Kang Chen1, Butong Sun1. 1. Department of Oncology and Hematology, China-Japan Union Hospital of Jilin University, Changchun 130031, China.
Abstract
BACKGROUND: Immune checkpoint inhibitors (ICIs) have good efficacy on most advanced tumors, which brings new hope to patients with advanced tumors. However, the immune system activated by ICIs may attack human normal tissues and organs, resulting in corresponding immunotoxicity, such as checkpoint inhibitor pneumonitis. This article carried out a meta-analysis on the incidence and risk of programmed cell death protein 1 (PD-1) and programmed cell death protein ligand 1 (PD-L1) inhibitor-associated pneumonia in advanced tumors patients. METHODS: The computer retrieval of PubMed, Cochrane Library, EMbase and CNKI was performed, and the studies on the occurrence rate of PD-1/PD-L1 inhibitor-associated pneumonia in terminal cancer patients were collected, with the retrieval time limit of January 2000 to January 2020. Statistical analysis was conducted by using Revman 5.3 and R 3.6.2 software to compare the occurrence rate of pneumonia under different circumstances. RESULTS: 15 studies were included, involving 8,642 patients, of which those with PD-1/PD-L1 inhibitor were treatment group, and those with chemotherapy were control group. The odds radio of all grades of immune pneumonia was 6.63, and that of high grade was 4.87. The occurrence rate of all grades of pneumonia in the ICI group with non-small cell lung cancer (NSCLC) was 1.658 times than other tumors, and that of high grade was 2.299 times. The occurrence rate of all grades of pneumonia in second-line or more treatment with ICI was 0.489 times than that in first-line, and that of high grade was 0.449 times than that in first-line or more treatment with ICI. CONCLUSIONS: Compared with chemotherapy, the risk of immune-associated pneumonia is higher in PD-1 and PD-L1 inhibitors, and its occurrence risk is high in the ICI group with NSCLC and the first-line treatment with ICI. This paper provides guidance for clinic treatment of terminal cancers and prevention of complications.
BACKGROUND: Immune checkpoint inhibitors (ICIs) have good efficacy on most advanced tumors, which brings new hope to patients with advanced tumors. However, the immune system activated by ICIs may attack human normal tissues and organs, resulting in corresponding immunotoxicity, such as checkpoint inhibitor pneumonitis. This article carried out a meta-analysis on the incidence and risk of programmed cell death protein 1 (PD-1) and programmed cell death protein ligand 1 (PD-L1) inhibitor-associated pneumonia in advanced tumorspatients. METHODS: The computer retrieval of PubMed, Cochrane Library, EMbase and CNKI was performed, and the studies on the occurrence rate of PD-1/PD-L1 inhibitor-associated pneumonia in terminal cancerpatients were collected, with the retrieval time limit of January 2000 to January 2020. Statistical analysis was conducted by using Revman 5.3 and R 3.6.2 software to compare the occurrence rate of pneumonia under different circumstances. RESULTS: 15 studies were included, involving 8,642 patients, of which those with PD-1/PD-L1 inhibitor were treatment group, and those with chemotherapy were control group. The odds radio of all grades of immune pneumonia was 6.63, and that of high grade was 4.87. The occurrence rate of all grades of pneumonia in the ICI group with non-small cell lung cancer (NSCLC) was 1.658 times than other tumors, and that of high grade was 2.299 times. The occurrence rate of all grades of pneumonia in second-line or more treatment with ICI was 0.489 times than that in first-line, and that of high grade was 0.449 times than that in first-line or more treatment with ICI. CONCLUSIONS: Compared with chemotherapy, the risk of immune-associated pneumonia is higher in PD-1 and PD-L1 inhibitors, and its occurrence risk is high in the ICI group with NSCLC and the first-line treatment with ICI. This paper provides guidance for clinic treatment of terminal cancers and prevention of complications.
过去十年间,肿瘤的免疫治疗飞速发展,其中免疫检查点抑制剂(immune checkpoint inhibitor, ICI)已成为肿瘤治疗领域的热点,并且在疗效方面取得令人瞩目的成绩。ICIs主要靶点包括细胞毒性T淋巴细胞抗原4(cytotoxic T lymphocyte-associated antigen 4, CTLA-4)和程序性细胞死亡受体1(programmed cell death protein 1, PD-1)及程序性细胞死亡受体配体1(programmed cell death protein ligand 1, PD-L1)。ICIs通过抑制CTLA-4、PD-1及PD-L1,解除免疫抑制,活化T淋巴细胞,达到清除肿瘤的目的[,其中PD-1抑制剂(Nivolumab、Pembrolizumab)和PD-L1抑制剂(Atezolizumab),已被作为多种类型晚期肿瘤的推荐用药写入指南。然而,随着ICIs在临床全面的应用,由此产生的免疫相关不良事件(immune related adverse events, irAE)也逐渐得到关注。这些irAEs主要包括皮肤、肺脏、内分泌、血液、消化系统等,其中免疫检查点抑制剂肺炎(checkpoint inhibitor pneumonitis, CIP)是ICI相关毒性的一种。CIP的总体发生率虽然不高,但是严重的CIP造成免疫治疗中断,甚至成为威胁生命的重要因素之一。因此临床医生需要对这一罕见但严重的不良事件给予更多的关注[。即使已有众多临床试验关于PD-1/PD-L1抑制剂相关肺炎发病率的报道,但上述临床试验中所获得的个体及队列数量较为局限。此外,虽然部分相关荟萃分析也研究了免疫相关肺炎的发生率,然而也面临着纳入分析的研究数量及人数有限的问题。基于此,我们对PD-1/PD-L1抑制剂在晚期癌症患者中相关肺炎的临床试验进行了荟萃分析,并比较了免疫相关性肺炎在不同肿瘤类型及治疗方案间的危险度和发病率,旨在为后期的临床治疗及相关并发症的预防提供指导依据。
资料与方法
检索策略
根据系统综述和荟萃分析优先报告的条目(Preferred Reporting Items for Systematic Reviews and meta-Analyses, PRISMA)原则,用计算机检索PubMed、Cochrane Library、EMbase、中国知网数据库,检索针对PD-1/PD-L1抑制剂在晚期肿瘤患者中相关肺炎的发生率及危险度的研究的相关文献,因avelumab和durvalumab发表报告较少,故检索词英文为:“immune checkpoint inhibitor”、“ICIs”、“nivolumab”、“pembrolizumab”、“atezolizumab”、“cancer”、“tumor”、“phase Ⅱ”、“phase Ⅲ”、“pneumonia”、“pneumonitis”,中文为:“PD-L1抑制剂”、“PD-1抑制剂”、“免疫检查点抑制剂”、“特瑞普利单抗”、“卡瑞利珠单抗”,搜索时间从2000年1月-2020年1月,文献类型限制为临床试验,为防止文献漏检,同时搜索相关参考文献,本文未进行手动检索。
Risk of bias graphs and summaries in various categories across all of the studies included. A: risk of bias graph; B: risk of bias summary.
在所有纳入的研究中,不同类别的偏倚风险图和总结。A:偏倚风险图;B:偏见风险总结。Risk of bias graphs and summaries in various categories across all of the studies included. A: risk of bias graph; B: risk of bias summary.
PD-1: programmed cell death protein 1; PD-L1: programmed cell death protein ligand 1.
Checkmate 017[3]
131
129
6
4.60%
1
0.80%
1
0.80%
1
0.80%
Checkmate 057[4]
287
268
8
2.80%
1
0.30%
3
1%
0
0
Checkmate 066[5]
206
205
3
1.50%
0
0
0
0
0
0
Checkmate 037[6]
268
102
6
2.20%
0
0
0
0
0
0
Checkmate 026[7]
267
263
7
2.60%
0
0
4
1.5%
0
0
Checkmate 141[8]
236
111
5
2.11%
1
0.90%
2
0.80%
0
0
Keynote 024[9]
154
150
9
5.80%
1
0.70%
4
2.60%
1
0.70%
Keynote 002[10]
357
171
6
1.70%
0
0
2
0.60%
0
0
Keynote 010[11]
682
309
31
4.50%
6
1.90%
14
2.10%
2
0.60%
Keynote 045[12]
266
255
11
4.10%
1
0.40%
6
2.30%
0
0
Keynote 042[13]
636
615
53
8.00%
3
0.50%
22
3%
1
0.20%
Keynote 061[14]
294
276
8
3.00%
0
0
2
0.70%
0
0
Keynote 040[15]
246
234
10
4.00%
3
1%
3
1%
3
1%
Oak[16]
609
578
6
1.00%
1
0.20%
4
0.70%
0
0
Poplar[17]
142
135
4
2.80%
0
0
1
0.70%
0
0
Total
4, 781
3, 801
173
3.60%
18
0.50%
68
1.40%
8
0.20%
图 3
PD-1/PD-L1治疗组所有级别肺炎发生率
Incidence of pneumonia at all levels in PD-1/PD-L1 treatment group
纳入研究的肺炎发生率Incidence of pneumonia in the studiesPD-1/PD-L1治疗组所有级别肺炎发生率Incidence of pneumonia at all levels in PD-1/PD-L1 treatment group
所有级别和高级别肺炎发生率的OR值
对实验组为PD-1抑制剂或PD-L1抑制剂,对照组为化疗组的研究的肺炎发生率进行评估,所有级别肺炎发生率OR值为6.63(95%CI: 4.22-10.40, P < 0.000, 01),此分析无异质性(I2=0%)(图 4A)。高级别肺炎发生率OR值为4.87(95%CI: 2.64-8.99, P < 0.000, 01),此分析同样无异质性(I2=0%)(图 4B)。
图 4
PD-1/PD-L1抑制剂治疗组与对照组肺炎OR值。A:所有级别肺炎;B:高级别肺炎。
OR value of pneumonia in PD-1/PD-L1 inhibitor treatment group and control group. A: all-grade of pneumonia; B: high-grade pneumonia. OR:odds ratio.
PD-1/PD-L1抑制剂治疗组与对照组肺炎OR值。A:所有级别肺炎;B:高级别肺炎。OR value of pneumonia in PD-1/PD-L1 inhibitor treatment group and control group. A: all-grade of pneumonia; B: high-grade pneumonia. OR:odds ratio.亚组分析:同时评估了不同类型免疫检查点抑制剂(Nivolumab、Pembrolizumab、Atezolizumab)相关肺炎的风险(图 5A和图 5B)。与对照组相比,Nivolumab的应用增加了所有级别肺炎的风险(OR=6.27, 95%CI: 2.35-16.73, P=0.000, 2),但高级别肺炎的风险未显示增加(OR=3.89, 95%CI: 0.99-15.34, P=0.05)。Pembrolizumab治疗组增加所有级别肺炎的风险(OR=6.72, 95%CI: 3.95-11.43, P < 0.000, 01),同时高级别肺炎的风险增加(OR=5.06, 95%CI: 2.45-10.42, P < 0.000, 1)。Atezolizumab治疗组所有级别肺炎的风险增加(OR=6.75, 95%CI: 1.22-37.45, P=0.03),但高级别肺炎的风险未增加(OR:5.74, 95%CI: 0.69-47.86, P=0.11)。最后Nivolumab、Pembrolizumab和Atezolizumab在所有级别组(P=0.99)和高级别组(P=0.93)均未发现明显差异。
图 5
不同类型ICIs相关肺炎的OR值。A:所有级别肺炎;B:高级别肺炎。
OR values for pneumonia associated with different types of ICIs. A: all-grade of pneumonia; B: high-grade pneumonia. ICIs: immune checkpoint inhibitors.
不同类型ICIs相关肺炎的OR值。A:所有级别肺炎;B:高级别肺炎。OR values for pneumonia associated with different types of ICIs. A: all-grade of pneumonia; B: high-grade pneumonia. ICIs: immune checkpoint inhibitors.
Incidence of pneumonia in non-small cell lung cancer in the ICIs group (A: all-grade of pneumonia; C: high-grade pneumonia); Incidence of pneumonia in other tumors of ICIs group (B: all-grade of pneumonia; D: high-grade pneumonia)
ICIs组非小细胞肺癌中肺炎发生率(A:所有级别肺炎;C:高级别肺炎);ICIs组其他瘤种中肺炎发生率(B:所有级别肺炎;D:高级别肺炎)Incidence of pneumonia in non-small cell lung cancer in the ICIs group (A: all-grade of pneumonia; C: high-grade pneumonia); Incidence of pneumonia in other tumors of ICIs group (B: all-grade of pneumonia; D: high-grade pneumonia)
Incidence of pneumonia in front-line ICIs (A: all-grade of pneumonia; C: high-grade pneumonia); Incidence of pneumonia in second-line and above (B: all-grade of pneumonia; D: high-grade pneumonia)
一线应用ICIs中肺炎的发生率(A:所有级别肺炎;C:高级别肺炎);二线及以上应用ICIs中肺炎的发生率(B:所有级别肺炎;D:高级别肺炎)Incidence of pneumonia in front-line ICIs (A: all-grade of pneumonia; C: high-grade pneumonia); Incidence of pneumonia in second-line and above (B: all-grade of pneumonia; D: high-grade pneumonia)
发表偏倚
纳入的15项研究在所有级别肺炎和高级别肺炎的OR值的荟萃分析中均表现为对称,无发表偏倚(图 8)。
图 8
相关肺炎的OR值的漏斗图。A:所有级别肺炎;B:高级别肺炎。
The funnel plot of the OR of pneumonia. A: all-grade of pneumonia; B: high-grade pneumonia.
相关肺炎的OR值的漏斗图。A:所有级别肺炎;B:高级别肺炎。The funnel plot of the OR of pneumonia. A: all-grade of pneumonia; B: high-grade pneumonia.
讨论
近年来,ICIs已成为最受欢迎的各种类型晚期肿瘤治疗方案之一,但免疫相关的不良反应是完全不同于传统的放疗、化疗及靶向治疗的不良反应。它们具有独特的特征,包括特定器官的倾向性、没有剂量依赖的相关性以及潜在的迟发性[。在这些与免疫相关的不良反应中,CIP往往是严重的,并可能危及生命。因此,对CIP的深入了解,对临床治疗及并发症防治有重要意义。CIP主要的临床症状是咳嗽、呼吸困难及呼吸急促,发生CIP的中位时间为2.5个月(范围:0.8个月-11.0个月)。在所有CIP病例中,72%的患者为1级-2级。与甲状腺炎和肝炎等自限性免疫反应不同,大部分的CIP需要激素或免疫抑制剂的治疗,据文献[报道,超过85%的患者可以通过停药和免疫抑制治疗得到缓解或治愈,但也有一部分患者使用激素治疗之后得不到缓解,一项meta分析[显示,PD-1/PD-L1抑制剂致死的主要原因为CIP(35%)。有关报道[指出:不同瘤种的CIP的发生率存在差异,相比恶性黑色素瘤患者,使用ICIs的非小细胞肺癌及肾恶性肿瘤患者CIP的发生率明显更高,分别为4.1% vs 1.6%(P=0.002)和4.1% vs 1.6%(P < 0.001),本荟萃分析同样发现,非小细胞肺癌患者中CIP的发生率较高。不同瘤种的CIP发生率差异的具体机制尚不明确,可能的原因:肿瘤微环境、免疫浸润、适应性免疫反应和新抗原的形成可能受到组织学的影响[。同时,ICIs治疗前接受过胸部放疗,接受EGFR-TKI联合ICIs治疗的驱动基因敏感突变阳性的非小细胞肺癌患者[,先前存在慢性阻塞性肺疾病(chronic obstructive pulmonary disease, COPD)、肺纤维化等或目前存在肺部活动性感染的患者为发生CIP的高危人群[,应用ICIs时需提高警惕。CIP的临床症状多种多样,所以一旦接受PD-1和PD-L1抑制剂的患者出现呼吸系统症状,均应行胸部电子计算机断层扫描(Computed Tomography, CT)检查。影像学特征包括提示急性间质性肺炎、隐源性组织肺炎、超敏性肺炎或非特异性间质性肺炎[。如果CT结果不确定,应行支气管镜检查并行支气管肺泡灌洗,若发生CIP,灌洗液中可见大量淋巴细胞。与化疗方案相比,免疫检查点抑制剂可显著增加任何级别和高级别的肺炎的发生率和风险。Nivolumab、Pembrolizumab、Atezolizumab、Avelumab、Durvalumab等虽然具有显著的疗效和良好的安全性并且已被美国食品药品监督管理局(Food and Drug Administration, FDA)批准用于治疗不同类型的癌症,然而,使用这些免疫检查点抑制剂治疗的患者发生CIP的风险是一致的,临床医生在选择此类药物时应始终考虑到这一点[。本研究分析的局限性首先表现为其他ICIs和其他肿瘤类型的覆盖不充分。因为在本研究收集数据时缺乏相关研究的数据,此文章的分析集中在PD-1/PD-L1抑制剂的相关肺毒性,而不包括其他ICI,如CTLA-4抑制剂。其次,文章的研究集中于黑色素瘤、非小细胞肺癌、胃癌、头颈部鳞癌和尿路上皮癌,对于其他类型的肿瘤,如结直肠癌或淋巴瘤,由于发表的报告数量有限、样本量小,所以没有被包括在内。另外,我们荟萃分析的研究中选择的患者是一组经过筛选的活动状态良好的患者,但是在临床实践中,器官功能障碍患者的实际毒性发生率可能更高。最后,荟萃分析的研究中有一部分患者先前存在肺间质疾病,这都可能造成研究的异质性,通过亚组分析和敏感性分析找到异质性来源,可证明荟萃分析的稳健性。值得注意的是,相关报道[指出CIP发生率与PD-1抑制剂的剂量无关,所以我们将同一研究的中不同剂量的PD-1抑制剂相关肺炎发生人数进行了合并,并不影响荟萃分析最终结果。针对PD-1/PD-L1通路的不同药物的肺炎发病率,还需要更多数据以及进一步的研究。根据此荟萃分析,我们建议在临床采用PD-1/PD-L1抑制剂治疗晚期肿瘤患者时,应注意患者的个体化管理,根据不良反应发生率的差异性,做好预防和相应的治疗准备。
Authors: Caroline Robert; Georgina V Long; Benjamin Brady; Caroline Dutriaux; Michele Maio; Laurent Mortier; Jessica C Hassel; Piotr Rutkowski; Catriona McNeil; Ewa Kalinka-Warzocha; Kerry J Savage; Micaela M Hernberg; Celeste Lebbé; Julie Charles; Catalin Mihalcioiu; Vanna Chiarion-Sileni; Cornelia Mauch; Francesco Cognetti; Ana Arance; Henrik Schmidt; Dirk Schadendorf; Helen Gogas; Lotta Lundgren-Eriksson; Christine Horak; Brian Sharkey; Ian M Waxman; Victoria Atkinson; Paolo A Ascierto Journal: N Engl J Med Date: 2014-11-16 Impact factor: 91.245
Authors: Karthik Suresh; Khinh Ranh Voong; Bairavi Shankar; Patrick M Forde; David S Ettinger; Kristen A Marrone; Ronan J Kelly; Christine L Hann; Benjamin Levy; Josephine L Feliciano; Julie R Brahmer; David Feller-Kopman; Andrew D Lerner; Hans Lee; Lonny Yarmus; Franco D'Alessio; Russell K Hales; Cheng Ting Lin; Kevin J Psoter; Sonye K Danoff; Jarushka Naidoo Journal: J Thorac Oncol Date: 2018-09-26 Impact factor: 15.609
Authors: Achim Rittmeyer; Fabrice Barlesi; Daniel Waterkamp; Keunchil Park; Fortunato Ciardiello; Joachim von Pawel; Shirish M Gadgeel; Toyoaki Hida; Dariusz M Kowalski; Manuel Cobo Dols; Diego L Cortinovis; Joseph Leach; Jonathan Polikoff; Carlos Barrios; Fairooz Kabbinavar; Osvaldo Arén Frontera; Filippo De Marinis; Hande Turna; Jong-Seok Lee; Marcus Ballinger; Marcin Kowanetz; Pei He; Daniel S Chen; Alan Sandler; David R Gandara Journal: Lancet Date: 2016-12-13 Impact factor: 79.321
Authors: O Abdel-Rahman; D Helbling; J Schmidt; U Petrausch; A Giryes; A Mehrabi; O Schöb; M Mannhart; H Oweira Journal: Clin Oncol (R Coll Radiol) Date: 2016-11-25 Impact factor: 4.126
Authors: Martin Reck; Delvys Rodríguez-Abreu; Andrew G Robinson; Rina Hui; Tibor Csőszi; Andrea Fülöp; Maya Gottfried; Nir Peled; Ali Tafreshi; Sinead Cuffe; Mary O'Brien; Suman Rao; Katsuyuki Hotta; Melanie A Leiby; Gregory M Lubiniecki; Yue Shentu; Reshma Rangwala; Julie R Brahmer Journal: N Engl J Med Date: 2016-10-08 Impact factor: 91.245
Authors: Louis Fehrenbacher; Alexander Spira; Marcus Ballinger; Marcin Kowanetz; Johan Vansteenkiste; Julien Mazieres; Keunchil Park; David Smith; Angel Artal-Cortes; Conrad Lewanski; Fadi Braiteh; Daniel Waterkamp; Pei He; Wei Zou; Daniel S Chen; Jing Yi; Alan Sandler; Achim Rittmeyer Journal: Lancet Date: 2016-03-10 Impact factor: 79.321
Authors: Daniel Y Wang; Joe-Elie Salem; Justine V Cohen; Sunandana Chandra; Christian Menzer; Fei Ye; Shilin Zhao; Satya Das; Kathryn E Beckermann; Lisa Ha; W Kimryn Rathmell; Kristin K Ancell; Justin M Balko; Caitlin Bowman; Elizabeth J Davis; David D Chism; Leora Horn; Georgina V Long; Matteo S Carlino; Benedicte Lebrun-Vignes; Zeynep Eroglu; Jessica C Hassel; Alexander M Menzies; Jeffrey A Sosman; Ryan J Sullivan; Javid J Moslehi; Douglas B Johnson Journal: JAMA Oncol Date: 2018-12-01 Impact factor: 31.777
Authors: G R Oxnard; J C-H Yang; H Yu; S-W Kim; H Saka; L Horn; K Goto; Y Ohe; H Mann; K S Thress; M M Frigault; K Vishwanathan; D Ghiorghiu; S S Ramalingam; M-J Ahn Journal: Ann Oncol Date: 2020-01-24 Impact factor: 32.976