A malignus melanoma immunterápiájának lehetoségei

Despite their well-documented immunogenicity, malignant melanomas belong to the most aggressive tumor types. A potential explanation for this is the suboptimal activation of tumor infiltrating T cells. In order to boost immune responses against tumors, a variety of treatment modalities have been tested in animal models and in clinical setting. Antigen-nonspecific approaches (e.g., IFN-alpha and IL-2), as well as active specific immunotherapeutical modalities based on the use of autologous or allogeneic tumor cell-save been investigated in clinical trials of melanoma. The identification of melanoma-associated antigens has opened new avenues in antigen-specific immunotherapy. A promising alternative for the delivery of different forms of melanoma antigens is the application of dendritic cells, the most potent antigen presenting cells capable of eliciting efficient T-cell response. Beside active immunotherapy, immune response against melanoma antigens could be increased through the adoptive transfer of tumor infiltrating lymphocytes or antigen specific T-cell clones. The most important conclusion that can be drawn from the results of published immunotherapy studies is that these modalities are able to induce durable complete tumor regressions,mostly with reasonable toxicity; however, generally only in a minority of patients. This points to the importance of appropriate patient selection, with regard to the expression of the targeted antigens and HLA molecules, as well as to the general immunocompetence of the patients. A crucial and still unsolved question is monitoring immune activation during treatment, although there are promising new tools that could prove useful in this respect. The presence of tumor-reactive CTL in the circulation or in the tumors does not guarantee an efficient immune response. It is important to assess if these T cells are in an activated and functional state. Finally, in several single target antigen-based clinical studies a therapy-induced immunoselection of antigen-negative clones, leading to disease progression, was observed. This could be overcome with the use of antigen cocktails or whole tumor approaches. A better understanding of the mechanisms of action of immunotherapeutical modalities may enhance the success rate of these strategies.     

Lévy flights in evolutionary ecology

We are interested in modeling Darwinian evolution resulting from the interplay of phenotypic variation and natural selection through ecological interactions. The population is modeled as a stochastic point process whose generator captures the probabilistic dynamics over continuous time of birth, mutation, and death, as influenced by each individual's trait values, and interactions between individuals. An offspring usually inherits the trait values of her progenitor, except when a random mutation causes the offspring to take an instantaneous mutation step at birth to new trait values. In the case we are interested in, the probability distribution of mutations has a heavy tail and belongs to the domain of attraction of a stable law and the corresponding diffusion admits jumps. This could be seen as an alternative to Gould and Eldredge's model of evolutionary punctuated equilibria. We investigate the large-population limit with allometric demographies: larger populations made up of smaller individuals which reproduce and die faster, as is typical for micro-organisms. We show that depending on the allometry coefficient the limit behavior of the population process can be approximated by nonlinear Lévy flights of different nature: either deterministic, in the form of non-local fractional reaction-diffusion equations, or stochastic, as nonlinear super-processes with the underlying reaction and a fractional diffusion operator. These approximation results demonstrate the existence of such non-trivial fractional objects; their uniqueness is also proved.     

Kálmán filters for continuous-time movement models

We introduce fast implementations for the likelihood functions, telemetry error filters, probabilistic trajectory and velocity reconstructions, and movement-path simulations for a large class of continuous-time movement models. This class of models includes all of the basic continuous-time models that have been applied to animal movement. A diverse array of movement behaviors can be modeled from within this framework, including range residence, persistence of motion, migration, range shifting, and territorial patrol. The fast algorithms presented here, based upon the Kálmán filter, are critical for applying movement analyses to the evergrowing number of modern datasets that feature thousands or more observed animal locations, and they are key to the continuous-time movement modeling (ctmm) R package.