 I'm happy to present to you our studies on effectiveness of combination therapy with liposomal amphotericin B and milltefossing for visceral leshmaniasis in HIV co-infection. The DNDI MSF strategy for policy change in East Africa. This presentation will be structured as follows. I'll start with an overview of visceral leshmaniasis, then the methods, results, discussions, conclusions, recommendations, and lastly, acknowledgments. Visceral leshmaniasis, VL, or kalaza, is a parasitic disease which is often fatal if untreated. Ethiopia is one of the top six countries with high burden of VL. 3,400 to 5,000 VL cases occur every year. In North West Ethiopia, where this picture is taken, 20% of VL patients are also HIV-confected. This is the highest co-infection rate in the world. Migrant workers are the most at risk. And before MSF intervened in this area, the majority of the patients died. In VL and HIV co-infection, HIV increases VL severity and worsens VL treatment outcomes. High case fatality, high initial parasitological failure rates, and high relapse rates have been reported. What does initial parasitological failure rates mean? This means that when a patient receives the first initial treatment for VL, at the end of treatment, we still find parasites in tissue. And that means the patient will still have to receive more treatment, meaning the patient may stay up to two months or even longer in hospital. VL, in turn, promotes the progression of HIV infection. Initial treatment outcomes in VL-HIV is associated with high initial parasitological failure with all the available VL drugs. For instance, pentavalent anitimonials, they have been shown to cause severe, to cause fatal adverse events resulting in high case fatality rates of up to 33%. Miltiforcing, resulting in high initial parasitological failures of up to 18%. Liposomal amphotericin B, or ambison, results in initial parasitological failure rates of up to 33%. Because combination therapy may improve treatment efficacy, in 2011 MSF introduced combination treatment with ambison infusion at 30 milligrams total dose and miltiforcing orally for 28 days at 100 milligrams per day. As fast line treatment for VL and in HIV co-infected patients in its treatment center in Abdurafi Health Center in Northwest Ethiopia. A retrospective cohort study in Abdurafi Health Center was followed by an open label randomized clinical trial in Abdurafi and Gonda Hospital in Northwest Ethiopia. In the trial, the WHO protocol of ambison monotherapy at 40 milligrams total dose was compared with the MSF protocol of ambison and miltiforcing. We aimed to determine the effectiveness of this combination regimen. For the retrospective cohort study, ethical approval was obtained from the relevant ethical bodies. We included all adult co-infected patients who were treated between January 2011 and August 2014 with an initial treatment of ambison and miltiforcing. And we excluded patients who were who discontinued treatment, defaulted, transferred out, or had missing treatment outcome. Proportions of initial treatment outcome categories were calculated. We determined predictors of initial parasitological failure and death using multivariable logistic regression. Because of missing CD4 counts, we created a composite marker of advanced HIV defined as WHO stage four, or CD4 less than 50 cells per microliter. We also did a sensitivity analysis of the 173 patients included in the cohort study. The majority were male. They were young. Residents had advanced HIV. And most of them were on antiretroviral treatment before the VR episode. Initial treatment outcomes in the cohort study where it has presented, QA was 83.8%, death 12.7%, and parasitological failure was 3.5%. We found that tuberculosis Advil diagnosis was predictive of parasitological failure. Age greater than 40, hemoglobin less than 6.5%, primary VR, which means no prior history of treatment of VR were predictive of death. In comparison with the previous cohort study that was done in the same population by Ritme, I think most of you probably know him. These results compared to our initial treatment outcomes where it has presented here, we see that with combination treatment, the results are significantly better. 3.5% parasitological failure versus 32.8%. QA rates of 83.8% versus 60.4%. And the death rates were non-significant higher. These death rates are not related to the combination treatment. We believe that it is probably because of the high admission of late stage VR patients during the study period. In the randomized control trial that followed the cohort study, 39 patients were randomized to ambisome and multifaceted, and 20 to ambisome. The QA rates were significantly better. The QA rates were better with the combination treatment. On day 29, the end of the initial treatment, we see it's 67 versus 49.8%. On day 58, end of extended treatment for patients with parasitological failure, we see that the treatment is 91% with combination treatment versus 59% with ambisome. There were some limitations in the cohort study. First, as a retrospective study, we could only study the predictors that were collected. Diagnosis and QA were not systematically confirmed by parasitological tests. And there were many missing CD4 counts. We also did not do long-term follow-up of patients. And neither did we have capacity to perform autopsies. In conclusion, we found parasitological rates with combination therapy were low. We also determined predictors of poor outcome, predictors of death, and predictors of parasitological failure. Knowledge of these predictors may facilitate better management. The WHO, MSF, DNDI, and the Ethiopian Ministry of Health have agreed to implement compassionate use of combination therapy based on the results of the studies. It's also been recommended to do a meta-analysis of the two studies. And lastly, once both studies are published, WHO will launch a consultative meeting to eventually endorse the change in VL HIV case management. I would like to thank several people from MSF, from the DNDI, from the Institute of Tropical Medicine, and from the University of Gondar, Ethiopia, WHO, and the Ethiopian Ministry of Health. This work has also been supported by the AFRI College project. Thank you very much for your attention.