Although there has been some improvement in the prognosis of ocular BD with the standard regimens (steroids, colchicine, cyclosporin A, azathioprine, cytotoxic agents, pentoxyphilline, benzathin penicillin, thalidomide), there still is a great risk of significant loss of vision in case of ocular involvement with retinal vasculitis irrespective of the kind of immunosuppresive or cytotoxic regimen used. The combination immunosuppressive regimens and the cytotoxic agents bear the longterm risk of secondary neoplasia. Thus, there still is a need to further improve the therapy of BD by instituting new therapeutic agents.
Interferon is a cytokine that was discovered more than 40 years ago by Isaacs and Lindenmann, who observed that virus-infected cell cultures produced a protein that rendered cells resistant to infection by many viruses. Interferon-a belongs to the so-called type-1-interferons and can be produced by virtually all somatic cells after viral infection. By inducing the release of intracellular enzymes such as 2`5`-oligoadenylate synthetase and double-stranded RNA-dependent protein kinase, it causes degradation of viral messenger RNA and inhibits protein synthesis. IFN-a additionnally has various immunomodulatory effects: increased expression of major histocompatibility complex antigens, increased natural killer and cytotixic T cell activity, shift of the T-cell response towards a TH1-type and many more. It also has antiproliferative and antiangiogenetic properties. There are two different human recombinant a-IFN`s in use for the treatment of viral hepatitis and myeloproliferative syndromes, as well as for certain solid tumours and lymphomas (IFN-a2a and IFN-a2b). They differ in one amino acid only and there probably is no great difference in their efficacy for the abovementioned disorders.
Interferon-a (2a) was instituted in 1986 in the treatment for BD by Tsambaos et al. from Patras in Greece. The rationale for using it was the possible viral etiology of BD and the antiviral properties of IFN-a. This group treated three patients with BD, one of whom had ocular involvement. The dosage was relatively high (9-12 Mill iU i.m./day) and the treatment period short (11-16 days). Anyhow, all symptoms (mucocutaneous, fever, thrombophlebitis and arthritis) remitted, except the ocular disease. Later on, 19 small studies and case reports appeared in the literature. Up to now, considering the latest abstracts from the 9th International Conference on Behçet`s Disease held in Seoul in May 2000, a total of 315 patients have been treated with IFN-a2a or a2b for BD, 40 of whom primarily had ocular involvement (mostly posterior uveitis with retinal vasculitis).
The results in all these studies and case reports from 17 different groups and 7 countries are very promising, although it is difficult to compare them, because different inclusion criteria, treatment regimens and outcome measures were used.
IFN-a clearly is effective for mucocutaneous lesions and arthritis, with a tendency towards lower efficacy for oral aphthae than for the other manifestations. As far as ocular manifestations are concerned, the first case reports on patients with refractory severe ocular BD who were successfully treated with IFN-a appeared in 1993 and 1994. In the 4 cases reported, the patients had retinal vasculitis and received IFN-a in addition to their (ineffective) immunosuppressive agents. All patients achieved complete remission of ocular inflammation.
After a pilot study with 7 patients, our group has treated 52 patients with severe ocular BD (posterior uveitis with retinal vasculitis) with IFN-a2a alone (maximum steroids 10 mg prednisolone) (3-6 Mill iU s.c. daily at the beginning, maintenance dosage 3 x 3 Mill iU / week, after 6-12 months tapering until discontinuation) from 1994 until now. The mean observation period is 26 months. Median BD activity score and posterior uveitis score fell from 7,3 to 3,7 (not significant) and from 3,7 to 0,4 (p<0,0001) respectively. Mean visual acuity significantly rose from 0,5 to 0,8 (p<0,005). 90% of the patients were responder. Time to response was 3,7 weeks for the posterior uveitis score (reduction of at least 50%). IFN could be discontinued after a mean treatment duration of 13,4 months in 36% of the patients without relapse of ocular disease for a mean period of 13,2 months. A recent, randomised controlled study on IFN-a2b plus colchicine plus benzathine penicillin versus colchicine plus benzathine penicillin alone that appeared in The Lancet (Demiroglu H et al, Lancet 355: 605-609, 2000), unfortunately had to be retracted from publication by the editor, due to fraud.
The side effects of IFN-a described in BD patients roughly are the same known from patients with hepatitis C or chronic myelogenous leukemia (flu-like syndrome in the first 2 weeks, reddening at the site of injection, itching, alopecia, depression, leukopenia, thyroiditis, exacerbation of psoriasis, occurrence of autoantibodies and anti-interferon-antibodies). Other side effects, mimicking BD symptoms, as retinal infiltrates (described in hepatitis C) and pathergy phenomenon (described in CML) have not occurred in the BD patients treated with IFN-a by now. To the contrary, pathergy phenomenon disappeared under IFN, as did retinal infiltrates due to vasculitis (own observations).
In conclusion, IFN-a is a promising agent in the treatment especially of severe ocular BD, where the results of the immunosuppressive regimens still are unsatisfactory. The dosage necessary to achieve optimal results probably is 3-6 Mill iU daily for 4-8 weeks, maintenance dosage 3 Mill iU 3x/week, with a treatment duration of at least 6 months before discontinuation of the drug. We would like to underline that in our hands it is not possible to discontinue azathioprine or cyclosporin A in severe ocular BD without relapse or even rebound phenomena - thus, IFN in this respect may be superior to the standard immunosuppressants. The time to response also seems to be shorter than with immunosuppressants. The mechanism of action of IFN-a in BD is still unclear and should be examined in further trials. A controlled randomised crossover study of IFN versus CSA should urgently be performed in order to be able to clearly determine the significance of IFN in the treatment of BD.