much better as treatments are more effective. But how do you identify patients with early-stage tumors who may not even be symptomatic? Researchers at Cancer Center Amsterdam are developing a non-invasive tumor detection test based on the
analysis of blood samples, together with international collaborators. A new report published in Cancer Cell demonstrates their liquid biopsy technique, thromboSeq, has the potential to detect up to 18 different cancer types.
The thromboSeq technology was previously developed in the lab of Prof. Tom Würdinger. It is based on AI-guided sequence analysis of mRNA found in platelets. While platelets, the smallest of our blood cells, are generally known for their role in blood clotting, they have many other functions as well, including involvement in immune responses and cancer progression.
Although platelets do not contain a nucleus, they possess RNA and the machinery to splice and express thousands of mRNAs. Importantly, platelets isolated from cancer patients have a population with a distinct phenotype: they contain tumor cell-derived RNA molecules. Termed ‘tumor-educated-platelets’ (TEP), this discovery opened the door to using platelets as a biomarker for the detection of cancer.
Getting it right
The current study presents the results of a large-scale retrospective evaluation study of this method in which blood samples from more than 2300 people were examined. The blood samples were collected in a national and international collaboration involving more than 100 scientists from 13 research institutes.
Among 1,628 individuals with a positive cancer diagnosis, the thromboSeq correctly predicted the presence of cancer in about two-thirds of more than 1,000 stage I-IV cancer patients and in half of 352 stage I-III tumors. Moreover, thromboSeq determined the tumor site of origin in five different tumor types correctly in over 80% of the cancer patients.
Data obtained with the refined thromboSeq liquid biopsy test revealed that 99% of healthy controls were correctly identified as not having cancer.
Sjors in ‘t Veld, PhD candidate and first author, explains: “For a cancer detection screen, it is paramount that false positive test results are prevented as much as possible, as this leads to unnecessary worry and clinical follow-up.”
Cancer detection in a blood sample
The correct identification of the early-stage tumors shows the potential of thromboSeq for early cancer detection to improve treatment outcome. In addition, two tumor types were identified among the screened samples that were not part of the Artificial Intelligence training set.
“This underlines that we have developed a truly universal TEP RNA profile for cancer detection in general, which could also facilitate the diagnosis of very rare tumors,” says Dr. Myron Best. “However, when our thromboSeq test indicates the presence of a tumor, this still must be confirmed by other diagnostic tests, such as an imaging scan or tissue biopsy. Since the possible origin and location of a tumor was predicted in over 80% of cases, this can guide follow-up diagnostic tests.”
Further development is still needed
At this time, the thromboSeq test is still in a preclinical development phase. Prof. Würdinger: “Platelet RNA patterns also change in patients with other conditions, such as multiple sclerosis and pulmonary hypertension. In the current study, we see that patients with these conditions are more often labeled as false-positive. This observation is important as it offers opportunities to improve thromboSeq’s specificity.”
Sjors in ‘t Veld: “Although liquid biopsy cancer screening is either still at a development stage or being tested in clinical trials, it is our anticipation that more and more blood tests will be developed and marketed for the early detection of cancer in the near future.”
Liquid Biopsy Center
The Liquid Biopsy Center at Amsterdam UMC plays a pivotal role in collecting blood samples that enable development of non-invasive diagnostic assays to detect disease. Already for several years liquid biopsies have been collected and stored for research purposes like the development of thromboSeq.
In ’t Veld, S.G.J.G., et al. (2022) Detection and localization of early- and late-stage cancers using platelet RNA. Cancer Cell. https://doi.org/10.1016/j.ccell.2022.08.006
Funding for this work was provided by: European Research Council (ERC), the Dutch Organisation for Scientific Research (NWO), Stichting STOPHersentumoren.nl, the Dutch Cancer Society (KWF), the Bennink Foundation, a Marie Sklodowska-Curie grant by the European Union’s Horizon 2020 program, the National Science Centre, and the Medical University of Gdansk statutory, the Netherlands Cardiovascular Research Initiative, an initiative supported by the Dutch Heart Foundation, DOLPHIN-GENESIS, and KWF/Alpe d’Huzes.
Researchers involved: Sjors G.J.G. In ’t Veld, Mohammad Arkani, Edward Post, Mafalda Antunes-Ferreira, Silvia D’Ambrosi, Daan C.L. Vessies#, Lisa Vermunt, Adrienne Vancura, Mirte Muller#, Anna-Larissa N. Niemeijer, Jihane Tannous#, Laura L. Meijer, Tessa Y.S. Le Large, Giulia Mantini, Niels E. Wondergem, Kimberley M. Heinhuis#,Sandra van Wilpe#, A. Josien Smits, Esther E.E. Drees, Eva Roos, Cyra E. Leurs, Lee-Ann Tjon Kon Fat#, Ewoud J. van der Lelij, Govert Dwarshuis, Maarten J. Kamphuis, Lisanne E. Visser, Romee Harting, Annemijn Gregory, Markus W. Schweiger, Laurine E. Wedekind, Jip Ramaker, Kenn Zwaan, Heleen Verschueren, Idris Bahce, Adrianus J. de Langen#, Egbert F. Smit#, Michel M. van den Heuvel#, Koen J. Hartemink#, Marijke J.E. Kuijpers#, Mirjam G.A. oude Egbrink#, Arjan W. Griffioen, Rafael Rossel#, T. Jeroen N. Hiltermann#, Elizabeth Lee-Lewandrowski#, Kent B. Lewandrowski#, Philip C. De Witt Hamer, Mathilde Kouwenhoven, Jaap C. Reijneveld, William P.J. Leenders#, Ann Hoeben#, Irma M. Verdonck-de Leeuw, C. René Leemans, Robert J. Baatenburg de Jong#,Chris H.J. Terhaard#, Robert P. Takes#, Johannes A. Langendijk#, Saskia C. de Jager#, Adriaan O. Kraaijeveld#, Gerard Pasterkamp#, Minke Smits, Jack A. Schalken#, Sylwia Łapińska-Szumczyk#, Anna Łojkowska#, Anna J. Żaczek#, Henk Lokhorst, Niels W.C.J. van de Donk, Inger Nijhof, Henk-Jan Prins, Josée M. Zijlstra, Sander Idema, Johannes C. Baayen, Charlotte E. Teunissen, Joep Killestein, Marc G. Besselink, Lindsay Brammen, Thomas Bachleitner-Hofmann#, Farrah Mateen#, John T.M. Plukker#, Michal Heger#, Quirijn de Mast#, Ton Lisman#, D. Michiel Pegtel, Harm-Jan Bogaard, Jacek Jassem#, Anna Supernat#, Niven Mehra#, Winald Gerritsen#, Cornelis D. de Kroon#, Christianne A.R. Lok#, Jurgen M.J. Piek#, Neeltje Steeghs#, Winan J. van Houdt#, Ruud H. Brakenhoff, Gabe S. Sonke, Henk M. Verheul#, Elisa Giovannetti, Geert Kazemier, Siamack Sabrkhany#, Ed Schuuring#, Erik A. Sistermans, Rob Wolthuis, Hanne Meijers-Heijboer, Josephine Dorsman, Cees Oudejans, Bauke Ylstra, Bart A. Westerman, Daan van den Broek#, Danijela Koppers-Lalic, Pieter Wesseling, R. Jonas A. Nilsson#, W. Peter Vandertop, David P. Noske, Bakhos A. Tannous#, Nik Sol, Myron G. Best, and Thomas Wurdinger.
#: scientist not affiliated with Amsterdam UMC.
Text adapted by Henri van de Vrugt.
This article was created for Amsterdam UMC - Cancer Center Amsterdam.
©2022 - NHBC - Your partner in science communication.