Adaptive therapy exploits the self-organization of tumor cells to delay the outgrowth of resistant subpopulations successfully. When the tumor has aggressive resistant subpopulations, the outcome of adaptive therapy was not superior to maximum tolerated dose therapy (MTD). To explore methods to improve the adaptive therapy’s performance of this case, the tumor system was constructed by osimertinib-sensitive and resistant cell lines and illustrated by the Lotka-Volterra model in this study. Restore index proposed to assess the system reachability can predict the duration of each treatment cycle. Then the threshold of the restore index was estimated to evaluate the timing of interrupting the treatment cycle and switching to high-frequency administration. The introduced reachability-based adaptive therapy and classic adaptive therapy were compared through simulation and animal experiments. The results suggested that reachability-based adaptive therapy showed advantages when the tumor has an aggressive resistant subpopulation. This study provides a feasible method for evaluating whether to continue the adaptive therapy treatment cycle or switch to high-frequency administration. This method improves the gain of adaptive therapy by taking into account the benefits of tumor intra-competition and the tumor control of killing sensitive subpopulation.

Background: Alzheimer’s disease (AD) reserve theory is based on specific individual characteristics that are associated with a higher resilience against neurodegeneration and its symptoms. A given degree of AD pathology may contribute to varying cognitive decline levels in different individuals. Although this phenomenon is attributed to reserve, the biological mechanisms that underpin it remain elusive, which restricts translational medicine research and treatment strategy development. Methods: Network-based approaches were integrated to identify AD reserve related genes. Then, AD brain transcriptomics data were clustered into co-expression modules, and a Bayesian network was developed using these modules plus AD reserve related phenotypes. The directed acyclic graph suggested that the module was strongly associated with AD reserve. The hub gene of the module of interest was filtered using the topological method. Validation was performed in the multi-AD brain transcriptomic dataset. Results: We revealed that the RALYL (RALY RNA Binding Protein-like) is the hub gene of the module which was highly associated with AD reserve related phenotypes. Pseudo-time projections of RALYL revealed the changes in relative expression drivers in the AD and control subjects over pseudo-time had distinct transcriptional states. Notably, the expression of RALYL decreased with the gradual progression of AD, and this corresponded to MMSE decline. Subjects with AD reserve exhibited significantly higher RALYL expression than those without AD reserve. Conclusion: The present study suggests that RALYL may be associated with AD reserve, and it provides novel insights into the mechanisms of AD reserve and highlights the potential role of RALYL in this process.

The risk of diabetic cardiovascular complications is closely linked to the length of hyperglycemia exposure. Mitophagy plays a significant role in vascular endothelial injury. However, the specific mechanisms by which mitophagy contributes to endothelial injury during sustained hyperglycemia remain unclear. In diabetic ApoE^−/− mice and human umbilical vein endothelial cell (HUVEC) models, mitophagy is enhanced following short‐term and long‐term high‐glucose exposure. Short‐term high‐glucose exposure promotes Parkin‐mediated mitophagy and upregulates mitochondrial fission protein 1 (Fis1) expression, whereas long‐term high‐glucose exposure suppresses Parkin‐mediated mitophagy and downregulates Fis1. With prolonged high‐glucose exposure, Syntaxin 17 (STX17) translocates from the endoplasmic reticulum to the mitochondria, activating STX17‐mediated mitophagy. Silencing STX17 alleviates mitochondrial degradation, decreases reactive oxygen species (ROS) levels, enhances endothelial nitric oxide synthase (eNOS) phosphorylation, and reduces apoptosis. Silencing Fis1 accelerates the switching to STX17‐mediated mitophagy, worsening endothelial dysfunction, whereas Fis1 overexpression prevents this switching, reducing ROS and apoptosis and enhancing eNOS phosphorylation. In summary, these findings suggest that the switching from Parkin‐mediated to STX17‐mediated mitophagy drives vascular endothelial injury following long‐term hyperglycemic exposure, providing valuable insights into therapeutic strategies for diabetic cardiovascular complications.

It is well accepted by the scientific community that the accumulation of beta-amyloid (Aβ) may be involved in endothelial dysfunction during Alzheimer’s disease (AD) progression; however, anti-Aβ anti-bodies, which remove Aβ plaques, do not improve cerebrovascular function in AD animal models. The reasons for these paradoxical results require investigation. We hypothesized that Aβ exposure may cause persistent damage to cerebral endothelial cells even after Aβ is removed (referred to as cerebrovascular endothelial damage memory). In this study, we aimed to investigate whether cerebrovascular endothelial damage memory exists in endothelial cells. hCMEC/D3 cells were treated with Aβ_1–42 for 12 h and then Aβ_1–42 was withdrawn for another 12 h incubation to investigate whether cerebrovascular endothelial damage memory exists in endothelial cells. A mechanism-based kinetics progression model was developed to investigate the dynamic characters of the cerebrovascular endothelial damage. After Aβ_1–42 was removed, the sirt-1 levels returned to normal but the cell vitality did not improve, which suggests that cerebrovascular endothelial damage memory may exist in endothelial cells. Sirt-1 activator SRT2104 and NAD⁺ (Nicotinamide Adenine Dinucleotide) supplement may dose-dependently relieve the cerebrovascular endothelial damage memory. sirt-1 inhibitor EX527 may exacerbate the cerebrovascular endothelial damage memory. Kinetics analysis suggested that sirt-1 is involved in initiating the cerebrovascular endothelial damage memory; otherwise, NAD⁺ exhaustion plays a vital role in maintaining the cerebrovascular endothelial damage memory. This study provides a novel feature of cerebrovascular endothelial damage induced by Aβ.

Objective: The treatment plans designed with the guidance of the mathematical model and adaptive strategy can trap tumor subpopulations in a periodic and controllable loop. But this process requires detailed information about the tumor system, which is difficult to obtain. Therefore, we wondered whether the fixed periodic treatment plans designed with the typical values of population parameters could be applied to a similar tumor system without complete information. Methods: A binary tumor system constructed by an EGFR-mutant and a KRAS-mutant cell line was used to explore the applicability of the fixed periodic treatment plans. The dynamics of this system were described by combining the Lotka-Volterra model with the framework of the nonlinear mixed-effects model. The typical values of population parameters were used to design the plans, and the robust plans were screened through parameter variation. These screened plans were examined their applicability in animal experiments and simulations. Results: In animal experiments where system parameters vary from − 30% to 30%, the “osimertinib administration, withdrawal, FK866 administration and withdrawal” plan can trap subpopulations of the system in periodic cycles. In simulation, when there was an unknown resistant subpopulation, the screened fixed periodic treatment plans can still delay the evolution of resistance. The median outcomes of screened plans were better than conventional sequential treatment in most cases. There was no significant difference between the outcomes of the screened plan with median stability and the optimal therapy. The evolutionary trajectories of these two plans were similar. Conclusion: According to the results, these fixed periodic plans should be tried in treatment even the information of the tumor system was incomplete.

The impact of mitochondrial haplogroups on Alzheimer’s disease (AD) risk has not been fully elucidated and warrants further investigation at the subgroup level. The aim of this research is to evaluate the association between mitochondrial haplogroups and AD risk in subgroups level. In total, 809 AD Neuroimaging Initiative subjects were assessed using mtDNA sequencing, the AD Assessment Scale-Cognitive Subscale (ADAS-cog), hippocampal volume measurements, the hypometabolic convergence index (HCI), and MCI-to-AD conversion proportion measurements. The frequency of haplogroup J was significantly higher than that of other haplogroups in the AD group (p = 0.013). According to the correlation between haplogroup J-specific SNPs and ADAS-cog, haplogroup J was divided into four subgroups harboring exacerbating SNPs, protective SNPs, both exacerbating and protective SNPs, or irrelevant SNPs. The subgroups harboring exacerbating SNPs exhibited higher AD risk represented by the levels of ADAS-cog, hippocampal volume, HCI, and MCI-to-AD conversion proportion than other subgroups. Heterogeneity existed among the subgroups of haplogroup J, which suggested that different subgroups exhibited different levels of AD risk. This study provides novel insights into the correlation between mitochondrial haplogroups and AD risk.