Background Relatives of patients with familial interstitial pneumonia (FIP) are at increased risk for pulmonary fibrosis. We assessed the prevalence and risk factors for preclinical pulmonary fibrosis (PrePF) in first-degree relatives of patients with FIP and determined the utility of deep learning in detecting PrePF on CT. Methods First-degree relatives of patients with FIP over 40 years of age who believed themselves to be unaffected by pulmonary fibrosis underwent CT scans of the chest. Images were visually reviewed, and a deep learning algorithm was used to quantify lung fibrosis. Genotyping for common idiopathic pulmonary fibrosis risk variants in MUC5B and TERT was performed. Findings In 494 relatives of patients with FIP from 263 families of patients with FIP, the prevalence of PrePF on visual CT evaluation was 15.6% (95% CI 12.6 to 19.0). Compared with visual CT evaluation, deep learning quantitative CT analysis had 84% sensitivity (95% CI 0.72 to 0.89) and 86% sensitivity (95% CI 0.83 to 0.89) for discriminating subjects with visual PrePF diagnosis. Subjects with PrePF were older (65.9, SD 10.1 years) than subjects without fibrosis (55.8 SD 8.7 years), more likely to be male (49% vs 37%), more likely to have smoked (44% vs 27%) and more likely to have the MUC5B promoter variant rs35705950 (minor allele frequency 0.29 vs 0.21). MUC5B variant carriers had higher quantitative CT fibrosis scores (mean difference of 0.36%), a difference that remains significant when controlling for age and sex. Interpretation PrePF is common in relatives of patients with FIP. Its prevalence increases with age and the presence of a common MUC5B promoter variant. Quantitative CT analysis can detect these imaging abnormalities.

中文翻译：

---

MUC5B 变异与视觉和定量检测到的临床前肺纤维化有关

背景 家族性间质性肺炎 (FIP) 患者的亲属患肺纤维化的风险增加。我们评估了 FIP 患者的一级亲属临床前肺纤维化 (PrePF) 的患病率和危险因素，并确定了深度学习在 CT 上检测 PrePF 的效用。方法 对自认为未受肺纤维化影响的 40 岁以上 FIP 患者的一级亲属进行胸部 CT 扫描。对图像进行视觉审查，并使用深度学习算法来量化肺纤维化。对 MUC5B 和 TERT 中常见的特发性肺纤维化风险变异进行了基因分型。结果 在来自 263 个 FIP 患者家族的 494 名 FIP 患者亲属中，视觉 CT 评估中 PrePF 的患病率为 15.6%（95% CI 12.6 至 19.0）。与视觉 CT 评估相比，深度学习定量 CT 分析在区分具有视觉 PrePF 诊断的受试者方面具有 84%（95% CI 0.72 至 0.89）和 86% 灵敏度（95% CI 0.83 至 0.89）。患有 PrePF 的受试者比没有纤维化的受试者（55.8 SD 8.7 岁）年龄更大（65.9，SD 10.1 岁），更可能是男性（49% 对 37%），更有可能吸烟（44% 对 27%）等等可能有 MUC5B 启动子变体 rs35705950（次要等位基因频率 0.29 vs 0.21）。MUC5B 变异携带者具有更高的 CT 纤维化定量评分（平均差异为 0.36%），当控制年龄和性别时，这一差异仍然显着。解释 PrePF 在 FIP 患者的亲属中很常见。其患病率随着年龄和常见 MUC5B 启动子变体的存在而增加。