Massachusetts General Hospital, Department of Dermatology, Boston.
Harvard Medical School, Department of Dermatology, Boston, Massachusetts.
JAMA Dermatol. 2022 Feb 1;158(2):189-193. doi: 10.1001/jamadermatol.2021.5476.
Despite the efficacy of immune checkpoint inhibitors (ICIs), cutaneous immune-related adverse events (cirAEs) occur in 20% to 40% of all treated patients. To our knowledge, little is known about the predictive value of these cutaneous eruptions and their subtypes regarding cancer survival.
To determine the association of developing cirAEs following treatment with anti-programmed cell death 1 (PD-1) or anti-programmed cell death ligand 1 (PD-L1) therapy with patient survival.
DESIGN, SETTING, AND PARTICIPANTS: This retrospective cohort study used data from the TriNetX Diamond Network, a database of health records and claims data from more than 200 million US and European patients, to conduct a population-level cohort analysis. The study included 7008 eligible patients who developed cirAEs after treatment with anti-PD-1 or anti-PD-L1 therapy for malignant neoplasms of digestive organs, bronchus or lung, melanoma of skin, and urinary tract who were identified through the TriNetX Diamond Network along with 7008 matched controls.
Development of cirAEs within 6 months following anti-PD-1 or anti-PD-L1 therapy.
A 6-month analysis using a Cox proportional hazards model was performed to determine the association of cirAEs with overall survival after adjusting for demographic characteristics, cancer type, and cancer stage.
A total of 7008 patients (3036 women [43.3%]; mean [SD] age, 68.2 [11.2] years) were matched to 7008 (3044 women [43.4%]; mean [SD] age, 68.3 [11.1] years) controls. Pruritus (hazard ratio [HR], 0.695; 95% CI, 0.602-0.803; P < .001), drug eruption (HR, 0.755; 95% CI, 0.635-0.897; P = .001), xerosis (HR, 0.626; 95% CI, 0.469-0.834; P = .001), nonspecific rashes (HR, 0.704; 95% CI, 0.634-0.781; P < .001), and appearance of any cirAE (HR, 0.778; 95% CI, 0.726-0.834; P < .001) were significantly protective of mortality using a Benjamini-Hochberg correction with a significance level of .05. Additionally, psoriasis (HR, 0.703; 95% CI, 0.497-0.994; P = .045) and lichen planus/lichenoid dermatitis (HR, 0.511; 95% CI, 0.279-0.939; P = .03) were significant. Eczematous dermatitis (HR, 0.612; 95% CI, 0.314-1.195), vitiligo (HR, 0.534; 95% CI, 0.254-1.123), bullous pemphigoid (HR, 0.524; 95% CI, 0.140-1.956), and Grover disease (HR, 0.468; 95% CI, 0.115-1.898) were all associated with strong protective clinical effects.
The results of this cohort study suggest that the development of cirAEs is strongly associated with response to ICI therapy and patient survival.
尽管免疫检查点抑制剂(ICIs)有效,但所有接受治疗的患者中有 20%至 40%会出现皮肤免疫相关不良反应(cirAEs)。据我们所知,对于这些皮肤发作及其亚型与癌症生存之间的预测价值,人们知之甚少。
确定治疗后出现 cirAEs 与接受抗程序性细胞死亡 1(PD-1)或抗程序性细胞死亡配体 1(PD-L1)治疗的患者生存之间的关联。
设计、设置和参与者:这项回顾性队列研究使用了来自 TriNetX Diamond 网络的数据库中的数据,该数据库包含来自 200 多万美国和欧洲患者的健康记录和索赔数据,进行了人群水平的队列分析。该研究包括 7008 名符合条件的患者,他们在接受抗 PD-1 或抗 PD-L1 治疗恶性消化系统肿瘤、支气管或肺、皮肤黑色素瘤和尿路肿瘤后 6 个月内出现 cirAEs,并与 7008 名匹配的对照患者一起通过 TriNetX Diamond 网络确定。
在接受抗 PD-1 或抗 PD-L1 治疗后 6 个月内发生 cirAEs。
使用 Cox 比例风险模型进行了 6 个月的分析,以确定 cirAEs 与调整人口统计学特征、癌症类型和癌症分期后总生存之间的关联。
共匹配了 7008 名患者(3036 名女性[43.3%];平均[SD]年龄 68.2[11.2]岁)和 7008 名(3044 名女性[43.4%];平均[SD]年龄 68.3[11.1]岁)对照患者。瘙痒(风险比[HR],0.695;95%CI,0.602-0.803;P<0.001)、药物疹(HR,0.755;95%CI,0.635-0.897;P=0.001)、干燥症(HR,0.626;95%CI,0.469-0.834;P=0.001)、非特异性皮疹(HR,0.704;95%CI,0.634-0.781;P<0.001)和任何 cirAE 的出现(HR,0.778;95%CI,0.726-0.834;P<0.001)在使用 Benjamini-Hochberg 校正的显著性水平为 0.05 时,对死亡率具有显著的保护作用。此外,银屑病(HR,0.703;95%CI,0.497-0.994;P=0.045)和扁平苔藓/苔藓样皮炎(HR,0.511;95%CI,0.279-0.939;P=0.03)具有显著意义。湿疹性皮炎(HR,0.612;95%CI,0.314-1.195)、白癜风(HR,0.534;95%CI,0.254-1.123)、大疱性类天疱疮(HR,0.524;95%CI,0.140-1.956)和 Grover 病(HR,0.468;95%CI,0.115-1.898)均与强烈的保护临床效果相关。
这项队列研究的结果表明,cirAEs 的发生与 ICI 治疗反应和患者生存密切相关。