TO THE EDITOR: de Meer et al. recently published a retrospective study comparing diagnostic whole-body scintigraphy (DxWBS) to stimulated thyroglobulin measurement in patients with high-risk differentiated thyroid cancer (1). The authors concluded that DxWBS offered no additional information compared with recombinant human thyroid-stimulating hormone (rhTSH)–stimulated thyroglobulin levels for this cohort of patients. We would argue that shortcomings in the study design and DxWBS methodology invalidate the authors’ conclusion.
The authors defined high-risk patients as those with either T3 or T4 tumors or cervical lymph node metastases (N1) based on American Joint Committee on Cancer (AJCC) TNM version 7 (2). Interestingly, they chose to exclude all patients with distant metastases (M1). However, the authors’ high-risk definition is inconsistent with both American and European guidelines, which include patients with M1 disease in their definitions of high-risk patients (3,4). Moreover, the inclusion and exclusion criteria may have been stated incorrectly: we were surprised to see AJCC TNM stage II patients listed in Table 1 of the article. AJCC TNM stage II is defined as either M1 disease in patients less than 45 y old or T2N0M0 for patients more than 45 y old. Since both of these subsets of patients were purportedly excluded, stage II patients should not have appeared in the group analyzed. This inconsistency warrants explanation or correction.
The authors did not take into account age when risk stratifying their patients since European treatment guidelines are independent of age. However, age is considered the most important prognostic variable for mortality by the American National Comprehensive Cancer Network guidelines, with higher mortality in patients over 40 y old (5). For papillary and follicular thyroid carcinoma, AJCC TNM staging defines all patients less than 45 y old as either stage I or stage II, including patients with distant metastases. In a study validating AJCC TNM classification and group staging for patients with papillary thyroid carcinoma, age was an independent predictor of both disease-free survival and cause-specific survival (6). Not using the age for risk stratification in such a study renders the results irrelevant to nuclear medicine practices in the United States.
In their study, the authors performed DxWBS with thyroid hormone withdrawal (THW) between January 1998 and December 2004 and then switched to rhTSH exclusively from January 2005 to January 2009. However, the original phase III clinical trial comparing rhTSH and THW preparation for DxWBS concluded that rhTSH DxWBS was less sensitive than DxWBS using THW (7). In this phase III trial, rhTSH DxWBS was inferior to THW DxWBS in 18 (29%) of 62 patients and failed to detect metastatic disease in 8 (13%) of 62 patients with positive scans. A second phase III clinical trial, again comparing rhTSH and THW preparation for DxWBS, also showed that rhTSH DxWBS was inferior to THW DxWBS in 8 (16%) of 49 of patients with metastatic disease (8). Although this difference was not statistically significant (p = 0.109), the trend favored THW DxWBS. As a result, we believe that rhTSH DxWBS should be reserved for low-risk patients only, an approach supported by the package insert for rhTSH itself. Moreover, the authors waited 7 d after 131I administration to perform the DxWBS (European guidelines recommend between 2 and 5 d) and used twice the recommended 131I activity; these factors make their technique impossible to compare with standard practices.
Relying on rhTSH-stimulated thyroglobulin for detection of recurrence has its perils. Comparing 131I rhTSH-stimulated DxWBS to thyroglobulin during routine follow-up evaluations, Robbins et al. found metastatic thyroid carcinoma on DxWBS in 13.7% of patients of all risk categories with stimulated thyroglobulin of 2 μg/L or less (9). The authors of the present paper mentioned the conclusion of Robbins et al. but left a huge gap in the discussion by offering no explanation as to why their own results and conclusions were so different from those of Robbins et al. Only the DxWBS can show iodine avidity and guide the decision on whether to treat the patients with 131I or with surgery. In the United States, DxWBS remains the gatekeeper to more advanced imaging with 18F-FDG PET, as most insurance companies require both elevated thyroglobulin levels and negative DxWBS findings to reimburse PET. Omitting the DxWBS would make 18F-FDG PET unavailable to patients in the United States.
The deMeer et al. study addressed a very narrow segment of high-risk thyroid carcinoma patients, did not follow accepted risk stratification guidelines, and does not help many of us who use age in this stratification. The striking deficiencies of this work are the application of rhTSH-stimulated DxWBS to a high-risk group in which it is considered to be inferior to THW DxWBS, the use of nonstandard imaging techniques, and the notion that disease location and iodine avidity are irrelevant to patient management. The readers are encouraged to keep these issues in mind as they evaluate high-risk thyroid cancer patients.
Footnotes
Published online Feb. 8, 2012.
- © 2012 by the Society of Nuclear Medicine, Inc.