Premium
Thyroglobulin antibody levels do not predict disease status in papillary thyroid cancer
Author(s) -
SmookePraw Stephanie,
Ro Kevin,
Levin Olga,
Ituarte Philip H. G.,
Harari Avital,
Yeh Michael W.
Publication year - 2014
Publication title -
clinical endocrinology
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.055
H-Index - 147
eISSN - 1365-2265
pISSN - 0300-0664
DOI - 10.1111/cen.12421
Subject(s) - thyroglobulin , medicine , papillary thyroid cancer , thyroid cancer , gastroenterology , linear regression , surrogate endpoint , endocrinology , thyroid , machine learning , computer science
Summary Background Thyroglobulin antibodies ( T g A b) are present in approximately 20% of patients with papillary thyroid cancer ( PTC ) and invalidate the serum thyroglobulin ( T g) level as a tumour marker. We examined whether trends in the T g A b level could serve as a surrogate marker of disease status in the surveillance of patients with PTC . Methods All patients found to have a least one positive postoperative T g A b level (determined by the B eckman‐ C oulter A ccess A ssay) after undergoing initial surgery for PTC from 2000 to 2010 at a single institution were included. Log–log transformation and linear regression were applied to longitudinal T g A b levels, yielding patient‐specific regression coefficients that categorized as follows: highly negative, moderately negative and positive/no trend. The recurrence rate in each category was then assessed. Results Ninety‐three of 967 patients with PTC were included. Recurrent disease was detected in 19 patients (20%) after a mean follow‐up time of 51 months. Regression coefficients in the highly negative and moderately negative groups were not different, and hence these groups were pooled. The proportion of recurrent cases in the negative trend group was similar to that in the positive/no trend group (19·7% vs 21·9%, NS ). The mean regression coefficients were similar for recurrent and nonrecurrent cases within both the negative trend group (−0·89 vs −0·80, NS ) and the positive/no trend group (0·08 vs 0·33, NS ). Conclusion Trends in the T g A b level do not predict disease status in PTC in our experience. In the context of most commercially available T g A b assays, surveillance of T g A b‐positive patients will hinge on high‐quality imaging until a valid alternative serum marker to T g is identified.