IgG4-related disease (ER-IgG4) is a fibroinflammatory condition characterised by lymphoplasmacytic infiltration with a predominance of IgG4+ plasma cells, fibrosis and in the majority of cases high levels of IgG4 in serum.1 It may essentially affect any organ, with a variable range of clinical manifestations, the most frequent of which are pancreatic and salivary gland compromise and adenopathies.2

Diagnosis is essentially histopathological, with characteristic findings of lymphoplasmacytic infiltration with a >40% proportion of IgG4/IgG, storiform fibrosis and obliterative phlebitis. Other diagnostic tools have also been used, such as imaging and IgG4 serum levels.1–4 There are 2 diagnostic scales (Umehara criteria5 and Okazaki criteria6) and an international consensus7 that makes it possible to reach the diagnosis. A series of organ-specific criteria have also been established, according to the organ that is involved (pancreatitis, sclerosing cholangitis, dacryoadenitis, ophthalmic, renal and pulmonary criteria associated with IgG4).8

Treatment is a field that has still hardly been explored. Steroids have formed the basis of treatment from the time it was described, although the relapse rate is high at 23–52%, especially when they are suspended.9,10 Disease-modifying anti-rheumatic drugs are not effective in inducing remission, although they may be beneficial in maintenance therapy.11 Recent case reports, series of cases and cohort studies have shown the possible favourable effects of therapy with rituximab (RTX), although the evidence is not conclusive. This systematic review has the aim of evaluating the efficacy and response to treatment with RTX in patients with ER-IgG4 who do not respond to glucocorticoids (GCC) and relapse with other immunomodulating therapies.

The search criteria recovered 65 results. After applying the inclusion and exclusion criteria and eliminating duplicated papers, 33 studies were selected. 2 studies were then eliminated because they were of duplicated cohorts, so that 31 papers were left for review, and they were subjected to quality analysis. 4 studies were excluded from the case reports and series of cases16–18 as they fulfilled fewer than 80% of CARE13 strategy criteria, while the observational studies met the CASP14 criteria except for the evaluation of confounding variables, while according to CONSORT15 the quality evaluation of the clinical trials gave optimum results. Finally 27 papers remained (with a total of 264 patients) for the final review, of which 23 corresponded to case reports and series of cases (105 patients), 3 corresponded to observational follow-up studies of a cohort (129 patients) and there was one clinical trial (30 patients). Table 2 describes the characteristics of the patients included in the studies, the type of clinical involvement depend on the affected system, previous treatments, type of RTX therapy, response to the same and relapses.

Although ER-IgG4 was described around the year 2000, it is still an emerging condition. In recent years more publications have appeared in association with the clinical, serological and histopathological characteristics of the disease, although no strong recommendations have been made regarding its first and second line treatment. Although therapy with GCC is considered to be a first-line therapeutic option, the high relapse rate is also known (at 23–52%), especially when they are suspended.9,10,21 Due to this reason the literature describes a large number of immunosuppressor drug combinations, where the use of RTX seems to stand out as a single or combined therapy. It was decided to evaluate the existing evidence for the therapeutic efficacy of using RTX to treat patients with ER-IgG4, with the aim of supporting or rejecting its use in cases refractory to GCC and other therapies.

The International Consensus on the use of ER-IgG421 recommends GCC as the first line therapy, with a 94% interexpert agreement on a dose of 30–40mg/day prednisolone during 2–4 weeks, with a gradual levelling off over 3–6 months after starting the treatment. Considering the high rate of recurrence with this treatment and its high toxicity over the medium to long term, the consensus recommends the use of steroid sparing agents, especially when it is impossible to perform appropriate levelling due to the persistence of disease activity or relapses, although interexpert agreement on this point amounts to only 46%. A study of patients with relapsed autoimmune pancreatitis treated with azatioprine, micophenolate or 6 mercaptopurine compared to GCC as a monotherapy, showed no differences in relapse-free survival between the two groups.11

A systematic review of therapeutic interventions in ER-IgG4 that included 62 studies with 3034 patients found steroid use in 74%, non-pharmacological therapies such as radiotherapy and surgery in 11%, no intervention in 13% and other therapies in 2%. Treatment with steroids as monotherapy attained an efficacy of 97%, although there were 33% of relapses in spite of the short follow-up periods. Relapses were treated with GCC, which attained a response rate of 95%, azatioprine at 81% and RTX in 9 patients with a 100% response rate.22

Given that the majority of patients with ER-IgG4 have to use GCC to remain in remission, other therapeutic options have been used as GCC sparing agents with results that are not very encouraging. In a series of cases of 21 patients with ER-IgG4 the use of sparing agents such as methotrexate, cyclosporine and azatioprine did not attain the goal of levelling off GCC.23 These findings are consistent with those obtained by this systematic review, where 65.4% of the patients included in the papers reviewed received steroids as the initial therapy and in 36% of the studies reviewed sparing agents were included in different protocols. There was a 10% response rate to the latter therapeutic strategies. In patients who were refractory to these strategies RTX was used.

Large series of cases and cohort studies24,45,47 have shown a high response rate to RTX associated with fewer relapses of the disease and a major possibility of levelling off GCC over the long term. The results of the review confirm the benefits of therapy with RTX as the second line in patients who are refractory to other treatments such as GCC and disease-modifying antirheumatic drugs, among others, emphasising the response rate of 90.7% when it is used as a second line therapy. Analysis of the total number of cases reported even showed that 10% of patients were treated with RTX as the first line therapy, obtaining a response in 100% of the patients. This finding means that it is now possible to think of using RTX from early phases after diagnosis.

Few of the studies analysed in the review reported relapses with RTX, although it should be underlined that the short periods of follow-up may influence this finding. This contrasts with the proportions described in the literature, in ranges that run from 10% to 41%.19,24,27 Some factors have been found to be associated with the risk of relapse, including high concentrations in serum of IgG4, IgE and eosynophiles, the absence of regular doses of the medication and most especially a response rate index in ER-IGG4>9 prior to the use of RTX.27 This index was only applied and reported in 3 of the 27 studies included in the review.

A multicentre study undertaken in France found that the patients who received treatment, not necessarily associated with relapses, achieved greater survival free of the same (41 vs 21 months).27 Although the data of the present study do not make it possible to evaluate the effectiveness of RTX in terms of relapse-free time, it would be important to answer this question in a future work within the context of long-term maintenance.

The systematic review analysed the response to treatment with RTX depending on the form of presentation of the disease. Eye socket involvement was found in 30% of cases, glandular involvement in 53% and ganglion involvement in 32.5% of all the patients, and this is similar to the finding of other series.28 Although a high proportion of patients with these clinical manifestations responded to treatment, as they are the most common forms of presentation of the disease it is not possible to conclude that RTX has a greater effect in this group. A comparative analysis was also performed between pure vasculitic forms, in which similar response rates were found. Nevertheless, it is not possible to categorically conclude that primary vasculitic forms respond to treatment with RTX, as the majority of cases had received previous or concomitant surgery and therapies with GCC before or during the administration of RTX for severe vascular complications such as aneurisms. However, the good response to this multimodal approach in the majority of cases stands out.

Although it was not a research aim, it is important to mention the heterogeneity that was found in how the different authors defined improvement or remission after treatment with GCC, RTX or other therapies, together with the absence of objective measurements of therapeutic response throughout the studies. This was shown by the fact that the ER-IgG4 response index was only applied in 2 observational cohort follow-up studies and in one clinical trial. This instrument was designed to evaluate disease activity, and it considers the organs which are potentially involved and the levels of IgG4 in mg/dl. To validate this index it was applied retrospectively to a registry of 15 patients in Massachusetts Hospital, obtaining a Pearson coefficient of 0.93 between the 2 evaluating rheumatologists.29 Including IgG4 levels in the score has led to debate, given that the pathological role of the said molecule has not been clearly established, and that it seems to behave as an epiphenomenon of the disease, so that it could be explained by an unknown primary inflammatory stimulus.1 It is therefore necessary to generalise the detection of the response, especially in future clinical trials, to give results that make it possible make comparisons and draw more precise conclusions.

It may be concluded from this review that therapy with RTX in patients with ER-IgG4 who have relapsed with while using GCC or other disease-modifying therapies has benefits in terms of the therapeutic response. Its effect on relapses is less clear due to the duration of follow-up in the studies, although over the mid-term it gave better results than other therapies. It has also been shown to be effective as a first line therapy.

Randomised controlled clinical trials are required that make it possible to confirm what has been reported to date in the literature. Therapy with anti-CD20 seems to be one of the best therapeutic strategies in ER-IgG4, not only for recurrent and refractory disease, but also as the initial treatment, given the potential benefits in terms of therapeutic response, reducing the number of relapses and lowering GCC treatment toxicity.

The authors have no conflict of interests to declare.