2023 Volume 70 Issue 10 Pages 969-976
The operative procedure in the surgical treatment of parathyroid carcinoma differs from that of benign hyperparathyroidism. However, preoperative differentiation is often difficult. This study elucidated how clinicians diagnose parathyroid carcinoma and the relationship between preoperative diagnosis and the operative course. Using a retrospective chart review, twenty cases of parathyroid carcinoma from nine participating centers were examined. In 11 cases with preoperative suspicion of malignancy, at least one of these three features was found: elevated serum calcium level (>14 mg/dL), palpable mass, and irregular margin on ultrasonography. Although an intact parathyroid hormone (iPTH) threshold to suspect malignancy has not been established, six cases showed marked iPTH elevation exceeding 8.0 times the upper limit of normal. One case was excluded from analysis due to hemodialysis. Compared with the four cases that showed calcium elevation, the iPTH threshold might represent better sensitivity. Among 9 cases of benign preoperative diagnosis, six cases were performed with pericapsular resection. In three cases where malignancy was suspected in the middle of the operation, the recommended en bloc resection with ipsilateral thyroid lobectomy was not performed but a parathyroidectomy with surrounding soft tissue. In contrast, 10 preoperatively suspected cases underwent en bloc resection, and one case underwent pericapsular resection followed by supplementary ipsilateral hemithyroidectomy due to the uncertain pre- and intraoperative findings to determine the diagnosis. In conclusion, the surgical procedure for parathyroid carcinoma strongly depends on the preoperative diagnosis. The presence of excessive iPTH levels might contribute to improved preoperative diagnostic sensitivity for parathyroid carcinoma.
PARATHYROID CARCINOMA is a rare disease, with an incidence ranging from 1–5% in patients with primary hyperparathyroidism [1]. Similar to benign primary hyperparathyroidism (PHPT), parathyroid carcinoma often presents with symptoms of hypercalcemia and related dysfunctions. However, clinical differentiation of these tumors is often difficult. Although several clinical clues have been proposed that might help suspect malignancy, no standardized diagnostic protocol has been established. Fine needle aspiration cytology (FNAC) is not routinely performed for diagnosis due to the risk of carcinoma spread on the needle tract [2, 3]. Additionally, distinguishing between benign and malignant parathyroid tumors using cytological diagnosis can be challenging, and differentiating a parathyroid lesion from a thyroid lesion can be difficult [4]. Thus, a surgeon has to plan the initial surgery without a definite diagnosis.
The existence of carcinoma should be examined midway through the operation, because the recommended excision area differs by the diagnosis. Pericapsular excision is commonly performed to treat PHPT. In cases of parathyroid carcinoma, complete resection of the primary lesion is essential at the time of the initial operation [1]. The agreed complete resection method is en bloc tumor resection [5], which comprises tumor resection, ipsilateral thyroid lobectomy, and adjacent cervical muscle resection. If the carcinoma involves the recurrent laryngeal nerve, muscular layer of the esophagus, or tracheal cartilage, the involved segments should also be excised. However, intraoperative diagnosis is based on surgeon’s findings such as invasive characteristics of the tumor [5], and intraoperative frozen section histology is of little value in diagnosing parathyroid carcinoma [1]. Although the recommended strategies for excision differ between benign and malignant tumors, surgeons have to perform proper procedures without the aid of preoperative cytology and intraoperative histopathology.
Here, we analyzed 20 cases of parathyroid carcinoma reported from multiple centers, focusing on preoperative diagnoses and surgical strategies based on them. This study aimed to clarify how a surgeon’s prediction of malignancy contributes to curative surgical treatment.
We performed a retrospective data review of 20 parathyroid carcinoma cases treated from 2012 to 2020 in nine participating medical institutes, associated with Kyoto University and its Affiliated Hospitals—Head and Neck Oncology Group (Kyoto–HNOG) in Japan. The inclusion criterion for this study was the presence of a histopathologically confirmed parathyroid carcinoma. However, we included a case in which the tumor recurred after excision of a parathyroid adenoma, leading to the diagnosis of parathyroid carcinoma. The protocol of this multi-site project was approved by the administrative single institutional review board and was agreed to by each participating institute (Institutional review board of Kyoto University; Approval ID_R3255). This observational research employed an opt-out style, allowing all subjects to withdraw from the study at any time.
In this study, the TNM classification was based on the 8th edition of the American Joint Committee on Cancer guidelines introduced in 2017. Cases treated before the guidelines were published were reviewed and reclassified.
Preoperative diagnosis was based on the clinical records of each patient. We divided the patient cohort by preoperative diagnosis to investigate how the diagnosis was made and how that affected the choice of surgical procedure.
Pericapsular excision is usually performed for a preoperative diagnosis of benign parathyroid adenoma. The recommended treatment for parathyroid carcinoma is en bloc excision of the tumor, ipsilateral thyroid lobectomy, and adjacent paratracheal tissue [5]. However, many carcinomas are excised during the diagnosis of benign lesions. Some surgeons underwent supplementary ipsilateral hemithyroidectomy after the histological diagnosis of carcinoma, while others did not perform hemithyroidectomy even though they suspected malignancy intraoperatively, resecting a tumor with a small amount of surrounding soft tissue such as adipose tissue and thyroid capsule. We defined this type of surgery as “extensive resection” in this article.
Data were analyzed using JMP pro 15.2 software (SPSS, Chicago, IL, USA). All analyses were performed on the entire cohort of 20 patients, with the exception of an analysis specifically evaluating the correlation between serum calcium and iPTH levels.
Patient demographics are summarized in Table 1. One patient initially diagnosed with a benign lesion was found to be malignant after a recurrent tumor appeared. This case was referred to a participating institute after the recurrent tumor was detected on imaging examinations, and classified as Tx due to the absence of records from the initial treatment. The remaining 19 patients were diagnosed during the initial treatment. Case #8 received hemodialysis, and case #18 was administered with a parathyroid hormone suppressor.
Patient demographic
| Gender | Male | 12 |
| Female | 8 | |
| Median Age (range) | 61.5 (23–76) | |
| Performance Status | 0 | 16 |
| 1 | 3 | |
| 2 | 1 | |
| 3 or 4 | 0 | |
| pTNM-T | Tx | 1 |
| T0 | 0 | |
| Tis | 1 | |
| T1 | 12 | |
| T2 | 3 | |
| T3 | 3 | |
| T4 | 0 | |
| pTNM-N | Nx | 4 |
| N0 | 16 | |
| N1a | 0 | |
| N1b | 0 | |
| pTNM-M | M0 | 20 |
| M1 | 0 | |
| Familial Historya | No | 15 |
| Unknown | 5 | |
| Congenital Diseaseb | No | 20 |
a. Familial history indicates the familial history of parathyroid disease, including benign tumor. b. Congenital disease includes multiple endocrine neoplasia (type 1 and 2a), hyperparathyroidism-jaw tumor syndrome, familial isolated hyperparathyroidism, and familial hypocalciuric hypercalcemia.
The overall survival and recurrence-free survival rates are displayed in Fig. 1. The median observation period was 1,284 days (28–3,677 days), and no deaths occurred during this period. Of the 20 cases, eight were censored before 3 years of observation, 11 before 5 years, and 19 before 10 years. The 3- and 5-year recurrence-free survival rates were 100% and 90.9%, respectively. Only one recurrent case was the Tx case described above. Apart from this case, none of the postoperative cases showed an increase in calcium or intact PTH levels, nor did they display structural lesions on cross-sectional imaging.
Overall survival and recurrence-free survival rates
We divided the cohort into two subgroups based on preoperative diagnosis: benign and possible malignancy. The preoperative diagnoses and clinical features are summarized in Table 2. The serum calcium and intact parathyroid hormone (iPTH) levels shown in the table are the maximum values obtained before the initial surgery.
Pre-Operative features and diagnoses
| Case | Pre-Op. Diagnosis | TNM | Diametera (mm) | Serum Cab (mg/dL) | iPTHb (pg/mL) | ALPb (U/L) | Palpable mass | Tumor appearance in ultrasonography | Radiographyc | Ca-related symptom | FNAC | ||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Irregular margin | Heterogeneous texture | Calcification | Osteolytic lesion | Subjective symptomd | Bone complicatione | Lithiasisf | |||||||||
| #1 | BENIGN | X | 23 | 16.3 | 1,145 | 284 | + | + | |||||||
| #2 | BENIGN | is | 20 | 12.9 | 321 | 333 | + | + | + | ||||||
| #3 | BENIGN | 1 | 20 | 8.8 | 17 | 356 | + | + | |||||||
| #4 | BENIGN | 1 | 36 | 10.8 | — | 435 | BENIGN | ||||||||
| #5 | BENIGN | 1 | 13 | 11.6 | 352 | 325 | |||||||||
| #6 | BENIGN | 1 | 16 | 12 | 95 | 213 | + | ||||||||
| #7 | BENIGN | 1 | 16 | 12 | 190 | 550 | + | + | |||||||
| #8 | BENIGN | 3 | 18 | 10 | 830 | 421 | + | ||||||||
| #9 | BENIGN | 3 | 28 | 12.6 | 333 | 242 | + | + | |||||||
| #10 | Possible MALIGNANT | 1 | 12 | 12 | 399 | 1,252 | + | + | |||||||
| #11 | Possible MALIGNANT | 1 | 24 | 13.1 | 437 | 508 | + | + | |||||||
| #12 | Possible MALIGNANT | 1 | 23 | 14.3 | 664 | 459 | + | + | + | + | |||||
| #13 | Possible MALIGNANT | 2 | 31 | 13.6 | 1,044.3 | 323 | + | + | BENIGN | ||||||
| #14 | Possible MALIGNANT | 2 | 21 | 14.3 | 579 | 560 | + | + | + | + | |||||
| #15 | Possible MALIGNANT | 1 | 30 | 11.5 | 708 | 623 | + | MALIGNANT | |||||||
| #16 | Possible MALIGNANT | 1 | 19 | 11.6 | 484 | 383 | + | + | + | + | |||||
| #17 | Possible MALIGNANT | 1 | 30 | 11.9 | 372 | 284 | + | + | UNDETERMINED SIGNIFICANCE | ||||||
| #18 | Possible MALIGNANT | 1 | 26 | 21.9 | 2,374 | 2,050 | + | + | + | ||||||
| #19 | Possible MALIGNANT | 2 | 22 | 11.4 | 75 | 1,149 | + | + | + | MALIGNANT | |||||
| #20 | Possible MALIGNANT | 3 | 16 | 12.5 | 345 | 477 | + | + | + | ||||||
a. Diameter indicates the maximum diameter of each tumor. b. Bold letters are used in excessively elevated serum calcium (>14 mg/dL), intact PTH (>520 pg/mL), and ALP (Alkaline phosphatase: >322 U/L). The 520 pg/mL of intact PTH is 8.0 times bigger than the upper limit of normal intact PTH range (=65 pg/mL). ALP was measured with the standard of Japan Society of Clinical Chemistry, which has the normal range of 106–322 U/L. c. Osterolytic finding in radiography indicates a bone resorption lesion known as brown tumor or osteitis fibrosa, which occurs in 1–2% of hyperparathyroidism. d. Ca-related subjective symptom includes headache, lethargy, confusion, depression, anorexia, weight loss, fatigue, muscle weakness, bone pain, polyuria, polydipsia, constipation, nausea and vomiting. e. Bone complication indicates a past or present history of osteoporosis, repeating bone fracture or osteitis fibrosa. f. Lithiasis consists of urolith, gallstone, pancreatolith or pancreatitis.
The aforementioned recurrent case was case #1, and the representative data were obtained before the initial surgery. In case #3, an anterior neck tumor was incidentally found, but without increased iPTH or serum calcium levels. Therefore, surgery was performed for diagnostic purposes, revealing nonfunctional parathyroid carcinoma. Case #4 was treated as a benign thyroid tumor with preoperative FNAC examination, which turned out to be a parathyroid carcinoma on postoperative histopathology. Case #8 underwent hemodialysis, and total parathyroidectomy was performed with a diagnosis of secondary hyperparathyroidism. A parathyroid hormone suppressor (cinacalcet hydrochloride) was administered only to case #18 before surgery.
Table 2 indicates that at least one of the following features was present in all preoperatively suspected carcinomas: elevated serum calcium levels (>14 mg/dL), palpable mass, and irregular margins on ultrasonography. Pearson’s chi-square test revealed a significant association between the preoperative diagnosis and each malignant feature. Notably, a palpable mass (p = 0.0146) and irregular margins on ultrasonography (p = 0.0006) exhibited the strongest correlations. However, some features, such as excessive iPTH levels, heterogeneous texture, and calcification on ultrasonography, were overlooked even though they are known characteristics of malignancy. FNAC showed the possibility of malignancy in two out of five cases. The FNAC result of case #19 suggested the existence of malignancy, but it was uncertain whether the tumor originated from the thyroid or parathyroid gland.
In this study, excessive elevation of iPTH was defined as eight times or higher than the upper normal limit (UNL) [6]. As the UNL of iPTH was 65 pg/mL in the participating institutes, an iPTH level ≥520 pg/mL was regarded as excessive elevation. A strong relationship was found between serum calcium and iPTH levels, as indicated by a Pearson correlation coefficient of 0.755 (0.431–0.9907; 95% confidence interval). It is important to note that this analysis excluded the hemodialysis case and the case where a parathyroid hormone suppressor was administered. However, some cases in this cohort showed modest calcium elevation compared to the excessive iPTH value, such as case #15. Alkaline phosphatase elevation, which reflects osteolysis due to hyperparathyroidism, was not routinely observed or related to bony complications in this cohort.
Choice of the surgical plan and postoperative treatmentFig. 2 demonstrates how preoperative diagnoses can influence surgical plans. En bloc resection was generally chosen in most cases with a preoperative diagnosis of possible malignancy. In contrast, pericapsular resection was chosen in patients with a benign preoperative diagnosis. However, in cases where malignancy was suspected intraoperatively (three cases), extensive excision involving tumor resection with surrounding soft tissue was performed. In these cases, suspicion of malignancy was based on the invasive nature of the tumors (three cases) and comparatively harder palpation (one case).
Diagnosis and treatment course
The schema shows the treatment flow with the details of the surgically excised area and diagnosis at the concerned time points. *The patient was treated as a benign thyroid tumor, but parathyroid carcinoma was incidentally noted on postoperative histopathology. This case is identical to case #4 in Table 2.
In six cases, pericapsular excision was performed during the initial operation, and two of those patients underwent supplementary ipsilateral hemithyroidectomy within 6 months after the initial surgery. No signs of tumor relapse or remaining tumor were detected, and no residual carcinoma was found in the supplementary excised specimens in these cases. In one preoperatively suspected case, pericapsular excision was performed because the surgeon was uncertain whether the tumor was infiltrating or if a nonspecific adhesion was present.
Radiation was performed in two cases when histological examination indicated a close margin of the tumor or possible microscopic tumor remnant around the recurrent laryngeal nerve.
Parathyroid carcinoma is a rare disease accounting for 0.005% of all malignancies [7]. Diagnosing parathyroid adenoma and early-stage parathyroid carcinoma is often difficult as they share similar clinical features resulting from excessive secretion of parathyroid hormone. Several criteria have been suggested to enable a preoperative suspicion of parathyroid carcinoma [1]: Excessively elevated calcium levels (typically >14 mg/dL), hypercalcemia-related symptoms, palpable mass, renal and/or skeletal involvement, in addition to the difference in gender distribution (a 1:1 male-to-female ratio for carcinoma and 1:3–4 for adenoma) and predominant age (carcinoma is frequent in the fifties, and adenoma is in fifties and sixties). An ultrasonography study indicated that parathyroid carcinoma is characterized by a higher incidence of heterogeneous echotexture, intra-nodular calcifications, and invasive or irregular mass [8]. In this study, preoperatively suspected cases shared at least one of these three features: elevated serum calcium (>14 mg/dL), palpable mass, and irregular margins on ultrasonography. However, the participating clinician did not focus on heterogeneous echotexture and intranodular calcification. Heterogeneous echotexture is observed in 78% of adenomas, making this characteristic a non-decisive criterion [8]. On the other hand, intra-nodular calcification appears in only 3% of adenomas [8], and may be a good hallmark of malignancy. Several recent studies have investigated the clinical utility of elastography in discriminating parathyroid carcinoma [9, 10]. Elastography is a non-invasive imaging technique that measures tissue stiffness based on its response to mechanical pressure. Since carcinomas are typically harder than benign adenomas, the measurement of stiffness using elastography holds promise as a diagnostic technique. As the radiologically found osteolytic lesions were observed in only one case, in addition to maximum serum calcium levels and other characteristics to suspect malignancy. This suggests that osteolytic lesions occur in extreme hypercalcemia cases, and their sensitivity in predicting malignancy is limited. Hypercalcemia-related symptoms were found in 70.0% (14/20) of the cases, but each symptom can also occur commonly in benign hyperparathyroidism. Their contribution is unclear to distinguish between benign and malignant cases.
Furthermore, iPTH elevation is reported to be more severe in carcinoma, and a 222-case review showed that the median iPTH value of carcinoma is 8.0 times higher than the UNL [6]. In our study, participating institutes applied an upper normal iPTH limit of 65 pg/mL, and the median iPTH value was 6.1 times the UNL (399 pg/mL). Considering that benign adenomas usually represent <5 times the UNL [11], excessively high iPTH values might be a clue to suspect malignant tumor characteristics. Although several reports have suggested different iPTH thresholds for distinguishing between benign and malignant parathyroid lesions [12, 13], the sample sizes in these reports might be insufficient to adequately represent the entire carcinoma population. Moreover, iPTH values could vary depending on the measurement method [14]. It might be more appropriate to determine the threshold based on the upper normal limit, rather than the absolute amount of iPTH. In the present study, seven cases exhibited iPTH values exceeding eight times the UNL. Even when excluding the hemodialyzed case, six cases still demonstrated elevated iPTH levels. Compared with >14 mg/dL serum calcium represented in four cases, suspecting a malignancy with ≥8 times the UNL of iPTH might enable the preoperative identification of more cases. Further validation of this threshold is eagerly awaited, particularly involving both benign and malignant parathyroid lesions, as its effectiveness has not yet been evaluated in benign cases.
Of note, few parathyroid carcinomas are functionally inactive and do not show calcium elevation. A review article calculated the prevalence of nonfunctioning carcinoma within reported parathyroid carcinoma cases to be 1.9% [15]. Our survey included a rare case (Case #3 in Table 2).
In our investigation, there was a clear link between the preoperative diagnosis and the surgical procedure performed. Although en bloc resection at first presentation contributes to local disease control and long term-survival [15], en bloc resection was not performed in any case with a preoperative diagnosis of benign PHPT. Even in intra-operatively suspected cases, only extensive excision was performed without hemithyroidectomy. In contrast, en bloc resection with ipsilateral thyroid lobectomy was performed in 90.9% (10/11) of cases when malignancy was suspected preoperatively. Several reasons could explain why en bloc resection is difficult without a preoperative diagnosis. First, some institutes performed focused parathyroidectomy for benign adenomas under local anesthesia, while general anesthesia was applied to perform en bloc resection in all institutes. It would be difficult to expand the excision area under local anesthesia, even if malignancy is suspected during surgery. Another reason is the difficulty of a definitive intraoperative pathological diagnosis. It is well known that frozen section histology contributes little to the diagnosis of parathyroid carcinoma [1]; consequently, surgeons must rely on their intraoperative findings. Koea et al. reported that invasion of the surrounding tissues was found in almost half of intra-operatively suspected carcinomas [15]. However, even an experienced surgeon has fewer chances to treat parathyroid carcinoma than adenoma. It would be a formidable challenge to diagnose a carcinoma with confidence and alter the surgical plan in the middle of the operation. Finally, some patients were not provided with sufficient information about how the surgical plan may change if a benign lesion was discovered to be malignant during surgery, including potential complications such as postoperative hypothyroidism from a hemithyroidectomy, despite being informed about the possibility of such a diagnosis. Thus, a preoperative suspicion of malignancy strongly affects the determination of the operative site.
Although supplementary ipsilateral hemithyroidectomies were conducted for two patients whose initial treatments were pericapsular excision, its efficacy in prognosis has not yet been established. Re-exploration of the neck is indicated when gross operative findings indicate a pathologically aggressive carcinoma or when the patient remains hypercalcemic [1]. In addition, two patients with suspected microscopic residual tumor received radiation therapy. Parathyroid carcinoma is usually considered radio-insensitive, and no standard radiation therapy has been established [16]; however, a recent report suggested the possible benefit of adjuvant radiotherapy [17].
Regarding survival outcomes, a study of 286 cases reported the 5- and 10-year survival rates of the disease as 85.5% and 49.1%, respectively [18]. Our current study showed 100% 3- and 5-year overall survival rates. The 3- and 5-year recurrence-free survival rates were 100% and 90.9%, respectively.
This study had some limitations. First, we recognize that the examined cohort was not followed up long enough to represent the prognosis of parathyroid carcinoma. The average time between surgery and first recurrence has been reported to be approximately 3 years, and a recurrence after 20 years of interval could occur [1]. A previous report even recommended lifetime follow-up for parathyroid carcinoma [19]. In this study, only one patient was followed up for longer than 10 years (3,677 days). Considering that many studies have applied the 10-year survival ratio to the prognosis of parathyroid carcinoma, this was a limitation. Second, each institution made the histopathological diagnoses individually in this study. It is widely recognized that the histopathological criteria for parathyroid carcinoma lacks full standardization, even among pathology professionals. However, we opted not to review every sample. The rationale behind this decision is that it would have minimal impact on the primary endpoints of this survey, which focused on the preoperative diagnosis of the clinician and the administered treatment. The clinical treatment decisions were based on the histopathological diagnosis available at the time of treatment. Third, review articles that summarized the outcomes of parathyroid carcinoma often regarded both extensive and en bloc resections as oncologic resection, comparing them with pericapsular resection to exhibit better local control [6, 15]. This might indicate that the advantage of authentic en bloc resection over extensive resection is not well established. Basceken et al. questioned the efficacy of en bloc resection compared with extensive resection [20].
In conclusion, the current study showed that the surgical procedure for parathyroid carcinoma strongly depends on the preoperative diagnosis. Recommended en bloc resection during the initial surgery was only performed when a preoperative suspicion of malignancy was made. Careful identification of the hallmarks of carcinoma will contribute to an accurate preoperative diagnosis. Our survey suggests that excessive iPTH elevation (>eight times the UNL) may be a threshold for suspecting parathyroid carcinoma.
We would like to thank Yoshiki Watanabe for data collection, and Editage (www.editage.com) for English language editing.
The authors declare that no funds, grants, or other support were received during the preparation of this manuscript.
None of the authors have any potential conflicts of interest associated with this research.
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by Yoshitaka Kawai, Yo Kishimoto, Hisanobu Tamaki, Takashi Fujiwara, Ryo Asato, Koji Ushiro, Shogo Shinohara, Shinpei Kada, Shinji Takebayashi, Tsuyoshi Kojima, Shuya Otsuki, Masakazu Miyazaki, Yohei Kumabe, and Koichi Omori. The first draft of the manuscript was written by Yoshitaka Kawai and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
The protocol of this multi-site project was approved by the administrative single institutional review board and was agreed to by each participating institute (Institutional review board of Kyoto University; Approval ID_R3255).
This observational research employed an opt-out style, allowing all subjects to withdraw from the study at any time.
