Perivascular tissue models to overcome MGMT-mediated temozolomide resistance in glioblastoma

摘要 胶质母细胞瘤(GBM)是最常见和致命的脑癌。标准的护理是手术切除 然后用烷化剂替莫唑胺(TMZ)处理。然而，两大挑战使GBM 目前无法治疗：1)弥漫性侵袭超出手术切缘；2)对TMZ的耐药性 与DNA损伤修复蛋白MGMT的表达有关。而血管周围壁龛(PVN)从 肿瘤进入周围实质被认为可以调节侵袭、复发和低存活率。 大多数动物胶质瘤模型对TMZ敏感，大多数不表达MGMT，因此很难 在没有表现出TMZ耐药性的动物模型中评估新的治疗方法。这种肿瘤组织工程学 合作项目将开发并彻底表征一种多维工程PVN生物材料， 研究推动GBM侵袭和TMZ耐药的病理生理过程，并加速评估 新型TMZ衍生物，用于靶向弥漫的GBM细胞，而不考虑MGMT状态。我们将使用高级 用微流控技术建立含有模拟边缘透明质酸(HA)的微型明胶水凝胶文库 和嵌入的血管周围网络。我们还使用一种新的合成管道来创建TMZ衍生品 产生无法被MGMT移除的替代DNA修改，我们假设MGMT在MGMT中起作用- 独立时尚。融合这些技术，我们将对使用以下技术形成的工程PVN平台进行基准测试 原代脑神经血管细胞用于快速评估GBM侵袭、MGMT表达和TMZ耐药性 可用于分析具有不同MGMT图谱的细胞系和患者来源的GBM样本。要做到这点， 我们将首先构建并彻底描述一个工程化的血管周围生态位(目标1)。我们将使用这个 基底膜细胞系侵袭和MGMT表达模式基准的新型生物材料(目标2)。最后， 我们将建立TMZ变异体在工程血管周围利基中的预测效果(目标3)。在一起，我们 将开发、表征和基准组织工程化PVN以检查微环境的作用 肿瘤边缘的选择压力与侵袭、MGMT介导的TMZ耐药、 复发，存活率低。与CTEC计划的分数驱动标准一致，我们将制定和 彻底描述一种工程PVN生物材料，表明它符合现有癌症的连续体 模型，使用它来检查无法实现持久生存的潜在现象，并获得可操作的 关于具有有效靶向边缘GBM细胞的潜力的新型TMZ衍生物的洞察 MGMT状态。

ABSTRACT Glioblastoma (GBM) is the most common and lethal form of brain cancer. Standard of care is surgical resection followed by treatment with the alkylating agent temozolomide (TMZ). However, two major challenges make GBM currently untreatable: 1) its diffuse invasion beyond the surgical margin; and 2) TMZ resistance that is tightly linked to expression of the DNA damage repair protein MGMT. While perivascular niches (PVNs) extending from the tumor into the surrounding parenchyma are believed to regulate invasion, recurrence, and poor survival, the majority of animal glioma models are sensitive to TMZ and most do not express MGMT, making it difficult to assess novel therapeutics in animal models that don’t display TMZ resistance. This Cancer Tissue Engineering Collaborative project will develop and thoroughly characterize a multidimensional engineered PVN biomaterial, study pathophysiological processes driving GBM invasion and TMZ resistance, and accelerate the evaluation of novel TMZ derivatives created to target diffuse GBM cells regardless of MGMT status. We will use advanced microfluidics to create libraries of miniaturized gelatin hydrogels containing margin-mimetic hyaluronic acid (HA) and an embedded perivascular network. We also use a novel synthetic pipeline to create TMZ derivatives that generate alternate DNA modifications that cannot be removed by MGMT that we hypothesize work in an MGMT- independent fashion. Merging these technologies, we will benchmark an engineered PVN platform formed using primary brain neurovascular cells for rapid evaluation of GBM invasion, MGMT expression, and TMZ resistance amenable to analysis of cell lines and patient-derived GBM specimens with disparate MGMT profiles. To do this, we will first construct and thoroughly characterize an engineered perivascular niche (Aim 1). We will use this novel biomaterial to benchmark patterns of invasion and MGMT expression in GBM cell lines (Aim 2). Finally, we will establish predictive efficacy of TMZ variants in an engineered perivascular niche (Aim 3). Together, we will develop, characterize, and benchmark a tissue engineered PVN to examine the role of microenvironmental selection pressures in the tumor margin on behaviors related to invasion, MGMT-mediated TMZ resistance, recurrence, and poor survival. Consistent with score-driving criteria of the CTEC program, we will develop and thoroughly characterize an engineered PVN biomaterial, show it fits within the continuum of existing cancer models, use it to examine phenomena underlying the failure to achieve durable survival, and gain actionable insight regarding novel TMZ derivatives with potential to effectively target GBM cells in the margins independent of MGMT status.