Background:Idiopathic multicentric Castleman disease (iMCD) is a rare and heterogeneous plasmacytic lymphoproliferative disorder that involves multiple regions of lymphadenopathy and cytokine‐driven systemic inflammatory symptoms. Many patients experience life‐threatening multi‐organ failure. Siltuximab, an anti‐interleukin (IL)‐6 monoclonal antibody, is the only European Medicines Agency‐approved drug treatment for iMCD based on a 34% response in a randomized phase 2 clinical trial. Better understanding of the etiology of iMCD and identification of effective treatments is urgently needed. ACCELERATE is a two‐armed natural history registry of Castleman disease (CD) comprised of a patient‐powered arm (PPA), operating out of the United States, and a physician directed arm (PDA), operating out of 9‐sites in Europe. Preliminary data from the PPA demonstrated the heterogeneity of the disease and identified dozens of off‐label drug treatments.Aims:We use the ACCELERATE registry to characterize iMCD in a European cohort and to identify the treatments most frequently used by European physicians.Methods:In the PDA, ACCELERATE consists of nine European study sites. Patients with a pathology report suggestive of Castleman disease are eligible to enroll. Site physician and study staff collect key demographic, clinical, and treatment data. Data is summarized in cases with sufficient extracted data. Quantitative data is presented as median (25th, 75th percentiles).Results:To date, 33 iMCD patients have met inclusion criteria (Germany 10 (30%), France 8 (24%), Italy 4 (12%), Spain 6 (18%), and Norway 5 (15%). The cohort is 48% female, and median age at diagnosis is 47.5 (38.1,57.1). Among patients with available diagnostic lymph node pathology reports (N = 22), histopathological features are strongly consistent with the histopathological characteristics defined in the 2017 International Consensus Diagnostic Criteria (Fajgenbaum et al., Blood, 2017). The most frequently noted features include interfollicular plasmacytosis (82%), vascular proliferation (68%), and atrophic germinal centers (55%). Histopathological subtypes are hyaline vascular (41%), plasmacytic (36.4%), and mixed (9.1%) with the remaining cases unspecified. Clinical and laboratory characteristics at diagnosis are also consistent with Consensus Criteria with 85% of assessed patients demonstrating constitutional symptoms, 28% organomegaly, and 19% fluid retention; laboratory abnormalities include hypoalbuminemia (33.0 (27.0, 35.0) g/L), inflammation (CRP 115.0 (12.2, 127.0) mg/L), anemia (hemoglobin 10.1 (8.5,12.1) g/dL), and hypergammaglobulinemia (17.3 (10.1, 33.4) g/L)). In this cohort, twenty unique drugs have been used to treat iMCD, including antineoplastic agents, corticosteroids, immunosuppressive agents, anti‐IL6 therapy, and immunoglobulins. Rituximab and siltuximab are the most frequently administered drugs, administered to 50% and 39% of treated patients, respectivelySummary/Conclusion:European iMCD patients in the ACCELERATE registry demonstrate clinical, laboratory, and pathological features consistent with iMCD found in other regions of the world. Despite there only being one EMA‐approved therapy for iMCD, many other agents are being used off‐label. Rituximab is the most frequently administered drug in patients in both the US and European centers, and siltuximab is among the top three drugs across both arms of the registry. Data collection is ongoing, and further work and additional patients are needed to assess the effectiveness of treatments in this cohort.

7550 Background: Since the introduction of combination antiretroviral therapy (ART) the outcome of HIV-associated lymphoma has significantly improved. However, aggressive lymphomas remain the most frequent AIDS-defining event leading to death. We aimed to analyse lymphoma-related mortality in the German HIV-Lymphoma Cohort Study. Methods: This prospective multicenter cohort study includes adult HIV- 1 infected patients with biopsy or cytologically proven HIV-related lymphoma diagnosed at 33 participating centers in Germany and Austria since January 2005. Data on HIV-infection and lymphoma characteristics, treatments and outcomes recorded until December 2014 were analyzed. Pts with T-cell lymphomas, indolent lymphomas and primary central nervous system lymphomas were excluded from the present analysis. Results: Of 499 pts (463 males, 36 females) 394 had aggressive NHL and 105 HL. At the time of lymphoma diagnosis the median age was 44.7 years (range, 22–74.7) and the median CD4-cell count was 210/μl (0–1586). 344 of 499 pts (69%) were diagnosed with advanced stage (III/IV) disease. 214 pts (43%) were ART-naïve, 175 pts (35%) had a viral load below the detection limit of < 50 HIV RNA copies/ml at lymphoma diagnosis while on ART, and 110 pts (22%) had a viral load > 50 copies/ml while on ART or while discontinuing ART. After a median follow-up of 1.8 years 25 of 105 HL (24%) and 122 of 394 NHL pts (31%) have died with an overall death rate of 30% (147/499). 84 of 147 pts (57%) died of relapsed/refractory lymphoma, 75 of 122 pts with NHL (62%) and 9 of 25 pts with HL (36%). In 30 of 147 pts (20%) the cause of death were infections during or shortly after primary chemotherapy (7/25 pts with HL [28%] and 23/122 of pts with NHL [19%]). CD4-cell count was 164/μl in pts who died of infections compared to 293/μl in pts alive (P< 0.001). Further causes of death were secondary malignancies (n = 6; 4%), AIDS-defining events other than NHL (n = 6; 4%) including 4 cases of progressive multifocal leukoencephalopathy, and miscellaneous (n = 21; 14%). Conclusions: HIV-related lymphomas mainly occur in ART-naïve or insufficiently treated patients. The major cause of death is relapsed/refractory lymphoma followed by infections during or shortly after primary chemotherapy.

OBJECTIVES:Dolutegravir (DTG), a second-generation integrase strand transfer inhibitor (INSTI), is now among the most frequently used antiretroviral agents. However, recent reports have raised concerns about potential neurotoxicity.METHODS:We performed a retrospective analysis of a cohort of HIV-infected patients who had initiated an INSTI in two large German out-patient clinics between 2007 and 2016. We compared discontinuation rates because of adverse events (AEs) within 2 years of starting treatment with dolutegravir, raltegravir or elvitegravir/cobicistat. We also evaluated factors associated with dolutegravir discontinuation.RESULTS:A total of 1950 INSTI-based therapies were initiated in 1704 patients eligible for analysis within the observation period. The estimated rates of any AE and of neuropsychiatric AEs leading to discontinuation within 12 months were 7.6% and 5.6%, respectively, for dolutegravir (n = 985), 7.6% and 0.7%, respectively, for elvitegravir (n = 287), and 3.3% and 1.9%, respectively, for raltegravir (n = 678). Neuropsychiatric AEs leading to dolutegravir discontinuation were observed more frequently in women [hazard ratio (HR) 2.64; 95% confidence interval (CI) 1.23-5.65; P = 0.012], in patients older than 60 years (HR: 2.86; 95% CI: 1.42-5.77; P = 0.003) and in human leucocyte antigen (HLA)-B*5701-negative patients who initiated abacavir at the same time (HR: 2.42; 95% CI: 1.38-4.24; P = 0.002).CONCLUSIONS:In this large cohort, the rate of discontinuation of dolutegravir because of neuropsychiatric adverse events was significantly higher than for other INSTIs, at almost 6% within 12 months. Despite the limitations of this retrospective study, the almost three-fold higher discontinuation rates observed amongst women and older patients underscore the need for further investigation, especially in patient populations usually underrepresented in clinical trials.

We thank Capetti and colleagues for their interesting comments 1 which suggest that morning dosing may help resolve dolutegravir (DTG)-related insomnia and sleep disorders. In our retrospective study 2, we had no information on the timing of DTG administration in most of our patients. However, in our experience, patterns of adverse events leading to DTG discontinuation were heterogeneous and not limited to sleep disorders. They included dizziness, depression, headache, paraesthaesia, poor concentration and slow thinking. Although relatively rare, reversible and usually mild to moderate, these events were less frequently seen with elvitegravir or raltegravir. Since our initial publication, we have seen additional patients complaining of dizziness or cognitive problems 1–2 h after DTG intake, suggesting that peak plasma levels may be too high in these individuals. Other groups from the Netherlands 3, Italy 4, 5 and France 6 have also reported unexpected adverse events with DTG. These include anxiety, irritability and depression but also headache and musculoskeletal pain. In these patients, it seems unlikely that a switch to morning dosing would resolve neuropsychiatric symptoms. Of note, Capetti and colleagues have confirmed our observations that neuropsychiatric events were more frequent in older and in female patients, suggesting a pharmacokinetic problem. Indeed, recent pharmacokinetic studies have found higher drug exposure in older patients 7, 8 but also in patients treated with atazanavir or cobicistat 9, 10. We believe that there is an urgent need for more data on drug levels in these patient populations, especially in individuals who develop neuropsychiatric events. Capetti and colleagues' univariate analysis did not find a higher event rate with abacavir use. This is in contrast to our findings and those of other studies 3, 4 and it is debatable whether currently available data are sufficient to exclude a drug−drug interaction between abacavir and DTG 11, 12. More recently, a preliminary study from Japan reported an association between DTG plasma trough concentrations and neuropsychiatric events 13. However, it remains to be seen whether neuropsychiatric events are mainly driven by increased drug exposure and whether they represent a class effect, as such symptoms have also been reported, albeit at a lower incidence, with other integrase strand transfer inhibitors (INSTIs). Simple re-analysis of randomized clinical trials 14 in which patients and investigators may have been reluctant to report side effects (because they knew that patients could be removed from the study) does not sufficiently address these emerging concerns. Hinting at a high background rate of psychiatric conditions among people living with HIV (PLWH) 14 is stigmatizing and unhelpful as, in our experience, many patients without any prior psychiatric morbidity were affected. Given the widespread use and the anticipated long-term exposure to this drug class, we would urge the manufacturers to perform further studies evaluating neuropsychiatric events associated with INSTI use. This should include not only pharmacokinetics and pharmacogenetics, but also analysis of the sleep architecture in PLWH and neuropsychological testing in asymptomatic patients on INSTIs, in order to elucidate subclinical cognitive changes.

Objectives The incidence of sexually transmitted hepatitis C virus (HCV) reinfection is on the rise in HIV-infected men who have sex with men (MSM). Data on natural history of acute hepatitis C and possible factors associated with spontaneous clearance are limited. The aim of this study was to analyse the outcome of HCV reinfections in HIV-positive MSM. Methods A retrospective analysis was carried out on patients with more than one sexually acquired HCV infection who were diagnosed at four major German HIV and hepatitis care centres. Reinfection was defined by genotype or phylogenetic clade switch, detectable HCV RNA after a sustained virological response (SVR) or after spontaneous clearance (SC). Results In total, 48 HIV-positive MSM were identified with HCV reinfection, among them 11 with a third episode and one patient with four episodes. At the first episode, 43 and five patients had an SVR and SC, respectively. The second episode was accompanied by a genotype switch in 29 patients (60%). Whereas 30 and nine patients showed an SVR and SC, respectively, eight patients developed chronic hepatitis. Neither HCV genotype switch nor interleukin-28B genotype was associated with SC. However, SC rates at the second episode were higher for patients with SC at the first episode compared with patients without SC (60 vs. 14%, respectively; P=0.03). Two patients with SC at the first episode were reinfected with the same genotype. Conclusions Multiple reinfections in HIV-infected MSM do occur, with or without genotype switch, and with prior SC of previous episodes. In this large case series, except for SC at the first episode, no factor was of value in clinical decision-making for early therapeutic intervention in acute HCV reinfection.

Overall survival (OS) of patients with acquired immunodeficiency syndrome (AIDS)-related Burkitt lymphoma (BL), diffuse large B-cell lymphoma (DLBCL) and plasmablastic lymphoma (PBL) was analysed in the German AIDS-related-Lymphoma-Cohort-Study. Of 291 patients prospectively included between January 2005 and December 2012, 154 had DLBCL, 103 BL and 34 PBL. Two-year OS rates were similar between BL (69%) and DLBCL patients (63%) but lower for PBL patients (43%). Intermediate (Hazard ratio [HR] 4·1 95% confidence interval [CI] 1·98-8·49) or high (HR 4·92 95% CI 2·1-11·61) International Prognostic Index, bone marrow involvement (HR 1·69 95% CI 1·00-2·84) and PBL histology (HR 2·24 95% CI 1·24-4·03) were independent predictors of mortality.

The incidence of HIV‐related non‐Hodgkin lymphoma (NHL) but not that of Hodgkin lymphoma (HL) has been declining. The aim of the study was to compare HIV‐infected patients with NHL and HL with respect to antiretroviral therapy (ART) exposure at the time of lymphoma diagnosis.

INTRODUCTION:There was a growing need for practical guidelines for the most common OIs in Germany and Austria under consideration of the local epidemiological conditions.MATERIALS AND METHODS:The German and Austrian AIDS societies developed these guidelines between March 2010 and November 2011. A structured Medline research was performed for 12 diseases, namely Immune reconstitution inflammatory syndrome, Pneumocystis jiroveci pneumonia, cerebral toxoplasmosis, cytomegalovirus manifestations, candidiasis, herpes simplex virus infections, varizella zoster virus infections, progressive multifocal leucencephalopathy, cryptosporidiosis, cryptococcosis, nontuberculosis mycobacteria infections and tuberculosis. Due to the lack of evidence by randomized controlled trials, part of the guidelines reflects expert opinions. The German version was accepted by the German and Austrian AIDS Societies and was previously published by the Arbeitsgemeinschaft der Wissenschaftlichen Medizinischen Fachgesellschaften (AWMF; German Association of the Scientific Medical Societies).CONCLUSION:The review presented here is a translation of a short version of the German-Austrian Guidelines of opportunistic infections in HIV patients. These guidelines are well-accepted in a clinical setting in both Germany and Austria. They lead to a similar treatment of a heterogeneous group of patients in these countries.

ObjectivesAntiretroviral therapy reduces mortality and morbidity in HIV‐infected individuals most markedly when initiated early, before advanced immunodeficiency has developed. Late presentation for diagnosis and care remains a significant challenge. To guide public health interventions effectively it is crucial to describe the factors associated with late presentation.MethodsCase surveillance data for all individuals newly diagnosed with HIV infection in Germany in the years 2001–2010 and data for the years 1999–2010 from the German Clinical Surveillance of HIV Disease (ClinSurv) cohort study, a large multicentre observational study, were analysed. Factors associated with late presentation (CD4 count < 350 cells/μL or clinical AIDS) were assessed using descriptive statistics and multivariable logistic regression methods.ResultsAmong 22 925 eligible patients in the national surveillance database, 49.5% were late presenters for HIV diagnosis. Among 6897 treatment‐naïve patients in the ClinSurv cohort, 58.1% were late presenters for care. Late presenters for care were older (median 42 vs. 39 years for early presenters), more often heterosexuals from low‐prevalence countries (18.1% vs. 15.5%, respectively) and more often migrants (18.2% vs. 9.7%, respectively; all P < 0.005). The probability of late presentation was >65% throughout the observation period in migrants. The probability of late presentation for care clearly decreased in men who have sex with men (MSM) from 60% in 1999 to 45% in 2010.ConclusionsIn Germany, the numbers of late presenters for HIV diagnosis and care remain high. The probability of late presentation for HIV diagnosis seems to be particularly high for migrants. These results argue in favour of targeted test promotion rather than opt‐out screening. Late presentation for care seems to be an additional problem after HIV diagnosis.