Abstract

Abstract Background: Cancer diagnosis is frequently hampered by limited access to adequate tissue of the primary tumor or of metastatic lesions. To overcome such limitations, the use of blood-based liquid biopsies has been suggested. Blood represents a biosource of tumor-educated platelets (TEPs) that sequester biomolecules during tumor growth, thereby altering the platelet mRNA profile. Methods: Blood platelet samples of 175 patients with cancer covering five tumor types (40 non-small cell lung cancer, 39 glioblastoma, 37 colorectal cancer, 35 pancreatic cancer, and 24 breast cancer) and of 52 healthy donors were isolated from whole blood by differential centrifugation. Total RNA was isolated, subjected to SMARTer mRNA amplification and submitted for whole transcriptome mRNA sequencing on the Illumina platform. Healthy donors, pan-cancer, and individual cancer classes were distinguished by a self-learning support vector machine (SVM) algorithm, using transcripts with moderate to high expression. Results: The 227 blood platelet samples were successfully sequenced and demonstrated a good intersample correlation of the detected mRNAs. Based on mRNA profiles, all tumor samples were clearly distinguished from healthy individuals: the pan-cancer SVM-supported classification test reached a sensitivity of 97% and a specificity of 92% to distinguish cancer patients from healthy controls. Also, all patients without overt metastases were correctly predicted as cancer patients. Moreover, a multiclass cancer diagnostics TEP-test, to distinguish multiple tumor subclasses and healthy controls provided an overall accuracy of 70%, far exceeding random classification. In addition, we distinguished HER2-positive, and mutant KRAS and EGFR tumors from their wild-type counterparts. Also, patients with metastatic tumors in lung, brain, and liver were accurately diagnosed according to the tumor in the tissue of origin. Conclusion: Molecular interrogation of TEP-based liquid biopsies may leverage cancer diagnostics. TEPs provide a circulating biosource for pan-cancer, multiclass, and molecular cancer classification. Of interest, this tool might also allow for blood-based highly sensitive early-stage cancer screening. Citation Format: Myron Best, Nik Sol, Irsan Kooi, Jonas Nilsson, Bart Westerman, Bauke Ylstra, Josephine Dorsman, Egbert Smit, Henk Verheul, Jaap Reijneveld, Bakhos Tannous, Pieter Wesseling, Thomas Würdinger. Tumor-educated platelets allow for multiclass liquid biopsy-based diagnosis of cancer. [abstract]. In: Proceedings of the 106th Annual Meeting of the American Association for Cancer Research; 2015 Apr 18-22; Philadelphia, PA. Philadelphia (PA): AACR; Cancer Res 2015;75(15 Suppl):Abstract nr LB-124. doi:10.1158/1538-7445.AM2015-LB-124

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call