Introduction: Solitary fibrous tumors (SFTs) are rare mesenchymal neoplasms. Most follow a benign course, but a subset will recur or metastasize. Various risk stratification schemes have been proposed for SFTs, but none has been universally endorsed and few have focused on pleuropulmonary SFTs. Methods: Histologic sections from surgically resected pleuropulmonary SFTs were examined, with confirmatory immunohistochemistry. Patients were risk‐stratified by using four prediction models as proposed by de Perrot, Demicco (original and modified), and Tapias. Kaplan‐Meier analysis was used to estimate overall survival (OS) and progression‐free survival (PFS). Results: The 147 study patients included 78 females (53.1%) with a median age of 61.5 years (range 25–87). The median follow‐up was 5.5 years (range 0–33). Recurrence or metastasis occurred in 15 patients (10.2%), with five deaths from disease. Significant predictors of worse OS included male sex, age at least 55 years, tumor size at least 10 cm, nonpedunculated growth, severe atypia, necrosis, and mitotic count of at least four per 10 high‐power fields. Predictors of recurrence included tumor size of at least 10 cm, severe atypia, necrosis, at least four mitoses per 10 high‐power fields, and Ki67 labeling index of at least 2%. All systems predicted PFS, but only the Demicco and Tapias systems significantly predicted OS. The modified Demicco system provided the best discrimination for PFS (C‐statistic = 0.80 compared with 0.78). Conclusion: The risk scoring systems proposed by Tapias et al. and Demicco et al. were both predictive of OS and PFS. The Demicco system has the advantages of simplicity and applicability to SFTs from other sites, as well as provision of the best discrimination for PFS, and thus may be the best system to apply in general practice.

中文翻译：

---

预测胸膜肺孤立性纤维瘤复发和生存的风险分层模型的比较

简介：孤立性纤维瘤（SFTs）是一种罕见的间叶性肿瘤。大多数遵循良性过程，但一个子集会复发或转移。已经为 SFT 提出了各种风险分层方案，但没有一个得到普遍认可，而且很少有人关注胸膜肺 SFT。方法：检查来自手术切除的胸膜肺 SFT 的组织学切片，并进行确认性免疫组织化学。使用 de Perrot、Demicco（原始和修改）和 Tapias 提出的四种预测模型对患者进行风险分层。Kaplan-Meier 分析用于估计总生存期（OS）和无进展生存期（PFS）。结果：147 名研究患者包括 78 名女性 (53.1%)，中位年龄为 61.5 岁（范围 25-87）。中位随访时间为 5.5 年（范围 0-33）。15 名患者（10.2%）发生复发或转移，其中 5 人死于疾病。OS 较差的重要预测因素包括男性、年龄至少 55 岁、肿瘤大小至少 10 厘米、无蒂生长、严重异型性、坏死和有丝分裂计数至少为每 10 个高倍视野 4 个。复发的预测因素包括肿瘤大小至少为 10 cm、严重异型性、坏死、每 10 个高倍视野至少有 4 个有丝分裂，以及 Ki67 标记指数至少为 2%。所有系统都预测了 PFS，但只有 Demicco 和 Tapias 系统显着预测了 OS。修改后的 Demicco 系统为 PFS 提供了最佳区分（C 统计量 = 0.80，而 C 统计量为 0.78）。结论：Tapias 等人提出的风险评分系统。和 Demicco 等人。均能预测 OS 和 PFS。