Lawrence Livermore National Laboratory, Livermore, CA, United States of America.
Cancer Registry of Norway, Oslo, Norway.
PLoS One. 2020 Nov 16;15(11):e0241225. doi: 10.1371/journal.pone.0241225. eCollection 2020.
Oncology is a highly siloed field of research in which sub-disciplinary specialization has limited the amount of information shared between researchers of distinct cancer types. This can be attributed to legitimate differences in the physiology and carcinogenesis of cancers affecting distinct anatomical sites. However, underlying processes that are shared across seemingly disparate cancers probably affect prognosis. The objective of the current study is to investigate whether multitask learning improves 5-year survival cancer patient survival prediction by leveraging information across anatomically distinct HPV related cancers. Data were obtained from the Surveillance, Epidemiology, and End Results (SEER) program database. The study cohort consisted of 29,768 primary cancer cases diagnosed in the United States between 2004 and 2015. Ten different cancer diagnoses were selected, all with a known association with HPV risk. In the analysis, the cancer diagnoses were categorized into three distinct topography groups of varying specificity. The most specific topography grouping consisted of 10 original cancer diagnoses differentiated by the first two digits of the ICD-O-3 topography code. The second topography grouping consisted of cancer diagnoses categorized into six distinct organ groups. Finally, the third topography grouping consisted of just two groups, head-neck cancers and ano-genital cancers. The tasks were to predict 5-year survival for patients within the different topography groups using 14 predictive features which were selected among descriptive variables available in the SEER database. The information from the predictive features was shared between tasks in three different ways, resulting in three distinct predictive models: 1) Information was not shared between patients assigned to different tasks (single task learning); 2) Information was shared between all patients, regardless of task (pooled model); 3) Only relevant information was shared between patients grouped to different tasks (multitask learning). Prediction performance was evaluated with Brier scores. All three models were evaluated against one another on each of the three distinct topography-defined tasks. The results showed that multitask classifiers achieved relative improvement for the majority of the scenarios studied compared to single task learning and pooled baseline methods. In this study, we have demonstrated that sharing information among anatomically distinct cancer types can lead to improved predictive survival models.
肿瘤学是一个高度专业化的研究领域，亚学科专业化限制了不同癌症类型的研究人员之间信息的共享。这可以归因于影响不同解剖部位的癌症在生理学和癌变方面的合理差异。然而，跨看似不同的癌症共享的潜在过程可能会影响预后。本研究的目的是调查是否通过利用跨解剖学上不同的 HPV 相关癌症的信息，多任务学习可以提高 5 年生存率癌症患者的生存预测。数据来自监测、流行病学和最终结果 (SEER) 计划数据库。研究队列包括 2004 年至 2015 年期间在美国诊断的 29768 例原发性癌症病例。选择了十种不同的癌症诊断，所有这些诊断都与 HPV 风险有关。在分析中，癌症诊断被分为三个不同的特定解剖部位分组，特异性逐渐增强。最特定的解剖部位分组由 10 种原始癌症诊断组成，这些诊断通过 ICD-O-3 解剖部位代码的前两位数字区分。第二个解剖部位分组由分为六个不同器官组的癌症诊断组成。最后，第三个解剖部位分组只有两组，头颈部癌症和肛门生殖器癌症。任务是使用 14 个预测特征预测不同解剖部位组的患者的 5 年生存率，这些特征是从 SEER 数据库中可用的描述性变量中选择的。任务之间通过三种不同的方式共享预测特征的信息，从而产生三个不同的预测模型：1）不同任务分配的患者之间不共享信息（单任务学习）；2）无论任务如何，所有患者之间都共享信息（池化模型）；3）仅在分配给不同任务的患者之间共享相关信息（多任务学习）。使用 Brier 分数评估预测性能。在每个三个不同的解剖部位定义的任务上，分别对所有三个模型进行评估。结果表明，与单任务学习和池化基线方法相比，多任务分类器在大多数研究场景中都实现了相对改善。在这项研究中，我们已经证明了在解剖学上不同的癌症类型之间共享信息可以导致改进的预测生存模型。
