DNA methylation classifier to diagnose pancreatic ductal adenocarcinoma metastases from different anatomical sites

Teodor G Calina ^1,2, Eilís Perez ^3,4, Elena Grafenhorst ⁵

¹TGC Ventures UG, Berlin, Germany.
²Faculty of Physics, Babeș-Bolyai University, Cluj-Napoca, Romania.
³Department of Neuropathology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin, Germany.
⁴Berlin School of Integrative Oncology (BSIO), Charite - Universitätsmedizin Berlin (CVK), Berlin, Germany.
⁵Institute of Pathology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Charitéplatz 1, 10117, Berlin, Germany.
⁶Department of Pathology and Laboratory Medicine, Memorial Sloan Kettering Cancer Center, New York, NY, USA.
⁷German Cancer Consortium (DKTK), Partner Site Berlin, and German Cancer Research Center (DKFZ), Heidelberg, Germany.
⁸State Key Laboratory of Oncology in South China, Department of Radiation Oncology, Collaborative Innovation Center of Cancer Medicine, Sun Yat-sen University Cancer Center, Guangzhou, Guangdong, People's Republic of China.
⁹Guangdong Esophageal Cancer Research Institute, Guangzhou, Guangdong, People's Republic of China.
¹⁰Berlin Institute of Health, Berlin, Germany.
¹¹Institute of Pathology, Heidelberg University Hospital, Heidelberg, Germany.
¹²Institute of Pathology and Neuropathology, Hospital RKH Kliniken Ludwigsburg, Ludwigsburg, Germany.
¹³Institute of Tissue Medicine and Pathology (ITMP), University Bern, Bern, Switzerland.
¹⁴Department of Surgical Oncology and Robotics, Krankenhaus Waldfriede, Lehrkrankenhaus der Charité, Berlin, Germany.
¹⁵Department of Translational Molecular Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁶The RNA Interference and Non-Coding RNA Center, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
¹⁷Department of Hematology, Oncology and Tumor Immunology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Berlin, Germany.
¹⁸Department of Medical Biology, Faculty of Health Sciences, UiT The Artic University of Norway, Tromsø, Norway.
¹⁹Institute of Pathology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität zu Berlin, Charitéplatz 1, 10117, Berlin, Germany. mihnea.dragomir@charite.de.
²⁰German Cancer Consortium (DKTK), Partner Site Berlin, and German Cancer Research Center (DKFZ), Heidelberg, Germany. mihnea.dragomir@charite.de.
²¹Berlin Institute of Health, Berlin, Germany. mihnea.dragomir@charite.de.

⁰TGC Ventures UG, Berlin, Germany.

Clinical Epigenetics +

|

November 10, 2024

View abstract on PubMed

Summary

A new DNA methylation classifier accurately identifies pancreatic ductal adenocarcinoma (PAAD) metastases from various sites. This tool aids in diagnosing PAAD, improving patient outcomes for cancers of unknown primary origin.

Area Of Science

Oncology
Genomics
Molecular Diagnostics

Background

Pancreatic ductal adenocarcinoma (PAAD) is a leading cause of cancer of unknown primary, often diagnosed by exclusion.
Accurate classification of PAAD metastases is crucial for effective treatment and patient management.
Existing diagnostic methods can be challenging, necessitating novel approaches for precise identification.

Purpose Of The Study

To evaluate the efficacy of an updated DNA methylation classifier (PAAD-iCCA-Classifier) in identifying PAAD metastases from diverse anatomical locations.
To determine if the classifier can distinguish PAAD metastases from other carcinoma types, including intrahepatic cholangiocarcinoma (iCCA).
To assess the classifier's performance on both biological and technical validation cohorts.

Main Methods

The PAAD-iCCA-Classifier was enhanced by incorporating 8 additional mimicker carcinomas into its negative class, totaling 10.
The updated classifier was validated on a large biological cohort (n=3579) and a technical cohort (n=15).
Performance was further assessed on a test set comprising 16 PAAD metastases (positive control) and 124 other carcinoma metastases (negative control).

Main Results

The updated classifier achieved high accuracy: 98.21% on biological validation and 100% on technical validation.
It correctly identified 93.75% of PAAD metastases and 98.39% of negative control samples, with an overall test set accuracy of 97.85%.
Analysis revealed distinct molecular characteristics of PAAD liver metastases compared to primary PAAD and peritoneal carcinomatosis.

Conclusions

The updated PAAD-iCCA-Classifier demonstrates high accuracy in classifying PAAD samples from various metastatic sites.
This classifier can serve as a valuable diagnostic aid, particularly for cases of cancer of unknown primary.
The findings support the use of DNA methylation profiling for precise cancer diagnosis and understanding metastasis patterns.

Keywords:

Cancer of unknown primary DNA methylation Epigenetics Molecular diagnosis Pancreatic ductal adenocarcinoma