BackgroundTofacitinib is an oral JAK inhibitor for the treatment of RA. There is no direct comparison of tofacitinib monotherapy vs tofacitinib +MTX in MTX inadequate responders (IR) and limited data comparing tofacitinib (±MTX) vs adalimumab (ADA) +MTX in patients (pts) with RA.ObjectivesTo compare efficacy and safety of tofacitinib monotherapy, tofacitinib+MTX, and ADA+MTX in a head-to-head, non-inferiority trial in MTX-IR pts.MethodsIn this randomised, triple-dummy, active-controlled, 1-year, Phase 3b/4 trial (ORAL Strategy; NCT02187055), pts had active RA (≥4 tender/painful joints on motion and ≥4 swollen joints [28-joint count] at baseline [BL]) inadequately controlled with MTX. Pts were randomised 1:1:1 to receive tofacitinib 5 mg twice daily (5 mg mono BID), tofacitinib 5 mg BID +MTX (5 mg BID+MTX) or subcutaneous ADA 40 mg every other week +MTX (ADA+MTX); MTX dose: 15–25 mg/wk. The primary endpoint was ACR50 at Month (Mo) 6. Non-inferiority between treatments was declared if the lower bound of 98.34% two-sided confidence intervals of the difference of ACR50 response at Mo 6 was larger than -13% (based on meta analysis of ADA trials1), and superiority if it was larger than 0%. Other endpoints included: ACR20/50/70 and least-squares mean changes from BL in SDAI, DAS28-4(ESR) and HAQ-DI at Mos 6 and 12. Safety was assessed throughout the trial.Results1146 pts were randomised and treated (5 mg mono BID: n=384; 5 mg BID+MTX: n=376; ADA+MTX: n=386). Demographics and BL disease characteristics were similar across groups. Most pts were female (82.7–83.1%), white (75.9–77.1%), with a mean age of 49.7–50.7 years, median disease duration of 5.4–6.1 years and mean HAQ-DI score of 1.6. Across groups, 80.2–81.6% of pts completed the study. ACR50 response rate at Mo 6 was 38.3% for 5 mg mono BID, 46.0% for 5 mg BID+MTX and 43.8% for ADA+MTX. Non-inferiority was demonstrated for 5 mg BID+MTX vs ADA+MTX (P<0.0001) but not for 5 mg mono BID vs ADA+MTX (P=0.0512) or 5 mg mono BID vs 5 mg BID+MTX (P=0.2101) which, although numerically different, were not statistically different (Figure). Tofacitinib monotherapy achieved the efficacy expected of an effective immunomodulator in this pt population. Secondary efficacy analyses were generally consistent with the primary analysis (Table). Adverse event (AE), serious AE and discontinuation due to AE rates were generally clinically similar across groups, though numerically fewer pts had increased alanine aminotransferase with 5 mg mono BID vs 5 mg BID+MTX or ADA+MTX.ConclusionsTofacitinib 5 mg BID+MTX was as effective as ADA+MTX in MTX-IR pts with RA. However, clinical outcomes of all 3 regimens, including tofacitinib 5 mg BID monotherapy, were comparable. There were no new or unexpected safety issues.References Machado et al. Rev Bras Reumatol 2013;53:419–430. AcknowledgementsThis study was funded by Pfizer Inc. Editorial support provided by D Binks of CMC.Disclosure of InterestR. Fleischmann Grant/research support from: Abbott, Amgen, Astellas, Bristol-Myers Squibb, Boehringer Inge...

The magnitude of risks associated with NSAIDs is small and similar in celecoxib-, nsNSAID- and placebo-treated patients. This analysis provides safety information that will allow physicians to make informed treatment decisions for patients who are appropriate candidates for celecoxib use.

Incorporation of the RSD in AD trials is feasible. With proper trial setup and statistical procedures, this design could support the detection of a disease-modifying effect. In our opinion, a two-phase RSD with a stepwise hypothesis testing procedure could be a reasonable option for future studies.

Novel experimental assays now simultaneously measure lineage relationships and transcriptomic states from single cells, thanks to CRISPR/Cas9-based genome engineering. These multimodal measurements allow researchers not only to build comprehensive phylogenetic models relating all cells but also infer transcriptomic determinants of consequential subclonal behavior. The gene expression data, however, is limited to cells that are currently present ("leaves" of the phylogeny). As a consequence, researchers cannot form hypotheses about unobserved, or "ancestral", states that gave rise to the observed population. To address this, we introduce TreeVAE: a probabilistic framework for estimating ancestral transcriptional states. TreeVAE uses a variational autoencoder (VAE) to model the observed transcriptomic data while accounting for the phylogenetic relationships between cells. Using simulations, we demonstrate that TreeVAE outperforms benchmarks in reconstructing ancestral states on several metrics. TreeVAE also provides a measure of uncertainty, which we demonstrate to correlate well with its prediction accuracy. This estimate therefore potentially provides a data-driven way to estimate how far back in the ancestor chain predictions could be made. Finally, using real data from lung cancer metastasis, we show that accounting for phylogenetic relationship between cells improves goodness of fit. Together, TreeVAE provides a principled framework for reconstructing unobserved cellular states from single cell lineage tracing data.

BackgroundTofacitinib is an oral JAK inhibitor for the treatment (tx) of rheumatoid arthritis (RA). Studies have shown diminishing response to tx in RA patients (pts) when cycling through TNF inhibitors. Prior analyses assessed tofacitinib in csDMARD-inadequate response (IR) pts vs overall bDMARD-IR pts.ObjectivesTo compare tofacitinib safety and efficacy in RA pts who have previously failed tx (lack of efficacy and/or safety reasons) with csDMARDs, with pts who failed tx with either 1 or ≥2 prior bDMARDs.MethodsData from pts who received ≥1 dose of tofacitinib in 19 RA studies up to 96 months (2 Phase [P] 1; 9 P2; 6 P3; 2 LTE studies [1 LTE ongoing; data as of March 2015]) were used in this analysis. Data were pooled across all 19 studies for safety assessments in the All RA population: csDMARD-IR, n=4377; bDMARD-IR, n=838 (1 bDMARD-IR, n=533; ≥2 bDMARD-IR n=305). Safety was also assessed up to 24 months in pts randomised to tofacitinib 5 or 10 mg BID or placebo (PBO) in a pooled P2/P3 randomised controlled trial (RCT) population (8 P2, 6 P3 studies; csDMARD-IR, n=3328; bDMARD-IR, n=782). Incidence rates (pts with events/100 pt-years) were calculated for serious AEs (SAEs), serious infections (SIs) and herpes zoster (HZ). Efficacy was assessed by pts achieving ACR20 response and DAS28-4(ESR) ≤3.2 at Month 3 in a pooled P3 RCT population (csDMARD-IR, n=2375; bDMARD-IR, n=664).ResultsPrior to tofacitinib tx, bDMARD-IR pts had longer RA duration, greater disease burden and more corticosteroid use vs csDMARD-IR pts. SAEs were more common among bDMARD-IR vs csDMARD-IR pts in both the P2/P3 RCT and the All RA populations; SAE rates were not higher in pts failing ≥2 bDMARDs vs 1 bDMARD (Table). Incidence rates for SIs were generally greater in pts with IR to bDMARDs vs csDMARDs in the All RA population, but generally lower in pts with IR to 1 or ≥2 bDMARDs vs csDMARDs in the P2/P3 RCT population; incidence with 5 mg BID was lower for 1 vs ≥2 bDMARDs in the P2/P3 RCT population. Incidence rates for HZ were similar between pts with IR to csDMARDs or 1 bDMARD, but appeared numerically greater in pts with IR to ≥2 bDMARDs in both the P2/P3 RCT and the All RA populations. A similar pattern was observed across tofacitinib and PBO groups. Efficacy at Month 3 in the P3 RCT population was greater with both tofacitinib doses vs PBO. Although absolute response was smaller in pts with IR to bDMARDs vs csDMARDs, generally similar efficacy was observed in pts with IR to 1 or ≥2 bDMARDs (Table).ConclusionsSAEs and SIEs were more common in tofacitinib- and PBO-treated pts with IR to bDMARDs vs csDMARDs; increased risk was generally not observed with increasing number of prior bDMARDs. HZ risk appeared greater with ≥2 bDMARDs vs 1 bDMARD. Efficacy was greater with tofacitinib vs PBO for csDMARD-IR and bDMARD-IR pts, with similar response observed in pts with IR to ≥2 or 1 bDMARD. These data support the use of tofacitinib in different lines of therapy, although the analysis is limited by smaller sample size in some groups.Acknowledgement...