In this randomized, double-blind, placebo-controlled trial, although there was no statistically significant difference in the primary endpoint (the number of fever attacks), the TCZ group was more effective than the placebo group in most endpoints. During the double-blind period, 10 of 13 fever attacks in the TCZ group and 11 of 20 in the placebo group were identified by week 8, and there was a trend toward suppression of fever attacks in the TCZ group compared to the placebo group as the treatment period increased. This result may indicate that the attacks appeared before the blood concentration of TCZ stabilized and the drug effect was fully exerted. In line with this observation, the percentage of patients who entered the rescue phase was higher in the placebo group compared to that of the TCZ group. In the placebo group, there were more patients who required intervention for increased fever attacks after week 14. The failure to validate the effect of TCZ in the primary endpoint was impacted by the lower-than-expected estimated number of fever attacks in the placebo group (0.113 per week). This was influenced by the small number of patients, given that half of the patients in the placebo group were attack-free.
The results of the subsequent open-label extension study showed that the numbers of fever attacks and accompanying symptoms, such as pleurisy and peritonitis, tended to be generally controlled in the long term both in patients who continued to receive TCZ from the preceding study and in those who switched to TCZ. Consistent with the results of this study, there are many reports of efficacy in the literature, including case reports of TCZ [19–22, 30, 31].
Currently, IL-1 inhibitors, including canakinumab, have been shown to be effective in treating crFMF [32, 33], but in some cases, they are not effective enough or cannot be used due to adverse events. IL-1 inhibitors have also been shown to be effective against amyloidosis [34, 35]. It has been proposed that IL-6 inhibitors may have a higher potential to normalize SAA than other biologic agents [36] and IL-6 inhibitors have been suggested to be useful in preventing the progression of amyloidosis and improving amyloid deposition [21, 37].
In the double-blind period, 2 serious adverse events were observed in 2 patients in the TCZ group (myocarditis, headache) and 1 in the placebo group (hypoglycemia). The following is the detailed history of a case of myocarditis. After 4 weeks of treatment with the investigational drug, the patient developed chest pain, mild ST-segment elevation at V1-4 on electrocardiogram, and cardiac enlargement on plain chest radiograph. Therefore, the patient was evaluated by cardiac catheterization. The catheterization showed no evidence of infarction and myocarditis was suspected, so a myocardial biopsy was performed, which showed neutrophilic inflammation. The cardiologist suggested systemic administration of steroids for progressive myocarditis. Since it was difficult to evaluate this study under moderate doses of steroids, the investigator decided to discontinue the study drug. The myocarditis improved with prednisolone 30 mg oral administration, and prednisone was tapered off. Except for the case of myocarditis, there was no causal relationship with the investigational drug, and the patients recovered with treatment. Although the duration of treatment and the numbers of patients in the prior and continuing studies are limited, the safety profile of TCZ in our research is similar to that in the information available to date [38, 39], and TCZ is considered to be well tolerated in patients with FMF.
There are several limitations to this study. First, the sample size was small, resulting in low statistical power and uncertainty in the results. Second, for patients who did not have fever attacks during the observation period, it was assumed that the frequency of fever attacks would remain low after entry, and therefore, reconsent for the same patients should not be allowed. Third, we should have considered an analysis in which attacks up to 4–8 weeks, when blood levels of TCZ were stable, were not included in the evaluation. Fourth, the high response to placebo could be attributed to differences in the baseline colchicine doses or the number of fever attacks during 24 weeks prior to study entry. Fifth, the percentage of Japanese patients with MEFV exon 10 variants is lower than those in Western countries [40–42], and the number of participants with exon 10 variants was small in this study. The overall low frequency of attacks in this study may be due to the small number of cases with the pathogenic variants in exon 10, which reflects the genetic characteristics of FMF in Japan. Despite these limitations, this is the first double-blind, randomized, placebo-controlled trial to demonstrate the long-term efficacy of TCZ in reducing recurrent fever attacks, and the results of this study provide useful insight into the management of FMF.