This study was designed as an exploratory single-center study, approved by the Ethics Committee of the Saarland Medical Association (votum number 73/20) and performed in accordance with the Declaration of Helsinki. Written informed consent was obtained from all participants.

Study population

Patients with newly diagnosed histologically confirmed GBM WHO grade 4 over 18 years of age with no additional known cancerous disease or pregnancy that underwent tumor resection were eligible for study inclusion.

Study samples

Glioblastoma tumor tissue was obtained during tumor resection. Blood samples were drawn from an arterial or intravenous line intraoperatively before opening of the dura. Reference blood samples from healthy donors were previously drawn and analyzed again in the present study [14]. Histopathological diagnosis, as well as O-6-methylguanine-DNA methyltransferase (MGMT) promoter- and isocitrate dehydrogenase 1 (IDH1) mutation status were compiled by the Institute of Neuropathology of the Saarland University Medical Center.

Primary glioblastoma cell cultures

Cultivation of primary glioblastoma cell cultures was performed according to the standard protocol of our group [29, 30]. The tumor tissue was minced with forceps and scissors and suspended with Dulbecco’s Modified Eagle Medium (DMEM; GIBCO®, Life Technologies, Darmstadt, Germany) containing 10% fetal calf serum, 1% non-essential amino acids and 1% penicillin/streptomycin. The cell suspension was distributed to cell culture flasks and placed in a CO2 incubator at 37 °C with 5% CO2 for 24 h. A change of the cell culture medium was performed twice per week. After the formation of a cell monolayer was registered, a change of cell culture medium was carried out every two to three days and incubation was continued in a CO2 incubator at 37°C with 5% CO2. The cell cultures were split when a confluence cell monolayer in the flasks was reached. An early passage (1 or 2) was utilized for further experiments.

Assessment of cell viability

When a confluent cell layer of 70–90% per flask was reached, cells were trypsinized and evaluated regarding cell count and cell viability utilizing the LUNA 2™ Automated Cell Counter (BioCat GmbH, Heidelberg, Germany) in conjunction with Trypan Blue staining. For every sample, 10 µl of cell suspension were examined.

Application of TTFields in vitro

TTFields were applied to the primary glioblastoma cell cultures with the inovitro™ TTFields Lab Bench System (Novocure, Haifa, Israel) according to a modified protocol after Porat et al. [31]. In brief, 500 μl of the cell culture suspension containing 10.000–20.000 cells were distributed into 24 specifically designed ceramic petri dishes for the inovitro™ system each containing a 22 mm glass coverslip, before 500 µl of DMEM were added. The petri dishes were placed in a CO2 incubator at 37 °C with 5 % CO2 for 24 h, before the culture medium was removed and replaced with 2 ml of fresh DMEM. Sterilized parafilm was then used as coverage for all petri dishes. The ceramic dishes were placed onto a base plate connected to a TTFields generator to be treated with TTFields in a CO2 incubator with 5 % CO2. Due to heat development from TTFields, the CO2 incubator was refrigerated to 18 °C, with a stable temperature of 37 °C within the petri dishes. TTFields were applied with the optimal TTFields frequency of 200 kHz for GBM treatment and a low intensity of 2 V/cm continuously for 72 h. For every tumor, control cultures were prepared identically to TTFields-treated cultures, but were placed in a separate CO2 incubator at 37°C with 5 % CO2 for 72 h without TTFields-treatment. After 72 h, treated and control cell cultures were trypsinized and evaluated for cell viability as described before.

Evaluation of TTFields treatment response

After 72 h of TTFields treatment, absolute cell count and cell viability were assessed again as described before. The relative individual treatment response was calculated based on the cell viability in the treated tumor cell cultures and the corresponding control cultures using the following formula:

$$Relative Reduction of Viability = \frac Viability TTFields Culture \left( \% \right)}} x 100$$

MicroRNA analysis

MicroRNA (miRNA) expression levels of miRNAs-21, -26a, -34a, -181c, -181d and -485-5p were assessed as relative expression in corresponding native tumor tissue, TTFields-treated cell cultures, untreated control cell cultures and blood samples.

Plasma miRNAs were isolated using the miRNeasy Serum Plasma Kit (Qiagen, Hilden, Germany) based on the manufacturer’s instructions, while isolation of miRNA from native tumor tissue and cell cultures was conducted with the miRNeasy Mini kit (Qiagen, Hilden, Germany). RNA quantification was performed with the NanoDrop spectrophometer (Thermo Fisher Scientific inc. Kandel, Germany). All RNA was diluted to a concentration of 20 ng/µl.

First, reverse transcription was conducted using the TaqMan™ MicroRNA Reverse Transcription Kit (Applied Biosystems Life Technologies, Darmstadt, Germany) along with MiRNA-specific Stem-Loop-Primers (TaqMan™ Small RNA Assay, Applied Biosystems Life Technologies, Darmstadt, Germany), adhering to the manufacturer’s guidelines. A master mix was prepared for each reaction, comprising 4.16 μl of PCR Water, 1.5 μl of RT-Buffer, 1 μl of Multiscribe™ Reverse Transcriptase, 0.19 μl of RNase Inhibitor, and 0.15 μl of dNTP Mix (All TaqMan™ MicroRNA Reverse Transcription Kit, Applied Biosystems™, Life Technologies, Darmstadt, Germany). For each reaction, 5 µl of diluted isolated RNA, along with 3 µl of the specific miRNA Stem-Loop-primer, were combined and centrifuged for 10 to 20 s. Subsequently, the samples underwent incubation in a thermal cycler (PTC-200, MJ Research, Hessisch Oldendorf, Germany) following this protocol: 30 min at 16 °C, 30 min at 42 °C, and 5 min at 85 °C. Post-incubation, the samples were stored at − 20 °C until further processing.

The quantitative PCR (qPCR) was executed using the Taqman™ Gene Expression Mastermix (Applied Biosystems™, Life Technologies, Darmstadt, Germany) and fluorescence labeled miRNA-specific Taqman™ miRNA assay primers were used according to the manufacturer’s instructions. Seven primer-specific master mixes were prepared (one for each miRNA-primer, one for RNU48), each consisting of 5 μl of Taqman ™ Gene Expression Mastermix, 0.5 μl of Taqman ™ miRNA primer, and 3.5 μl of PCR-grade water per sample. The qPCR was performed with the StepOne System (Applied Biosystems™, Life Technologies, Darmstadt, Germany). All reactions were executed in triplicate along with a negative control, using RNU48 as an internal reference as described before [14].

The expression of miRNA was normalized to the RNU48 expression in the corresponding tumor, plasma and cell culture sample, respectively. The quantification of miRNA expression was conducted by calculating the relative expression with the comparative 2-(ΔCT) method as follows:

$$\begin & Relative expression tumor tissue = 2^ \right) Ct \left( \right)} \right)}} \\ & Relative expression plasma = 2^ \right) Ct \left( \right)} \right)}} \\ & Relative expression TTFields cell culture = 2^ \right) Ct \left( \right)} \right)}} \\ & Relative expression control cell culture = 2^ \right) Ct \left( \right)} \right)}} \\ \end$$

Statistical analysis

Statistical analysis was carried out with GraphPad Prism 9.5 (GraphPad Software, Boston, Massachusetts, USA). Shapiro Wilk Normality test was used to evaluate Gaussian’ distribution. Kruskal-Wallis test was used to determine significance of differences of cell viability between cell cultures before and after TTFields treatment and control cultures, respectively. Correlation was evaluated using Spearman’s rank correlation coefficient and linear regression analysis. Mann Whitney U test was carried out to determine differences in miRNA expression levels of patient plasma samples and healthy donor plasma samples. A p-value < 0.05 was considered statistically significant.