获取胚系和体细胞性肿瘤基因组数据对于肿瘤个体化治疗有何意义？Mcleod教授在《肿瘤瞭望》的采访中指出，患者每个DNA都会对患者特征有所影响，比如影响头发和眼睛的颜色。细胞内的DNA对患者是否受到药品毒性影响也很重要。癌症治疗的药物可能对患者造成损害，因为DNA发生了变化。当癌症形成时，DNA的变化可否用于识别癌症？或者是否有望确定哪些药物可使患者获益？DNA的变化是可以衡量的，不仅有助于识别患者自身情况，也有望使癌症治疗药物更加安全。

　　如何根据体细胞变异的信息做出临床决策，要取决于治疗目标。如果目标只是挑选在缩小肿瘤方面效果最好的药物，不考虑药物毒性，那就只关注肿瘤的DNA。如果想平衡药物疗效和不良反应，则需要关注两方面。目前一种肿瘤往往有多种药物选择，当只有一种药物时，我们只需要考虑药物疗效，有多种有效药物时，需要根据多种可靠的数据做出选择。

　　基因检测技术的价格大幅降低，将成为临床可用的技术。如何根据这种技术所获得的大量基因信息做出更好的临床决策？Mcleod教授指出：我们能获取大量的数据，但并不一定意味着这些数据可转化为临床决策。即使对患者实施全基因组测序，也往往只关注几千个基因组，因为这些基因信息可影响临床决策。对大数据的进行分析，找出哪些基因变异对药物毒性和药物疗效有影响，哪些基因变异无关重要应该忽略。

　　在癌症管理方面，目前重要的是从各种治疗方案中做出选择。过去我们通常选择熟悉的药物、临床可获得的药物。关注的重点不是患者获益可能性有多大。如果某疾病的疗效不够好，我们则希望能改善疗效。但是，今天获得的数据不是明天就能用得上，我们会设计临床数据库，记录患者的信息，并使这些信息最终使以后每一次治疗获益。所以目前无用的数据、花费的时间对以后的患者管理非常重要。

　　Howard L. Mcleod博士，药学博士，来自美国北卡罗来纳大学药学院个体化及药物基因组学研究所。长期从事肿瘤药物基因组学研究；过去20多年中，在世界顶级科学杂志上发表论文超过400篇，获得NIH基因支持超过2千万美元，主持、参与国家重大项目10余项，并在全球37个国家受邀做报告超过900场。

访谈原文：

　　Oncology Frontier: Thank you for your time and join with us for the interview. Can you talk about Germline and somatic data and explain what it means for individualized therapy?

　　Howard L. Mcleod: Yes， I think we all know that the patients have DNA， and each DNA has important for the outlook features， such as hair color， eye color. But the DNA inside each cell is important for whether someone get the toxicity from medicine. So um， the intended medicine may cause harm because of some DNA have changes. And when the cancer forms， further DNA changes occur， (that) can they be used as identity for cancer? Or， hopefully， also to identify which medicine that patient may benefit from. So the DNA changes could be measured， not only helps identify or what’s going on with the patient， but also hopefully help us to make it  safe ineffective medicine.

　　Oncology Frontier: So， some somatic variants are clinically significant and provide genetic information that cannot be obtained from the Germline data， is it possible to fully rely on one or the other when provide comprehensive treatment?

　　Howard L. Mcleod: It depends on your goal. If your goal is only pick which medicine will have good chance to shrink the tumor， without regard to the toxicity， they more can focus just on the tumor DNA. If you also want to balance the adverse events， then you need to look at both DNAs. Then the thing especially important now is that there are so many medicines to choose from. Back to when it is only one medicine， benefit is the only thing we could think about. But now there are multiple different medicines， it is a demon song; you have many different (medicines) to choose from. So you need to have many types of reliable data to make this dicision.

　　Oncology Frontier: So， pharmacogenetics is a field still in its infancy. Recently costs for whole genome sequencing have been reduced making it a more feasible treatment. How can we investigate the amount of genetic information we can now obtain and translate that into better care?

　　Howard L. Mcleod: well， you reached a very important issue， that we can generate a lot of data， but that doesn’t necessary mean it is easy to turn that into a clinical decision. So， even we sequence the entire genome of patient， we often just focus down on the few thousand locations of genome， where we know we can act on that information. So the important part of this analysis is having big data analysis basis， and analyze these large big data basis， in which to understand when is the genetic variants have action for toxicity， for efficacy， for some other important point; and when it is just a noise， that we just need to try to ignore in order to look forward.

　　Oncology Frontier: Where do you see this area of cancer management developing?

　　Howard L. Mcleod: Well， I think the important part what is happening now is choosing from the amount of therapeutic options. And that is important， because in the past we chose the medicine we are almost familiar with， or the ones that we think we can obtain for the patients. Whereas that hasn’t already focused on probability of patients or benefit from that medicine， and that is now changing with genetic information. We all know is that ten percent chance or nighty percent chance， is the big difference， obviously. Um， if it is ten percent， try to move to something with the higher chance of benefit. So if we give the data today， we don’t always use it tomorrow. And we need to be designing clinical databases， so that we will obtain the information from the patients. It benefits them every single time to getting care， so the both the useless data or time is something will be very important for helping management of patient more effectively.