Objectives

Exhaled breath analysis by electronic nose (eNose) has shown to be a potential predictive biomarker before start of anti-PD-1 therapy in patients with non-small cell lung carcinoma (NSCLC). We hypothesized that the eNose could also be used as an early monitoring tool to identify responders more accurately at early stage of treatment when compared to baseline. In this proof-of-concept study we aimed to definitely discriminate responders from non-responders after six weeks of treatment.

Materials and Methods

This was a prospective observational study in patients with advanced NSCLC eligible for anti-PD-1 treatment. The efficacy of treatment was assessed by the Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 at 3-month follow-up. We analyzed SpiroNose exhaled breath data of 94 patients (training cohort n = 62, validation cohort n = 32). Data analysis involved signal processing and statistics based on Independent Samples T-tests and Linear Discriminant Analysis (LDA) followed by Receiver Operating Characteristic (ROC) analysis.

Results

In the training cohort, a specificity of 73% was obtained at a 100% sensitivity level to identify objective responders. The Area Under the Curve (AUC) was 0.95 (CI: 0.89–1.00). In the validation cohort, these results were confirmed with an AUC of 0.97 (CI: 0.91–1.00).

Conclusion

Exhaled breath analysis by eNose early during treatment allows for a highly accurate, non-invasive and low-cost identification of advanced NSCLC patients who benefit from anti-PD-1 therapy.

中文翻译：

---

用于非小细胞肺癌早期免疫治疗反应监测的 eNose 分析

目标

在非小细胞肺癌 (NSCLC) 患者开始抗 PD-1 治疗之前，通过电子鼻 (eNose) 进行的呼出气分析已被证明是一种潜在的预测性生物标志物。我们假设，与基线相比，eNose 也可以用作早期监测工具，在治疗的早期阶段更准确地识别反应者。在这项概念验证研究中，我们旨在明确区分治疗六周后有反应者和无反应者。

材料和方法

这是一项针对适合抗 PD-1 治疗的晚期 NSCLC 患者的前瞻性观察研究。在 3 个月的随访中，通过实体瘤反应评估标准 (RECIST) 1.1 版评估治疗效果。我们分析了 94 名患者的 SpiroNose 呼气数据（训练队列 n = 62，验证队列 n = 32）。数据分析涉及基于独立样本 T 检验和线性判别分析 (LDA) 的信号处理和统计，然后是接收器操作特性 (ROC) 分析。

结果

在训练队列中，在 100% 的灵敏度水平下获得了 73% 的特异性，以识别客观响应者。曲线下面积 (AUC) 为 0.95（CI：0.89–1.00）。在验证队列中，AUC 为 0.97（CI：0.91–1.00）证实了这些结果。

结论

在治疗早期通过 eNose 进行呼气分析，可以高度准确、无创和低成本地识别受益于抗 PD-1 治疗的晚期 NSCLC 患者。