Patients recruited in Asia were also at a higher risk of SIEs than were those recruited outside of Asia. Because nearly all Asian patients were recruited in the Asian region, we were unable to distinguish between geographic effects and ethnicity. In addition, the low number of SIEs in the DBPC period meant that we had limited statistical power in the analyses of interactions of risk factors, such as Asian region with treatment. Confounding factors may have contributed to the higher incidence of opportunistic infections such as endemic areas for histoplasmosis in the United States, tuberculosis in Mexico, and hepatitis B in Japan. In addition, the patient with Candida infections was receiving highdose steroid treatment for concurrent medical conditions. The clinical GDC-0879 development of OCR was initiated in part with the aim of evaluating the potential safety advantage of a humanized molecule over chimeric antibodies. Humanization may be expected to reduce the incidence of anti-drug antibody responses. The incidence of HAHAs was low across the 4 trials and, in general, comparable between the pooled OCR+MTX and PBO+ MTX groups. There was no association between IRRs and development of HAHAs. In addition, there were no clear differences in the incidence of HAHAs when single-infusion and dual-infusion OCR were compared, although, because the patient numbers in FEATURE were small, the question of whether a difference exists between single- and dual-infusion OCR remains open. In a previous pooled analysis of approximately 2500 patients in the rituximab RA clinical trial program, 11% of those treated with rituximab developed human anti-chimeric antibodies. As expected, both doses of OCR rapidly depleted B cells shortly after infusion. The question was whether the higher rates of serious infections seen in patients treated with OCR500+MTX could have been explained, in part, by differences in B-cell depletion/ repletion profiles between the higher and lower doses. It should be noted that evaluation of B-cell levels in clinical trials is limited by measurement of peripheral CD19 counts only; however, the analyses suggested that there was no difference in time to peripheral B-cell repletion between the OCR500 and OCR200 doses. Moreover, the number of repeat treatment courses also did not seem to have a clinically meaningful effect on time to B-cell repletion. The conclusion that the two doses of OCR, in combination with MTX tested in the RA clinical trials did not demonstrate a superior benefit-risk profile compared with available treatments led to the termination of the clinical development program of OCR in RA. OCR500+MTX demonstrated clinical benefit by improving signs and symptoms of RA and radiographic outcomes; however this dose was associated with an increased incidence of SIEs. OCR200+MTX did not show superior efficacy compared with existing therapies, but was safe and well-tolerated. The clinical development of OCR is continuing in multiple sclerosis, for which there remains an unmet need for more effective therapies and background immunosuppressant therapy is not used.