BackgroundMelanoma is a heterogeneous tumour, but the impact of this heterogeneity upon therapeutic response is not well understood.MethodsSingle cell mRNA analysis was used to define the transcriptional heterogeneity of melanoma and its dynamic response to BRAF inhibitor therapy and treatment holidays. Discrete transcriptional states were defined in cell lines and melanoma patient specimens that predicted initial sensitivity to BRAF inhibition and the potential for effective re-challenge following resistance. A mathematical model was developed to maintain competition between the drug-sensitive and resistant states, which was validated in vivo.FindingsOur analyses showed melanoma cell lines and patient specimens to be composed of >3 transcriptionally distinct states. The cell state composition was dynamically regulated in response to BRAF inhibitor therapy and drug holidays. Transcriptional state composition predicted for therapy response. The differences in fitness between the different transcriptional states were leveraged to develop a mathematical model that optimized therapy schedules to retain the drug sensitive population. In vivo validation demonstrated that the personalized adaptive dosing schedules outperformed continuous or fixed intermittent BRAF inhibitor schedules.InterpretationOur study provides the first evidence that transcriptional heterogeneity at the single cell level predicts for initial BRAF inhibitor sensitivity. We further demonstrate that manipulating transcriptional heterogeneity through personalized adaptive therapy schedules can delay the time to resistance.FundingThis work was funded by the . The funder played no role in assembly of the manuscript.

Cyclic mechanical loading of articular cartilage results in a complex biomechanical environment at the scale of the chondrocytes that strongly affects cellular metabolic activity. Under dynamic loading conditions, the quantitative relationships between macroscopic loading characteristics and solid and fluid mechanical variables in the local cellular environment are not well understood. In this study, an axisymmetric multiscale model of linear biphasic cell-matrix interactions in articular cartilage was developed to investigate the cellular microenvironment in an explant subjected to cyclic confined compressive loading. The model was based on the displacement-velocity-pressure (u-v-p) mixed penalty weighted residual formulation of linear biphasic theory that was implemented in the COMSOL Multiphysics software package. The microscale cartilage environment was represented as a three-zone biphasic region consisting of a spherical chondrocyte with encapsulating pericellular matrix (PCM) that was embedded in a cylindrical extracellular matrix (ECM) subjected to cyclic confined compressive loading boundary conditions. Biphasic material properties for the chondrocyte and the PCM were chosen based on previous in vitro micropipette studies of cells or chondrons isolated from normal or osteoarthritic cartilage. Simulations performed at four loading frequencies in the range 0.01-1.0 Hz supported the hypothesized dual role of the PCM as both a protective layer for the cell and a mechanical transducer of strain. Time varying biphasic variables at the cellular scale were strongly dependent on relative magnitudes of the loading period, and the characteristic gel diffusion times for the ECM, the PCM and the chondrocyte. The multiscale simulations also indicated that axial strain was significantly amplified in the range 0.01-1.0Hz, with a decrease in amplification factor and frequency insensitivity at the higher frequencies. Simulations of matrix degradation due to osteoarthritis indicated that strain amplification factors were more significantly altered when loss of matrix stiffness was exclusive to the PCM. The findings of this study demonstrate the complex dependence of dynamic mechanics in the local cellular environment of cartilage on macroscopic loading features and material properties of the ECM and the chondron.

Summary This study investigates the mechanism of action behind the long-term responses (12–16 months) of two BRAF WT melanoma patients to the AKT inhibitor MK-2206 in combination with paclitaxel and carboplatin. Although single agent MK-2206 inhibited phospho-AKT signaling, it did not impact in vitro melanoma growth or survival. The combination of MK-2206 with paclitaxel and carboplatin was cytotoxic in long-term colony formation and 3D spheroid assays, and induced autophagy. Autophagy was initially protective with autophagy inhibitors and deletion of ATG5 found to enhance cytotoxicity. Although prolonged autophagy induction (>6 days) led to caspase-dependent apoptosis, drug resistant clones still emerged. Autophagy inhibition enhanced the cell death response through reactive oxygen species and could be reversed by anti-oxidants. We demonstrate for the first time that AKT inhibition in combination with chemotherapy may have clinical activity in BRAF WT melanoma and show that an autophagy inhibitor may prevent resistance to these drugs. Significance Approximately 30% of all cutaneous melanomas are wild-type for both BRAF and NRAS. As yet, no targeted therapy strategies exist for this sub-set of tumors. Constitutive signaling through the PI3K/AKT pathway is a common occurrence in cutaneous melanoma, irrespective of the driver mutation. Here we report durable responses to the AKT inhibitor MK-2206 in combination with carboplatin and paclitaxel in two patients with BRAF wild-type melanoma. Through mechanistic studies, we demonstrate a role for autophagy induction in the response to the AKT inhibitor/chemotherapy combination and suggest that autophagy inhibitors may be one strategy to enhance efficacy in the clinical setting.

Adaptive therapy is an evolution-based treatment approach that aims to maintain tumor volume by employing minimum effective drug doses or timed drug holidays. For successful adaptive therapy outcomes, it is critical to find the optimal timing of treatment switch points in a patient-specific manner. Here we develop a combination of mathematical models that examine interactions between drug-sensitive and resistant cells to facilitate melanoma adaptive therapy dosing and switch time points. The first model assumes genetically fixed drug-sensitive and -resistant popul tions that compete for limited resources. The second model considers phenotypic switching between drug-sensitive and -resistant cells. We calibrated each model to fit melanoma patient biomarker changes over time and predicted patient-specific adaptive therapy schedules. Overall, the models predict that adaptive therapy would have delayed time to progression by 6−25 months compared to continuous therapy with dose rates of 6−74% relative to continuous therapy. We identified predictive factors driving the clinical time gained by adaptive therapy, such as the number of initial sensitive cells, competitive effect, switching rate from resistant to sensitive cells, and sensitive cell growth rate. This study highlights that there is a range of potential patient-specific benefits of adaptive therapy and identifies parameters that modulate this benefit.

We present an integrated study to understand the key role of senescent fibroblasts in driving melanoma progression. Based on the hybrid cellular automata paradigm we developed an in silico model of normal skin. The model focuses on key cellular and microenvironmental variables that regulate interactions among keratinocytes, melanocytes and fibroblasts, key components of the skin. The model recapitulates normal skin structure, and is robust enough to withstand physical as well as biochemical perturbations. Furthermore, the model predicted the important role of the skin microenvironment in melanoma initiation and progression. Our in vitro experiments showed that dermal fibroblasts, which are an important source of growth factors in the skin, adopt a secretory phenotype that facilitates cancer cell growth and invasion when they become senescent. Our co-culture experiments showed that the senescent fibroblasts promoted the growth of non-tumorigenic melanoma cells and enhanced the invasion of advanced melanoma cells. Motivated by these experimental results, we incorporated senescent fibroblasts into our model and showed that senescent fibroblasts transform the skin microenvironment and subsequently change the skin architecture by enhancing the growth and invasion of normal melanocytes. The interaction between senescent fibroblasts and the early stage melanoma cells leads to melanoma initiation and progression. Of microenvironmental factors that senescent fibroblasts produce, proteases are shown to be one of key contributing factors that promoted melanoma development from our simulations. Although not a direct validation, we also observed increased proteolytic activity in stromal fields adjacent to melanoma lesions in human histology. This leads us to the conclusion that senescent fibroblasts may create a pro-oncogenic skin microenvironment that cooperates with mutant melanocytes to drive melanoma initiation and progression and should therefore be considered as a potential future therapeutic target. Interestingly, our simulations to test the effects of a stroma-targeting therapy that negates the influence of proteolytic activity showed that the treatment could be effective in delaying melanoma initiation and progression.