20. Tocilizumab: is out of sight really out of mind?

INTRODUCTION: Giant cell arteritis (GCA) is the most common form of vasculitis in adults with visual complications, in particular loss of vision, constituting the most feared manifestation. Tocilizumab (TCZ), an IL-6 receptor blocker, has been shown to induce lasting remission. However, the optimal duration of treatment with tocilizumab and risk of relapse once treatment is stopped, is currently unknown. We present an unusual case of a patient with GCA who developed visual loss after stopping a three year course of tocilizumab for surgical indications. CASE DESCRIPTION: A 73-year-old lady presented to her GP in 2011 with headache and lower limb stiffness. At that time CRP was greater than 200 and ESR was 126. The GP commenced her on high dose prednisolone with an excellent response. A temporal artery biopsy was not done. She was then referred 14 months later to the rheumatology clinic due to bilateral hip pain and difficulty weaning off steroids, with an increase in inflammatory markers and return of symptoms, upon dose reduction. She was still on 30mg prednisolone at the time of referral. Methotrexate was introduced as a steroid sparing agent, but was stopped a few months later due to side effects. In 2012, while on prednisolone, she developed avascular necrosis of both hips requiring bilateral total hip replacements. An IFR was granted to start TCZ in 2014. She remained well on it with good clinical and serological response and eventually came off steroids completely in 2016. In July 2017, TCZ was stopped for aortic valve replacement for 4 months. This was then restarted at the beginning of December 2017. In February 2018, she developed frontal headaches and PMR-like symptoms. CRP was 2 and ESR 27 at this time. She was commenced on 15 mg daily of prednisolone and TCZ was continued at 8mg/kg. Unfortunately, a few days after her TCZ infusion, she was admitted to hospital with anterior ischaemic optic neuropathy (AION) in the left eye, resulting in partial blindness. Inflammatory markers were found to be completely normal. Intravenous methyl prednisolone was commenced. The patient was referred to a tertiary referral centre and TCZ 162 mg subcutaneously weekly was commenced in combination with leflunomide while tapering the dose of prednisolone. She is currently well with no new visual symptoms. DISCUSSION: This is an unusual case as the patient developed AION whilst on 15 mg of prednisolone following cessation and then re-institution of tocilizumab. The GiACTA trial showed that treatment with TCZ, while extremely efficacious, did not prevent disease flares in all patients, with 24% of patients in TCZ groups experiencing at least 1 or more disease flares. The trial proposed that the flares might be attributable to Th-1 cells, which are not affected directly by IL-6 receptor blockade, and may play an important role in some patients with GCA. Recently, the long-term safety and maintenance of efficacy in GCA patients was assessed in a 2-year extension of the GiACTA trial. Among the 81 patients in the once weekly TCZ arm, who were in clinical remission at week 52, 38 (47%) maintained clinical remission through part 2. However, only 25 of these 38 patients (66%) were treatment-free (no TCZ and no glucocorticoid treatment) at the end of the trial. Similarly, the risk of relapse after discontinuation of TCZ in GCA was assessed in a study by Adler et al. 17 of 20 patients randomised to the TCZ treatment arm were in remission after 12 months. Only half stayed in clinical remission after cessation of TCZ. No clinical or laboratory marker was found to be a reliable predictor of relapse. These two studies show that the odds of relapse after discontinuation are very high, as seen in our case. Furthermore, Adler et al reported that the mean time to relapse was 6.3 months. This shows that the risk of relapse is highest in the first few months, as is the case in our patient. KEY LEARNING POINTS: The optimal strategies for monitoring disease progress in GCA remain uncertain, and currently rely on assessing clinical features and measuring inflammatory markers. However, this has become even more challenging with the use of TCZ, which suppresses the acute phase response via IL-6 receptor blockade. CRP and ESR remain low in nearly all TCZ-treated patients who experience disease flares. This underlines the importance of clinical assessment, particularly poignant history-taking and clinical experience in assessing whether GCA activity might be present. Clinicians should have a very low threshold for suspecting a flare-up, as even the mildest of symptoms could be a harbinger of a full blown flare. As discussed above, studies have shown that effectively half the patients relapse after TCZ is discontinued and most do so within the first few months. This emphasizes the need for careful clinical observation during this time and immediate resumption of treatment in those with new symptoms. Furthermore, it raises the possibility that patients may need to be on TCZ long term and in certain cases indefinitely, given the very high risk of recurrence once the patient is off the drug. Further research and deeper understanding is needed of the pathways that lead to a flare of GCA while on optimal treatment. Finally, this case underscores the dire need for advances in the development of clinically useful biomarkers to help monitor disease activity and predict relapses in patients on TCZ. This case brings to light the fact that AION is possible following cessation and re-institution of tocilizumab. It is axiomatic that clinicians keep a close eye on patients stopping TCZ after the NICE guideline 12 month recommended course. CONFLICT OF INTEREST: The authors declare no conflicts of interest.