FDA Arthritis Panel Split on the Efficacy and Safety of Baricitinib in RA Save

On Monday, April 23rd the FDA convened the Arthritis Advisory Committee (AAC) to evaluate Lilly’s resubmitted NDA for the approval of the JAK inhibitor baricitinib for use in rheumatoid arthritis (RA). The panel included 15 voting members that included 7 rheumatologists, 3 epidemiologist-statisticians, a hematologist, 2 patient representatives and 2 pharmacologists.

The sponsor was seeking FDA approval for the use of baricitinib in the treatment of adult patients with moderately to severely active rheumatoid arthritis (RA) who have had an inadequate response to methotrexate.

The FDA presentations favored the efficacy of baricitinib 2 mg and 4 mg doses based on clinical (ACR 20/50/70, DAS, SDAI and CDAI) and functional (HAQ) scores. They also noted the “robust” radiographic data favoring the 4 mg dose, but found less certain X-ray benefits with the 2 mg dose.  By contrast, FDA concerns about the safety of the baricitinib dominated much of the discussion, especially the imbalance of venous thromboembolic events (VTE) with the 4 mg dose (6 VTE) compared to none on placebo or 2 mg baricitinib in the first 16 weeks of randomized therapy. Other safety concerns, some of which were dose related, were also covered.

The Voting

After detailed data presentations by both Lilly and the FDA, the AAC voted on the efficacy, safety and overall benefit: risk approval of the 2 mg and 4 mg daily dose of baricitinib in RA.

The FDA asked the panel to consider the efficacy of the 2 mg and 4 mg dose against placebo and consider the evidence for added efficacy when comparing 4 mg and 2 mg dosing.

Efficacy: There were 14 votes in favor and 1 vote against the EFFICACY of baricitinib 2 mg per day dose.  There were 15 votes in favor and none (zero) against the EFFICACY of the 4 mg per day dose.

The safety of baricitinib was presented including long term extension data, comparisons to currently marketed biologics and tofacitinib and real world claims data on VTE risk. 

Safety: There were 9 votes in favor and 6 votes against adequate evidence to support the SAFETY of baricitinib 2 mg; but there were only 4 votes in favor and 11 votes against evidence to support the SAFETY of the 4 mg dose for the requested indication.

The totality of issues was considered and the panel was asked to vote on whether baricitinib should be approved at either the proposed dosing of baricitinib 2 mg or 4 mg per day.

Approval: Based on an overall benefit: risk analysis there were 10 votes in favor and 5 votes against the APPROVAL of baricitinib at the 2 mg dose. Much less support was seen for the 4 mg dose with only 5 votes in favor but 10 votes against the APPROVAL of baricitinib at the 4 mg dose. 

Background

Baricitinib is a novel Jak 1 and Jak 2 inhibitor that has been developed for use in RA.  The drug has a ½ life of 12 hours and 75% excreted in the urine.  After a series of phase 2 dose ranging studies the sponsor developed a phase 3 program with 4 randomized controlled trials using 4 mg in all trials, 2 mg in 2 trials and another trial compared baricitinib 4 mg to adalimumab (ADA) in MTX incomplete responders (IR).

The 4 pivotal phase 3 trials are shown below.

Throughout the day the FDA reviewers noted that both “baricitinib doses (2 mg, 4 mg) are effective in reducing the signs and symptoms and improving physical function”, pointing to the efficacy endpoints of ACR 20, 50, 70 and HAQ.  While the X-ray benefits of 4 mg were “robust”, the FDA was less certain about the protective effects offered by the 2 mg dose.

Therefore, the problem of approvability was not with efficacy data, it was with the safety findings (described below) and with the applicants resubmitted NDA asking for a new wording for the drugs indication. The newly proposed indication stated, “baricitinib is indicated for the treatment of adult patients with moderately to severely active rheumatoid arthritis who have had an inadequate response or intolerance to methotrexate.

The problem was that the 4 mg data did not consistently outperform the 2 mg dose groups.  In support of the notion that some patients would need the 4 mg dose over the 2 mg dose, Lilly showed data from the JADW study with an incremental advantage of 4 mg over 2 mg (especially for change in DAS28-CRP and SDAI). While the JADX study showed no difference in response rates between 2 mg and 4 mg groups, the company showed that non-differences were observed in those who had failed a single prior DMARD but that those who had failed 2 or more DMARDs had significantly better responses at 4 mg per day. 

Lastly the company showed what happened to 4 mg/day patients (n 874) who rolled over into the long term extension study (JADY), they were either continued on 4 mg or blindly tapered to 2 mg/day (n 439). While the majority of those tapered remained in remission or LDAS for up to 48 weeks, the 4 mg group was 12% better than the 2 mg group by week 48; suggesting a small but meaningful differential effect.

The FDA did not support the need for 2 doses, citing the efficacy data showing the differences were nominal or infrequent.  Moreover, the safety review suggested more VTE events and lab abnormalities in the higher 4 mg dose treated patients. 

The background risk of VTE (either PE or DVT) in the RA population is likely to be 5-6 VTE per 1000 Patient-years.  During the first 16 weeks of the placebo controlled trials there were 6 VTE in the 4 mg group, but not in the placebo or 2 mg treated patients. This calculated to 17 VTE/1000 PY for short term exposure of 4 mg baricitinib.  However when all patients were followed between 0-52 weeks this rate fell to 8 per 1000 PY and after week 52 fell to 5 per 1000PY.  

A total of 42 VTE events were observed in their database of 3464 treated patients (7860 pt-yrs exposure). Lilly’s analysis of the issue failed to identify a pathogenic link whereby a Jak 1, Jak 2 inhibitor could result in thrombosis. Moreover they showed that in their data set, risk factors for VTE were; prior VTE, Cox-2 inhibitor therapy, increasing age and BMI (all known risk factors for VTE).

Also quizzical, but unrelated to VTE, was the observation that baricitinib initiation was followed by a 2 week rise in platelet counts, generally rising no more than 60,000 cells/mm³.  The percentage of patients with platelet counts > 450,000 was only 9.9-15.8% and this did not correlate with VTE or any other clinically identifiable toxicity.  Other safety concerns, similar to that seen with tofacitinib, were seen including modest LFT or CPK elevations and hyperlipidemia.

When comparing the safety of the 2 and 4 mg doses there were no differences found in overall adverse events (AE), serious AE, serious infections (SIE), and deaths.  The SIE rate was 3.8 per 100 PY. There were 87 cases of  H. zoster, at a rate of 3.7/100PY. Overall 11 case of TB were described (0.14/100PY), opportunistic infection in 0.3 per 100PY and cancer in 0.7 per 100 PY.

Hence the finding of more VTE events (early on) in the 4 mg dose group and a few instances of dose related increases in labs (platelets, CPK) shed some safety concern over the 4 mg dose; the dose that Lilly felt was the more effective dose and was best studied within their portfolio.

While the voting indicates greater overall confidence in the safety and efficacy of the 2 mg dose, there are only 2 studies and fewer patients treated at this dose. Only 403 patients received 2 mg in the JADX and JADW studies. Overall 723 patients were exposed to 2 mg/day (1275 patient-yrs exposure).   

It remains to be seen what decisions will be made by the FDA following this AAC advisory hearing, in which they stated they found the discussions, comments and voting of the committee to be most helpful.