TPS5146 Background: The androgen receptor (AR) is a key driver of prostate cancer progression. AR pathway inhibitors (ARPIs) are standard of care for metastatic prostate cancer, but acquired resistance is inevitable and often involves reactivation of AR signaling via AR gene alterations such as amplifications or ligand binding domain mutations. The need remains for alternative strategies to target the AR. AZD9750 is a novel oral proteolysis-targeting chimera that potently degrades wild-type, mutant, and amplified forms of the AR. Preclinical data also showed significant antitumor benefit with AZD9750 as monotherapy and combined with the PARP1 selective inhibitor saruparib (AZD5305). ANDROMEDA (NCT07336446) is a first-in-human, phase I/II, open-label, multicenter study investigating the safety, tolerability, pharmacokinetics, pharmacodynamics, and preliminary efficacy of AZD9750 as monotherapy and combined with saruparib in patients (pts) with metastatic castration-resistant prostate cancer. Methods: This study follows a modular design in which pts receive AZD9750 as monotherapy (Module 1) or combined with saruparib (Module 2; Table). Each module has 2 parts: monotherapy dose escalation or combination dose finding (Part A) and dose optimization and expansion (Part B). Eligible pts are ≥18 years of age with histologically or cytologically confirmed adenocarcinoma of the prostate and documented metastatic disease with serum testosterone level ≤50 ng/dL, evidence of disease progression, and ECOG performance status of 0 or 1. Pts should have previously received ARPIs and taxane-based chemotherapy, with the exception of no prior taxane-based chemotherapy in Module 1 Part B3. Primary endpoints include occurrence of dose-limiting toxicities (Part A only), incidence of adverse events (AEs) and AEs leading to treatment discontinuation (Parts A and B), and preliminary efficacy (ie, ≥50% decrease in prostate-specific antigen [PSA]; Part B only). Secondary endpoints include preliminary efficacy (ie, ≥50% and ≥90% decrease in PSA, objective response rates and progression-free survival per RECIST v1.1 and PCWG3 criteria, change in target lesion size per RECIST v1.1 criteria, time to PSA response, and time to PSA progression per PCWG3 criteria) and pharmacokinetic parameters (Parts A and B). This study is currently recruiting for monotherapy dose escalation. Clinical trial information: NCT07336446 . Study part Module 1: AZD9750 monotherapy Module 2: AZD9750 plus saruparib Part A A1. Dose escalationA2. Backfill cohorts Combination dose finding a Part B B1. Dose optimization a B2. Dose expansion b B3. Dose expansion b Combination dose expansion c a Recommended doses determined in Module 1 Part A; b To be opened once recommended phase II dose identified in Module 1 Part B1; c To be opened once combination dose identified in Module 2 Part A.

177 Background: Poly(ADP-ribose) polymerase inhibitors (PARPi) + ARPI + ADT have improved clinical outcomes versus ARPI + ADT alone in pts with metastatic castration-resistant prostate cancer (mCRPC), particularly those with BRCA mutations. Interim efficacy results from the Phase 1/2 PETRANHA study (NCT05367440) showed high rates of undetectable prostate specific antigen (uPSA) levels in pts with mHSPC who received saruparib, a PARP1 selective inhibitor, + ARPI + ADT, irrespective of homologous recombination repair mutation (HRRm) status (Azad A, et al. ESMO 2025 [2384MO]). We report updated efficacy and safety results for pts with mHSPC. Methods: Pts received oral saruparib 60 mg once daily + physician’s choice of ARPI (enzalutamide, abiraterone acetate or darolutamide) + ADT. ADT for up to 6 months prior to consent was permitted. Prior chemotherapy for metastatic prostate cancer was not allowed. Treatment continued until disease progression or intolerable toxicity. Results: At data cutoff (June 10, 2025), 93 pts with mHSPC were treated with saruparib + ARPI (enzalutamide [n=3], abiraterone acetate [n=14] or darolutamide [n=76]) + ADT, with a median follow-up of 16.4 months (min–max, 0.0–34.8). Overall, 55.9% (52/93) of pts had high volume disease, 12.9% (12/93) had visceral metastasis, and the baseline median PSA level was 2.5 ng/mL. In response evaluable pts (34/93), the objective response rate (ORR) was 82.4% (28/34; 80% CI, 71.1–90.5), including 5 complete responses (14.7%). The confirmed uPSA rate at any time was 69.9% (65/93; 80% CI, 63.0–76.1) and confirmed 52-week uPSA rate was 76.7% (46/60; 80% CI, 68.2–83.7). For pts with HRRm versus non-HRRm, ORR was 100% (4/4) and 85.7% (12/14); confirmed uPSA rate at any time was 71.4% (10/14) and 70.6% (24/34); and confirmed 52-week uPSA rate was 77.8% (7/9) and 73.9% (17/23), respectively. The combination had a manageable safety profile (Table). Conclusions: In pts with mHSPC, saruparib + ARPI + ADT induced high ORRs and high 52-week uPSA rates. Efficacy was observed regardless of HRRm status. The safety profile of the combination was manageable with no new safety signals. These findings warrant confirmation in the ongoing Phase 3 EvoPAR-Prostate01 trial. Clinical trial information: NCT05367440 . Safety summary (N=93). Median total duration of saruparib / ARPI exposure, months (min–max) 16.3 (0.7–35.6) / 16.5 (0.7–35.6) Safety parameter, n (%) Any AECausally related to saruparib 92 (98.9)83 (89.2) Any Grade ≥3 AECausally related to saruparib 45 (48.4)29 (31.2) Any serious AECausally related to saruparib 24 (25.8)8 (8.6) Saruparib / ARPI discontinuation due to AE* 7 (7.5) / 3 (3.2) Saruparib / ARPI dose reduction due to AE* 24 (25.8) / 3 (3.2) Saruparib / ARPI interruption due to AE* 52 (55.9) / 36 (38.7) *Irrespective of the action taken on other drugs. AE, adverse event.

Prostate cancer follows a long and heterogeneous disease course with incompletely understood aetiology1. Here we dissect the mutational processes shaping the genomes of 959 donors from the Pan Prostate Cancer Group and assess their clinical relevance. By integrating de novo extracted single-base substitution, insertion–deletion and copy-number signatures with six novel complex structural variant signatures, we identify eight integrated mutational footprints (IMFs) that collectively explain the mutational processes in 85% of primary prostate cancer genomes. IMFs were strongly influenced by regional biases in the genome, most prevalently androgen receptor-mediated mutagenesis and replication stress. Four IMFs, present in 37% of primary tumours, were significantly associated with shorter time to metastasis. These included reactive oxygen-species-driven mutagenesis and both canonical and non-canonical homologous recombination deficiency, the latter being enriched in patients of African ancestry. Extending to the metastatic setting, we found that IMFs predicted sensitivity to androgen receptor pathway inhibitors. Taken together, our study delineates the aetiologies and mutational processes that drive the genomic and clinical heterogeneity of prostate cancer, introduces IMFs as a unifying framework, and highlights their potential to improve both risk stratification and biomarker-guided treatment selection. Eight integrated mutational footprints collectively explain the mutational processes in 85% of primary prostate cancer genomes.

Prostate cancer is the most common cancer diagnosed in men and the incidence is rising globally. Disease-related mortality however remains comparatively low. There is now irrefutable evidence that many men do not need treatment if diagnosed with early cancer and can instead be safely managed conservatively. Active surveillance is therefore now an increasingly popular management option for these men. A minority of men on surveillance however will experience disease progression to a point where radical treatment is necessary. It is therefore logical to consider interventions that might slow down or abrogate this natural history. This is particularly important for subgroups of men with early cancer who are at a higher risk of progression and where the risk-benefit of therapeutic intervention is much more favourable. In this narrative review we explore the literature on known molecular and genetic events in prostate cancer which may drive progression. Our principal focus was to consider mechanisms that could be realistically targeted by therapeutics. We further consider key attributes that early cancer therapeutic trials should incorporate in their design. These include risk-stratified patient selection, bespoke dosing schedules and the importance of unambiguous, clinically meaningful endpoints in this new trial space.

Interleukin-23 (IL23) has been reported to drive androgen receptor (AR) and JAK2-STAT3 signaling, promoting treatment resistance and disease progression in advanced prostate cancer (PC). We evaluated the safety, tolerability, and antitumor activity of the anti-IL23 monoclonal antibody tildrakizumab in combination with the AR pathway inhibitor (ARPI) abiraterone acetate (AA) in men with ARPI-resistant metastatic castration-resistant PC (mCRPC). mCRPC patients of ECOG performance status (PS) ≤ 2, who had previously progressed on first-line ARPI therapy, were treated with tildrakizumab (100 mg, 300 mg, 600 mg; 4-weekly) in combination with AA (YonsaTM, 500 mg daily). The primary objective was to determine the recommended phase 2 dose. Secondary endpoints were elucidation of pharmacokinetics (PK), pharmacodynamics (PD), and antitumor activity. No dose limiting toxicities (DLTs) were observed, nor any grade ≥ 3 adverse effects (AEs). The most common treatment-related AEs attributable to tildrakizumab were grade 1—2 fatigue (n = 3/12; 25.0

3095 Background: GRPR is overexpressed in many solid tumors. NeoB binds to GRPR with high affinity and can be radiolabeled for theranostics. NeoRay is the first-in-human study of 177 Lu-NeoB in pts with advanced solid tumors overexpressing GRPR. Phase (Ph) I reported favorable organ dosimetry and safety, and a recommended Ph II dosage (RP2D) of 9.25 GBq. Here, we report the Ph IIa primary analysis. Methods: This Ph IIa, open-label, multicenter, dosage expansion study enrolled 5 adult cohorts with confirmed [ 68 Ga]Ga-NeoB tumor uptake: A) HR+/HER2- breast cancer; B) prostate cancer (PCa); C) gastrointestinal stromal tumor (GIST); D) impaired renal function; and E) pts eligible for Cohorts A–C who also received sacubitril/valsartan (49/51 mg) at Cycle 1 to assess drug-drug interaction (DDI). There was no formal sample size calculation; the target was ~12 pts in Cohorts A–C, ≤6 pts in D, and ~3 pts in E. 177 Lu-NeoB 9.25 GBq was to be administered every 6 wks for ≤9 cycles, except in DDI Cohort E (5.55 GBq in Cycle 1, then 9.25 GBq). Primary endpoints (descriptive statistics) were individual response (centrally assessed by RECIST v1.1) in Cohorts A–C, and 177 Lu-NeoB PK and dosimetry in Cohort E. Secondary endpoints included safety/tolerability and QoL. Results: Overall, 18 pts (A n = 4; B n = 7; C n = 2; D n = 2; E n = 3) received treatment (median age 65 y; 56% male). Median (range) 177 Lu-NeoB exposure was 9 (6–60) wks; median (range) cumulative activity administered was 12.1 (5.7–81.4) GBq. At data cutoff (7 Jan 2025), 4 pts (B n = 2; C n = 2) had completed treatment (≥3 cycles); 1 pt (GIST) received 9 cycles. Of 10 pts with centrally assessed post-baseline response data, 1 had partial response (PCa), 5 had stable disease (of whom 2 had stable disease ≥20 wks [GIST]), and 4 had progressive disease. Adverse events (AEs) occurred in 17 (94%) pts (serious AEs in 3 [17%] pts; Gr ≥3 AEs in 6 [33%] pts). Treatment-related AEs (TRAEs) occurred in 7 (39%) pts (all non-serious; Gr ≥3 in 2 [11%] pts). The most common TRAEs were fatigue (n = 5, 28%), anemia (n = 2, 11% [Gr ≥3 in 1 pt, 6%]), and bone pain (n = 2, 11%). One pt died of hepatic failure related to disease progression (not a TRAE). No AEs led to dosage reductions/interruptions/discontinuations. Dosimetry evaluated in 12 pts showed favorable biodistribution, similar to Ph I. Projected cumulative absorbed doses in target organs were below EBRT limits; lesion absorbed doses were generally consistent across cohorts. Mean ± SD EORTC QLQ-C30 global health scores in Cohorts A–C were 70 ± 22 at baseline (n = 12) and 73 ± 16 at Cycle 2 (n = 4), suggesting stable QoL. Conclusions: NeoRay Ph IIa data at the 177 Lu-NeoB RP2D (9.25 GBq) reinforce Ph I dosimetry and safety results and support further clinical evaluation. Most AEs were mild/moderate and no new safety signals were identified. Despite the limited sample size, preliminary signs of antitumor activity were observed. Clinical trial information: NCT03872778 .

Gastrin-releasing peptide receptor (GRPR) is overexpressed in a range of tumor types, making it an attractive candidate for novel treatment approaches. NeoB binds to GRPR with high affinity and can be radiolabeled with 68Ga ([68Ga]Ga-NeoB) for imaging or 177Lu ([177Lu]Lu-NeoB, hereafter 177Lu-NeoB) for therapy, making it suitable for theranostics. Methods: NeoRay is a prospective, phase 1/2a, open-label, multicenter, first-in-human study of 177Lu-NeoB. Patients with selected advanced solid tumors with GRPR expression (confirmed by [68Ga]Ga-NeoB lesion uptake) were enrolled. Here, we report preliminary data (cutoff, April 29, 2024) from phase 1, which aimed to identify the maximum tolerated dose and/or recommended phase 2 dose of 177Lu-NeoB. Patients were scheduled to receive at least 3 cycles of 177Lu-NeoB at an interval of at least 6 wk. A Bayesian optimal interval design was used, with dose-escalation decisions based on dose-limiting toxicities (DLTs) during cycle 1 of each dose level. The primary endpoint was the incidence and nature of DLTs. Safety was assessed before and throughout each cycle. Dosimetry was assessed after the first administration. Results: Seventeen patients (median age, 65 y; 71% male) with advanced gastrointestinal stromal tumors, prostate cancer, glioblastoma, or breast cancer received 177Lu-NeoB activities of 1.85 GBq (cycle 1) and then 5.55 GBq (cycles 2-4) (n = 3), 9.25 GBq (n = 9), or 11.1 GBq (n = 5). Four DLTs were observed in 3 patients who received 11.1 GBq: grade 3 anemia (n = 2), grade 4 neurologic decline (n = 1), and grade 3 encephalopathy (n = 1). No DLTs were observed at lower administered activities. Overall, 4 of 17 patients (23.5%) had treatment-related adverse events of grade 3 or higher. Among patients with at least 1 evaluable dosimetry measurement (n = 16), the mean absorbed dose coefficient was 0.10 Gy/GBq (SD, 0.056 Gy/GBq) in the kidneys, 0.055 Gy/GBq (SD, 0.039 Gy/GBq) in the pancreas, and 0.018 Gy/GBq (SD, 0.0076 Gy/GBq) in the red marrow. Conclusion: 177Lu-NeoB has a favorable organ dosimetry profile in patients with advanced solid tumors expressing GRPR, with a large safety margin compared with accepted external beam radiation therapy thresholds for organ toxicity. The maximum tolerated dose of 177Lu-NeoB was identified as 9.25 GBq, and the recommended phase 2 dose for the phase 2a dose expansion is 9.25 GBq.

Second line treatments for advanced oesophageal adenocarcinoma (OAC) have limited efficacy, and biomarkers for early relapse are needed. We evaluated the use of a Next Generation Sequencing (NGS) based circulating tumour DNA (ctDNA) assay to predict the efficacy of PD-L1 second line monotherapy in patients with advanced OAC in a prospective, single-centre, pilot study. 19 participants received Durvalumab every four weeks. ctDNA was analysed with Guardant 360® CDx at baseline, weeks 4 and 7. Radiological response at week 26 was assessed by RECIST v1.1. Concordance between ctDNA and CT results was pre-defined as either ≥50% increase in ctDNA mean variant allele frequency (VAF) with progressive disease, or ≥50% decrease with radiological response or stable disease. The primary endpoint was to explore whether early changes in ctDNA levels can predict 6-month response to durvalumab by RECIST criteria and considered positive with ≥70% concordance, Tissue biomarkers were assessed with 30X Whole Genome Sequencing. At week four, 6/17 patients (35.3%) have ctDNA VAF changes ≥50%, of which 4/17 (23.5%) show concordance. At week seven, 8/12 (66.7%) have ctDNA changes ≥50%, with concordance in 6/12 participants (50.0%). In post-hoc analyses, 9/12 (75%) of patients show concordance with any ctDNA change at week seven, and the participants with any increase in ctDNA all have progression (5/5 week four, and 7/7 week seven). Although the study’s primary endpoint was not reached, our data supports assessing ctDNA for immunotherapy response prediction. We report that even a modest increase in ctDNA could indicate progressive disease, assisting decision-making to cease ineffective treatments. Further research for the use of ctDNA in OAC is needed. NCT03653052 (CALIBRATION); drug supplied by AstraZeneca; assay supplied by Guardant. Advanced oesophageal adenocarcinoma is a cancer that started in the oesophagus and has spread to other parts of the body. Patients with advanced oesophageal adenocarcinoma usually have poor outcomes, with cancers continuing to grow following treatment. Immunotherapy, where the patient’s own immune system is boosted to recognise and attack cancer cells, can be effective. However, it is hard to see the effect of treatment on early medical imaging scans as tumour responses can be delayed or misleading. Early tools, such as blood tests, to identify when treatments are not working are needed. We found that measuring DNA released by the cancer (circulating tumour DNA, ctDNA) into the bloodstream 7 weeks following treatment predicted the response to treatment, especially where the cancer had grown and increases in ctDNA were measured. This result suggests early increases in ctDNA could be used to determine when to discontinue ineffective immunotherapy in people with advanced oesophageal cancer, but larger validation studies are needed before clinical practise is changed. Linossi, Azizi et al. report a pilot study testing whether early ctDNA changes predict 6-month response to durvalumab monotherapy in advanced oesophageal adenocarcinoma. Week 7 ctDNA dynamics outperformed week 4, with any ctDNA increase at week 7 identifying all progressors, showing ctDNA is an indicator of immunotherapy discontinuation.

PURPOSE:The purpose was to investigate combined PARP and androgen inhibition in primary prostate cancer and understand the biological mechanisms underlying clinical efficacy, especially in the absence of mutations in homologous recombination (HR) repair pathways. PATIENTS AND METHODS:The primary objective was to measure PARP inhibition, and the secondary objectives were to assess safety and feasibility. Participants received olaparib for 2 weeks before prostatectomy and were randomly assigned or not assigned (1:1) to degarelix. We analyzed diagnostic biopsy and radical prostatectomy samples for PARylated protein expression using IHC. Exploratory analyses included tumor gene sequencing, mutation analysis, and RNA sequencing (RNA-seq) using both bulk and single-cell RNA-seq performed on pretreatment and posttreatment tissues. RESULTS:PARylated protein expression was significantly reduced in both cohorts, with no drug-related delays in radical prostatectomy. The gene set enrichment analysis identified distinct treatment response signatures related to olaparib in both cohorts and showed downregulation of androgen response genes after olaparib + degarelix treatment.Transcript profiling revealed an upregulation of the p53 hallmark, which was more pronounced with the combination treatment. Canonical cell-cycle progression hallmarks, including E2F targets and the G2-M checkpoint, were suppressed across all cases, correlating with a HR-deficient transcriptional signature. Single-nuclear RNA-seq indicated a greater increase in inflammatory response pathway activity within tumor epithelia after combination treatment. CONCLUSIONS:Transcriptomic analysis identified common hallmark alterations reflecting the combined impact of PARP inhibitor and androgen blockade on cell-cycle progression. We observed a shared phenotypic response to combination therapy across prostate cancers without known HR repair gene alterations. This suggests alternative mechanisms rather than antiandrogen-induced HR deficiency.

Abstract Background: The investigator-initiated, prospective, single-arm CALIBRATION trial (NCT03653052) was designed to evaluate early (weeks 4 and 7) liquid biopsies for concordance with meaningful treatment response (week 26) in patients with advanced esophageal cancer. Participants were given durvalumab (a PD-L1 inhibitor) following at least one prior systemic therapy. Methods: Participants received 1500 mg of durvalumab every four weeks. Circulating tumor DNA (ctDNA) was analyzed with Guardant 360® CDx at baseline, week 4, and week 7. Tumor biopsies were taken for 30X whole-genome sequencing. CT scans at week 26 were assessed using RECIST v1.1 criteria. The trial's primary endpoint was concordance between ctDNA changes and response. Concordance was pre-defined and ctDNA was compared to baseline: a ≥50% increase in mean variant allele frequency (VAF) of ctDNA was concordant with disease progression or a ≥50% decrease in VAF was concordant with radiologic response or stable disease. Results: By week 26, 13 of 18 (72.2%) evaluable patients experienced disease progression (PD), 2 of 18 (11.1%) had stable disease (SD), and 3 of 18 (16.7%) achieved a partial response (PR). 2 participants (11.1%) with PR have remained on treatment for over two years post-enrollment. Due to clinical deterioration or non-compliance, not all planned ctDNA samples were collected. At week 4, 4 of 17 patients (23.5%, 95% confidence interval [CI] 6.8–49.9) showed concordance; at week 7, this rose to 6 of 12 patients (50.0%, 95% CI 21.1–78.9). For subjects with a ≥50% increase in ctDNA at week 7, median survival was 9.7 months (n=4/12, 95% CI 7.3–NA), compared to 13.6 months (n=4/12, 95% CI 11.5–NA) for those with a ≥50% decrease. Patients with PR at week 26 showed mean ctDNA VAF decreases of -41% at week 4 and -85% at week 7, while patients with PD exhibited mean increases of +38% and +172%, respectively. Importantly, all patients with any increase in ctDNA VAF at either week 4 (5 patients, 29%) or week 7 (6 patients, 50%) experienced progression. Tumor biopsies, available for 15 patients, revealed numerous mutations and copy number aberrations, with MYC and AR gains common across samples. No specific mutation, aneuploidy state, nor tumor mutational burden was significantly associated with response, though the trial was not powered for this analysis. Safety was consistent with published information. In a post hoc analysis, we progressively reduced the threshold for ctDNA changes from ±50% to ±1%. The three clinical outcomes (PD, SD, and PR) were more clearly resolved using week 7 samples. At a ±1% threshold at week 7, 75% of cases (9 of 12, 95% CI 42.8–94.5) showed concordance. Conclusions: Our results align with the growing body of literature supporting early ctDNA dynamics as a predictive marker for immunotherapy response. Specifically, increases in ctDNA VAF could help guide the discontinuation of ineffective immunotherapies. These findings highlight week 4 and, particularly, week 7 as key timepoints to assess ctDNA dynamics and inform therapeutic decisions. Citation Format: Alexander Azizi, Constanza Linossi, Andrea Machin, Alimu Dayimu, Nikos Demiris, Ahmad Miremadi, Maria O'Donovan, Edmund Godfrey, Paulius D Mennea, Ze Zhou, Angela An, Elizabeth Smyth, Iris Faull, Simon J Dovedi, Rebecca C Fitzgerald, Amit Roshan, Simon Pacey. Utility of ctDNA assessment after six weeks of immunotherapy to predict radiological response in advanced esophageal cancer [abstract]. In: Proceedings of the AACR IO Conference: Discovery and Innovation in Cancer Immunology: Revolutionizing Treatment through Immunotherapy; 2025 Feb 23-26; Los Angeles, CA. Philadelphia (PA): AACR; Cancer Immunol Res 2025;13(2 Suppl):Abstract nr B064.

Proof-of-principle studies have identified circulating tumor DNA (ctDNA) in small volumes of whole blood, even dried blood spots (DBS) collected by finger-prick sampling from patients with advanced cancer. This minimally invasive sampling method offers ease, frequency and low cost of collection, simplified shipping and long-term room temperature stability. Adopting DBS for ctDNA collection and whole genome sequencing (WGS) analysis may support novel clinical trial design and increase real-world utility of liquid biopsies, especially where traditional phlebotomy is a barrier. We collected sample quartets comprising serial DBS, and matching plasma, tumor, and buffy coat from ∼125 patients with multiple cancer types including oesophageal, lung, head & neck, melanoma, ovarian, colorectal and pancreatic cancer. Collections were in diverse settings at hospitals in the UK and Ethiopia, and by patients at home in the UK. DNA isolated from DBS was size-selected using bead-based enrichment to remove genomic DNA. Libraries were generated from DBS size-selected DNA, and DNA extracted from plasma, tumor and buffy coat samples had paired-end WGS at depth of ∼1X, 10X, and 50X respectively. Using bioinformatic pipelines for tumor-informed and tumor-agnostic detection, WGS data was analyzed for somatic copy number aberrations (SCNAs), fragment length, and tumor-specific single nucleotide variants (SNVs). We detected SNVs and SCNAs in DBS reflective of genomic changes in matching tumor DNA and plasma, from patients with Stage II-IV cancers. We observed concordant changes to the levels of those SNVs and SCNAs in plasma and DBS samples collected before and after systemic treatment compared with clinical and radiological measures. With tumor-informed analysis, utilizing patient-specific lists of 5, 930-51, 927 SNVs identified from tumor DNA, limit of detection for ctDNA was 34 ppm in plasma, and 95 ppm in DBS, comparable to commercially available assays for ctDNA. DBS collected in hospital by healthcare professionals, remotely in Ethiopia and at-home by patients in the UK with transportation at room temperature showed ctDNA can be extracted for analysis from diverse settings. Our data highlights the promising role of DBS in advancing non-invasive cancer diagnostics and monitoring. The versatility of DBS in capturing ctDNA from various cancer types, coupled with its accessibility for collecting samples, is a significant step forward in blood-based cancer detection. This offers potential to enhance cancer monitoring and early detection strategies in both tumor-informed and tumor-agnostic settings. DBS is a cost-effective, user-friendly sampling strategy allowing at-home collection, enabling large-scale research and development programs needed to bring biomarker methods into widespread use. Amit Roshan, Fazlur R. Talukdar, Angela An, Emma-Jane Ditter, Ze Zhou, Paulius D. Mennea, Linfang Wu, Maria C. Neofytou, Anita Balakrishnan, Claire M. Connell, Gabriel Funingana, Giovanni Codacci-Pisanelli, Girma Mulisa, Tamrat Abebe, Constanza Linossi, Hui Zhao, Wendy N. Cooper, Huiqi Yang, Rajesh Jena, James D. Brenton, Pippa Corrie, Rebecca C. Fitzgerald, Simon Pacey, Richard Baird, Nitzan Rosenfeld. Dried blood sot sampling as a minimally invasive and accessible method for investigating circulating tumor DNA in diverse cancer settings [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 4550.

Abstract Aim: Recent phase 3 trials (PROPEL, TALAPRO2) have demonstrated survival benefits when combining PARP inhibitors (PARPi) with androgen pathway inhibitors (API) in prostate cancer (PC) treatment. Importantly, this benefit is observed regardless of homologous repair deficiency (HRD) status, suggesting alternative underlying mechanisms beyond synthetic lethality. However, the complete identification of all the PC patients who can benefit from this combination therapy remains an unmet need. Therefore, further investigation into the distinct molecular pathways involved and identification of biomarkers of response is critical and is the aim of this study. Methods: We have performed an exploratory transcriptomic analysis including RNA-seq, differential expression analysis and gene set enrichment analyses (GSEA) of pre and post treatment biopsies of localized PC patients (pts) with higher risk of relapse and that were included in a window of opportunity (WoO) study. We have included pts recruited to the CANCAP-3 study that were exposed to either two weeks of olaparib alone or in combination with degarelix (1:1 randomization). A corresponding transcriptomic analysis of pre and post enzalutamide biopsies of patients recruited to the WoO study DARANA was also included. Results: GSEA results have shown p53 hallmark upregulation more pronounced in the combination cohort. In the combination cohort the androgen response genes were downregulated. Additionally, canonical cell cycle progression hallmarks (E2F and G2M checkpoint) were suppressed with a significant downregulation of the Prolaris signature. In the olaparib monotherapy cohort pts with the greatest PSA decline demonstrated large reductions in the Prolaris score. Furthermore, in the olaparib monotherapy cohort the two patients that relapsed within 4 years of follow up demonstrated greater Prolaris scores. Interestingly, a significant positive correlation between HR deficiency signatures and reduced prostate cancer proliferation index was observed across the study cohorts, although there was only minimal overlap in the genes constituting these two signatures. This correlation was maintained excluding HRD pts (n=2). Finally, this relationship was also observed following analysis of the DARANA trial. Conclusions: Tissue transcriptomic data analysis of PC pts can provide further insights on the possible biological mechanisms that underlie the effect of this drug combination. In this study the expression of HRD signature appeared to be in direct proportion to the reduction in cellular proliferation (Prolaris). This was seen in the combination treatment cohort, as well as in a separate API monotherapy study (DARANA). This suggests that androgen inhibition is the main driver for the reduction in HR signature. However, this phenomenon may represent a consequence of cell cycle arrest occurring more readily in cells with the greatest HR defect. Further studies are needed to further explain the unique upstream processes that contribute to an AR-supported DNA damage response, in the context of PARP inhibition. Citation Format: Ana Filipa Palma dos Reis, Toby Milne-Clark, Amit Dipak Amin, Liliya Nazlamova, Simon Pacey, Harveer Dev. Pre and post treatment transcriptomic analysis provide new insights on the mechanisms underlying the efficacy of PARPi and Androgen blockage in prostate cancer [abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: DNA Damage Repair: From Basic Science to Future Clinical Application; 2024 Jan 9-11; Washington, DC. Philadelphia (PA): AACR; Cancer Res 2024;84(1 Suppl):Abstract nr A021.

TPS5109 Background: Progression to metastatic castration-resistant prostate cancer (mCRPC) occurs in most patients treated with androgen receptor signaling inhibitors (ARSi) for advanced disease. Preclinical studies demonstrate that AR and PI3K - AKT - mTOR (PAM) pathways interact through reciprocal negative feedback, whereby inhibition of one pathway activates the other. Thus, combining a PAM inhibitor with an ARSi may deliver improved anti-cancer activity in patients with mCRPC. A Phase 2 trial in 129 patients with mCRPC who progressed on abiraterone demonstrated improved median radiographic progression-free survival (rPFS) when samotolisib, a dual PI3K-mTOR inhibitor, was added to enzalutamide. These results form the basis for this clinical trial of gedatolisib, a potent PAM inhibitor, in combination with darolutamide in men with mCRPC who have previously progressed on ARSi. Methods: This open-label, multicenter, Phase 1/2 study will evaluate the safety and efficacy of gedatolisib in combination with darolutamide in men with mCRPC who have progressed on ARSi. In Phase 1, 36 patients will be randomized to one of two dose arms to evaluate dose limiting toxicities (DLTs) and determine the recommended Phase 2 dose (RP2D). Gedatolisib will be administered once weekly for 3-weeks-on/1-week-off: Arm 1 – 120 mg and Arm 2 – 180 mg, with darolutamide 600 mg orally administered twice daily. Arm 2 may be dose de-escalated depending on the number of DLTs observed. In Phase 2, 12 additional patients will be enrolled at the RP2D (n= 30). Key inclusion criteria include adult males (≥ 18 years) with mCRPC who have progressed on or after treatment with one next-generation ARSi. Key exclusion criteria include males with adenocarcinoma with a small cell component and with ≥10% neuroendocrine type cells; prior treatment with PI3K, AKT, or mTOR inhibitor; prior chemotherapy or radiopharmaceutical therapy for mCRPC; uncontrolled type 1/2 diabetes; or active brain or leptomeningeal metastases. Primary endpoints for Phase 1 are safety and tolerability (incidence of DLTs, adverse events, and determination of maximum tolerated dose) determination of the recommended Phase 2 dose (RP2D), PK, and Bayesian Optimal Interval utility score. Primary endpoints for Phase 2 are rPFS rate at 6 months based on Response Evaluation Criteria in Solid Tumors (RECIST) v1.1 with modifications as specified in Prostate Cancer Working Group 3 criteria. Secondary endpoints include rPFS rates at 9 and 12 months, overall rPFS, prostate-specific antigen response of ≥50% decrease from baseline at 4, 8, 12, and 16 weeks, overall response rate, duration of response, clinical benefit rate, overall survival rate at 18 and 24 months, and safety. The trial is currently open for enrollment (NCT06190899). Clinical trial information: NCT06190899 .

TPS356 Background: Novel hormonal agents (NHAs), such as darolutamide, are the standard of care for treatment of metastatic prostate cancer (mPC). There is emerging evidence that combining PARP inhibitors with NHAs can further improve benefit in patients with mPC. AZD5305 is a potent and selective PARP inhibitor that specifically targets and inhibits PARP1 while sparing other PARP family enzymes and has the potential for demonstrating a better therapeutic index compared to first-generation PARP inhibitors that block both PARP1 and PARP2. The Phase 3 PROpel study showed that first-line combination treatment with olaparib and abiraterone significantly improved radiographic progression-free survival (rPFS) over abiraterone alone in pts with metastatic castration-resistant PC (mCRPC) (hazard ratio [HR], 0.66; 95% CI, 0.54 to 0.81; p<0.001). Similarly, the Phase 3 TALAPRO-2 study showed that first-line talazoparib with enzalutamide resulted in statistically significant improvement in rPFS over enzalutamide alone in pts with mCRPC (HR, 0.63; 95% CI, 0.51 to 0.78; p<0.0001). Both PARP1 and androgen receptor are involved in DNA repair, which may explain the additional benefit of combining PARP inhibitors with NHAs; however, direct demonstration of the mechanism behind the benefit of combination treatment have not been established in humans. ASCERTAIN is a window-of-opportunity study that will provide insights into the mechanism of action of combination treatment with PARP inhibitors and NHAs, further supporting their use in clinical practice. Methods: ASCERTAIN (NCT05938270) is a Phase 1 multi-center study enrolling pts aged ≥18 years with localized, unfavorable-intermediate or high-risk PC who are eligible for radical prostatectomy. Pts with prior treatment with any systemic or local anti-cancer treatment for localized PC are excluded. Eligible pts will either be randomly allocated to receive AZD5305 alone or in combination with darolutamide 600 mg twice daily (BD) or darolutamide alone (600 mg BD), for a period of 21 days; additional pts will be recruited into a no-treatment arm. Pts will undergo surgery on Day 22. The primary objective is to assess the fold change from baseline in the percentage of cells with phosphorylated-Ser139 histone H2AX, a marker of DNA damage, in tumor biopsy samples at diagnosis and in the surgical specimen. Key secondary objectives include safety and tolerability; surgery outcomes; and the change in percentage of Ki-67-positive cells in tumor samples. Exploratory analysis using comprehensive omics approaches are planned to elucidate further insights into the mechanism of action of combination therapy. Enrollment began in September 2023; sites across North America, Europe and Australia will enroll up to 120 patients. Clinical trial information: NCT05938270 .

123 Background: The combination of PARP inhibitors (PARPi) with novel hormonal agents (NHAs) has recently demonstrated significant improvement in progression-free survival in patients (pts) with metastatic castration-resistant prostate cancer (mCRPC) compared with NHAs alone. AZD5305, a PARP1-selective inhibitor, has the potential for an improved safety profile and limited drug-drug interactions (DDIs) compared with first-generation PARPi. PETRANHA, an open-label nonrandomized study, is evaluating the safety and DDIs of AZD5305 with physician’s choice of NHA (enzalutamide [enza], abiraterone acetate [abi] or darolutamide [daro]) in pts with metastatic prostate cancer. Methods: Pts were aged ≥18 years with histologically confirmed mCRPC or metastatic castration-sensitive prostate cancer (mCSPC), suitable for NHA treatment, with or without HRR mutations in tumor tissue. Key exclusion criteria were previous PARPi, platinum chemotherapy, or targeted radioligand therapy for pts with mCRPC or mCSPC; and previous NHA or docetaxel in pts with mCSPC. Pts were assigned to receive AZD5305 60 mg once daily (OD; first dose level tested) and either enza 160 mg OD (Arm 1), abi 1000 mg + 5 mg prednisone OD (Arm 2), or daro 600mg twice daily (Arm 3) until disease progression or any intolerable adverse event (AE) occurred. Primary objectives were safety and tolerability. DDIs were evaluated for each combination. Results: At data cutoff (July 10, 2023), 48 pts were included in the interim safety analysis (Arm 1, n=11; Arm 2, n=19; Arm 3, n=18). In total, 32 (66.7%) pts had mCRPC, 16 (33.3%) pts had mCSPC, and 8 (16.7%) pts were pre-exposed to NHAs. Median duration of AZD5305 exposure was 6.3 months (range, 1.05–12.58) across all arms. Safety data are presented in the Table. The most common AEs were anemia (52.1%; Gr ≥3, 16.7%), fatigue (50.0%; Gr ≥3, 2.1%), and neutropenia (33.3%; Gr ≥3, 6.3%). No dose-limiting toxicities or AE-related deaths occurred in any arm; AEs leading to discontinuations were uncommon. No clinically significant DDIs were measured with any of the combinations. Conclusions: Initial safety, tolerability, and DDI data from the PETRANHA study indicates that AZD5305 can be safely combined with three individual NHAs with low rates of dose interruptions or reductions. The study is ongoing in Australia, Italy, UK, and USA. Clinical trial information: NCT05367440 . [Table: see text]

Abstract Background NXP800 is an oral, small molecule that activates the integrated stress response through activation of the kinase GCN2. The drug causes growth inhibition and regression in various preclinical models, including ARID1A mutated ovarian and endometrial xenograft and cholangiocarcinoma PDX models. Methods We describe results of the dose-defining dose escalation cohort and the toxicity-pharmacokinetics and pharmacodynamic evaluation in a first-in-human phase I trial of NXP800. Dose escalation was guided by a Bayesian modified continual reassessment method that targeted a dose-limiting toxicity DLT probability (DLTp) closest to 30% but <33%. Toxicity was assessed using NCI.CTC.V5. Detailed sampling of blood for pharmacokinetic analysis and blood for mRNA and protein analysis in peripheral blood mononuclear cells was conducted on C1D-3 to -7 (run in prior to continuous dosing) and C2D1. Two schedules (once per day [QD] and twice per day [BID] and a dose range of 50 to 150 mg/day were evaluated. Results Eighteen patients were treated: at baseline, the median age was 65 years (range 42-77), 9 (50%) were female and 14/18 (78%) had an ECOG score of 1. The most common treatment emergent adverse events (more than 20% of patients) were nausea, vomiting, diarrhea, fatigue, decreased appetite, AST increase and thrombocytopenia. All these side effects were grade 1-2 except grade 3 nausea and grade 3 diarrhea reported in one patient each and grade 3 thrombocytopenia in 2 patients. The MTD for the QD schedule was 100 mg (DLTp=26%), no doses in the BID schedule had estimated DLTp of <33%. The average Cmax at 50 mg and 75 mg OD was 190 nM and 384 nM during the run-in and 495 nM and 656 nM in cycle 2 reached by 3 hours. ATF4 protein in PBMCs was elevated in 57% and 80% of patients during run in and Cycle 2 respectively, 6 hours after 50 mg and 75 mg treatment. The mRNA biomarkers in blood most consistently elevated included ATF4 transcriptional targets NUPR1, ULBP1 and TRIB3 at 2 - 24 hours after 50 mg and 75 mg treatment. Further, mRNA levels for these biomarkers were elevated in post-treatment biopsy samples in one of the two patients tested. It is envisaged that NXP800 will need to be dosed continuously and although the protocol defined MTD at 28 days at 100 mg QD, it was decided to take a dose of range of 50 mg and 75 mg OD which caused biomarker modulation, forward for further evaluation. Conclusions A tolerable dose/schedule of continuous dosing of NXP800 has been established with biomarker modulation in blood. Patients will be randomized to two dose levels 50 and 75 mg/day in a QD schedule to evaluate response in a population of ARID1A mutated, platinum-resistant, clear cell ovarian cancer (NCT05226507). Citation Format: Simon Pacey, Ching Leung, Simon Rodney, Ana Filipa Palma DosReis, Ruth Ruddle, Matthew Tall, Karen Swales, Robert te Poele, Paul Workman, Diane Marsolini, Megan Sardone, Enrique Poradosu, Shay Shemesh, Udai Banerji. Pharmacokinetic and pharmacodynamic evaluation of NXP800, a novel GCN2 activator, in a first in human clinical trial [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr CT111.

Abstract Background: Progression to metastatic castration-resistant prostate cancer (mCRPC) occurs in most patients treated with androgen receptor signaling inhibitors (ARSi) for advanced disease. Preclinical studies demonstrate that AR and PI3K - AKT - mTOR (PAM) pathways interact through reciprocal negative feedback, whereby inhibition of one pathway activates the other. Thus, combining a PAM inhibitor with an ARSi may deliver improved anti-cancer activity in patients with mCRPC. A Phase 2 trial in 129 patients with mCRPC who progressed on abiraterone demonstrated improved median radiographic progression-free survival (rPFS) when samotolisib, a dual PI3K-mTOR inhibitor, was added to enzalutamide. These results form the basis for this clinical trial of gedatolisib, a potent PAM inhibitor, in combination with darolutamide in men with mCRPC who have previously progressed on ARSi (NCT06190899). Methods: This open-label, multicenter, Phase 1/2 study will evaluate the safety and efficacy of gedatolisib in combination with darolutamide in men with mCRPC who have progressed on ARSi. In Phase 1, 36 patients will be randomized to one of two dose arms to evaluate dose limiting toxicities (DLTs) and determine the recommended Phase 2 dose (RP2D). Gedatolisib will be administered once weekly for 3-weeks-on/1-week-off: Arm 1 - 120 mg and Arm 2 - 180 mg, with darolutamide 600 mg orally administered twice daily. Arm 2 may be dose de-escalated depending on the number of DLTs observed. In Phase 2, 12 additional patients will be enrolled at the RP2D (n= 30). Key inclusion criteria include adult males (≥ 18 years) with mCRPC who have progressed on or after treatment with one next-generation ARSi. Key exclusion criteria include males with adenocarcinoma with a small cell component and with ≥10% neuroendocrine type cells; prior treatment with PI3K, AKT, or mTOR inhibitor; prior chemotherapy or radiopharmaceutical therapy for mCRPC; uncontrolled type 1/2 diabetes; or active brain or leptomeningeal metastases. Primary endpoints for Phase 1 are safety and tolerability (incidence of DLTs, adverse events, and determination of maximum tolerated dose) determination of the recommended Phase 2 dose (RP2D), PK, and Bayesian Optimal Interval utility score. Primary endpoints for Phase 2 are rPFS rate at 6 months based on Response Evaluation Criteria in Solid Tumors (RECIST) v1.1 with modifications as specified in Prostate Cancer Working Group 3 criteria. Secondary endpoints include rPFS rates at 9 and 12 months, overall rPFS, prostate-specific antigen response of ≥50% decrease from baseline at 4, 8, 12, and 16 weeks, overall response rate, duration of response, clinical benefit rate, overall survival rate at 18 and 24 months, and safety. The trial is currently open for enrollment. Citation Format: Adam Sharp, Alice Bernard-Tessier, Delphine Borchiellini, Johann de Bono, Karim Fizazi, Gwenaelle Gravis, David Lorente, Hakim Mahammedi, Begona Mellado, David Olmos, Simon Pacey, Jose Piulats, Sara P. Ramirez, Igor Gorbatchevsky, Elisabeth Heath. A phase 1/2, open-label, randomized, dose-finding and dose expansion study of gedatolisib in combination with darolutamide in metastatic castration-resistant prostate cancer (mCRPC) [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2024; Part 2 (Late-Breaking, Clinical Trial, and Invited Abstracts); 2024 Apr 5-10; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2024;84(7_Suppl):Abstract nr CT087.