Shuryak Igor, Wang Eric, Brenner David J
Center for Radiological Research, Columbia University Irving Medical Center, New York City, NY, United States.
Front Oncol. 2024 Aug 13;14:1422211. doi: 10.3389/fonc.2024.1422211. eCollection 2024.
Treating head and neck squamous cell carcinomas (HNSCC), especially human papillomavirus negative (HPV-) and locally advanced cases, remains difficult. Our previous analyses of radiotherapy-only HNSCC clinical trials data using mechanistically-motivated models of tumor repopulation and killing by radiotherapy predicted that hyperfractionation with twice-daily fractions, or hypofractionation involving increased doses/fraction and reduced treatment durations, both improve tumor control and reduce late normal tissue toxicity, compared with standard protocols using 35×2 Gy. Here we further investigated the validity of these conclusions by analyzing a large modern dataset on 3,346 HNSCC radiotherapy patients from the University Health Network in Toronto, Canada, where 42.5% of patients were also treated with chemotherapy.
We used a two-step approach that combines mechanistic modeling concepts with state-of-the-art machine learning, beginning with Random Survival Forests (RSF) for an exploratory analysis and followed by Causal Survival Forests (CSF) for a focused causal analysis. The mechanistic concept of biologically effective dose (BED) was implemented for the standard dose-independent (DI) tumor repopulation model, our alternative dose-dependent (DD) repopulation model, and a simple model with no repopulation (BED). These BED variants were included in the RSF model, along with age, stage, HPV status and other relevant variables, to predict patient overall survival (OS) and cause-specific mortality (deaths from the index cancer, other cancers or other causes).
Model interpretation using Shapley Additive Explanations (SHAP) values and correlation matrices showed that high values of BED or BED, but not BED, were associated with decreased patient mortality. Targeted causal inference analyses were then performed using CSF to estimate the causal effect of each BED variant on OS. They revealed that high BED (>61.8 Gy) or BED (>57.6 Gy), but not BED, increased patient restricted mean survival time (RMST) by 0.5-1.0 years and increased survival probability (SP) by 5-15% several years after treatment. In addition to population-level averages, CSF generated individual-level causal effect estimates for each patient, facilitating personalized medicine.
These findings are generally consistent with those of our previous mechanistic modeling, implying the potential benefits of altered radiotherapy fractionation schemes ( 25×2.4 Gy, 20×2.75 Gy, 18×3.0 Gy) which increase BED and BED and counteract tumor repopulation more effectively than standard fractionation. Such regimens may represent potentially useful hypofractionated options for treating HNSCC.
治疗头颈部鳞状细胞癌（HNSCC），尤其是人乳头瘤病毒阴性（HPV-）和局部晚期病例，仍然具有挑战性。我们之前使用基于机制的肿瘤再增殖模型和放射治疗杀伤模型，对仅采用放射治疗的HNSCC临床试验数据进行分析，预测与使用35×2 Gy的标准方案相比，每日两次分割的超分割放疗，或涉及增加每次分割剂量和缩短治疗疗程的低分割放疗，均可提高肿瘤控制率并降低晚期正常组织毒性。在此，我们通过分析来自加拿大多伦多大学健康网络的3346例HNSCC放疗患者的大型现代数据集，进一步研究这些结论的有效性，其中42.5%的患者还接受了化疗。
我们采用了一种两步法，将机制建模概念与最先进的机器学习相结合，首先使用随机生存森林（RSF）进行探索性分析，然后使用因果生存森林（CSF）进行重点因果分析。将生物学有效剂量（BED）的机制概念应用于标准剂量无关（DI）肿瘤再增殖模型、我们的替代剂量依赖（DD）再增殖模型以及无再增殖的简单模型（BED）。这些BED变体与年龄、分期、HPV状态和其他相关变量一起纳入RSF模型，以预测患者的总生存期（OS）和特定病因死亡率（死于指数癌症、其他癌症或其他原因）。
使用Shapley加性解释（SHAP）值和相关矩阵进行的模型解释表明，高BED或BED值（而非BED值）与患者死亡率降低相关。然后使用CSF进行靶向因果推断分析，以估计每个BED变体对OS的因果效应。结果显示，高BED（>61.8 Gy）或BED（>57.6 Gy）（而非BED）可使患者受限平均生存时间（RMST）增加0.5 - 1.0年，并使治疗后数年的生存概率（SP）提高5 - 15%。除了群体水平的平均值外，CSF还为每位患者生成了个体水平的因果效应估计值，有助于个性化医疗。
这些发现总体上与我们之前的机制建模结果一致，这意味着改变放疗分割方案（25×2.4 Gy、20×2.75 Gy、18×3.0 Gy）可能具有潜在益处，这些方案比标准分割更有效地增加了BED和BED，并更有效地抵消肿瘤再增殖。此类方案可能代表了治疗HNSCC潜在有用的低分割选择。
