Predictive Modeling for Voxel-Based Quantification of Imaging-Based Subtypes of Pancreatic Ductal Adenocarcinoma (PDAC): A Multi-Institutional Study.
Predictive Modeling for Voxel-Based Quantification of Imaging-Based Subtypes of Pancreatic Ductal Adenocarcinoma (PDAC): A Multi-Institutional Study.
复制标题
DOI：
10.3390/cancers12123656
复制
发表时间：
2020-12-05
期刊：
影响因子：
5.2
通讯作者：
Koay EJ
中科院分区：
文献类型：
--
作者：
Zaid M;Widmann L;Dai A;Sun K;Zhang J;Zhao J;Hurd MW;Varadhachary GR;Wolff RA;Maitra A;Katz MHG;Herman JM;Wang H;Knopp MV;Williams TM;Bhosale P;Tamm EP;Koay EJ
之前，我们证明了胰腺导管腺癌（PDAC）肿瘤在计算机断层扫描（CT）扫描（delta）上的定性评分具有生物学和临床相关性，因此与边界不明显（低delta）的肿瘤相比，边界明显（高delta）的肿瘤显示出更具侵袭性的生物学，并且与更差的临床结局相关。然而，在某些情况下，视觉分类可能具有挑战性和主观性。在这里，我们使用机器学习和多机构数据集的定量方法来构建生物学和临床相关模型，该模型可以从常规CT扫描中定量识别这些基于成像的PDAC亚型。我们的研究结果表明，定量分类（q-δ）与内部和外部数据集的金标准定性评分具有高度相关性。此外，q-δ分类被证明与PDAC肿瘤的临床结果和基质异质性相关。较高的评价者内和评价者间一致性分数表明结果的可重复性。以前，我们的特点是定性成像为基础的亚型胰腺导管腺癌（PDAC）的计算机断层扫描（CT）。与不明显的（低δ）肿瘤相比，明显的（高δ）PDAC肿瘤更可能具有侵袭性生物学和较差的临床结局。在这里，我们开发了这种基于成像的亚型的定量分类（定量delta; q-delta）。将PDAC患者的三个队列（队列1 = 101，队列2 = 90和队列3 = 16 [外部验证]）的逆行基线胰腺方案CT扫描定性分类为高和低Δ。我们使用基于体素的方法来定量肿瘤增强，同时参考正常胰腺实质，并使用基于机器学习的分析来建立预测模型。此外，我们使用苏木精和伊红染色的未经治疗的PDAC切片定量基质含量。分析显示，PDAC定量增强值可预测定性delta评分，并用于构建分类模型（q-delta）。与高q-δ相比，低q-δ肿瘤与预后改善相关，q-δ分类是生存的独立预后因素。此外，与高q-δ肿瘤相比，低q-δ肿瘤具有较高的基质含量和较低的细胞结构。我们的研究结果表明，q-δ分类提供了一个临床和生物学相关的工具，可以整合到正在进行的和未来的临床试验。
Previously, we demonstrated that qualitative scoring of pancreatic ductal adenocarcinoma (PDAC) tumors on computed tomography (CT) scans (delta) is biologically and clinically relevant, whereby tumors with a conspicuous border (high delta) show more aggressive biology and are associated with worse clinical outcomes when compared to those with an inconspicuous border (low delta). However, in some cases, a visual classification can be challenging and subjective. Here, we used machine learning and quantitative approaches for a multi-institutional dataset to build a biologically and clinically relevant model that can quantitatively identify these imaging-based subtypes of PDAC from routine CT scans. Our results showed that the quantitative classification (q-delta) had high correlation with the gold standard qualitative scoring in internal and external datasets. Further, q-delta classification was demonstrated to be associated with the clinical outcome and the stromal heterogeneity of the PDAC tumors. High intra- and interrater agreement scores indicate the reproducibility of the results. Previously, we characterized qualitative imaging-based subtypes of pancreatic ductal adenocarcinoma (PDAC) on computed tomography (CT) scans. Conspicuous (high delta) PDAC tumors are more likely to have aggressive biology and poorer clinical outcomes compared to inconspicuous (low delta) tumors. Here, we developed a quantitative classification of this imaging-based subtype (quantitative delta; q-delta). Retrospectively, baseline pancreatic protocol CT scans of three cohorts (cohort#1 = 101, cohort#2 = 90 and cohort#3 = 16 [external validation]) of patients with PDAC were qualitatively classified into high and low delta. We used a voxel-based method to volumetrically quantify tumor enhancement while referencing normal-pancreatic-parenchyma and used machine learning-based analysis to build a predictive model. In addition, we quantified the stromal content using hematoxylin- and eosin-stained treatment-naïve PDAC sections. Analyses revealed that PDAC quantitative enhancement values are predictive of the qualitative delta scoring and were used to build a classification model (q-delta). Compared to high q-delta, low q-delta tumors were associated with improved outcomes, and the q-delta class was an independent prognostic factor for survival. In addition, low q-delta tumors had higher stromal content and lower cellularity compared to high q-delta tumors. Our results suggest that q-delta classification provides a clinically and biologically relevant tool that may be integrated into ongoing and future clinical trials.
登录
查看更多内容
影响因子：
6.2
作者：
Amer AM;Zaid M;Chaudhury B;Elganainy D;Lee Y;Wilke CT;Cloyd J;Wang H;Maitra A;Wolff RA;Varadhachary G;Overman MJ;Lee JE;Fleming JB;Tzeng CW;Katz MH;Holliday EB;Krishnan S;Minsky BD;Herman JM;Taniguchi CM;Das P;Crane CH;Le O;Bhosale P;Tamm EP;Koay EJ
通讯作者：
Koay EJ
影响因子：
3.7
作者：
Katz MH;Varadhachary GR;Fleming JB;Wolff RA;Lee JE;Pisters PW;Vauthey JN;Abdalla EK;Sun CC;Wang H;Crane CH;Lee JH;Tamm EP;Abbruzzese JL;Evans DB
通讯作者：
Evans DB
影响因子：
45.3
作者：
Evans, Douglas B.;Varadhachary, Gauri R.;Wolff, Robert A.
通讯作者：
Wolff, Robert A.
影响因子：
5.9
作者：
Chockalingam, Arun;Duran, Rafael;Lin, MingDe
通讯作者：
Lin, MingDe
影响因子：
4.4
作者：
Clark, Kenneth;Vendt, Bruce;Prior, Fred
通讯作者：
Prior, Fred