Results C

[C] Initial diagnosis and therapy of clinically apparent disease

Fine-needle aspiration biopsy (FNA) of thyroid nodules is one of the most useful, safe, and accurate tools in the diagnosis of thyroid pathology. Chang and colleagues (186) investigated the pitfalls in the diagnosis of MTC by FNA. Cytomorphology was reviewed in the FNA slides of 34 patients with proven MTC. Eighty-two percent of cases were diagnosed correctly as MTC by FNA, three cases were misdiagnosed as follicular neoplasm and one as desmoid, and two cases were suspicious for MTC. Thus, FNA would have indicated the need for surgery due to lack of benign findings in essentially all of these patients. Similarly, Papaparaskeva et al. (187) reported that FNA findings indicated the need for surgery in 99% of their MTC cases, and diagnosed MTC in 89%. They reported that the most important cytologic criteria of MTC with FNA were dispersed cell-pattern of polygonal or triangular cells, azurophilic cytoplasmic granules, and extremely eccentrically placed nuclei with coarse granular chromatin and amyloid. Bugalho et al. (188) reported the sensitivity of FNA for MTC as 63%, compared to a sensitivity of 98% for serum Ct. However, while only 9% of patients might have escaped surgery based on FNA results, attention to the central neck compartment may have been diminished in a greater number due to the lack of suspected MTC.

Elisei and colleagues (189) reported the results of Ct screening in 10,864 patients with thyroid nodular disease. The prevalence of MTC found by Ct screening was 0.40%. A positive Ct test had a higher diagnostic sensitivity and specificity compared with FNA. Ct screening allowed the diagnosis of MTC at an earlier stage compared to an unmatched control group diagnosed with MTC that did not undergo Ct screening. Normalization of serum Ct levels (undetectable) after surgery was more frequently observed in the Ct-screened group. At the end of follow-up, complete remission was observed in 59% of the Ct-screened group and in 2.7% of the control group (p = 0.0001).

Currently, the use of serum Ct to complement FNA in making the diagnosis of MTC remains controversial. It is recognized that basal (>20–100 pg/mL) and pentagastrin stimulated serum Ct levels (>100–500 pg/mL) may be used to identify patients with MTC, but with imperfect cut-off levels to optimize sensitivity and specificity (190,191). Karges et al. (3) recommended that to exclude MTC, serum Ct should be determined in patients with nodular thyroid disease. If basal serum Ct exceeds 10 pg/mL, Ct should be analyzed by pentagastrin stimulation testing, after renal insufficiency and proton pump inhibitor medication have been excluded as confounding contributors to the Ct level. Karges et al. (3) report that the risk for MTC is higher than 50% in patients with stimulated Ct values >100 pg/mL. Conversely, they reported that pentagastrin-stimulated Ct values <100 pg/mL are associated with a low risk of MTC, or very rarely, nonmetastasizing micro-MTC (size <10mm) (3). However, pentagastrin is not available for stimulation testing in the United States and many other countries. Costante et al. (190) reported that the positive-predictive value of basal Ct to diagnose MTC with values ≥ 20 and ≤50 pg/mL, ≥50 and ≤100 pg/mL, and ≥ 100 pg/mL were 8.3%, 25%, and 100%, respectively.

Given that MTC is present in about 0.3–1.4% of patients with thyroid nodules (189,191–199), routine serum Ct measurement in all patients with thyroid nodules has raised concerns of cost-effectiveness, especially when many of the operated patients would not have disease based on the imperfect specificity if a cut-off was chosen that optimized sensitivity. Additionally, the clinical significance and natural history of MTC diagnosed by Ct screening is unknown. However, cost-effectiveness studies in the United States have been favorable (200,201), although these models are quite sensitive to multiple important decision analysis variables, including cost of the Ct test, such that assuming a relatively small increase in cost could substantially alter the conclusions. Recent studies utilizing Ct measurement from thyroid nodule FNA washings may significantly improve this testing accuracy (202,203), although some questions of cost-effectiveness are likely to remain. However, the low rate of cure once disease is present outside of the thyroid gland bolsters interest in early diagnosis and treatment.

    This Guideline defers the recommended approach to thyroid nodules, including FNA and serum Ct testing, to the ATA Guideline that addresses thyroid nodules (5). However, if obtained, a basal or stimulated* serum Ct level ≥100 pg/mL should be interpreted as suspicious for MTC and further evaluation and treatment should ensue (Fig. 2). Grade: A Recommendation

[C1] Preoperative laboratory testing for presumed MTC when an FNA or Ct level is diagnostic or suspicious for MTC (Fig. 2). Preoperative laboratory testing has three purposes: first, to predict the extent of metastatic disease because this will determine the extent of preoperative imaging and may alter the surgical approach (Fig. 2); second, to identify the comorbid conditions of PHPT and/or PHEO in MEN 2 that would alter the surgical approach and surgical priorities; and third, to identify RET mutation carriers so that testing of appropriate family members can allow for early diagnosis and treatment of affected individuals.

Although a variety of substances including carcinoembryonic antigen (CEA), chromogranin A, Ct gene-related peptide, adrenocorticotropic hormone, amyloid, somatostatin, serotonin, and vasoactive intestinal peptide can be produced by MTC tumor cells, Ct is the main biochemical marker used for detection, staging, postoperative management, and prognosis in patients with MTC. Very rarely Ct may not be elevated in the serum despite clinically apparent MTC (204).

Concomitant PHPT is unlikely in the absence of MEN 2A, however, because measurement of serum calcium is readily available and because RET oncogene testing results may not be available before surgery (although when possible, the germline RET mutation status of the patient should be determined prior to surgery), preoperative measurement of serum calcium is prudent. Similarly, PHEO is unlikely in the absence of MEN 2. However, because PHEO would alter operative planning it is critical that this tumor be excluded or treated prior to surgery for MTC. Completion of both the adrenal and neck surgeries during the same anesthesia induction may be considered.

    Preoperative evaluation of patients presumed to have MTC (when an FNA or Ct level is diagnostic or suspicious for MTC) should include serum measurements of basal Ct,CEA, and calcium (albumin-corrected or ionized); and RET protooncogene analyses. PHEO preoperative screening should begin by age 8 years for those with MEN 2B, and mutated RET codons 634 and 630; and by age 20 years for other RET mutations. Exclusion of PHEO may include any of the following tests: 1) negative RET protooncogene analysis and family history; 2) negative plasma free metanephrines and normetanephrines, or negative 24-hour urine metanephrines and normetanephrines; 3) negative adrenal CT or magnetic resonance imaging (MRI). Grade: A Recommendation

[C2] Evaluation and treatment of PHEO.   Routine PHEO screening was discussed above. The high risk of anesthesia and surgery in the presence of a PHEO dictates that its management takes the highest surgical priority. In the setting of MTC, the presence of an extra-abdominal PHEO is rare.

    Patients with MTC and elevated plasma free metanephrines or normetanephrines, or elevated 24-hour urine metanephrines or normetanephrines should undergo adrenal imaging for PHEO with MRI or CT scan. Grade: A Recommendation
    In the setting of MTC, routine extra-abdominal imaging for PHEO is not indicated. Grade: E Recommendation
    PHEO should be surgically resected after appropriate preoperative preparation and prior to surgery for MTC or PHPT, preferably by laparoscopic adrenalectomy. Grade: A Recommendation

One study documented that 22% of patients experienced several episodes of Addisonian crisis, including a death, after bilateral adrenalectomy. The authors concluded that adrenal-sparing adrenalectomy and close monitoring of the remnant may be the treatment of choice for hereditary bilateral PHEO in MEN 2A, since overall recurrence is low (205).

    Cortical-sparing adrenal surgery may be considered in patients requiring surgery when there is only one remaining adrenal gland, or when bilateral PHEOs are present. Grade: C Recommendation

[C3] Preoperative imaging for presumed MTC when an FNA or Ct level is diagnostic or suspicious for MTC (Fig. 2).   Preoperative imaging is indicated because local neck or distant metastatic disease may change the operative approach. The sensitivity of intra-operative palpation to detect lymph node metastases by experienced surgeons is only 64% (117). Lymph node metastases are present in >75% of patients with palpable MTC (117,119). In the setting of an experienced ultrasonographer, neck US is the most sensitive test to detect local metastases in the cervical compartments and upper aspect of the superior mediastinum (206). However, it is common that a higher number of malignant lymph nodes are removed surgically during compartmental lymph node dissections than were visualized preoperatively with US, which demonstrates the reduced sensitivity of all diagnostic maneuvers to localize the smallest lymph node metastases.

Patients with distant metastases are viewed as incurable, and the goals of locoregional surgery may differ from the goals of surgery in patients with less extensive disease. Distant metastases most commonly affect the bones/bone marrow, liver, and lungs (207). Metastases to brain and skin are less common and associated with multisystemic disease and poor 1-year survival (208). Liver metastases often appear similar to hepatic hemangiomas with calcifications (209). Unfortunately, radiographic detection of distant metastatic disease is unlikely when the preoperative Ct level is <250 pg/mL (210). Machens et al. (94) found that radiographically identifiable distant metastases began to appear in the primary surgery setting at a preoperative basal serum Ct level of 400 pg/mL and at primary tumor diameters of 12mm. In the setting of the primary surgery, the risk of radiographically detectable distant metastases exceeded 50% at preoperative basal serum Ct levels of 15,000 pg/mL, and primary tumor diameters of 50mm (94). Distant metastases were almost always present when preoperative basal serum Ct levels were >100,000 pg/mL or the primary tumor diameter was >60mm (94). The cumulative risks of distant metastases did not differ between sporadic and hereditary MTC (94).

Giraudet et al. (206) reported that the most sensitive methods to detect metastases in the neck was US followed by contrast-enhanced CT. CT was most sensitive to detect lung and mediastinal lymph node metastases. Contrast-enhanced MRI was the most sensitive to detect liver metastases. Axial MRI and bone scintigraphy were complementary and most sensitive to detect bone metastases. Fluorodeoxyglucose positron emission tomography (FDG PET) was less sensitive than these modalities to identify metastases. Oudoux et al. (211) also found that CT was more sensitive than FDG PET for the lung and liver, and that MRI of the spine and pelvis was more sensitive than FDG PET to detect bone and bone marrow metastases. However, FDG PET was more sensitive than CT to detect disease in the neck and mediastinum in their series. While correlated, Ct doubling time (DT) and the CEA DT are better predictors of tumor progression than is the FDG PET maximum standardized uptake value (SUVmax) (211,212).

Unfortunately, no single test provides optimal whole-body imaging. This Task Force concluded that a comprehensive preoperative imaging strategy was not practical, and probably was not necessary to guide initial therapy as nearly all patients with residual disease postoperatively can be identified biochemically and selected then for further evaluation.

    Preoperative neck US is recommended for all patients when an FNA or Ct level is diagnostic or suspicious for MTC. Grade: A Recommendation
    Preoperative chest CT, neck CT, and three-phase contrast-enhanced multidector liver CT or contrast-enhanced MRI is recommended for all patients with suspected MTC when local lymph node metastases are detected (N1), or the serum Ct is >400 pg/mL. Grade: C Recommendation
    FDG PET imaging and somatostatin receptor imaging are not recommended for routine initial screening for MTC metastases in patients when an FNA and/or Ct level is diagnostic or suspicious for MTC. Grade: E Recommendation

[C4] Surgery for MTC patients without advanced local invasion or cervical node or distant metastases (Fig. 2).   These patients have not undergone prior thyroidectomy, have no evidence of cervical lymph node metastases by physical examination and cervical US. MTC has a high rate of lymph node metastases (117,119) that are suboptimally detected preoperatively in the central compartment by US or intra-operatively by the surgeon (117), and re-operation is associated with a higher rate of surgical complications (119). For these reasons, most authors advocate for a total thyroidectomy and prophylactic central neck dissection in the setting of clinically detected MTC (12).

    Patients with known or highly suspected MTC with no evidence of advanced local invasion by the primary tumor, no evidence cervical lymph node metastases on physical examination and cervical US, and no evidence of distant metastases should undergo total thyroidectomy and prophylactic central compartment (level VI) neck dissection. Grade: B Recommendation

Because of the low rate of biochemical cure in patients with lymph node metastases or large primary tumors, there is diminished enthusiasm for prophylactic lateral neck dissections. Indeed, Machens et al. (94) reported that the cumulative rates of biochemical remission (basal and pentagastrin-stimulated serum Ct <10 pg/mL) in node-negative MTC patients declined to 50% when the preoperative basal serum Ct levels was>300 pg/mL, or the primary tumor measured more than 10mm. Overall, 38% of node-negative MTC patients who undergo extensive surgery failed to achieve normal postoperative serum Ct levels, suggesting early radiographically occult distant metastases (94). In node-positive patients, only 10% achieved postoperative basal and pentagastrin-stimulated serum Ct levels <10 pg/mL, which did not happen when the preoperative basal Ct level was >3000 pg/mL or the tumor was >40mm in diameter (94). The correlation with biochemical remission was better for basal than for pentagastrin-stimulated serum Ct levels. About 3.3% of patients that achieve biochemical remission are likely to demonstrate biochemical recurrence over the subsequent 0.7 to 7.5 years (213). In addition, lateral neck compartmental dissection can be associated with long-term cosmetic and functional consequences. Thus, in the current era of high resolution neck imaging, lateral neck dissection (levels IIA, III, IV, V) may be best reserved for patients with positive preoperative imaging, although a minority of the Task Force favored prophylactic lateral neck dissection when lymph node metastases were present in the adjacent paratracheal central compartment.

[C5] Surgery for MTC patients with limited local disease and limited or no distant metastases (Fig. 2).   Limited local disease is considered ≤ T3 and ≤ N1b lymph node status with subcentimeter lymph node metastases including those with minor extra-nodal extension (Table 4). Limited distant metastases are typically subcentimeter in size but may also include macroscopic distant metastases when they are few in number. Significant differences in survival times are present between patients who achieve complete remission, those with biochemically persistent disease postoperatively, and those with distant metastases (214). Unfortunately, most MTC patients with metastases to regional lymph nodes are not biochemically cured despite aggressive surgery to include bilateral neck dissection. Modigliani et al. (40) demonstrated in multivariate analysis that age and stage were independent predictive factors of survival, whereas the type of surgery was not. However, in patients with persistent elevations in Ct levels, survival was still good: 80.2% and 70.3% at 5 and 10 years, respectively. Similarly, Pelizzo et al. (12) demonstrated in multivariate analysis that age, stage, and extent of surgery were independent predictive factors of survival; with more extensive surgery correlating with a worse prognosis. Leggett et al. (215) demonstrated that an increased number of lymph nodes resected was associated with improved survival in node-positive patients by categorical (1 lymph node versus >1 lymph node), but not continuous, multivariate analysis. This finding was interpreted to indicate a finite benefit to increasing the number of lymph nodes resected with patient outcome being dominated by patient age and tumor size. Machens et al. (94) reported a 10% rate of normalization of postoperative basal Ct levels in nodepositive MTC patients. Metastases in 10 or more lymph nodes, or involvement of more than two lymph node compartments nearly precludes normalization of serum Ct (119,216,217). Unfortunately, lymph node involvement is common and the incidence of lateral compartment lymph node metastases is related to the incidence of central compartment lymph node metastases. Machens et al. (218) reported that the rate of ipsilateral lateral compartment lymph node metastases when no central compartment lymph node metastases were present, 1–3 central lymph node metastases were present, or when ≥4 central lymph nodes were present was 10.1%, 77%, and 98%, respectively. The rate of contralateral lateral compartment lymph node metastases when no central compartment lymph node metastases were present, 1–9 central lymph node metastases were present, or when ≥10 central lymph nodes were present was 4.9%, 38%, and 77%, respectively. However, resection of local disease may decrease the risk of local recurrence (13,119,219), and clearance of the central compartment may prevent future complications such as invasion into the recurrent laryngeal nerve or aerodigestive track with resulting loss of speech or swallowing (12). For these reasons, most authors suggest that if metastastic lymph nodes are identified, then a compartment- oriented lymph node dissection should be done (12,81,94,220–224).

    MTC patients with suspected limited local metastatic disease to regional lymph nodes in the central compartment (with a normal US examination of the lateral neck compartments) in the setting of no distant (extracervical) metastases, or limited distant metastases should typically undergo a total thyroidectomy and level VI compartmental dissection. A minority of the Task Force favored prophylactic lateral neck dissection when lymph node metastases were present in the adjacent paratracheal central compartment. Grade: B Recommendation
    MTC patients with suspected limited local metastatic disease to regional lymph nodes in the central and lateral neck compartments (with US-visible lymph node metastases in the lateral neck compartments) in the setting of no distant metastases, or limited distant metastases should typically undergo a total thyroidectomy, central (level VI), and lateral neck (levels IIA, III, IV, V) dissection. Grade: B Recommendation
    In the presence of distant metastatic disease, less aggressive neck surgery may be appropriate to preserve speech, swallowing, and parathyroid function while maintaining locoregional disease control to prevent central neck morbidity. Grade: C Recommendation

[C6] Surgery for MTC patients with advanced local disease or extensive distant metastases (Fig. 2).   In most patients, the surgical goal is thyroidectomy, level VI compartmental dissection, and therapeutic (clinical or image-positive) lateral neck dissection. However, in the presence of extensive distant metastases or advanced local features, the goals of surgical therapy are typically more palliative with attention to minimizing complications, such as hypoparathyroidism, and maintaining normal speech and swallowing. These patients should additionally be considered for clinical trials, and other palliative therapies including surgery, external beam radiation therapy (EBRT), and hepatic embolization. In patients with infiltrative central neck disease that does not involve the trachea or esophagus, the extent of local surgery—especially with respect to unilateral resection of the recurrent laryngeal nerve when the tumor or associated nodal disease can not be completely separated from the nerve—remains controversial even among the expert panel. In patients with high volume extra-cervical metastatic disease, preservation of speech and swallowing is the major goal of therapies (to include surgery) directed at the neck. In patients with small volume extra-cervical metastatic disease, or disease confined to the neck, a more aggressive approach to locally invasive central compartment disease may be more appropriate to prevent local tumor recurrence; in some patients, resection of a unilateral recurrent laryngeal nerve and the use of adjuvant EBRT may be appropriate.

In patients with locally invasive MTC that involves the trachea, thyroid cartilage, and/or esophagus, the extent of extirpative surgery performed in the neck (palliative debulking, laryngectomy, esophagectomy, laryngopharyngectomy, or observation alone) is influenced by an assessment of the patient’s life expectancy based on the extent of extra-cervical metastatic disease and other medical comorbidities. The desire for the short-term maintenance of speech and swallowing is balanced by the concern for longterm locoregional disease control and critical to this balance is the estimated duration of survival. These decisionsmay be best individualized by an experienced multidisciplinary treatment team.

    In the presence of advanced local or distant disease, less aggressive neck surgery may be appropriate to maintain local disease control while preserving speech, swallowing, and parathyroid function. Grade: C Recommendation
    In patients with extensive distant metastases a palliative neck operation may still be needed when there is pain, or evidence of tracheal compromise and the need to maintain a safe airway. Otherwise, in the setting of moderate to high volume extra-cervical disease, neck disease may be observed and surgery deferred (Task Force opinion was not unanimous). Grade: C Recommendation

[C7] Thyrotropin suppression therapy in MTC.   Differentiated epithelial cell–derived thyroid cancers are thyrotropin (TSH)-dependent and suppression of TSH is desirable to arrest or retard the rate of cell growth (5). By contrast, C-cell tumors are not TSH-dependent and there is no evidence that thyroxine (T4) therapy aimed at TSH suppression reduces recurrences or improves survival of patients with MTC. Serum TSH should be measured 2–3 months postoperatively, and T4 dose adjusted to keep the TSH in the normal range.

    Replacement rather than suppressive T4 therapy with target serum TSH levels between 0.5 and 2.5 mIU/L is recommended for patients with MTC. Grade: B Recommendation 

[C8] Somatic RET testing in sporadic MTC.   The presence of somatic RET mutations in sporadic MTC cells is about 40–50% (35–37), however, these tumors are heterogeneous and meticulous microdissection studies have shown somatic RET mutations in up to 80% of sporadic MTCs, most commonly 918ATG → ACG (225,226), and less frequently at codon 634 and others (37).

There are two main reasons to consider genotyping sporadic MTCs. First, is that tumor mutation status may predict its response to therapy. Many of the new small molecule therapeutics are designed to target RET, or its downstream pathways, thus theoretically patients whose MTCs contain somatic RET mutations, compared to those whose MTCs do not, might be more likely to respond to these compounds. So far, however, this hypothesis has not been proven in MTC. Moreover, because many of the tyrosine kinase inhibitors entering clinical trials at present inhibit multiple kinases, the precise mechanisms by which they may inhibit tumor growth and/or progression may not be predicted by RET mutations.

A second reason to consider analyzing sporadic MTCs for somatic RET mutations is that some studies have demonstrated that tumors with an identifiable RET mutation have a more aggressive course than those without RET mutations (36,37,227). However, others have not found this association (226). Recently, Elisei et al. (37) reported that the presence of a somatic RET mutation in sporadic MTC was associated with larger tumor size, the presence of nodal and distant metastases, and advanced stage at diagnosis. On multivariate analysis of all prognostic factors correlated with persistent disease or death, only advanced stage at diagnosis and the presence of a RET mutation showed an independent correlation. However, it is not clear how knowledge of the somatic RET mutational status would change patient management, or how this predictor of the clinical course compares to others such as the Ct or CEA DTs.

    Currently, sporadic MTC tumors should not be routinely evaluated for somatic RET mutations, although agreement amongst the Task Force was not unanimous. Grade: D Recommendation