- Email: [email protected]
Prevalence and biological behaviour of variants of papillary thyroid carcinoma experience at a single institute
Pathology (October 2008) 40(6), pp. 617–622
ANATOMICAL PATHOLOGY
Prevalence and biological behaviour of variants of papillary thyroid carcinoma: experience at a single institute YASUHIRO ITO*, MITSUYOSHI HIROKAWA{, TAKASHI URUNO*, MINORU KIHARA*, TAKUYA HIGASHIYAMA*, YUUKI TAKAMURA*, AKIHIRO MIYA*, KAORU KOBAYASHI*, FUMIO MATSUZUKA* AND AKIRA MIYAUCHI* Departments of *Surgery and {Pathology, Kuma Hospital, Kobe City, Japan
Summary Aims: There are many variants of papillary carcinoma, and some of these variants have been reported to show biological behaviours differing from that of conventional papillary carcinoma. In this study, we present our experience regarding the prevalence and prognoses of these variants of papillary carcinoma. Methods: H&E sections from 1521 patients who underwent initial surgery for papillary carcinoma in Kuma Hospital between 1987 and 1995 were re-reviewed and classified into conventional papillary carcinoma and various histological variants. We investigated the biological behaviours of these lesions, including prognoses. Results: Follicular, tall cell and oncocytic variants were observed in comparably high incidences: 6.6%, 3.9%, and 1.9%, respectively. Patients with tall cell variant showed significantly worse disease-free survival (DFS) and causespecific survival (CSS) rates than those with conventional papillary carcinoma. The prognoses of patients with follicular variant did not differ from those of patients with conventional papillary carcinoma. Patients with oncocytic variant have not shown carcinoma recurrence. Among the rare variants, which accounted for less than 1%, columnar cell carcinoma showed a worse prognosis. Conclusions: Since patients with some variants show different clinical outcomes from those with conventional papillary carcinoma, classification of variants might be helpful to predict patient prognosis.
preoperatively detected by imaging studies, independently affects not only disease-free survival (DFS) but also causespecific survival (CFS) of patients, while pathologically diagnosed node metastasis (pN) that was only microscopically detected does not significantly affect outcome.2 Furthermore, clinical and pathological massive extrathyroid extension (T4 and pT4) and distant metastasis at diagnosis (M1) independently predicted poor prognosis in patients.2,3 Furthermore, some histological findings have been reported to affect patient prognosis. Previous studies demonstrated that poorly differentiated carcinoma showed a poor clinical outcome in patients4,5 and this was adopted by the World Health Organization (WHO) classification as an independent entity.6 Furthermore, there are many other histological variants in papillary carcinoma as shown in the most recent edition of the WHO classification.7 Although studies on the prognosis of some variants have been performed, our knowledge about the prevalence and biological behaviour of many variants remains unclear because of the lack of studies involving a large number of patients. In this study, we investigated the prevalence of these histological variants using a series of patients who underwent surgery in our institute, and compared their biological behaviours, including patient prognoses, with those of conventional papillary carcinoma.
Key words: Papillary carcinoma, thyroid, variants, prognosis.
PATIENTS AND METHODS
Received 6 July, revised 12 and 24 September, accepted 3 October 2007
Patients
INTRODUCTION Papillary carcinoma is the most common malignancy originating from the thyroid. Generally, it is less virulent and shows a good prognosis if competently resected. However, tumours showing certain clinicopathological parameters often grow constantly and show a worse clinical outcome. In order to evaluate these parameters, the International Union Against Cancer (UICC) TNM classification system is the most prominent.1 We previously demonstrated that, among the components of TNM classification, clinically apparent lateral node metastasis (N1b), which is metastasis in the lateral compartment
Between 1987 and 1995, 1707 patients underwent initial surgical treatment with primary papillary carcinoma at Kuma Hospital, Japan. At total of 186 patients were diagnosed as poorly differentiated carcinoma, proposed by Sakamoto et al.5 (15 of these patients were also classified as having poorly differentiated carcinoma by the WHO classification6) and excluded from the study. Therefore, 1521 (1707 minus 186) patients were enrolled in this study. These patients consisted of 109 males and 1412 females and the mean patient age was 47.6 + 13.3 years. Post-operative follow-up periods ranged from 0.5 to 19 years (median follow-up time 12.9 years). Surgical designs of thyroidectomy and lymph node dissection are summarised in Table 1. Pathological examination Formalin fixed and paraffin embedded blocks were made from the maximal cut surface of each tumour. H&E sections from the blocks of 1521 cases were reviewed by one endocrine pathologist (MH). Specimens were
Print ISSN 0031-3025/Online ISSN 1465-3931 # 2008 Royal College of Pathologists of Australasia DOI: 10.1080/00313020802320630
618
Pathology (2008), 40(6), October
ITO et al.
classified into conventional papillary carcinoma and various histological variants: follicular, tall cell, oncocytic, columnar cell, macrofollicular, diffuse sclerosing, cribriform morular, clear cell, and nodular-fascilitis variants.
Post-operative evaluation of recurrence We normally follow the patients by ultrasonography with chest X-ray and/or computed tomography (CT) scan once or twice per year. In this study, we analysed patients who showed recurrence on imaging studies. To date, 170 patients (11.3%) have shown recurrence and the organs of recurrence are summarised in Table 2. Furthermore, 24 patients (1.6%) have died of thyroid carcinoma.
Statistical analyses Fisher’s exact test was used to compare variables. Kaplan–Meier method and log rank test were adopted to analyze time-dependent variables. Cox regression model was also adopted for multivariate analysis. These analyses were performed using StatView-J 5.0 (SAS Institute, USA). A p value less than 0.05 was regarded as significant.
RESULTS The H&E sections of 1521 papillary carcinomas were rereviewed by one endocrine pathologist and classified into conventional papillary carcinoma and various histological variants as shown in Table 3. Of 1521 patients, 1313 (86.3%) were diagnosed as having conventional papillary carcinoma (Fig. 1A). Follicular (Fig. 1B), tall cell (Fig. 1C) and oncocytic variants (Fig. 1D) were observed in 100 (6.6%), 60 (3.9%) and 29 patients (1.9%), respectively. Twenty-six of 29 patients with oncocytic variant were TABLE 1
Surgical designs of 1521 patients
Surgery Thyroidectomy Total Near total Subtotal Lobectomy or isthmectomy Lymph node dissection Bilateral MND Unilateral MND Central only
n
%
785 17 172 547
51.6 1.1 11.3 36.0
187 1138 196
12.3 74.8 12.8
MND, modified radical neck dissection.
TABLE 2
diagnosed as having Warthin-like tumour. These three variants were observed in comparably high incidences in our series. Furthermore, columnar cell (Fig. 1E), macrofollicular, diffuse sclerosing, cribriform morular, clear cell and nodular-fascilitis variants were found to be rare, accounting for less than 1%. We then compared the incidence of factors indicating aggressive biological characteristics such as male gender, advanced age, large tumour, massive extrathyroid extension (pT4a), distant metastasis at diagnosis (M1), and clinically apparent lateral node metastasis (N1b) as well as pathological lymph node metastasis (pN) between conventional papillary carcinoma and the three representative variants indicated above. Furthermore, we compared their differences in disease-free survival (DFS) (for M0 patients) and cause-specific survival (CSS). Patients having follicular variant were significantly older. Although the incidence of pN in patients with follicular variant was significantly lower than in patients with conventional papillary carcinoma, the incidences of N1b did not differ (Table 4). The DFS and CSS of follicular variant did not significantly differ from those of conventional papillary carcinoma (data not shown). Patients with tall cell variant were older and more frequently demonstrated pathological node metastasis than patients with conventional papillary carcinoma (Table 5). The incidences of N1b, M1 and pT4 were similar to those of conventional papillary carcinoma. However, as shown in Fig. 2A and B, DFS and CSS of patients with tall cell variant were significantly worse than those of conventional papillary carcinoma patients. There was no significant difference in the incidences of clinicopathological parameters between oncocytic variant and conventional papillary carcinoma (data not shown). However, none of the oncocytic variant showed distant metastasis at diagnosis. Two cases that were not Warthinlike were classified into N1b. However, there was no Warthin-like tumour classified as N1b and there was only one patient with Warthin-like who showed massive extrathyroid extension. None of the other 25 patients showed such extension. Regarding patient prognosis, there were no patients with oncocytic variant, including Warthinlike tumour, who showed recurrence or died of carcinoma to date (data not shown). Among the rare variants, columnar cell variant showed significantly worse DFS and CSS than conventional
Recurrence in 170 patients (11.3%)
Recurrence 1. Lymph node Previously dissected compartments Compartments that had not been dissected Both compartments 2. Locoregional organs Thyroid Others 3. Distant organs* Lung Bone Brain
n
%
123 54 58 11
8.1
33 18
2.2 1.2
25 2 1
1.6 0.1 0.06
TABLE 3 Prevalence of conventional and various histological variants in 1521 papillary carcinomas Papillary carcinomas
*24 patients died of carcinoma; 15 showed recurrence in two or more organs.
Conventional Follicular variant Tall cell variant Oncocytic variant Columnar cell variant Macrofollicular variant Diffuse sclerosing variant Cribriform morular variant Clear cell variant Nodular-fascilitis variant
No. patients
%
1313 100 60 29 6 5 4 2 1 1
86.3 6.6 3.9 1.9 0.4 0.3 0.3 0.1 0.06 0.06
PREVALENCE OF VARIANTS OF PAPILLARY THYROID CARCINOMA
619
FIG. 1 H&E staining of (A) conventional papillary carcinoma, (B) follicular variant, (C) tall cell variant, (D) Warthin-like tumour, and (E) columnar cell variant.
papillary carcinoma, as shown in Fig. 3A and B. Three of five patients with M0 showed recurrence and one patient with M1 died of carcinoma only 22 months after surgery. Regarding other rare variants, all patients but one with the diffuse sclerosing variant showed carcinoma recurrence but none of these patients have died of carcinoma to date.
DISCUSSION In this study, we retrospectively investigated the prevalence and biological characteristics of variants of papillary carcinoma at one institute. Similar trials have been published by other institutes,8,9 but with a smaller number of cases.
620
Pathology (2008), 40(6), October
ITO et al.
Our series consisted of 109 males and 1412 females, indicating that the female to male ratio is more than 10 to 1, which is definitely more than usual. In Japan, this is TABLE 4 Comparison in clinicopathological parameters between conventional papillary carcinoma and follicular variant Parameter Gender M/F Age (years) 455/55 Tumour size (cm) 44/4 pT pT4/pT1–3 M classification* M1/M0 N classification* N1b/N0 or 1a pN classification* pN1/pN0
Follicular
Conventional
7/93
97/1216
Total
p values
104/1309
NS
41/59
403/910
444/969
0.0437
6/94
158/1155
164/1249
NS
11/89
165/1148
176/1237
NS
3/97
15/1298
18/1395
NS
14/86
192/1121
206/1207
NS
60/40
1009/304
1069/344
0.0001
*According to UICC classification.1
TABLE 5 Comparison in clinicopathological parameters between conventional papillary carcinoma and tall cell variant Parameter Gender M/F Age (years) 455/55 Tumour size (cm) 44/4 pT pT4/pT1–3 M classification* M1/M0 N classification* N1b/N0 or 1a pN classification* pN1/pN0
Tall cell
Conventional
3/57
97/1218
Total
p values
100/1273
NS
29/31
403/910
432/941
0.0064
8/52
158/1155
165/1207
NS
11/49
165/1150
176/1197
NS
1/59
15/1297
16/1356
NS
8/52
192/1121
200/1173
NS
54/6
1009/304
*According to UICC classification.1
1063/310
0.0169
possibly because females are more likely to have thyroid carcinomas detected by ultrasonography of the thyroid performed simultaneously during ultrasonography of the breast during mass screening. Thus, at least in Japan, thyroid carcinoma in females may be detected more easily than that in males. In our series, there were three variants that showed comparably high prevalences: follicular, tall cell and oncocytic variants. In contrast to the findings of previous studies,10–12 we demonstrated that the prognoses of patients with follicular variant and conventional papillary carcinoma were similar. This may be reasonable because the incidences of clinicopathological features affecting prognosis such as N1b, pT4, and M1 did not differ between the follicular variant and conventional papillary carcinoma, except for patient age. The discrepancy in the prognosis of follicular variant patients between previous studies and our study may be due to some endemic factors. Indeed, in Japan, follicular carcinoma also showed better prognosis than that in most Western countries.13 Patients with tall cell variant showed significantly worse DFS and CSS than those with conventional papillary carcinoma, which is not discrepant with previous findings.14–17 Interestingly, the incidence of clinicopathological findings affecting prognosis such as pT4, M1 and N1b did not differ between tall cell variant and conventional papillary carcinoma. Therefore, it is suggested that tall cell variant shows a dire prognosis with a high independence. Our series included 29 cases of oncocytic variant, accounting for 1.9%. Of these, 26 were diagnosed as Warthin-like tumour, which shows abundant chronic inflammatory cells and is frequently associated with chronic thyroiditis.18 This incidence was significantly higher than that previously reported.9 Such a difference may be due to some endemic factors, but the reason remains to be elucidated. Although one group reported Warthin-like tumour showing dedifferentiation,19 previous reports showed that it has a mild character.20–23 Ludvikova et al. investigated 12 patients with Warthinlike tumour and showed that this variant has a mild
FIG. 2 (A) Comparison of DFS between patients with tall cell variant and conventional papillary carcinoma. (B) Comparison of CSS between patients with tall cell variant and conventional papillary carcinoma.
PREVALENCE OF VARIANTS OF PAPILLARY THYROID CARCINOMA
621
FIG. 3 (A) Comparison of DFS between patients with columnar cell variant and conventional papillary carcinoma. (B) Comparison of CSS between patients with columnar cell variant and conventional papillary carcinoma.
character.23 However, the average follow-up period in their series was rather short and they did not show its prevalence or compare its biological aggressiveness with that of conventional papillary carcinoma. In this study, we demonstrated that there were no Warthin-like tumours showing clinically apparent lateral node metastasis, although two of three oncocytic variants that were not Warthin-like demonstrated such metastases. In addition, only one of the Warthin-like tumours showed massive extrathyroid extension, while none of the others did. Regarding patient prognosis, none of the patients with Warthin-like tumour have died of carcinoma or even shown recurrence to date. Therefore, it is suggested that Warthin-like tumour is very indolent and shows an excellent prognosis compared with conventional papillary carcinoma. Columnar cell variant accounted for 0.4% of our series, which is similar to the incidence demonstrated in a previous study.24 Among the rare variants, only this carcinoma showed a dire prognosis in our series, with three of five (60.0%) M0 patients showing recurrence. Interestingly, only one patient was classified as having N1b and pT4, while four were classified as pN0. Therefore, similar to tall cell variant, columnar carcinoma has a biologically aggressive behaviour independently, not because of the frequency of aggressive clinicopathological findings. This finding is in partial contrast to that of a previous study describing 16 cases. Further studies involving a larger number of cases and long-term follow-up are necessary to elucidate this issue. Previous studies investigating the patient prognosis of diffuse sclerosing variant have shown controversial findings. Chow et al. and Lam et al. showed that diffuse sclerosing variant has a prognosis similar to that of conventional papillary carcinoma,25,26 while Falvo et al. indicated a poor prognosis of this variant.27 Macrofollicular and cribriform morular variants were reported to have an indolent character.28–31 In this study, there was no evidence that any of the rare variants except for columnar cell carcinoma have an aggressive behaviour because all patients except for one with the diffuse sclerosing variant
showed carcinoma recurrence but none of the patients have died of carcinoma to date. In this study, we presented the experience of a single institute regarding the prevalence and biological behaviour of variants of papillary carcinoma. Since patients with some variants have shown different outcomes than those with conventional papillary carcinoma, classification of variants may be useful to predict prognosis of patients with papillary carcinoma. Address for correspondence: Dr Y. Ito, Department of Surgery, Kuma Hospital, 8-2-35, Shimoyamate-Dori, Chuo-ku, Kobe City, 650-0011 Japan. E-mail: [email protected]
References 1. Sobin LH, Wittekind CH, editors. UICC: TNM Classification of Malignant Tumors. 6th ed. New York: Wiley-Liss, 2002. 2. Ito Y, Miyauchi A, Jikuzono T, et al. Risk factors contributing to a poor prognosis of papillary thyroid carcinoma; Validity of UICC/ AJCC TNM classification and stage grouping. World J Surg 2007; 31: 838–48. 3. Ito Y, Hirokawa M, Jikuzono T, et al. Extranodal tumor extension to adjacent organs predicts a worse cause-specific survival in patients with papillary thyroid carcinoma. World J Surg 2007; 31: 1196–203. 4. Carcangiu ML, Zampi G, Rosai J. Poorly differentiated (‘insular’) thyroid carcinoma. A reinterpretation of Langhans’ ‘Wuchernde Struma’. Am J Surg Pathol 1984; 8: 655–68. 5. Sakamoto A, Kasai N, Sugano H. Poorly differentiated carcinoma of the thyroid. A clinicopathologic entity for a high risk group of papillary and follicular carcinomas. Cancer 1983; 52: 1849–55. 6. Sobrinho-Simoes M, Carcangiu ML, Albores-Saavedra J, et al. Poorly differentiated carcinoma. In: DeLeillis RA, Lloyd RV, Heitz PU, et al., editors. Pathology and Genetics of Tumous of Endocrine Organs. Lyon: IARC Press, 2004; 73–6. 7. LiVolsi VA, Mazzaferri EL, Albores-Saavedra J, et al. Papillary carcinoma. In: DeLeillis RA, Lloyd RV, Heitz PU, et al., editors. Pathology and Genetics of Tumous of Endocrine Organs. Lyon: IARC Press, 2004; 57–66. 8. Khan AR, Abu-Eshy SA. Variants of papillary carcinoma of the thyroid: experience at Asir Central Hospital. J R Coll Surg Edinb 1998; 43: 20–5. 9. Lam AK, Lo CY, Lam KS. Papillary carcinoma of thyroid: a 30-yr clinicopathological review of the histologic variants. Endocr Pathol 2005; 16: 323–30. 10. Chag HY, Lin JD, Chou SC, et al. Clinical presentations and outcomes of surgical treatment of follicular variant of the papillary thyroid carcinomas. Jpn J Clin Oncol 2006; 36: 688–93.
622
ITO et al.
11. Hagag P, Hod N, Kummer E, et al. Follicular variant of papillary thyroid carcinoma: clinical-pathological characterization and longterm follow-up. Cancer J 2006; 12: 275–82. 12. Liu J, Singh B, Tallini G, et al. Follicular variant of papillary thyroid carcinoma: a clinicopathologic study of a problematic entity. Cancer 2006; 107: 1255–64. 13. Ito Y, Hirokawa M, Higashiyama T, et al. Prognosis and prognostic factors of follicular carcinoma: Importance of postoperative pathological examination. World J Surg 2007; 31: 1417–24. 14. Johnson TL, Lloyd RV, Thompson NW, et al. Prognostic implications of the tall cell variant of papillary thyroid carcinoma. Am J Surg Pathol 1988; 12: 22–7. 15. Cortese F, Boffo V, Gargiani M, et al. Tall cell variant of papillary thyroid cancer. J Exp Clin Cancer Res 1998; 17: 523–6. 16. Egea AM, Gonzalez JMR, Perez JS, et al. Prognostic value of the tall cell variety of papillary cancer of the thyroid. Eur J Surg Oncol 1993; 19: 517–21. 17. Michels JJ, Jacques M, Henry-Amar M, et al. Prevalence and prognostic significance of tall cell variant of papillary thyroid carcinoma. Hum Pathol 2007; 38: 212–9. 18. Apel RL, Asa SL, LiVolsi VA. Papillary Hurthle cell carcinoma with lymphocytic stroma. ‘Warthin-like tumor’ of the thyroid. Am J Surg Pathol 1995; 19: 810–4. 19. Lam KY, Lo CY, Wei WI. Warthin tumor-like variant of papillary thyroid carcinoma: a case with dedifferentiation (anaplastic change) and aggressive biological behavior. Endocr Pathol 2005; 16: 83–9. 20. Urano M, Abe M, Kuroda M, et al. Warthin-like tumor variant of papillary thyroid carcinoma: case report and literature review. Pathol Int 2001; 51: 707–12. 21. Baloch ZW, LiVolsi VA. Warthin-like papillary carcinoma of the thyroid. Arch Pathol Lab Med 2000; 124: 1192–5.
Pathology (2008), 40(6), October
22. Yousef O, Dichard A, Bockalge T. Aspiration cytology features of the warthin tumor-like variant of papillary thyroid carcinoma. A report of two cases. Acta Cytol 1997; 41: 1361–8. 23. Ludvikova M, Ryska A, Korabecna M, et al. Oncocytic papillary carcinoma with lymphoid stroma (Warthin-like tumour) of the thyroid: a distinct entity with favourable prognosis. Histopathology 2001; 39: 17–24. 24. Wenig BM, Thomson LDR, Adair CF, et al. Thyroid papillary carcinoma of columnar cell type; a clinicopathologic study of 16 cases. Cancer 1998; 82: 40–53. 25. Chow SM, Chan JK, Law SC, et al. Diffuse sclerosing variant of papillary thyroid carcinoma—clinical features and outcome. Eur J Surg Oncol 2003; 29: 446–9. 26. Lam AK, Lo CY. Diffuse sclerosing variant of papillary carcinoma of the thyroid: a 35-year comparative study at a single institution. Ann Surg Oncol 2006; 13: 176–81. 27. Falvo L, Giacomelli L, D’Andrea V, et al. Prognostic importance of sclerosing variant in papillary thyroid carcinoma. Am Surg 2006; 75: 438–44. 28. Lugli A, Terracciano LM, Oberholzer M, et al. Macrofollicular variant of papillary carcinoma of the thyroid: a histologic, cytologic, and immunohistochemical study of 3 cases and review of the literature. Arch Pathol Lab Med 2004; 128: 54–8. 29. Albores-Saavedra J, Gould E, Vardaman C, et al. The macrofollicular variant of papillary thyroid carcinoma: A study of 17 cases. Hum Pathol 1991; 22: 1195–205. 30. Tomoda C, Miyauchi A, Uruno T, et al. Cribriform-morular variant of papillary thyroid carcinoma: clue to early detection of familial adenomatous polyposis-associated colon cancer. World J Surg 2004; 28: 886–9. 31. Dalal KM, Moraitis D, Iwamoto C, et al. Clinical curiosity: cribriformmorular variant of papillary thyroid carcinoma. Head Neck 2006; 28: 471–6.
ANATOMICAL PATHOLOGY
Prevalence and biological behaviour of variants of papillary thyroid carcinoma: experience at a single institute YASUHIRO ITO*, MITSUYOSHI HIROKAWA{, TAKASHI URUNO*, MINORU KIHARA*, TAKUYA HIGASHIYAMA*, YUUKI TAKAMURA*, AKIHIRO MIYA*, KAORU KOBAYASHI*, FUMIO MATSUZUKA* AND AKIRA MIYAUCHI* Departments of *Surgery and {Pathology, Kuma Hospital, Kobe City, Japan
Summary Aims: There are many variants of papillary carcinoma, and some of these variants have been reported to show biological behaviours differing from that of conventional papillary carcinoma. In this study, we present our experience regarding the prevalence and prognoses of these variants of papillary carcinoma. Methods: H&E sections from 1521 patients who underwent initial surgery for papillary carcinoma in Kuma Hospital between 1987 and 1995 were re-reviewed and classified into conventional papillary carcinoma and various histological variants. We investigated the biological behaviours of these lesions, including prognoses. Results: Follicular, tall cell and oncocytic variants were observed in comparably high incidences: 6.6%, 3.9%, and 1.9%, respectively. Patients with tall cell variant showed significantly worse disease-free survival (DFS) and causespecific survival (CSS) rates than those with conventional papillary carcinoma. The prognoses of patients with follicular variant did not differ from those of patients with conventional papillary carcinoma. Patients with oncocytic variant have not shown carcinoma recurrence. Among the rare variants, which accounted for less than 1%, columnar cell carcinoma showed a worse prognosis. Conclusions: Since patients with some variants show different clinical outcomes from those with conventional papillary carcinoma, classification of variants might be helpful to predict patient prognosis.
preoperatively detected by imaging studies, independently affects not only disease-free survival (DFS) but also causespecific survival (CFS) of patients, while pathologically diagnosed node metastasis (pN) that was only microscopically detected does not significantly affect outcome.2 Furthermore, clinical and pathological massive extrathyroid extension (T4 and pT4) and distant metastasis at diagnosis (M1) independently predicted poor prognosis in patients.2,3 Furthermore, some histological findings have been reported to affect patient prognosis. Previous studies demonstrated that poorly differentiated carcinoma showed a poor clinical outcome in patients4,5 and this was adopted by the World Health Organization (WHO) classification as an independent entity.6 Furthermore, there are many other histological variants in papillary carcinoma as shown in the most recent edition of the WHO classification.7 Although studies on the prognosis of some variants have been performed, our knowledge about the prevalence and biological behaviour of many variants remains unclear because of the lack of studies involving a large number of patients. In this study, we investigated the prevalence of these histological variants using a series of patients who underwent surgery in our institute, and compared their biological behaviours, including patient prognoses, with those of conventional papillary carcinoma.
Key words: Papillary carcinoma, thyroid, variants, prognosis.
PATIENTS AND METHODS
Received 6 July, revised 12 and 24 September, accepted 3 October 2007
Patients
INTRODUCTION Papillary carcinoma is the most common malignancy originating from the thyroid. Generally, it is less virulent and shows a good prognosis if competently resected. However, tumours showing certain clinicopathological parameters often grow constantly and show a worse clinical outcome. In order to evaluate these parameters, the International Union Against Cancer (UICC) TNM classification system is the most prominent.1 We previously demonstrated that, among the components of TNM classification, clinically apparent lateral node metastasis (N1b), which is metastasis in the lateral compartment
Between 1987 and 1995, 1707 patients underwent initial surgical treatment with primary papillary carcinoma at Kuma Hospital, Japan. At total of 186 patients were diagnosed as poorly differentiated carcinoma, proposed by Sakamoto et al.5 (15 of these patients were also classified as having poorly differentiated carcinoma by the WHO classification6) and excluded from the study. Therefore, 1521 (1707 minus 186) patients were enrolled in this study. These patients consisted of 109 males and 1412 females and the mean patient age was 47.6 + 13.3 years. Post-operative follow-up periods ranged from 0.5 to 19 years (median follow-up time 12.9 years). Surgical designs of thyroidectomy and lymph node dissection are summarised in Table 1. Pathological examination Formalin fixed and paraffin embedded blocks were made from the maximal cut surface of each tumour. H&E sections from the blocks of 1521 cases were reviewed by one endocrine pathologist (MH). Specimens were
Print ISSN 0031-3025/Online ISSN 1465-3931 # 2008 Royal College of Pathologists of Australasia DOI: 10.1080/00313020802320630
618
Pathology (2008), 40(6), October
ITO et al.
classified into conventional papillary carcinoma and various histological variants: follicular, tall cell, oncocytic, columnar cell, macrofollicular, diffuse sclerosing, cribriform morular, clear cell, and nodular-fascilitis variants.
Post-operative evaluation of recurrence We normally follow the patients by ultrasonography with chest X-ray and/or computed tomography (CT) scan once or twice per year. In this study, we analysed patients who showed recurrence on imaging studies. To date, 170 patients (11.3%) have shown recurrence and the organs of recurrence are summarised in Table 2. Furthermore, 24 patients (1.6%) have died of thyroid carcinoma.
Statistical analyses Fisher’s exact test was used to compare variables. Kaplan–Meier method and log rank test were adopted to analyze time-dependent variables. Cox regression model was also adopted for multivariate analysis. These analyses were performed using StatView-J 5.0 (SAS Institute, USA). A p value less than 0.05 was regarded as significant.
RESULTS The H&E sections of 1521 papillary carcinomas were rereviewed by one endocrine pathologist and classified into conventional papillary carcinoma and various histological variants as shown in Table 3. Of 1521 patients, 1313 (86.3%) were diagnosed as having conventional papillary carcinoma (Fig. 1A). Follicular (Fig. 1B), tall cell (Fig. 1C) and oncocytic variants (Fig. 1D) were observed in 100 (6.6%), 60 (3.9%) and 29 patients (1.9%), respectively. Twenty-six of 29 patients with oncocytic variant were TABLE 1
Surgical designs of 1521 patients
Surgery Thyroidectomy Total Near total Subtotal Lobectomy or isthmectomy Lymph node dissection Bilateral MND Unilateral MND Central only
n
%
785 17 172 547
51.6 1.1 11.3 36.0
187 1138 196
12.3 74.8 12.8
MND, modified radical neck dissection.
TABLE 2
diagnosed as having Warthin-like tumour. These three variants were observed in comparably high incidences in our series. Furthermore, columnar cell (Fig. 1E), macrofollicular, diffuse sclerosing, cribriform morular, clear cell and nodular-fascilitis variants were found to be rare, accounting for less than 1%. We then compared the incidence of factors indicating aggressive biological characteristics such as male gender, advanced age, large tumour, massive extrathyroid extension (pT4a), distant metastasis at diagnosis (M1), and clinically apparent lateral node metastasis (N1b) as well as pathological lymph node metastasis (pN) between conventional papillary carcinoma and the three representative variants indicated above. Furthermore, we compared their differences in disease-free survival (DFS) (for M0 patients) and cause-specific survival (CSS). Patients having follicular variant were significantly older. Although the incidence of pN in patients with follicular variant was significantly lower than in patients with conventional papillary carcinoma, the incidences of N1b did not differ (Table 4). The DFS and CSS of follicular variant did not significantly differ from those of conventional papillary carcinoma (data not shown). Patients with tall cell variant were older and more frequently demonstrated pathological node metastasis than patients with conventional papillary carcinoma (Table 5). The incidences of N1b, M1 and pT4 were similar to those of conventional papillary carcinoma. However, as shown in Fig. 2A and B, DFS and CSS of patients with tall cell variant were significantly worse than those of conventional papillary carcinoma patients. There was no significant difference in the incidences of clinicopathological parameters between oncocytic variant and conventional papillary carcinoma (data not shown). However, none of the oncocytic variant showed distant metastasis at diagnosis. Two cases that were not Warthinlike were classified into N1b. However, there was no Warthin-like tumour classified as N1b and there was only one patient with Warthin-like who showed massive extrathyroid extension. None of the other 25 patients showed such extension. Regarding patient prognosis, there were no patients with oncocytic variant, including Warthinlike tumour, who showed recurrence or died of carcinoma to date (data not shown). Among the rare variants, columnar cell variant showed significantly worse DFS and CSS than conventional
Recurrence in 170 patients (11.3%)
Recurrence 1. Lymph node Previously dissected compartments Compartments that had not been dissected Both compartments 2. Locoregional organs Thyroid Others 3. Distant organs* Lung Bone Brain
n
%
123 54 58 11
8.1
33 18
2.2 1.2
25 2 1
1.6 0.1 0.06
TABLE 3 Prevalence of conventional and various histological variants in 1521 papillary carcinomas Papillary carcinomas
*24 patients died of carcinoma; 15 showed recurrence in two or more organs.
Conventional Follicular variant Tall cell variant Oncocytic variant Columnar cell variant Macrofollicular variant Diffuse sclerosing variant Cribriform morular variant Clear cell variant Nodular-fascilitis variant
No. patients
%
1313 100 60 29 6 5 4 2 1 1
86.3 6.6 3.9 1.9 0.4 0.3 0.3 0.1 0.06 0.06
PREVALENCE OF VARIANTS OF PAPILLARY THYROID CARCINOMA
619
FIG. 1 H&E staining of (A) conventional papillary carcinoma, (B) follicular variant, (C) tall cell variant, (D) Warthin-like tumour, and (E) columnar cell variant.
papillary carcinoma, as shown in Fig. 3A and B. Three of five patients with M0 showed recurrence and one patient with M1 died of carcinoma only 22 months after surgery. Regarding other rare variants, all patients but one with the diffuse sclerosing variant showed carcinoma recurrence but none of these patients have died of carcinoma to date.
DISCUSSION In this study, we retrospectively investigated the prevalence and biological characteristics of variants of papillary carcinoma at one institute. Similar trials have been published by other institutes,8,9 but with a smaller number of cases.
620
Pathology (2008), 40(6), October
ITO et al.
Our series consisted of 109 males and 1412 females, indicating that the female to male ratio is more than 10 to 1, which is definitely more than usual. In Japan, this is TABLE 4 Comparison in clinicopathological parameters between conventional papillary carcinoma and follicular variant Parameter Gender M/F Age (years) 455/55 Tumour size (cm) 44/4 pT pT4/pT1–3 M classification* M1/M0 N classification* N1b/N0 or 1a pN classification* pN1/pN0
Follicular
Conventional
7/93
97/1216
Total
p values
104/1309
NS
41/59
403/910
444/969
0.0437
6/94
158/1155
164/1249
NS
11/89
165/1148
176/1237
NS
3/97
15/1298
18/1395
NS
14/86
192/1121
206/1207
NS
60/40
1009/304
1069/344
0.0001
*According to UICC classification.1
TABLE 5 Comparison in clinicopathological parameters between conventional papillary carcinoma and tall cell variant Parameter Gender M/F Age (years) 455/55 Tumour size (cm) 44/4 pT pT4/pT1–3 M classification* M1/M0 N classification* N1b/N0 or 1a pN classification* pN1/pN0
Tall cell
Conventional
3/57
97/1218
Total
p values
100/1273
NS
29/31
403/910
432/941
0.0064
8/52
158/1155
165/1207
NS
11/49
165/1150
176/1197
NS
1/59
15/1297
16/1356
NS
8/52
192/1121
200/1173
NS
54/6
1009/304
*According to UICC classification.1
1063/310
0.0169
possibly because females are more likely to have thyroid carcinomas detected by ultrasonography of the thyroid performed simultaneously during ultrasonography of the breast during mass screening. Thus, at least in Japan, thyroid carcinoma in females may be detected more easily than that in males. In our series, there were three variants that showed comparably high prevalences: follicular, tall cell and oncocytic variants. In contrast to the findings of previous studies,10–12 we demonstrated that the prognoses of patients with follicular variant and conventional papillary carcinoma were similar. This may be reasonable because the incidences of clinicopathological features affecting prognosis such as N1b, pT4, and M1 did not differ between the follicular variant and conventional papillary carcinoma, except for patient age. The discrepancy in the prognosis of follicular variant patients between previous studies and our study may be due to some endemic factors. Indeed, in Japan, follicular carcinoma also showed better prognosis than that in most Western countries.13 Patients with tall cell variant showed significantly worse DFS and CSS than those with conventional papillary carcinoma, which is not discrepant with previous findings.14–17 Interestingly, the incidence of clinicopathological findings affecting prognosis such as pT4, M1 and N1b did not differ between tall cell variant and conventional papillary carcinoma. Therefore, it is suggested that tall cell variant shows a dire prognosis with a high independence. Our series included 29 cases of oncocytic variant, accounting for 1.9%. Of these, 26 were diagnosed as Warthin-like tumour, which shows abundant chronic inflammatory cells and is frequently associated with chronic thyroiditis.18 This incidence was significantly higher than that previously reported.9 Such a difference may be due to some endemic factors, but the reason remains to be elucidated. Although one group reported Warthin-like tumour showing dedifferentiation,19 previous reports showed that it has a mild character.20–23 Ludvikova et al. investigated 12 patients with Warthinlike tumour and showed that this variant has a mild
FIG. 2 (A) Comparison of DFS between patients with tall cell variant and conventional papillary carcinoma. (B) Comparison of CSS between patients with tall cell variant and conventional papillary carcinoma.
PREVALENCE OF VARIANTS OF PAPILLARY THYROID CARCINOMA
621
FIG. 3 (A) Comparison of DFS between patients with columnar cell variant and conventional papillary carcinoma. (B) Comparison of CSS between patients with columnar cell variant and conventional papillary carcinoma.
character.23 However, the average follow-up period in their series was rather short and they did not show its prevalence or compare its biological aggressiveness with that of conventional papillary carcinoma. In this study, we demonstrated that there were no Warthin-like tumours showing clinically apparent lateral node metastasis, although two of three oncocytic variants that were not Warthin-like demonstrated such metastases. In addition, only one of the Warthin-like tumours showed massive extrathyroid extension, while none of the others did. Regarding patient prognosis, none of the patients with Warthin-like tumour have died of carcinoma or even shown recurrence to date. Therefore, it is suggested that Warthin-like tumour is very indolent and shows an excellent prognosis compared with conventional papillary carcinoma. Columnar cell variant accounted for 0.4% of our series, which is similar to the incidence demonstrated in a previous study.24 Among the rare variants, only this carcinoma showed a dire prognosis in our series, with three of five (60.0%) M0 patients showing recurrence. Interestingly, only one patient was classified as having N1b and pT4, while four were classified as pN0. Therefore, similar to tall cell variant, columnar carcinoma has a biologically aggressive behaviour independently, not because of the frequency of aggressive clinicopathological findings. This finding is in partial contrast to that of a previous study describing 16 cases. Further studies involving a larger number of cases and long-term follow-up are necessary to elucidate this issue. Previous studies investigating the patient prognosis of diffuse sclerosing variant have shown controversial findings. Chow et al. and Lam et al. showed that diffuse sclerosing variant has a prognosis similar to that of conventional papillary carcinoma,25,26 while Falvo et al. indicated a poor prognosis of this variant.27 Macrofollicular and cribriform morular variants were reported to have an indolent character.28–31 In this study, there was no evidence that any of the rare variants except for columnar cell carcinoma have an aggressive behaviour because all patients except for one with the diffuse sclerosing variant
showed carcinoma recurrence but none of the patients have died of carcinoma to date. In this study, we presented the experience of a single institute regarding the prevalence and biological behaviour of variants of papillary carcinoma. Since patients with some variants have shown different outcomes than those with conventional papillary carcinoma, classification of variants may be useful to predict prognosis of patients with papillary carcinoma. Address for correspondence: Dr Y. Ito, Department of Surgery, Kuma Hospital, 8-2-35, Shimoyamate-Dori, Chuo-ku, Kobe City, 650-0011 Japan. E-mail: [email protected]
References 1. Sobin LH, Wittekind CH, editors. UICC: TNM Classification of Malignant Tumors. 6th ed. New York: Wiley-Liss, 2002. 2. Ito Y, Miyauchi A, Jikuzono T, et al. Risk factors contributing to a poor prognosis of papillary thyroid carcinoma; Validity of UICC/ AJCC TNM classification and stage grouping. World J Surg 2007; 31: 838–48. 3. Ito Y, Hirokawa M, Jikuzono T, et al. Extranodal tumor extension to adjacent organs predicts a worse cause-specific survival in patients with papillary thyroid carcinoma. World J Surg 2007; 31: 1196–203. 4. Carcangiu ML, Zampi G, Rosai J. Poorly differentiated (‘insular’) thyroid carcinoma. A reinterpretation of Langhans’ ‘Wuchernde Struma’. Am J Surg Pathol 1984; 8: 655–68. 5. Sakamoto A, Kasai N, Sugano H. Poorly differentiated carcinoma of the thyroid. A clinicopathologic entity for a high risk group of papillary and follicular carcinomas. Cancer 1983; 52: 1849–55. 6. Sobrinho-Simoes M, Carcangiu ML, Albores-Saavedra J, et al. Poorly differentiated carcinoma. In: DeLeillis RA, Lloyd RV, Heitz PU, et al., editors. Pathology and Genetics of Tumous of Endocrine Organs. Lyon: IARC Press, 2004; 73–6. 7. LiVolsi VA, Mazzaferri EL, Albores-Saavedra J, et al. Papillary carcinoma. In: DeLeillis RA, Lloyd RV, Heitz PU, et al., editors. Pathology and Genetics of Tumous of Endocrine Organs. Lyon: IARC Press, 2004; 57–66. 8. Khan AR, Abu-Eshy SA. Variants of papillary carcinoma of the thyroid: experience at Asir Central Hospital. J R Coll Surg Edinb 1998; 43: 20–5. 9. Lam AK, Lo CY, Lam KS. Papillary carcinoma of thyroid: a 30-yr clinicopathological review of the histologic variants. Endocr Pathol 2005; 16: 323–30. 10. Chag HY, Lin JD, Chou SC, et al. Clinical presentations and outcomes of surgical treatment of follicular variant of the papillary thyroid carcinomas. Jpn J Clin Oncol 2006; 36: 688–93.
622
ITO et al.
11. Hagag P, Hod N, Kummer E, et al. Follicular variant of papillary thyroid carcinoma: clinical-pathological characterization and longterm follow-up. Cancer J 2006; 12: 275–82. 12. Liu J, Singh B, Tallini G, et al. Follicular variant of papillary thyroid carcinoma: a clinicopathologic study of a problematic entity. Cancer 2006; 107: 1255–64. 13. Ito Y, Hirokawa M, Higashiyama T, et al. Prognosis and prognostic factors of follicular carcinoma: Importance of postoperative pathological examination. World J Surg 2007; 31: 1417–24. 14. Johnson TL, Lloyd RV, Thompson NW, et al. Prognostic implications of the tall cell variant of papillary thyroid carcinoma. Am J Surg Pathol 1988; 12: 22–7. 15. Cortese F, Boffo V, Gargiani M, et al. Tall cell variant of papillary thyroid cancer. J Exp Clin Cancer Res 1998; 17: 523–6. 16. Egea AM, Gonzalez JMR, Perez JS, et al. Prognostic value of the tall cell variety of papillary cancer of the thyroid. Eur J Surg Oncol 1993; 19: 517–21. 17. Michels JJ, Jacques M, Henry-Amar M, et al. Prevalence and prognostic significance of tall cell variant of papillary thyroid carcinoma. Hum Pathol 2007; 38: 212–9. 18. Apel RL, Asa SL, LiVolsi VA. Papillary Hurthle cell carcinoma with lymphocytic stroma. ‘Warthin-like tumor’ of the thyroid. Am J Surg Pathol 1995; 19: 810–4. 19. Lam KY, Lo CY, Wei WI. Warthin tumor-like variant of papillary thyroid carcinoma: a case with dedifferentiation (anaplastic change) and aggressive biological behavior. Endocr Pathol 2005; 16: 83–9. 20. Urano M, Abe M, Kuroda M, et al. Warthin-like tumor variant of papillary thyroid carcinoma: case report and literature review. Pathol Int 2001; 51: 707–12. 21. Baloch ZW, LiVolsi VA. Warthin-like papillary carcinoma of the thyroid. Arch Pathol Lab Med 2000; 124: 1192–5.
Pathology (2008), 40(6), October
22. Yousef O, Dichard A, Bockalge T. Aspiration cytology features of the warthin tumor-like variant of papillary thyroid carcinoma. A report of two cases. Acta Cytol 1997; 41: 1361–8. 23. Ludvikova M, Ryska A, Korabecna M, et al. Oncocytic papillary carcinoma with lymphoid stroma (Warthin-like tumour) of the thyroid: a distinct entity with favourable prognosis. Histopathology 2001; 39: 17–24. 24. Wenig BM, Thomson LDR, Adair CF, et al. Thyroid papillary carcinoma of columnar cell type; a clinicopathologic study of 16 cases. Cancer 1998; 82: 40–53. 25. Chow SM, Chan JK, Law SC, et al. Diffuse sclerosing variant of papillary thyroid carcinoma—clinical features and outcome. Eur J Surg Oncol 2003; 29: 446–9. 26. Lam AK, Lo CY. Diffuse sclerosing variant of papillary carcinoma of the thyroid: a 35-year comparative study at a single institution. Ann Surg Oncol 2006; 13: 176–81. 27. Falvo L, Giacomelli L, D’Andrea V, et al. Prognostic importance of sclerosing variant in papillary thyroid carcinoma. Am Surg 2006; 75: 438–44. 28. Lugli A, Terracciano LM, Oberholzer M, et al. Macrofollicular variant of papillary carcinoma of the thyroid: a histologic, cytologic, and immunohistochemical study of 3 cases and review of the literature. Arch Pathol Lab Med 2004; 128: 54–8. 29. Albores-Saavedra J, Gould E, Vardaman C, et al. The macrofollicular variant of papillary thyroid carcinoma: A study of 17 cases. Hum Pathol 1991; 22: 1195–205. 30. Tomoda C, Miyauchi A, Uruno T, et al. Cribriform-morular variant of papillary thyroid carcinoma: clue to early detection of familial adenomatous polyposis-associated colon cancer. World J Surg 2004; 28: 886–9. 31. Dalal KM, Moraitis D, Iwamoto C, et al. Clinical curiosity: cribriformmorular variant of papillary thyroid carcinoma. Head Neck 2006; 28: 471–6.