Computational modelling of lung cancer evolution

癌症是一种以细胞亚群逐渐发展为侵袭性恶性肿瘤为特征的疾病。这一过程通常以获得不同的基因突变为标志，但虽然一些特定的突变对细胞适应性有已知的影响，但遗传进化对癌症轨迹的总体影响尚不清楚。解开细胞遗传特征与其表型之间的关系对于理解和治疗癌症至关重要。由于癌症的发展还不能在体外完全模拟，许多理论和计算模型已经建立来研究肿瘤形成的机制。这些模型的输出通常与实验数据进行比较，以评估其基本前提的有效性。然而，由于需要对突变对细胞适应性的影响做出限制性假设，这些模型的应用迄今受到限制。在这里，我们设计了基于主体的模型来严格检验关于突变对肺肿瘤形成影响的特定假设。我们将这种方法应用于非小细胞肺癌（NSCLC）肿瘤发生的类器官模型，表明EGFR-L858R突变诱导细胞向邻近细胞发出不分裂的信号。在进一步的工作中，我们将使用更大和更复杂的模型来模拟肺肿瘤的发展。通过将该模型的输出与TRACERx队列（取自人类患者的未经治疗的肺肿瘤的大型数据集）的遗传和转录组数据进行比较，将对该模型进行拟合和测试。我们将使用这个模型来推断控制肺癌发展的可能进化模式。具体来说，我们会问肺癌细胞的适应度是否由基因突变的积累决定，是否独立于基因突变而进化；细胞适应度通常应该被认为是静态的还是时变的；以及控制患者疾病的进化模式是否可以通过已知的生物标志物来预测。这种对进化动力学的理解对于新疗法的发展至关重要，特别是那些旨在控制进化模式和影响疾病轨迹的疗法。研究方向：数学生物学

Cancer is a disease characterised by the gradual development of invasive malignancy by subpopulations of cells. This process is usually marked by the acquisition of distinct genetic mutations, but whilst some specific mutations have known effects on cell fitness, the general influence of genetic evolution on the trajectory of cancer is unclear. Untangling the relationship between a cell's genetic profile and its phenotype is of crucialimportance to understanding and treating cancer. As the development of cancer cannot yet be fully simulated in vitro, many theoretical and computational models have been built to investigate the mechanisms of tumour formation. The outputs of these modelsare usually compared with experimental data to assess the validity of their underlying premises. However, the application of these models has so far been limited by the need to make restrictive assumptions about the influence of mutations on cell fitness.Here, we design agent-based models to rigorously test specific hypotheses about the influence of mutations on lung tumour formation. We apply this approach to an organoid model of tumorigenesis in non-small cell lung cancer (NSCLC) to suggest that the EGFR-L858R mutation induces cells to signal their neighbours not to divide.In further work, we will use larger and more complex models to simulate the development of lung tumours. This model will be fitted and tested by comparing its outputs to genetic and transcriptomic data from the TRACERx cohort (a large dataset of untreated lung tumours taken from human patients). We will use this model to infer possible modes of evolution governing lung cancer development. Specifically, we will ask whether the fitness of lung cancer cells is determined by the accumulation of genetic mutations, orevolves independently of them; whether cell fitness should generally be considered static or time-varying; and whether the mode of evolution governing a patient's disease can be predicted by known biomarkers. This understanding of evolutionary dynamics is crucial to the development of new treatments, especially those which aim to control patterns of evolution and affect the trajectory of disease.Research Area: Mathematical Biology