IC: Given the role of penumbral imaging in the expansion of the time window for acute stroke therapy， do you think that the time window should be extended to more than 6 hours?
《国际循环》：根据半影成像在急性中风治疗时间窗延长中的作用，您认为时间窗应该被延长至6小时以上吗？
Obviously we want to try to successfully treat as many people as possible and minimize potential harm. With tPA， it was initially developed to be given very early， within 3 hours of stroke onset. The idea behind acute stroke therapy is to treat or preserve parts of the brain that are at risk of dying. On average， a smaller infarct means there should be less functional deficit in the patient and the tissue most at risk is the penumbra. If you treat early， most patients will have extensive amounts of penumbra.  As time elapses， there are certainly patients that will still have salvageable penumbra， but there are a lot of other patients that will not have penumbra. If you limit yourself to clinical criteria in these later time window trials， you are going to have a lot of uninformative patients that cannot respond to any treatment， because the whole ischemic lesion is already dead. Therefore it would be difficult to demonstrate that there is a clinical benefit for the patients enrolled in the trial. If we want to help those that can still benefit from treatment， then it makes sense to identify the target tissue of the treatment， which is the penumbra. That is the only way that I can see of extending the time window. This does not have to be six hours， you can treat people much longer. There are some patients that have extensive penumbra many hours after the stroke onset. However， by just using clinical criteria， you cannot identify them.
Right now， the EXTEND trail from Australia and the ECAS 4 trial from Europe are taking patients with substantial mismatch for up to 9 hours. Instead of basic treatment enrolment criteria， where decisions are based on time， you can hopefully prove penumbral imaging identifies patients who can respond for many hours after stroke onset. This is going to be a whole paradigm shift in how we approach treating people.
I personally think that the benefits of tPA in the 3 to 4.5 hour window are pretty modest. About a month ago， the FDA refused Genentech’s request to extend tPA’s time window of approval to after 4.5 hours in the United States. Apparently they did this because they felt that the data in ECAS 4 were not robust enough to persuade them to extend the time window. A study that I’d like to see done， and maybe the FDA’s lack of approval will move this forward， is to treat everybody in a 3-4.5 hour window， and obtain advanced imaging either just before treatment starts or while it is being infused. Then， they could look at the patterns on images to predict the responders and the non-responders. I am confident that we will be able to identify patterns in response.
Diffuse data from Gregory Albers has identified patterns that actually predict harm， such as increased risk of bleeding and poor outcomes.  My analogy for this is the beginning of the antibiotic era in the 1940s， when penicillin was first used. Back then， they were just treating everybody because they did not have other options. They assumed that many people would benefit， but over time， they realized that doing cultures and sensitivities made sense because then you could target your specific antibiotic for those who had responsive bacteria. In a gross analogy， advanced imaging is like microbiology for antibiotic therapy: we need to move it to the next phase where everybody agrees that the target of acute stroke therapy is to salvage some of the penumbra. If we have ways to approximately identify the penumbra， then why not try to identify patients who have a target tissue.
And for my last example， the June issue of Stoke featured an analysis of the desmoteplase trials .  They showed that in patients with no documented occlusion， on CT angiograms or MR angiograms， there was absolutely no benefit for the treatment of this other lytic. This illustrates that if you cannot observe a clot， then the lytic was not working because the patient either did not have a clot or it was so far distal that it had little benefit. This is where I see future progress， we went through an initial era of showing the we actually had a functional treatment and now we need to find ways to include more people and to maximize treatment benefits.
Dr. Fisher: 很明显我们希望治疗尽可能多的患者并同时把潜在的危险最小化。在使用tPA时，最初的设计是很早期就给药，即在发生中风后的3小时内给药。急性中风的治疗目的是挽救那些即将坏死的脑组织。平均来讲，更小面积的梗死意味着患者功能缺失的程度更少，并且更容易进展为坏死的组织大部分都在半暗带内。如果你及早的进行治疗，大部分患者的半暗带面积会明显增大。随着时间的延长，肯定还有患者存在可挽救的半暗带，但是很多其他患者的半暗带是消失的。如果你在这些晚时间窗试验中完全依赖于临床标准，你就会纳入很多资料不清并对任何治疗都没有反应的患者，因为这些患者的整个缺血病变部位已经全部死亡了。因此就很难在试验纳入患者中显示出治疗的临床获益。如果我们想帮助那些仍然可能从治疗中获益的患者，就必须明确治疗的靶组织，也就是半暗带。这是我认为能延长时间窗的唯一办法。这个时间限制并非必须是6小时，你也可以治疗超过这一时间的患者。有些患者在中风发生很多个小时后依然存在很大的半暗带。然而，你不能仅仅通过临床标准来确定这种情况。
现在，澳大利亚的EXTEND试验和欧洲的ECAS 4试验都纳入了最长时间为9小时的较大影像差异患者。这些决定基于时间而非基础治疗纳入标准，你能通过半暗带影像来识别那些在中风发生很多个小时后依然对治疗有反应的患者。这将是我们治疗患者方法的一种根本性改变。
我个人认为tPA的3至4.5小时有效时间窗是非常适当的。大约在一个月前，FDA拒绝了Genentech公司的请求，他们希望把tPA在美国应用的时间窗延长至4.5小时以后。很明显，FDA拒绝了这一请求是因为他们认为ECAS 4中的试验数据并不足以说服他们延长时间窗。我希望看到的试验是，也许缺少FDA的批准会促进这些试验的完善，使他们能在3-4.5小时的时间窗内治疗所有的患者，并获得在治疗前或在给药同时的进一步的影像资料。那时，他们就能通过影像的特征来预测对治疗有反应或没有反应的患者。我坚信我们将会明确那些对治疗有反应患者的特征。
Gregory Albers获得的弥散数据已经确定出了实际上能预测危害的特征，比如增加出血风险和不良后果的情况。我把这比喻为1940年的抗生素时代的开始时期，当青霉素首次被使用的时候。那时，青霉素被用来治疗每个患者，因为别无选择。人们假定很多人会从中获益，但是随着时间的流逝，人们认识到进行病原体培养和抗生素敏感性实验更合理，因为通过这些手段能靶向应用特殊的抗生素来治疗那些携带有反应的细菌的患者。从总体上来比喻，先进的成像就像微生物的抗生素治疗：我们需要跨越到下一阶段，在这一阶段所有的人都认为急性中风治疗的目标是半暗带的挽救。如果我们有粗略确定半暗带的方法，那我们为什么不试着确定哪些患者存在靶向组织呢？
最后我再举个例子，Stoke杂志的六月刊发表了一项去氨普酶的分析。研究表明在那些没有CT或MR血管造影显示存在明确闭塞的患者中，明显也是没有溶栓治疗获益的。这也表明如果你不能观察到血栓，那么溶栓就不会起效，因为患者可能没有血栓或血栓在血管远端，获益太小。这就是我认为将来的进展，我们经历的最初时代显示我们确实有很有效的治疗，而现在我们需要发现扩大受益患者的方法来最大化治疗的获益。

采访稿整理

延长r-tPA治疗时间窗时机尚未成熟

——麻省大学医学院Marc Fisher教授专访

延长时间窗，明确半影区

我们希望治疗尽可能多的患者并将潜在危险最小化。如果及早进行治疗，大部分患者的半影区面积会明显增大。随时间延长，肯定还有患者存在可挽救的半影带。如果我们想帮助那些仍然可能从治疗中获益的患者，就必须明确治疗的靶组织，即半影带，这是我认为能延长时间窗的唯一办法。这个时间限制并非必须是6小时。有些患者在卒中发生很多个小时后依然存在很大的半影区。

现在，澳大利亚的EXTEND试验和欧洲的ECAS 4试验都纳入了最长时间为9小时的较大影像差异患者。这些决定基于时间，而非基础治疗纳入标准，你能通过半影区影像来识别那些在卒中发生很多个小时后依然对治疗有反应的患者。这将是治疗方法的一种根本性改变。

扩大受益者范围，最大化治疗获益

对于r-tPA，我个人认为3~4.5小时的有效时间窗是非常恰当的。大约在1个月前，FDA拒绝了Genentech公司将tPA时间窗延长至4.5小时以上的请求，因为他们认为ECAS 4中的试验数据并不足以说服他们延长时间窗。

我们的工作需要跨越到下一阶段：确定哪些患者存在治疗靶组织。Stoke 6月刊发表的一项去氨普酶的分析表明，在CT或磁共振血管造影显示无明确血管闭塞的患者中，溶栓治疗是没有获益的。我们将来的研究方向应是发现扩大受益患者的方法，以使治疗获益最大化。

Obviously， we want to try to successfully treat as many people as possible and minimize potential harm. r-tPA was initially developed to be given very early， within 3 hours of stroke onset. The idea behind acute stroke therapy is to treat or preserve parts of the brain that are at risk of dying. On average， a smaller infarct means there should be less functional deficit in the patient and the tissue most at risk is the penumbra. If you treat early， most patients will have extensive amounts of penumbra. As time elapses， there are certainly patients that will still have salvageable penumbra， but there will also be some without any. If you limit yourself to clinical criteria in these later time window trials， you are going to have a lot of uninformative patients that cannot respond to any treatment， as the whole ischemic lesion is already dead. Therefore it would be difficult to demonstrate that there is a clinical benefit for the patients enrolled in the trial. If we want to help those that can still benefit from treatment， then it makes sense to identify the target tissue of the treatment， which is the penumbra. That is the only way that I can see of extending the time window. This does not have to be six hours: you can treat people much longer. There are some patients that have extensive penumbra many hours after the stroke onset. However， by just using clinical criteria， you cannot identify them.

Right now， the EXTEND trail from Australia and the ECAS 4 trial from Europe are taking patients with substantial mismatch for up to 9 hours. Instead of basic treatment enrolment criteria， where decisions are based on time， you can hopefully prove that penumbral imaging identifies patients who can respond for many hours after stroke onset. This is going to be a whole paradigm shift in how we approach treating people.

  I personally think that the benefits of r-tPA in the 3 to 4.5 hour window are pretty modest. About a month ago， the FDA refused Genentech’s request to extend r-tPA’s time window of approval to after 4.5 hours in the United States. Apparently they did this because they felt that the data in ECAS 4 were not robust enough to persuade them to extend the time window. A study that I’d like to see done， and maybe the FDA’s lack of approval will move this forward， is to treat everybody in a 3-4.5 hour window， and obtain advanced imaging either just before treatment starts or while it is being infused. Then， they could look at the patterns on images to predict the responders and the non-responders. I am confident that we will be able to identify patterns in response.

The next phase will be where everybody agrees that the target of acute stroke therapy is to salvage some of the penumbra. If we have ways to approximately identify the penumbra， then why not try to identify patients who have a target tissue. The June issue of Stoke featured an analysis of the desmoteplase trials. They showed that in patients with no documented occlusion， on CT angiograms or MR angiograms， there was absolutely no benefit for the treatment of this other lytic. This illustrates that if you cannot observe a clot， then the lytic was not working， because the patient either did not have a clot or it was so far distal that it had little benefit. This is where I see future progress， we went through an initial era of showing that we actually had a functional treatment and now we need to find ways to include more people and to maximize treatment benefits.