Asian Journal of Surgery Excerpta Medica Asia Ltd Prognostic Factors for Well-Differentiated Thyroid Cancer in an Endemic Area A.K. Sarda, Shweta Aggarwal, Durgatosh Pandey, Gagan Gautam, Department of General Surgery, Maulana Azad Medical College and Lok Nayak Hospital, New Delhi, India. A retrospective analysis of 215 differentiated thyroid cancers was undertaken to identify the various prognostic variables. There were 132 papillary and 83 follicular cancers; both histologic types occurred at an earlier age, with a male to female ratio of 1: 1.1 (signifies near parity in both sexes, in contrast to marked female preponderance in most of the reports and amongst benign thyroid disorders). There was a significant difference in the size of the primary tumours; 60/132 (45%) of papillary and 30/83 (36%) of follicular cancers were early T 0-1 lesions and 20 (15%) papillary and 24 (29%) follicular cancers were advanced T 3 lesions. Age did not affect the size of the primary tumour. Regional lymph node and pulmonary metastases were common in both types of cancer while distant metastases occurred more frequently in follicular cancers (p < 0.005). Following surgery, contralateral lobe recurrence in the remaining lobe was more common in the follicular cancer group, while loco-regional recurrence after near-total was more frequent in the papillary cancer group (p > 0.05). Mortality in 26/132 (20%) papillary and 28/83 (34%) follicular cancer patients was high in both groups, but significantly higher in the follicular cancer group (p < 0.05). Although mortality among patients with papillary cancer was higher in patients older than 40 years of age (p < 0.05), age did not affect survival in patients with follicular cancer. Gender did not affect survival in either group. The extent of the disease at presentation was the most important determinant of survival, with mortality significantly higher among patients with T 3 lesions (p < 0.001). (Asian J Surg 2002;25(4):325 9) INTRODUCTION The behaviour of differentiated thyroid cancers in endemic areas is reportedly different from that in nonendemic areas. In endemic areas, not only is there a preponderance of follicular and anaplastic cancers, both recognized for their virulent nature, even papillary cancers reportedly have more aggressive behaviour. 1,2 We conducted a retrospective study of 289 thyroid cancer cases from an area endemic for goitres in an effort to identify the various prognostic factors for differentiated thyroid cancers. Address reprint requests to Dr. A. K. Sarda, 27 RPS, Triveni, New Delhi-110017, India. Date of acceptance: 30 th July 2002 MATERIALS AND METHODS A total of 1,576 patients with thyroid disorders underwent surgery in a general hospital unit between 1975 and 2000. Of these, 289 thyroid cancers were analyzed for various prognostic indices. Of the 289 thyroid cancers analyzed, there were 215 differentiated thyroid cancers, which were histologically classified according to the World Health Organization (WHO) classification. 1 Thus, there were 132 papillary cancers (63 in males and 69 in females) and 83 follicular cancers (35 in males and 48 in females). The age at diagnosis ranged from 12 to 75 years in the papillary cancer group, with a mean of 37.5 years. In the follicular cancer group, the age at diagnosis ranged from 4 to 75 years, with a mean of 43.2 years (Figure 1). The duration of symptoms at presentation was short in both histologic types with a mean of 27.9 months among patients with papillary cancer and 30.1 months among those with follicular cancer. Overt cancers, i.e., tumours Asian Journal of Surgery 325
SARDA AND OTHERS No. of patients (%) 100 80 60 40 20 Figure 1. Age distribution. Papillary Follicular < 19 20 29 30 39 40 49 50 59 > 60 Age at chagnosis (years) with local spread and/or distant metastases, were present in 63 papillary and 44 follicular cancer patients. Papillary cancers in 48 patients and follicular cancers in 21 occurred as clinically benign, solitary thyroid nodules. Eighteen papillary and 17 follicular cancers presented as clinically benign, euthyroid multinodular goitres. Four cancers were encountered in patients who had undergone operations for Grave s disease; of these, three were papillary and one was a follicular cancer. Assessment of the extent of the disease was made by clinical examination, radiological investigation, radionuclide thyroid scan (performed routinely until 1988), radioimmunoassay of serum T 4, T 3 and TSH, and fine-needle aspiration cytology (FNAC) of the primary tumour and of the metastatic deposit (whenever possible). Categorizing of the tumours was done according to the TNM staging system (Figure 2). 1 Thus, there were 60 patients with papillary cancersof which 38 were < 40 years of age. There were 30 patients with follicular cancers who presented at this early stage of the disease, with 17 of them being < 40 years of age. There were 34 patients with papillary and eight with follicular cancers at stage N 1 2, of which 24 patients with papillary and three with follicular cancers were less than 40 years of age. Advanced stage (T 3 ) cancer was encountered among 38 patients with papillary and 45 with follicular cancers, of which 16 patients with papillary and 14 with follicular cancer were less than 40 years of age. Anaplastic Distant metastasis at presentation were found in 17 papillary and 29 follicular cancers. Patients with pulmonary metastasis were typically asymptomatic at the time of detection of the lung of involvement. Pulmonary involvement was found in 11 papillary and 16 follicular cancers. Skeletal metastases were usually associated with bony pain and/or pathological fracture and occurred in five papillary and 26 follicular cancers. Surgery for the primary tumour aimed at near-total while a hemi was performed in patients who did not have a definite preoperative diagnosis and who refused a completion at a second stage (Figure 3). The patients follow-up ranged from 3 to 20 years (average, 12.6 years). All patients were treated with suppressive doses of thyroid hormone. In patients with complete remission after the initial treatment of the tumour, clinical examination, chest x-ray, and serum TSH and thyroglobulin estimations were performed at 6, 12, 18 Figure 2. Papillary < 40y 40y Tumour Size T 0 1 T 2 T 3 Nodal Involvement N 1 N 2 3 Metastases Clinical presentation. Follicular < 40y 40y 326 Vol 25 No 4 October 2002
Figure 3. Hemi 23.8% Sub-total 5.3% Hemi 15.6% Sub-total 12.5% Papillary Cancers Biopsy 2.1% Follicular Cancers Biopsy 6.2% Near-total 69.5% Near-total 65.6% Surgery for differentiated thyroid cancer. and 24 months and then annually. Whole body 131 I scan was performed 3 to 6 weeks after near-total to detect and treat any residual and/or metastatic disease, followed 6 months later by a repeat whole body scan to ensure total ablation of the disease. Recurrences were defined as a separate appearance of the disease after apparent previous control, regression or removal. Progressive disease course without remission or a diseasefree interval was not classified as a recurrence. Thus, contralateral lobe recurrence after hemi occurred in 5/30 papillary and 3/13 follicular cancers. Loco-regional recurrence, which was categorized as a recurrence in the thyroid bed, residual thyroid tissue, ipsilateral or contralateral lymph nodes or lymph nodes of the upper mediastinum, or recurrence in the other cervical structures, was seen in 32/99 patients with papillary and 13/65 patients with follicular cancer undergoing near-total or subtotal. Mortality due to thyroid cancer occurred in 26 patients with papillary and 28 patients with follicular cancer. Significance of the various parameters was calculated using the chi-square test. RESULTS Although papillary cancer was the predominant tumour in the patients studied and occurred in 46% of the patients, follicular cancer in 29% and anaplastic and squamous cell cancers in 13% also constituted large percentages. Papillary cancers occurred at all ages, with the majority seen in the third, fourth and fifth decades of life (53.8% of patients < 40 years of age). Nonetheless, of interest was the occurrence of follicular cancers in the younger age groups (33.7% in patients < 40 years of age and nearly 7% in patients < 19 years of age). Females predominated, with 52.3% papillary and 57.8% follicular cancers seen in females. The striking features, therefore, regardless of histological type, were the occurrence of tumours at an earlier age, the higher percentage of follicular cancers in younger patients and the male to female parity. The commonest presentation of both histologic types occurred as overt lesions (p > 0.05). There was a significant difference in the size of the primary tumour at presentation. There was no relationship to the age and sex of the patient. Lymph node involvement was present in 52 papillary and 33 follicular cancers (p > 0. 1). Age and sex of the patient did not affect nodal involvement. Distant metastases were more common in patients with follicular cancer (p < 0.001), were significantly higher in the older age groups in both cancers and were not affected by the gender of the patient Although there was a significantly higher prevalence of skeletal metastases in follicular cancers, there was no difference in the metastases to the lungs (p > 0.1). If is taken as an early stage of the disease, 45.5% of papillary and 36.1% of follicular cancers were early cancers (p > 0.5). Advanced T 3 lesions occurred in 28.8% of papillary and 54.2% of follicular cancers (p < 0.001). Although a significantly higher number of papillary cancers occurred as early lesions, there was no statistical difference in the occurrence of follicular cancers among the various stages of the disease. There was no significant difference in the development of contralateral lobe recurrence following hemi or the development of loco-regional recurrence following subtotal or near-total between the papillary and follicular cancers (p > 0. 5). Mortality was high among patients with both histologic types of cancer, but was significantly higher among patients with follicular carcinoma (p < 0.05). Although Asian Journal of Surgery 327
SARDA AND OTHERS mortality was significantly higher in patients 40 years with papillary cancers (p < 0.05), age did not affect the survival of patients with follicular cancer (Table). Gender of the patient also did not reflect on the mortality in both histological types of tumours. Mortality was significantly higher in both histologic groups if the cancers at presentation were advanced T 3 lesions (p < 0.05); there was no difference in the mortality of early lesions in both histologic types. Only two patients with early size of the primary tumour but N1 2 lymph node status died of the disease, suggesting a better prognosis for patients with small tumours metastasizing to the regional lymph nodes. Death due to local recurrence of the tumour occurred in 50% of patients with papillary and 69% of patients with follicular cancers; thus, 55% of patients with differentiated thyroid cancers with loco-regional recurrence died of the disease. Distant metastases at presentation signified a poor prognosis, with 34% of patients with papillary and 48% of patients with follicular cancers succumbing to the metastatic disease. Metastases to the lungs and the skeletal system resulted in death in 31% and 27% of the patients, respectively; all patients with metastasis to the brain, vertebral column and liver died. DISCUSSION The patients studied were from a geographic area endemic for goitres, where there was a relatively high proportion of follicular cancers, generally implying a worse prognosis. These patients were older, with Table. Prognostic factors Histological type Papillary, n (%) Follicular, n (%) 26 (132) 28 (83) Age, years < 40 9 (71) 11 (28) 40 17 (61) 27 (61) Sex Males 9 (63) 13 (35) Females 17 (69) 15 (48) Stage 5 (60) 3 (30) N 1 2 1 (34) 1 (8) T 3 20 (38) 24 (45) Distant metastases 9 (17) 13 (29) larger size primary tumours and fewer lymph node metastases. 3,4 Papillary cancers of the thyroid, on the other hand, are recognized for a more favourable prognosis and a greater incidence of lymph node involvement. Nonetheless, these observations were not confirmed by our study; we found no distinction between the two histologic cancer types, and the disease had a short duration of symptoms at presentation and displayed an aggressive biological behaviour irrespective of the histology. These findings are comparable to reports from other areas endemic for goitres. 2 Follicular cancers resulted in higher mortality than did papillary carcinomas (p < 0.05), although mortality was high in both groups. Besides the histology of the tumour, extent of the disease and distant metastases at presentation were considered in the univariate analysis as factors affecting survival in our patients. These results are similar to other reports stating that survival rates for papillary and follicular carcinomas are similar among patients of comparable age and disease stage. 5,6 The age of the patient at the time of diagnosis is an important consideration, as it influences outcome of the disease. 5,7 15 Follicular cancers are reported to occur in older patients, with a median age at diagnosis typically in the sixth decade of life, 16 while the mean age of patients with papillary cancers is reportly in the fifth decade. 17 Our patients in both groups presented a decade earlier than the ages reported in the literature. Except among patients with papillary cancer, where the prognosis was significantly better in patients less than 40 years of age, age did not affect survival of patients with follicular cancers. Gender as a prognostic factor is currently being debated, 5,7 9,18 with some authors claiming the importance of sex as one of the major determinants of survival in differentiated thyroid cancers, with males having a poorer prognosis. 7 18 However, we could not confirm this finding because survival among our patients was not different between the two sexes, a finding that is confirmed by other authors 5,10 18 The extent of the disease also contributed to the prognosis; advanced lesions conferred a worse prognosis. 3 7,11,16 18 A study of 157 differentiated thyroid cancers included in the multicentric EORTC trial, places tumour stage in first place as a prognostic factor. 12 Our findings of the extent of the disease as a predictor of death are in agreement with these reports. Nonetheless, the presence of lymphadenopathy did not have an 328 Vol 25 No 4 October 2002
unfavourable effect on the prognosis of the disease in our study. This was also noted by several other authors. 3,5,8 15,17,19 Finally, distant metastasis at presentation was also an important prognostic variable. Distant metastasis at presentation in 21% of our patients occurred more frequently than in reports in the literature 5,13,14 Of our patients with distant metastases, 47% died, confirming the findings of other authors. 3,4,14,17,19,20 Near total or total is the surgical treatment of choice for thyroid carcinoma. 1,5 The extent of surgery is a good indicator of the outcome. In the present study, 32/99 (32.3%) papillary and 13/65 (20%) follicular cancer patients who underwent near total developed loco-regional recurrence during the 3 to 20-year follow-up period (p > 0.05). Death due to local recurrence of the tumour occurred in 50% of patients with papillary and 69% of patients with follicular cancers; thus, 55% of patients with differentiated thyroid cancers with loco-regional recurrence died of the disease. In our study, the extent of the disease, especially the presence of distant metastases at presentation, was the most important determinant of survival, followed by histology of the tumour and age, which was prognostically significant only in cases of papillary cancer. REFERENCES 1. Sarda AK, Kapur MM. Thyroid carcinoma: a report of 206 cases from an area of endemic goitre. Acta Oncol 1990;29: 863 7. 2. Riccabona G, Ladurner D, Steiner E. Changes in thyroid surgery during iodine prophylaxis of endemic goitre. World J Surg 1983;7:195 200. 3. Cady B, Sedgwick CE, Meissner WA, et al. Changing clinical, pathologic, therapeutic, and survival patterns in differentiated thyroid carcinomas. Ann Surg 1976;184:541 53. 4. Schelfhout LJDM, Creutzberg CL, Hamming JF, et al. Multivariate analysis of survival in differentiated thyroid cancers: the prognostic significance of the age factor. Eur J Cancer Clin Oncol 1988;24:331 37. 5. Mazzaferri EL. Thyroid cancer: Impact of therapeutic modalities on prognosis. In: JA Fagin, ed. Thyroid Cancer. Boston: Kluwer Academics Publications, 1998:255 84. 6. Balan KK, Raouf AH, Critchley M. Outcome of 249 patients attending a nuclear medicine department with welldifferentiated thyroid cancer: a 23-year review. Br J Radiol 1994;67:501 58. 7. Samaan NA, Schultz PN, Haynie TP, Ordonez MG. Pulmonary metastasis of differentiated thyroid carcinomas: treatment results in 101 patients. J Clin Endocrinol Metab 1985;60:376 80. 8. Fransilia KO. Is the differentiation between papillary and follicular cancers valid? Cancer 1973:32:853 64. 9. Simpson WJ, McKinney SE, Carruthers JS, et al. Papillary and follicular thyroid cancers. Prognostic factors in 1578 patients. Am J Med 1987;83:479 88. 10. Smith SA, Hay ID, Goellner JR, et al. Mortality in papillary thyroid carcinomas: a case-control study of 56 lethal cases. Cancer 1988;62:1381 8. 11. Tscholl-Ducommun J, Hedinger C. Papillary thyroid carcinomas: morphology and prognosis. Virschows Arch (A) 1982;396:19 39. 12. Wanebo J, Andrews W, Kaiser D. Thyroid cancer. Some basic considerations. Am J Surg 1981;142:472 9. 13. Schlumberger M, Tubiana M, Vathaire F, et al. Long-term results of treatment of 283 patients with lung and bone metastases from differentiated thyroid carcinoma. J Clin Endocrinol Metab 1986;63:960 7. 14. Ruegemer JJ, Hay ID, Bergstralh EJ, et al. Distant metastasis in differentiated thyroid carcinoma: a multivariate analysis of prognostic variables. J Clin Endocrinol Metab 1988;67: 501 8. 15. Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. Am J Med 1994;97:418 28. 16. Cooper DS, Schneyer CR. Follicular and Hurthle cell carcinoma of the thyroid. Clin Endocrinol Metab North Am 1990;19:577 91. 17. Hay ID. Papillary thyroid carcinoma. Clin Endocrinol Metab North Am 1990;19:545 76. 18. Tennvall DS, Biorklund A, Moller T, et al. Is the EORTC prognostic index of thyroid cancer valid in differentiated thyroid carcinomas? Retrospective multicentric analysis of differentiated thyroid carcinomas with long follow up. Cancer 1986;57:1405 14. 19. Rossi PL, Cady B, Silverman ML, et al. Surgically incurable well-differentiated thyroid carcinoma. Arch Surg 1988;123: 569 74. 20. Yamashita H, Noguchi S, Murakami N, et al. Extracapsular invasion of lymph node metastasis is an indicator of distant metastasis and poor prognosis in patients with thyroid papillary carcinoma. Cancer 1997;80:2268 72. Asian Journal of Surgery 329