Development of PET imaging biomarkers to predict enhanced glioblastoma radiotherapy by a novel H-NOX oxygen carrier

1.项目摘要/摘要 低氧(低氧)是已知的肿瘤侵袭性和肿瘤抗辐射能力的关键驱动因素 治疗(RT)在各种实体癌症。鉴于大约50%的癌症患者接受了RT(~800,000 每年在美国)，Omniox已经开发出一种新的氧气载体蛋白OMX，它被设计成 特异性地将氧气输送到肿瘤的低氧区域，并增强放射治疗的效果。Omniox由 NCI SBIR(第一、第二、第二阶段)和Wellcome Trust翻译奖，以测试其主要候选药物作为 用于治疗新诊断的胶质母细胞瘤(GB)患者的标准护理RT的辅助。Omniox已经 已获得FDA Pre-IND支持，目前正在进行支持IND的制造和GLP 计划在加州大学旧金山分校进行的首例人类临床试验的毒理学研究 (加州大学旧金山分校)2017年。重要的是，使用低氧选择剂的非侵入性PET成像显示，低氧 GB肿瘤的体积与疾病进展迅速和生存不良密切相关，这是因为它 对标准护理RT的钝化效应。之前在人体临床试验中为肿瘤充氧的努力 结果不明确，部分原因是肿瘤缺氧负担的异质性和缺乏针对性 选择病人的方法。因此，这项提议的目标是发展预测性非侵入性 PET成像生物标记物，将识别可实质上受益的GB患者群体 OMX疗法。 Omniox和加州大学戴维斯分校兽医学院完成了一项犬只0期临床试验，以评估 OMX在犬体内的安全性、药代动力学和药效学 脑瘤。作为试验的一部分，Omniox与Brain Biosciences合作使用他们设计的PET扫描仪 专门用于脑成像(CerePET™)。Omniox和这笔AIP赠款的主要合作者Simon博士 加州大学戴维斯分校的Cherry(目标1)和Allison Zwingenberger(目标2)，以及脑生物科学的David Beylin(目标1 和2)正在开发对犬脑患者进行PET成像的方案，初步扫描已经成功 已生成。在这项建议中，我们将以我们对啮齿动物和犬类的积极初步结果为基础来评估 肿瘤乏氧和OMX肿瘤积聚的显像可作为预测OMX活性的生物标志物 大鼠颅内GB模型和犬脑胶质瘤患者。然后这些数据将被用来通知IND 应用和人类生物标记物的临床试验。 OMX由加州大学伯克利分校发现的耐热H-NOX蛋白家族发展而来，是一种三聚体， 聚乙二醇化H-NOX氧(O2)结合突变体通常被设计为向缺氧肿瘤组织深处扩散 与血管渗漏有关。H-NOX部分被设计成对O2有很高的亲和力 因此，它在常氧组织中保留氧气，并在严重缺氧的区域特别释放氧气。 Omniox已经完成了一系列支持这一提议的临床前研究： 1)OMX具有良好的生化特性，目前正在以高产量和 GLP毒理学和人类GB临床试验的纯度 2)OMX已安全地用于1000多只小鼠和大鼠，几只健康的Beagle狗， 全球缺氧性羔羊，以及来自我们的犬类临床试验的15名犬脑癌患者(表1) 3)OMX通过渗漏的血管渗出，并聚集在缺氧的肿瘤组织中，如图所示： A)使用免疫组织化学(IHC)(图1)和酶联免疫吸附试验(数据说明 B)荷瘤小鼠和大鼠使用89Zr标记的OMX的PET成像(图5)；以及c)在 犬脑癌患者肿瘤的免疫组化和酶联免疫吸附试验(图6) 4)OMX给药使肿瘤氧化，表现为：a)直接测量增加的 使用OxyliteTM探针插入低氧小鼠肿瘤组织的氧分压(PO2)(图3)。 2)；低氧调节的转录因子和下游信号的下调 IHC、FACS(数据未显示)和ELISA(图3)；以及c)减少大鼠颅内肿瘤的缺氧。 通过18F-FMISO信号，一种PET低氧示踪剂，在试点成像研究中测量(图5)。 5)荷有皮下肿瘤的小鼠在放疗前接受OMX治疗后，肿瘤明显延迟 与仅接受RT治疗的小鼠相比，生长速度和治愈率均为50%(图4) 6)CerePET可用于对大鼠和犬脑进行高分辨率成像(图7) 在这项建议中，我们提出了一种创新的翻译策略来开发和临床前测试PET 可潜在用作候选药物预测标记物的影像伴随诊断 在未来的人体临床试验中的有效性。我们将在现有的颅内大鼠和小鼠实验结果的基础上再接再厉 肿瘤模型和犬脑胶质瘤患者优化和验证非侵入性临床相关成像 如果在人类临床研究中得到证实，这种方法将能够选择目标患者群体，其 通过OMX介导的肿瘤氧合，RT治疗可能会得到实质性的加强。

1. Project Summary/Abstract Hypoxia (low oxygen) is a known key driver of tumor aggressiveness and tumor resistance to radiation treatment (RT) in a variety of solid cancers. Given that about 50% of cancer patients receive RT (~800,000 cases per year in the US), Omniox has developed a novel oxygen carrier protein, OMX, which is engineered to specifically deliver oxygen to low oxygen regions of tumors and enhance RT efficacy. Omniox is supported by NCI SBIR (Phase I, II, IIb) and Wellcome Trust translational awards to test its lead drug candidate as an adjuvant to standard-of-care RT for the treatment of newly diagnosed glioblastoma (GB) patients. Omniox has received FDA pre-IND support for, and is currently conducting, IND-enabling manufacturing and GLP toxicology studies for the first-in-human clinical trial scheduled at the University of California, San Francisco (UCSF) for 2017. Importantly, non-invasive PET imaging with hypoxia-selective agents has shown that hypoxic volume in GB tumors is strongly associated with rapid disease progression and poor survival1 due to its blunting effects on standard-of-care RT. Previous efforts to oxygenate tumors in human clinical trials have shown ambiguous results due in part to heterogeneity in tumor hypoxic burden and the lack of targeted approach for patient selection. Therefore, the goal of this proposal is to develop predictive non-invasive PET imaging biomarkers that will identify the GB patient population that can substantially benefit from OMX therapy. Omniox and the UC Davis School of Veterinary Medicine completed a Phase 0 canine clinical trial to evaluate the safety, pharmacokinetics (PK), and pharmacodynamics (PD) of OMX in canine patients presenting with brain tumors. As part of the trial, Omniox partnered with Brain Biosciences to use their PET scanner designed specifically for brain imaging (CerePET™). Omniox and the key collaborators on this AIP grant, Drs. Simon Cherry (Aim 1) and Allison Zwingenberger (Aim 2) at UC Davis, and David Beylin of Brain Biosciences (Aims 1 and 2) are developing protocols for PET imaging of canine brain patients, with successful initial scans already generated. In this proposal, we will build on our positive preliminary results in rodents and canines to evaluate imaging of tumor hypoxia and OMX tumor accumulation as predictive biomarkers of OMX activity in an intracranial rat GB model and in canine glioma patients. These data will then be used to inform an IND application and a human biomarker clinical trial. Developed from the thermostable H-NOX protein family discovered at UC Berkeley, OMX is a trimerized, PEGylated H-NOX oxygen (O2)-binding variant engineered to diffuse deep into hypoxic tumor tissue commonly associated with leaky blood vessels. The H-NOX moiety has been engineered to have a high affinity for O2 whereby it retains O2 in normoxic tissue and specifically releases it in regions of severe hypoxia. Omniox has completed a series of preclinical studies that support this proposal: 1) OMX is biochemically well-characterized and is currently being manufactured at high yields and purity for GLP toxicology and human GB clinical trials 2) OMX has been safely administered to over a thousand mice and rats, several healthy Beagle dogs, globally hypoxic lambs, and fifteen canine brain cancer patients from our canine clinical trial (Table 1) 3) OMX extravasates across the leaky vasculature and accumulates in hypoxic tumor tissue as seen in: a) mouse and rat intracranial tumors using immunohistochemistry (IHC) (Fig. 1) and ELISA (data not shown); b) tumor bearing mice and rats using PET imaging of 89Zr-labeled OMX (Fig. 5); and c) in tumors of canine brain cancer patients as seen by IHC and ELISA (Fig. 6) 4) OMX administration oxygenates tumors as demonstrated by: a) the direct measurement of increased oxygen partial pressure (pO2 ) using an OxyliteTM probe inserted into hypoxic mouse tumor tissue (Fig. 2); downregulation of hypoxia-regulated transcription factors and downstream signaling as measured by IHC, FACS,(data not shown), and ELISA (Fig. 3); and c) hypoxia reduction in rat intracranial tumors as measured by 18F-FMISO signal, a PET hypoxia tracer, in a pilot imaging study (Fig. 5). 5) Mice bearing subcutaneous tumors treated with OMX prior to RT showed significant delays in tumor growth and a 50% cure rate compared to mice treated with RT alone (Fig. 4) 6) CerePET can be used to image rat and canine brains with high resolution (Fig. 7) In this proposal, we present an innovative translational strategy to develop and preclinically test a PET imaging companion diagnostic that can potentially be used as a predictive marker of drug candidate efficacy in future human clinical trials. We will build on our existing results in intracranial rat and mouse tumor models and canine glioma patients to optimize and validate a non-invasive clinically relevant imaging method that, if confirmed in human clinical studies, will enable selection of a targeted patient population whose RT treatment is likely to be substantially enhanced by OMX-mediated tumor oxygenation.