Identification of a high-risk immunogenic prostate cancer patient subset as candidates for t-cell engager immunotherapy and the introduction of a novel albumin-fused anti-cd3 × anti-psma bispecific design

ABSTRACT BACKGROUND Cancer immunotherapies such as bispecific T-cell engagers have seen limited adoption in prostate cancer (PC), possibly due to differing levels of cancer receptor

expression and effector T-cell infiltration between patients and inherent defects in T-cell engager design. METHODS CD8+ T-cell infiltration and PSMA expression were determined by RNA

sequencing of primary PC tissue samples from 126 patients with localised PC and 17 patients with metastatic PC. Prognostic value was assessed through clinical parameters, including CAPRA-S

risk score. A panel of albumin-fused anti-CD3 ×  anti-PSMA T-cell engagers with different neonatal Fc receptor (FcRn) affinity were characterised by flow cytometry, Bio-Layer Interferometry

and functional cellular assays. RESULTS A subset of patients with localised (30/126 = 24%) and metastatic (10/17 = 59%) PC showed both high PSMA expression and high CD8+ T-cell enrichment.

The High/High phenotype in localised PC associated with a clinically high-risk cancer subtype, confirmed in an external patient cohort (_n_ = 550, PRAD/TCGA). The T-cell engagers exhibited

tunable FcRn-driven cellular recycling, CD3 and PSMA cellular engagement, T-cell activation and PSMA level-dependent cellular cytotoxicity. CONCLUSION This work presents an albumin-fused

bispecific T-cell engager with programmable FcRn engagement and identifies a high-risk PC patient subset as candidates for treatment with the T-cell engager class of immuno-oncology

biologics. SIMILAR CONTENT BEING VIEWED BY OTHERS NOVEL IMMUNE ENGAGERS AND CELLULAR THERAPIES FOR METASTATIC CASTRATION-RESISTANT PROSTATE CANCER: DO WE TAKE A BITE OR RIDE BIKES, TRIKES,

AND CARS? Article 25 May 2021 IMMUNOGENOMIC PROFILES ASSOCIATED WITH RESPONSE TO LIFE-PROLONGING AGENTS IN PROSTATE CANCER Article 13 July 2023 TARGETING THE TUMOUR CELL SURFACE IN ADVANCED

PROSTATE CANCER Article 01 April 2025 BACKGROUND Multiple cancers have seen long-term death rate declines; however, prostate cancer (PC) remains the second leading cause of cancer death in

men in Western countries [1, 2]. Curatively intended treatments are offered with localised disease, but recurrence occurs in ~30% of the patients, who, therefore, require additional

therapeutic intervention [3, 4]. Palliative treatments are available such as androgen deprivation therapy (ADT) for metastatic PC, however, resistance eventually develops, and cancer becomes

classified as metastatic castration-resistant prostate cancer (mCRPC) with very poor prognosis [5]. Immunotherapy has yet to be successfully adopted for PC, with immune checkpoint

inhibition failing to improve overall survival in clinical trials of unselected patient populations, and the Sipuleucel-T cancer vaccine failing to widely impact clinical practice due to

modest benefit and high cost [6,7,8]. This likely reflects different cancer evasion mechanisms and immune environments between patients [9]. Bispecific T-cell engagers provide a conduit

between the patient’s cytotoxic CD8+ T-cells, commonly through CD3, and the tumour cells resulting in the formation of an immunological synapse and concomitant effector T-cell-mediated

killing of the cancer cell [10]. Interaction between CD3 and tumour antigen is independent of the canonical MHC-TCR pathway, thus, overcoming immune evasion mechanisms such as MHC-1

downregulation [11]. Efficacy is dependent on tumour resident CD8+ T-cells, the ratio of effector to target cells, and raised levels of target antigen in malignant tissue [12,13,14].

Multiple bispecific T-cell engagers targeting a variety of antigens are under development, with Blinatumomab (Blincyto®) approved for the treatment of acute lymphoblastic leukaemia [15]. The

design involves the removal of the fragment crystallisable (Fc) region to avoid adverse non-specific immune induction such as cytokine release syndrome (CRS) mediated by Fc interaction with

Fcƴ receptor-bearing immune cells [16]. This, however, results in loss of Fc engagement and neonatal Fc receptor (FcRn)-driven cellular recycling in the vasculature endothelium resulting in

a short plasma half-life and requirement for continual infusion into the patient [17]. The long circulatory half-life of human serum albumin (HSA) of ~19 days, is also facilitated by

engagement with FcRn [18] but unlike IgGs does not react with Fcƴ receptors, and has been utilised as a safe half-life extension technology in marketed albumin-based drug designs [19, 20].

Single-point amino acid mutations in albumin domain III, lying within the main binding interface with FcRn, has been shown to alter binding affinity and modulate the circulatory half-life

[21]. Our lab has recently engineered a panel of anti-CD3 × anti-EGFR bispecific T-cell engager-albumin genetic fusions engineered with different FcRn affinity for programming the half-life

as a method to maximise efficacy [22]. Prostate-specific membrane antigen (PSMA, encoded by _FOLH1_) is upregulated in most prostate cancers and utilised as a therapeutic and diagnostic

target [1, 23,24,25]. Several bispecific constructs targeting PSMA are currently under evaluation in Phase-1 clinical trials for PC. These include CC-1 [26], TNB-585 [27], CCW702 [28], and

AMG 160 [29] with selection criteria not based on PSMA and effector T-cell levels, that likely results in the inclusion of patients with either immunologically “cold” or PSMA negative

tumours. AMG 212 is an anti-CD3 × anti-PSMA scFv bispecific with promising Phase-1 clinical results, however, requirement of continuous infusion was necessary due to lack of Fc region and

the project terminated [30]. AMG 160 is a continuation of the AMG 212 bispecific design with the addition of a silenced Fc region improving half-life to 6.1 days [31]. In a recent Phase-1

clinical trial, 27% of patients treated with AMG 160 had confirmed Prostate-specific antigen (PSA) response, yet overall, the PSA response remained heterogeneous with ~a third of patients

being non-responders and 84.4% of patients experiencing CRS [29, 32]. This suggests that shortcomings remain and potentially could be overcome by modifying the bispecific T-cell engager

design and improving the selection of patients included in clinical trials [33]. PSMA levels combined with CD8+ effector T-cell infiltration presently remains underexplored. Identifying

PSMA/CD8+ T-cell correlation in PC tumours could lead to the personalised application of bispecific T-cell engagers in pre-selected patient subsets. In this work, we have designed, produced,

and characterised a panel of novel scFv anti-CD3 × human domain antibody anti-PSMA HSA-genetically fused bispecific T-cell engagers (termed AlproTox) engineered with either null-binding

(NB), wild-type (WT) or high-binding (HB) FcRn affinity to potentially tune circulatory half-life and give long-lasting effects. Furthermore, we have identified a high-risk immunogenic PC

patient subset as potential candidates for T-cell engager immunotherapy. METHODS PATIENT SAMPLES Total RNA sequencing was performed on 7 normal (N) prostate tissue samples, 31

adjacent-normal (AN) prostate tissue samples, 126 primary PC tissue samples (LPC) from patients who underwent curatively intended radical prostatectomy (RP) for clinically localised PC, and

17 primary tumour samples (MPC) from patients with metastatic PC who underwent palliative transurethral resection of the prostate (TURP). The normal prostate tissue samples were from

patients who underwent cystoprostatectomy due to bladder cancer and were histopathologically confirmed to be free of PC. AN samples were from LPC patients, who underwent RP [34]. All samples

were collected at Department of Urology, Aarhus University Hospital or Holstebro Hospital, as described previously [34]. Immediately following surgery, fresh prostate tissue biopsies were

obtained and stored at −80 °C in TissueTek. For RNA extraction, TissueTek samples were cut in ~40 sections (20-µm thickness) and the first and last sections stained with haematoxylin and

eosin (H&E) and evaluated by a trained pathologist to assess areas of malignant (PC) and benign (AN/N) prostate tissue. Total RNA was extracted from the remaining ~38 sections using the

RNeasy Plus Mini Kit (Qiagen, Hilden, Germany), as described previously [35]. RNA concentration and quality was assessed using a NanoQuant and an Agilent 2100 Bioanalyzer (RIN ≥ 7),

respectively. RNA SEQUENCING AND XCELL ANALYSES Sequencing libraries were generated using either the ScriptSeq RNA-seq Library Kit with the Ribo-Zero Magnetic Gold Kit from Illumina (San

Diego, California, USA) (N = 7; AN = 11; LPC = 52) or the KAPA RNA HyperPrep Kit with KAPA RiboErase Kit from Roche (Basel, Switzerland) (AN = 20; LPC = 74; MPC = 17). All libraries were

sequenced paired-end on either an Illumina HiSeq 2000, NextSeqTM 500, or NovaSeq 600 (~25 million reads per sample). Transcript expression was quantified using Kallisto (v.46.2) [36] with

GRCh38/hg38 as reference transcriptome, and subsequently aggregated to gene level using tximport (v1.16.1) [37]. Trimmed Means of M-values (TMM) normalisation, filtering, and log2

transformation was performed on gene level counts using edgeR (v.3.30.3) [38]. Batch effect correction was performed using the removeBatchEffect function in the R package Limma (v3.44.3)

[39], and batch effect adjusted data was used for subsequent analyses. Relative cell-type enrichment of CD8+ T-cells was estimated from the total RNA sequencing data using xCell in

R(v.1.1.0) [40]. xCell is a gene expression signature-based approach to determine cell-type enrichment from bulk tissue RNA expression data, used here to explore the enrichment of a CD8+

T-cell signature. High expression of a specific signature results in a high enrichment score for the corresponding cell type. SPATIAL TRANSCRIPTOMICS AND DATA ANALYSIS Spatial

transcriptomics (ST) was performed on three representative prostate tissue samples from three patients included in the MPC cohort. An experienced pathologist confirmed the presence of

malignant lesions within the tissue sections based on H&E staining. ST was performed on an adjacent tissue section (10 μm) using the Visium spatial gene expression methodology following

the manufacturer’s instructions (10X genomics, Pleasanton, California, USA, #CG000239). A detailed description of the ST methodology is available elsewhere [41]. The 10X Visium spatial gene

expression slide applies thousands of 55-μm barcoded spots within a 6.5 × 6.5 mm area to capture mRNA. The captured mRNA was sequenced on an Illumina NextSeq platform. Space Ranger (10X

genomics) pipeline was used to align the reads to the human reference genome (hg38) and for obtaining the coordinates for each transcript. Data analysis was performed in R [42] using the

Seurat package [43]. Spots with less than three detected genes were filtered out, and the data were normalised. The presence or absence of CD8A (CD8 alpha chain) and FOLH1 (PSMA) was

determined using this dataset and visualised using the SpatialDimPlot function. Venn diagrams were generated using the ggvenn [44] R package. STATISTICAL ANALYSIS Recurrence-free survival

analysis was conducted using the survminer package [45] (v.0.4.8) in RStudio (v. 1.0.153) using the Kaplan–Meier methods. Log-rank tests were used to determine significance with the clinical

endpoint being biochemical recurrence (BCR), defined as PSA ≥ 0.2 ng/mL following RP. _P_-values below 0.05 were considered significant. PSMA expression and CD8+ T-cell infiltration in

prostate tissue samples were compared with Wilcoxon rank-sum test [46]. When patients were dichotomised (high/high versus remaining), comparisons to clinical parameters (BCR, surgical margin

status, pre-surgery PSA, CAPRA-S risk Score, T-stage and Gleason grade) were conducted using Fisher’s exact test [47]. To determine the correlation between PSMA expression and CD8+ T-cell

infiltration, Pearson’s correlation coefficient [48] was calculated. Experimental data were analysed using the Origin 2018 software and GraphPad Prism software v.8. The respective used

statistical analysis is indicated in the corresponding figure and table legends. A minimum value of _P_ < 0.05 was considered statistically significant. All experiments were conducted at

least two times. EXTERNAL PC PATIENT DATASET For external validation, the Prostate adenocarcinoma (PRAD) tumour RNA-seq dataset from The Cancer Genome Atlas (TCGA) [49] was downloaded from

the TCGA data portal [50], as described previously [51]. PSMA (FOLH1) expression data and clinical data were available for 498 LPC and 52 AN prostate tissue specimens. CELL LINES AND CULTURE

CONDITIONS LNCaP (ATCC, CRL—1740), Du-145 (ATCC, CRL—HTB—81), PC3 (ATCC, CRL—1435), C4-2 (ATCC, CRL—3314), HEK293E (ATCC, CRL—10852) and Jurkat cells (ATCC, TIB-152) were cultured in RPMI

medium 1640 (Gibco, Waltham, Massachusetts, USA #61870-010) supplemented with 10% foetal bovine serum (FBS) (Gibco, #10500-064) and 1% penicillin–streptomycin (Gibco, #15140-122). Human

dermal microvascular endothelial cells (HMEC-1) stably overexpressing human FcRn (HMEC-1-FcRn) [18] were cultivated in complete medium consisting of MCDB131 (Life Technologies, Waltham,

Massachusetts, USA #10372-019), Recombinant human EGF (PeproTech, Cranbury, New Jersey, USA #AF-100-15), Hydrocortisone (Sigma-Aldrich, St. Louis, Missouri, USA #H0888), G418 solution

(Sigma-Aldrich, #G418-RO), Puromycin (Life Technologies, #A11138-03), 10% FBS, and 2 mM l-glutamine (Lonza, Basel, Switzerland #BE17-605E). Cells were cultivated following ATCC protocols,

and mycoplasma tested when new stock were thawed. Cell lines were routinely authenticated by short tandem profiling. MORPHOLOGICAL CHARACTERISATION BY ATOMIC FORCE MICROSCOPY (AFM) AlproTox

(WT, NB, or HB) solution (1 µg/ml) was deposited on the positively charged mica surface, pretreated by (3-aminopropyl) triethoxysilane vapour (APTES; Sigma-Aldrich, #919-30-2), and, scanned

in peak force tapping mode in the fluid using a MultiMode VIII (Bruker, Santa Barbara, California, USA) by SNL-10C (Bruker) cantilever with a spring constant of 0,24 N/m and a tip radius of

2 nm. All parameters were optimised during imaging to avoid protein deformation. Data were processed using SPIP (Image Metrology, Lyngby, Denmark). NEONATAL FC RECEPTOR ENGAGEMENT KINETICS

AND CELLULAR RECYCLING Bio-Layer Interferometry studies were conducted using the Octet Red96 system (FortéBio, Fremont, California, USA). Biotinylated human FcRn (hFcRn) (Immunitrack,

Copenhagen, Denmark # ITF02) was immobilised on streptavidin-coated biosensors (Sartorius AG) in 0.01% PBST, pH 7.4. Wild-type (WT), null-binding (NB) or high-binding (HB) recombinant human

albumin (rHSA) variants were used as controls for the albumin fusions. Kinetic measurements were performed in 25 mM NaOAc, 25 mM NaH2PO4, 150 mM NaCl, 0.01% PBST, pH 5.5 with sevenfold

serial dilutions at maximum 3 µM concentration. A baseline was established in buffer before sensors were transferred to buffer containing analyte for 300 s followed by 600 s of

disassociation in sample-free association buffer and 240 s of regeneration in PBST. Data analysis was performed using the Octet data analysis software ver. 10.0.1.6 (FortéBio) using a 1:1

binding model for curve fitting to estimate the kinetic parameters with all data referenced with FcRn-streptavidin sensors in wells containing only buffer. Cellular recycling of the AlproTox

or rHSA panel (0,1 µM) was performed in HMEC-1-FcRn overexpressing cells using a cellular recycling assay previously described [18]. CELLULAR ANTIGEN RECOGNITION LNCaP, C4-2, Du-145, PC3 or

Jurkat cells were cultivated, and 1 × 105 cells were seeded into a U-bottom NunclonTM plate and incubated on ice with dilution series of ice-cold assay buffer (PBS + 1% FBS) supplemented

with either AlproTox, mouse anti-human CD3 FITC-conjugated antibody (Immunotools, Friesoythe, Germany #21850033), or PE-conjugated mouse monoclonal anti-PSMA antibody (Abcam, #ab77228).

Following incubation, cells were pelleted and washed two times in 200 µL assay buffer before the addition of secondary antibody: rabbit polyclonal antibody to human albumin (Abcam, #ab2406).

Plates were incubated and the wash routine was repeated before the tertiary antibody, goat anti-rabbit IgG FITC-conjugated antibody (Invitrogen, #A11034), was added with subsequent 30 min

of incubation on ice. Cells were washed and resuspended in 200 µL 1:200 7-AAD live/dead cell stain (Invitrogen, Waltham, Massachusetts, USA #S10274) before samples were run on a Novocyte

flow cytometer (ACEA Biosciences Inc, Santa Clara, California, USA). Control wells consisted of secondary and tertiary antibody. Data analysis was performed using FlowJo™ v10.0.7 Software

(BD Life Sciences, Franklin Lakes, New Jersey, USA). T-CELL ACTIVATION AND CD69 UPREGULATION Anti-human CD3 antibody clone OKT3 (Biolegend, Waltham, Massachusetts, USA #317302) was added to

a 96 Nunc MicroWell plate (Thermo Fisher Scientific, Waltham, Massachusetts, USA #163320) at 10 µg/mL and left overnight at 4 °C. In all, 4 × 104 LNCaP, C4-2, Du-145 or PC3 cells were seeded

in triplicates together with a serial dilution of the AlproTox panel at a maximum 100 nM together with 2 × 105 Jurkat cells. As a positive control, Jurkat cells seeded in the OKT3-coated

wells was used. Following 24 h of incubation, cells were spun down and resuspended in 50 µL assay buffer (PBS + 1% FBS) before washing and addition of 100 µL anti-CD69 FITC (ImmunoTools,

#21620693×2) at 1:100 dilution. Cells were incubated at 30 min, before 2× washing and resuspending in 200 µL 1:200 7-AAD before running the samples on a Novocyte flow cytometer (ACEA

Biosciences Inc). Data analysis was performed using FlowJo™ v10.0.7 Software (BD Life Sciences). LDH CYTOTOXICITY ASSAY In total, 4 × 105 target cells (LNCaP, C4-2, Du-145 or PC3) were

seeded onto 96-well TC-treated flat-bottomed plate and incubated at 37 °C, 5% CO2 for 24 h. Freshly isolated PBMCs from healthy donors were added at a 5:1 concentration to target cells along

with a dilution series of AlproTox WT, NB or HB at maximum 100 nM in assay media (RPMI 1640 + Glutamax, 10% FBS, 1% penicillin–streptomycin) and co-incubated for 48 h. Separate experimental

repeats used different PBMC donors. 1% final concentration Triton X-100 (Cell Biolabs Inc, San Diego, California, USA #124102) was added to high control wells for 15 min for cell lysis

verified by light microscopy before spinning down cells at 600×_g_ and transferring 100 µL supernatant to a 96-well Nunc MaxiSorpTM flat-bottomed plate in duplicates for a total of two

technical replicates for each experimental repeat. As low control, untreated wells containing target and effector cells only were used. The LDH cytotoxicity detection kit (TaKaRa, Kusatsu,

Japan #MK401) was used according to the manufacturer’s instructions. Absorbance was measured at 620 nm with 492 nm as background by a CLARIOstarTM (BMG LABTECH, Ortenberg, Germany).

Cytotoxicity was calculated following the equation, cytotoxicity (%) = ((experiment value – low control)/(high control – low control)) × 100. REPORTING SUMMARY Further information on

research design is available in the Nature Research Reporting Summary linked to this article. RESULTS CD8+ T-CELL INFILTRATION AND PSMA EXPRESSION IN PROSTATE TUMOURS Total RNA sequencing

data from seven normal (N) prostate tissue samples, 31 adjacent-normal (AN) prostate tissue sample as well as primary PC tissue samples from 126 patients with clinically localised PC (LPC)

and 17 patients with metastatic PC (MPC) was used for quantification of PSMA expression. In addition, the xCell algorithm was applied to the RNA-seq data for cell-type enrichment analysis to

estimate the levels of infiltrating CD8+ T-cells. To visualise the spatial distribution of CD8A + (gene encoding CD8) and FOLH1 + (gene encoding PSMA) spots, we performed ST on three

representative prostate tissue samples from the MPC patient cohort. For external validation, we used RNA-seq data from the TCGA-PRAD cohort, including 498 LPC samples and 52 AN samples. For

patient characteristics, refer to Supplementary Table 1. PSMA was overexpressed in both MPC and LPC samples, as compared to AN prostate tissue samples (_P_ = 0.018 and _P_ < 0.001

respectively, Wilcoxon test, Fig. 1a), and as compared to normal (N) prostate tissue samples (_P_ = 0.011 and _P_ < 0.001, respectively, Wilcoxon test, Fig. 1a). There was no significant

difference in PSMA expression between LPC and MPC samples in this patient cohort (_P_ = 0.39, Wilcoxon test, Fig. 1a). PSMA was also overexpressed in LPC as compared to AN samples in the

TCGA dataset (_P_ < 0.001; Wilcoxon test; Supplementary Fig. s1a). CD8+ T-cell enrichment score was increased in MPC samples as compared with all other sample types analysed (_P_ < 

0.05, Wilcoxon test, Fig. 1b). There was no significant difference in CD8+ T-cell enrichment score between N, AN and LPC samples (_P_ > 0.05, Wilcoxon test, Fig. 1b). Similarly, there was

no significant difference in CD8+ T-cell enrichment between AN and LPC samples in the external TCGA cohort (_P_ > 0.05; Wilcoxon test; Supplementary Fig. s1b). There was no significant

correlation between overall PSMA expression and CD8+ T-cell infiltration in MPC or LPC samples (_P_ = 0.17 and _P_ = 0.3, respectively; Pearson, Supplementary Fig. s2), and a similar result

was found in the TCGA dataset (Supplementary Fig. s3). A focus was to identify patients with a high PSMA expression and a high CD8+ T-cell infiltration profile as potential candidates for

bispecific T-cell engager treatment. Hence a score designated High/High defined as > median PSMA expression and > median CD8+ T-cell infiltration was determined (cut-offs based on

median in tumour samples from LPC patients). 59% of patients with MPC (_n_ = 10/17) scored as High/High, as well as 24% of patients with LPC (_n_ = 30/126) (Fig. 1c). Similarly, in the TCGA

cohort 25% (_n_ = 126/498) of patients with LPC scored as High/High (Supplementary Fig. s1c). ST validated the presence of CD8+ cells in the prostate tissue from all three MPC patients (Fig.

 1d–f), corroborating the enrichment of CD8+ T-cells (xCell) as identified from bulk tumour RNA-seq in this study. FOLH1 (PSMA) expression was heterogeneous in the analysed tissue but seemed

predominant in the High/High samples with total proportion of FOLH1+ spots being 91.8% (e) and 17.1 % (f) compared to the 9.1% in tissue samples belonging to remaining group (d). Whilst the

total overlap between FOLH1+ and CD8A+ spots appears low (2.6–3.9%), the proportion of double positive spots, with respect to CD8A+ spots, was considerable (9.1–95.5%). Furthermore, the two

largest overlaps were found in the High/High subgroup, 95.5% (e) and 21.2% (f), while the lowest overlap was seen in the remaining group, 9.1% (d). ASSOCIATION OF CD8+ T-CELL INFILTRATION

AND PSMA EXPRESSION WITH HIGH-RISK PROSTATE CANCER Next, we assessed the prognostic value of having both high CD8+ T-cell infiltration and high PSMA expression in LPC patients (High/High

phenotype, _n_ = 30). The remaining patients (_n_ = 96), with below median expression of PSMA and/or below median CD8+ T-cell infiltration, were assigned to a separate group. Biochemical

recurrence status (BCR, defined as PSA > 0.2 ng/mL), surgical margin status, and pre-surgery serum PSA levels (PSA < 10 ng/mL = low, PSA ≥ 10 ng/mL = high) were not statistically

different between the two groups (Fisher’s exact test, _P_ > 0.05, Fig. 2). Kaplan–Meier analysis of BCR-free survival also did not show significant differences between the two subgroups

(Supplementary Fig. s4). Similar results were observed in the TCGA cohort (Supplementary Figs. s5 and s6). However, for all these factors, there was a moderate trend towards more high-risk

disease characteristics in the High/High subgroup. In support of this, there was a significant association between the High/High subgroup and more advanced T stage (_P_ = 0.025, Fisher’s

exact test) as well as a significant association with higher Gleason Grade _(P_ = 0.0143, Fisher’s exact test, Fig. 2). To improve statistical power, we used the established clinical

risk-stratification tool, USCF Cancer of the Prostate Risk Assessment Post-Surgical (CAPRA-S) score [52], to categorise the patients into a high-, intermediate-, and low-risk PC subgroup.

Here, there was a significant association between a higher CAPRA-S score and the High/High phenotype, defined by high PSMA expression and high CD8+ T-cell infiltration (_P_ = 0.004, Fisher’s

exact test). These findings were also supported in the TCGA cohort, where the High/High phenotype was significantly associated with a higher CAPRA-S risk score (_P_ < 0.001, Fisher’s

exact test, Supplementary Fig. s6) and higher Gleason Grade (_P_ < 0.001, Fisher’s exact test; Supplementary Fig. s6) as well as borderline associated with advanced T stage (_P_ = 0.099,

Fisher’s exact test, Supplementary Fig. s6). To determine if CD8+ T-cell infiltration correlated to clinical parameters independently of PSMA expression, LPC patients were stratified into a

High vs. Remaining subgroup, based solely on median CD8+ T-cell infiltration. No clinical parameters were significant for the LPC cohort, although the same trend towards more high-risk

disease characteristics, as seen for the High/High versus Remaining subset (Fig. 2), was apparent (Supplementary Fig. s7). For the larger TCGA cohort, higher CAPRA-S risk score (_P_ = 0.044,

Fisher’s exact test), higher Gleason Grade (_P_ < 0.001, Fisher’s exact test) and positive margin status (_P_ = 0.038, Fisher’s exact test) were all significantly associated with high

CD8+ T-cell infiltration (Supplementary Fig. s8). In summary, results from two independent PC patient cohorts indicate that a subset of patients with localised and metastatic disease express

high levels of PSMA and have high CD8+ T-cell infiltration levels (High/High phenotype). The High/High phenotype seemed to be common in advanced MPC. Furthermore, the patients with

localised disease in this subset had clinical characteristics known to be associated with higher-risk cancer, providing rational for treatment with bispecific T-cell engagers following

radical prostatectomy. EXPRESSION OF ALPROTOX AlproTox vector constructs containing an anti-CD3 × anti-PSMA bispecific antibody were genetically fused to the N-terminus of wild-type (WT) HSA

(AlproTox WT), or mutated HSA variants for null-binding (NB) (AlproTox NB) or high-binding (HB) (AlproTox HB) to FcRn. These were transiently transfected into HEK293E cells and cultivated

in serum-free media. The AlproTox fusions were purified using affinity chromatography (Supplementary Fig. s9) and collected eluate fractions were analysed with ELISA (Supplementary Fig. 

s10). Final samples analysed with Coomassie staining (Supplementary Fig. s10), and western blotting confirmed high purity of samples (Supplementary Fig. s11). Protein yields of 0.59 mg/L,

0.76 mg/L and 0.54 mg/L were obtained for AlproTox WT, NB and HB, respectively. MORPHOLOGY, FCRN AFFINITY AND CELLULAR RECYCLING The panel of fusions (Fig. 3a–c) displayed uniform discrete

particle characteristics of similar nanoscale size ranging from: 1.10 ± 0.51 nm, 1.58 ± 0.59 nm, and 1.84 ± 0.77 nm for AlproTox WT, AlproTox NB and AlproTox HB, respectively (Fig. 3d–f).

The AlproTox HB and WT fusions displayed a clear pH-dependent affinity towards human FcRn (hFcRn) (KD 5.3 × 10−9 M and KD 1.1 × 10−8 M, respectively) with a profile comparable to that of

recombinant HB and WT HSA non-fused albumin controls (Fig. 3g–j). Both AlproTox HB and AlproTox WT, as well as the respective recombinant HSA (rHSA) variant counterparts, could be fitted to

a 1:1 binding model with results displayed in Table 1. AlproTox NB and HSA NB, as expected, could not be fit to a 1:1 binding model, and as such are not shown. The level of AlproTox recycled

in endothelial HMEC-1-FcRn cells correlated with measured hFcRn affinity, following the trend AlproTox HB (0.460 nM) > AlproTox WT (0.2090 nM) > AlproTox NB (0.054 nM) (Fig. 3k–l).

CELLULAR ANTIGEN RECOGNITION The AlproTox panel bound specifically to PSMA-expressing PC cell lines (LNCaP, C4-2) and CD3 expressing T-lymphocyte (Jurkat) cells at concentrations of 1 µg/mL

demonstrated by a shift in fluorescent intensity using commercial antibodies (Fig. 4). No binding was detected for the PSMA negative PC cell lines (PC3, Du-145). The AlproTox fusions

displayed similar shifts in fluorescence, independent of hFcRn affinity, and shifts in fluorescence were detected only for PSMA and CD3 expressing cell lines, demonstrating binding

specificity. T-CELL ACTIVATION AND T-CELL-MEDIATED CYTOTOXICITY CD69 expression as a measure of T-cell activation strongly correlated with the concentration of AlproTox and PSMA expression

on the target cells. The target cells displayed concentration-dependent cell lysis when incubated with AlproTox WT, NB or HB in the presence of PBMCs, whilst no cell lysis was detected for

PSMA negative cell lines incubated with AlproTox WT or NB (Fig. 5d–f). The EC50 ranged from 0,44 nM (NB), 0.39 nM (WT) to 0,16 nM (HB), with AlproTox HB displaying some degree of cell lysis

of the PSMA negative Du-145 and PC3 cell lines (Fig. 5f). DISCUSSION Bispecific T-cell engagers that drive redirection of the effector T-cell response, provides a potential strategy to

overcome the multiple resistance mechanisms seen in advanced PC [9, 30]. However, the investigation into selecting PC patients most likely respondent to T-cell engagers is lacking and is one

of the main hurdles to bispecific T-cell engager adoption. In this study, we aimed to identify patients who could be candidates for T-cell engager immunotherapy and introduce a novel

anti-CD3 x PSMA albumin-fused bispecific T-cell engager with tunable FcRn-driven cellular recycling for programming the circulatory half-life and consequent duration of effect. RNA

sequencing of tumour samples from PC patients showed increased PSMA expression in the prostate of PC-diagnosed patients and increased CD8+ T-cell infiltration in patients with metastatic

disease (Fig. 1a–c). Whilst the clear prognostic value of PSMA is established [23], it is not the case for T-cell infiltration. Some studies associate CD8+ T-cell infiltration to longer

biochemical recurrence-free survival [53], others finding no significant effect [54], and others like Ness et al. finding clear association with shorter biochemical recurrence-free survival

following radical prostatectomy [55]. _Hitherto_, no studies have investigated the co-localisation of PSMA and CD8+ T-cell infiltration in PC tumours and the combined impact on disease

aggressiveness and biochemical recurrence in localised disease patients. From the cohort of patients investigated, we found a subgroup of MPC patients (59%, _n_ = 10/17), expressing high

PSMA and high CD8+ T-cell infiltration in the primary tumour. ST analysis confirmed CD8+ cell infiltration in MPC prostate with 9.1–95.5% of CD8A+ positive spots overlapping with FOLH1+ 

spots, indicating the proximity of effector T-cells and PSMA-expressing cancer cells in the tumour tissue (Fig. 1d–f). Unfortunately, it was not possible to supplement the ST images with

immunohistochemistry for single-cell image resolution to verify spatial location. However, The ST images alone suggest clinical utility for bispecific T-cell engagers that are reliant on the

formation of an immunological synapse between cancer and T cell, and, thus, require direct cell-cell interaction. Furthermore, the motility of infiltrating CD8+ T-cells within tumour tissue

[56] may allow for sequential cell-cell interaction in a high PSMA-expressing environment, predominant in the High/High subset. A total of 24% (_n_ = 30/126) of LPC patients exhibited a

similar High/High profile and were stratified according to these parameters with the High/High subgroup showing significantly higher CAPRA-S Score, advanced T stage and higher Gleason grade,

indicative of a high-risk form of prostate cancer (Fig. 2). The findings were validated in an independent PC tumour RNA-seq dataset (TCGA-PRAD) [49], where a similar High/High subgroup was

identified (25%, _n_ = 126/498, Supplementary Fig. s1), that also exhibited significantly higher CAPRA-S Score, (borderline) T stage and Gleason grade (Supplementary Fig. s6). A similar

association with clinical parameters was found for the TCGA cohort when stratified only on CD8+ T-cell infiltration albeit with higher _P_-values for all parameters except margin status,

which was now significant (Supplementary Fig. s8). For the LPC patients, however, no significant associations with the clinical parameters were found (Supplementary Fig. s7), indicating that

the combination of PSMA expression and CD8+ T-cell infiltration can more accurately identify the high-risk patient subtype. The inconclusive influence on tumour characteristics and

inconsequential effect on BCR of tumour infiltrating CD8+ T-cells is in line with the inconsistent nature of prior study results on the subject [57]. T-cell infiltration has been shown

correlated to mutational burden in multiple cancers, with high mutational burden as well as mutations in mismatch repair genes correlating to high T-cell infiltration [58, 59]. While genomic

analysis is beyond the scope of this paper, further characterisation of the identified high-risk prostate cancer subset in regard to specific mutations and mutational load are warranted and

should be included in future studies. Current European clinical guidelines [60] recommend multimodal treatment of localised high-risk prostate cancer and ADT, commonly used to treat locally

advanced and metastatic cancer, has been shown to cause an influx of CD8+ T-cells in the primary tumour [61]. This could, in part explain the higher level of CD8+ T-cell infiltration seen

in the prostate tissue of the MPC patients when compared to the LPC patients (Fig. 1) that were all treatment naïve prior to RP. The impact of immunotherapy in combination with ADT on

treatment efficacy is currently being evaluated in a Phase II/III clinical trial of metastatic patients [62], and whilst metastatic patients should indeed be considered the highest treatment

priority for novel drugs, the identified subset of LPC patients (High/High) also stand to benefit from additional treatment and may be responsive to treatment with an anti-CD3 x PSMA

bispecific T-cell engager. A limitation of the study was the inability to examine anti-tumour efficacy based on pretreatment CD8+ T-cell infiltration. While it is plausible that the

localised presence of readily available CD8+ T-cells may potentiate treatment with bispecific T-cell engagers in solid tumours, this hypothesis has only been evaluated in a few studies of

non-haematological malignancies. Ströhlein et al. found that relative lymphocyte count was a direct positive prognostic parameter in peritoneal carcinomatosis patients treated with

catumaxomab, a CD3 targeting bispecific antibody, in a Phase II clinical trial [63]. In support of this, Belmontes et al. found that the most important factor for bispecific T-cell engager

efficacy in multiple solid-tumour in vivo models was pretreatment T-cell density in the tumour, with CD8+ T-cells being the most important mediator of bispecific T-cell engager activity

[14]. Finally, although based on different classes of immunotherapeutics, parallels can be drawn to immune checkpoint inhibitors that also rely on CD8+ T-cell cytotoxicity to mediate effect

on which there is a more extensive body of literature. In a multiomics analysis of 21 types of cancer, including prostate adenocarcinoma, a study found that the best indicator, out of 36

variables, to anti-PD-1/PD-L1 therapy was CD8+ T-cell abundance in the tumour [64]. While the success of checkpoint inhibitors has been modest in the treatment of prostate cancer, Graff et

al. in a small Phase II trial (_n_ = 10) described two mCRPC patients that showed robust response to anti-PD-1 treatment with baseline tumours having CD8+ T-cell infiltration [65]. The

higher rate of immunotherapeutic response rates seen in tumours with CD8+ T-cell infiltration, coupled with the heightened risk profile identified in our study, therefore, lends plausibility

to the identified patient subset in our work being potential treatment candidates. However, further studies on a possible role of tumour infiltrating CD8+ T-cells in bispecific T-cell

engager treatment would be interesting. Another limitation was the inability to explore the PSMA expression and CD8+ T-cell infiltration in metastasis biopsies, despite these sites being

relevant targets for successful disease treatment. Both Queisser et al. and Wright et al. demonstrated increased PSMA expression in metastatic lesions compared to localised tumour [66, 67].

Furthermore, Sartor et al. demonstrated in their Phase III clinical trial of 177Lu-PSMA-617 that ~86.6% (_n_ = 869/1003) of mCRPC patients imaged were PSMA-positive, indicative of the high

prevalence of PSMA-positive lesions at later stages of PC [68]. Still, correlation with CD8+ T-cell infiltration was unknown [66, 67]. The results in our work, however, have identified, for

the first time, a high-risk PC cohort, encompassing patients with localised and metastatic prostate cancer that could benefit from treatment with an anti-CD3 ×  PSMA bispecific engager.

While better stratification may improve therapeutic effect, inherent defects in bispecific T-cell engager design, namely short blood circulatory half-life, and adverse effects in the form of

CRS are hindering clinical development. Despite Fc silencing, residual binding to the Fcγ receptor has been shown [69] and could potentially lead to overzealous immune stimulation.

Furthermore, the large number of mutations required to silence the Fc domain may increase the risk of anti-drug antibody generation and protein instability [70]. Albumin is an attractive

half-life extension alternative that has reached the clinic [19, 20] driven by FcRn-mediated cellular recycling [18] without induction of Fcγ innate immunity. We introduce a novel anti-CD3 x

PSMA albumin-fused bispecific T-cell engager with tunable FcRn affinity that may offer programmable pharmacokinetics (PK) and control of therapeutic effects. The panel of AlproTox fusions

(null-binding (NB), wild-type (WT) or high-binding (HB)) were expressed in HEK293E cells with yields ranging from 0.54 mg/L to 0.76 mg/L. FcRn affinity was dependent on the incorporated

albumin sequences shown previously to modulate FcRn binding [71], which followed the trend AlproTox HB (KD 5.3 × 10−9 M) > AlproTox WT (KD 1.1 × 10−8 M) > AlproTox NB. This correlated

with levels recycled in the endothelial cellular recycling assay (Fig. 3k–l). Vasculature endothelial cellular recycling and concomitant extended circulatory half-life of albumin is mediated

by an cellular endosomal sorting process diverting albumin from lysosomal breakdown facilitated by higher FcRn engagement at low endosomal pH and release at physiological pH at the cell

surface [18]. Results from our group have shown that FcRn affinity and FcRn-driven cellular recycling correlates with the circulatory half-life of other bispecific (EGFR × CD3) albumin

fusions with the inclusion of identical albumin sequences to the AlproTox fusions in this study [22]. It is, therefore, expected that the HB > WT > NB circulatory half-life trend will

be observed in vivo with the AlproTox design. Specific binding against PSMA and CD3 was verified using flow cytometry (Fig. 4), with the panel exhibiting a similar binding profile. All

albumin fusions were able to activate T cells when co-incubated with PSMA-expressing cell lines (Fig. 5a–c) and mediate T-cell cytotoxicity against PSMA-expressing prostate cancer cell lines

(Fig. 5d–f). EC50 was within one order of magnitude across the panel, being highest for AlproTox NB (0.44 nM) and lowest for AlproTox HB (0.16 nM). The T-cell-mediated cell-killing efficacy

is comparable to published AMG 160 results where EC50 values ranged from 6 to 42 pmol/mL, dependent on cell line [31]. While NB and WT were highly specific, cell killing was detected for

AlproTox HB in PSMA negative cell lines, that may indicate binding towards endogenous FcRn expressed at low levels in the Du-145 and PC3 prostate cancer cell lines [72]. AFM analysis

revealed discrete AlproTox constructs at the nanoscale that supports the possibility of these systems to penetrate solid tumours (Fig. 3d–f). Furthermore, exploitation of PSMA-targeting may

mediate delivery to both primary sites and metastatic lesions. Future work should investigate the PK in a physiologically relevant double transgenic human FcRn+/+/HSA+/+ mouse model

introduced by our group [73] that abolishes competition from endogenous mouse albumin associated with wild-type strains. Furthermore, we have developed a _RAG1_ knockout FcRn+/+/HSA+/+

strain that we have used to investigate T-cell engager efficacy [22] that should be applied to AlproTox anti-PC tumour investigations. The availability of a panel of HSA constructs with

different PK may offer a tool to tune the dosage requirements and therapeutic effects. The well-documented entry of HSA at tumour sites by both passive [74] and active [75] processes also

offers the advantage of improved accumulation at tumour sites. In summary, we have investigated the PSMA expression and CD8+ T-cell infiltration in primary tumour samples of PC patients from

two independent cohorts and identified a patient subset with high-risk local disease or metastatic cancer that may be susceptible to T-cell engager therapy. Precise stratification of

patients may improve clinical response rates and, in combination with novel bispecific designs, lead to the successful adoption of bispecific T-cell engagers in PC. DATA AVAILABILITY All

data are available from the corresponding author upon reasonable request. CODE AVAILABILITY Code can be made available by making a request to the corresponding author. REFERENCES * Parker C,

Castro E, Fizazi K, Heidenreich A, Ost P, Procopio G, et al. Prostate cancer: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Annal Oncol. 2020;31:1119–34. Article

  CAS  Google Scholar  * Siegel RL, Miller KD, Jemal A. Cancer statistics, 2020. CA Cancer J Clin. 2020;70:7–30. * Kolodziej M. Management of biochemically recurrent prostate cancer

following local therapy. Am J Managed Care. 2014;20:S273–81. Google Scholar  * Lin X, Kapoor A, Gu Y, Chow MJ, Xu H, Major P, et al. Assessment of biochemical recurrence of prostate cancer

(review). Int J Oncol. 2019;55:1194–212. CAS  PubMed  PubMed Central  Google Scholar  * He L, Fang H, Chen C, Wu Y, Wang Y, Ge H, et al. Metastatic castration-resistant prostate cancer:

academic insights and perspectives through bibliometric analysis. Medicine. 2020;99:e19760. Article  PubMed  PubMed Central  Google Scholar  * Topalian SL, Hodi FS, Brahmer JR, Gettinger SN,

Smith DC, McDermott DF, et al. Safety, activity, and immune correlates of anti-PD-1 antibody in cancer. New Engl J Med. 2012;366:2443–54. Article  CAS  PubMed  Google Scholar  * Gulley JL,

Borre M, Vogelzang NJ, Ng S, Agarwal N, Parker CC, et al. Phase III trial of PROSTVAC in asymptomatic or minimally symptomatic metastatic castration-resistant prostate cancer. J Clin Oncol.

2019;37:1051–61. Article  CAS  PubMed  PubMed Central  Google Scholar  * Kwon ED, Drake CG, Scher HI, Fizazi K, Bossi A, van den Eertwegh AJ, et al. Ipilimumab versus placebo after

radiotherapy in patients with metastatic castration-resistant prostate cancer that had progressed after docetaxel chemotherapy (CA184-043): a multicentre, randomised, double-blind, phase 3

trial. Lancet Oncol. 2014;15:700–12. Article  CAS  PubMed  PubMed Central  Google Scholar  * Bou-Dargham MJ, Sha L, Sang Q-XA, Zhang J. Immune landscape of human prostate cancer: immune

evasion mechanisms and biomarkers for personalized immunotherapy. BMC Cancer. 2020;20:572. Article  CAS  PubMed  PubMed Central  Google Scholar  * Ellerman D. Bispecific T-cell engagers:

towards understanding variables influencing the in vitro potency and tumor selectivity and their modulation to enhance their efficacy and safety. Methods. 2019;154:102–17. Article  CAS 

PubMed  Google Scholar  * Baeuerle PA, Kufer P, Bargou R. BiTE: teaching antibodies to engage T-cells for cancer therapy. Curr Opin Mol Therapeut. 2009;11:22–30. CAS  Google Scholar  * Wolf

E, Hofmeister R, Kufer P, Schlereth B, Baeuerle PA. BiTEs: bispecific antibody constructs with unique anti-tumor activity. Drug Discov Today. 2005;10:1237–44. Article  CAS  PubMed  Google

Scholar  * Yuraszeck T, Bartlett D, Singh I, Reed M, Pagano S, Zhu M. A quantitative systems pharmacology (QSP) model to assess the action of blinatumomab in NHL patients (pts). Clin Oncol.

2016;34:e14511. * Belmontes B, Sawant DV, Zhong W, Tan H, Kaul A, Aeffner F, et al. Immunotherapy combinations overcome resistance to bispecific T cell engager treatment in T cell–cold solid

tumors. Sci Transl Med. 2021;13:eabd1524. Article  CAS  PubMed  Google Scholar  * Newman MJ, Benani DJ. A review of blinatumomab, a novel immunotherapy. J Oncol Pharmacy Practice.

2016;22:639–45. Article  CAS  Google Scholar  * Kang TH, Jung ST. Boosting therapeutic potency of antibodies by taming Fc domain functions. Exp Mol Med. 2019;51:1–9. PubMed  Google Scholar 

* Portell CA, Wenzell CM, Advani AS. Clinical and pharmacologic aspects of blinatumomab in the treatment of B-cell acute lymphoblastic leukemia. Clin Pharmacol. 2013;5:5–11. CAS  PubMed 

PubMed Central  Google Scholar  * Schmidt EGW, Hvam ML, Antunes F, Cameron J, Viuff D, Andersen B, et al. Direct demonstration of a neonatal Fc receptor (FcRn)-driven endosomal sorting

pathway for cellular recycling of albumin. J Biol Chemistry. 2017;292:13312–22. Article  CAS  Google Scholar  * Larsen MT, Kuhlmann M, Hvam ML, Howard KA. Albumin-based drug delivery:

harnessing nature to cure disease. Mol Cell Ther. 2016;4:3. Article  PubMed  PubMed Central  Google Scholar  * Pilati D, Howard KA. Albumin-based drug designs for pharmacokinetic modulation.

Expert Opin Drug Metabol Toxicol. 2020;16:783–95. Article  CAS  Google Scholar  * Andersen JT, Dalhus B, Viuff D, Ravn BT, Gunnarsen KS, Plumridge A, et al. Extending serum half-life of

albumin by engineering neonatal Fc receptor (FcRn) binding*. J Biol Chemistry. 2014;289:13492–502. Article  CAS  Google Scholar  * Mandrup OA, Ong SC, Lykkemark S, Dinesen A, Rudnik-Jansen

I, Dagnæs-Hansen NF, et al. Programmable half-life and anti-tumour effects of bispecific T-cell engager-albumin fusions with tuned FcRn affinity. Commun Biol. 2021;4:310. Article  CAS 

PubMed  PubMed Central  Google Scholar  * Silver DA, Pellicer I, Fair WR, Heston WD, Cordon-Cardo C. Prostate-specific membrane antigen expression in normal and malignant human tissues. Clin

Cancer Res. 1997;3:81–5. CAS  PubMed  Google Scholar  * Heck MM, Retz M, Tauber R, Knorr K, Kratochwil C, Eiber M. [PSMA-targeted radioligand therapy in prostate cancer]. Der Urologe Ausg

A. 2017;56:32–9. Article  CAS  PubMed  Google Scholar  * Hofman MS, Emmett L, Sandhu S, Iravani A, Joshua AM, Goh JC, et al. [177Lu]Lu-PSMA-617 versus cabazitaxel in patients with metastatic

castration-resistant prostate cancer (TheraP): a randomised, open-label, phase 2 trial. Lancet. 2021;397:797–804. Article  CAS  PubMed  Google Scholar  * Heitmann JS, Walz JS, Pflügler M,

Kauer J, Schlenk RF, Jung G, et al. Protocol of a prospective, multicentre phase I study to evaluate the safety, tolerability and preliminary efficacy of the bispecific PSMAxCD3 antibody

CC-1 in patients with castration-resistant prostate carcinoma. BMJ Open. 2020;10:e039639. * Buelow B, Dalvi P, Dang K, Patel A, Johal K, Pham D, et al. TNB585.001: a multicenter, phase 1,

open-label, dose-escalation and expansion study of tnb-585, a bispecific T-cell engager targeting PSMA in subjects with metastatic castrate resistant prostate cancer. Clin Oncol.

2021;39:TPS5092-TPS. * Markowski MC, Kilari D, Eisenberger MA, McKay RR, Dreicer R, Trikha M, et al. Phase I study of CCW702, a bispecific small molecule-antibody conjugate targeting PSMA

and CD3 in patients with metastatic castration-resistant prostate cancer (mCRPC). Clin Oncol. 2021;39:TPS5094-TPS. * Tran B, Horvath L, Dorff TB, Greil R, Machiels J-PH, Roncolato F, et al.

Phase I study of AMG 160, a half-life extended bispecific T-cell engager (HLE BiTE) immune therapy targeting prostate-specific membrane antigen (PSMA), in patients with metastatic

castration-resistant prostate cancer (mCRPC). J Clin Oncol. 2020;38:TPS261–TPS. Article  Google Scholar  * Kamat NV, Yu EY, Lee JK. BiTE-ing into prostate cancer with bispecific T-cell

engagers. Clin Cancer Res. 2021;27:2675–7. Article  CAS  PubMed  PubMed Central  Google Scholar  * Deegen P, Thomas O, Nolan-Stevaux O, Li S, Wahl J, Bogner P, et al. The PSMA-targeting

half-life extended BiTE therapy AMG 160 has potent antitumor activity in preclinical models of metastatic castration-resistant prostate cancer. Clin Cancer Res. 2021;27:2928. Article  CAS 

PubMed  Google Scholar  * Tran B, Horvath L, Dorff T, Rettig M, Lolkema MP, Machiels JP, et al. 609 O Results from a phase I study of AMG 160, a half-life extended (HLE), PSMA-targeted,

bispecific T-cell engager (BiTE®) immune therapy for metastatic castration-resistant prostate cancer (mCRPC). Annal Oncol. 2020;31:S507. Article  Google Scholar  * Middelburg J, Kemper K,

Engelberts P, Labrijn AF, Schuurman J, van Hall T. Overcoming challenges for CD3-bispecific antibody therapy in solid tumors. Cancers. 2021;13:287. Article  CAS  PubMed  PubMed Central 

Google Scholar  * Hansen EB, Fredsøe J, Okholm TLH, Ulhøi BP, Klingenberg S, Jensen JB, et al. The transcriptional landscape and biomarker potential of circular RNAs in prostate cancer.

Genome Med. 2022;14:8. Article  CAS  PubMed  PubMed Central  Google Scholar  * Salachan PV, Rasmussen M, Fredsøe J, Ulhøi B, Borre M, Sørensen KD. Microbiota of the prostate tumor

environment investigated by whole-transcriptome profiling. Genome Med. 2022;14:9. Article  CAS  PubMed  PubMed Central  Google Scholar  * Bray NL, Pimentel H, Melsted P, Pachter L.

Near-optimal probabilistic RNA-seq quantification. Nat Biotechnol. 2016;34:525–7. Article  CAS  PubMed  Google Scholar  * Soneson C, Love MI and Robinson MD. Differential analyses for

RNA-seq: transcript-level estimates improve gene-level inferences [version 2; peer review: 2 approved]. F1000Research 2016;4:1521. * Robinson MD, McCarthy DJ, Smyth GK. edgeR: a Bioconductor

package for differential expression analysis of digital gene expression data. Bioinformatics. 2009;26:139–40. Article  PubMed  PubMed Central  Google Scholar  * Ritchie ME, Phipson B, Wu D,

Hu Y, Law CW, Shi W, et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 2015;43:e47. Article  PubMed  PubMed Central  Google

Scholar  * Aran D, Hu Z, Butte AJ. xCell: digitally portraying the tissue cellular heterogeneity landscape. Genome Biol. 2017;18:220. Article  PubMed  PubMed Central  Google Scholar  * Ståhl

PL, Salmén F, Vickovic S, Lundmark A, Navarro JF, Magnusson J, et al. Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science. 2016;353:78–82.

Article  PubMed  Google Scholar  * R Core Team (2018). R: a language and environment for statistical computing. R Foundation for Statistical Computing, Vienna, Austria. 2018;

https://www.R-project.org/. * Hao Y, Hao S, Andersen-Nissen E, Mauck WM 3rd, Zheng S, Butler A, et al. Integrated analysis of multimodal single-cell data. Cell. 2021;184:3573–87. e29 Article

  CAS  PubMed  PubMed Central  Google Scholar  * Yan L. ggvenn: Draw Venn Diagram by ‘ggplot2’. R package version 0.1.9 ed2021. Available from https://CRAN.R-project.org/package=ggvenn. *

Kassambara A, Kosinski M, Biecek P. survminer: Drawing Survival Curves using 'ggplot2'. 2021. R package version 0.4.9. Available from https://CRAN.R-project.org/package=survminer.

* Rey D, Neuhäuser M. Wilcoxon-signed-rank test. In: Lovric M, editor. International Encyclopedia of Statistical Science. Berlin, Heidelberg: Springer Berlin Heidelberg; 2011. p. 1658–9. *

Sprent P. Fisher Exact test. In: Lovric M, editor. International Encyclopedia of Statistical Science. Berlin, Heidelberg: Springer Berlin Heidelberg; 2011. p. 524–5. * Freedman DR, Pisani R,

Purves R. Statistics (international student edition). Chapter 8: Correlation. In: Pisani, R Purves, editors. 4th edn. New York: WW Norton & Company; 2007. * Abeshouse A, Ahn J, Akbani

R, Ally A, Amin S, Andry CD, et al. The molecular taxonomy of primary prostate cancer. Cell. 2015;163:1011–25. * Zhu Y, Qiu P, Ji Y. TCGA-assembler: open-source software for retrieving and

processing TCGA data. Nat Methods. 2014;11:599–600. Article  CAS  PubMed  PubMed Central  Google Scholar  * Schmidt L, Møller M, Haldrup C, Strand SH, Vang S, Hedegaard J, et al. Exploring

the transcriptome of hormone-naive multifocal prostate cancer and matched lymph node metastases. Br J Cancer. 2018;119:1527–37. Article  CAS  PubMed  PubMed Central  Google Scholar  *

Cooperberg MR, Hilton JF, Carroll PR. The CAPRA-S score: a straightforward tool for improved prediction of outcomes after radical prostatectomy. Cancer. 2011;117:5039–46. Article  PubMed 

Google Scholar  * Yang Y, Attwood K, Bshara W, Mohler JL, Guru K, Xu B, et al. High intratumoral CD8( + ) T-cell infiltration is associated with improved survival in prostate cancer patients

undergoing radical prostatectomy. Prostate. 2021;81:20–8. Article  CAS  PubMed  Google Scholar  * McArdle PA, Canna K, McMillan DC, McNicol AM, Campbell R, Underwood MA. The relationship

between T-lymphocyte subset infiltration and survival in patients with prostate cancer. Br J Cancer. 2004;91:541–3. Article  CAS  PubMed  PubMed Central  Google Scholar  * Ness N, Andersen

S, Valkov A, Nordby Y, Donnem T, Al-Saad S, et al. Infiltration of CD8 + lymphocytes is an independent prognostic factor of biochemical failure-free survival in prostate cancer. Prostate.

2014;74:1452–61. Article  CAS  PubMed  Google Scholar  * Bougherara H, Mansuet-Lupo A, Alifano M, Ngô C, Damotte D, Le Frère-Belda M-A, et al. Real-time imaging of resident T cells in human

lung and ovarian carcinomas reveals how different tumor microenvironments control T lymphocyte migration. Front Immunol. 2015;6:500. * Hendry S, Salgado R, Gevaert T, Russell PA, John T,

Thapa B, et al. Assessing tumor-infiltrating lymphocytes in solid tumors: a practical review for pathologists and proposal for a standardized method from the international immuno-oncology

biomarkers working group: part 2: TILs in melanoma, gastrointestinal tract carcinomas, non-small cell lung carcinoma and mesothelioma, endometrial and ovarian carcinomas, squamous cell

carcinoma of the head and neck, genitourinary carcinomas, and primary brain tumors. Adv Anat Pathol. 2017;24:311–35. Article  PubMed  PubMed Central  Google Scholar  * Le DT, Durham JN,

Smith KN, Wang H, Bartlett BR, Aulakh LK, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. 2017;357:409–13. Article  CAS  PubMed  PubMed Central

  Google Scholar  * Smyrk TC, Watson P, Kaul K, Lynch HT. Tumor-infiltrating lymphocytes are a marker for microsatellite instability in colorectal carcinoma. Cancer. 2001;91:2417–22. Article

  CAS  PubMed  Google Scholar  * Froehner M, Koch R, Graefen M. Re: Nicolas Mottet, Roderick C.N. van den Bergh, Erik Briers, et al. EAU-EANM-ESTRO-ESUR-SIOG guidelines on prostate

cancer-2020 update. Part 1: screening, diagnosis, and local treatment with curative intent. Eur Urol 2021;79:243-62: comorbidity measurement in patients with prostate cancer. Eur Urol.

2021;79:e138. Article  PubMed  Google Scholar  * Sommer U, Ebersbach C, Beier AK, Baretton GB, Thomas C, Borkowetz A, et al. Influence of androgen deprivation therapy on the PD-L1 expression

and immune activity in prostate cancer tissue. Front Mol Biosci. 2022;9:878353. Article  CAS  PubMed  PubMed Central  Google Scholar  * Arija JAA, Valderrama BP, Gordoa TA, Diaz EG, Sanchez

JMS, Fernandez-Parra E, et al. 894TiP - PROSTRATEGY: a Spanish Genitourinary Oncology Group (SOGUG) multi-arm multistage (MAMS) phase III trial of immunotherapy strategies in high-volume

metastatic hormone-sensitive prostate cancer. Annal Oncol. 2019;30:v352–v3. Article  Google Scholar  * Ströhlein MA, Lefering R, Bulian DR, Heiss MM. Relative lymphocyte count is a

prognostic parameter in cancer patients with catumaxomab immunotherapy. Medical Hypotheses. 2014;82:295–9. Article  PubMed  Google Scholar  * Lee JS, Ruppin E. Multiomics prediction of

response rates to therapies to inhibit programmed cell death 1 and programmed cell death 1 ligand 1. JAMA Oncol. 2019;5:1614–8. Article  CAS  PubMed  PubMed Central  Google Scholar  * Graff

JN, Alumkal JJ, Drake CG, Thomas GV, Redmond WL, Farhad M, et al. Early evidence of anti-PD-1 activity in enzalutamide-resistant prostate cancer. Oncotarget. 2016;7:52810–7. Article  PubMed

  PubMed Central  Google Scholar  * Queisser A, Hagedorn SA, Braun M, Vogel W, Duensing S, Perner S. Comparison of different prostatic markers in lymph node and distant metastases of

prostate cancer. Modern Pathol. 2015;28:138–45. Article  CAS  Google Scholar  * Wright GL Jr., Haley C, Beckett ML, Schellhammer PF. Expression of prostate-specific membrane antigen in

normal, benign, and malignant prostate tissues. Urologic Oncol. 1995;1:18–28. Article  Google Scholar  * Sartor O, de Bono J, Chi KN, Fizazi K, Herrmann K, Rahbar K, et al.

Lutetium-177–PSMA-617 for metastatic castration-resistant prostate cancer. New Engl J Med. 2021;385:1091–103. Article  CAS  PubMed  Google Scholar  * Vafa O, Gilliland GL, Brezski RJ, Strake

B, Wilkinson T, Lacy ER, et al. An engineered Fc variant of an IgG eliminates all immune effector functions via structural perturbations. Methods. 2014;65:114–26. Article  CAS  PubMed 

Google Scholar  * Schlothauer T, Herter S, Koller CF, Grau-Richards S, Steinhart V, Spick C, et al. Novel human IgG1 and IgG4 Fc-engineered antibodies with completely abolished immune

effector functions. Protein Eng Design Selection. 2016;29:457–66. Article  CAS  Google Scholar  * Bern M, Nilsen J, Ferrarese M, Sand KMK, Gjølberg TT, Lode HE, et al. An engineered human

albumin enhances half-life and transmucosal delivery when fused to protein-based biologics. Sci Transl Med. 2020;12:eabb0580. * Expression Atlas: gene and protein expression across multiple

studies and organisms: PC-3 and DU-145 [cited 2022 1702].

https://www.ebi.ac.uk/gxa/experiments/E-MTAB-2770/Results?specific=true&geneQuery=%255B%257B%2522value%2522%253A%2522FCGRT%2522%252C%2522category%2522%253A%2522symbol%2522%257D%255D&filterFactors=%257B%2522CELL_LINE%2522%253A%255B%2522DU%2520145%2522%252C%2522PC-3%2522%255D%257D&cutoff=%257B%2522value%2522%253A0.5%257D&unit=%2522TPM%2522.

* Viuff D, Antunes F, Evans L, Cameron J, Dyrnesli H, Thue Ravn B, et al. Generation of a double transgenic humanized neonatal Fc receptor (FcRn)/albumin mouse to study the pharmacokinetics

of albumin-linked drugs. J Controlled Release. 2016;223:22–30. Article  CAS  Google Scholar  * Matsumura Y, Maeda H. A new concept for macromolecular therapeutics in cancer chemotherapy:

mechanism of tumoritropic accumulation of proteins and the antitumor agent smancs. Cancer Res. 1986;46:6387–92. CAS  PubMed  Google Scholar  * Desai N, Trieu V, Yao Z, Louie L, Ci S, Yang A,

et al. Increased antitumor activity, intratumor paclitaxel concentrations, and endothelial cell transport of cremophor-free, albumin-bound paclitaxel, ABI-007, compared with cremophor-based

paclitaxel. Clin Cancer Res. 2006;12:1317–24. Article  CAS  PubMed  Google Scholar  Download references ACKNOWLEDGEMENTS The authors gratefully acknowledge Dr. Benedicte Parm Ulhøi and Dr.

Jørgen Bjerggaard Jensen for patient sample collection and evaluation. FUNDING This research was funded by the Novo Nordisk Foundation, Grant; CEMBID (Center for Multifunctional Biomolecular

Drug Design), Grant Number: NNF17OC0028070. This work was also funded in part by the Independent Research Fund Denmark (Grant Numbers: 6110-00450B and 9039-00084B). AUTHOR INFORMATION

AUTHORS AND AFFILIATIONS * Interdisciplinary Nanoscience Center (iNANO), Department of Molecular Biology and Genetics, Aarhus University, 8000, Aarhus, Denmark Eske N. Glud, Yonghui Zhang, 

Ole A. Mandrup & Kenneth A. Howard * Department of Molecular Medicine, Aarhus University Hospital & Department of Clinical Medicine, Aarhus University, 8200, Aarhus, Denmark Martin

Rasmussen, Paul Vinu Salachan & Karina Dalsgaard Sørensen * Department of Urology, Aarhus University Hospital & Department of Clinical Medicine, Aarhus University, 8200, Aarhus,

Denmark Michael Borre Authors * Eske N. Glud View author publications You can also search for this author inPubMed Google Scholar * Martin Rasmussen View author publications You can also

search for this author inPubMed Google Scholar * Yonghui Zhang View author publications You can also search for this author inPubMed Google Scholar * Ole A. Mandrup View author publications

You can also search for this author inPubMed Google Scholar * Paul Vinu Salachan View author publications You can also search for this author inPubMed Google Scholar * Michael Borre View

author publications You can also search for this author inPubMed Google Scholar * Karina Dalsgaard Sørensen View author publications You can also search for this author inPubMed Google

Scholar * Kenneth A. Howard View author publications You can also search for this author inPubMed Google Scholar CONTRIBUTIONS EG drafted and revised the manuscript, carried out statistical

analysis and conceived and planned most experiments. MR conducted the statistical analysis, data analysis and revised the manuscript. OAM designed the AlproTox constructs. YZ carried out AFM

analysis. PVS carried out ST analysis and revised the manuscript. MB was responsible for patient inclusion. KDS aided in the interpretation of the results, developed the theoretical

framework, and revised the manuscript. KH conceived the main conceptual idea, directed the project and involved in the manuscript writing and manuscript revision. CORRESPONDING AUTHOR

Correspondence to Kenneth A. Howard. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no competing interests. ETHICS APPROVAL AND CONSENT TO PARTICIPATE This study conforms to the

principles of the Helsinki Declaration and was approved by The Central Denmark Region Committees on Health Research Ethics [#2000/0299, #1-10-72-361-18, #1-10-72-367-13] and The National

Committee on Health Research Ethics [#1603543/66451] and notified to The Danish Data Protection Agency [#2013–41-2041, #1-16-02-330-13, #1-16-02-23-19, #1-16-02-248-14]. Written consent was

obtained from all participants prior to their donation of tissue samples for a research biobank, while the requirement for patient consent to the specific analyses in this study was waived.

CONSENT TO PUBLISH Not applicable. ADDITIONAL INFORMATION PUBLISHER’S NOTE Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional

affiliations. SUPPLEMENTARY INFORMATION SUPPLEMENTARY FIGURES REPORTING SUMMARY SUPPLEMENTARY METHODS RIGHTS AND PERMISSIONS OPEN ACCESS This article is licensed under a Creative Commons

Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original

author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the

article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use

is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit

http://creativecommons.org/licenses/by/4.0/. Reprints and permissions ABOUT THIS ARTICLE CITE THIS ARTICLE Glud, E.N., Rasmussen, M., Zhang, Y. _et al._ Identification of a high-risk

immunogenic prostate cancer patient subset as candidates for T-cell engager immunotherapy and the introduction of a novel albumin-fused anti-CD3 × anti-PSMA bispecific design. _Br J Cancer_

127, 2186–2197 (2022). https://doi.org/10.1038/s41416-022-01994-1 Download citation * Received: 01 April 2022 * Revised: 14 September 2022 * Accepted: 16 September 2022 * Published: 15

October 2022 * Issue Date: 07 December 2022 * DOI: https://doi.org/10.1038/s41416-022-01994-1 SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content:

Get shareable link Sorry, a shareable link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative