A Functional Spatial Analysis Platform for Discovery of Immunological Interactions Predictive of Low-Grade to High-Grade Transition of Pancreatic Intraductal Papillary Mucinous Neoplasms

被引:32
作者
Barua, Souptik [1 ,2 ]
Solis, Luisa [3 ]
Parra, Edwin Roger [3 ]
Uraoka, Naohiro [3 ]
Jiang, Mei [3 ]
Wang, Huamin [4 ]
Rodriguez-Canales, Jaime [3 ]
Wistuba, Ignacio [3 ]
Maitra, Anirban [4 ]
Sen, Subrata [3 ]
Rao, Arvind [1 ,2 ,5 ]
机构
[1] Rice Univ, Dept Elect & Comp Engn, POB 1892, Houston, TX 77251 USA
[2] Univ Michigan, Dept Computat Med & Bioinformat, 100 Washtenaw Ave, Ann Arbor, MI 48109 USA
[3] Univ Texas MD Anderson Canc Ctr, Dept Translat Mol Pathol, Houston, TX 77030 USA
[4] Univ Texas MD Anderson Canc Ctr, Dept Anat Pathol, Houston, TX 77030 USA
[5] Univ Michigan, Dept Radiat Oncol, 100 Washtenaw Ave, Ann Arbor, MI 48109 USA
关键词
IPMN; multiplex immunofluorescent (mIF) imaging; spatial statistics; functional data analysis; machine learning;
D O I
10.1177/1176935118782880
中图分类号
R73 [肿瘤学];
学科分类号
100214 [肿瘤学];
摘要
Intraductal papillary mucinous neoplasms (IPMNs), critical precursors of the devastating tumor pancreatic ductal adenocarcinoma (PDAC), are poorly understood in the pancreatic cancer community. Researchers have shown that IPMN patients with high-grade dysplasia have a greater risk of subsequent development of PDAC in the remnant pancreas than do patients with low-grade dysplasia. In this study, we built a computational prediction model that encapsulates the spatial cellular interactions in IPMNs that play key roles in the transformation of low-grade IPMN cysts to high-grade cysts en route to PDAC. Using multiplex immunofluorescent images of IPMN cysts, we adopted algorithms from spatial statistics and functional data analysis to create metrics that summarize the spatial interactions in IPMNs. We showed that an ensemble of models learned using these spatial metrics can robustly predict, with high accuracy, (1) the dysplasia grade (low vs high grade) and (2) the risk of a low-grade cyst progressing to a high-grade cyst. We obtained high classification accuracies on both tasks, with areas under the curve of 0.81 (95% confidence interval: 0.71-0.9) for task 1 and 0.81 (95% confidence interval: 0.7-0.94) for task 2. To the best of our knowledge, this is the first application of an ensemble machine learning approach for discovering critical cellular spatial interactions in IPMNs using imaging data. We envision that our work can be used as a risk assessment tool for patients diagnosed with IPMNs and facilitate greater understanding and investigation of the cellular interactions that cause transition of IPMNs to PDAC.
引用
收藏
页数:8
相关论文
共 36 条
[1]
spatstat: An R package for analyzing spatial point patterns [J].
Baddeley, A ;
Turner, R .
JOURNAL OF STATISTICAL SOFTWARE, 2005, 12 (06) :1-42
[2]
Baddeley A., 2008, ANALYSING SPATIAL PO
[3]
Spatial interaction of tumor cells and regulatory T cells correlates with survival in non-small cell lung cancer [J].
Barua, Souptik ;
Fang, Penny ;
Sharma, Amrish ;
Fujimoto, Junya ;
Wistuba, Ignacio ;
Rao, Arvind U. K. ;
Lin, Steven H. .
LUNG CANCER, 2018, 117 :73-79
[4]
A Revised Classification System and Recommendations From the Baltimore Consensus Meeting for Neoplastic Precursor Lesions in the Pancreas [J].
Basturk, Olca ;
Hong, Seung-Mo ;
Wood, Laura D. ;
Adsay, N. Volkan ;
Albores-Saavedra, Jorge ;
Biankin, Andrew V. ;
Brosens, Lodewijk A. A. ;
Fukushima, Noriyoshi ;
Goggins, Michael ;
Hruban, Ralph H. ;
Kato, Yo ;
Klimstra, David S. ;
Kloeppel, Guenter ;
Krasinskas, Alyssa ;
Longnecker, Daniel S. ;
Matthaei, Hanno ;
Offerhaus, G. Johan A. ;
Shimizu, Michio ;
Takaori, Kyoichi ;
Terris, Benoit ;
Yachida, Shinichi ;
Esposito, Irene ;
Furukawa, Toru .
AMERICAN JOURNAL OF SURGICAL PATHOLOGY, 2015, 39 (12) :1730-1741
[5]
Depletion of tumor-associated macrophages switches the epigenetic profile of pancreatic cancer infiltrating T cells and restores their anti-tumor phenotype [J].
Borgoni, Simone ;
Iannello, Andrea ;
Cutrupi, Santina ;
Allavena, Paola ;
D'Incalci, Maurizio ;
Novelli, Francesco ;
Cappello, Paola .
ONCOIMMUNOLOGY, 2018, 7 (02)
[6]
Random forests [J].
Breiman, L .
MACHINE LEARNING, 2001, 45 (01) :5-32
[7]
Functional approaches for predicting land use with the temporal evolution of coarse resolution remote sensing data [J].
Cardot, H ;
Faivre, R ;
Goulard, M .
JOURNAL OF APPLIED STATISTICS, 2003, 30 (10) :1185-1199
[8]
Spatial computation of intratumoral T cells correlates with survival of patients with pancreatic cancer [J].
Carstens, Julienne L. ;
de Sampaio, Pedro Correa ;
Yang, Dalu ;
Barua, Souptik ;
Wang, Huamin ;
Rao, Arvind ;
Allison, James P. ;
LeBleu, Valerie S. ;
Kalluri, Raghu .
NATURE COMMUNICATIONS, 2017, 8
[9]
Role of immune cells in pancreatic cancer from bench to clinical application An updated review [J].
Chang, Jae Hyuck ;
Jiang, Yongjian ;
Pillarisetty, Venu G. .
MEDICINE, 2016, 95 (49) :e5541
[10]
Intraductal Papillary Mucinous Neoplasm: Clinical Surveillance and Management Decisions [J].
Chin, Joanna Y. ;
Pitman, Martha B. ;
Hong, Theodore S. .
SEMINARS IN RADIATION ONCOLOGY, 2014, 24 (02) :77-84