Endocrine Abstracts

The observed association between thyroid disorders and breast cancer has provoked many investigations. One area where there is some, although far from universal, agreement is the finding of an increased prevalence of anti-thyroid autoantibodies in breast cancer patients compared to controls. Whether this is a feature of a generalised autoimmune response to breast cancer, or provoked by the enzyme thyroid peroxidase acting as an autoantigen, remains unclear. In this study we looked at thyroid antibodies in terms both of their positivity, either detectability or elevation, and their relationship to serum TSH in patients with breast cancer (n=907) compared to 179 postmenopausal controls. TPOAb detectability was defined as >2.0 and <20.0 kIU/l; and positivity >20.0 kIU/l. TPOAb was undetectable (≤2.0 kIU/l in 82.1% of controls compared to 51.8% of breast cancers; P<0.001). Detectable TPOAb values were (controls 3.9%; breast cancer 24.5%; P<0.001); elevated values (controls 14.0%; BrCa 22.9%; P<0.01). In the case of elevated TPOAb serum TSH was >4.0 mU/l in 61.5% of BrCa compared to 20.0% of controls. Interestingly, both TPOAb positivity and detectability were increased in BrCa patients (33.1 and 24.0%), even when TSH was within the upper 50% of the reference range (2.1–4.0 mU/l). Comparable values for controls were 11.0 and 4.8% respectively (P<0.001 in both cases). Thus a tendency towards higher TSH applied to patients with BrCa who had either detectable or elevated TPOAb. This finding suggests that even the marginally detectable TPOAb levels described in this study may represent more than so called ‘assay noise’. They indicate, on the basis of serum TSH distribution, that TPOAb positivity may be associated with a subtle thyroid dysfunction which we and others have suggested may be beneficial in terms of breast cancer outcome.