PROTOCOL AMENDMENT 1 • Added statistical analysis of efficacy and safety of induction phase.•Reorganized sections on analysis of maintenance phase • Removed secondary analysis of response in induction • Replaced randomization by minimization algorithm with randomization by stratification • Clarified number of subjects required for induction phase • Added description of Steering Committee, Clinical Endpoints CEC, and Publications Committee • Added sparse sampling population pharmacokinetics analysis • Clarified preparation of IV cyclophosphamide • Clarified calculation of azathioprine dosing • Rationalized and clarified allowed and prohibited medications • Clarified instructions on contraception • Clarified criteria for use of IV prednisone • Amended Screening Visit -7 days to -1 days prior to Baseline to -10 days to -1 day prior to Baseline • Clarified Visit 9 (end of induction phase) requirements • Clarified requirements for hematology laboratory monitoring per label requirements for induction and maintenance phases • Added hormone tests • Corrected overlap of requirements for urinary protein • Amended reference to registry study • Amended number and location of centers from 50 to 100 worldwide • Added rationale for including patients with Class V lupus nephritis • Revised schedule of assessments for usability and to reflect protocol revisions • Clarified instructions for reporting of serious adverse events • Replaced lower case Roman numeral page numbers in synopsis with continuous numeric page numbers • Renumbered references • Amended clinical phase of study from Phase III /IV to Phase III • Changed Project Leader and Project Statistician • Correction of minor typographic errors

PROTOCOL AMENDMENT 2 • Study title and overall study objectives were updated • Duration of induction phase treatment has been fixed at 24 weeks for all patients • The description of the primary efficacy analysis for the end of induction phase was clarified • Maintenance phase will continue until sufficient subjects have experienced treatment failure to maintain the power of the study, or until the last subject has been followed for 36 months, whichever is earlier • Treatment with pulse IV corticosteroids, plasmapheresis, or intravenous immunoglobulin, formerly permitted within the protocol, is now prohibited • Requirement for discussions with medical monitor were clarified • Required follow-up of subjects withdrawn was added • Inclusion and exclusion criteria were clarified. • Text was added regarding dosing of subjects aged 12-18 years • Low dose aspirin is added as an allowed medication • Serum pregnancy test must be completed within 7 days of Baseline • Description of pharmacokinetic sampling and analysis was expanded • Frequency of the SELENA-SLEDAI was reduced to 3-month intervals • Frequency of some laboratory tests were reduced • ECGs will be determined at Baseline, end of induction/beginning of maintenance, and end of maintenance • GFR will be determined at screening and all visits for which a 24-hour urine will be collected • Conversion of value for serum creatinine >300 umol/l was corrected in suggested dose adjustments for IVC dosing in renal insufficiency • Determination of urine protein:creatinine ratio was clarified • Urinalysis and urine microscopy were added to Schedule of Assessments • Schedule of CBCs was amended during the induction phase to create a consensus schedule fulfilling monitoring requirements for both study treatments, so that timing of CBCs would not reveal treatment assignments to blinded reviewers • Instructions for preparation of IV cyclophosphamide were clarified • Justification for sample size

PROTOCOL AMENDMENT 3 • Amended sample size and statistical analysis according to FDA Special Protocol Assessment requirement that primary efficacy analysis be for superiority rather than non-inferiority • Amended long title to reflect induction phase being unblinded • Clarified definition of treatment response (primary endpoint of induction phase and required for entry into maintenance) • Clarified IVC dosing, number of doses, target dose, and adjustment of doses • Clarified that secondary efficacy endpoints during the induction phase would be analysed descriptively • Corrected footnote in Table of Assessments for induction regarding frequency of pregnancy tests • Amended reference to BILAG from 'organ' to ‘system’ in secondary endpoints • Corrected glossary of abbreviations • Corrected typographic error in Appendix 3 • Corrected arithmetic error in conversion of serum creatinine units

PROTOCOL AMENDMENT 4 • A full pharmacokinetic (PK) profiling for a subset of subjects was added.

PROTOCOL AMENDMENT 5 • Temporary dose stoppage of IVC was allowed for no more than 7 days in total. • Urine protein to creatinine ratio was clarified to specify that the ratio was derived from the 24 hour urine protein and creatinine measurements. • Hierarchical testing for the key secondary endpoints in the induction phase was included in the statistical analysis following feedback from the FDA. • The interpretation of what constitutes high dose phenobarbital was clarified to reflect that it was the judgment of the investigators. • The Systemic Lupus Erythematosus Disease Activity Index (SLEDAI) flare instrument was removed and the procedure for calculating the total for the SLEDAI was corrected. The flare instrument that was part of the Safety of Exogenous Estrogens in Lupus Erythematosus National Assessment (SELENA)-SLEDAI scale, which appeared in Appendix 7 of the protocol but was not used for the study, was deleted due to the potential for confusion with study flare definitions. • The visit window at the beginning of induction was increased. Due to practical difficulties regarding the availability of laboratory results required for randomization, the window for the beginning of the induction visit (Visit 2) was increased from 10 to 14 days from Screening and the end of induction (Visit 9) was increased from 7 to 10 days. • The BILAG worksheet in Appendix 5 of the protocol was updated to conform to the latest version that was used in the clinical study.

PROTOCOL AMENDMENT 6 Protocol Amendment 6, dated April 12, 2007 was approved by the IECs/IRBs before implementation. Last subject, last visit on March 16, 2007 was completed but the database had not been locked prior to implementation of this amendment. • The primary analysis for the induction phase of protocol was changed from a Fisher’s Exact Test comparing the proportion of responders in the two treatment groups, to a logistic regression comparing the proportion of responders in the two treatment groups. The rationale for this change in statistical test was that the logistic regression included the covariates race (Caucasian/Asian/Other), and World Health Organization (WHO) LN Class (V only, others) (the stratification variables used for the randomization), and geographical region (US/Canada, Latin America, Asia and Rest of the World). Analysis with these covariates is common in LN trials and provided a more meaningful analysis of the response to treatment than the Fisher’s Exact Test. Fisher’s Exact Test is used primarily when the expected number of subjects in a treatment by response cell is less than five, which was unlikely to be the case in this study. • Partial remission was one of the secondary efficacy parameters for efficacy assessments at the end of the induction phase. This parameter of partial remission was removed from the protocol, since it was felt that the primary endpoint of response was more clinically meaningful and more robust. • For the purpose of clarification, the complete remission criteria were amended to “proportion of subjects meeting remission criteria for each individual parameter AND proportion of subjects meeting at least one criterion for complete remission”. The secondary efficacy parameter was changed from the “number of” subjects to the “proportion of” subjects. • Clarification of the hierarchical testing of the secondary efficacy variables in the induction phase was added to the synopsis.