Sergio Bracarda， MD

意大利托斯卡纳肿瘤研究所

　　肾癌的治疗历史正在改写。几年前，肾癌的免疫治疗还是干扰素和白细胞介素2治疗方案，肾癌患者的存活率非常低，最长的生存期只有12～14个月。自酪氨酸激酶抑制剂（TKI）和mTOR抑制剂问世后，现在可获得70％的疾病控制率，患者的预期寿命由12个月提高到36个月。Bracarda博士指出，目前肾癌的一线、二线甚至三线治疗均有可以选择的药物，不足的是如何在合适的时机为患者选择适合的药物，以实现肾癌的个体化治疗。

　　预测性生物标志物的可能是实现个体化治疗的重要因素，在合适的时机为正确的患者选择正确的药物，使疗效最大化，进而延长患者预期的生命。通过借鉴乳腺癌（雌激素受体和HER-2状态）、黑色素瘤（BRAF突变）、结肠直肠癌和某些类型肺癌的治疗模式，已确定了一些预测因素。此外，纪念斯隆-凯特琳癌症中心（Memorial Sloan-Kettering Cancer Center， MSKCC）模型和国际转移性肾细胞癌联合数据库（International Metastatic Renal-Cell Carcinoma Database Consortium， IMDC）模型也提供了一些预后因素。目前，学者们正尝试研究出一种新的肾癌治疗决策的预算办法。

　　现在可以分析的主要是临床要素，如高血压。然而，并没有试验证据证实高血压可作为一个预测因素，而只是一种预后因素。另一个要素是循环中的糖蛋白。在一项相前瞻性随机对照研究（RECORD-3）中，应用了28项生物标记物分析，并加入到一系列的指数，以帮助患者选择合适的靶向药物。

　　此外，也有来自DNA和RNA的元素，主要是单核苷酸多态性（SNP），其可以对两种药物在具体患者治疗中的疗效进行比较。例如，目前已发现一个与白细胞介素-8相关的阴性预测因素。对于mRNA，有一些可以控制肿瘤细胞基因组特征的转录元件，这些基因组特征将使得治疗更加个体化。目前，可获得的信息还非常少，因为数据来自于一个单一的小样本，但是当添加了来自于mRNA和DNA的突变状态的后续对照后，就可以获得一个复杂的系统，该系统可以允许部分特定组织学类型的患者个体化治疗，也可实现单个患者的个体化治疗。

　　在采访最后，Bracarda博士也表达了对肾癌个体化治疗的期望。希望未来能够发现更多的预测标记物，从患者的角度，综合考虑疗效、耐受性等因素，在合适的时机为患者选择合适的药物，提高生活质量，延长生存期，实现肾癌的个体化治疗。

　　Oncology Frontier: Your presentation concerned the personalization of targeted therapy. Is this a future possibility?

　　Dr. Bracarda: What we have now are a number of agents that allow us to modify the history of kidney cancer. Just a few years ago， we had only the older immunotherapy options (interferon and interleukin-2) and the survival rates of patients with kidney cancer were very low with a maximum life expectancy of 12-14 months. With the TKIs and mTOR inhibitors， we can now achieve disease control in 70% of patients for a significant number of months and the life expectancy has improved from 12 months to 36 months. We have agents released and authorized as first-line options and after progression we can utilize other agents in the second-line and still others for third-line therapy. What is lacking now is how to select these agents for the individual patient.

　　Learning from the model of breast cancer (estrogen receptors and HER-2 status) and more recently， melanoma (BRAF mutation)， colorectal and some types of lung cancer， some predictive factors have been identified. We have been studying a new algorithm for decision-making in kidney cancers. We have some prognostic classifications from the old model from the Memorial Sloan-Kettering prognostic score and the International Consortium score， but these are independent of the treatment that we have. The difference in the predictive biomarkers is the evaluation of an element that might drive the selection of the right drug at the right moment for the right patient. This is a way to identify some elements that may be important in determining the drug that is more efficacious in a single patient and hopefully translates into a longer life expectancy for that patient.

　　The elements that we can analyze now are clinical factors like hypertension. There is however no trial evidence for hypertension being a predictive factor， only a prognostic one. Another element is circulating glycoproteins， and we have a model derived from the biological studies associated with the prospective randomized study， RECORD-3， in which a group of 28 biomarkers were analyzed and added to a pooled index which may help to identify agents (which in the case of the trial were everolimus and sunitinib) for a group of patients. This will be an important source of information for the near future.

　　There are elements derived from DNA and RNA mainly from single nucleotide polymorphisms (SNPs) that may identify one agent ahead of another. For example， at the moment we have a negative predictive factor associated with interleukin-8. For mRNA， there are transcriptional elements that may control the genomic characteristics of the tumor cell， which will allow treatment to be more personalized. We currently have very little information as the data is derived from a single， small sample size trial， but when we add the epigenetic control that can be derived from mRNA and DNA mutation status， we can end up with a complex system that may allow personalization for a group of histological types of patients and maybe for the single patient.