What are the benefits and risks of rituximab (a medicine that reduces inflammation) for the treatment of thyroid-associated ophthalmopathy (a disease affecting the tissues surrounding the eyes)

Key messages
- We did not find enough evidence to show that people with moderate to severe thyroid-associated ophthalmopathy who have treatment with rituximab have any different outcome in terms of signs and symptoms of disease compared to those treated with steroid or placebo (dummy treatment). 

- Rituximab may cause more adverse (unwanted) effects compared to steroids or a placebo, but we are not certain of this.

- More studies are needed to strengthen the body of evidence and investigate whether rituximab should be used for the treatment of thyroid-associated ophthalmopathy.

What is thyroid-associated ophthalmopathy?
Thyroid-associated ophthalmopathy is an eye disease that affects half of people that have Graves' disease, a condition where the body's immune system attacks the thyroid gland and makes it overactive. In thyroid-associated ophthalmopathy the body's immune system also attacks the tissues surrounding the eyes. Symptoms include eye pain, redness, swelling, eye protrusion (proptosis), double vision, and in severe cases, reduction in vision. 

How is thyroid-associated ophthalmopathy treated?
Treatment options for severe thyroid-associated ophthalmopathy include anti-inflammation medicines called steroids, and radiotherapy.  However, relapses are common. Surgery to relieve the pressure that builds up behind the eye (orbital decompression) is usually reserved for sight-threatening cases. 

A possible alternative treatment is a medicine called rituximab, given as an intravenous infusion (injection into the vein), which aims to stop the body’s defence system (immune system) from mistakenly attacking its healthy tissues and causing inflammation.

What did we want to find out?
We wanted to compare the benefits and risks of rituximab against those of other treatments for thyroid-associated ophthalmopathy.

What did we do?
We searched for studies that compared rituximab against placebo (dummy) treatment or steroid treatment for thyroid-associated ophthalmopathy. We compared and summarised the results of these studies and rated our confidence in the evidence, based on factors such as study methods and sizes of participant groups.

What did we find?
We found two studies, conducted in Italy and the USA, involving a total of 56 people with thyroid-associated ophthalmopathy. Everyone taking part in the study was aged 18 to 80 years (average age 55 years) and 77% were women. One study (31 people) compared intravenous rituximab (2 x 1000 mg infusions given 2 weeks apart or a single infusion of 500 mg) to intravenous methylprednisolone (steroid). The other study (25 people) compared intravenous rituximab (2 x 1000 mg infusions given 2 weeks apart) to intravenous placebo (dummy) infusions. As the studies compared rituximab to different alternative treatments, it was not possible to combine the studies' results to analyse them together.

What were the main results of our review?
The evidence suggests that rituximab, compared to intravenous methylprednisolone, may result in an improvement in how 'active' the thyroid-associated ophthalmopathy is at 24 weeks following treatment when measured using one particular scale, but the evidence for whether there was a true difference was uncertain. Furthermore, there was little to no difference when thyroid-associated ophthalmopathy activity was measured using a different scale. There was no evidence of a difference in the following outcomes, measured at 24 weeks following treatment:

- reduction in eye protrusion 

- reduction in how wide open the eyelids are

- improvement in eye movement and double vision

- improvement in quality of life

The evidence suggests that rituximab, compared to placebo, results in little to no difference in any of the outcomes outlined above, measured at 24 weeks following treatment.

The number of adverse events was higher for rituximab compared to both intravenous methylprednisolone and placebo, but the evidence for whether there was a true difference was uncertain.

What are the limitations of the evidence?
The evidence is only based on two studies with few participants, and both of these studies were stopped earlier than originally planned, which limits our confidence in the findings. 

How up-to-date is the evidence?
The evidence is current to February 2022.

Authors' conclusions: 

There is currently insufficient evidence to support the use of RTX in people with TAO. Future studies investigating RTX in people with active TAO may need to be multi-centre in order to recruit enough participants to make an adequate judgement on the efficacy and safety of this novel therapy.

Read the full abstract...

Thyroid-associated ophthalmopathy (TAO) is the most frequent extrathyroidal manifestation of Graves' disease, affecting up to 50% of patients. It has a great impact on quality of life. Rituximab (RTX) is a human/murine chimeric monoclonal antibody that targets the CD20 receptor on B-lymphocytes. Preliminary work has shown that blocking this CD20 receptor with RTX may affect the clinical course of TAO by reducing inflammation and the degree of proptosis. 


This review update, originally published in 2013, assesses the efficacy and safety of using RTX for the treatment of TAO.

Search strategy: 

We searched the Cochrane Central Register of Controlled Trials (CENTRAL; 2022, Issue 2), which contains the Cochrane Eyes and Vision Trials Register, Ovid MEDLINE, Ovid Embase, Latin American and Caribbean Health Science Information database (LILACS), the ISRCTN registry, clinicaltrials.gov and the WHO International Clinical Trials Registry Platform (WHO ICTRP). There were no language restrictions in the electronic search for trials. We last searched the electronic databases on 22 February 2022. 

Selection criteria: 

We included randomised controlled trials (RCTs) of RTX administered by intravenous infusion using any dosage regimen for the treatment of active TAO in adults, compared to placebo or glucocorticoids treatment. 

Data collection and analysis: 

We used standard methodological procedures expected by Cochrane. Two review authors independently scanned titles and abstracts, and screened full-text reports of potentially relevant studies. The outcomes of interest in this review were: clinical activity score (CAS), NOSPECS severity scale, proptosis (mm), palpebral aperture (mm), extraocular motility (degrees or diplopia rating scale), quality of life and adverse effects.

Main results: 

We identified two studies that met the inclusion criteria in this updated review. Across both studies, the mean age of participants was 55 years and 77% were women.

RTX compared to intravenous methylprednisolone (IVMP)

One study, conducted in Italy, compared RTX (n = 15 after one participant withdrew) with IVMP (n = 16) for active TAO (CAS ≥ 3 out of 7 or 4 out of 10). We judged this study to be at low risk of bias in most domains, but it was stopped early because of disease reactivation in the comparator group (5/16 participants). This study provided low-certainty evidence that RTX may result in CAS improvement at 24 weeks compared to IVMP (15/15 versus 12/16 improved by ≥ 2 points; risk ratio (RR) 1.32, 95% confidence interval (CI) 0.98 to 1.78). Only very low-certainty evidence was available for the other outcomes: NOSPECS improvement by 2 or more classes (3/15 versus 3/16; RR 1.07, 95% CI 0.25 to 4.49); proptosis improvement by 2 mm or more (0/15 versus 1/16; RR 0.35, 95% CI 0.02 to 8.08); palpebral aperture improvement by 3 mm or more (2/15 versus 0/16; RR 5.31, 95% CI 0.28 to 102.38); motility improvement by 1 class or more (3/15 versus 3/16; RR 1.07, 95% CI 0.25 to 4.49); and improvement on the Graves’ ophthalmopathy QoL scale by at least 6 points for "functioning" (5/14 versus 8/13; RR 0.58, 95% CI 0.25 to 1.32), and “appearance” (9/14 versus 6/13; RR 1.39, 95% CI 0.69 to 2.82). Adverse events were more common in the RTX group (RR 1.39, 95% CI 0.90 to 2.13; low-certainty evidence). Minor adverse effects (mild infusion reactions) were observed in most people receiving RTX at first infusion. Two participants experienced a major infusion reaction, likely cytokine release syndrome.

RTX compared to placebo

One study, conducted in the USA, enrolled 25 participants with active TAO (CAS ≥ 4 out of 7), comparing RTX (13 participants) to placebo. We judged this study to be at low risk of bias in most domains, but it was stopped early due to recruitment issues. It provided very low-certainty evidence on the following outcomes at 24 weeks: CAS improvement by 2 or more points (4/13 RTX versus 3/12 placebo; RR 1.23, 95% CI 0.34 to 4.40); NOSPECS improvement by 2 or more classes (2/13 versus 2/12; RR 0.92, 95% CI 0.15 to 5.56); proptosis improvement by 2 mm or more (2/13 versus 4/12; RR 0.46, 95% CI 0.10 to 2.08); palpebral aperture median change (0 mm in RTX group, in both eyes separately, versus -0.5 mm and 0.5 mm in placebo group right and left eye, respectively); motility median diplopia score (3 versus 2.5); SF-12 physical component median score (45.9 versus 40.3) and mental component median score (52.8 versus 46.1). More participants in the RTX group experienced adverse effects (8/13 versus 3/12; RR 2.46, 95% CI 0.84 to 7.18).