**4. Immune interventions**

Several immune interventions have been tried since the 1970ies we tried plasmapheresis in Linköping, Sweden, with the aim to preserve residual beta cell function, but so far all different approaches have shown insufficient efficacy and/or given unacceptable adverse effects [21-28]. Broad immunosuppressive or immunoblocking therapies with steroids, cytostatics, high doses of immunoglubulins, anti-lymphocyte globulins have shown some but unfortunately limited efficacy, and adverse events have lead to restrictions both in dose and time. Our studies using photopheresis did show some efficacy, and although the treatment was very laborious it has regained some interest. However, most encouraging is the use of monocloncal antibodies, especially against CD-3 [29-31] but also against CD-20 [32]. Unfortunately treatment with monoclonal antibodies in doses large enough to give efficacy also cause rather common and occasionally serious adverse events.Therefore such therapies are rarely justified as preventive interventions in healthy children with increased risk of developing T1D except for children with extremely high risk of developing T1D close in time.

**Figure 1.** At a Nordic symposium in connection with Annual meeting of Scandinavian Society for the Study of Diabe‐ tes, Linköping 1981, the author showed this slide. Type 1diabetes was proposed to develop after a long autoimmune process destroying the beta cells. Events during pregnancy and the importance of breast-feeding was suggested, and later shown to be relevant, and regeneration of beta cells was proposed as a possibility.

After encouraging Phase II trials two different Phase III trials using antiCD3 failed to reach their primary endpoints [33,34], but one of them, the Protégé study, did show efficacy in younger patients age 8-18 years when a reasonably high dose of antiCD-3 was used [34]. This was especially true in certain patient populations (mainly patients in USA, but also in Europe) who had rather well preserved C-peptide, often near-normal HbA1c and low insulin require‐ ment. Further studies are needed to learn what doses are efficient without severe adverse events, and in what patient populations the treatment works best. The old policy defended by many diabetologists to treat all so called T1D in the same way irrespective of age, ethnic back‐ ground, severity of disease at diagnosis etc may probably have to be left.
