Antitumor Activity of Radiation Therapy Combined With Checkpoint Kinase Inhibition in SHH/p53-Mutated Human Medulloblastoma

1. Introduction

Medulloblastoma (MB) is one of the most common malignant brain tumors in children with median age between five and nine years [1]. According to the 2021 WHO classification of CNS tumors, four molecular MB subgroups are defined differing in prognosis and theron adapted therapy regimes [2].

Sonic hedgehog-activated and TP53-mutant (SHH/p53-mut) MB is a high-risk subgroup showing enhanced therapy resistance and has one of the worst prognosis among the MB subgroups [3,4].

Current treatment for SHH/p53-mut MB patients consists of maximum safe surgical resection, TP53 mutation-dependent (somatic or germline) radiation therapy (RT) and chemotherapy (ChT) according to the SIOP-PNET5-MB-SHH-TP53 trial protocol [5]. High-dose-intensity regimes, RT as well as ChT, cause devastating (long-term) adverse effects, which might decrease the quality of life. Disruption of white matter developement and reduced hippocampal neurogenesis after high-intensity therapy are known biological causes for the neurological and sensory impairments, and endocrine deficits in adolescent patients (reviewed in [6]); [7].

Therefore, there is a urgent clinical need to identify new approaches for the treatment of high-risk MB improving patient outcome while minimizing adverse therapy effects. An appropriate strategy is to increase (tumor cell-specific) RT efficiency by inhibiting repair of RT-induced DNA damage (DD). As consequence, lower RT doses and/or prolonged OS could be achieved.

The unrestricted G1/S transition in TP53-mutated cells circumvents DNA repair in G1 and enforces it at later cell cycle stages [8]. Therefore the SHH/p53-mut MB subgroup is particularly vulnerable for DD repair blockade by G2/M checkpoint abrogation leading to DD accumulation and enhanced cell death. The checkpoint kinases Chk1 and Chk2 are responsible for DD-induced cell cycle arrest/prolongation allowing DNA repair [9]. Notably, Chk1 plays a crucial role in activating the G2/M checkpoint after phosphorylation by ATR (ataxia telangiectasia and Rad3-related protein), a DD sensor protein. When inhibited, the cell enters mitosis with fragmented chromosomes resulting in cell death. Additionaly, Chk1 inhibition leads to unscheduled increase of DNA replication forks, generating more DD and resulting in replication catastrophe [10,11]. Several Chk-i were examined preclincally but did not enter clinical trials mainly due to off-target activity/toxicity issues. Here we compare the in vitro and in vivo characteristics of two Chk-i with emphasis on Chk1 inhibition.

Prexasertib (PRX) is an ATP-competitive inhibitor of Chk1 and to a lesser extent of Chk2 [10]. PRX monotherapy induces DD and tumor cell death in vitro and tumor growth delay in vivo [10,12]. Combination of PRX with RT and cisplatin in head and neck squamous cell carcinoma (HNSCC) showes enhanced antitumor effects [11,13]. In MB, PRX sensitizes tumor cells to genotoxic drugs like cyclophosphamide, cisplatin, or gemcitabine in vitro and in vivo. The strong chemosensitizing activity of PRX has been shown in high-risk MB-bearing mice reducing tumor burden and increasing mouse survival [14]. However, the effect of PRX combined with RT in MB is still unknown. In clinical trials, PRX is already under investigation administered as monotherapy or in combination with cytotoxic agents. Also, in advanced MB patients, SHH subgroup included, the efficiency of PRX in combination with gemcitabine or cyclophosphamide is currently evaluated [15].

In contrast to PRX, SAR-020106 (SAR) is highly selective for Chk1 inhibition [16]. In colon carcinoma xenografts, it is able to enhance the efficiency of genotoxic drugs like irinotecan and gemcitabine [16,17]. Furthermore, in HNSCC and colorectal cancer, SAR sensitizes tumor cells to IR-induced damage in vitro and in vivo [16,17,18,19]. In glioblastoma cells, we have already shown a chemo-and radiosenzitizing effect of SAR prolonging genotoxic-induced DD repair and reducing clonogenic long-term survival [20].

In this study, we investigated for the first time the combinatory effects of RT with the Chk-i PRX and SAR in human SHH/p53-mut MB. In vitro, the influence on DNA damage repair, long-term clonogenic tumor cell survival, apoptosis, proliferation, and metabolic activity was analyzed in two SHH/p53-mut cell lines. In vivo, we used an orthotopic patient-derived SHH/p53-mut MB xenograft mouse model and assessed tumor growth by bioluminiscence imaging (BLI) and proliferation index by Ki-67 staining of tumor tissue.

Our studies will outline a possible clinicial opportunity for the therapy of high-risk SHH/p53-mut MB patients with adjuvant Chk1 inhibition.

2. Materials and Methods

2.1. Cell Lines

UW228 (SHH/p53-mut MB) cells were kindly provided by Dr. Hendrik Witt (DKFZ Heidelberg/Germany) and maintained in DMEM with 4.5 g/l glucose and L-glutamine (Biozym). DAOY (SHH/p53-mut MB) cells were purchased from ATCC cell biology collection (Manassas VA/USA) and maintained in MEM and 2 mM L-glutamine (Sigma-Aldrich). Cell culture media were supplemented with 10 % fetal calf serum (Sigma-Aldrich), 100 U/ml penicillin, and 100 U/ml streptomycin (Biozym).

2.2. Drugs

For in vitro use, stock solutions of 10 mM prexasertib HCl (LY2606368; Selleckchem) and 20 mM SAR-020106 (SYNkinase) were prepared in DMSO and stored at -20°C. Working solutions were diluted freshly in cell culture medium with a final DMSO concentration of 0.01 % (PRX) or 0.005 % (SAR).

2.3. Cell Culture Treatment and Assays

Cells were seeded and allowed to attach for 24 h, PRX or SAR was added 1 h prior to RT (single-dose). For fractionation experiments examining clonogenicity, half of medium complemented with drugs was renewed daily. An X-ray machine (X-Strahl 200, Xstrahl GmbH, Ratingen) with dose rates of 1.3 - 1.9 Gy/min was used at 150 kV. Appropriate DMSO-treated and sham-irradiated controls were implemented.

Metabolic activity was detected using WST-1 reagent (Roche). Cell proliferation was measured using colorimetric BrdU cell proliferation ELISA (Sigma-Aldrich). Cell death induced by apoptosis was detected by Annexin-V-FLUOS Staining Kit (Sigma-Aldrich) and analyzed by fluorescence cytometry as previously described [20]. To determine long-term survival of clonogenic cells, cells were seeded in 6-well plates and treated (daily; 4x) 24 h later. At d14, colonies were stained and surviving fraction (SF) was determined according to [21].

2.4. Fluorescence-Microscopic Analyses of DNA Damage in S Phase Cells

To determine DD in cell cycle S phase, double-staining of gH2AX protein and EdU-incorporation were examined by immunofluorescence. Prior to cell fixation at 1, 24, and 72 h, cells were exposed to 10 µM EdU (Click iT Plus EdU Flow Cytometry Assay Kit, Invitrogen #C10632) for 5 h. Staining of gH2AX protein was performed as previously described [20] with following amendments: Prior to DAPI counterstaining, gH2AX-stained cells were fixed with EdU Click-iT Fixation solution for 15 min and washed once with PBS + 1% BSA + 0.5% Tween20 (wash buffer). Permabilization with EdU Click-iT- Permeabilisation and Wash solution for 15 min and one washing step with wash buffer followed. The EdU Click-iT Reaction cocktail was mixed and cells incubated according to manufacturers’ instructions. Cells were washed three times, DAPI-counterstained and mounted according to [22].

Microscopic imaging (BZ-9000; BZ-II Viewer; Keyence) of at least 50 non-peripheral nuclei (DAPI) were taken for Alexa Fluor 488 (EdU) and Alexa Fluor 568 (gH2AX) using identical exposure parameters. Overlay pictures were analyzed using hybrid cell count – fluorescence – double extraction application (BZ-II analyzer) and identical conditions (Table 1).

2.5. Western Blot of DNA Damage Proteins

Western blot (WB) analysis of gH2AX, phospho-RPA, Histone H3, and GAPDH was adapted from [20,23]. In detail, cells suspension in assay buffer was treated with protease inhibitor (cOmpleteTM; Roche), sonicated 3 times (HTU SONI-130 MiniFIER, G. Heinemann, Schwäbisch Gmünd; 10 s on - 20 s off; amplitude 30 %), and 25 µg protein loaded onto a 15 % polyacrylamide gel together with ScanLaterTM Protein Ladder (Molecular Devices). Following antibodies were used: mouse anti-phospho-Histone H2A.X (Ser139), clone JBW301, Millipore, 1:500; rabbit phospho-RPA32/RPA2 (Ser8) clone E5A2F, Cell Signaling Technology, 1:1000; mouse Anti-GAPDH Loading Control Monoclonal Antibody (GA1R), Invitrogen Antibodies,1:1000; rabbit anti-histone H3 clone D1H2, XP® ChIP formulated, Cell Signaling Technology, 1:1000; secondary antibodies IRDye 680RD goat anti-mouse, 1:15000; and IRDye 800CW goat anti-rabbit, 1:8000; Li-COR Biosciences.

2.6. Mouse Model, Treatment, and Imaging

Female NSGTM mice (NOD.Cg-PrkdcscidIl2rgtm1Wjl/SzJ mice) were bred and housed as previously described [24]. All experiments had been approved in advance by the local authorities (Landesdirektion Sachsen TVV36/19). Orthotopic patient-derived xenograft (PDX) SHH/p53-mutated, MYCN-amplified (BT084), and luciferase-transfected MB mouse models were created as already described [24].

Mice were treated daily at d7-11 after tumor cell transplantation (Figure 4D). Drug solutions were prepared daily using 40 % DMSO in sterile water (PRX) or 10 % DMSO + 5 % Tween20 in 0.9 % NaCl (SAR). 2 ml/kg body weight were given s.c. (PRX 1 or 5 mg/kg) or i.p. (SAR 40 mg/kg) 2 h prior to RT. Whole-brain irradiation (daily, 5x) and bioluminescence imaging (BLI) (weekly) were performed using antagonizable narcosis as described [24]. For BLI, total flux (photons/second) was measured 13 min after luciferin injection with automatic exposure time. Relative total flux was calculated to mean of all animals at d4.

2.7. Tissue Preparation and Staining

Mice were euthanized 4 weeks after treatment. Tissue preparation, freezing, histological staining (H/E) and Ki-67 (tumor proliferative index; purified Mouse Anti-Ki-67, BD Pharmingen, CloneB56, 250 µg/mL, 1:150) were performed as described [24]. Number of Ki-67 positive cell nuclei per field of view (800×800 µm, ocular counting grid) was manually counted in the three most proliferative tumor areas (Axiolab, Zeiss).

2.8. Statistics

Statistical analysis between two treatment groups was conducted by one-sided Student’s t-test using Microsoft Excel 2016 software. P-values ≤ 0.05 (*; #) and ≤ 0.01 (**; ##) were considered as statistically significant; p-values ≤ 0.001 (***; ###) as highly statistically significant.

3. Results and Discussion

To determine in vitro effects of Chk inhibition by PRX (Chk1/2) or SAR (Chk1) combined with RT in SHH/p53-mut MB cells, we measured metabolic changes, proliferation, and short-term cell death 72 h, and clonogenic long-term survival 14 d after treatment.

Both, PRX and SAR, lead to a concentration-dependent decrease of metabolic activity and proliferation in UW228 and DAOY cells. Combination of Chk-i with RT led to a stronger decrease than single treatments (Figure 1A, B).

In accordance with the known high therapy-resistance of SHH MB [25], only minor RT-induced short-term cell death induction was found. Chk-i enhanced the apoptotic cell fraction in UW228/DAOY cells by 24/32 % at 10 nM PRX and 11/10 % at 1 µM SAR versus control (16/27 % apoptotic cells). Combinatorial treatment resulted in strongest effects. Thereby, PRX and SAR enhanced apoptotic fraction versus 15 Gy RT (21/50 % in UW228/DAOY) by 42/19 and 34/16 %, respectively (Figure 1C, D, E).

Our data, showing for the first time radioadditive antitumor effects of Chk-i in SHH/p53-mut MB cells, are in line with recent in vitro and preclinical findings demonstrating single-agent as well as sensitizing effectivity of PRX and SAR when combined with genotoxic treatment [10,11,12,13,16,18,26,27,28,29,30,31] including DNA-damaging ChT in MB [14]. Thereby genotoxic replication stress combined with simultaneous Chk1 inhibition often leads to enhanced cell death especially in p53-deficient cells where the ATM–Chk2–p53 signaling pathway is already disabled impeding DNA repair by non-homologous end-joining [10,18,32,33,34]. Furthermore, in p53-deficient cells, p53-independent apoptosis can be induced via Caspase-2, which is enhanced by Chk1 inhibition [35,36]. Accordingly, we previously demonstrate that the Chk1-i SAR enhance cytotoxic effects of genotoxic drugs and/or radiation stronger in p53-deficient than in p53-wildtype glioblastoma cells [20].

In UW228 MB cells, we could demonstrate radiosensitizing activity of PRX (5nM) and SAR (0.05 µM) on clonogenicity. Repeated PRX/SAR alone was effective to strongly induce clonogenic tumor cell death (Figure 2), most likely by accumulation of unrepaired, sublethal DD during ongoing cell cycles as shown in Figure 3A. Here, after Chk-i treatment, enhanced area of gH2AX expression/nucleus (PRX 39 ± 5 %; SAR 30 ± 4 %) coexisted with decreased foci number (PRX 21.6 ± 3.1; SAR 21.9 ± 3.6) compared to control (area 14 ± 2 %; foci 35.1 ± 2.4) at 72 h indicating large areas of condensed DD foci after 1-2 cell cycles (doubling time approx. 35 h); (Figure 3E). Similarly, combinatorial treatment with RT further enhanced DD (area gH2AX expression/nucleus PRX 70 ± 4 %; SAR 53 ± 4 %) compared to RT alone (13 ± 2 %) at 72 h and decreased foci numbers (Figure 3A).

By comparison of EdU-positive to non-labeled cells, we observed a slight shortening of S/G2M phase by SAR and PRX at 1 h, as already described for UW228 cells after lovastatin treatment [37]. At later time points, PRX (5 nM) induced S/G2M phase arrest at 24 and 72 h possibly as a result of cyclin-dependent kinase (CDK) inhibition, described for high-dose Chk1-i [38] (Figure 3C). Others described a Chk1-i-dependent enhancement of DD mainly in S phase [10,16,18], which was seen here only for SAR (Figure 3D).

Figure 1. Metabolic activity (A), proliferation (B), and apoptosis (C, D, E) of UW228 and DAOY cells 72 h after combined Chk inbibition (PRX or SAR) and RT. Data in line charts (A, B) present mean ± standard error of mean (SEM) of three independent experiments performed in triplicates. Apoptotic fraction (C) is presented as mean ± SEM measured in three independent experiments. Statistical significance compared to RT-only (also sham-RT) is indicated by hashtag (#, p ≤ 0.05; ##, p ≤ 0.01; ###, p ≤ 0.001) and compared to drug-only (also placebo) by asterisks (*, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001. (D) Representative dot blots from Annexin-V assay. (E) Representative photographs of cell growth and detached dead cells, 20-fold magnification.

Figure 1. Metabolic activity (A), proliferation (B), and apoptosis (C, D, E) of UW228 and DAOY cells 72 h after combined Chk inbibition (PRX or SAR) and RT. Data in line charts (A, B) present mean ± standard error of mean (SEM) of three independent experiments performed in triplicates. Apoptotic fraction (C) is presented as mean ± SEM measured in three independent experiments. Statistical significance compared to RT-only (also sham-RT) is indicated by hashtag (#, p ≤ 0.05; ##, p ≤ 0.01; ###, p ≤ 0.001) and compared to drug-only (also placebo) by asterisks (*, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001. (D) Representative dot blots from Annexin-V assay. (E) Representative photographs of cell growth and detached dead cells, 20-fold magnification.

Figure 2. Clonogenic survival of UW228 cells 14 d after combined Chk inbibition (PRX or SAR) and RT. (A) Long-term clonogenicity of UW228 cells decreased after fractionated RT (SF = 0.81 ± 0.04 at 4 × 0.5 Gy and 0.60 ± 0.03 at 4 × 1 Gy). PRX reduced SF to 0.75 ± 0.04 (1 nM) and 0.018 ± 0.004 (5 nM). SAR diminished SF to 0.87 ± 0.06 (0.05 µM) and 0.82 ± 0.04 (0.1 µM). Strongest radioadditive effects were found at 4 × 0.5 Gy for 5 nM PRX (SF = 0.006 ± 0.001; 135-fold) and for 0.1 µM SAR (SF = 0.59 ± 0.04; 1.4-fold) versus RT. Data present mean ± SEM of three independent experiments performed in sextuplicates. Statistical significance compared to RT-only group is indicated by asterisks (*, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001). Bars shows SF of drug-only treated groups. Significance compared to untreated control is indicated by hashtag (#, p ≤ 0.05; ##, p ≤ 0.01; ###, p ≤ 0.001) (B, C) Representative photographs of grown colonies, 100 cells were initially seeded. (C) 4-fold magnification, bar = 500 µM.

Figure 2. Clonogenic survival of UW228 cells 14 d after combined Chk inbibition (PRX or SAR) and RT. (A) Long-term clonogenicity of UW228 cells decreased after fractionated RT (SF = 0.81 ± 0.04 at 4 × 0.5 Gy and 0.60 ± 0.03 at 4 × 1 Gy). PRX reduced SF to 0.75 ± 0.04 (1 nM) and 0.018 ± 0.004 (5 nM). SAR diminished SF to 0.87 ± 0.06 (0.05 µM) and 0.82 ± 0.04 (0.1 µM). Strongest radioadditive effects were found at 4 × 0.5 Gy for 5 nM PRX (SF = 0.006 ± 0.001; 135-fold) and for 0.1 µM SAR (SF = 0.59 ± 0.04; 1.4-fold) versus RT. Data present mean ± SEM of three independent experiments performed in sextuplicates. Statistical significance compared to RT-only group is indicated by asterisks (*, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001). Bars shows SF of drug-only treated groups. Significance compared to untreated control is indicated by hashtag (#, p ≤ 0.05; ##, p ≤ 0.01; ###, p ≤ 0.001) (B, C) Representative photographs of grown colonies, 100 cells were initially seeded. (C) 4-fold magnification, bar = 500 µM.

RT-induced DD are known to activate Chk1 initiating S/G2M cell cycle arrest and DD repair [39]. Accordingly, cells accumulated in S/G2M phase during 24 h after IR and normalize to cell cycle distribution of control levels at 72 h. Chk-i pretreatment reduced the number of S/G2M phase cells versus RT alone (Figure 3C) after 24 h. RT and to a higher degree Chk-i/RT-treated cells showed largely unrepaired DD mainly in S/G2M phase nuclei (Figure 3 D). Despite normalized cell cycle distribution in the Chk-i treated cells after 72 h, DD was still present. WB of gH2AX and pRPA proteins (Figure 3E, F) and previous studies in glioblastoma cells treated with SAR and RT [20] are in line with these findings.

We conclude that induction of Chk1 by RT can be efficiently inhibited by PRX and SAR resulting in abrogation of S/G2M arrest and DD repair and enhanced tumor cell death.

To validate the promising in vitro data in vivo, we performed a proof-of-principle study with at least three mice per treatment group using an orthotopic SHH/p53-mut MB PDX model (Figure 4D).

Estimated in vivo doses were a maximal brain concentration of 13.7 nM for PRX (1 mg/kg; i.v.) [40] and a plasma concentration of about 50 µM for SAR (40 mg/kg; i.p.) [16]. For PRX, we also tested 5 and 10 mg/kg s.c. to compensate the possibly lower bioavailibility of s.c. versus i.v. application. For SAR, brain barrier penetration capability is still unknown, this is the first assessment of its potential intracranial antitumor activity.

Figure 3. DNA damage in UW228 cells after combined Chk inbibition (5 nM PRX or 0.5 µM SAR) and RT (8 Gy). (A) Area of gH2AX expression/nucleus and number of gH2AX foci 1, 24, and 72 h after treatment. Data present mean ± SEM of one experiment with at least 50 analyzed cell nuclei. Statistical significance compared to untreated/sham-irradiated control is indicated by asterisks (*, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001) and compared to RT-only group by hashtag (##, p ≤ 0.01; ###, p ≤ 0.001). (B) Representative photographs of immunofluorescence-stained nuclei (blue), EdU (green), and gH2AX protein (red). Lowest picture shows nuclei/EdU/gH2AX detected by BZ-II analyzer software (see method section). (C) Distribution of EdU-positive S/G2 phase cells and EdU-negative cells 1, 24, and 72 h after treatment. Mean of one experiment with at least 50 analyzed cell nuclei is shown. (D) Area of gH2AX expression/nucleus and number of gH2AX foci in EdU-positive S/G2 phase and and EdU-negative cells 1, 24, and 72 h after treatment. Data are presented as box-and-whisker plots, including minimum/maximum, lower/higher quartile, and median/mean of one experiment with at least 50 analyzed cell nuclei. Statistical significance between S/G2 and EdU-negative cells is indicated by asterisks (*, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001). (E) Representative photographs of gH2AX foci in control and PRX+RT-treated cells at 72 h. (F) Quantitative Western blot analyses of pRPA and gH2AX fluorescence intensities. Data present mean ± SEM of two blotting membranes of one experiment. Data are corrected to appropriate loading controls (Histone H3 and GAPDH) and normalized to untreated control. (G) Representative Western blots of pRPA, gH2AX, and loading controls (Histone H3, GAPDH).

Figure 3. DNA damage in UW228 cells after combined Chk inbibition (5 nM PRX or 0.5 µM SAR) and RT (8 Gy). (A) Area of gH2AX expression/nucleus and number of gH2AX foci 1, 24, and 72 h after treatment. Data present mean ± SEM of one experiment with at least 50 analyzed cell nuclei. Statistical significance compared to untreated/sham-irradiated control is indicated by asterisks (*, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001) and compared to RT-only group by hashtag (##, p ≤ 0.01; ###, p ≤ 0.001). (B) Representative photographs of immunofluorescence-stained nuclei (blue), EdU (green), and gH2AX protein (red). Lowest picture shows nuclei/EdU/gH2AX detected by BZ-II analyzer software (see method section). (C) Distribution of EdU-positive S/G2 phase cells and EdU-negative cells 1, 24, and 72 h after treatment. Mean of one experiment with at least 50 analyzed cell nuclei is shown. (D) Area of gH2AX expression/nucleus and number of gH2AX foci in EdU-positive S/G2 phase and and EdU-negative cells 1, 24, and 72 h after treatment. Data are presented as box-and-whisker plots, including minimum/maximum, lower/higher quartile, and median/mean of one experiment with at least 50 analyzed cell nuclei. Statistical significance between S/G2 and EdU-negative cells is indicated by asterisks (*, p ≤ 0.05; **, p ≤ 0.01; ***, p ≤ 0.001). (E) Representative photographs of gH2AX foci in control and PRX+RT-treated cells at 72 h. (F) Quantitative Western blot analyses of pRPA and gH2AX fluorescence intensities. Data present mean ± SEM of two blotting membranes of one experiment. Data are corrected to appropriate loading controls (Histone H3 and GAPDH) and normalized to untreated control. (G) Representative Western blots of pRPA, gH2AX, and loading controls (Histone H3, GAPDH).

Tumor mass of placebo-treated animals grew continually over time and was enhanced 301-fold at d42 (euthanasia) versus d4 (Figure 4A, B, C). SAR monotherapy initially enhanced tumor growth at d14 (12.4-fold; p ≤ 0.05) presumably due to cell cycle acceleration by S/G2M checkpoint abrogation. This was followed by growth inhibition, compared to control (d28-42; n = 3; n.s.), assumingly due to DD accumulation as shown in vitro. PRX alone induced similar effects albeit to a lesser extent (n = 3; n.s.). Fractionated RT inhibited the tumor growth leading to reduced tumor mass until end of experiment (>d28; n = 6; p ≤ 0.05), which is also reflected by the reduction of proliferative cells (225 ± 9 to 185 ± 6 cells/FOV; n = 9; p ≤ 0.001), (Figure 5). Nevertheless, RT alone did not cure the animals as BLI evaluation revealed progression of tumor growth on d35 (Figure 4C). Combination of SAR/PRX and RT showed additive effects and further reduced tumor mass compared to monotherapy, except of SAR at d35/42 and high-dose PRX at d42 where Chk-i therapy apparently induces radioresistance. Accordingly, combination of RT and PRX reduced proliferation index by 24 % (n = 3; p ≤ 0.05) only at low concentration (1 mg/kg), (Figure 5), indicating a persistent effect. Onset of bulky head was monitored as a marker of growing tumor but was found to be too insensitive to reveal any differences between treatment groups (Figure 6).

Toxicity of therapy was assessed by mouse body weight (BW). After tumor inoculation, BW decreased by 5 % at d1-2 due to anesthesia and intracranial surgery but stabilized at d3-4 (Figure 6). Preliminary dose-testing experiments revealed that intense treatment regime with PRX (5 × 10 mg/kg) and/or RT (5 × 1 Gy) leads to a strong decrease (10 – 20 %) of BW, requiring analgesia and was therefore abandoned. Lower PRX doses (5 × 1 or 5 x 5 mg/kg), SAR, and/or RT (5 × 0.5 Gy) were well tolerated. Our data show for the first time intracranial antitumor activity suggesting brain barrier penetration of SAR although, combinations of SAR with RT/ChT have already been shown to delay tumor growth in HNSCC and colorectal cancer mouse models [16,18]. For PRX, CNS penetration is already proven [14,40] and is hereby supported in NSG mice. Several recently published clinical trials confirm PRX tolerability and efficiency [41,42,43,44,45,46,47,48,49] and one phase 1 trial examining the combination of PRX and chemotherapy in MB is currently active [15]. Also in pediatric patients, PRX was under clinical examination [50]. Our study is limited by the number of animals used in the experiments and possibly a suboptimal dosing of SAR. Further investigations combining Chk-i not only with genotoxic therapies but also with DD repair-affecting drugs, like PARP inhibitors [51], are warranted.

Figure 4. Tumor growth of SHH/p53-mut MB after fractionated therapy (5x, daily) with PRX (1 or 5 mg/kg/d) or SAR (40 mg/kg/d) and RT (0.5 Gy/d). (A) Representative bioluminescence images. (B) Relative total flux at d14, 28, 35, and 42 after tumor inoculation. Raw values were normalized to d4 of all evaluated mice (n = 20). Data presented as box-and-whisker plots, including minimum/maximum, lower/higher quartile, and median/mean. Number of analyzed animals/treatment group: control, RT n =6; other groups n = 3. Statistical significance compared to appropriate sham-RT group is indicated by asterisks (*, p ≤ 0.05), compared to RT-only by hashtag (#, p ≤ 0.05), and compared to placebo-treated/sham-irradiated control by rhomb (◊, p ≤ 0.05). (C) Absolute total flux of single mouse BLI measurements and mean of all (red line). Treatment window is indicated by a black bar at x-axis. (D) Workflow of in vivo procedures. Treatment comprised PRX (1 or 5mg/kg/d, s.c.) or SAR (40 mg/kg/d, i.p.) and RT (0.5 Gy/d) 2 h later. Created with BioRender.com.

Figure 4. Tumor growth of SHH/p53-mut MB after fractionated therapy (5x, daily) with PRX (1 or 5 mg/kg/d) or SAR (40 mg/kg/d) and RT (0.5 Gy/d). (A) Representative bioluminescence images. (B) Relative total flux at d14, 28, 35, and 42 after tumor inoculation. Raw values were normalized to d4 of all evaluated mice (n = 20). Data presented as box-and-whisker plots, including minimum/maximum, lower/higher quartile, and median/mean. Number of analyzed animals/treatment group: control, RT n =6; other groups n = 3. Statistical significance compared to appropriate sham-RT group is indicated by asterisks (*, p ≤ 0.05), compared to RT-only by hashtag (#, p ≤ 0.05), and compared to placebo-treated/sham-irradiated control by rhomb (◊, p ≤ 0.05). (C) Absolute total flux of single mouse BLI measurements and mean of all (red line). Treatment window is indicated by a black bar at x-axis. (D) Workflow of in vivo procedures. Treatment comprised PRX (1 or 5mg/kg/d, s.c.) or SAR (40 mg/kg/d, i.p.) and RT (0.5 Gy/d) 2 h later. Created with BioRender.com.

Figure 5. Proliferation index of SHH/p53-mut MB four weeks after fractionated therapy (5x, daily) with PRX and RT (0.5 Gy/d). (A) Mean number of Ki-67-positive cells per field of view (FOV) in the most proliferative tumor area. Data present mean ± SEM from three mice with three analyzed FOVs. Statistical significance of RT over all groups is indicated by asterisks (*, p ≤ 0.05), statistical significance of single groups to placebo-treated group is indicated by hashtag (#, p ≤ 0.05). (B) Representative photographs of Ki-67-stained cells, 400-fold magnification, scale bars = 50 µm.

Figure 5. Proliferation index of SHH/p53-mut MB four weeks after fractionated therapy (5x, daily) with PRX and RT (0.5 Gy/d). (A) Mean number of Ki-67-positive cells per field of view (FOV) in the most proliferative tumor area. Data present mean ± SEM from three mice with three analyzed FOVs. Statistical significance of RT over all groups is indicated by asterisks (*, p ≤ 0.05), statistical significance of single groups to placebo-treated group is indicated by hashtag (#, p ≤ 0.05). (B) Representative photographs of Ki-67-stained cells, 400-fold magnification, scale bars = 50 µm.

Figure 6. Body weight and health condition of SHH/p53-mut MB-bearing mice before/during/after fractionated therapy (5x, daily) with PRX or SAR and RT. (A) Daily BW of single mice over time normalized to d0. Onset of bulky head is marked by yellow circle, necessity of analgesia is marked by green triangle. Treatment window is shaded in blue. (B) Photographs of MB-bearing mice over experimental period.

Figure 6. Body weight and health condition of SHH/p53-mut MB-bearing mice before/during/after fractionated therapy (5x, daily) with PRX or SAR and RT. (A) Daily BW of single mice over time normalized to d0. Onset of bulky head is marked by yellow circle, necessity of analgesia is marked by green triangle. Treatment window is shaded in blue. (B) Photographs of MB-bearing mice over experimental period.

4. Conclusions

Our data emphasize the clinical benefit of integration of Chk-i in SHH/p53-mut MB therapy regimes including RT. This might hold true also in MYC-driven Group 3 MB, as MYC is a highly predictive marker for PRX sensitivity [14,51]. However, a great deal of attention should be taken on dosing regimens as high-dose Chk-i may compromise the RT effect on tumor growth possibly by inhibition of proliferation. We demonstrate for the first time the intracranial antitumor activity of the Chk1-specific inhibitor SAR in mouse.