Subscribe to our newsletter

Tocilizumab: Answering Current Questions

CorEvitas is now part of Thermo Fisher Scientific. Read More.

by Nancy Walsh Senior Staff Writer, MedPage Today: As treatments for rheumatoid arthritis have multiplied, questions about their optimal use remain. Two experts reported on new findings for the interleukin-6 receptor inhibitor tocilizumab (Actemra) in an online supplement to Arthritis & Rheumatology.

Seoyoung Kim, MD, of Brigham and Women’s Hospital and Harvard Medical School in Boston, addresses the uncertainty as to whether lipid changes with tocilizumab would translate into an increase in cardiovascular events.

(Click here to download full poster)

And Leslie Harrold, MD, of the University of Massachusetts Medical School in Worcester, reports on the question of how tocilizumab monotherapy stacks up against a tumor necrosis factor (TNF) inhibitor plus methotrexate after anti-TNF failure.

(Click here to download abstract)

Tocilizumab and Cardiovascular Disease

Kim: As you know, patients with rheumatoid arthritis have an increased risk of cardiovascular morbidity and mortality. The excess cardiovascular risk seen in patients with rheumatoid arthritis is thought to be related to traditional cardiovascular risk factors, as well as their rheumatoid arthritic activity or severity.

Tocilizumab, which is an interleukin-6 receptor antagonist, is an effective biologic drug that reduces systemic inflammation in patients with rheumatoid arthritis.

In clinical trials of tocilizumab, patients who were treated with the tocilizumab had increased levels of LDL cholesterol. The objective of this study was to determine comparative cardiovascular safety of tocilizumab compared to a tumor necrosis factor inhibitor in patients with rheumatoid arthritis.

To examine cardiovascular safety, we conducted a cohort study of RA patients who newly started tocilizumab versus TNF inhibitors.

We used claims data from Medicare, IMS PharMetrics, and Truven MarketScan. All patients were adults aged 18 years or above, and they had to have either one in-patient diagnosis of rheumatoid arthritis or two outpatient diagnoses of rheumatoid arthritis. Patients were also required to have previously used a different type of [medication] such as different TNF inhibitors, abatacept, or tofacitinib, prior to initiating tocilizumab or TNF inhibitors.

The primary outcome of interest in this study was a composite cardiovascular endpoint of hospitalization for MI (myocardial infarction) or a stroke. This was defined using previously validated claims-based algorithms, which had high specificity with a positive predictive value over 94%.

For the primary as-treated analysis, the follow-up time started the day after initiation of the tocilizumab or TNF inhibitor, and the follow-up time ended on treatment discontinuation plus 30 days as a grace period, and plan disenrollment, outcome occurrence, death, or the study database end date.

To control for a number of different potential confounders such as demographic factors, RA medication use at baseline, their comorbidities, other medication use, and their healthcare utilization intensity, we used a propensity score matching method.

Tocilizumab starters were propensity score matched to TNF initiators with a 1:3 variable ratio within each database. We estimated incidence rates … we estimated the incidence rate with a 95% confidence interval in each database.

We also estimated hazard ratios in each database and combined all three hazard ratios into one using an inverse variance-weight fixed-effects model. The final study cohort included a total of 9,218 tocilizumab starters and 18,810 TNF-inhibitor initiators matched on propensity score.

The mean age was 72 in Medicare, 51 in IMS PharMetrics, and 52 in MarketScan databases. Over 80% of our study patients were women.

At baseline, 72% of patients in Medicare, 73% in IMS PharMetrics, and 66% in Truven MarketScan had a concomitant use of methotrexate. The median follow-up time was about seven months.We observed a total of 36 events in the tocilizumab group and 89 events in the TNF-inhibitor group across three databases.

The combined incidence rate was 0.52 per 100 person-years in the tocilizumab group and 0.59 per 100 person-years in the TNF-inhibitor group across three databases. The risk of composite cardiovascular event associated with the tocilizumab use compared to TNF-inhibitor use was similar across three databases, and the combined hazard ratio was 0.84 with a 95% confidence interval from 0.56 to 1.26.

Our study had limitations. First, even though we controlled for multiple potential confounders using propensity score matching methods, we did not have data on patients’ rheumatoid arthritis duration, activity, seropositivity, family history of cardiovascular disease, or BMI and physical activity.

Second, even though we used the previously validated claims-based algorithm to define the primary outcome, there is still a potential for outcome misclassification.

Thirdly, the median duration of this study was less than one year. Therefore, the long-term cardiovascular safety of tocilizumab versus TNF-inhibitor cannot be answered in this study.

The takeaway message of this study is based on this large, multi-database, population-based cohort study, we did not find evidence that suggests an increased risk of cardiovascular events in tocilizumab starters compared to tumor necrosis factor inhibitor starters. In other words, tocilizumab use is associated with an increased level in LDL cholesterol in patients with rheumatoid arthritis. It does not appear to be related to an increased risk of clinically relevant cardiovascular events.

As a future step, our team plans to conduct several subgroup analysis to determine comparative cardiovascular safety of tocilizumab versus tumor necrosis factor inhibitors in clinically important subgroups such as those with a baseline cardiovascular disease or diabetes, and if so, we will conduct the group analysis stratified by their methotrexate use and other DMARD-use pattern.

Tocilizumab vs Anti-TNF

Harrold: Both patients and providers want to know what’s the best treatment for them. So currently it’s not known if you’ve failed one TNF, what’s the best thing to go on next.

So we compared patients with prior experience to TNF, whether they initiated tocilizumab as monotherapy versus TNF in combination with methotrexate.

We used a national cohort of patients, the Corrona registry, and we identified patients who had prior TNF exposure. They had to initiate either tocilizumab as monotherapy or a TNF in combination with methotrexate at less than 10 mg/week. They also had to have a follow-up appointment at 6 months.

Among that population we used propensity score methodology. We trimmed patients who were very different from each other so we had a very homogeneous cohort of patients and then we did the analyses.

We also used a matched cohort where we matched 1:1 a tocilizumab patient to a TNF patient on methotrexate. In both of those groups, we looked at outcomes at 6 months. That included measures of disease activity, including change in CDAI, change in the mHAQ, and achievement of low disease activity based on the CDAI.

We actually found the patients both did well, the patients in both groups. Those who were on tocilizumab monotherapy were not any different in terms of their responses to those on TNF with methotrexate at less than 10 mg/week.

In terms of limitations, we only included one group of TNF with methotrexate patients at the less than 10 mg dose. We are subsequently now doing analyses with the higher doses of methotrexate and that’s something we are submitting for the European meetings. We just hadn’t gotten to it by the time we submitted to the American meetings.

With the greater amount of methotrexate there may be a difference and so that’s what we’re looking at. Clinically there might be a difference.

We are going to do further comparisons using patients in the tocilizumab monotherapy as composed with TNF with methotrexate, but in the higher methotrexate groups. So the plan is to do the 10 to 15 mg people, the 15 to 20, and the over 20. We’ll do the comparisons in those different groups and see whether there are differences.


Kim and colleagues disclosed financial relationships with Genentech, Pfizer, Bristol-Myers Squibb, AstraZeneca, Lilly, Amgen, Aetion, and Boehringer Ingelheim.

Harrold and colleagues disclosed financial relationships with Pfizer, Roche, Genentech, AbbVie, Amgen, Bristol-Myers Squibb, GlaxoSmithKline, Lilly, Medimmune, Pfizer, and Sanofi.


For any questions contact us at or call us at (508) 408-5435.