Search for a command to run...
Cryoglobulinemia is defined as the presence of immunoglobulins that precipitate at cold temperatures and dissolve upon rewarming. According to the classification by Brouet et al. [1], there are three types of cryoglobulinemia, among them types II and III, collectively referred to as mixed cryoglobulinemia [2]. Mixed cryoglobulinemia is generally associated with connective tissue disorders, chronic infections, or lymphoproliferative disorders. When no underlying cause is identified, the condition is referred to as essential cryoglobulinemia. Cryoglobulinemia can be asymptomatic but may also lead to systemic vasculitis (CryoVas), patients presenting with symptoms ranging from skin involvement (purpura, necrosis) and arthralgia to more severe organ damage such as kidney, neurological, digestive, or pulmonary manifestations [3]. Due to its past predominance with up to 90% of cases [2], most available data on CryoVas are related to hepatitis C virus (HCV)-associated disease. However, with direct-acting antiviral therapies, the incidence of HCV-related cryoglobulinemia has significantly declined [4], and it is no longer the leading cause of the disease. In contrast, data on noninfectious mixed cryoglobulinemia vasculitis (NIMCV) remain limited [5]. There is an unmet need to describe long-term outcomes of patients with CryoVas and identify factors associated with vasculitis relapse, and complications such as hematological malignancies [6]. Our main objective was to describe long-term prognosis of patients with noninfectious mixed cryoglobulinemia vasculitis and to identify factors associated with relapse. We included 304 patients (67% women, median age at diagnosis 64 [53–74] years) with non-infectious mixed cryoglobulinemia vasculitis (methods in Appendix S1, Figure S1, Table 1). Associated conditions were connective tissue disorders in 97 (32%) patients (Sjögren's disease in 81 patients, associated with systemic lupus in 10 patients, rheumatoid arthritis in 4 patients and systemic sclerosis in 2 patients), lymphoid hematological malignancies in 79 (26%) patients (marginal zone lymphoma in 38 patients, follicular lymphoma in 21 patients, chronic lymphocytic leukemia in 6 patients, Waldenström macroglobulinemia in 5 patients, mucosae associated lymphoid tissue in 4 patients, mantle cell lymphoma in 3 patients, and monoclonal gammopathy of undetermined significance in 2 patients), and essential in 128 (42%) patients. Clinical manifestations at diagnosis were predominantly skin involvement in 244 (80%) patients, peripheral neuropathy in 156 (51%) patients and kidney involvement in 92 (30%) patients. Vasculitis was diagnosed as severe in 215 (71%) patients. C4 levels were low in 290 (95%) patients, and type II mixed cryoglobulinemia was the predominant type, found in 240 (79%) patients. After CryoVas diagnosis, 240 (79%) patients were treated either with steroids alone or in association with rituximab in 93 (31%) or cyclophosphamide in 65 (21%) patients. Plasmapheresis was initiated in 43 (14%) (Table S1). After first-course CryoVas treatment, clinical and immunological remission were obtained in 194 (64%) and 105 (35%) patients, respectively. Vasculitis relapse or progression occurred in 132 (43%) patients after first-course treatment. Relapses were diagnosed 18 [5–49] months after first-course regimen ending (Figure 1). Organ involvement at vasculitis relapses were mostly similar to those at vasculitis diagnosis (Figure 2A). First relapse was defined as severe in 41 (14%) patients, including 6 (2%) patients with a non-severe vasculitis at CryoVas diagnosis (Figure 2B). In multivariate analysis, connective-tissue disorders associated with cryoglobulinemia (OR 1.80 CI95 [1.05–3.09], p = 0.03) and absence of complete immunological remission (OR 0.45 CI95 [0.24–0.83], p = 0.01) were associated with CryoVas relapse (Table S2). During follow-up, 132 patients experienced 398 vasculitis relapses ranging from 1 to 11 relapses per patient and resulting in 702 treatment lines (Table S3). To account for intra-patient correlation due to repeated treatment lines, a generalized linear mixed model was performed and revealed that joint involvement (1.64 [1.07–2.51], p = 0.02), absence of complete immunological response (0.57 [0.37–0.86], p = 0.008) and absence of rituximab use (OR [95%] for relapse 0.63 [0.43–0.93], p = 0.002) were independently associated with vasculitis relapse (Figure S2). During a median follow-up of 70 [34–144] months, 70 (23%) patients died. Death causes were major adverse cardiovascular events in 33 (47%) patients, infection in 29 (41%) patients, CryoVas flare in 4 (6%) patients and lymphoma in 4 (6%) patients. Severe infection occurred in 92 (30%) patients, including 48 bacterial pneumonia, 18 cellulitis, 6 Pneumocystis pneumonia, 6 bloodstream infections, 5 urinary tract infections, 2 cryptococcosis, and 1 progressive multifocal leukoencephalopathy. At the end of follow-up, 59 (20%) patients had chronic kidney disease and among them, 9 (15%) received kidney-replacement therapy by dialysis and 3 (5%) received kidney transplantation. Those 12 patients with kidney failure had CryoVas associated with connective tissue disorder in six patients, essential in three patients and hematological malignancy in three patients. Among patients with non-hematological malignancy CryoVas (i.e., connective tissue disorders and essential at diagnosis), a lymphoma occurred in 35 (16%) patients, including 20 in Sjögren's syndrome patients and 15 in patients with essential cryoglobulinemia vasculitis. In this largest cohort of non-infectious CryoVas with long-term follow-up, our main findings were that 43% of patients relapsed within 18 months after a first-course treatment. Relapses were independently associated with connective tissue disorder associated CryoVas and persistence of cryoglobulinemia and C4 consumption after treatment. One out of six patients with connective-tissue disorders or essential cryoglobulinemia will develop lymphoid hematological malignancy during follow-up. This study is the first to identify a correlation between the underlying etiology of mixed cryoglobulinemia and relapse patterns. Our results showed that in connective tissue disorders CryoVas, relapses occurred more frequently and earlier. These findings could lead to tailored treatment and monitoring regarding the underlying cause of CryoVas. Our results confirm that C4 consumption and cryoglobulin persistence are associated with relapses, as previously described by our group and others [7]. Relapses occurred from a few weeks to several years after the end of first-course treatment and mostly involved the same symptoms as those observed at CryoVas diagnosis. Moreover, most relapses were not qualified as severe, but some patients with a non-severe vasculitis at diagnosis relapsed with a severe form of vasculitis. These results highlight the need for long-term monitoring and follow-up in this population. Whether pre-emptive maintenance therapy based on biological markers persistence is still to be determined. We also provided data to show that patients mainly relapse within their initial organ involvement. Follow-up intensity and treatment maintenance could be discussed based on initial presentation at CryoVas diagnosis. When comparing all therapeutic lines received by patients, we showed that rituximab was independently associated with relapse-free survival. Strongest data on rituximab use in CryoVas are based on HCV-related CryoVas [8, 9]. In non-infectious mixed CryoVas, several studies suggested benefits from rituximab treatment without statistical demonstration and our results strengthen the use of this treatment as first-course therapy [9-12]. We still need further studies for patients with multiple and early relapses after rituximab treatment [6, 13]. One promising treatment option, currently under investigation, is belimumab, a human monoclonal antibody that targets BlyS. Small series have demonstrated the potential efficacy of belimumab in rituximab-refractory CryoVas patients [14, 15]. A prospective clinical trial (NCT04629144) is currently underway to systematically evaluate the safety and efficacy of belimumab in this patient population. Our study provides valuable data on long-term outcomes in this population. We report a mortality nearly as high as one quarter of the patients included, mostly from major adverse cardiovascular events. We also report high rates of serious infections, requiring hospital management, as previously described [5, 6, 16]. These findings raise the question of infection prophylaxis, including Pneumocystis pneumonia prevention, and raise the question of whether reduced dosing of rituximab could be considered in selected patients to potentially lower the risk of infection [17]. Hematological malignancies development in patients with HCV-related CryoVas and in patients with Sjögren syndrome is well reported [18, 19]. In our study, one out of six patients with connective-tissue disorders or essential mixed cryoglobulinemia developed lymphoid hematological malignancy during follow-up. This includes 12% of essential mixed cryoglobulinemia with lymphoma occurring during follow-up. This result highlights that in essential cryoglobulinemia with a negative auto-immune and malignancy work-up at diagnosis, a B-cell lineage clonal proliferation is highly probable and should guide surveillance and/or refractory diseases treatment. This cohort has some limitations. Data are retrospective, and some data may be missing. Patients were included over a long time period with changes in standard of care over time. Due to the retrospective design, patients' care and follow-up were not homogenous. However, two-thirds of treatment lines were administered during the rituximab standard-of-care era, and we believe our study provides robust data on a disease whose therapeutic landscape has significantly evolved over the past decades. To conclude, in this long-term cohort of non-infectious mixed CryoVas, relapse was significantly associated with connective tissue disease and persistent biological immune activation. This condition is at high risk of lymphoid hematological malignancy development. Rituximab treatment was the only one to be associated with flare reduction. This study provides new insights into relapse patterns and long-term risks, supporting a need for stratified treatment and follow-up strategies in non-infectious CryoVas. N.D. and A.M. made substantial contributions to the conception and design of the work, the acquisition, analysis, and interpretation of data for the work, drafted the work and reviewed it critically for important intellectual content, and gave final approval of the version to be published and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. D.S. and P.C. made substantial contributions to the analysis and interpretation of data for the work, reviewed the work critically for important intellectual content, gave final approval of the version to be published, and agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. A.L.J., C.D.R., E.P., E.E., I.M., G.C., T.Q., L.S., C.M., A.P., F.B., B.T., M.G., T.Z., D.L., made substantial contributions to the acquisition of data for the work, reviewed the work critically for important intellectual content, and gave final approval of the version to be published, and agreed to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The authors would like to acknowledge the valuable contributions of the following physicians, who were involved in patient care and study inclusion but could not be listed as authors in accordance with editorial guidelines: Isabelle Durieu, Anne-Laure Fauchais, Jacques-Eric Gottenberg, Antoine Huart, Aurélie Hummel, Gilles Kaplanski, Guillaume Le Guenno, Claire Larroche, Nicolas Limal, Raphaele Nove-Josserand, Patricia Rullier, Raphaèle Seror. Authors would also like to acknowledge historical CryoVas study group: Laurent Aaron, Sébastien Abad, Redouane Bakir, Pauline Belenotti, Lucas Benarous, Nathalie Beneton, Gilles Blaison, Claire Blanchard-Delaunay, Fabrice Bonnet, Pascal Cathébras, Fabrice Carrat, Laurent Chiche, Olivier Chosidow, Bernard Combe, Christian Combe, Nathalie Costedoat- Chalumeau, Pierre Cougoul, Bernard Cribier, Gilles Defuentes, Elisabeth Diot, Isabelle Durieu, Philippe Evon, Thibault Fraisse, Camille Francès, Hélène Francois, Thierry Frouget, Loik Geffray, Helder Gil, Jean Gutnecht, Eric Hachulla, Olivier Hermine, Olivier Hinschberger, Arnaud Hot, Jean-Emmanuel, Kahn, Alexandre Karras, Evguenia Krastinova, Adeline Lacraz, Estibaliz Lazaro, Véronique Leblond, Jean-Marc Léger, Eric Liozon, Catherine Lok, Véronique Loustau-Ratti, Kim Ly, Hervé Maillard, Raifah Makdassi, Xavier Mariette, Olivier Meyer, Philippe Modiano, David Noel, Céline Onno, Christian Pagnoux, Caroline Pajot, Géraldine Perceau, Antoinette Perlat, Anne-Marie Piette, Emmanuelle Plaisier, Jean-François Pouget-Abadie, Viviane Queyrel, Joseph Rivalan, Eric Rondeau, Luc de Saint-Martin, Elisabeth Salles-Thomasson, Françoise Sarrot-Reynauld, Thierry Schaeverbeke, Gwendoline Sebille, Patricia Senet, Jacques Serratrice, Alexandre Somogyi, Pascale Soria, Marie-Francoise, Thiercelin-Legrand, François Truchetet, Jean-Paul Vagneur, Philippe Vanhille, Bérengère Vivet, Pierre-Jean Weiller, Bernadette Woehl-Kremer, Jean-Marie Woehl. The authors have nothing to report. The authors declare no conflicts of interest. The data that support the findings of this study are available from the corresponding author upon reasonable request. Appendix S1: Methods. Table S1: characteristics of 43 patients treated with plasma exchange as first-course treatment. Table S2: Univariate and multivariate analysis of factors associated with relapse after first-course of treatment. Table S3: Univariate and Multivariate analysis associated with total relapses among 702 treatment lines. Figure S1: Flowchart of 304 patients with non-infectious mixed cryoglobulinemia vasculitis included in the study. Figure S2: Forest Plot of Adjusted Odds Ratios from the Generalized Linear Mixed Model (GLMM) of factors associated with CryoVas relapse. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.