Chopra Arvind
National Center for Biotechnology Information, NLM, Bethesda, MD 20894
The human epidermal growth factor receptor-2 (HER2, ErbB2) modulates its activity through a tyrosine kinase (TK) signaling pathway and is involved in the development of a variety of cancers (1, 2). Overexpression or amplification of the HER2 gene occurs in many cancer types (e.g., 20% of breast cancer (BC)) and indicates a poor prognosis for the patient. Invasive methods such as biopsies, in conjunction with immunohistochemistry and fluorescence hybridization, are often used to assess the HER2 status of the primary and metastasized neoplastic tumors; however, because of sampling bias and tumor heterogeneity, results obtained with these procedures are not completely reliable (2). In the clinic, F-labeled fluorodeoxyglucose is commonly used with positron emission tomography (PET) to detect and determine the tumor burden of a patient, but this imaging agent cannot distinguish between benign and malignant lesions, does not differentiate tumors that overexpress HER2 from those that have a low or no expression of the receptor, and often identifies inflammation as a false-positive neoplasm (1). Therefore, much effort has been devoted to produce and evaluate new imaging probes, such as radiolabeled anti-HER2 antibodies (3) and Affibodies (4), that can be used with single-photon emission computed tomography and PET to detect and diagnose cancers that overexpress HER2. An Affibody molecule is a chain of 58 amino acids (6.5 kDa) that contains a modified B domain of the staphylococcal protein A and can be obtained either chemical synthesis or produced in bacteria with recombinant DNA technology (5). The Affibody scaffold consists of 3-helix peptide chains: helix chains 1 and 2 are each composed of thirteen randomized amino acids and contain the receptor-binding moieties of the Affibody (6), and the third helix chain functions as a stabilizer of the Affibody molecule (7). Affibody molecules have a high affinity and specificity of binding to the target, such as a receptor, and are considered to be extremely suitable for the noninvasive imaging of solid tumors (8). The radionuclide-labeled Affibody Z (and its derivatives), directed against the HER2 (Z = 22 pM for HER2), has been used successfully with molecular imaging techniques to screen for BC patients who can probably benefit most from treatment with trastuzumab (a monoclonal antibody that targets HER2) or lapatinib (a small molecule drug that inhibits the TK activity of the HER2 signaling pathway) (8, 9). To further improve the imaging properties of radiolabeled Affibodies, it has been hypothesized that reducing the size of the molecule to a 2-helix structure will probably facilitate rapid clearance of the tracer from circulation, allow high penetration into solid tumors, ameliorate the uptake of label in non-targeted organs (such as the liver and lungs), and generate increased tumor/background ratios compared with the 3-helix molecule (10). To test this hypothesis, the third (stabilizing) helix chain was removed from the Z molecule; it was determined that the new 2-helix Affibody (MUT-DS; ~4.6 kDa) had a binding affinity of 5 nM for HER2 (10). In another study, Ga-labeled MUT-DS ([Ga]-DOTA-MUT-DS) was evaluated for the detection of human ovarian carcinoma SKOV3 cell xenograft tumors, which overexpress HER2, in nude mice (11). Although the radioactivity from [Ga]-DOTA-MUT-DS was rapidly cleared from circulation and had a high binding specificity for the tumors, the uptake of label in the neoplastic lesions was considerably lower (4.12 ± 0.83% of injected dose per gram tissue (ID/g) at 2 h postinjection (p.i.)) (11) than the accumulation with [Ga]-ABY-002, another Ga-labeled HER2-binding 3-helix Affibody (12.4 ± 3.8% at 2 h p.i.) (12). This indicated that there was no particular advantage in using a smaller Affibody compared with the larger parent 3-helix molecule to detect tumors that overexpress HER2. The comparison between the two Affibody types was considered to be inaccurate because the data obtained with the two labeled Affibodies were generated in two separate and independent studies (6). In addition, the two SKOV3 cell subclones used to grow the tumors in mice in the two studies may not have produced identical lesions in the animals, and the same amounts of HER2 may not have been expressed in the neoplasms (6). Therefore, to ascertain the similarities or dissimilarities between MUT-DS and ABY-002 Affibodies, it was necessary to investigate the biological characteristics of the 2-helix and 3-helix Affibodies simultaneously using identical handling techniques and by performing similar and studies with the two Affibodies (6). The receptor targeting and biodistribution characteristics of In-labeled MUT-DS (denoted as [In]PEP09239) and In-labeled ABY-002 ([In]ABY-002) were investigated by Rosik et al. in a side-by-side study with mice bearing xenograft tumors (6). It was observed that although the accumulation of label from [In]ABY-002 in the tumors was higher than with [In]PEP09239, the tumor/blood ratio was higher with the latter tracer. This indicated that both Affibody types are suitable for the imaging of tumors that expressed HER2. However, these probes can be easily degraded by circulating exoproteases because they are unprotected at the terminals, and, due to the presence of disulfide bonds in their structures, only a limited number of methods can be used to radiolabel either Affibody (7). In a continuing effort to develop a smaller 2-helix scaffold, it was shown that a native chemical ligation technique can be used to produce such a molecule (denoted as Z) from the parental 2-helix Affibody (13). The characteristics of Z and the advantages of generating a molecular imaging probe based on the Z platform are discussed elsewhere (13). Hence, a backbone-cyclized version of the HER2-binding Z Affibody molecule, designated Z, was generated with the native chemical ligation technique and evaluated for its thermal stability, affinity, and selectivity for HER2 (7). Subsequently a DOTA-Z conjugate was prepared for labeling with In and Ga, and the tumor-targeting and biodistribution characteristics of [In]Z and [Ga]Z were compared with those of [In]PEP09239 and [Ga]PEP09239 (7).
人表皮生长因子受体2(HER2,ErbB2)通过酪氨酸激酶(TK)信号通路调节其活性,并参与多种癌症的发生发展(1,2)。HER2基因的过表达或扩增在许多癌症类型中都有发生(例如,约20%的乳腺癌(BC)),这表明患者预后不良。活检等侵入性方法,结合免疫组织化学和荧光杂交,常用于评估原发性和转移性肿瘤的HER2状态;然而,由于取样偏差和肿瘤异质性,这些方法获得的结果并不完全可靠(2)。在临床上,F标记的氟脱氧葡萄糖通常与正电子发射断层扫描(PET)一起用于检测和确定患者的肿瘤负荷,但这种成像剂无法区分良性和恶性病变,无法区分过表达HER2的肿瘤与低表达或不表达该受体的肿瘤,并且常常将炎症识别为假阳性肿瘤(1)。因此,人们致力于研发和评估新的成像探针,如放射性标记的抗HER2抗体(3)和亲和体(4),它们可与单光子发射计算机断层扫描和PET一起用于检测和诊断过表达HER2的癌症。亲和体分子是由58个氨基酸组成的链(约6.5 kDa),包含葡萄球菌蛋白A的修饰B结构域,可通过化学合成或利用重组DNA技术在细菌中生产获得(5)。亲和体支架由3条螺旋肽链组成:螺旋链1和2各由13个随机氨基酸组成,包含亲和体的受体结合部分(6),第三条螺旋链作为亲和体分子的稳定剂(7)。亲和体分子对靶点(如受体)具有高亲和力和特异性,被认为非常适合实体瘤的无创成像(8)。放射性核素标记的针对HER2的亲和体Z(及其衍生物,HER2的Z值为22 pM)已成功用于分子成像技术,以筛选可能从曲妥珠单抗(一种靶向HER2的单克隆抗体)或拉帕替尼(一种抑制HER2信号通路TK活性的小分子药物)治疗中获益最大的BC患者(8,9)。为了进一步改善放射性标记亲和体的成像特性,有人推测将分子大小减小到2螺旋结构可能有助于示踪剂从循环中快速清除,使示踪剂能够高渗透进入实体瘤,减少在非靶向器官(如肝脏和肺)中的摄取,并与3螺旋分子相比提高肿瘤/背景比值((10)。为了验证这一假设,从Z分子中去除了第三条(稳定)螺旋链;结果确定新的2螺旋亲和体(MUT-DS;约4.6 kDa)对HER2的结合亲和力为5 nM(10)。在另一项研究中(11),评估了Ga标记的MUT-DS([Ga]-DOTA-MUT-DS)在裸鼠中检测过表达HER2的人卵巢癌SKOV3细胞异种移植瘤的情况。尽管[Ga]-DOTA-MUT-DS的放射性从循环中快速清除,并且对肿瘤具有高结合特异性,但肿瘤病变中的摄取量明显较低(注射后2小时(p.i.)为每克组织注射剂量的4.12±0.83%(ID/g))(11),低于另一种Ga标记的HER2结合3螺旋亲和体[Ga]-ABY-002(注射后2小时为12.4±3.8%)((12)。这表明与较大的亲本3螺旋分子相比,使用较小的亲和体检测过表达HER2的肿瘤并没有特别优势。两种亲和体类型之间的比较被认为不准确,因为两种标记亲和体获得的数据是在两项独立的研究中产生的(6)。此外,在两项研究中用于在小鼠体内生长肿瘤的两个SKOV3细胞亚克隆可能在动物体内产生的病变并不相同,并且肿瘤中HER2的表达量可能也不一样(6)。因此,为了确定MUT-DS和ABY-002亲和体之间相同或不同之处,有必要使用相同的处理技术并对这两种亲和体进行类似的研究,同时研究2螺旋和3螺旋亲和体的生物学特性(6)。Rosik等人通过对携带异种移植瘤的小鼠进行并行研究,研究了In标记的MUT-DS(表示为[In]PEP09239)和In标记的ABY-002([In]ABY-002)的受体靶向和生物分布特性(6)。观察到尽管[In]ABY-002在肿瘤中的摄取高于[In]PEP09239,但后者示踪剂的肿瘤/血液比值更高。这表明两种亲和体类型都适用于对表达HER2的肿瘤进行成像。然而,这些探针在循环中容易被外切蛋白酶降解,因为它们在末端没有保护结构,并且由于其结构中存在二硫键,只能使用有限的方法对亲和体进行放射性标记(7)。为了继续研发更小的2螺旋支架,研究表明可以使用天然化学连接技术从亲本2螺旋亲和体中生产出这样的分子(表示为Z)(13)。Z的特性以及基于Z平台生成分子成像探针的优势在其他地方进行了讨论(13)。因此,使用天然化学连接技术生成了HER2结合Z亲和体分子的主链环化版本,命名为Z,并评估了其对HER2的热稳定性、亲和力和选择性(7)。随后制备了用于In和Ga标记的DOTA-Z缀合物,并将[In]Z和[Ga]Z的肿瘤靶向和生物分布特性与[In]PEP09239和[Ga]PEP09239进行了比较(7)。