2021 Volume 68 Issue 8 Pages 871-880
Current histopathological diagnosis methods cannot distinguish the two types of thyroid carcinoma: clinically significant carcinomas with a potential risk of recurrence, metastasis, and cancer death, and clinically insignificant carcinomas with a slow growth rate. Both thyroid tumors are diagnosed as “carcinoma” in current pathology practice. The clinician usually recommends surgery to the patient and the patient often accepts it because of cancer terminology. The treatment for these clinically insignificant carcinomas does not benefit the patient and negatively impacts society. The author proposed risk stratification of thyroid tumors using the growth rate (Ki-67 labeling index), which accurately differentiates four prognostically relevant risk groups based on the Ki-67 labeling index, ≥30%, ≥10 and <30%, >5 and <10%, and ≤5%. Indolent thyroid tumors with an excellent prognosis have the following four features: young age, early-stage (T1-2 M0), curatively treated, and low proliferation index (Ki-67 labeling index of ≤5%), and are unlikely to recur, metastasize, or cause cancer death. Accurate identification of these indolent tumors helps clinicians select more conservative treatments to avoid unnecessary aggressive (total thyroidectomy followed by radio-active iodine) treatments. Clinicians can alleviate the fears of patients by confirming these four features, including the low proliferation rate, in a pathology report immediately after surgery when patients are most concerned.
A marked increase in the prevalence of thyroid carcinoma has been observed worldwide. However, the number of deaths from thyroid carcinoma remains unchanged. Based on these two observations, epidemiologists suggested a high risk of overdiagnosis and overtreatment of thyroid carcinomas in current clinical practice, which does not benefit the patient and negatively impacts society as a whole [1-3]. This opinion paper introduces a prognostic risk classification of thyroid tumors using the Ki-67 labeling index [4, 5] and a strategy to integrate it with the AJCC (American Joint Committee on Cancer) TNM (tumor-node-metastasis) staging system [6] to identify indolent thyroid tumors that are unlikely to recur, metastasize, or cause cancer death. It reliably distinguishes clinically significant carcinomas with the potential risk of recurrence, metastasis, and cancer death from clinically insignificant carcinomas with a slow growth rate. A high Ki-67 labeling index also identifies carcinomas with a high risk of recurrence and death [7-13], and the Ki-67 labeling index was previously recommended as an essential component in pathology reporting [5, 14].
Welch classified cancer into four types (fast-growing, slow-growing, very slow-growing, and non-progressive) based on growth speed [1]. Fast-growing cancer develops local and systemic symptoms due to rapid tumor growth and metastasis, and patients die of tumors in a short period after the onset of the disease. Among thyroid carcinomas, anaplastic carcinoma of the thyroid corresponds to fast-growing cancer. Slow-growing cancer grows and manifests over more than several years. If left untreated, the patient will die of the tumor years later. If it can be detected/treated at an early stage, it is curable. Very slow-growing cancer does not cause death and the patient dies of another illness before the tumor causes any symptom. Welch emphasized that very slow-growing cancer does not harm the patient and does not require treatment. The fourth group is cancer that grows but spontaneously disappears and is termed non-progressive cancer. However, many thyroid carcinomas, including symptomatic carcinomas and carcinomas with lymph node metastasis, do not cause death [15-22]. The 10-year disease-specific survival rate (DSSR) of AJCC stage I papillary thyroid carcinoma (PTC) was reported to be 99.6% by Kim et al. in a Korean patient cohort [19], 99.7% in a Chinese patient cohort by Tam et al. [18], and 99.6% in a European population by van Velsen et al. [20]. An international multi-institutional study by Nixon et al. of 9.484 thyroid cancer patients revealed a 10-year DSSR of 99.5% [21]. The 20-year DSSR rate of AJCC stage I PTC was 99.3% in a Japanese patient cohort reported by Ito et al. [22]. Therefore, cancer mortality does not occur in more than 99% of patients with AJCC stage I PTC after complete excision regardless of lymph node metastasis and extrathyroid invasion.
Incorporating the concept of “tumor growth speed determines the prognosis” advocated by Welch, the author published a risk classification of thyroid follicular cell tumors based on the tumor growth speed in 2015 (Table 1) [4]. This classification is characterized by the Ki-67 labeling index to evaluate the proliferative capacity of tumor cells. Cell growth activity is one of the two characteristic features (abnormal cells divide without control and can invade nearby tissues) of carcinoma [23]. It is an essential feature for assessing the biological behavior of low-grade malignancies. The most useful marker to evaluate cell proliferative activity is Ki-67, which is expressed in the cell nuclei in all cells except for those in the G0 phase [24]. It is immunohistochemically evaluated as the labeling index using formalin-fixed paraffin-embedded tissue specimens. There are two potential applications of the Ki-67 labeling index in human neoplasms: 1) differential diagnosis between benign and malignant lesions, and 2) risk classification of low-grade malignancies, which is currently performed in clinical practice for grading diagnoses of neuroendocrine tumors, breast carcinomas, malignant lymphoma, and gastrointestinal stromal tumors [24, 25]. No major staging systems of thyroid tumors (AMES, American Thyroid Association (ATA), EORTIC, Japan Association of Endocrine Surgery (JAES), MACIS, and TNM) include one of the two essential cancer definition criteria, tumor growth speed, in their classification schema. The reason for adopting cutoff values at 3%, 5%, 10%, and 30% is because the World Health Organization (WHO) classification previously defined the Ki-67 labeling index of normal follicular cells as <3%, well-differentiated carcinoma as <10%, and anaplastic thyroid carcinoma as >30% [18]. Furthermore, the cutoff values of 5% and 10% are used to maintain consistency with a study from Kuma Hospital by Miyauchi et al., which demonstrated a significant prognostic difference in PTC using these values [12]. Other previous studies reported lower cutoff values (1.85% by Kjellman et al., 2.5% by Tang et al., and 3% by Ito et al.) [9, 11, 13]. These values were not incorporated in our prognostic classification because they were elucidated from studies with less sensitive Avidin-Biotin Complex immunohistochemical methods [4].
| Classification of Tumors | Ki-67 | Incidence | Clinical managements |
|---|---|---|---|
| 1 Benign tumors | |||
| Follicular adenoma | <3% | na | No further treatment if resected. |
| 2 Borderline tumors (pN0, pEx0 and M0) | |||
| 1) Encapsulated tumors | |||
| HTT, WDT-UMP, FT-UMP, NIFTP, NEPRAS | <3% | na | No further treatment if resected. |
| 2) Non-encapsulated tumors (<1 cm) | |||
| Papillary microcarcinoma | <3% | na | Active Surveillance or lobectomy. |
| 3 Malignant tumors (invasive carcinoma) | |||
| 1) Low-risk | ≤5% | 312/390 (80%) | Lobectomy is sufficient for stage I PTCs. |
| 2) Moderate-risk | >5 and ≤10% | 48/390 (12%) | Multidiciplinary discussion to decide treatment modality. |
| 3) High-risk | >10 and ≤30% | 30/390 (8%) | Total thyroidectomy with adjuvant RAI. |
| 4) Anaplastic thyroid carcinoma | >30% | na | Systemic treatments after multidiciplinary discussion. |
The author suggests appropriate clinical management for each risk group in the right-most column. The incidence (second from the right) of PTCs in Ki-67 low-risk, moderate-risk, and high-risk groups treated by total thyroidectomy at Kuma Hospital reported by Miyauchi [12] is shown. HTT, hyalinizing trabecular tumor; WDT-UMP, well-differentiated tumor of uncertain malignant potential; FT-UMP, follicular tumor of uncertain malignant potential; NIFTP, noninvasive follicular thyroid neoplasm with papillary-like nuclear features; NEPRAS, noninvasive encapsulated papillary RAS-like thyroid tumor; RAI, radioactive iodine; na, not applicable.
As risk stratification of surgically treated cases is highly important in clinical practice, the author modified Welch’s original classification for natural history of carcinomas (Table 1). The author excluded non-progressive tumors (shown by a purple dotted line in Fig. 1 for comparison) (Fig. 1). The highest risk group is anaplastic carcinoma. Although the number is small among surgical cases, it was defined as a tumor with a Ki-67 labeling index of >30%, which corresponds to anaplastic thyroid carcinoma [4, 7, 26, 27]. After excluding anaplastic thyroid carcinoma, surgically treated follicular cell-derived thyroid carcinomas were subdivided into the following three (high-, moderate-, and low-) risk groups based on the Ki-67 labeling index, 1) >10 and ≤30%, 2) >5 and ≤10%, and 3) ≤5%, respectively (Table 1). High-risk thyroid carcinoma of follicular cell origin is a tumor with a high Ki-67 labeling index of >10% and ≤30%, which often recurs and metastasizes, and can lead to death. It corresponds to poorly differentiated carcinomas [28] (Fig. 2), thyroid carcinomas with high-grade histology [29], and cases with distant metastases [30]. The 5-year tumor-specific survival rate was estimated to be 40–80%. Moderate-risk thyroid carcinoma of follicular cell origin is defined by a Ki-67 labeling index of >5 and ≤10%. It is termed moderate-risk carcinoma because the value is between those of low-risk and high-risk carcinomas. It corresponds to widely invasive FTC, aggressive variant PTC, and PTC with a gross extrathyroid extension (Fig. 3). Low-risk thyroid carcinoma of follicular cell origin is a group of carcinomas that have a Ki-67 labeling index of ≤5% (Fig. 4), which is a very slow growth rate that overlaps significantly with those of benign and borderline tumors (Ki-67 labeling index of <3%) (Fig. 5).
Classification of thyroid carcinomas of follicular cell origin in four prognostic risk groups based on the Ki-67 index (proliferation rate) and Welch’s non-progressive cancer.
Based on the tumor growth rate (Ki-67 labeling index), the natural history of thyroid carcinomas of follicular cell origin is classified into four risk groups (using cutoff values, ≤5%, >5 and ≤10%, >10% and ≤30%, and >30%). 1) Low-risk (Ki-67 labeling index of ≤5%, black arrow), 2) moderate-risk (Ki-67 labeling index of >5 and ≤10%, blue arrow), 3) high-risk (Ki-67 labeling index of >10 and ≤30%, green arrow), and 4) anaplastic (Ki-67 labeling index of >30%, red arrow). Welch’s non-progressive cancer was added as a purple dotted line for comparison. Both non-progressive cancer and low-risk carcinoma do not cause cancer death because of their very slow growth rate. In addition to very slow-growing cancer in Welch’s classification, the Ki-67 low-risk carcinomas are symptomatic, and may develop small-volume lymph node metastasis and minor extrathyroid extension.
An example of high-risk carcinoma of follicular cell origin found in a 56-year-old Chinese male patient who developed lung metastasis after total thyroidectomy. A: Poorly differentiated carcinoma with a solid growth pattern. Note a small focus of tumor necrosis on the upper right corner. B: Ki-67 labeling index of greater than 20%. Please note the negative staining in a necrotic area at the upper right corner of B. (A: HE staining, ×20, B: Ki-67 immunohistochemistry, ×40)
An example of moderate-risk carcinoma of follicular cell origin found in a 59-year-old Japanese female. Papillary carcinoma, pT4a, Ex2 (trachea), pN1b. The patient underwent total thyroidectomy at an advanced stage with trachea invasion. After one year, she had an increased serum thyroglobulin level (101 ng/mL) and multiple lung metastases on CT examination. A: Invasive papillary carcinoma invading extrathyroid fibrous connective tissue and B: Ki-67 labeling index of approximately 9%. (A: HE staining, ×10, B: Ki-67 immunohistochemistry, ×20)
An example of low-risk thyroid carcinoma of follicular cell origin found in a 32-year-old Japanese female. A: Papillary carcinoma showing expansive growth in the left lobe of the thyroid gland. B: Classic-type papillary carcinoma exhibiting a papillary growth pattern. Note ground glass nuclei, irregular nuclear contours, and few nuclear grooves characteristic of papillary carcinoma. C: There are three brown nuclei in this field, but only one was judged as a tumor nucleus. As the other two are in the glandular space, they were judged as reactive histiocytes. The Ki-67 labeling index in this field is <1%. (A: HE staining, loupe image, B: HE stain ×20, C: Ki-67 immunohistochemistry, ×20)
An example of a borderline tumor, NIFTP (noninvasive follicular thyroid neoplasm with papillary-like nuclear features), found in a 59-year-old Japanese male patient. A: Cut surface of the thyroid demonstrates a well-demarcated solid nodule after formalin fixation. B: Higher magnification reveals nuclear irregularity and pale chromatin in tumor cells with a follicular pattern. C: Ki-67 immunohistochemistry demonstrated a low index of <1%. (A: Gross photo, B: HE staining ×40, C: Ki-67 immunohistochemistry, ×10)
The first study of the Ki-67 index focusing on PTCs was by Kjellman et al., and they demonstrated an association between a high Ki-67 index and more aggressive disease (recurrence, metastasis, and cancer death) [9], which was later confirmed by several authors [11-13, 31-36]. The Ki-67 labeling index was reported to be an independent prognostic factor for recurrence [11, 24, 33, 34] and cancer death [11, 24]. Pan et al. and Guadagno et al. conducted meta-analyses separately. They found significant associations between the Ki-67 labeling index and patient age, tumor size, nodal metastasis, distant metastasis, extrathyroid extension, and TNM stage. Furthermore, they confirmed that patients with a high Ki-67 labeling index had a poorer disease-free survival and increased risk of mortality than patients with a low Ki-67 labeling index [32, 36]. However, earlier studies used less sensitive immunohistochemical methods and their absolute values are no longer comparable with those in recent studies, although their conclusions are overall consistent with our conclusion that the Ki-67 examination can identify curable carcinomas as cases with a low Ki-67 index. A high Ki-67 index also accurately identifies high-grade carcinomas, such as poorly differentiated carcinomas, with a Ki-67 labeling index of >10 and ≤30%. Considering this, high-risk thyroid carcinoma of follicular cell origin was listed in the 4th edition WHO classification of thyroid tumors as a synonym for poorly differentiated carcinoma [25].
Kitayama, a thyroid cancer survivor and one of the authors, wondered if pathologists can differentiate indolent thyroid tumors that are curable by simple excision from clinically significant cancers with a potential risk of recurrence and cancer death [37]. She believes that this distinction will help relieve the fear of recurrence because without such support, patients will fear recurrence for the rest of their lives [37]. Kitayama found many reported cases in which a complete cure was expected, even in the intermediate-risk group of PTC defined by the JAES and the ATA clinical guidelines [16, 17]. However, it required a long time and effort for her to reach this conclusion. She is concerned that many patients, who worry about recurrence and metastasis, are deprived of their peace of mind [37].
As a dermatologist, Kitayama noted the same situation in the dermatology field. There are two types of skin tumors, squamous cell carcinoma (cancer with a risk of recurrence) and basal cell carcinoma (cancer cured by simple excision). In dermatology practice, complete excision is initially performed. After excluding squamous cell carcinoma, the dermatologist usually tells the patient, “I was able to confirm that it was a basal cell carcinoma. It was completely removed and has been cured.” This explanation is prepared in order for the patient to feel assured. The terms for skin cancer clearly distinguish carcinomas with a potential risk of recurrence from those cured by excision. If the pathologist makes this distinction clear in thyroid tumor classification, the patient will feel greater peace of mind [37]. The distinction between slow-growing cancer and very slow-growing cancer by Welch fulfills Kitayama’s request; however, Welch did not discuss how to distinguish very slow-growing cancer from slow-growing cancer clinically [1]. The risk classification using the Ki-67 labeling index, as shown in the Table 1, is the first and only histopathological classification of thyroid tumors enabling this distinction [4]. The attending physician can accurately predict the risk of recurrence and death of the patient using the Ki-67 labeling index-based classification. As most patients have very low-risk carcinoma with a Ki-67 labeling index of ≤5%, the attending physician can relieve their anxiety immediately after surgery when they are most concerned about recurrence.
Ito et al. reported that the 20-year DSSR of stage I PTC was 99.3% in a Japanese patient cohort [22] even though the AJCC stage I PTC in young patients (<55 years old) includes a broad disease spectrum. It includes very low-risk diseases, such as intrathyroidal papillary microcarcinomas (ATA and JAES low-risk), to clinically significant carcinomas such as PTCs with >5 lymph node metastasis (ATA and JAES intermediate-risk) and PTCs with gross extrathyroidal extension (ATA and JAES high-risk). This suggested that there were many thyroid carcinomas with lymph node metastasis and extrathyroid extension that did not recur or metastasize. By excluding older patients (≥55 years old) and Ki-67 moderate- and high-risk carcinomas in the young age group, the author estimated the DSSR of early-stage (T1-2 M0) PTC to approach almost 100% when curative treatment is carried out successfully. The author further proposes integrating four features, young age (<55 years old), early-stage (T1-2 M0), curatively treated, and low tumor growth rate (Ki-67 labeling index ≤5%), to identify curable carcinomas that are unlikely to recur, metastasize, or cause cancer death. This method is almost equivalent to a strategy reported by Ghaznavi et al., which integrated three clinical features, young patients (<55 years old), ATA low-risk categories, and AJCC stage I [38]. They reported a DSSR of 100% for well-differentiated thyroid carcinomas when they fulfilled these three features [38]. However, our proposed method covers more patients (both the ATA low-risk and intermediate-risk patients) than Ghaznavi’s method, as there are still many curable PTCs among ATA intermediate-risk carcinomas, such as cases with regional lymph node metastasis and minor extrathyroid extension, which accounted for 38.4% of differentiated thyroid carcinomas by Grani et al., 56% of PTCs by Ghaznavi et al. and 68.0% of PTCs by Lee et al. [38-40].
After the re-stratification based on the dynamic risk stratification system, among patients who had an initial excellent response, the probability of having structural incomplete response was reduced from 3 to 2% in ATA low-risk patients and from 18 to 2% in ATA intermediate risk patients by a thorough review by Pitoia and Jerkovich [41]. Their conclusion is in good agreement with our conclusion, that is integrating four clinical features, (young age, early-stage (T1-2 M0), curatively treated, and low Ki-67 labeling index of <5%), to identify very low-risk PTCs that are unlikely to recur, metastasize, or cause cancer death immediately after initial thyroid surgery.
Randolph et al. reviewed the literature and concluded that small-volume microscopic lymph node metastases in PTC are often of little clinical significance [42]. In their interpretation, PTC is often associated with subclinical microscopic lymph node metastases, and these lesions usually do not progress and seldom become clinically relevant [42]. In our interpretation, the tumor growth speed determines the clinical significance of microscopic lymph node metastases, and tumors with a high Ki-67 index develop into clinically significant diseases, whereas those with a low Ki-67 index remain stable and clinically insignificant. Midorikawa et al. observed growth arrest (self-limiting cancer) in many cases of pediatric PTC with lymph node metastases on ultrasound during clinical follow-up of histologically confirmed PTC [43]. Furthermore, several authors emphasized that minor extrathyroid extension did not impact the prognosis [44-46], and it was excluded from the T3 criteria by the 8th edition of the AJCC staging system (AJCC) [6]. Based on these observations, the author concludes that cases of small-volume lymph node metastasis and minor extrathyroid extension have an excellent prognosis as long as the Ki-67 is very low. The patient age, younger than 55 years old, was integrated into this risk classification to identify curable thyroid carcinoma with a negligible risk of recurrence, metastasis, and cancer death. It was designed to exclude ATA and JAES high-risk patients often found in older age groups; however, Ito et al. reported that the JAES low-risk patients in the older age group also had a favorable prognosis equivalent to that of the young age group [47, 48]. As this study concentrated on the young age group, future studies should focus on identifying curable PTCs in older age groups.
In summary, a Ki-67 labeling index of ≤5% can identify a significant number of curable carcinomas in ATA intermediate-risk and high-risk patients. This conclusion is significant for clinicians to avoid the overtreatment of intermediate-risk and high-risk patients, as most clinical guidelines recommended total thyroidectomy and adjuvant RAI treatment [49-51]. Avoiding unnecessary completion thyroidectomy (aggressive cancer surgery), particularly for young patients, is essential because a significant percentage has a low Ki-67 index (Fig. 4) and an excellent prognosis. Please refer to clinical management in the Table 1, as recommended by the author. Aggressive cancer treatments (total thyroidectomy followed by RAI) should be applied only to patients with a high (>10%) Ki-67 index and advanced-stage disease.
A limitation of the Ki-67-based risk classification is the requirement of surgical specimens. Thus, the tumor growth speed is evaluable only after surgery. It is desirable to determine the growth speed before surgery to prevent unnecessary surgery for benign tumors, borderline tumors, and low-risk carcinomas [23]. Fine-needle aspiration biopsy cytology or core needle biopsy specimens may be necessary [10]. However, biopsy specimens to not always represent hot spots, suggesting that these procedures miss lesions with high proliferative activity. The second limitation is that it predicts a long-term prognosis of more than 20 years with only a single growth data point. As the proliferation capacity of thyroid carcinoma changes with age [52-59], the long-term prognosis cannot be accurately predicted when cancer progression occurs and the growth rate varies. Proper adjustment is clinically desirable when possible. Clinical methods for assessing tumor growth potential include: 1) Estimating the growth rate from size changes on tumor images (primary thyroid tumor and recurrence when clinical follow-up is selected) [43, 56-58]. 2) Measurement of the Ki-67 index using tissue specimens (primary thyroid tumors and recurrent tumors). 3) Measurement of the serum thyroglobulin doubling time and serum thyroglobulin doubling rate if the patient has recurrence/metastasis after total thyroidectomy [43, 53, 60].
As there are several pitfalls in Ki-67 immunohistochemistry and measurement of the proliferation index, it is important to confirm that there are no conflicting clinical features when predicting the absence of recurrence, metastasis, and cancer death in patients with a Ki-67 index of ≤5%. Ki-67 staining is negative (false negative) in specimens with tumor necrosis or low in samples with poor antigen preservation, such as decalcified specimens. On the other hand, Ki-67-positive cells may be found among non-cancer cells such as vascular endothelial cells, histiocytes, and lymphocytes; the measured value increases when erroneously counted as positive tumor cells. In addition to these technical issues, there is tumor diversity in the proliferation index, and relatively low- and high-labeled areas may coexist within a single tumor or between the primary tumor and metastatic tumor [61]. There is no consensus on which to consider as the representative value. The author is currently reporting the highest value (hotspot) as the representative value. Scoring larger numbers of tumor cells may improve the accuracy of the Ki-67 index and is desirable, but scoring all cells on a whole-tissue section is impractical for most pathologists. We recommend counting positive cells among at least 500 tumor cells in hot spot areas to assess the Ki-67 index. Measurement of the cell proliferation fraction using Ki-67 immunohistochemistry, despite these pitfalls and technical difficulties, is an inexpensive method that can be performed at all pathology laboratories worldwide, differing from costly gene panel tests. However, most Western reporting systems do not recommend the Ki-67 index in the diagnosis of thyroid tumors [25, 62]. They emphasize increased mitotic counts and tumor necrosis (reflecting high proliferation capacity but lower sensitivity) to identify high-grade thyroid carcinomas and poorly differentiated thyroid carcinomas [25, 28, 29, 62]. However, their absence does not guarantee the indolent nature of thyroid carcinomas, as the majority (>95%) of PTC and FTC cases do not demonstrate increased mitoses or tumor necrosis.
The prognostic classification of thyroid tumors based on the Ki-67 index (Table 1) can predict the prognosis immediately after surgery. It accurately distinguishes low-risk thyroid carcinomas from moderate- and high-risk thyroid carcinomas. Advanced-stage carcinomas and non-curatively treated cases should be excluded from low-risk carcinoma of follicular cell origin, even with a Ki-67 labeling index ≤5%. Thus, thyroid carcinomas with a low (≤5%) Ki-67 labeling index can be further divided into two groups. 1) Curatively treated thyroid carcinomas at early-stage (T1-2 M0) for which a complete cure and long life expectancy are expected, and 2) non-curatively treated advanced stage (T3-4) thyroid carcinomas (incomplete resection) and stage II (with distant metastasis), which are persistent and have a potential risk of metastasis and cancer death. In conclusion, the author proposes integrating four clinical features, young age, early-stage (T1-2 M0), curatively treated, and low tumor growth rate (Ki-67 labeling index of ≤5%), to identify very low-risk PTCs that are unlikely to recur, metastasize, or cause cancer death. For young patients with carcinoma with a low Ki-67 index after curative surgery, the clinician can provide significant relief after surgery if the patient fears recurrence [63, 64]. Most PTCs have a Ki-67 labeling index of ≤5% based on the author’s experience (Table 1). It can be further confirmed with a dynamic risk assessment proposed by Tuttle et al. and Rosario et al. in patients who achieve an excellent response (absence of elevated serum thyroglobulin or structural disease detected by imaging tests) during clinical follow-up [41, 65, 66].
Recently Odate et al. published 10 cases of PTC operated on at Ito Hospital and Yamanashi University for 40 years [67]. In these 10 cases, the tumor recurred in the lymph nodes, developed anaplastic transformation, and eight patients died of tumor [67]. The Ki-67 labeling index was low at <5% in 8 cases of primary thyroid PTC, and it was not available in the other 2. As there were seven patients older than 55 years old and 7 subjects with advanced disease (pT3–4a), the proposed criteria successfully excluded all ten fatal PTCs from indolent thyroid tumors with an excellent prognosis.
All eight authors have nothing to disclose.
