SLE and Antiphospholipid Syndrome [223–231]

Background: Anti-CD20 (rituximab) therapy has been very successful in SLE patients resistant to conventional therapy, with about 80% of patients responding in most open label case series. However, about 30% of patients develop anti-drug antibodies (ADA), which may have a detrimental effect on the efficacy of the drug. Identifying ADA is, therefore, important to predict non-responders and to explain loss of efficacy, but low levels of ADA may be missed because of formation of immune complexes with the drug. A novel approach to this problem is the acid disruption of immune complexes immediately before analysis, which has been implemented by GE Healthcare in the Immunogenicity Package on the T100 BIAcore instrument. We have used this technology to characterize anti-rituximab responses in a series of SLE patients. Methods: We identified 9 patients treated with rituximab in dedicated lupus clinics in Birmingham, all with a confirmed diagnosis of SLE. All were treated with rituximab after failing to respond to one or more conventional immunosuppressants. Standardized doses of either 750 mg or 1 g of rituximab were given on two occasions, 2 weeks apart. 9 patients had undergone more than one course of rituximab treatment. Blood samples were collected for up to 51 months. A BIAcore CM5 chip was prepared by coupling rituximab to one flow cell and a rat anti-rituximab to another. Using the Immunogenicity Package on the BIAcore T100, samples were acidified with hydrochloric acid and then neutralized with Tris-HCl simultaneously with their injection over the coupled chip. In paired analyses the HCl and Tris-HCl were replaced with Hepes/saline pH 7.4, so that the same analysis occurred without acid disruption. Results: 25 serum samples from the 9 patients were assessed. When the analysis was done without acid pre-treatment 2 samples from one patient were positive for ADA against rituximab and 5 samples from 4 patients were positive for the drug. However, when acid disruption was used 5 samples from 2 patients were shown to be positive for ADA while 8 samples were positive for rituximab. Most notably, one patient was negative for both drug and ADA when assessed at pH 7 but were positive for both when acid treatment was used. One patient had high levels of ADA on three separate occasions and in each case this was masking the presence of drug which was revealed by the acidification. Conclusions: We have shown that two patients from this small cohort of 9 had developed ant