医药魔方 / 最新资讯 / 正文

Nature热议:CAR-T疗法淋巴瘤“大战”,谁将是赢家?

曼话 曼话 来源:医药魔方
2019-03-15
Nature 系列 CAR-T
原文

随着FDA批准了两款靶向CD19的CAR-T细胞产品——Yescarta和Kymriah——作为治疗复发和/或难治性大B细胞淋巴瘤的三线或三线以上疗法,这类癌症进入了激动人心的靶向抗癌新时代。

 

去年12月,有关Yescarta的关键临床试验ZUMA-1的两年数据在The Lancet Oncology杂志上公布[1];而有关Kymriah的关键临床试验JULIET的初步报告也于今年1月在NEJM上披露[2]。


图片来源:The Lancet Oncology

 

ZUMA-1于2015年5月19日至2016年9月15日招募了119名患者,其中108名患者接受了Yescarta回输治疗。截至2018年8月11日,对II期试验中可接受评估的101名患者进行了中位27.1个月的随访。客观缓解率为83%,其中CR为58%。中位响应持久时间为11.1个月。中位总生存期未达到,中位无进展生存期为5·9个月。在I期和II期108名可评估的患者中,48%的人发生了≥3级的严重不良事件:11%的患者发生了≥3级的细胞因子释放综合征(CRS),32%的患者发生了≥3级的神经毒性。除了先前报道的2例,之后的随访中没有发生新的与治疗相关的死亡。

 

图片来源:NEJM

 

JULIE试验中,共93名患者接受了Kymriah回输治疗,并纳入疗效评估。从回输治疗到数据截至的中位时间为14个月。总响应率为52%,其中40%为CR,12%为PR。不同预后亚组的响应率是一致的。距初始响应12个月后,无复发生存率估计为65%,这其中有79%的患者为CR。最常见的3级或4级特殊不良事件包括CRS(22%)、神经毒性(12%)、持续28天以上的全血细胞减少(32%)、感染(20%)和发热性中性粒细胞减少症(14%)。3名患者在回输后30天内死于疾病进展。没有死亡是由Kymriah、CRS或脑水肿引起的。

 

图片来源:Nature Reviews Clinical Oncology

 

2月27日,针对以上数据,美国华盛顿大学的一位科学家在Nature Reviews Clinical Oncology上发表一篇题为“Anti-CD19CAR T cell therapy for lymphoma — off to the races!”的观点性文章,对这两个CAR-T产品进行了详细的对比分析。

 

文章称,这两个多中心、单臂试验均显示,接受CD19-CAR-T疗法的患者缓解率较高,且生存期超过了“历史纪录”(该记录为中位生存期6.2个月,来自对难治性弥漫性大B细胞淋巴瘤患者的回顾性SCHOLAR-1分析)。


两项试验都招募了漫性大B细胞淋巴瘤(DLBCL)患者以及从滤泡性淋巴瘤或其它高等级B细胞淋巴瘤转化为DLBCL的患者。不过,ZUMA-1还招募了原发性纵隔B细胞淋巴瘤患者,这类患者JULIET未涉及。

 

尽管有很多相似处,但临床试验中的具体操作以及两个CAR-T细胞产品之间的差异妨碍了数据之间的直接对比。

 

治疗方法对比

 

这两个产品都由经白细胞除去法采集而来的自体CD3+ T细胞生成,Yescarta未经冷冻,运输给制造商;而Kymriah冷冻保存后,运输给制造商。


Yescarta使用的是逆转录病毒载体,Kymriah使用的是慢病毒载体。


Yescarta的给药剂量为2×106cells/kg,而Kymriah的给药剂量范围为0.1–6.0×108 cells(中位3.0 × 108 cells)。

 

CAR-T细胞通常在化疗后被回输到患者体内。化疗的作用是清除竞争性的内源性T细胞,增加CAR-T细胞的移植成活率;同时,也减少或延缓了机体对CAR-T细胞的免疫排斥反应。

 

在ZUMA-1中,化疗剂量更高:每天500mg/m2环磷酰胺和30 mg/m2氟达拉滨,持续3天;在JULIET中,化疗剂量为:每天250 mg/m2环磷酰胺和25 mg/m2氟达拉滨,持续3天。此外,在JULIET中,出现严重细胞减少症的患者(7%)省略了化疗,部分患者接受了苯达莫司汀治疗。化疗的差异可能影响早期总响应率,但不太可能影响长期的完全缓解率。


CAR-T细胞的设计(图片来源:Nature Reviews ClinicalOncology)[3]

 

安全性对比

 

Yescarta和Kymriah的CAR结构包含了相同的CD19靶向域(CD19-targeting domain)以及CD3ζ信号域,但共刺激域不同,前者是CD28,后者是4-1BB。与4-1BB共刺激域相比,CD28共刺激域与更早出现T细胞扩增高峰有关,这可能解释了为什么Yescarta治疗会更早出现毒性。在ZUMA-1中,CRS(包含任何级别)的中位发生时间为2天,而在JULIET中,对应的时间为3天。Yescarta治疗引发的较早发生的毒性需要住院治疗。


CRS与细胞增殖和细胞因子的产生有关,部分可通过单独使用托珠单抗(tocilizumab)或皮质类固醇阻断IL-6受体来控制。

 

神经毒性是CD19-CAR-T疗法另一种有时会危机生命,但未能被很好理解的不良反应,通常发生在CRS发生之后,也用皮质类固醇进行控制。

 

由于ZUMA-1和JULIET中使用了不同的分级系统,因此,不太好比较这两个试验中观察到的毒副作用。

 

在ZUMA-1中,根据Lee等[4]提出的分级标准的修订版,≥3级的CRS发生在11%的患者中,≥3级的神经毒性发生在32%的患者中。

 

在JULIET中,研究者们使用了宾夕法尼亚大学的分级量表,22%的患者出现了≥3级的CRS,12%的患者出现了≥3级的神经毒性。如果根据Lee等提出的分级标准对JULIET中的毒性作用进行重新分类,那么,≥3级的CRS的发生率有所降低,为17%。

 

尽管严重CRS发生频率相似,但与Kymriah相比,Yescarta与较高的CRS总体发生率和更频繁地使用托珠单抗有关。

 

图片来源:Nature Reviews Clinical Oncology

 

疗效对比

 

在疗效方面,预后不良的患者中,Yescarta和Kymriah分别与58%和40%的完全缓解有关。客观缓解可发生在治疗4周后。之后的几个月里,许多患者病情会复发;不过,未复发的患者,响应会在随后的几个月继续增强。在3-6个月时获得完全缓解的患者复发的风险较低。

ZUMA-1显示,近40%接受Yescarta治疗的患者在中位随访2年后依然处于缓解状态,很少出现晚期复发,这表明这些患者中有许多可能被治愈了。

 

在ZUMA-1试验中,接受CAR-T细胞回输的患者,中位总生存期>2年;在JULIET试验中,对应的数据为~1年。

用Yescarta治疗时,CAR-T细胞在前28天的扩增(血液中的峰值浓度)以及总暴露(累积浓度)与持续缓解有关。但用Kymriah治疗时没有观察到这种关联,同时,也不存在剂量与响应之间的关联。

 

比较这些疗效数据表明,Yescarta比Kymriah更加有效,但两者的临床试验设计和资格标准(eligibility criteria)使得认定谁更好变得更加复杂。


Apheresis:一种技术,患者的血液通过一个设备,分离出一种特殊的成分,剩余的部分再返回到体内。(图片来源:维基百科)

 

在ZUMA-1中,因为不允许“桥接化疗”(bridging chemotherapy),因此只有能够等待一定细胞生产周期的患者才能被选择,从apheresis到制成CAR-T产品的中位周期为17天;在JULIET中,患者在apheresis时就被招募进来,但从apheresis到回输CAR-T细胞进行治疗需要等待~54天。因此,92%的患者需要“桥接化疗”来控制他们的病情,有相当数量的患者接受了apheresis,但没有等到CAR-T回输(有的在等待过程中去世,有的不满足继续接受治疗的条件)。

 

长期缓解的可能性和严重毒性的风险似乎与疾病负担有关。在ZUMA-1中,肿瘤负荷四分位数最低的患者,1年缓解率为67%,≥3级CRS和神经毒性的发生率分别仅为4%、7%;而在肿瘤负荷四分位数最高的患者中,三个对应的数值分别为27%、31% 和12%。因此,在临床实践中,在患者发展为难治性疾病之前进行治疗,可能会改善预后。


 

图片来源:摄图网

 

总结

 

总结来说,来自ZUMA-1和JULIET的试验数据表明,靶向CD19的CAR-T细胞疗法能够在30–40%复发和/或难治性侵袭性大B细胞淋巴瘤患者中提供持久的缓解,且病情复发或进展的风险较低。ZUMA-1的两年随访没有发现任何意料之外的毒性,但还需要进一步的安全监测。

 

不过,文章作者表示,尽管,ZUMA-1和JULIET的试验结果令人印象深刻,但CD19-CAR-T疗法在改善持久缓解率和降低毒性方面还有很大的空间。

 

该文章还指出,除了这两个产品外,一个名为lisocabtagene maraleucel的CD19–4-1BB CAR-T细胞产品(按照特定的CD4+ T细胞/CD8+ T细胞的比率进行给药)的临床试验结果显示出了令人鼓舞的结果,试验中观察到的CRS和神经毒性发生率低于ZUMA-1和JULIET中的数据。

 

作者认为,未来,CAR-T细胞的试验应该使用相同的毒性标准,以便进行交叉试验比较。在这方面,目前,美国血液和骨髓移植协会的CRS和神经毒性分级标准是可用的。此外,理想情况下,细胞采集的protocols应该进行改进,使得淋巴细胞的采集能够在更多的地方完成,而不是集中在一个特定的制造商。

 

最终,这些产品的商业接受程度将取决于临床疗效、安全性和保健成本,包括药物成本、门诊或住院的管理费、住院时间以及因严重不良反应需要在ICU待的时间。


相关论文:

[1]FrederickL Locke et al. Long-term safety and activity of axicabtagene ciloleucel inrefractory large B-cell lymphoma (ZUMA-1): a single-arm, multicentre, phase 1–2trial. The Lancet Oncology(2018).

[2]Stephen J. Schuster etal. Tisagenlecleucel in Adult Relapsed or Refractory Diffuse Large B-CellLymphoma. NEJM(2019).

[3] Hollie J. Jackson etal. Driving CAR T-cells forward. Nature Reviews Clinical Oncology(2016).

[4]Lee, D. W. et al. Current concepts inthe diagnosis and management of cytokine release syndrome. Blood (2014).

参考资料:

1# Anti-CD19 CAR T celltherapy for lymphoma — off to the races!

机器翻译

With the FDA's approval of two CD19-targeted CAR-T cell products — Yescarta and Kymriah — as third- or more lines of therapy for relapsed and/or refractory large B-cell lymphomas, this type of cancer has entered an exciting new era of targeted anticancer therapy.

In December last year, two-year data from the pivotal clinical trial ZUMA-1 on Yescarta were published in The Lancet Oncology [1].A preliminary report on Kymriah's pivotal clinical trial, JULIET, was also presented at NEJM in January [2].

Picture source: The Lancet Oncology

ZUMA-1 enrolled 119 patients between 19-May-2015 and 15-Sep-2016, 108 of whom received Yescarta reinfusion.As of 11 August 2018, a median of 27.1 months of follow-up has been conducted for the 101 patients evaluable in the Phase II trial.The objective response rate was 83%, with a CR rate of 58%.The median duration of response was 11.1 months.Median overall survival was not reached and median progression-free survival was 5.9 months.Of the 108 evaluable patients in Phase I and Phase II, 48% had Grade ≥ 3 serious adverse events: 11% had Grade ≥ 3 cytokine release syndrome (CRS), and 32% had Grade ≥ 3 neurotoxicity.With the exception of two previously reported cases, no new treatment-related deaths occurred during follow-up.

Picture source: NEJM

In the JULIE trial, a total of 93 patients were reinfused with Kymriah and included in the efficacy assessment.The median time from reinfusion to data cutoff was 14 months.The overall response rate was 52%, with 40% CR and 12% PR.Response rates were consistent across prognostic subgroups.After 12 months from initial response, recurrence-free survival was estimated at 65%, with 79% of patients in CR.The most common Grade 3 or 4 specific adverse events included CRS (22%).Neurotoxicity (12%).Pancytopenia lasting more than 28 days (32%).Infections (20%) and febrile neutropenia (14%).Three patients died within 30 days of reinfusion due to disease progression.No deaths were reported by Kymriah.Caused by CRS or cerebral edema.

Image source: Nature Reviews Clinical Oncology

On February 27, a University of Washington scientist published an article in Nature Reviews Clinical Oncology titled "Anti-CD19CAR T cell therapy for lymphoma-off to the races!"The author made a detailed comparative analysis of these two CAR-T products.

The article states that these two multicenter.Single-arm trials have shown that patients receiving CD19-CAR-T therapy have a high response rate and survival exceeds the "historical record" (this record is the median survival of 6.2 months, from the retrospective SCHOLAR-1 analysis of patients with refractory diffuse large B-cell lymphoma).

Both trials enrolled patients with diffuse large B-cell lymphoma (DLBCL) and patients with DLBCL transformed from follicular lymphoma or other high-grade B-cell lymphomas.However, ZUMA-1 also enrolled patients with primary mediastinal B-cell lymphoma, which JULIET did not involve.

Although there are many similarities, the specific operations in clinical trials and the differences between the two CAR-T cell products prevent a direct comparison between the data.

Treatment Method Comparison

Both products are generated from autologous CD3 + T cells collected by leukapheresis and shipped unfrozen to the manufacturer.Kymriah is shipped to the manufacturer after cryopreservation.

Yescarta uses retroviral vectors and Kymriah uses lentiviral vectors.

Yescarta was administered at 2 × 106 cells/kg, while Kymriah was administered at doses ranging from 0.1 – 6.0 × 108 cells (median 3.0 × 108 cells).

CAR-T cells are usually returned to the patient after chemotherapy.The effect of chemotherapy is to eliminate the endogenous T cells and increase the survival rate of CAR-T cells.At the same time, it also reduced or delayed the immune rejection of CAR-T cells.

In ZUMA-1, chemotherapy doses were higher: 500 mg/m2 cyclophosphamide and 30 mg/m2 fludarabine per day for 3 days.In JULIET, chemotherapy doses were: 250 mg/m2 cyclophosphamide and 25 mg/m2 fludarabine per day for 3 days.In addition, in JULIET, chemotherapy was omitted in patients with severe cytoreduction (7%) and some patients received bendamustine.Differences in chemotherapy may affect early overall response rates but are unlikely to affect long-term complete response rates.

Design of CAR-T cells (Source: Nature Reviews ClinicalOncology) [3]

Safety Comparison

The CAR constructs of Yescarta and Kymriah contain the same CD19 targeting domain (CD19-targeting domain) as well as the CD3ζ signaling domain, but differ in the costimulatory domain, the former being CD28 and the latter 4-1BB.The CD28 costimulatory domain was associated with an earlier peak of T-cell expansion than the 4-1BB costimulatory domain, which may explain why Yescarta treatment was associated with an earlier onset of toxicity.In ZUMA-1, the median time to onset of CRS (any grade included) was 2 days, whereas in JULIET, the corresponding time was 3 days.Earlier onset of toxicity with Yescarta treatment requires hospitalization.

CRS is associated with cell proliferation and cytokine production, which can be controlled in part by blocking the IL-6 receptor with tocilizumab or corticosteroids alone.

Neurotoxicity is another, sometimes life-threatening, but poorly understood adverse effect of CD19-CAR-T therapy, which usually follows CRS and is also controlled with corticosteroids.

Because of the different grading systems used in ZUMA-1 and JULIET, it is not easy to compare the toxic side effects observed in these two trials.

In ZUMA-1, according to the revised version of the grading criteria proposed by Lee et al. [4], grade ≥ 3 CRS occurred in 11% of patients, and grade ≥ 3 neurotoxicity occurred in 32% of patients.

In JULIET, the investigators used the University of Pennsylvania grading scale, and 22% of patients experienced grade ≥ 3 CRS and 12% experienced grade ≥ 3 neurotoxicity.If the toxicity in JULIET was reclassified according to the grading criteria proposed by Lee et al., the incidence of CRS ≥ grade 3 was reduced to 17%.

Although the frequency of severe CRS was similar, Yescarta was associated with a higher overall incidence of CRS and more frequent use of tocilizumab than Kymriah.

Image source: Nature Reviews Clinical Oncology

Efficacy vs.

In terms of response, Yescarta and Kymriah were associated with 58% and 40% complete responses, respectively, in patients with a poor prognosis.Objective response may occur after 4 weeks of treatment.In the following months, many patients will relapse.However, in patients who do not relapse, the response will continue to increase over the next few months.Patients who achieve complete remission at 3-6 months have a lower risk of relapse.

ZUMA-1 showed that nearly 40% of patients treated with Yescarta remained in remission after a median follow-up of 2 years, with few late relapses, suggesting that many of these patients may be cured.

In the ZUMA-1 trial, patients who received CAR-T cell reinfusion, median overall survival ampgt.2 years.In the JULIET trial, the corresponding data were ~ 1 year.

When treated with Yescarta, expansion of CAR-T cells in the first 28 days (peak concentration in blood) as well as total exposure (cumulative concentration) were associated with sustained remission.However, this association was not observed with treatment with Kymriah, nor was there an association between dose and response.

Comparing these efficacy data suggests that Yescarta is more effective than Kymriah, but the design of clinical trials and eligibility criteria for both make it more complicated to determine who is better.

Apheresis: A technique in which the patient's blood is passed through a device that separates a specific component and returns the remainder to the body.(Source: Wikipedia)

In ZUMA-1, because "bridging chemotherapy" is not allowed, only patients who can wait for a certain cell production cycle can be selected, and the median period from apheresis to making CAR-T products is 17 days.In JULIET, patients are recruited at the time of apheresis, but wait ~ 54 days from apheresis to reinfusion of CAR-T cells for treatment.Therefore, 92% of patients required "bridging chemotherapy" to control their condition, and a significant number of patients received apheresis but did not wait for CAR-T reinfusion (some died while waiting, and some did not meet the conditions for continued treatment).

The likelihood of long-term remission and the risk of severe toxicity appear to be related to disease burden.In ZUMA-1, patients in the lowest quartile of tumor burden had a 1-year response rate of 67%, and the incidence of grade ≥ 3 CRS and neurotoxicity was only 4%, respectively.7%.Among patients in the highest quartile of tumor burden, the three corresponding values were 27%.31% and 12%.Therefore, in clinical practice, treatment before patients develop refractory disease may improve outcomes.

Photos from

Summary

In summary, data from the ZUMA-1 and JULIET trials indicate that CD19-targeted CAR-T cell therapy provides durable responses in 30 – 40% of patients with relapsed and/or refractory aggressive large B-cell lymphoma with a low risk of disease recurrence or progression.Two-year follow-up of ZUMA-1 did not reveal any unexpected toxicity, but further safety monitoring is needed.

However, despite the impressive results of the ZUMA-1 and JULIET trials, there is still much room for CD19-CAR-T therapy to improve durable remission rates and reduce toxicity, the authors note.

In addition to these two products, the results of a clinical trial of a CD19 – 4-1BB CAR-T cell product called lisocabtagene maraleucel (administered at a specific CD4 + T cell/CD8 + T cell ratio) showed encouraging results, with lower rates of CRS and neurotoxicity observed in the trial than in ZUMA-1 and JULIET.

The authors concluded that in the future, the same toxicity criteria should be used in CAR-T cell assays to enable cross-trial comparisons.In this regard, the American Society for Blood and Marrow Transplantation grading criteria for CRS and neurotoxicity are currently available.In addition, ideally, the protocols for cell collection should be improved so that the collection of lymphocytes can be done in more places than focused on a specific manufacturer.

Ultimately, the commercial acceptance of these products will depend on clinical efficacy.Safety and health care costs, including drug costs.Outpatient or inpatient management fees.Length of hospital stay and time spent in the ICU due to serious adverse reactions.

Related Papers:

1] FrederickL Locke et al.Long-term safety and activity of axicabtagene ciloleucel infracyclic large B-cell lymphoma (ZUMA-1): a single-arm, multicentre, phase 1 – 2 trial.The Lancet Oncology (2018).

2] StephenJ.Schuster etal.Tisagenlecleucel in Adult Relapsed or Refractive Diffuse Large B-CellLymphoma.NEJM (2019).

3] HollieJ.Jackson etal.Driving CAR T-cells forward.Nature Reviews Clinical Oncology (2016).

4] Lee, D.W. et al.Current concepts in the diagnosis and management of cytokine release syndrome.Blood (2014).

References:

1anti-CD19 CAR Tcelltherapy for lymphoma-off to the races!

扫码实时看更多精彩文章

版权及免责声明 本文由医药魔方原创,版权归医药魔方所有。未经许可,严禁任何媒体或个人以任何形式摘编、改写、复制、转载本文内容。对于恶意侵权行为,医药魔方保留采用法律手段追究的权利。媒体内容及商务合作请联系医药魔方工作人员(微信号:medicube)

魔方微信公众号