Efficacy and safety of telitacicept in primary Sjögren’s syndrome: a randomized, double-blind, placebo-controlled, phase 2 trial

Abstract

Objective

To evaluate the efficacy and safety of telitacicept in adult patients with primary SS (pSS) in a phase II randomized double-blind placebo-controlled trial.

Methods

Patients with pSS with positive anti-SSA antibody and ESSDAI ≥ 5 were randomly assigned, in a 1:1:1 ratio, to receive weekly subcutaneous telitacicept 240 mg, 160 mg, or placebo for 24 weeks. The primary end point was the change from baseline in the ESSDAI at week 24. Safety was monitored.

Results

A total of 42 patients were enrolled and randomized (n = 14 per group). Administration of telitacicept 160 mg resulted in a significant reduction in ESSDAI score from baseline to week 24 compared with placebo (P < 0.05). The placebo-adjusted least-squares mean change from baseline was –4.3 (95% CI –7.0, –1.6; P = 0.002). While, mean change of ESSDAI in telitacicept 240 mg was –2.7(–5.6–0.1) with no statistical difference when compared that in placebo group (P = 0.056). In addition, MFI-20 and serum immunoglobulins decreased significantly (P < 0.05) at week 24 in both telitacicept groups compared with placebo. No serious adverse events were observed in the telitacicept treating group.

Conclusion

Telitacicept showed clinical benefits and good tolerance and safety in the treatment of pSS.

Trial registration

ClinicalTrials.gov, https://clinicaltrials.gov, NCT04078386

Open in new tabDownload slide

Introduction

Primary Sjögren’s syndrome (pSS) is a systemic autoimmune disease characterized by lymphocytic infiltration of exocrine glands, leading to sicca syndrome. ∼30–40% of patients with pSS have systemic manifestations [1]. B cells play an important role in the pathogenesis of pSS. The presence of ectopic germinal centres in salivary glands highlights that B-cell activation is a characteristic of pSS. B-cell activating factor (BAFF) [also called B lymphocyte stimulator (BlyS)] and proliferation-inducing ligand (APRIL), as TNF family members, are critical factors in the maintenance of the B-cell pool and humoral immunity [2, 3] and are aberrantly expressed in pSS, which may explain the activation and survivial of pathogenic B cells in this condition. Recent studies have suggested increased levels of BAFF in serum and salivary glands [4, 5] and higher serum APRIL levels in patients with pSS [6]. Several clinical trials targeting B cells, including rituximab, belimumab and ianalumab, have been published, and the results were controversial.

Telitacicept is a soluble fusion protein composed of the transmembrane activator, calcium modulator, cyclophilin ligand interactor (TACI), and fragment crystallizable (Fc) domain of human immunoglobulin G (IgG) [7]. Blockade of APRIL and BAFF by use of a recombinant serotype 2 adeno-assosiated virus encoding TACI-Fc had showed a significantly reduced number of inflammatory foci and level of IgG and IgM in the salivary gland of non-obese diabetic mice[8]. Telitacicept has been approved to treat patients with SLE by the National Medical Products Administration in China in 2021. A phase 3 study of SLE in China had finished, the result showed 82.6% of SRI-4 response rate in the telitacicept 160 mg group at week 52 and good safety and tolerance[9]. In 2019, telitacicept received approval from the US FDA to conduct a global multicentre, phase-II clinical trial for SLE. Clinical studies of telitacicept in several other indications, including rheumatoid arthritis, IgA nephropathy, multiple sclerois, myasthenia gravis, neuromyelitis optica spectrum disorders are underway in China. This study reports a multicentre, randomized, double-blind, placebo-controlled phase 2 trial to assess the efficacy and safety of telitacicept in patients with pSS (NCT04078386).

Patients and methods

Patients

Patients with pSS who fulfilled the 2016 ACR/EULAR classification criteria for SS [10], aged 18–65 years old, anti-SSA antibody positivity, and ESSDAI score ≥ 5 points were recruited in this trial. Patients with secondary SS, or treated with glucocorticoid, immunosuppressive agents (cyclophosphamide, leflunomide, methotrexate, etc.) and traditional Chinese medicines (tripterygium wilfordii polyglycosides and total glucosides of paeony) due to pSS within four weeks, biological agents (rituximab, belimumab, etanercept, infliximab, etc.) within six months, drugs increasing salivation (pilocarpine and anethole trithione, etc.) within one week before randomization, and patients without stable use of sodium hyaluronate eye drops and artificial tears within four weeks before randomization were excluded. Patients treated with HCQ were not eligible except for those who had received a stable dose treatment >12 weeks before randomization.

Methods

A phase 2, randomized, double-blind, placebo-controlled trial was conducted to compare telitacicept with placebo in terms of efficacy and safety in patients with primary Sjögren’s syndrome. In our trial, no formal calculation was conducted for the sample size as it was determined at a time when there is no available data on possible effect of telitacicept in primary Sjögren’s syndrome. As a preliminary assessment, a relatively small number of patients was preferred and based on practical considerations, the planed sample size was n = 14 in each group.

Eligible patients were randomly assigned (1:1:1) to receive subcutaneously an injection of 240 mg telitacicept, 160 mg telitacicept or placebo for 24 weeks, once a week, according to a block andomization scheme (block size of six) using an interactive web response system by Shanghai BioGuider Medical Technology Co., Ltd. Efficacy and safety were assessed at week 24.

Endpoints

The primary end point was the change from baseline in the ESSDAI at week 24 [11]. Key secondary endpoints included changes from baseline in ESSDAI at week 12 and changes from baseline at week 12 and week 24 in the following items: ESSPRI, Physician Global Assessment (PGA), Patients Global Assessment (PaGA), 36-item Short-Form (SF-36), Multidimensional Fatigue Inventory (MFI-20). In addition, we statistically analysed other items including unstimulated whole saliva (UWS) flow rate, Schirmer’s test and immunological markers including IgG, IgA, IgM, C3, C4, CD19⁺ B cells, CD4⁺ T cells and CD8⁺ T cells. We also explored the percentage of patients who achieved MCII, which was defined as an improvement in the ESSDAI score of ≥3 points and an improvement in the ESSPRI score of ≥1 point or ≥15% [12].

Safety

Safety assessments included the incidence and severity of adverse events (AEs). Adverse events were recorded and graded for severity using the Common Terminology Criteria for Adverse Event (CTCAE) 4.03.

Statistical analysis

Statistical analyses were performed using SAS V9.4. All statistical tests were performed using two-sided tests, and statistical significance was set as P ≤0.05. Quantitative variables are presented as mean (s.d.) or median (for non-normal distribution), minimum, and maximum values. Categorical variables were described as frequencies and percentages. The incidence of adverse events was calculated from the total number of subjects in the safety dataset. The mixed-effects model repeated measure (MMRM) was used to model the primary endpoints, mean change in the ESSDAI score at week 24 from baseline. Sensitivity analysis of the primary efficacy variable was also performed using analysis of covariance (ANCOVA) with the covariance of ESSDAI at baseline. Missing data were imputed using the last observation carried forward (LOCF) method. The secondary efficacy endpoints were analysed using analysis of variance (ANOVA) to compare differences among the groups. When P ≤0.05, Fisher's least significant differences (LSD) method is used in ANOVA to create confidence intervals for all pairwise differences between factor level means while controlling the individual error rate to a significant level. LSD tests are performed for multiple comparison of means when the overall model F test is significant.

Patient and public involvement

The patients and the public were not involved in the design and conduct of the trial. The study was conducted following the Declaration of Helsinki and International Conference on Harmonization Good Clinical Practice guidelines. The study was approved by the Institutional Review Board of the Peking Union Medical College Hospital and other participating study sites (a full list of IRBs is provided in Supplementary Data S1, available at Rheumatology online). Written informed consent was obtained from all patients.

Results

Patient characteristics

A total of 57 patients were screened for eligibility from 14 medical centres in China during November 2019 to June 2021. Forty-two were enrolled and randomized (1:1:1) into telitacicept 240 mg, telitacicept 160 mg and placebo. All 42 patients were included in the FAS and SS. The mean age of patients was 49.4 years. Of these, 40(95.2%) were female. For detailed information, refer to Fig. 1. The mean and median ESSDAI scores in all patients at baseline were 8.7 and 7, individually. At baseline, patients in each group were balanced in their general condition, disease duration, laboratory examination test, ESSDAI score, ESSPRI score, PGA, PaGA, SF-36, MFI-20 (Table 1). For detailed data at baseline, refer to the supplementary material, available at Rheumatology online.

Among the 42 randomized patients, eight patients in the telitacicept 240 mg group, 12 patients in the telitacicept 160 mg group and 10 patients in the placebo group finished the whole trial. Twelve patients discontinued the treatment before the end (Fig. 1). Among those, four patients quit in advance due to AEs in the telitacicept treating groups. One patient in the teletacicept 160 mg group quit because of a drug allergy (CTCAE Grade 1). The other three in the teletacicept 240 mg group quit for local injection reactions (CTCAE Grade 1–2).

Primary end point

At week 24, mean changes of ESSDAI from baseline in the placebo group, the 160 mg group and the 240 mg group were 0.6 ± 4.55 [mean (s.d.)], –3.3 ± 2.73, –1.3 ± 4.14 individually. By MMRM, changes of ESSDAI decreased obviously compared with the placebo group. The placebo-subtracted LSMEANS of the 160 mg group is –4.3 (95% CI –7.0, –1.6; P = 0·002) while mean change of ESSDAI in telitacicept 240 mg was –2.7(–5.6–0.1) with no statistical difference when compared with that in the placebo group (P = 0.056). ESSDAI score at each visit is shown in Fig. 2A.

Similar results were observed by sensitivity analysis using ANCOVA with the covariate of ESSDAI at baseline. Statistical difference of mean ESSDAI changes were observed by the comparation between the telitacicept 160 mg group and the placebo group (P = 0.010), but not in the telitacicept 240 mg group (P > 0.05).

Second endpoints

ESSDAI change at week12

Mean changes of ESSDAI at week 12 from baseline in the placebo group, the 160 mg group and the 240 mg group are 0.4 ± 4.67, -3.1 ± 2.80, -1.9 ± 3.43, respectively. By ANOVA, mean ESSDAI of the telitacicept 160 mg group at week 12 from baseline, not the telitacicept 240 mg group, decreased obviously compared with the placebo group, P = 0.015. For variation of ESSDAI, see Fig. 2A. For data about each ESSDAI domain at baseline and change in each ESSDAI domain at week 12 and 24, see Supplementary Table S1–S4, available at Rheumatology online. Further, we analysed the change of each ESSDAI domain, and found the biological domain changed as statistically different (P = 0.047) in the group of telitacicept 160 mg vs placebo at week 24. The constitutional and articular domain also changed, but did not reach statistical difference (P = 0.079 and 0.072, respectively) (Supplementary Table S4, available at Rheumatology online).

ClinESSDAI change

At week 24, by MMRM, the placebo-subtracted LSMEANS of clinESSDAI in the 160 mg group is -4.7 (95% CI -7.8, -1.5; P = 0.004), while the placebo-subtracted LSMEANS of clinESSDAI in 240 mg group is -2.5 (95% CI -5.7, 0.8) with no statistical difference (P = 0.139). Similar results were observed by sensitivity analysis using ANCOVA. Statistical difference of placebo-subtracted LSMEANS were observed in the telitacicept 160 mg group (P = 0.013), but not in the telitacicept 240 mg group (P = 0.285).

MFI-20

Statistically significant greater reduction from baseline in MFI-20 to weeks 12 and 24 compared with placebo (P < 0.05) was observed in both telitacicept treating groups. Mean MFI-20 of patients in the telitacicept 160 mg group decreased from 58.9 at baseline to 55.6 at week 12 and to 53.1 at week 24; it decreased from 59.6 at baseline to 53.7 at week 12 and to 50.9 at week 24; while it was 50.7 at baseline, 53.3 at week 12 and 52.8 at week 24 in placebo group (see Fig. 2B).

Immunoglobulins and complements

Compared with placebo, telitacicept induced significant reductions in serum IgG, IgA and IgM levels at each visit throughout the 24 weeks (P < 0.001). At week 24, the mean change of IgG level increased from 16.26 ± 4.64 g/l to 16.89 ± 6.39 g/l in the placebo group (1.02%); it decreased from 19.65 ± 6.09 g/l to 15.63 ± 5.44 g/l and decreased from 15.89 ± 4.45 g/l to 12.74 ± 3.98 g/l seperately in telitacicept 160 mg group (-23.91%) and telitacicept 240 mg group (−28.40%). Compared with the changed mean change of IgA level from 3.52 ± 1.64 g/l to 3.65 ± 1.65 g/l in the placebo group (-1.29%), it decreased from 3.04 ± 1.24 g/l to 1.47 ± 0.63 g/l and from 3.46 ± 1.54 g/l to 2.13 ± 1.34 g/l seperately in the telitacicept 160 mg group (−47.55%) and telitacicept 240 mg group (49.18%). Meanwhile, the mean change of IgM decreased from 1.29 ± 0.70 g/l to 0.97 ± 0.72 g/l in the placebo group (−7.87%), it decreased from 1.22 ± 0.44 g/l to 0.55 ± 0.28 g/l and 0.99 ± 0.42 g/l to 0.32 ± 0.13 g/l seperately in the telitacicept 160 mg group (−55.74%) and the telitacicept 240 mg group (−61.61%). See Fig. 3 and Supplementary Table S5, available at Rheumatology online.

C4, not C3, varied statistically at the end of the treatment (P < 0.05) by the comparison of patients in both telitacicept treating groups and the palcebo group. At week 24, mean level of C4 increased from 0.19 ± 0.04 g/l to 0.22 ± 0.06 g/l for patients in the telitacicept 160 mg group (P = 0.017), from 0.23 ± 0.07 g/l to 0.27 ± 0.09 g/l for patients in the telitacicept 240 mg group (P = 0.015), while it was maintained in the placebo group (see Fig. 3; Supplementary Table S5, available at Rheumatology online).

MCII

The percentage of patients who achieved MCII in the ESSDAI at week 24 was significantly higher in the 160 mg telitacicept group than in the placebo group (71.4% vs 14.3%, P = 0.006). However, there was no significant difference in the percentage of patients who achieved MCII in the ESSPRI at week 24 among the three groups (P > 0.05).

No statistical differences were observed in each telitacicept group compared with the placbo group in the following items: ESSPRI, SF-36, UWS flow rate, PGA, PaGA, Schirmer’s test (Supplementary Table S1, available at Rheumatology online). For CD19⁺ B cells count, CD4⁺ T cells count, and CD8⁺ T cells count, see Supplementary Table S5 and Supplementary Fig. S1, available at Rheumatology online.

Safety

During the whole treatment phase, no death or serious adverse events (SAE) happened in the telitacicept treating groups. The only patient suffering a serious adverse event occurred in the placebo group due to aggravation of pSS.

Most adverse events were mild or moderate in severity. Local injection reaction (CTCAE grade 1–2) is the main adverse reaction. There is a statistically different occurence of it between each telitacicept treating group and the placebo group. Two patients (14.3%) in the telitacicept 240 mg group suffered from severe adverse events (CTCAE Grade3), presenting with acute pyelonephritis and leukopoenia. After the suspension of investigational drug administration, the acute pyelonephritis was recovered with treatment of antibiotic and the leukopoenia was persistence. No adverse events graded at CTCAE 4 or 5. We didn’t observe increased risk of infection in view of the incidence of infection in each group. There were 8 (57.1%), 9 (64.3%) and 5 (35.7%) patients suffering from infection individually in the placebo group, the telitacicept 160 mg group and telitacicept 240 mg (Table 2).

Discussion

B lymphocytes play a crucial role in pSS pathogenesis [13]. Elevated levels of BAFF were observed in the saliva, serum and salivary glands of SS patients [14–16]. BAFF is related to multiple disease characteristics of pSS, such as the production of autoantibodies, facilitating class switching to the IgG isotype, and ectopic germinal centre formation in Sjögren’s syndrome is driven, at least in part, by elevated BAFF levels [16, 17]. It is reported that BAFF is also significantly involved in pSS and pSS-related lymphoma pathogenesis [18, 19]. APRIL is implicated in plasma cell survival and autoantibody production in mice [20–22] and humans [23]. Another study found that increased serum BAFF and APRIL levels were related to focal scores and IgG levels, suggesting a possible association between BAFF and APRIL and disease activity in SS [24].

Currently, the treatment of Sjögren’s syndrome is mainly symptomatic and involves immunotherapy. Based on the lymphoproliferative background, direct or indirect targeting of B lymphocytes has become a cornerstone for developing therapeutic strategies for pSS. Several clinical trials targeting B cells have been conducted. Although supported by several open and registry studies and case series studies for improving some extraglandular organ involvement, rituximab did not reach the primary end point in large RCT studies [25–28]. Belimumab achieved positive results in an open-label prospective phase II trial (BELLIS). A total of 18/30 (60%) patients with pSS reached the primary end point at week 28. The ESSDAI and ESSPRI scores were significantly lower and transitional and naive B cell subsets reduced [29–31]. Recently, the first large RCT trail in pSS that met its primary end point was published. A new biologic ianalumab, targeting B cells by ADCC and BAFF inhibition, had finished a global multicentre phase 2 b dose-finding study. At week 24, it showed the efficacy in the treatment of pSS by a dose-related decrease in ESSDAI [32].

Compared with other B-cell targeting pharmaceuticals, characteristics of telitacicept are: (i) in contrast to anti-BLyS therapy, telitacicept can neutralize both BlyS and APRIL heterotrimers; (ii) it contains a longer TACI fragment; and (iii) in contrast to classical B-cell targeted drugs, such as CD20 monoclonal antibodies, telitacicept can inhibit the survival of long-lived PCs that have developed into potential autoantibody-producing cells. BLyS/APRIL-targeted drugs might be an avenue to improve B-cell inhibition with improved safety [7]. Previous preclinical studies have shown that telitacicept regulates T- and B lymphocytes function and ameliorates joint and spleen pathology in CIA mice [33]. In a randomized, double-blind, multicentre, placebo-controlled phase-II clinical trial in SLE patients, telitacicept 240 mg induced a higher SLE response index (SRI) compared with placebo and a significant reduction in B cell and Ig levels as well as an increase in complement [7].

In this pilot RCT, we observed that telitacicept 160 mg significantly reduced the ESSDAI score compared with placebo at week 24. Given telitacicept is known to decrease immunoglobulin levels as a function of its targeting of B cells, we further analysed the clinESSDAI as an important outcome measure independent of the biological effect of the drug and found similar results. A significant difference was already shown from week 12 onwards, suggesting that treatment with telitacicept significantly reduced disease activity in patients with pSS. In order to enhance the robustness of the result, both mean imputation and multiple imputation were performed as sensitivity analysis (Supplementary Tables S6–S9, available at Rheumatology online). As in the primary analysis, we observed the effectiveness of telitacicept 160 mg by mean imputation (P = 0.026), but no statistical differences were observed in each telitacicept group compared with the placebo group at week 24 when using multiple imputation, though decreasing of mean ESSDAI at telitacicept 160 mg group at week 24 from baseline was more numerically compared with the placebo group. More patients and less missing data were needed for further study. The statistical difference between the telitacicept 240 mg group and the placebo group was not found in the study. The following factors may contribute to the differences in efficacy between the two telitacicept groups.

The previous study showed that the efficacy of telitacicept is correlated with its blood concentration (not shown publicly). However, the administrating dose for each patient was not completely correlated with the blood concentration. There might be a great variety in the blood concentrations within subjects in the same group due to the small sample size. Moreover, the outbreak of the COVID-19 epidemic in China made it difficult for some patients to visit hospital once a week, and so some patients withdrew from the trial. It is difficult to find significant associations that are misaligned with the expected dose-dependent efficacy of a drug with a low number of patients. In summary, the likely reasons are the small sample size, the high proportion of dropouts in the 240 mg telitacicept group, and the high likelihood of type 1 and 2 errors.

MFI-20 evaluates fatigue symptoms from multiple dimensions, which can more accurately reflect fatigue symptoms than the VAS in ESSPRI. The study showed that telitacicept significantly improved fatigue symptoms assessed in a multi-dimensional manner at week 24; however, the fatigue, pain and dryness symptoms assessed in the ESSPRI score did not significantly improve.

Data in this trial showed good safety in the treatment of patients with pSS by similar incidence of infection in both the telitacicept and the placebo treating population. The main adverse reaction is local injection reaction and it is mild (CTCAE grade 1–2). It is consistent with other clinical trials investigating telitacicept. For example, in a phase 2b RCT, 249 patients with SLE were treated with telitacicept or placebo. Incidence of adverse drug reactions are similar: 66.8% for the telitacicept treating groups and 64.5% for the placebo administrating group, according to the package inserts of Telitacicept for injection in China [34]. The stable incidence of the telitacicept treating population could be more or less explained by the stable T- and B-cell counts. In contrast, telitacicept increased the level of C4, a risk factor for the development of lymphoma in patients with pSS, suggesting that telitacicept may prevent the progression of pSS to lymphoma.

The limitations of this study included a small number of patients in this study, and the relatively large number of patients withdrawn from the trial because of the outbreak of the COVID-19 epidemic in China, which made it difficult for some patients to visit hospital once a week.

Telitacicept has shown clinical benefits for the treatment of pSS. Compared with placebo, telitacicept significantly improved ESSDAI and MFI-20 in pSS patients at weeks 24 and 12 and reduced the level of Igs. It showed a favourable safety profile in patients with pSS.

Supplementary material

Supplementary material is available at Rheumatology online.

Data availability

The data that support the findings of this study are available on request from the corresponding author. Some data are not publicly available due to the privacy protection of patients.

Funding

This study was funded by RemeGen Co., Ltd. It is the only sponsor of this trial and it developed telitacicept. All investigational drugs and placebos were provided by the sponsor.

Disclosure statement: The authors have declared no conflicts of interest.

References

Author notes