Elimusertib (BAY1895344), a Novel ATR Inhibitor, Demonstrates in vivo Activity in ATRX Mutated Models of Uterine Leiomyosarcoma

Abstract

Introduction.

Uterine leiomyosarcoma (uLMS) is a rare, highly aggressive malignancy. Recent data suggest 50% of uLMS may harbor alterations in the ATRX gene and such mutations may confer sensitivity to ataxia-telangiectasia-and-Rad3-related (ATR) kinase inhibitors. We sought to investigate the in vivo activity of Elimusertib (BAY1895344), a novel ATR-inhibitor, against ATRX-mutated uLMS patient-derived xenografts (PDXs).

Methods.

Two fully characterized uLMS (i.e., LEY-11 and LEY-16) were grafted into female CB-17/SCID mice. Treatments with control vehicle or BAY1895344 (20 mg/kg dosed twice daily 3 days on 4 days off) were given via oral gavage and tumor measurements as well as weights obtained twice weekly. Tumor volume differences were calculated with a two-way ANOVA. Mechanistic studies were performed ex vivo using BAY1895344 treated uLMS tumor samples by western blot analysis.

Results.

Both PDX LEY-11 and PDX LEY-16 harboring ATRX gene mutations demonstrated an aggressive behavior in vivo (i.e., control mice were euthanized on average at day 12.5 for PDX LEY-11 and at day 33 for PDX LEY-16). In both tumor models BAY1895344 20mg/kg dosed with an intermittent oral schedule was able to induce significant growth inhibition compared to vehicle control treatment (p<0.001 for both LEY-11 and LEY-16) and prolong median overall survival [PDX LEY-11 (12.5 vs. 42 days, p <0.001) and PDX LEY-16 (33 vs. 60 days, p < 0.001)]. There were not significant changes in weight between treatment and controls. By western blot assays BAY1895344 exposure decreased phosphorylated-ATR and increased expression of apoptotic molecules in LMS PDXs.

Conclusions.

BAY1895344 demonstrates promising in vivo activity against biologically aggressive PDX models of uLMS harboring ATRX mutations, with no significant toxicity. Clinical trials of BAY1895344 in uLMS patients are warranted.

Introduction

Uterine leiomyosarcoma (uLMS) is a rare but highly aggressive gynecologic malignancy that has an estimated 5-year overall survival of 50% for even early-stage disease. Though they are rare and account for only 3–7% of uterine malignancies, uLMS is the most common gynecologic sarcoma, accounting for 80% of cases [1]. Management consists of extirpative surgery followed by observation for early-stage disease (stage I-II) and adjuvant therapy for advanced-stage disease [1]. However, given that uLMS is a rare disease, evidence guiding recommendations for systemic therapy for recurrent or metastatic disease is poor., Several prospective studies demonstrate doxorubicin, gemcitabine, ifosfamide, and docetaxel to be active in advanced-stage disease, but in widespread metastatic disease outcomes are universally poor [2–5].

Given the lack of effective adjuvant therapies and poor overall survival, recent studies have examined the genetic landscape of uLMS to determine if there are common and potentially actionable genetic mutations. However, these studies had samples sizes of <30 and/or included leiomyosarcomas from extra-gynecologic anatomical sites [6–8]. In 2021, our research group published the largest genomic analysis to date of uLMS [9]. Using whole-genome-sequencing (WGS), whole exome sequencing (WES) and RNA-Sequencing (RNA-Seq), 56 uLMS along with 27 matched uLMS samples obtained through The Cancer Genome Atlas (TCGA) were evaluated and results demonstrated frequently mutated genes including TP53, ATRX, PTEN, and MEN1 genes [9]. Somatic Copy Number Variant (CNV) analysis revealed evidence of C-MYC copy-number gains and homologous recombination deficiency (HRD) signatures as well in a subset of uLMS. Lastly, in vivo monotherapy with olaparib, GS-626510, and copanlisibdemonstrated tumor growth inhibition and prolonged survival in SCID mice implanted with uLMS xenografts [9].

The ATRX gene is associated with a specific route of telomere lengthening in cancer cells, termed alternative lengthening of telomeres (ALT), which allows 10–15% of human tumors to achieve “immortalization” and unlimited tumor division through this mechanism [10]. Tumors of mesenchymal origin (such as uLMS) have been found to more commonly utilize ALT; inactivating mutations of ATRX and its associated death-domain associated protein (DAXX) are more common in tumor cells that utilize ALT [10,11]. The proteins encoded by the ATRX and DAXX genes facilitate binding of the histone variant H3.3 into telomeric chromatin and stalling of cell division in the presence of replicative stress, thus functioning as a tumor suppressor [10]. A similar and related protein, ataxia telangiectasia and Rad3-related protein (ATR), is a key component of DNA Damage Repair (DDR) for replicative stress, though not a tumor-suppressor [11]. ATR, along with ataxia telangiectasia mutated (ATM) and DNA dependent protein kinase (DNA-PK) form the cell’s main response to DDR; ATM and DDK are primarily involved in double-strand break repair and, as aforementioned, ATR is involved in repair of replicative stress [10–12]. As such, ATRX/DAXX and ATM mutations may lead to increased susceptibility/dependency to ATR for a cancer cell’s rescue from mitotic catastrophe, making it an appealing target for inhibition in the fight against cancer. Consistent with this view, ATRX mutations have been found to confer increased sensitivity to ATRi [13]. Currently, there are several ATR inhibitors (ATRi) in Phase I-II trials that have demonstrated promise, including BAY1895344, a selective inhibitor of ATR kinase activity [12]. In preclinical models of various human cancers with varying DDR mutations (i.e. ATM, ATR, ATRX, BRIP1, MLH1, etc.) BAY1895344 demonstrated significant activity both as monotherapy and in combination with varying agents including olaparib and rucaparib; however no sarcomas were studied [12]. Given that the prevalence of ATRX mutations in leiomyosarcoma is estimated to be over 30% [14] and the potential of ATRX as a novel therapeutic target, we sought to determine the in vivo activity of BAY1895344 against two recently established and fully sequenced ATRX-mutated uLMS PDXs.

Materials and Methods

2.1. Patient and Specimen Acquisition

The study protocol was approved by the Yale Human Investigation Committee and was conducted in accordance with the Declaration of Helsinki. DNA and RNA was extracted from two PDXs, which were established from two patients after informed consent from the participant and/or their legally authorized representative was obtained prior to initiating any research activities. Both patients underwent staging at Yale New Haven Hospital. Benign myometrial tissue was similarly collected from patients after informed consent from the participant and/or their legally authorized representative was obtained.

2.2. BAY1895344

BAY1895344 was obtained from Bayer AG and prepared as previously described [12]. Briefly, it was prepared into a stock weekly with a vehicle of PEG 400, water, and 100% ethanol and kept at 4°C shielded from light. Based on data published on the bioavailability of the compound in preclinical models, dosages of 10mg/kg twice daily and 20mg/kg twice daily were chosen for evaluation [12].

2.3. Whole-exome DNA sequencing and CNV analysis

Whole-exome sequencing and CNV analysis on uLMS tumors were undertaken as previously described [9].

2.4. Real-time (RT) reverse transcription PCR

RNA isolation was performed on two uLMS samples (LEY-11 and LEY-16) and 11 randomly selected uterine myometrium control fresh-frozen samples was performed using AllPrep DNA/RNA/Protein Mini Kit (Qiagen), according to the manufacturer’s instructions. Quantitative PCR was carried out with a 7500 Real-Time PCR System using the manufacturer’s recommended protocol (Applied Biosystems) to evaluate the expression of ATRX, ATR, and DAXX. The primers and probes were obtained from Applied Biosystems (i.e., ATRX, Assay ID: Hs00997529_m1; ATR Assay ID: Hs00992123_m1; DAXX Assay ID: Hs00154692_m1). The comparative threshold cycle method was used to determine gene expression in each sample, relative to the value of glyceraldehyde-3-phosphate dehydrogenase (assay ID Hs99999905_m1) RNA as an internal control.

2.5. Leiomyosarcoma PDX model and Treatment

Fresh frozen samples of the two fully characterized uLMS (LEY-11 and LEY-16) were xenografted into female CB17/lcrHsd-Prkdc/SCID mice into the subcutaneous tissue of the lower abdomen. Mice were triaged into treatment groups when the tumor was established at ≥ 0.2 cm³, which included a control group,BAY1895344 10mg/kg, and BAY1895344 20mg/kg. Animals were treated with BAY1895344 or normal saline via oral gavage twice daily for 3 days on/4 days off. Lower abdominal tumors were measured in length and width using calipers twice weekly. Animals were humanely euthanized when tumor volume reached 1.0 cm³ using the formula length × (width)²/2. Once all control group animals were euthanized, treatment group animals were followed for survival. Toxicity was assessed via mouse weight, which was recorded twice weekly; treatment was held if an animal’s weight dropped below 10% of their starting weight. All mice were housed and treated in accordance with the policies set forth by the Institutional Animal Care and Use Committee (IACUC) at Yale University.

2.6. Ex vivo tumor tissue culture and treatment

Tumor tissues were dissected from mice and cut into 3×3×3mm pieces. Tumor pieces were then incubated in RPMI 1640 medium (Life Technologies, Grand Island, NY) containing 10% fetal bovine serum (Seradigm, Radnor, PA) and penicillin/streptomycin (Life Technologies) at 37°C and 5% CO₂. After 1 hour of incubation, tumor pieces were treated with medium containing BAY1895344 (0.3 μM or 3 μM) or the DMSO solvent (0.03%) as untreated control. At two hours, tumor pieces were homogenized and used for preparing protein lysates.

2.7. Western Blot

Tumor tissues were homogenized using a BeadBug Homogenizer (Benchmark Scientific, Sayreville, NJ). Protein lysates were prepared with lysis buffer (1% Triton X-100, 0.05% SDS, 100 mM Na₂HPO4, and 150 mM NaCl). Protein lysate was electrophoresed on a 4–20% pre-cast SDS-polyacrylamide gel (Bio-Rad, Hercules, CA) and transferred onto Amersham Hybond 0.45 PVDF membranes (GE Healthcare, Chicago, IL). After blocking with 5% non-fat milk in PBS-0.05% Tween 20, the membranes were incubated with primary antibodies at 4°C overnight, and then secondary antibodies for 1 h at room temperature. Antibodies include anti-p-ATR antibody (#2853, Cell Signaling Technology), anti-ATR antibody (#13934, Cell Signaling Technology), p-CHK1 antibody (#2348, Cell Signaling Technology, Danvers, MA), cleaved-caspase-3 antibody (#9664, Cell Signaling Technology), HRP-conjugated anti-β-actin antibody (#HRP-6008, ProteinTech), anti-rabbit secondary antibody (#7074, Cell Signaling Technology). The blots were developed using Clarity or Clarity Max Western ECL Blotting Substrates (Bio-Rad).

2.8. Statistical Analysis

Statistical analysis was performed using GraphPad Prism version 7 (GraphPad Software, Inc. San Diego, CA). ANOVA was used to evaluate significant differences in tumor volumes at specific time points for the in vivo experiments. Overall survival (OS) data (OS defined as time of enrollment to either death or tumor volume of 1.0 cm³) were analyzed and plotted using Kaplan-Meier survival curves, which were compared using the log-rank test. Differences in all comparisons were considered statistically significant at p-values <0.05.

Results

LEY-11 and LEY-16 genetic landscape

Analyses of the genetic characteristics of the two uLMS prior to treatment or implantation as PDX (LEY-11 and LEY-16), including somatic single nucleotide variant (SNV) and CNV mutations, were highly consistent with the genetic landscape results recently reported by our group using a large set of fresh and formalin fixed uLMS (ie, 83 samples) [10]. As depicted in Table 1.A, we found SNV alterations in multiple driver genes including TP53, PTEN, ATRX, RB1 and KTM2B as well as CNV in C-MYC and MAP2K4 in LEY-11. Similarly, LEY-16 demonstrated SNV alterations in driver genes including TP53, SMARCA4 and NAV3, CNV in PTEN, MAP2K4 and ATRX genes, as well as multiple fusions/translocations (Table 1.B).

Table 1.

Genetic landscape of sequenced tumors. A) Genetic landscape of LEY-11. B) Genetic landscape of LEY-16.

Table 1A. LEY-11 Genetic Landscape
i. Somatic SNVs in Cancer190 genes or HR48 gene
Chr	Pos	Ref	Alt	Gene	Type	AA	dbSNP	ExAC	MetaSVM	CADD	Ref:Alt
17	7574003	G	A	TP53	stopgain	p.R303X	.	.	.	37.0	24,302
10	89711950	C	-	PTEN	frameshift deletion	p.P363Qfs*9	.	.	.	.	9,84
X	76855019	T	-	ATRX	frameshift deletion	p.D1940lfs*15	.	.	.	.	99,57
13	49039469	T	-	RB1	frameshift deletion	p.L819Cfs*7	.	.	.	.	30,232
19	36220087	C	T	KMT2B	stopgain	p.Q1603X	.	.	.	44.0	11,96
ii. Somatic CNVs covering Cancer190 or HR48 genes
Chr	Start	End	CNV type	Copy number	#Genes	Cancer190 genes
1	149901457	166839177	Duplication	3	365	B4GALT3 (S100A7, S100A8, S100A9)
4	1219049	8952176	Duplication	4	70	CRIPAK, FGFR3
7	23730942	35013303	Duplication	3	64	NFE2L3
8	48114004	146279624	Duplication	3	402	EPPK1, RAD21, SOX17 (MYC, NBN, RAD54B)
10	5919996	38407815	Deletion	1	130	GATA3
11	100665721	109295003	Duplication	3	51	ATM
16	97334	33965671	Duplication	3	434	AXIN1, CREBBP, SOCS1, TRAF7 (PALB2)
16	47117063	52060975	Duplication	3	22	CYLD
16	68771193	76592630	Duplication	3	76	CDH1
17	909249	10233910	Duplication	3	213	TP53 (RPA1)
17	10608193	16137459	Duplication	3	28	MAP2K4, NCOR1
18	18531196	78005326	Duplication	3	198	BCL2, SETBP1, SMAD2, SMAD4 (RBBP8)
20	68253	26189020	Duplication	3	176	FOXA2
20	35788394	62905003	Duplication	3	253	TSHZ2
21	14982479	45514442	Duplication	3	162	RUNX1, U2AF1
X	5827024	57936904	Duplication	3	295	BCOR, GATA1, KDM5C, KDM6A, SMC1A, USP9X
X	85631746	155240124	Deletion	1	365	PHF6, STAG2 (BRCC3, RPA4)
11	89402478	96125438	Duplication	3	41	(MRE11A)
17	16593655	18291667	Duplication	3	27	(TOP3A)
iii. Fusions/Translocations in Kinase518/Cancer190/Cosmic567/HR48 genes None found
Table 1B. LEY-16 Genetic Landscape
i. Somatic SNVs in Cancer190 genes or HR48 gene
Chr	Pos	Ref	Alt	Gene	Type	AA	dbSNP	ExAC	MetaSVM	CADD	Ref:Alt
17	7578406	C	T	TP53	Nonsynonymo us SNV	p.R136H	8.24E-06	8.24E-06	D	31.0	16,46
19	11101886	G	T	SMARCA4	Nonsynonymo us SNV	p.A436S	.	.	T	15.4	70,41
12	78443857	T	C	NAV3	Nonsynonymo us SNV	p.L703P	.	.	T	25.4	37,4
ii. Somatic CNVs covering Cancer190 or HR48 genes
Chr	Start	End	CNV type	Copy number	# Genes	Cancer190 genes
1	156643133	164762038	Duplication	4	135	B4GALT3 (S100A7, S100A8, S100A9)
1	145368425	147608093	Duplication	3	32	(BCL9)
6	121544283	157100664	Duplication	3	154	ARID1B, TNFAIP3
9	107456607	113734510	Duplication	3	31	KLF4
10	88635664	93871025	Deletion	1	44	PTEN
14	58598133	82000184	Deletion	1	178	TSHR (RAD51B)
17	8296018	12921314	Duplication	3	33	MAP2K4
X	76845218	77131148	Deletion	1	2	ATRX
15	40993171	41672467	Duplication	3	18	RAD51
iii. Fusions/Translocations in Kinase518/Cancer190/Cosmic567/HR48 genes
Gene1	Gene2	Genelclass	Gene2class	Breakpoint1	Breakpoint2	Type	Splitreads1	Splitreads2	Discordantmates	Coverage1	Coverage2	Confidence	Reading-frame	Calledby
ZCCHC4	TEC	NA	Kinase	4:25363923	4:48152959	Inversion	2	2	1	47	36	High	Inframe	Arriba, Starfusion
HEG1	CAMTA1	NA	Cosmic567	3:124724109	1:7796404	Translocation	0	1	3	247	15	High	Inframe	Arriba
NUSAP1	RAD51	NA	HR	15:41669502	15:40990955	Duplication	2	4	11	1495	232	High	Inframe	Amba, Starfusion
POLD2	EEPD1	HR	NA	7:44154631	7:36334034	Deletion/5’-5’	0	2	1	1	1	medium	-	Arriba

GISTIC significant duplications or deletions

Low ATRX expression by RT-PCR in PDX LEY-11 and PDX LEY-16

As both LEY-11 and LEY-16 demonstrated alterations in the ATRX gene, we used real-time PCR to evaluate ATRX RNA expression as well as the expression of other genes of this pathway such as ATR and DAXX in the two available PDX samples, along with 11 fresh frozen myometrium control samples obtained from benign hysterectomy specimens.. As demonstrated in Table 2, similar to our previous report studying 37 fresh-frozen uLMS (which included both LEY-11 and LEY-16) [9], we found both PDXs to express lower levels of ATRX transcripts when compared to benign uterine myometrium, while no differences in expression were noted in ATR and DAXX genes (Table 2).

Table 2.

RT-PCR expression of ATRX, ATR and DAXX in PDX LEY-16 and PDX LEY-11 and controls (benign myometrium).

Sample	ATR Ct	ATR ΔCt	ATRX Ct	ATRX ΔCt	DAXX Ct	DAXX ΔCt
LEY-11	29.18	12.41	26.89	10.091	25.77	8.97
LEY-16	28.53	8.41	35.77	15.978	26.39	6.59
Leiomyoma controls	32.80	11.08	30.22	6.54	30.69	6.46

In vivo Treatment of ATRX-mutated uLMS PDXs in SCID mice with BAY1895344 Produces a Cytostatic Effect and Prolongs Overall Survival

We next evaluated the impact of BAY1895344 in vivo in xenograft models generated by subcutaneously grafting fresh-frozen LEY-11 and LEY-16 samples into female CB17/lcrHsd-Prkdc/SCID mice. In both models, PDX LEY-11 and PDX LEY-16, BAY1895344 dosed intermittently demonstrated a significant cytostatic effect with a smaller tumor volume (Figure 1.A, 2.1.A, and 2.2.A). In PDX LEY-16, this reached statistical significance on day 22 of treatment (p<0.001); in PDX LEY-11, two separate experiments were undertaken given the more pronounced effect in this model compared to PDX LEY-16 and to also test BAY1895344 at a lower dose (10 mg/kg) (Figure 2). At a dose of 20mg/kg given twice daily (BID), the tumor volume inhibition reached significance at day 11 (p<0.001) in the first experiment (Figure 2.1.A) and day 8 (p<0.001) in the second experiment. However, at a lower dose of 10mg/kg BID, tumor volume inhibition was not significantly different than control (p=0.54). Additionally, the tumor growth rate for PDX LEY-11 was significantly slower in the 20mg/kg treatment groups: 0.093 cm³/day vs. 0.028 cm³/day in the control group and 20 mg/kg group, respectively in experiment 1 (p=0.002) and 0.112 cm³/day vs. 0.017 cm³/day in experiment 2 (p=0.009). The tumor growth rate for PDX LEY-11 treated with 10mg/kg of drug was not significantly different than control in experiment 2 (p=0.94). Overall survival curves demonstrated a significant median OS difference between BAY1895344 20mg/kg BID and control in PDX LEY-16 (Figure 1.B) and between 20mg/kg BID and control in both PDX LEY-11 experiments (Figure 2.1.B and 2.2.B). For PDX LEY-16, median OS for control vs. 20mg/kg BID was 33 vs. 60 days (p=0.0013) (Figure 1.B). For PDX LEY-11, median OS for control vs. 20mg/kg BID in experiment 1 was 12.5 vs. 42 days (p<0.001) (Figure 2.1.B); In experiment 2, median OS for control vs. 10mg/kg vs. 20mg/kg BID was 8 vs. 11 vs. 42 days; the difference in survival between control and 10mg/kg was not statistically significant (p=0.0016 in control vs. 20mg/kg; p=0.06 in control vs. 10mg/kg) (Figure 2.2.B). Twice daily oral doses for three days weekly (10 mg/kg or 20 mg/kg) were well tolerated with no impact on body weight compared with vehicle (Figure 1.C, 2.1.C, and 2.2.C).

Figure 1.

Antitumor activity and overall survival in mice inoculated with uLMS xenograft tumor model (PDX LEY-16) after treatment with BAY1895344 (20mg/kg) compared to vehicle control. A) BAY1895344 (20mg/kg) demonstrates significant tumor growth inhibition compared to vehicle control (p<0.001). B) Overall survival was significantly prolonged among mice treated with BAY1895344 (20mg/kg) (p=0.0013). C) Xenograft mice weight changes by treatment group did not differ.

Figure 2.

Antitumor activity and overall survival in mice inoculated with uLMS xenograft tumor model (PDX LEY-11) after treatment with BAY1895344 (20mg/kg or 10mg/kg) compared to vehicle control over two experiments. Experiment 1) Vehicle control vs. BAY1895344 (20mg/kg) 2.1.A) BAY1895344 (20mg/kg) demonstrates significant tumor growth inhibition compared to vehicle control (p<0.001) 2.1.B) Overall survival was significantly prolonged among mice treated with BAY1895344 (20mg/kg) (p<0.001). 2.1.C) Xenograft mice weight changes by treatment group. Experiment 2) Vehicle control vs. BAY1895344 (20mg/kg) vs. BAY1895344 (10mg/kg) 2.2.A) BAY1895344 (20mg/kg) demonstrated significant tumor growth inhibition compared to vehicle control (p<0.001); 10mg/kg dosage did not have significant inhibition compared to control (p=0.54) 2.2.B) Overall survival was significantly prolonged among mice treated with BAY1895344 (20mg/kg) (p=0.0016); Mice treated with BAY1985344 (10mg/kg) did not have significantly prolonged overall survival (p=0.06). 2.2.C) Xenograft mice weight changes by treatment group did not differ.

PDX LEY-11 BAY1895344 exposure Reduces Phosphorylated-ATR and increases cleaved caspase-3

As representatively demonstrated in Figure 3, in ex-vivo culture model of PDX LEY-11 tumors, treatment with ATR inhibitor BAY1895344 for 2 hours significantly reduced the level of phosphorylated ATR. The checkpoint kinase CHK1 undergoes ATR-mediated phosphorylation and activation. Accordingly, the inhibition of ATR by BAY1895344 treatment also led to the reduction of phosphorylated CHK1. As a marker of cell death induction, the activated caspase-3 (17 or 19 kDa) resulting from cleavage adjacent to Asp175 was progressively detected in the BAY1895344-treated tumors demonstrating increased tumor cell apoptosis.

Figure 3.

Representative western blot of ex-vivo tumor culture model of PDX LEY-11 treated with BAY1895344 for 2 hours. Treatment with BAY1895344 significantly reduced the level of phosphorylated ATR and phosphorylated CHK1; Levels of cleaved-caspase3 increase with prolonged treatment demonstrating increased cell apoptosis.

Discussion

There are several major challenges in the treatment of uLMS, including its relative resistance to traditional cytotoxic therapies, its rarity, and the relative lack of available targeted therapeutics. In our in vivo studies, BAY1895344 demonstrated significant cytostatic effect in both available PDX models. The aggressive biologic nature of these tumor models is highlighted by the low median overall survival in the vehicle group: 8–12.5 days in PDX LEY-11 and 33 days in PDX LEY-16. Additionally, the treatment was well tolerated with no differences in body weight between control-group and treatment-group mice. This is mirrored in a recent Phase I trial of BAY1895344, wherein the majority of dose-limiting toxicities were hematologic, which recovered with treatment holds [15]. Experiments were performed at two dosages (20mg/kg and 10mg/kg BID, 3 days on 4 off schedule): only the higher dosage was efficacious for tumor size reduction and survival benefit.

A recent publication from our laboratory with the largest WGS, WES, and RNA-Seq data to date demonstrated that 51% of samples harbored ATRX gene alterations, which result in decreased ATRX RNA expression when compared to wild type ATRX uLMS [9]. In this comprehensive genetic evaluation, we also found that ATRX gene derangement was significantly associated with decreased survival in uLMS patients [9]. Given the link between ATRX mutations and sensitivity to ATR inhibitors, we sought to investigate this novel targeted treatment in uLMS where an unmet clinical need exists [15].

ATR and its cascade demonstrate a promising target for novel clinical therapeutics. Preclinical data demonstrate many uLMS carry these alterations as well as additional alterations in DAXX and ATM genes, increasing the number of patients potentially responsive to novel treatment modalities targeting ALT. The recent Phase I trial and first-in-human trial of BAY 1895344 in patients with advanced solid tumors confirmed that the drug had promising antitumor activity, especially among those with ATM gene mutations [15]. In this regard, recent work from our lab demonstrated that tumors harboring ATRX alterations express lower levels of gene expression, which is associated with the aforementioned ALT pathway activation. It is worth noting that patients with ATRX/DAXX mutations were characterized by a significantly shorter survival, supporting the view that alterations in this DNA repair pathway trigger a more aggressive uLMS phenotype [9]. These clinical findings are further supported by our current preclinical study as demonstrated by the rapid in vivo growth of both uLMS PDX models harboring ATRX alterations in SCID mice. More importantly, these data suggest that novel ATR inhibitors currently in clinical trials such as BAY1895344, may represent new, potentially effective treatment options for the large subset of patients with advanced/recurrent uLMS carrying ATRX/DAXX mutations [9,16,17] Importantly, detection of ATRX and DAXX alterations is now practical with commercially available genomic profiling tests that sample tumoral specimens, making detection of these prevalent alterations more feasible [18]. Additionally, recent studies demonstrate that SMARCA4/BRG1 mutations can also confer susceptibility to ATR inhibition [19–21]; it is notable that LEY-16 was found to have alternations in SMARCA4 driver gene as well as ATRX.

.. Mechanistically, in vivo RT-PCR of pre-treatment PDX tissue demonstrated lower expression of ATRX gene transcripts in the PDX LEY-11 and PDX LEY-16 models, as expected given their known ATRX mutations. Moreover, western blotting assays performed ex vivo after exposure of uLMS to BAY1895344 demonstrated dephosphorylation of both ATR as well as CHK1, with a progressive increase in the induction of markers of apoptosis (i.e., caspase-3 proteins), compared to controls. Taken together these results demonstrate the impressive activity of BAY1895344 in inhibiting ATR activity and inducing apoptosis in uLMS.

This paper has a few important weaknesses. Without an appropriate in vitro model, we were not able to confirm effectivity of BAY1895344 on uLMS tumor cell lines outside of our PDX experiments. Additionally, we were not able to evaluate hematologic toxicity or other end organ toxicity directly in mice, only weight-based toxicity. Given that drug studies in humans are underway, more relevant toxicity of BAY1895344 is being evaluated [15].

In conclusion, we report for the first time the remarkable in vivo preclinical anti-tumor activity and the acceptable toxicity profile of BAY1895344 (Elimusertib) against biologically aggressive ATRX-mutated uterine leiomyosarcoma. Given that upwards of 51% of uLMS may express ATRX alterations, and an additional 19% may harbor DAXX gene mutations [10], ALT might represent a common feature of uLMS. Accordingly, we recommend all patients be offered genomic tumor profiling and that women with uLMS be included in early phase trials of ATR inhibitors in solid organ malignancies [22,23].

• Elimusertib (BAY1895344) is a novel ATR kinase inhibitor effective against uterine leiomyosarcoma

• Elimusertib decreased tumor growth in uterine leiomyosarcoma PDX mouse models with ATRX mutations

• Elimusertib prolonged overall survival in PDX mouse models of uterine leiomyosarcoma

• Tumor cells exposed to Elimusertib showed decreased phosphorylated-ATR and increased apoptotic molecules on western blot