Abstract
Abstract Intro. Glioblastoma is the most common and aggressive adult brain cancer, with a 5-year survival of less than 15%. Treatment decisions are guided by imaging, which at early timepoints cannot distinguish tumor pseudoprogression from non-response. Our work seeks to identify circulating biomarkers that can classify response as an alternative or adjunct to imaging. Here we present preliminary data exploring the potential of circulating cytokines and tumor cells collected before, during, and after treatment with the goal to identify potential makers for treatment adaptation. Methods. After consent, patient blood was collected the week before initiating chemoradiotherapy, weekly during treatment, and three weeks after treatment. Serum was tested by multiplex bead-based immunoassay (Legendplex Human Inflammation Panel 1, Biolegend inc.) targeting 13 cytokines and ELISA targeting soluble Human GFAP (Glial fibrillary acidic protein) (Biovendor R&D). The Circulogix filtration system was also used to capture cells of size consistent with circulating tumor cells (CTCs). Initial efforts focused on cells co-staining for GFAP and Olig2 and negative for CD45 (hematopoietic marker). All patient tumor specimens and cell lines were verified for GFAP and Olig2 expression by IHC. We selected 13 patients for the current analysis: 5 with growth due to non-response (NR), 6 stable disease (S) and two pseudo progression (PP). These classifications were evaluated radiologically at the first follow up after chemoradiotherapy (S or no) and based on whether changes observed stabilized or resolved spontaneously within six months or continued to progress (NR or PP). Results. The immunoassays detected approximately 70% of the analytes in our samples. Among those, IL10 and IL33 were higher before treatment on patients later determined to have stable disease (IL10 p=0.09; IL33 p=0.043 KW test; IL10 median S=22.93 ng/mL, NR=13.415ng/mL; IL33 median S= 164.10ng/mL, NR=55.57ng/mL). Interestingly, IL33 is reduced on stable patients during week 5 of treatment (PRE vs. W5 p=0.033; median PRE=164.1ng/mL, W5=57.57ng/mL). GFAP presented a slightly lower level after treatment in stable patients (S vs. NR p=0.014; median S=0.441 ng/ml, NR=0.473 ng/ml). As proof of concept, 50 GBM cells from each of U87 adherent cells, or 0913 and 0821 neurosphere cells (CD45 negative) glioblastoma cell lines were mixed in separate vials with 106 hematopoietic cells (RAJI; CD45 positive) and filtered per the procedure to identify CTCs in patient blood. We separated GBM cells from hematopoietic cells on the filtration system. Conclusion. Our preliminary data suggest that blood obtained during chemoradiotherapy can identify biomarkers associated with response. For example, IL10, IL33, GFAP, and CTCs might be used to classify treatment response at very early timepoints. We continue to recruit additional patients for additional analyses. Citation Format: Paulo Roberto Del Valle, Jonathan B. Bell, Scott M. Welford, Kaylie Kullison, Alessandro Valderrama, Gregory A. Azzam, Jessica J. Meshman, Macarena I. De La Fuente, Eric A. Mellon. Preliminary analysis of circulating biomarkers of glioblastoma response during chemoradiotherapy [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2023; Part 1 (Regular and Invited Abstracts); 2023 Apr 14-19; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2023;83(7_Suppl):Abstract nr 2162.
