Abstract: Rituximab (RTX), an antibody targeting CD20, is widely used as a first-line therapeutic strategy in B cell–mediated autoimmune diseases. However, a large proportion of patients either do not respond to the treatment or relapse during B cell reconstitution. Here, we characterize the cellular basis responsible for disease relapse in secondary lymphoid organs in humans, taking advantage of the opportunity offered by therapeutic splenectomy in patients with relapsing immune thrombocytopenia. By analyzing the B and plasma cell immunoglobulin gene repertoire at bulk and antigen-specific single-cell level, we demonstrate that relapses are associated with two responses coexisting in germinal centers and involving preexisting mutated memory B cells that survived RTX treatment and naive B cells generated upon reconstitution of the B cell compartment. To identify distinctive characteristics of the memory B cells that escaped RTX-mediated depletion, we analyzed RTX refractory patients who did not respond to treatment at the time of B cell depletion. We identified, by single-cell RNA sequencing (scRNA-seq) analysis, a population of quiescent splenic memory B cells that present a unique, yet reversible, RTX-shaped phenotype characterized by down-modulation of B cell–specific factors and expression of prosurvival genes. Our results clearly demonstrate that these RTX-resistant autoreactive memory B cells reactivate as RTX is cleared and give rise to plasma cells and further germinal center reactions. Their continued surface expression of CD19 makes them efficient targets for current anti-CD19 therapies. This study thus identifies a pathogenic contributor to autoimmune diseases that can be targeted by available therapeutic agents.

Abstract: Cancer-associated acute cholangitis (CAAC) are becoming more frequent and their characteristics may be changing with the evolution of cancer management. Our aim was to compare clinical, microbiological and outcome characteristics of CAAC to those of cancer-free acute cholangitis (CFAC). Methods All consecutive cases of acute cholangitis (AC) from November 2015 to March 2017 were collected retrospectively in a single tertiary care hospital in Clichy, France, specialized in gastroenterology. Hospital stays referred as AC by coding were screened. Patients fulfilling the 2018 Tokyo Guidelines diagnostic criteria for definite AC were included. Data were collected using a standardized form. CAAC were defined as AC that occurred in patients who had active cancer or history of cancer in the five previous years. CFAC were defined as AC in patient who no history of cancer, or in remission for more than 5 years. Comparison was made using Fisher or Student’s t-test. P 〈 0.05 was considered as significant. Results 156 episodes of AC in 130 patients were analyzed. 101 had CAAC and 55 had CFAC. Age and sex did not differ (table 1), but CAAC had a higher Charlson’s comorbidity index (4.4 vs. 1.7, P 〈 0.0001). Despite similar clinical presentation, CAAC had more pronounced cholestasis (Gamma GT 659 vs. 391UI/L; Alkaline phosphatases 526 vs. 309 UI/L; P 〈 0.0001 for both) and C-reactive protein level (133 vs. 97mg/L, P = 0.008, Table 2). E. coli was more common in CFAC (72.4% vs. 54% of positive blood cultures, P = 0.004). In bile cultures, Enterococci and multi-drug-resistant Gram negatives tended to be more frequent in CAAC than in CFAC (63 vs. 17%, P = 0.07 and 9.1% vs. 4.1%, P = 0.33, Table 2), respectively. CAAC more frequently required drainage (86.1% of cases vs. 43.6% in CFAC (P 〈 0.0001), including radiological drainage (42.5% vs. 12.5%; P = 0.008) and with multiple sessions (28.7% vs. 8.3%, P 〈 0.0001, Table 3). Antibiotherapy duration did not differ between the two groups. Despite similar initial severity, only 51.5% of patients with CAAC were alive, without febrile recurrence or other biliary drainage at day 28, vs. 85.5% of patients with CFAC (P 〈 0.0001, Table 3). Conclusion Despite comparable initial clinical presentation, management is more complex and outcome less favorable in CAAC vs. CFAC. Disclosures All authors: No reported disclosures.

Abstract: Introduction: The incidence of immune thrombocytopenia (ITP) is high in the elderly. Observational studies demonstrated that age above 65 years impacts ITP presentation and management. However, data are lacking in very elderly patients (VEP) with ITP. The aim of this study was to describe the presentation and the management of primary ITP that occurs in VEP (≥80 year-old) in comparison with elderly patients (EP; ≥65-79 year-old). Methods: Data source was the CARMEN-France registry. The CARMEN (Cytopénies Auto-immunes : Registre Midi-PyréneEN) registry is aimed at the prospective follow-up of all incident ITP adult patients in the French Midi-Pyrénées region (South-West of France, 3 million inhabitants) since June 2013. Each investigator follows all adult patients (aged ≥18 years) newly diagnosed for ITP in routine visit or hospital stay and detailed information on patients' characteristics and management of ITP are prospectively recorded. This registry has been implemented in the French referral center for ITP (Créteil) since October 2015, taking the name of CARMEN-France. Inclusion criteria in the present study were: inclusion in the CARMEN-France registry between June 2013 and December 2018; age ≥65 years at ITP onset; primary ITP defined by international criteria (platelets count 〈 100 x 109/L and exclusion of other causes of thrombocytopenia) and with normal bone marrow examination. Patients were then categorized by age groups (VEP vs EP). We compared patients' characteristics, ITP presentation including bleeding by platelet counts, exposure to first and second-line treatments. We also assessed the factors associated to any and severe (hematuria, gastro-intestinal tract or intracranial) bleeding at ITP onset in the VEP group using logistic regression models. Results: Out of 541 patients included in the CARMEN-France registry, 184 fulfilled inclusion criteria: 87 in the VEP group and 97 in the EP group. Mean age was 85.7 years in the VEP group versus 71.8 in the EP group. Male:female sex-ratio was similar (63.2% vs. 60.8%). Patients in the VEP group had more frequently comorbidities of the Charlson's Index (67.4% versus 47.9%) and polypharmacy (≥4 drugs; 63.3% versus 45.4%); they were more frequently exposed to antiplatelet drugs (37.9% versus 23.7%) and to anticoagulant (18.4% versus 10.3%). Median platelet counts at ITP onset were similar (22.0 versus 18.0 x 109/L). The frequencies of any bleeding were similar (58.6% versus 54.6%) as well as mucosal bleeding (25.3% versus 26.8%), but severe bleeding was more frequent in the VEP group (10.3% versus 4.1%). The frequencies of any bleeding and mucosal bleeding were higher in case of platelet count 〈 20 x 109/L in both groups. Severe bleeding occurred with platelet count 〈 20 x 109/L in all EP cases (n=4) while they occurred at any platelet count in the VEP group. In the VEP group, 85.1% of the patients were treated for ITP, versus 80.4% in the EP group. Among treated patients, 36.5% in the VEP group had steroids alone and 60.3% steroids plus intravenous immunoglobulin (IVIg), in comparison with 50.0% and 48.7% in the EP group, respectively. Second-line treatment was prescribed in 39.1% of VEP versus 37.5% of EP: thrombopoietin receptor agonists (13.8% versus 13.4%), dapsone (8.1% versus 8.2%), rituximab (6.9% versus 6.2%), and danazol (6.9% versus 4.1%). In univariate analysis, the factors associated with any bleeding in the VEP group were: female sex (odds ratio - OR: 1.97; 95% confidence interval - CI: 0.79-4.93), polypharmacy (OR: 2.64; 95% CI: 1.08-6.48); infection within the six weeks before ITP onset (OR: 2.60; 95% CI: 0.66-10.4) and platelet count 〈 20 x 109/L (OR: 10.0; 95% CI: 3.49-28.63). In univariate analysis, the main factors associated with severe bleeding were exposure to anticoagulant (OR: 7.61; 95% CI: 1.77-32.83), polypharmacy (OR: 5.28; 95% CI: 0.63-44.30), a Charlson's Comorbidity Index score ≥1 (OR: 4.32; 95% CI: 0.51-36.38); the OR for a platelet count 〈 20 x 109/L was 1.54 (95% CI: 0.38-6.16). Conclusion: VEP had more frequently severe bleeding at ITP onset. They were more frequently exposed to IVIg but did not require more frequently a second-line treatment. The pattern of second-line treatment was similar between VEP and EP. Platelet count 〈 20 x 109/L was a major risk factor for any bleeding in VEP. In contrast, exposure to anticoagulant, and not platelet count 〈 20 x 109/L, was highly associated with severe bleeding in VEP. Introduction: The incidence of immune thrombocytopenia (ITP) is high in the elderly. Observational studies demonstrated that age above 65 years impacts ITP presentation and management. However, data are lacking in very elderly patients (VEP) with ITP. The aim of this study was to describe the presentation and the management of primary ITP that occurs in VEP (≥80 year-old) in comparison with elderly patients (EP; ≥65-79 year-old). Methods: Data source was the CARMEN-France registry. The CARMEN (Cytopénies Auto-immunes : Registre Midi-PyréneEN) registry is aimed at the prospective follow-up of all incident ITP adult patients in the French Midi-Pyrénées region (South-West of France, 3 million inhabitants) since June 2013. Each investigator follows all adult patients (aged ³18 years) newly diagnosed for ITP in routine visit or hospital stay and detailed information on patients' characteristics and management of ITP are prospectively recorded. This registry has been implemented in the French referral center for ITP (Créteil) since October 2015, taking the name of CARMEN-France. Inclusion criteria in the present study were: inclusion in the CARMEN-France registry between June 2013 and December 2018; age ≥65 years at ITP onset; primary ITP defined by international criteria (platelets count 〈 100 x 109/L and exclusion of other causes of thrombocytopenia) and with normal bone marrow examination. Patients were then categorized by age groups (VEP vs EP). We compared patients' characteristics, ITP presentation including bleeding by platelet counts, exposure to first and second-line treatments. We also assessed the factors associated to any and severe (hematuria, gastro-intestinal tract or intracranial) bleeding at ITP onset in the VEP group using logistic regression models. Results: Out of 541 patients included in the CARMEN-France registry, 184 fulfilled inclusion criteria: 87 in the VEP group and 97 in the EP group. Mean age was 85.7 years in the VEP group versus 71.8 in the EP group. Male:female sex-ratio was similar (63.2% vs. 60.8%). Patients in the VEP group had more frequently comorbidities of the Charlson's Index (67.4% versus 47.9%) and polypharmacy (³4 drugs; 63.3% versus 45.4%); they were more frequently exposed to antiplatelet drugs (37.9% versus 23.7%) and to anticoagulant (18.4% versus 10.3%). Median platelet counts at ITP onset were similar (22.0 versus 18.0 x 109/L). The frequencies of any bleeding were similar (58.6% versus 54.6%) as well as mucosal bleeding (25.3% versus 26.8%), but severe bleeding was more frequent in the VEP group (10.3% versus 4.1%). The frequencies of any bleeding and mucosal bleeding were higher in case of platelet count 〈 20 x 109/L in both groups. Severe bleeding occurred with platelet count 〈 20 x 109/L in all EP cases (n=4) while they occurred at any platelet count in the VEP group. In the VEP group, 85.1% of the patients were treated for ITP, versus 80.4% in the EP group. Among treated patients, 36.5% in the VEP group had steroids alone and 60.3% steroids plus intravenous immunoglobulin (IVIg), in comparison with 50.0% and 48.7% in the EP group, respectively. Second-line treatment was prescribed in 39.1% of VEP versus 37.5% of EP: thrombopoietin receptor agonists (13.8% versus 13.4%), dapsone (8.1% versus 8.2%), rituximab (6.9% versus 6.2%), and danazol (6.9% versus 4.1%). In univariate analysis, the factors associated with any bleeding in the VEP group were: female sex (odds ratio - OR: 1.97; 95% confidence interval - CI: 0.79-4.93), polypharmacy (OR: 2.64; 95% CI: 1.08-6.48); infection within the six weeks before ITP onset (OR: 2.60; 95% CI: 0.66-10.4) and platelet count 〈 20 x 109/L (OR: 10.0; 95% CI: 3.49-28.63). In univariate analysis, the main factors associated with severe bleeding were exposure to anticoagulant (OR: 7.61; 95% CI: 1.77-32.83), polypharmacy (OR: 5.28; 95% CI: 0.63-44.30), a Charlson's Comorbidity Index score ³1 (OR: 4.32; 95% CI: 0.51-36.38); the OR for a platelet count 〈 20 x 109/L was 1.54 (95% CI: 0.38-6.16). Conclusion: VEP had more frequently severe bleeding at ITP onset. They were more frequently exposed to IVIg but did not require more frequently a second-line treatment. The pattern of second-line treatment was similar between VEP and EP. Platelet count 〈 20 x 109/L was a major risk factor for any bleeding in VEP. In contrast, exposure to anticoagulant, and not platelet count 〈 20 x 109/L, was highly associated with severe bleeding in VEP. Disclosures Comont: BMS: Consultancy, Membership on an entity's Board of Directors or advisory committees. Michel:Rigel: Consultancy; Amgen: Consultancy; Novartis: Consultancy. Beyne-Rauzy:Cellgene: Research Funding; Novartis: Research Funding. Godeau:Amgen: Consultancy, Speakers Bureau; Novartis: Consultancy, Speakers Bureau. Moulis:Amgen pharma: Research Funding, Speakers Bureau; Novartis pharma: Research Funding, Speakers Bureau; CSL Behring: Research Funding.

Abstract: Escherichia coli is frequently responsible for bloodstream infections (BSI). Among digestive BSI, biliary infections appear to be less severe. Respective roles of host factors, bacterial determinants (phylogroups, virulence and antibiotic resistance) and portal of entry on outcome are unknown. Methods Clinical characteristics and prognosis of 770 episodes of E. coli BSI were analyzed and isolates sequenced (Illumina technology) comparing phylogroups, MLST, virulence and resistance gene content. BSI isolates were compared with 362 commensal E. coli from healthy subjects. Results Among 770 episodes, 135 were biliary, 156 non-biliary digestive and 479 urinary. Compared to urinary, BSI of digestive origin occurred significantly more in men, comorbid and immunocompromised patients. Digestive portal of entry was significantly associated with septic shock and death. Among digestive infections, patients with biliary infections were dies less (P=0.032), despite comparable initial severity. Biliary E. coli resembled commensals (phylogroup distribution, ST group and few virulence-associated genes) whereas non-biliary digestive and urinary strains carried many virulence-associated genes. Conclusions E. coli strains responsible for biliary infections exhibit commensal characteristics and are associatedd with lower mortality rates, despite similar initial severity than other digestive BSI. Biliary drainage in addition to antibiotics in the management of biliary infections may explain improved outcome.

Abstract: Data about the presentation and the management of primary immune thrombocytopenia (ITP) in very elderly patients (VEPs; aged ≥80 years) are lacking. The aim of the present study was to describe ITP in this subgroup. The data source was the prospective CARMEN‐France registry. Patients included between 2013 and 2018 were selected. ITP presentation and management in VEPs was compared to elderly patients (EPs; aged 65–79 years). We assessed factors associated with bleeding at ITP onset in VEPs. Of 541 patients, 184 were included: 87 in the VEP group and 97 in the EP group. The mean age was 85·7 years in the VEP group. Comorbidities were more frequent in the VEP group (67·4% vs. 47·9%). The median platelet count at ITP onset was similar but severe bleeding tended to be more frequent in VEPs (10·3% vs. 4·1%, P  = 0·1) as well as mortality. Exposure to ITP drugs, response to first‐line treatment, need of second‐line treatment, evolution towards persistency, occurrence of bleeding, infection and thrombosis did not differ between groups. In VEPs, factors associated to bleeding were female sex [odds ratio (OR) 4·75, 95% confidence interval (CI) 1·31–17·32] and platelet count of 〈 20 × 10 9 /l (OR 10·05, 95% CI 4·83–67·39). Exposure to anticoagulants was strongly associated with severe bleeding (OR 7·61, 95% CI 1·77–32·83).