Cryoglobulins are immunoglobulins (Ig) that undergo precipitation at temperatures below 37°C and re-dissolve upon rewarming (1). Cryoglobulinemia is a heterogeneous condition defined by the presence of circulating cryoglobulins in the sera and their deposition into small to medium blood vessels, causing chronic inflammation and end-organ damage (2). According to Brouet et al., cryoglobulinemia can be classified into three subtypes: Type I cryoglobulins comprising isolated monoclonal Ig, usually IgM, and this can be associated with plasma cell dyscrasias; Type II cryoglobulins comprising polyclonal IgG and monoclonal IgM with rheumatoid factor (RF) activity; and Type III cryoglobulins comprising polyclonal IgG and polyclonal IgM with RF activity, commonly referred to as mixed cryoglobulinemia (MC) (2, 3). Cryocrit, defined as the relative volume of the precipitate as a percentage of the total serum volume, is elevated in Type 1 and lowest in Type III cryoglobulins (4). Rocatello et al. recently introduced a novel concept of idiopathic hypocryoglobulinemia in patients with clinically suspicious vasculitis who have negative cryoglobulins by standard techniques. Through a modified precipitation technique in a hypoionic medium, a subset of patients are found to have trace levels of circulating polyclonal Type III cryoglobulins (hypocryoglobulins), redefining the diagnostic pathways for cryoglobulinemia (5).

Cryoglobulinemic vasculitis (CV) is an immune complex-mediated small- to medium-vessel vasculitis. The prevalence of CV is rare, estimated at approximately 1 per 100,000 worldwide. It is commonly observed in patients between 45 and 65 years of age, with a predominance of a female-to-male ratio of 3:1 (3). CV can be classified into infectious CV (ICV), where Hepatitis C virus (HCV) infection accounts for 90% of cases, and non-infectious CV (NICV), which is associated with autoimmune diseases and hematological malignancies, mainly B-cell lymphoproliferative disorders (6).

A triad of clinical presentations, including fatigue, arthralgia, and palpable purpura, are typically found in type II and type III cryoglobulinemia. In comparison, type I cryoglobulinemia is usually associated with conditions of the skin such as Raynaud’s phenomenon, ulcers, and gangrene (3). The literature describing the clinical characteristics and effective treatment of NICV is limited. The pathogenesis of NICV is less understood compared to ICV, but it is well acknowledged that chronic inflammation resulting from autoimmune conditions stimulates B-cell proliferation through complement and macrophage activation, subsequently inducing cryoglobulinemia (2, 3).

Various treatment approaches have been used in CV, which involve the use of corticosteroids (CS), rituximab (RTX), azathioprine (AZA), and plasmapheresis (PLEX) either individually or in combination. However, the treatment response in CV is variable, and treatment failures have been reported. In type I cryoglobulinemia, targeted therapies addressing the underlying lymphoproliferative disorder did not affect relapses (7). In HCV-related mixed cryoglobulinemia (MC), successful eradication is achieved using direct-acting antivirals (8). The combined administration of pegylated interferon α or ribavirin and/or novel antiviral drugs has also been effective in inducing remission in most cases of MC (9). However, despite successful treatment, there are instances of treatment failures, which can result in persistent disease manifestations or relapses (8). These treatment failures emphasize the importance of understanding the underlying disease mechanisms in different subtypes of cryoglobulinemia and developing more effective therapeutic strategies.

Clinical relapse is defined as the reappearance or recurrence of vasculitic clinical manifestations following the achievement of a complete treatment response. A complete immunological response refers to the normalization of the C4 complement level and the reduction/disappearance of cryoglobulins in sera. Only one study examined possible predictors for early relapse, where the absence of achieving a complete immunological response influenced relapse (10). Pulmonary involvement, gastrointestinal involvement, renal insufficiency, and advancing age >65 years were notable factors independently associated with death (11). To date, there is limited data on risk factors for relapse in NICV. Therefore, this systematic review aims to explore possible risk factors for relapse to guide effective patient management.

This systematic review was conducted according to ‘Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) (12). We registered our protocol with the International prospective register of systematic reviews (PROSPERO) on 14 March 2023 (PROSPERO registration CRD42023408140).

A total of 3,724 records were retrieved from the database search ( Figure 1 ). Following deduplication, 1,615 records were screened through a review of titles. A total of 335 records were retained for full-text assessment following the exclusion of 1,280 records that did not fulfill our inclusion criteria. A total of 27 studies comprising randomized controlled trials (n = 1), non-randomized clinical trials (n = 1), observational studies (n = 23), and case series ≥4 (n = 2) were included in the final review. Due to the scarcity of clinical trials and large prospective cohort studies on NICV, we were unable to perform a meta-analysis.

For this review, the sample size has been adjusted with the NICV cohort segregated from the total population in 10 studies (16–25). The total population from the included studies was 1,519, with a mean of 60.76; the study population ranged between 1 and 266. This review encompassed several studies that included a broad spectrum of patient characteristics in the context of relapses in CV, shedding light on its clinical characteristics, therapeutic approaches, and prognosis. The patient population included age groups ranging from 21 to 91 years, with a female gender preponderance. Studies included a range of countries: France (fourteen studies), Italy (six studies), the USA (two studies), and one study each from Brazil, Greece, and Poland. Multiple countries were involved in two studies ( Table 1 ).

The most frequently reported non-infectious etiological associations were connective tissue diseases accounting for 41.8% (n = 636), predominantly primary Sjögren’s Syndrome (pSS) and systemic lupus erythematosus (SLE), and haematological diseases comprising 27.6% (n = 416) encompassing MGUS and haematological malignancies (HM). Approximately 26.1% consisted of EMC (n = 397), whose underlying etiology was unidentified.

The included studies consistently found prevalent cutaneous manifestations in the patient population at baseline, while articular involvement, neuropathy, and nephropathy were observed to varying degrees. Immunological findings such as elevated cryoglobulin levels, low complement C4 levels, and positive RF activity in MC were commonly detected. Treatment strategies varied and included CS, RTX, cyclophosphamide (CYC), AZA, PLEX, and others. An in-depth summary of the study and patient characteristics are outlined in Tables 2 , 3 .

Of the 27 studies included in this review, 20 provided insights into the risk factors associated with CV relapse. Four studies investigated relapse in type 1 CV-associated MGUS and HM (25–28), while the remaining twenty studies focused on relapse in mixed CV (10, 29–40). Interestingly, seven studies (14, 16, 20–22, 41) failed to identify any identifiable risk factors for relapse. Below is a comprehensive summary of risk factors for relapse in NICV, including type I cryoglobulinemia.

NICV is commonly associated with non-infectious underlying causes, such as autoimmune diseases and hematological malignancies, in contrast to HCV-related CV. Treatments aiming at eliminating HCV infection have reduced relapse risk. In our systematic review, by synthesizing and analyzing existing evidence, we categorized risk factors for relapse in NICV based on Ig isotype. Due to the heterogeneity of the included studies, we delineated risk factors for relapse by identifying common patterns through the correlation of their clinical and immunological responses to various therapeutic approaches across the literature. Notably, each underlying condition impacted the clinical severity and cryoprecipitation in NICV (42).

Our analysis showed that the mean age at diagnosis of NICV was ≥60 years, and the male gender had poor survival outcomes, particularly if there was renal involvement (33, 39, 42). Terrier et al. (2014) described earlier relapses within 12 months in patients with a GFR <60 ml/min/1.73 m² (10).

Type I cryoglobulinemia with an underlying lymphoproliferative disorder tended to be more frequently associated with severe skin manifestations ranging from purpuric lesions to skin necrosis to ulcers in comparison to MC (25–28). Type I cryoglobulinemia lacked RF activity by conventionally abstaining from complement activation (26). In contrast, Kolopp-Sarda et al. suggested possible complement engagement and complement-inducing vasculopathy in type I cryoglobulinemia (43). There is an increased prevalence of hyperviscosity in type I, although some studies have reported leukocytoclastic vasculitis on histopathology, which is typical for mixed CV (44).

A huge proportion of the study population had an underlying autoimmune disease in the spectrum of NICV, where pSS was the leading etiology, followed by SLE. These often present with cutaneous manifestations (17, 20, 21, 32, 40). Additionally, few studies have associated symptoms such as low-grade fever, purpuric lesions, skin ulcers, low C4 complement level, and higher frequency of CV and severity of disease activity in pSS with concomitant cryoglobulin in sera with an increased risk of lymphoma and/or death (17, 20, 45).

Interestingly, Type II MC with monoclonal IgM is more susceptible to precipitation upon cold exposure. This is probably due to the pentameric conformation of IgM exhibiting increased avidity and the presence of multiple Fc sites on the same IgM molecule, making it accessible for rapid and/or multiple C1q binding, subsequently activating the complement pathway. Furthermore, binding affinity properties, glycosylation, amino acid formation, and temperature-induced molecular changes favor cryoprecipitation of IgM (46).

The persistence of cryoglobulinemia despite treatment constitutes a significant risk factor for relapse in NICV. Potential mechanisms for inadequate clearance include defective complement C1q receptor (47), failure of crosstalk between antibodies and complement non-interference with immune precipitation via the classical pathway, and impaired solubilization of immune precipitates impeding clearance by the alternative complement pathway in SLE (48). Circulating cryoglobulins activate complements, forming immune complexes with subsequent complement C4 consumption. Another plausible mechanism postulated for low C4 levels is competent cryoglobulin interaction with the C1q protein (47).

More than 40% of cases were EMC with type III cryoglobulinemia in the GISC study (42), compared to the CryoVas French series (10, 32, 33). Furthermore, most patients with type III EMC progressed to type II during their follow-up, highlighting that ineffective treatment of EMC could impact the clearance of cryoglobulin (42). In our review, frequent relapses were encountered in patients with underlying EMC (23, 31, 37, 39).

A study by Fayed et al. identified predictors of relapses following sustained viral response in HCV-related CV. The main predictors determined the presence of severe clinical manifestations at the end of direct-acting antiviral therapy such as skin ulcers, renal disease, peripheral neuropathy, and male gender (49). These findings align with our review of NICV, which also demonstrated that severe skin manifestations were associated with relapse, while male gender and renal disease were linked to a poorer prognosis. Moreover, the study conducted by Fayed et al. did not find a significant association between relapse and B-cell lymphoma or immunological parameters due to the limited availability of data (49).

The reactivation of RF-expressing B cells occurs when B-cell receptor (BCR) binds to immune complexes and Toll-like receptor-7 (TLR-7) and Toll-like receptor-9 (TLR-9), in response to microbial nucleic acid (50). Unlike TLR7, which had a limited impact on relapses, TLR9 had a significant role in disrupting immune tolerance and promoting relapse, even in patients who had successfully eliminated the virus (51). HCV-related CV patients who initially responded to B-cell targeting agents such as RTX experienced relapses due to ongoing viral mutations, leading to the emergence of different B-cell clones (52). Elevated levels of TNF ligand superfamily member 13B (also known as the B-cell-activating factor of the TNF family [BAFF]) and TNF ligand superfamily member 13 (also known as a proliferation-inducing ligand [APRIL]), which are implicated in B-cell survival and autoimmunity, have been observed in both HCV infection and autoimmune diseases (53). Considering the underlying pathogenic mechanisms, B-cell proliferation is a potential risk factor for relapse. Furthermore, BAFF and APRIL may serve as potential biomarkers to predict disease progression and/or relapse (49).

Several studies explored various therapeutic regimes; conspicuously, the RTX and CS regimen on induction, followed by maintenance RTX, demonstrated better efficacy and safety than CS or AKA, particularly in relapsing and/or refractory NICV patients (10, 23–26, 30–34, 37–40). RTX-associated CV flares have been observed in some patients and may be considered a risk factor for relapse, highlighting the importance of individualized treatment strategies (22, 26, 31). Strikingly, patients who are refractory to RTX share similar characteristics such as female predominance, age >50 years, type II cryoglobulinemia with monoclonal IgM kappa component, pSS, and EMC (32, 36). Saadoun et al. demonstrated significant improvement with the belimumab and RTX combination regimen in refractory cases (54). Further studies are needed to investigate the safety and efficacy of plausible treatment regimens for RTX-refractory cases.

A study by Lobbes et al. observed that patients with type I cryoglobulinemia who received treatment with Rituximab and experienced a reduction in circulating cryoglobulin levels showed rapid clinical improvement (28). Factors including advancing age with renal impairment were associated with poor prognosis (19, 27), and monitoring of cryocrit levels in type I predicted symptom improvement, serving as a prognostic tool in guiding management (25).

Our study has limitations due to the rarity of NICV, cohort size, and heterogeneity of the included studies. Most studies were retrospective observational studies discussing relapses of NICV and the efficacy of multiple treatment regimens. RCTs and/or prospective studies on relapse with a larger sample size are necessary to achieve explicit results. Some studies had mixed cohorts inclusive of ICV, alluding to the possibility of segregation bias. Case reports and case series of less than four were excluded; articles in languages other than English were excluded, indicative of selection and language bias.

To our knowledge, this is the first systematic review exploring identifiable risk factors for relapse with an underlying immunopathological etiology in NICV in relation to Ig isotype. Given the disease’s rarity and lack of RCTs and prospective studies, we could not perform a meta-analysis. Nonetheless, the achievement of remission in HCV-related CV with direct-acting antiviral agents paves the way for NICV as the next leading cause of CV (21). Abating the likelihood of relapse at the time of diagnosis is cumbersome; however, identification of those at risk might be critical for efficiently treating patients with NICV.

In conclusion, type I NICV accompanied by lymphoproliferative disorder often shows a response-relapse pattern, particularly when treated with CS alone. In MC, baseline factors such as cutaneous necrosis, purpura, and joint involvement, along with incomplete clinical and immunological responses to treatment, increase the risk of relapse. CS monotherapy is associated with a higher relapse risk. RTX-associated CV flare is more common in patients with renal involvement, B-cell lymphoproliferation, elevated cryoglobulin levels, decreased IgG levels, and low C4 prior to RTX treatment. These findings emphasize the importance of achieving both clinical and immunological responses and identifying relapse risk factors in NICV. Further research is needed to optimize treatment strategies and improve long-term outcomes for patients with NICV and associated lymphoproliferative disorders.

Conceptualization: NR. Data curation: NR and PR. Formal analysis: NR, PR, and HA. Methodology: NR and PR. Validation: NR and PR. Visualization: NR and PR. Writing: original draft: NR and PMR. Writing: review and editing: NR, PR, and HA. All authors contributed to the article and approved the submitted version.