Background Tofacitinib is an oral Janus kinase inhibitor for the treatment of rheumatoid arthritis (RA). The limited size of the control groups (placebo and active comparator) and limited duration of treatment within the tofacitinib RA clinical trial programme do not permit precise direct comparative assessments for adverse events of long latency, including malignancies, therefore a meta-analytic approach was taken.
Objectives To compare the rate of malignancies in patients (pts) with moderately to severely active RA within the tofacitinib RA clinical trial programme with estimates from published trial data of approved biologic (b)DMARDS using meta-analytic methods.
Methods Incidence rates (IR; pts with events per 100 pt-years [yrs] exposure) for tofacitinib were calculated using pooled data from randomised controlled trials (RCT) and long-term extension (LTE) studies. Two Phase (P) 1 studies, 9 P2 studies, 6 P3 studies and 2 LTE studies (one study ongoing; database unlocked at March 2015 data cut off) constituted the P123LTE dataset; pts received 1, 3, 5, 10, 15, 30 mg BID or 20 mg QD of tofacitinib. A systematic literature review of published RCT and LTE studies (through August 2014) of approved bDMARDs (abatacept, adalimumab, certolizumab, etanercept, golimumab, infliximab, rituximab and tocilizumab) was conducted. IRs for the endpoint of interest (all malignancies [excluding non-melanoma skin cancers (NMSC)]) were calculated for each bDMARD utilising a random-effects meta-analytic model with a restricted Maximum Likelihood Estimator for between-study variances. Estimates of Q statistic and I2 were generated to determine the existence and degree of heterogeneity related to the study pool.
Results The tofacitinib P123LTE dataset included 6194 pts with a total exposure of 19 385 pt-yrs. A total of 64 bDMARD articles were extracted for analysis, representing 58 unique studies and approximately 27 000 pts. Study populations were generally consistent across studies and treatments: mean pt age ranged from 51–54 yrs; mean percentage of pts who were female was 74–84%; and mean baseline C-reactive protein level was 20–31 mg/L. The IR (95% confidence interval [CI]) of malignancy for tofacitinib in the P123LTE dataset was 0.89 (0.76, 1.04) (Figure). Estimated IRs (95% CI) of malignancy were 0.75 (0.56, 1.01) for abatacept, 1.06 (0.41, 2.74) for rituximab, 1.02 (0.69, 1.52) for tocilizumab and 0.95 (0.79, 1.14) for tumour necrosis factor inhibitors (adalimumab, certolizumab, etanercept, golimumab and infliximab) (Figure).
Conclusions The results of this meta-analysis indicate that the tofacitinib IR for all malignancies (excluding NMSC) in pts with moderately to severely active RA is within a similar range to those reported in published interventional studies of similar RA populations treated with approved bDMARDs.
Acknowledgements This study was sponsored by Pfizer Inc. Editorial support was provided by N Jones and C Viegelmann of CMC and was funded by Pfizer Inc.
Disclosure of Interest J. Gomez-Reino Grant/research support from: AbbVie, Novartis, Pfizer Inc, Roche, UCB, Consultant for: Pfizer Inc, Speakers bureau: AbbVie, Bristol-Myers Squibb, Janssen, MSD, Pfizer Inc, Roche, UCB, T. Checchio Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, J. Geier Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, M. Boy Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, S. Ahadieh Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, S. Menon Shareholder of: Pfizer Inc, Employee of: Pfizer Inc, S. Krishnaswami Shareholder of: Pfizer Inc, Employee of: Pfizer Inc