PGx7643: Efficacy pharmacogenetic analysis of SLE patients treated with Benlysta in study BEL112341

Trial description: RationaleSystemic lupus erythematosus (SLE) is a chronic autoimmune disorder characterized by autoantibody production and abnormal B lymphocyte function. The etiology of SLE is unknown, although genetics, sex hormones, and environmental conditions are thought to play a role. This disease is more common in women (~90% of patients) than men and more prevalent in African-Americans than Caucasians. Belimumab (also known as LymphoStat-B™; BENLYSTA®) is a B-lymphocyte stimulator (BLyS)-specific inhibitor that blocks the binding of soluble BLyS, a B-cell survival factor, to its receptors on B cells. Belimumab does not bind B cells directly, but by binding BLyS, belimumab inhibits the survival of B cells, including autoreactive B cells, and reduces the differentiation of B cells into immunoglobulin-producing plasma cells.Phase 3 studies of IV-administered belimumab in SLE (BEL110752/BLISS-52/C1056 and BEL11751/BLISS-76/C1057) were completed in 2009 and 2010 and formed the basis of the approval of IV belimumab in the US, Canada and the EU. Pharmacogenetic evaluations were performed on these two phase III studies that evaluated genetic variants across the genome for an association with belimumab efficacy. None of the genetic variants tested met pre-defined thresholds for statistical significance accounting for variation in efficacy response to 10mg/kg belimumab at 52 weeks in either study separately or when meta-analyzed together. The current evaluation will analyze genetic variants across the genome for an association with efficacy in subjects with SLE treated with subcutaneously administered belimumab from study BEL112341. These results will be incorporated into a meta-analysis of the three phase III clinical studies: BEL112341, BEL110752, BEL110751 and reported separately as Study 204901. Genetic Data Generation and ImputationGenetics data was generated using the Affymetrix Axiom Biobank array with GSK custom content v2. Standard quality control exclusions will be applied. Array genotypes will be aligned to the reference strand and phased by chromosome using sequence and genotype data to estimate haplotypes and unobserved genotypes with MaCH v1.0.18.c (Genomics and Human Genetics 2009, 10:387-406). The phased haplotypes will be used to impute genotype dosages using the 1000 Genomes reference haplotypes (phase1_release_v3.20101123 without singletons) and the minimac 2012-11-16 release (Nature Genetics 2012; 44: 955-9). EndpointsTreatment response at week 52 will be characterized by the SLE Response Index (SRI). SRIX response where X is between 4 and 8 is defined as: Responders show >=X point reduction from baseline in SELENA SLEDAI score AND no worsening (increase of <0.30 points from baseline) in PGA AND no new BILAG A organ domain score or 2 new BILAG B organ domain scores compared with baseline at the time of assessment (i.e. week 52).Non-responders show <X point reduction from baseline in SELENA SLEDAI score OR worsening (increase of >=0.30 points from baseline) in PGA OR new BILAG A organ domain score or 2 new BILAG B organ domain scores compared with baseline at the time of assessment (i.e. week 52).Five treatment response endpoints will be assessed:1. SRI48 response: SRI8 responders versus SRI4 non-responders2. SRI4 response: SRI4 responders versus SRI4 non-responders3. SRI ordinal response where a subject is 0 if SRI4 non-responder, 1 if SRI4 through SRI7 responder and 2 if SRI8 responder. 4. SRI4 response in subjects with high disease activity at baseline defined as having a positive anti-ds DNA (>=30 IU/mL at baseline) AND low C3 and/or C4 levels at baseline5. SRI ordinal response in subjects with high disease activity at baseline defined as having a positive anti-ds DNA (>=30 IU/mL at baseline) AND low C3 and/or C4 levels at baselineGenetic AnalysisA logistic regression model will be used to assess the association of genotype with the SRI endpoints assuming an additive genetic model and adjusting for 10 genetic ancestry principle components, baseline SELENA SLEDAI scores and baseline complement (C3, C4) levels. Fifteen candidate variants will be included based on functional significance or strong biological hypotheses. Genome-wide variants with minor allele frequencies >0.01 will be analyzed. Genomic control adjustments will be applied to control for test statistic inflation. Per-allele odds ratios (OR) with 95% confidence intervals will be generated. Pre-specified statistical significance thresholds will be set a p≤0.0013 for candidate variants; p<=3.0×10^-8 for genome-wide variants, which controls the family wise error rate at 0.05 per endpoint analyzed. Analyses of the above endpoints in placebo-treated subjects will also be conducted to allow for estimation of genotype-by-treatment (GxT) interaction effects using appropriate contrasts or linear combinations of the genetic effect size estimates obtained within treatment subgroups.