ANTIBODIES TO THE PROTEIN-TYROSINE PHOSPHATASES IAR AND IA-2 ARE ASSOCIATED WITH PROGRESSION TO INSULIN-DEPENDENT DIABETES (IDDM) IN FIRST-DEGREE RELATIVES AT-RISK FOR IDDM
Rs. Schmidli et al., ANTIBODIES TO THE PROTEIN-TYROSINE PHOSPHATASES IAR AND IA-2 ARE ASSOCIATED WITH PROGRESSION TO INSULIN-DEPENDENT DIABETES (IDDM) IN FIRST-DEGREE RELATIVES AT-RISK FOR IDDM, Autoimmunity (Print), 28(1), 1998, pp. 15-23
Insulin-dependent diabetes mellitus (IDDM) is preceded by the presence
of antibodies against islet proteins including a protein tyrosine pho
sphatase (PTP) designated IA-2 Recently we cloned a novel PTP named IA
R which shares 43% sequence identity with IA-2 and is recognised by an
tibodies from a majority of patients with IDDM, The aim of the present
study was to determine whether IAR antibodies (IAR Ab) or IA-2 antibo
dies (IA-2 Ab) are associated with progression to IDDM in first-degree
relatives ''at-risk'' for IDDM (operationally defined as those with i
slet cell antibodies [ICA] greater than or equal to 20JDFU or insulin
autoantibodies [IAA] greater than or equal to 100 nU/ml), and to exami
ne combinations of IAR Ab and IA-2 Ab in these subjects. The sensitivi
ty and specificity of these antibodies were also examined in patients
with recent-onset IDDM, Using Cox's Proportional Hazards Model, the nu
mber of siblings with IDDM was associated with progression to IDDM in
at-risk'' relatives, but other covariables (age, sex, number of affect
ed offspring or parents) were not significantly associated. Using numb
er of affected siblings as a covariable, both IAR and IA-2 antibodies
were significantly associated with progression to IDDM (p < 0.005), Co
mbinations of both antibodies, however, did not result in a significan
tly stronger association with progression to IDDM, The threshold of po
sitivity for IAR Ab (0.5 units) and IA-2 Ab (3.0 units) assays was adj
usted to give the same specificity (97.9%) for each assay in 144 healt
hy control subjects, to allow standardised comparisons. Levels of IAR
Ab and IA-2 Ab were strongly correlated in 53 recent-onset IDDM patien
ts (r = 0.70, p < 0.0001) but 11.3% had IAR Ab in the absence of IA-2
Ab and 16.9% had IA-2 Ab in the absence of IAR Ab. The sensitivity for
IDDM (defined as the proportion of IDDM patients positive) was 56.6%
for IAR Ab and 62.3% for:IA-2 Ab. We conclude that there is considerab
le overlap in IA-2 Ab and IAR Ab positivity, although tither antibody
can occur independently in IDDM patients. Both IAR Ab and IA-2. antibo
dies are associated with progression to IDDM in first-degree relatives
at-risk of IDDM, but the use of IAR and IA-2 antibodies in combinatio
n are not significantly more strongly associated with progression than
single antibodies. IAR Ab may play an important role in the predictio
n of IDDM.