Marco Songini
on behalf of the Sardinian IDDM Study Groups
The Type 1 Diabetes Sardinia (Hot&Cold Spot) Project:
what did we learn so far?
Dr. Marco Songini is the director of the Diabetes Unit at the S. Michele Hospital in Cagliari (Sardinia- Italy) He is also the vice-president of ASRIS (Association for the Study of Type 1 Diabetes in Sardinia)
The island of Sardina has the second highest incidence of type 1 diabetes in the world and a high prevalence of other autoimmune diseases (celiac disease, thyroid autoimmune diseases) has been reported.
The type 1 diabetes prevalence has rapidly increased after the second world war.
The island also offers a relatively small population with a homogenous genetic background selected by centuries of isolation.
Immune markes of type 1 diabetes (ICA, GADA, IA2, IAA) appear several years before the clinical onset of the disease, as was seen in family members of diabetic patients, but relatively little is know about the prevalence and prognosis of such markers in the general population. The aim of the Hot and Cold Spot Project is to investigate the prevalence of immune markers in Sardinian population and to evaluate their prognostic significance to develop a screening procedure.
History of the Hot&Cold Spot Project (1)
1995 1994 ending of the recruitment of the SSI (total cohort 10,000 children) first recruitment of newborn from the general population: launch of the Sardinian Newborn IDDM study (SNI) the Military Service approach (secular trend of type 1 diabetes among Sardinians) ICA assay in 1,800 serum samples of school children from the general population enrolled for an epidemiological study of goitre prevalence in Sardinia; first milestone of the Sardinian School children IDDM Study (SSI) Eurodiab - Ace: Sardinia is an ‘hot spot’ for Type 1 diabetes in Europe
1993 1990
1989
History of the Hot&Cold Spot Project (2)
1999 1998 1998 1997
1997 1997
HLA typing in ‘immunologically at risk’ children for type 1 diabetes the Post-partum Thyroiditis and Neonatal Hypothyiroidism studies the Gestational Diabetes (GDM) Study Coeliac Disease in the Northern Sardinia school children
The Sardinian Migrants IDDM study (SMI) in the province of Pavia stop of recruitment of SNI (total cohort 19,000 children) The Environmental / Veterinarian / Ecological studies first results
1996
Time trends in Type 1 Diabetes incidence
Incidence (per 100,000)
45 40 35 30 25 20 15 10 5 0 1960 1965 1970 1975 1980 1985 1990 1994
( 0 - 14 years)
Finland (40) Sardinia (35) Sweden (25.8) Norway (21.2) Denmark (16.4) Hungary (8.8)
Calendar year
The Sardinian Schoolchildren IDDM Study (SSI) (1)
Survival Function 1,0
,9 Cum Survival
Number of Abs
,8 ,7 ,6 ,5 0 20 40 60 80 LATENCY (months)
3 2 1 0 100
The Sardinian Schoolchildren IDDM Study (SSI) (2)
% 10
9 8 7 6 5 4 3 2 1 0
1 Ab
45/100,000/year 35/100,000/year
>= 2 Abs
Cagliari Sassari 38/100,000/year 30/100,000/year
Oristano Nuoro
The Sardinian Newborn-IDDM Study (SNI) (1)
18 Centres from the 4 Sardinian provinces involved in the Study
Number recruited ~ 19,000 ICA tested = 15,509 (cord blood)
ICA JDF-u ICA 6-19 JDF-u ICA JDF-u
2.3%
1.7%
0.6%
The Sardinian Newborn-IDDM Study (SNI) (2)
% 2.0
FOLLOW UP
prevalence
1.6 1.2 0.8 0.4 0.0
2,959 2,125 2,117 1,148 399 399
1st YEAR 2nd YEAR
388
3rd YEAR
ICA JDF-u
GADA
IA-2icA
Pr e v a l e n c e o f i s l e t - re l a t e d a u t o a n t i b o d i e s (ICA, GADA and IA - 2icA) according to the different age (from the S a r d i n i a n N e w b o r n and the S a r d i n i a n S c h o o l C h i l d r e n S t u d i e s) 7 %
6
prevalence
5 4 3
2 1 0
cord 1 blood
2
3
6
7
8
9 10 11 12 13 14 15
age (years)
GADA IA-2icA
ICA JDF-u
The COELIAC DISEASE study (1) ~ 1,600 Sardinian school children investigated for AGA-IgG, AGA-IgA and AEA Prevalence of coeliac disease
(confirmed by intestinal biopsy)
10.5/1000 children the highest reported so far in a background population
Coeliac Disease and pre -Type 1 Diabetes in Sardinian schoolchildren (2)
Isletrelated Abs CD-related Abs
79
* ICA + IgG-AGA = 8
11*
126
GADA + IgG-AGA = 1 GADA + IgG/IgA-AGA + AEA = 2
The MILITARY SERVICE approach
Secular trend of type 1 diabetes prevalence at 19 years among male conscripts in Sardinia Prevalence (per 1000)
7,00
6,00 5,00
4,00 3,00 2,00 1,00
1936 1938 1940 1942 1944 1946 1948 1950 1952 1954 1956 1958 1960 1962 1964 1966 1968 1970 1972 1974 1976 1978
0,00
Birth cohorts
The ENVIRONMENTAL and ECHOLOGICAL studies
Birth seasonality Onset seasonality Temp, pop. density, urban/rural Average rainfall Time- and space-clustering overlap with: - malaria - talassemia - G-6-P-D deficiency Nitrate intake (bottle and tap waters) Cow’s and breast milk feeding YES YES NO NO YES NO NO NO NO NO
The GESTATIONAL DIABETES (GDM) Study ~ 100 Sardinian mothers with GDM investigated for ICA, GADA and IA-2icA 8 women resulted positive for at least 1 isletrelated autoantibody at the time of OGTT
After a follow up of 4 yrs, 5 became diabetic (3 insulin-dependent and 2 non insulin-dependent)
AUTOIMMUNE THYROIDITIS in Sardinian school children (1) ~ 8,000 Sardinian schoolchildren from the general population
investigated for ATA
Overall, the prevalence of ATA was 3.7% and the prevalence of a subclinical thyroiditis was about 0.9% No correlations were found between prevalence of ATA and urinary iodine excretion or prevalence of goitre
ATA pos (%)
6 5
4 3
2 1
Males Females
0
6-7 7-8 8-9 9-1010-1111-12 12-13 13-14 >14
Age (years)
ATA and ICA in Sardinian school children (2)
ICA + 325
16
ATA + 211
AUTOIMMUNE THYROIDITIS and PREGNANCY (3) ~ 2,500 Sardinian mothers at delivery time were investigated for ATA and ICA At the time of delivery, the prevalence of ATA and ICA was 11.8 and 2.6%, respectively (0.4% with both specificities) Prevalence of ATA (%)
100 80 60 40 20 0 40 35 30 25 20 15 10 5 0
OR
CA
NU
SS
ATA at low titers ATA at high titers
ICA* 20
ICA 5-19
ICA <5
The SARDINIAN MIGRANTS study the prevalence of type 1 diabetes was assessed in ~ 2,200 born in Sardinia and migrated to Pavia In 10 individuals the diagnosis of T1D has been confirmed giving a prevalence (4/1000) similar to that registered in the island and 3 times higher than the ones registered in Northern Italy. 3 subjects were already diabetic at the time of migration, and 7 developed the disease after the migration to Pavia.
35 30 25 20 15 10 5 0
Age at onset of T1D (years)
Before migration After migration (Nr=3) (Nr=7)
What did we learn so far? (1)
Islet-related autoantibodies can appear very early in life (with particular reference to GADA) and they play a predictive role towards the future onset of type 1 diabetes. The appearance of islet-related autoantibodies progressively increases in the first years of life. The combination of more than 1 islet-related autoantibody (rather than which autoantibody) is the best predictor for the development of type 1 diabetes in the Sardinian school children population.
What did we learn so far? (2)
Even though some epidemiological evidences suggest a role of the environment on the etiopathogenesis of type 1 diabetes, none of the variables considered so far have shown their influence; however other variables need to be further investigated. Coeliac disease shows an high prevalence among Sardinian school children and then it deserves more large investigations. It seems not to play a relevant role on the etiopathogenesis of type 1 diabetes in Sardinia instead.
What did we learn so far? (3)
The prevalence of thyroid-related autoantibodies seems not to be as much as high among Sardinian school children, even though they live in an Island at high risk for other autoimmune diseases. The same findings appear among the pregnant mothers, in whom the prevalence of ATA is not significantly higher than the ones registered in other matched populations. However, the possible immunosuppressive role of pregnancy on these parameters needs to be further investigated.
What we are doing now (1)
Trying to further improve the prediction of type 1 diabetes in the general population by carrying out HLA genetic typing in children found ‘at immunological risk’ during the screening. Studying the immunological and genetic markers for type 1 diabetes and other autoimmune diseases in Sardinian migrants and their relatives. Investigating other putative environmental factors which can play a role towards the etiopathogenesis of type 1 diabetes (e.g., chemicals, toxins, vaccinations, viral infections, etc.).
What we are doing now (2)
Broadening the original investigation for type 1 diabetes on the prevalence of coeliac disease and autoimmune thyroid diseases and to study their associations. Comparing data from Sardinia and other areas (continental Italy, Finland, Sweden, Spain) by new collaborative studies. Investigating LADA within the Sardinian type 2 patients.
Cost of predicting T1-DM from birth and in school children background population
cost saved
$ 250,000
225,000 200,000 175,000 150,000 125,000 100,000 75,000 50,000 25,000 0
20 18
16 14
cost saved
cost saved
12 10
8
years
6
4 2 0
1 2 3 4
modified from Hahl et al. Diabetologia (1998) 41:79-85
Benefits of a predictive screening for Type 1 diabetes in the general population
1. Early diagnosis / Early insulin treatment
(In U.S.A. about 50 deaths yearly from DKA)
2. ? Prevention and/or delay of diabetic complications 3. Prevention of the onset of type 1 diabetes in
‘at risk’ individuals (? magic bullet)
4. Identification of ‘non at-risk’ individuals (>99%)
In spite of Gian ? Franco’s breakthrough twenty-five years ago ? with the ICA, the mystery of Type 1 diabetes still remains?deeply hidden ... ? ? ? ? ? … but we strongly believe that along ? with him ? this beautiful Island, we in have a very good chance to unravel the causes of the autoimmune diseases. ? ?
The IDDM-Sardinia Study Groups