| Treatment category | Specific agents/approaches | Mechanism of action | Evidence level / response rates | Key references / notes |
|---|---|---|---|---|
| Corticosteroids (systemic) | Oral prednisone 60–80 mg/day; intravenous methylprednisolone induction | Broad anti-inflammatory and immunosuppressive effects; suppress autoimmune retinal injury | Evidence mainly from retrospective series and case reports; no standardized regimen; IV methylprednisolone reported to have better outcomes than oral prednisone in some reports (pqac-00000011) | Common first-line therapy after excluding/treating underlying malignancy or systemic disease; early treatment considered important to limit irreversible retinal damage; no randomized controlled trials (pqac-00000009, pqac-00000011) |
| Corticosteroids (local) | Intravitreal triamcinolone; sub-Tenon triamcinolone; intravitreal sustained-release fluocinolone acetonide implant; topical/depot steroids | Local suppression of ocular inflammation with reduced systemic exposure | Brief intravitreal/sub-Tenon steroid trial has been proposed diagnostically/therapeutically; case reports describe restoration of retinal anatomy and vision improvement in CAR and benefit in MAR with fluocinolone implant; evidence limited to case reports/series (pqac-00000011) | Suggested by some authors before prolonged systemic steroids to verify steroid responsiveness; used for CME and local disease control; protocol not standardized (pqac-00000011) |
| Immunomodulators / steroid-sparing agents | Azathioprine, cyclosporine, mycophenolate mofetil, infliximab, methotrexate | Reduce adaptive immune activation and autoantibody-associated inflammation; steroid-sparing maintenance therapy | Retrospective triple-therapy series (cyclosporine + azathioprine + prednisone) in 30 patients reported overall response in 70%; among npAIR, 54% without CME and 73% with CME responded, but study limitations were substantial (pqac-00000011) | Frequently used when corticosteroids are insufficient or not tolerated; evidence is heterogeneous and vulnerable to publication bias; successful methotrexate use is not well documented in npAIR (pqac-00000009, pqac-00000011) |
| Biologic agents | Rituximab; tocilizumab; sarilumab; alemtuzumab; ipilimumab | Rituximab depletes CD20+ B cells and pro-inflammatory CD3+CD20+ T cells; IL-6 blockade (tocilizumab/sarilumab) may reduce refractory CME/inflammation; ipilimumab augments anti-tumor immunity in melanoma-associated disease | Rituximab has the strongest biologic signal in AIR from retrospective series and case reports; one 16-patient series reported stable or improved visual outcomes in 77% of eyes with rituximab-based combinations; anti-IL-6 therapies reported complete resolution of refractory CME in isolated case reports (pqac-00000011) | Biologics are generally used for refractory disease; evidence remains low-level and non-comparative; recent reviews emphasize lack of prospective trials and no standard treatment protocol (pqac-00000000, pqac-00000009, pqac-00000011) |
| IVIG | Intravenous immunoglobulin | Immunomodulation via Fc-mediated immune regulation, neutralization of pathogenic antibodies, and altered B/T-cell signaling | Evidence limited to case reports/small series; sometimes used in combination with steroids or other immunosuppressants; no robust response-rate estimate available (pqac-00000011) | Considered an option in refractory AIR, especially when humoral autoimmunity is suspected; included among available therapies in recent review tables (pqac-00000011) |
| Plasmapheresis / plasma exchange | Therapeutic plasma exchange, often combined with systemic immunosuppression | Removes circulating antiretinal autoantibodies and immune mediators | Evidence limited to case reports and small series; no standardized schedule or comparative efficacy data available (pqac-00000011) | Mechanistically attractive for antibody-mediated disease, but current support is anecdotal; should be interpreted cautiously due to absence of controlled trials (pqac-00000009, pqac-00000011) |
| Tumor-directed therapies (paraneoplastic AIR) | Surgical cytoreduction, chemotherapy, radiotherapy; treatment of underlying malignancy before ocular immunosuppression | Reduces tumor antigen burden that may drive molecular mimicry and autoantibody production | Evidence indirect but clinically important; early oncologic diagnosis in CAR was associated with better visual outcomes in a recent paraneoplastic review; treatment of malignancy is foundational in pAIR (pqac-00000001, pqac-00000011) | Recommended first in pAIR/CAR/MAR before or alongside ocular immunosuppression; MAR may indicate immune response to melanoma metastasis, and immunosuppression can theoretically worsen tumor control (pqac-00000001, pqac-00000010, pqac-00000011) |
| Overall treatment strategy / evidence gap | Multidisciplinary individualized care; combine oncology, ophthalmology, rheumatology/immunology; monitor ERG, OCT, FAF, visual fields | Tailors therapy to subtype (CAR, MAR, npAIR), systemic disease, and structural/functional progression | No international standard treatment protocol; evidence base is dominated by retrospective series and case reports; prospective randomized placebo-controlled trials are explicitly needed (pqac-00000000, pqac-00000009, pqac-00000011) | Current management is empirical; early diagnosis is repeatedly emphasized because retinal damage may become irreversible and pAIR can precede cancer detection (pqac-00000000, pqac-00000009, pqac-00000011) |


*Table: This table summarizes the main currently reported treatment approaches for autoimmune retinopathy, their rationale, and the level of supporting evidence. It is useful because AIR management remains non-standardized and is based largely on retrospective series and case reports rather than prospective trials.*