编者按：2019年9月6~8日，“第四届抗白血病·淋巴瘤国际高峰论坛暨CSCO抗白血病联盟&抗淋巴瘤联盟巡讲——哈尔滨站”隆重召开。本次会议设置了白血病及淋巴瘤两个主会场，另有淋巴瘤+多发性骨髓瘤专场、白血病+MDS专场、病理专场、儿科淋巴瘤专场、药监政策及青年专场，内容可谓极其全面。本刊记者在大会现场特邀采访了美国麻省总医院的癌症中心JeremyAbramson教授，针对CAR-T细胞疗法的未来与挑战等问题进行了精彩的分析。现整理如下，供读者参考。
《肿瘤瞭望》：近期，已经有2种CAR-T疗法被FDA批准用于复发难治NHL，展示出巨大前景。您能介绍一下，除了CD19，目前国际上还有哪些有潜力的CAR-T靶点未来能够脱颖而出吗？
Abramson教授：当前我们对于CAR-T细胞的认识，主要局限于CD19介导的CAR-T细胞在淋巴瘤中的作用，以及BCMA介导CAR-T细胞在多发性骨髓瘤中的作用，但这仅仅是茫茫“CAR-T细胞海洋”中的少数两种。现在我们正在努力发现更多可用于CAR-T治疗的靶点，这也是CAR-T细胞研究中最富有挑战性的工作，因为这些靶点对于肿瘤细胞应具有足够的特异性，以避免对于健康细胞的影响。就CD19而言，其CAR-T细胞对于B细胞淋巴瘤并没有完全的选择性，因为CD19几乎在所有成熟B细胞表面皆有表达，抗CD19的CAR-T细胞会消除体内几乎所有的B淋巴细胞。好在B细胞缺失在一段时间内不会有太大影响，且CAR-T治疗后9-12个月就可以恢复，期间还可以通过补充抗体来防止病人受到感染困扰。但总体上，通过抗CD19的CAR-T细胞治疗淋巴瘤，有“杀敌一千，自损八百”的特点。
基于这个问题，其他B细胞相关的CAR-T靶点也在研究当中。CD22是B细胞表面广泛表达的靶点，在淋巴瘤的免疫治疗领域被认为具有相当的潜能。而如同CD19，CD22当前也没有直接靶向的抗体药物面世，因而算是一种全新的CAR-T靶点。在一些小规模的研究中，抗CD22的CAR-T细胞显示出了极为显著的疗效。此外，CD19/CD22双特异性CAR-T细胞也得到了人们的广泛关注，已有临床研究开始考察其在淋巴瘤和急性淋巴细胞白血病中的效果，并发现双特异性CAR-T细胞较单特异性具有极大的优势。因此CD22和CD22/CD19双特异性CAR-T细胞都是淋巴瘤领域有潜质的发展方向。
此外，CD37是B细胞和T细胞共有的表面标记，因而其特异性CAR-T细胞可能对于B细胞和T细胞性淋巴瘤都有杀伤作用。还有CD30，此抗原在霍奇金淋巴瘤、EBV相关的大B细胞淋巴瘤、间变性大T细胞性淋巴瘤，以及其他B细胞分化相关的淋巴瘤中广泛表达，对于患有这类疾病的病患，抗CD30的CAR-T细胞无疑具有极大的治疗潜力，本妥昔单抗耦连药物（Brentuximabvedotin）治疗CD30阳性B/T细胞淋巴瘤的成功案例也提示了这一靶点的潜力。CD30介导的CAR-T细胞当前的研究中已被证明在治疗B细胞/T细胞淋巴瘤以及霍奇金淋巴瘤中的活性，但是其具体的疗效仍需要更大样本量临床试验确证。
在B细胞之外，我认为对于T细胞特异性的CAR-T细胞靶点的探索具有更为重要的意义。因为病患尚且可以在B细胞缺失的情况下生存一段时间，但T细胞缺失则是致命的，任何正常T细胞的“误伤”都会对病人造成严重影响。除此之外，CAR-T细胞的治疗领域仍在延伸，对于急性髓细胞白血病、急性淋巴细胞白血病在内的其他疾病CAR-T疗法都有触及。我认为在接下来的五到十年中，CAR-T疗法会呈现爆发式的发展趋势，不止新的靶点会不断涌现，而可治疗的疾病范围也会大大扩展。
IIt’s a great question. Currently， our armamentarium for CAR-T cells are dominated by CD19 directed CAR-T cells in lymphoma， and BCMA directed CAR-T cells in multiple myeloma， but these represent only two of potentially limitless targets to treat malignancies. Currently we’re looking at multiple other targets as well. Selecting a target is a critical element of CAR-T cell treatment. And in some ways， IT’s the most challenging. We have to select a target that is sufficiently specific for the cancer cell and won’t destroy a healthy cell that the patient requires to live healthy. So the anti-CD19 CAR-T cells are not entirely selective for B cell lymphoma. In fact， CD19 is expressed on all mature， healthy B lymphocytes as well. And so with anti CD19 CAR-T cells， we eliminate all healthy b lymphocytes in the human body of our patients. Now， the fact is that patients can live without their B cells for an extended period of time. And we see recovery of these cells happening usually nine to twelve months after their CAR-T cell treatment. And occasionally we can supplement antibodies to allow patients of fight off infection while their immune system is recovering， while their B cells recur， so the B cell， in that sense is a sacrificial target we can eliminate it for a period of time.
There are other B cell targets that are also under investigation. CD22， it was a target on the surface of the B lymphocyte is a very appealing target for immunotherapy， similar to CD19， CD22 is not yet been targeted with other agents in our patients likes a Rituximab has targeted CD20. And so it is a naïve antigen to direct against. And anti CD22 CAR-T cells have also shown significant evidence in small trials. In fact， what current clinical trials are evaluating in both lymphoma and in acute lymphoblastic leukemia is trying to target multiple antigens at the same time. And so early evidence suggests that not only targeting CD22， but also targeting CD19 at the same time might work better than either a single CD19 or a single CD22 directed CAR-T cell. So both CD22 CAR-T cells， and bi-specific CAR-T cells against multiple B cell antigens are under investigation.
We’re also looking at other lymphoid antigens. That includes CD37， which is expressed on a number of B cells and T cells. And so we’re going to be looking at this target in both B and T cell lymphoma that expresses CD37. We’re very interested in CD30. CD30 is widely expressed in classical hodgkin lymphoma， EBV-associated large B cell lymphomas， and on anaplastic large T cell lymphomas， as well as with other lymphomas with B cell differentiation， often times including things like primary affusion lymphoma. And so CD30 is an appealing target for those patients. And we previously have recorded of using a CD30 directed therapy， the antibody drug conjugate Brentuximab vedotin， and for the treatment of patients with CD30 positive B cell and T cell lymphomas. CD30 directed CARs are currently in development with early data demonstrating activity in both B cell lymphoma， T cell lymphoma， and hodgkin lymphoma， but those require much further investigation on a larger number of patients.
Beyond that， there are still additional B cell antigens that could be targeted. I think we have to continue identify additional T cell antigens. Unlike the B cell， which the human body can live without for a period of time， we can’t live without T lymphocytes due to the severe infectious risk. So finding a target among T cell lymphomas means finding one that won’t eliminate all healthy T cells in the body. Beyond lymphomas， there are CARs currently being directed for acute myelogenous leukemia， as well as newer drugs for acute lymphoblastic leukemia. So I think over the next five to ten years， I hope to see an explosion of CAR-T cells directed numerous antigens that broaden the availability of this technology to multiple diseases， which we can’t currently target with anti CD19 cells.
《肿瘤瞭望》：尽管CAR-T疗法已经显示了较好的缓解率，但是仍有一部分患者在CAR-T治疗后复发，目前的观点是，CAR-T疗法应该整合到淋巴瘤的整体治疗中。您能谈一下CAR-T治疗在患者整体治疗中的地位，以及您了解的，导致CAR-T耐药和复发的机制吗？
Abramson教授：CAR-T疗法的耐受现象是不得不提的话题。以弥漫性B细胞淋巴瘤为例，虽然绝大多数病患都对CAR-T疗法有响应，但仅有40%的患者获得了长期、持续的环节，而更多患者出现了耐受。
那么耐受原因是什么呢？肿瘤细胞CD19抗原的下调可能是导致耐受的原因之一，因为我们在抗体的治疗中已经广泛发现，肿瘤细胞抗原的下调现象，而在急性淋巴细胞白血病的CAR-T细胞疗法中，抗原下调也已经被证明是造成耐受的主要机制。但研究发现，弥漫性B细胞淋巴瘤对于CAR-T疗法的耐受案例中，仅有10%的患者丢失了CD19抗原，复发的患者中确实绝大多数CD19的表达是正常的。在这种情况下，一定是因为某些因素阻止了CAR-T细胞对于淋巴瘤细胞的进攻。
在我看来，耐受的机理是免疫相关的，肿瘤细胞产生了得以逃避免疫反应的变化，因此靶向免疫反应、重建CAR-T细胞功能是应对耐受行之有效的方法。免疫检查点抑制剂可能是解决这一问题的方式之一，因为它们可以在多种肿瘤中激活自身免疫系统以对抗肿瘤细胞。已有案例证明，免疫检查点抑制疗法恢复了耐受患者的中CAR-T细胞的功能，但多数情况下，单纯的使用免疫检查点抑制剂不能使肿瘤恢复对于CAR-T疗法的敏感性。解决这一问题的方法在与对于耐受机制有更深入的了解，例如对于病人肿瘤活检以确定PD-L1的表达水平，高表达的患者人群对于免疫检查点抑制疗法可能会做出更好的响应。
但免疫检查点不止PD-L1一种，我们也需要关注其他免疫检查点的过表达情况并做出正确的应对，在我们广泛的了解了这些检查点的过表达情况后，或许我们可以发现一种鸡尾酒疗法去广泛的发挥免疫重建的功能，这种疗法甚至可以做到针对不同病人的个体化用药。此外，一些其它的药物也可以恢复病人的免疫反应，例如免疫调节剂沙利度胺，其在滤泡性淋巴细胞瘤和其他淋巴瘤中的应用能够激活T细胞的免疫反应。对于沙利度胺在下调CAR-T细胞耗竭的分子标志、上调其活性方面的功能我们现在也有了一些初步证据。诸如BDK、IDK抑制剂对于重建CAR-T细胞的功能也有报导，依鲁替尼是BDK/IDK的抑制剂，在慢性淋巴细胞白血病的CAR-T治疗中，依鲁替尼可以下调CAR-T细胞耗竭的分子标志并上调其活性。
最后，我们也可以针对CAR-T细胞本身进行改造去解决耐受问题。当前我们所使用的CAR-T细胞属于第二代的技术，它们靶向CD19，且具有共激活的结构域，但是我们具有诸多的技术手段去改造CAR-T细胞，使其具有更好的活性。前面提到了多特异性的CAR-T细胞是发展的方向之一，通过使用更多的共激活结构域可以达到这一目的；同样我们可以使用例如CRISPR的基因编辑技术去删除表达免疫检查点的基因，去除CAR-T细胞耗竭的使动因素。我们还可以从健康的志愿者采集未产生耐药的CAR-T细胞，在删除T细胞受体和耐受相关的蛋白后，这些细胞对于细胞耗竭会有本质上的抵抗，且可以免除患者自身CAR-T细胞改造的较长周期。
因此，我认为想解决CAR-T疗法的耐受问题以及其中的免疫逃逸机制，针对CAR-T细胞本身的改造以及CAR-T疗法与其他免疫调节疗法的联用都有潜在的可能性，期待这些手段可以为病人带来福祉。
Resistance of CAR-T cell therapy is a critically important topic today. Clearly， we see the majority of patients with diffuse large B cell lymphoma will have a response to CAR-T cells， but that only about forty percent of them will have durable， long term remissions， so ultimately the majority of patients with diffuse large B cell lymphoma will be resistant to CAR-T cell therapy.
Why might that be? Well， one potential explanation is loss of the CD19 antigen. So we see this and other diseases targeted with monoclonal antibodies， for example， that the cell can down regulate the target antigen and thus be resistant to an immune directed target. That’s been a very common mechanism of resistance to CAR-T cells in acute lymphoblastic leukemia.
Interestingly， it appears to be less common in diffuse large B cell lymphoma， where only about 10% of patients who are resistant to CAR-T cells have lost the CD19 antigen. In fact， if you look at patients relapsing on CAR-T cells， we find that the majority of them still express the CD19 antigen and still have the CAR-T cell detected in their blood. Why isn’t it working? Well， if the target is there and the cell is there， then something is preventing it from attacking the lymphoma cell.
And so it seems to me that the primary mechanism of resistance is immune mediated， that the cells are finding a way to elude the immune response. And so targeting the immune response and restoring activity of the CAR-T cell is likely the way we’re going to overcome most mechanisms of resistance. How might we do that? Well， we have immune checkpoint inhibitors， which have shown the ability in lymphomas and multiple other cancers to restore an effective host immune attack against cancer cells. We’ve seen some patients can respond to immune checkpoint inhibitors and reactivate their CAR-T cells. However， most patients just treated with an immune checkpoint inhibitor alone will not restore sensitivity to CAR-T cells. I think we need to understand that better， and test patients biopsies for PD-L1 expression. And I think we’re likely to find a subset of patients with resistant therapy on CAR-T cells who up-regulated PD-L1 and are a good candidate for a PD-L1 targeted antibody， an immune checkpoint inhibitor.
But there are multiple other immune checkpoints beyond PD-L1. So we also have to look at other immune checkpoints， and targeting， and they’re over a dozen different immune checkpoints. And if we survey every patient for the increasing levels of expression of these checkpoints， we may be able to identify a cocktail of inhibitors that will help restore immune function for our patients. And it might be personalized to it by patient level. We can also look at other drugs that might help improve the immune response. We have immune-modulating drugs， such as Thalidomide. Thalidomide we know has been used in follicular lymphoma and other lymphomas and works in part by helping restore T cell immune response against lymphoma. We have early evidence in multiple different lymphomas that giving Thalidomide can down-regulate exhaustion markers in the CAR-T cell， up-regulate activity of the CAR-T cell and may restore effectiveness of that CAR-T cell.
We can then look at other drugs such as BDK and IDK inhibitors， and studies in chronic lymphoblastic leukemia have shown that patients treated with a BDK or IDK inhibitor called ibrutinib have shown down regulation of exhaustion markers， and also increased activation markers with better outcomes for those patients after treated with CAR-T cells.
And then finally， we can make better CAR-T cells. So currently， our CAR-T cells are what we call second generation. They target CD19， and they have a co-simulation domain. But are there other mechanisms we can use to make this CAR-T cell work better? Well， there are multiple potential options. We can target the CAR-T cell against multiple cancer antigens at once， and thus make it more powerful. We can use more powerful or additional co-stimulation domains. We can use CRISPR or gene editing technology to delete immune checkpoints and thus prevent the CAR-T cell from becoming exhausted in the first place， and potentially even use allogenetic healthy CAR-T cell donors， and create T cells from healthy donors that have not been previously exposed to chemotherapy. And again， use gene editing technology to delete the T cell receptor and immunogenic proteins and use effectively now and off the shelf CAR-T cell product， which is resistant to immune exhaustion to treat a patient without the necessary turnaround time required to create an autologous CAR-T cell product.
So I think both thinking about the CAR-T cell trial designs the CAR-T cell products themselves and using combination strategies with a different additional treatments be the immune-therapy immuno-modulating agents or small molecule inhibitors will hopefully help us overcome the primary mechanism of immune evasion on T cell exhaustion， for CAR-T cell patients.
京公网安备 11010502033352号