848 E D I T O R I A L Reply to Counterpoint Staging of Thyroid Carcinoma– Reply Steven I. Sherman, M.D. Section of Endocrine Neoplasia and Hormonal Disorders, The University of Texas M. D. Anderson Cancer Center, Houston, Texas. F See editorial counterpoint on pages 844 –7 and referenced original article on pages 1012–21, this issue. Address for reprints: Steven I. Sherman, M.D., University of Texas M. D. Anderson Cancer Center, 1515 Holcombe Blvd., Box 15, Houston, TX 77030. Received June 15, 1998; accepted June 19, 1998. © 1998 American Cancer Society or physicians who care for patients with thyroid carcinoma, one of the most frustrating yet critical early management functions is to attempt to predict a patient’s prognosis. The inescapable reality is that approximately 10% of patients eventually will die of their tumor and significantly more patients may face morbidity due to disease recurrence and metastasis.1 The ability to provide patients with an accurate estimate of their risk for morbidity and mortality and to use this information in planning treatment and follow-up is a fundamental responsibility for the treating physician. Furthermore, appropriate clinicopathologic information must be made available to the physician to assist in this task.2 Thus, the accompanying editorial regarding our analysis of clinical staging approaches using the data of the National Thyroid Cancer Treatment Cooperative Study (NTCTCS) is disappointing. The author’s preference for patient stratification as a binary function of low versus high risk is oversimplified, not recognizing the broad spectrum of morbidity and mortality that occurs with these diseases. For example, as patient follow-up extends out multiple decades from the initial diagnosis of papillary carcinoma, it becomes clear that there remains a finite risk of disease recurrence and disease-related death,3 and that decisions made even in those early years of treatment potentially can affect long term outcomes. Furthermore, the options available to clinicians for treatment are far from binary. Locoregional disease control likely depends on the extent of initial surgery, which requires decisions not only regarding the amount of thyroid resection but also the magnitude and location of lymph node dissection as well. Radioiodine ablation itself contains multiple options, including variations in the amount of administered activity to achieve ablation if appropriate. Decisions regarding adjuvant therapies also come with multiple choices; clinicians not only must select whether or not to use thyroidstimulating hormone (TSH)-suppressive levothyroxine therapy, but also must identify a degree of suppression appropriate for a patient’s risk status.4 Adjuvant external beam radiotherapy also may be appropriate for certain groups of patients at risk for locoregional recurrence.5 And with each of these choices, an individualized risk:benefit analysis is required to counterbalance the potential complications of a therapy versus the potential gain from treating the disease. Results from a recent survey of thyroid spe- Reply to Counterpoint/Sherman cialists6 and from an analysis of practice patterns within the NTCTCS7 support the concept that thyroid carcinoma management reflects a range of possibilities rather than a binary choice. Viewing a patient’s risk as occurring on only two levels also reduces the amount of prognostic information gained by applying a disease stage to a patient. As demonstrated by both the retrospective single hospital data set published in 19978 and our current analysis, use of a dichotomous risk assignment yields among the lowest predictive values for disease mortality from differentiated thyroid carcinoma. The limitations of a binary classification also are apparent in designing an appropriate follow-up strategy for the patient. Again, a continuous range of options exists, including inclusion or frequency of tests such as thyroglobulin determination, chest Xray, neck ultrasound, radioiodine imaging, or other diagnostic modalities.9 This flexibility of patient monitoring will only be augmented by the anticipated availability of recombinant human TSH for stimulating radioiodine uptake and thyroglobulin levels.10 Dr. Cady’s editorial takes issue with the inclusion of multiple histologies of thyroid carcinoma in the NTCTCS staging classification. In particular, it criticizes the inclusion of patients with medullary carcinoma of neuroendocrine lineage, as well as those with anaplastic carcinoma, an aggressive yet follicular cell-derived malignancy. Nonetheless, the core element of any commonly used oncologic staging remains anatomic extent rather than tissue type.11 For example, the TNM staging approach to both prostate and lung carcinomas completely ignores the cells of origin for these diverse diseases, applying the same criteria for the staging of both epithelial and neuroendocrine tumors within each of these respective organs.11 In contrast, the NTCTCS classification provides distinctly different criteria for the staging of differentiated, medullary, and anaplastic thyroid carcinomas. The editorial touts the TNM thyroid classification, but even this approach provides staging criteria for medullary and anaplastic carcinomas. Therefore, the argument to remove these two diseases from the analysis is inconsistent with the usual and accepted approach to cancer staging, and appears to be inconsistent internally with other aspects of the editorial. From a practical viewpoint, the analogy provided to gastric sarcoma and lymphoma suffers in that these diseases often are treated by different groups of specialists within cancer centers; in contrast, there usually is little difference in the caregiving team 849 involved in the management of patients with thyroid carcinoma, regardless of the tumor histology, and it is only sensible that we pool our patient experiences together in a cooperative study such as this. Issue also must be taken with the criticism regarding the inability to separate prognoses of patients with Stage I and II disease. Given the early stage of follow-up for this tumor registry, I believe it shortsighted to expect that the outcomes of such patients would yet diverge. It has been a traditional fallacy of many retrospective studies to draw conclusions regarding the results of treatment in these lower risk stages with limited follow-up. Instead, we fully anticipate that these patient cohorts will demonstrate disease-related morbidities and mortality in the years and decades to come, requiring an ongoing intensive effort to continue to collect and analyze these events prospectively. Finally, one of our primary responsibilities as clinicians is to the patient sitting there, desiring as accurate an answer as possible to the question, ‘‘Doctor, what is my prognosis?’’ As the generally poor predictive values in our analysis demonstrate, not only is the current ‘‘surfeit’’ of staging classifications inadequate, but newer approaches clearly will be necessary in the future. Only when we have developed and validated such improved methods will it be likely that clinicians observe Dr. Cady’s laudable goal of modifying disease management based on appropriate application of a prognostic staging classification. Until such time, we must use the best classifications we have available, including the NTCTCS, to guide communication among physicians, to assist in patient management, and to guide future clinical studies. REFERENCES 1. 2. 3. 4. 5. Robbins J, Merino MJ, Boice JD Jr., Ron E, Ain KB, Alexander HR, et al. Thyroid cancer: a lethal endocrine neoplasm. Ann Intern Med 1991;115(2):133– 47. Sherman SI. Toward a standard clinicopathologic staging approach for thyroid carcinoma. Semin Surg Oncol. In press. Mazzaferri EL, Jhiang SM. Long-term impact of initial surgical and medical therapy on papillary and follicular thyroid cancer. Am J Med 1994;97(5):418 –28. Pujol P, Daures J-P, Nsakala N, Baldet L, Bringer J, Jaffiol C. Degree of thyrotropin suppression as a prognostic determinant in differentiated thyroid cancer. J Clin Endocrinol Metab 1996;81(12):4318 –23. Farahati J, Reiners C, Stuschke M, Muller SP, Stuben G, Sauerwein W, et al. Differentiated thyroid cancer. Impact of adjuvant external radiotherapy in patients with perithyroidal tumor infiltration (stage pT4). Cancer 1996; 77(1):172– 80. 850 6. 7. 8. CANCER September 1, 1998 / Volume 83 / Number 5 Solomon BL, Wartofsky L, Burman KD. Current trends in the management of well differentiated papillary thyroid carcinoma. J Clin Endocrinol Metab 1996;81(1): 333–9. Cooper DS, Specker B, Ho M, Sperling M, Ladenson PW, Ross DS, et al. TSH suppression and disease progression in patients with differentiated thyroid cancer: results from the National Thyroid Cancer Treatment Cooperative Study Registry. Thyroid. In press. Brierley JD, Panzarella T, Tsang RW, Gospodarowicz MK, O’Sullivan B. A comparison of different staging systems predictability of patient outcome. Thyroid carcinoma as an example. Cancer 1997;79(12):2414 –23. 9. Schlumberger MJ. Papillary and follicular thyroid carcinoma. N Engl J Med 1998;338(5):297–306. 10. Ladenson PW, Braverman LE, Mazzaferri EL, Brucker-Davis F, Cooper DS, Garber JR, et al. Comparison of administration of recombinant human thyrotropin with withdrawal of thyroid hormone for radioactive iodine scanning in patients with thyroid carcinoma. N Engl J Med 1997;337(13):888 –96. 11. American Joint Committee on Cancer. AJCC cancer staging manual. Philadelphia: Lippincott-Raven, 1997.