Abstract While many Bayesian state-space models for infectious disease processes focus on population infection dynamics (e.g., compartmental models), in this work we examine the evolution of infection processes and the complexities of the immune responses within the host using these techniques. We present a joint Bayesian state-space model to better understand how the immune system contributes to the control of Leishmania infantum infections over the disease course. We use longitudinal molecular diagnostic and clinical data of a cohort of dogs to describe population progression rates and present evidence for important drivers of clinical disease. Among these results, we find evidence for the importance of co-infection in disease progression. We also show that as dogs progress through the infection, parasite load is influenced by their age, ectoparasiticide treatment status, and serology. Furthermore, we present evidence that pathogen load information from an earlier point in time influences its future value and that the size of this effect varies depending on the clinical stage of the dog. In addition to characterizing the processes driving disease progression, we predict individual and aggregate patterns of Canine Leishmaniasis progression. Both our findings and the application to individual-level predictions are of direct clinical relevance, presenting possible opportunities for application in veterinary practice and motivating lines of additional investigation to better understand and predict disease progression. Finally, as an important zoonotic human pathogen, these results may support future efforts to prevent and treat human Leishmaniosis. 1 AUTHOR SUMMARY The immune system is a complex network that involves organs, cells, and proteins working together with the main purpose of protecting the body against harmful microorganisms such as bacteria, viruses, fungi, and toxins. To explore and study the responses of the host immune system during the course of a disease, we modeled the interaction between pathogen load, antibody responses, and the clinical presentation of this complex system. Specifically, we focused on Canine Leishmaniasis (CanL), a vector-borne disease caused by a parasite that affects internal organs of the body and is known to be fatal if patients remain untreated. In addition, we also considered the impact of possible co-infections with other diseases, which could potentially interact with many disease processes and contribute to different outcomes for infected subjects. With CanL specifically, we consider the presence of Borrelia, Anaplasma, Ehrlichia ,, and Heartworm. In general, one limitation in vaccination strategies is a focus on neutralizing antibodies, without incorporating the broader complexities of immune responses. Here, we explore this complexity by jointly considering the interaction between pathogen and antibody development with the purpose of improving our understanding of the processes of disease progression and natural immunity. In this paper, we present a Bayesian model specification for immune responses to a Leishmania infection considering a tick-borne co-infection study. The model implementation is based on the general vector autoregressive (VAR) model, adapted to the problem under study. While the methodology around Bayesian VAR models is not new in the literature, in this work we adapt the more general VAR approach in a parsimonious way to a particular subclass of longitudinal problems. We believe our defined Bayesian model is useful to clinicians and veterinarians to better understand the immune responses and Leishmania infection control over time, which makes this work a novel application of Bayesian VAR models. We present evidence that pathogen load information from an earlier point in time influences its future value and that the size of this effect varies depending on the CanL clinical stage of the dog. In addition to characterizing evidence for the processes driving disease progression, we predict individual and aggregate patterns of CanL progression. The structure of this paper starts in Section 2 with an introduction to CanL infection as well as a discussion of possible co-infection with other pathogens. In Section 3, we include a description of the motivating prospective study along with the measured individual-level variables, a definition of the clinical signs of leishmaniosis infection, and a description of the available data coming from the study. In addition, this section explains the dynamic process and corresponding model specification via Bayesian methodology and a statement of contribution. A summary of prior distributions for model parameters, model implementation details, and convergence diagnostics are also included. In Section 4, we provide summary results from the posterior distribution as well as a summary of the corresponding disease progression forecasts. In Section 5, we discuss the results and describe future considerations to improve and extend the model.
