Search

The protocol used to induce cell death for generating vaccines from whole tumor cells is a critical consideration that impacts vaccine efficacy. Here we compared how different protocols used to induce cell death impacted protection provided by a prophylactic whole tumor cell vaccine in a mouse melanoma model. We found that melanoma cells exposed to γ-irradiation or lysis combined with UV-irradiation (LyUV) provided better protection against tumor challenge than lysis only or cells exposed to UV-irradiation. Furthermore, we found that the immunoregulatory cytokine, IL-27 enhanced protection against tumor growth in a dose-dependent manner when combined with either LyUV or γ-irradiated whole tumor cell vaccine preparations. Taken together, this data supports the use of LyUV as a potential protocol for developing whole tumor cell prophylactic cancer vaccines. We also showed that IL-27 can be used at low doses as a potent adjuvant in combination with LyUV or γ-irradiation treated cancer cells to improve the protection provided by a prophylactic cancer vaccine in a mouse melanoma model.

Additional file 2. Detailed DVH results table. Organ at risk and target volume DVH dose difference separated by individual parameters for male and female cohorts. Note: Some structure parameters are not included as sample size is too small for individual analysis.

Additional file 1. DVH data collection table. Organ at risk and target volume DVH parameters assessed by treatment site.

Medical image translation (e.g. CT to MR) is a challenging task as it requires I) faithful translation of domain-invariant features (e.g. shape information of anatomical structures) and II) realistic synthesis of target-domain features (e.g. tissue appearance in MR). In this work, we propose Manifold Disentanglement Generative Adversarial Network (MDGAN), a novel image translation framework that explicitly models these two types of features. It employs a fully convolutional generator to model domain-invariant features, and it uses style codes to separately model target-domain features as a manifold. This design aims to explicitly disentangle domain-invariant features and domain-specific features while gaining individual control of both. The image translation process is formulated as a stylisation task, where the input is "stylised" (translated) into diverse target-domain images based on style codes sampled from the learnt manifold. We test MDGAN for multi-modal medical image translation, where we create two domain-specific manifold clusters on the manifold to translate segmentation maps into pseudo-CT and pseudo-MR images, respectively. We show that by traversing a path across the MR manifold cluster, the target output can be manipulated while still retaining the shape information from the input.

Table S1. Participating centers in the VESPA audit and explanatory variables details for each center. Table S2. Statistical testing of the differences between audit results (GMV) for the explanatory variables. Results with asterisk indicate significant differences where Variable 1 (V1) has lower GMV than Variable 2 (V2). (DOCX 25 kb)