Oncolytic virotherapy provides a potent therapy option for squamous bladder cancer

Abstract

Prognosis for squamous cell carcinoma (SCC) of the bladder is limited mostly because of lack of effective treatment regimens. Oncolytic virotherapy represents a promising option for bladder cancer and received in 2024 FDA therapy designation for the treatment of non-invasive high-grade bladder cancer (BLCA). For muscle-invasive bladder cancer (MIBC), preclinical studies demonstrated high efficacy of the oncolytic adenovirus XVir-N-31 in urothelial carcinoma (UC). We analyzed the potency of XVir-N-31 virotherapy as a novel treatment option in SCC. Replication of XVir-N-31 has been described to be facilitated by high expression level of Y-Box binding protein 1 (YB-1). Increased YB-1-mRNA expression was detected in basal/squamous subtype in TCGA BLCA cohort compared to urothelial and luminal BLCA and correlated with patient outcomes. Furthermore, immunohistochemical staining of 89 SCC on a tissue microarray confirmed strong YB-1 expression in squamous BLCA (sq-BLCA). In vitro, XVir-N-31 showed in subtype-specific cell cultures high rates of infection, replication and cell-killing capacity. In a novel in ovo xenograft model, XVir-N-31 impaired growth of xenografts of patient-derived ex vivo cell lines (p-SCC, p-UC) with growth suppression rates of 39-49%. We provide preclinical evidence ex vivo and in ovo for high efficacy of XVir-N-31 based oncolytic virotherapy as novel SCC therapy.

Keywords: Bladder cancer; Oncolytic virus; Squamous cell carcinoma; Virotherapy; XVir-N-31; YB-1.

Fig. 1

YB-1 mRNA expression in BLCA subtypes predicts patients’ outcome. (A,B) Independent BLCA data sets of the TCGA network, with (A) YB-1 mRNA expression classified by molecular subtypes: luminal (n = 23), luminal-infiltrated (n = 58), luminal-papillary (n = 110), basal/squamous (ba/sq) (n = 108) and neuroendocrine (n = 16). ***P < 0.001 (Kruskal–Wallis and Dunn’s multiple comparison tests). (B) Histological subtyping of TCGA data confirmed high YB-1 mRNA expression in sq-BLCA (n = 44) compared to UC (n = 271). **P < 0.01 (Mann–Whitney U test). (C–F) Kaplan–Meier survival curves illustrate (C,E) overall survival (OS) and (D,F) relapse-free survival (RFS) of BLCA patients with moderate to high YB-1 mRNA expression (red curve) compared to low YB-1 mRNA expression (blue curve) classified by molecular subtypes: (C,D) basal/squamous-like (ba/sq) (n = 98) (E,F) and luminal type (n = 164) based on TCGA BLCA data sets. YB-1^{moderate/high} in basal/squamous UC with mean OS 1832.7 days ± 251.0, 95% CI 1340.9–2324.6 and mean RFS 2659.1 days ± 303.9, 95% CI 2063.5–3254.8. YB-1^low in basal/squamous UC with mean OS 2495.8 days ± 418.8, 95% CI 1675.0–3316.6 and mean RFS 2482.2 days ± 430.9, 95% CI 1637.6–3326.8. YB-1^{moderate/high} in luminal UC with mean OS 2219.1 days ± 221.6, 95% CI 17845.0–2653.5 and mean RFS 2962.9 days ± 242.0, 95% CI 2488.6–3437.1. YB-1^low in luminal UC with mean OS 1467.1 days ± 242.1, 95% CI 992.6–1941.7 and mean RFS 1581.3 days ± 287.2, 95% CI 1018.5–2144.2. For all illustrated box plots: Horizontal lines: grouped medians. Boxes: 25–75% quartiles. Vertical lines: range, maximum and minimum.

Fig. 2

YB-1 expression and localization in sq-BLCA patient samples. (A–B) Representative images of immunohistochemically YB-1 staining in sq-BLCA patient tissues: (A) (i) negative YB-1 protein expression (ii) low/minimal YB-1 protein expression (intensity + 1), (iii) moderate/intermediate YB-1 protein expression (intensity + 2), (iv) high/strong YB-1 staining (intensity + 3). Scale bar 100 μm. (B) (i) YB-1 staining of nucleus and (ii) YB-1 staining of cytoplasm. Scale bar 50 μm. (C,D) YB-1 expression and distribution after IHC staining of sq-BLCA (n = 89) FFPE-sections. IHC scoring was estimated by percentage of YB-1 positive cells and (C) the intensity of positive staining in corresponding sections of sq-BLCA tumors. (D) IRS score was calculated as product of IHC intensity and percentage of positively stained cells according to Remmele and Stegner.

Fig. 3

Efficacy of YB-1 selective oncolytic virotherapy in SCC and UC cell culture models. (A–D) Characterization of patient-derived UC model (p-UC). (A) Histological characterization of original UC tissue with (i) HE staining, (ii) immunohistochemical anti-KRT5/6 staining, (iii) immunohistochemical anti-GATA3 staining. p-UC morphology was characterized by light microscopy (iv–vi) at different magnifications (20×, 50×, 100×). Black scale bars correspond to 500 μm and white to 100 μm. (B–D) Confirmation of urothelial character of p-UC and squamous-like character of p-SCC with marker expression of (B) KRT5- (C) KRT6- and (D) GATA3-mRNA compared to the original UC and SCC tumor with GAPDH as control. (E) Overall YB-1 protein level in uninfected BLCA cell lines (RT112, p-UC, SCaBER, p-SCC) was measured by immunoblotting. DOWN: YB-1 (50 kDa) and GAPDH (36 kDa) membranes. The raw immunoblot is presented in (Supplementary Fig. S1). UP: protein bands normalized at SCaBER YB-1 protein level and GAPDH. (F–H) Viral infection process: (F) BLCA cell lines (RT112, p-UC, SCaBER, p-SCC) were infected with 20–100 MOI of indicated viruses and harvested 4, 24, 48 and 72 hpi. After DNA phenol-chloroform extraction viral replication was assessed by qPCR to amplify viral fiber DNA. Values were normalized to the 4 hpi values. (G) Hexon titer test with representative staining (left) and detected virus titers (right). BLCA cell lines (RT112, p-UC, SCaBER, p-SCC) cells were infected with 10-200 MOI and harvested with supernatant 72 hpi to quantify formation of infectious particles. (H) SRB assay (n = 3–6 independent experiments) with representative staining (left) and detected SRB absorbance (right). BLCA cell lines (RT112, p-UC, SCaBER, p-SCC) cells were infected with indicated MOIs and stained with SRB five to six days after infection to measure cell-killing effect.

Fig. 4

YB-1 translocation and formation of adenovirus replication compartments post infection in BLCA cells. Immunofluorescence staining of YB-1 and E1A to visualize adenovirus replication compartments (AdRC). UC in vitro and patient-derived ex vivo models (A) RT112 and (B) p-UC and SCC in vitro and patient-derived ex vivo models (C) SCaBER and (D) p-SCC were not infected or infected with the indicated virus. Cells were fixed 48 hpi. E1A staining served as infection control. Scale bar 20 μm.

Fig. 5

Oncolytic virotherapy suppresses in ovo tumor growth of patient-derived BLCA cells. (A) Schematic illustration of the experimental timeline. Fertilized eggs were incubated till ED8 to applicate patient-derived cell culture models (p-UC, n = 61; p-SCC, n = 72). Treatment started at ED11 by adding either XVir-N-31 (127.75 MOI) or H₂O as control. Xenografts were harvested at ED12 (24 hpi), ED14 (72 hpi) and ED16 (120 hpi) to measure luminescence intensity and tumor weight. (B) XVir-N-31 treatment of CAM tumors reduces tumor size. Quantification of tumor weight from harvested CAM tumors derived from p-UC and p-SCC cells treated with H₂O or XVir-N-31 24, 72 or 120 hpi. H₂O treatment served as control. n = 9–16 for each measurement. *P < 0.05 (Kruskal–Wallis and Dunn’s multiple comparison tests). Scale bar 1000 μm. (C) Quantification of luciferase activity by measurement of bioluminescence images (left) taken from CAM tumors derived from p-UC and p-SCC cells treated with H₂O or XVir-N-31 were harvested 24, 72 or 120 hpi. H₂O treatment served as control. n = 9–16 for each measurement. *P < 0.05, **P < 0.01 (Kruskal–Wallis and Dunn’s multiple comparison tests). (D) Immunohistochemistry for Hexon on representative FFPE-sections of p-UC (i–iv) and p-SCC (v–viii) untreated (i, ii, v, vi) and XVir-N-31 treated (iii, iv, vii, viii) xenografts harvested 72 hpi after CAM assay. Scale bars correspond to 50 μm (2× and 20× magnification).