Menck HR (1999) Two hundred eighty-six cases of parathyroid carcinoma treated in the U.S. between 1985–1995: a National Cancer Data Base Report. Cancer 86:538–544 28. Iacobone M, Lumachi F, Favia G (2004) Up-to-date on parathyroid carcinoma: analysis of an experience of 19 cases. J Surg Oncol 88:223–228