What Were The Key Prostate Cancer Abstracts at ASCO GU 2026?

On the first day of the 2026 American Society of Clinical Oncology (ASCO) Genitourinary Cancers Symposium, key abstracts highlighted potential advances across the prostate cancer continuum. These presentations examined emergent artificial intelligence (AI) models to predict radioligand efficacy in metastatic disease, safety outcomes with olaparib (Lynparza) when added to pembrolizumab (Keytruda)-based regimens in high-risk prostate cancer, and patient-reported outcomes (PROs) and tolerability of capivasertib (Truqap) and abiraterone acetate (Zytiga) in PTEN-deficient metastatic hormone-sensitive prostate cancer (HSPC).

Leveraging Machine Learning Models to Predict Overall Survival in Metastatic Disease

In a poster session at the conference, Lucas Edouard, MD, of Institut Bergonié, presented findings from a study evaluating a machine learning model (MLM) combined with ⁶⁸Ga-PSMA PET radiomics and clinical features to predict outcomes among patients with metastatic castration-resistant prostate cancer (CRPC).¹

Specifically, the retrospective monocentric study evaluated outcomes among patients treated with ¹⁷⁷Lu-PSMA (Pluvicto) between January 2022 and March 2024. Therein, 5 lesions were segmented for each patient on their pre-therapeutic ⁶⁸Ga-PSMA PET/CT, and radiomic features were extracted. MLMs were leveraged, integrating clinical genomic, biochemical, and imaging features to predict overall survival (OS).

Investigators found that the MLM-assisted approach offered a non-invasive, standardized, and accessible approach to predict survival in these patients, with a high-performing model achieving a 12-month AUC of 0.778 by combining 22 clinical, biochemical, genomic, and radiomic features. Furthermore, the study authors discovered that imaging-only modalities conferred comparable performance to clinical-biochemical models, supporting the predictive value of PSMA PET alone; the 12-month AUCs were 0.715 vs 0.766, respectively. Furthermore, intralesional heterogeneity was associated with shorter OS (P = .025), adding further independent prognostic value.

Does Adding Olaparib Impact Safety in Pembrolizumab-Based Regimens for High-Risk Disease?

Zin Myint, MD, associate professor of Internal Medicine specializing in genitourinary malignancies at the Markey Cancer Center of the University of Kentucky, presented findings from a safety analysis of a phase 2 study (NCT05568550) examining the incidence of adverse effects (AEs) among patients with high-risk prostate cancer treated with pembrolizumab and radiotherapy with or without olaparib.² Myint and colleagues sought to assess the safety of olaparib following potential synergistic effects with immune checkpoint inhibition and radiotherapy shown in preclinical data.

There were 17 patients included for enrollment on the study who received pembrolizumab and had NCCN-defined high- or very high-risk localized disease. They were randomly assigned 1:1 to receive pembrolizumab plus standard-of-care radiotherapy and androgen deprivation therapy (ADT) with 200 mg of olaparib twice daily (n = 7; Arm 1) or pembrolizumab/radiotherapy alone (n = 10; Arm 2). Safety was assessed per CTCAE v 5.0 criteria, and the primary end point of the study was PSA nadir less than or equal to 0.06 ng/mL within 6 months of radiotherapy.

Findings from the trial revealed that renal/urinary AEs were slightly more common with the addition of olaparib, at 83% with vs 75% without. However, they were mostly grade 1 or 2 and manageable. Moreover, grade 3 or greater AEs were more common in Arm 1, including lab-related (67%), cardiac, infection, and respiratory-related events (all 33%). Vascular-related events were more common in Arm 2 (63%), particularly hypertension.

Overall, pembrolizumab with radiotherapy and ADT was feasible and tolerable, and olaparib generally did not significantly increase high-grade toxicity. Additionally, all immune-mediated AEs, as well as a treatment-related death, were associated with pembrolizumab.

“[Patients with] localized, high-risk, and very high-risk prostate cancer substantially [increase their] risk of recurrent disease and biochemical recurrence. Subsequently, [they] can develop metastasis and prostate-specific mortality. The treatment intensification strategies are a need for these populations,” Myint said in an interview with CancerNetwork® at the conference. “There are limited data on the safety of the combination with pembrolizumab and radiation or pembrolizumab plus olaparib with the radiation in prostate cancer. This [was] a prespecified safety analysis for the population.”

Could Capivasertib Enhance PROs with Abiraterone in PTEN-Deficient Metastatic HSPC?

In an oral session at the conference, Daniel J. George, MD, Eleanor Easley Distinguished Professor in the School of Medicine at Duke University, discussed findings from the phase 3 CAPItello-281 trial (NCT04493853) evaluating capivasertib vs placebo combined with abiraterone among patients with PTEN-deficient metastatic HSPC.³ Although the primary end point was radiographic progression-free survival (PFS), his analysis primarily sought to examine PROs with each abiraterone-based regimen.

Patients with PTEN-deficiency, defined as 90% or more viable malignant cells with no specific cytoplasmic staining by immunohistochemistry, were included in the assessment. A total of 1012 patients were randomly assigned 1:1 to receive abiraterone at 5 mg daily for a total cumulative dose of 1000 mg plus ADT with capivasertib at 400 mg twice daily for 4 days on and 3 days off (n = 507) or placebo (n = 505) at the same schedule.

Key secondary end points included OS, time to first subsequent therapy, symptomatic skeletal-event free survival, and time to pain progression, with PRO/tolerability end points encompassing Functional Assessment of Cancer Therapy – Prostate (FACT-P) scores, AEs, and Patient Global Impression of Severity (PGI-S).

Findings revealed that the investigational arm showed a significant improvement in radiographic PFS, at a median value of 33.2 months (95% CI, 25.8-44.2) vs 25.7 months (95% CI, 22.0-29.9) with placebo (HR, 0.81; 95% CI, 0.66-0.98; P = .034). Moreover, investigators found that although AEs were numerically higher in the investigational arm (98.8% vs 92.0%), the addition of capivasertib did not impact functional aspects of patient life, including FACT-P physical wellbeing (LS mean estimated difference, –0.4; 95% CI, –0.89 to 0.14), FACT-P functional wellbeing (–0.3; 95% CI, –1.01 to 0.37), and total FACT-P score (0.4; 95% CI, –1.97 to 2.78).

“Despite more AEs in the capivasertib plus abiraterone arm, the addition of capivasertib did not affect other functional aspects of patient life [including work, sleep, enjoyment of life] and overall health-related quality of life during metastatic HSPC, allowing for continued treatment in [approximately] 80% of patients,” George concluded in the oral session. “Capivasertib in combination with abiraterone represents a potential first-in-class targeted treatment for patients with PTEN-deficient metastatic HSPC.”

References

1. Edouard L, Ferrer L, Papillon L, et al. Machine learning models combining 68Ga-PSMA PET radiomics and clinical features to predict overall survival for mCRPC patients treated with 177 Lu-PSMA radioligand therapy. J Clin Oncol. 2026;44(suppl 7):215. doi:10.1200/JCO.2026.44.7_suppl.215
2. Myint Z, Yan D, Strup S, et al. Interim safety analysis of a randomized phase II trial comparing pembrolizumab with radiation versus pembrolizumab, olaparib, and radiation in localized high risk prostate cancer. J Clin Oncol. 2026;44(suppl 7):362. doi:10.1200/JCO.2026.44.7_suppl.362
3. George DJ, Clarke NW, De Santis M, et al. Patient-reported outcomes and tolerability of capivasertib plus abiraterone versus placebo plus abiraterone in patients with PTEN-deficient metastatic hormone-sensitive prostate cancer: CAPItello-281. J Clin Oncol. 2026;44(suppl 7):14. doi:10.1200/JCO.2026.44.7_suppl.14