Development and Validation of an Artificial Intelligence Digital Pathology Biomarker to Predict Benefit of Long-Term Hormonal Therapy and Radiotherapy in Men With High-Risk Prostate Cancer Across Multiple Phase III Trials

Author(s): Andrew J. Armstrong, MD, ScM¹; Vinnie Y.T. Liu, MSc²; Ramprasaath R. Selvaraju, PhD²; Emmalyn Chen, PhD²; Jeffry P. Simko, MD, PhD³; Sandy DeVries, MA³; Oliver Sartor, MD⁴; Howard M. Sandler, MD⁵; Osama Mohamad, MD, PhD³; Huei-Chung Huang, MA²; Jacqueline Griffin, PhD²; Rikiya Yamashita, MD, PhD²; Andre Esteva, PhD²; Phuoc T. Tran, MD, PhD⁶; Daniel E. Spratt, MD⁷; John Hi Carson, MD⁸; Christopher Peters, MD⁹; Elizabeth Gore, MD^10,11; Steve P. Lee, MD, PhD¹²; Jedidiah M. Monson, MD¹³; Mark E. Augspurger, MD¹⁴; Ali El-Gayed, MD¹⁵; Joseph P. Rodgers, MS^16,17; Rana McKay, MD¹⁸; Todd Morgan, MD¹⁹; Felix Y. Feng, MD³; Paul L. Nguyen, MD²⁰;

Source: DOI: 10.1200/JCO.24.00365

Dr. Anjan Patel's Thoughts

A new MMAI-derived digital pathology biomarker was trained and prospectively validated across multiple NRG/RTOG phase III trials, including RTOG 9202 (N=1,192), to predict which high-risk/locally advanced PCa patients benefit from LT-ADT vs ST-ADT with RT. In the overall cohort, LT-ADT reduced DM (17% vs 26% at 15 years) and DDM (15% vs 23% at 15 years), but this benefit was limited to biomarker-positive patients (DM: 19% vs 33%; DDM: 19% vs 30%), with no advantage seen in biomarker-negative patients (DM: 11% vs 11%; DDM: 9% vs 10%). This tool could allow about a third of our "high-risk" patients to avoid two extra years of ADT without compromising metastasis outcomes, while ensuring we intensify for those most likely to benefit. In short, this is a practical step toward personalizing ADT duration and sparing toxicity for a significant subset of our patients.

PURPOSE

Long-term androgen deprivation therapy (ADT) improves survival in men with high-risk localized prostate cancer (PCa) receiving radiotherapy (RT). Predictive biomarkers are needed to guide ADT duration.

METHODS

A multimodal artificial intelligence (MMAI)???derived predictive biomarker was trained for long-term (LT) versus short-term (ST) ADT using pretreatment digital prostate biopsy images and clinical data (age, prostate-specific antigen, Gleason, and T stage) from six NRG Oncology phase III randomized radiotherapy trials. The novel MMAI-derived biomarker was developed to predict the differential benefit of LT-ADT on the primary end point, distant metastasis (DM). MMAI predictive utility was validated on a seventh randomized trial, RTOG 9202 (N = 1,192), which randomly assigned men to RT + ST-ADT (4 months) versus RT + LT-ADT (28 months). Fine-Gray and cumulative incidence analyses for DM, and secondarily, death with DM, were performed. Deaths without DM were treated as competing risks.

RESULTS

In the validation cohort (median follow-up, 17.2 years), LT-ADT significantly improved DM from 26% to 17% (subdistribution hazard ratio [sHR], 0.64 [95% CI, 0.50 to 0.82], P < .001). A significant biomarker-treatment predictive interaction was observed (P = .04) for DM, whereby MMAI biomarker???positive men (n = 785, 66%) had reduced DM with LT-ADT versus ST-ADT (sHR, 0.55 [95% CI, 0.41 to 0.73], P < .001), whereas no treatment benefit was observed for MMAI biomarker???negative men (n = 407; sHR, 1.06 [95% CI, 0.61 to 1.84], P = .84). The estimated 15-year DM risk difference between RT + LT-ADT and RT + ST-ADT was 14% in MMAI biomarker???positive men and 0% in MMAI biomarker???negative men. The MMAI biomarker was also prognostic for DM, irrespective of treatment (sHR, 2.35 [95% CI, 1.72 to 3.19], P < .001).

CONCLUSION

To our knowledge, the MMAI model is the first validated predictive biomarker to guide ADT duration with RT in localized/locally advanced PCa. Approximately one third of men with high-risk PCa could safely be spared the additional 24 months of ADT and the associated morbidity.

Author Affiliations

¹Duke University Medical Center, Durham, NC; ²Artera Inc, Los Altos, CA; ³University of California San Francisco, San Francisco, CA; ⁴Mayo Clinic, Rochester, MN; ⁵Cedars-Sinai Medical Center, Los Angeles, CA; ⁶University of Maryland, Baltimore, MD; ⁷Case Western Reserve University, Cleveland, OH; ⁸Saint John Regional Hospital, Saint John, New Brunswick, CA; ⁹Northeast Radiation Oncology Center, Dunmore, PA accrual for Regional Hospital of Scranton; ¹⁰Zablocki Veterans Administration Medical Center, Milwaukee, WI; ¹¹Medical College of Wisconsin, Milwaukee, WI, Accrual for Clarkson Hospital; ¹²Long Beach VA Healthcare System, Long Beach, CA; ¹³Saint Agnes Medical Center, Fresno, CA; ¹⁴Baptist Medical Center South, Jacksonville, FL; ¹⁵Saskatoon Cancer Centre, Saskatoon, SK; ¹⁶NRG Oncology Statistics and Data Management Center, Philadelphia, PA; ¹⁷American College of Radiology, Philadelphia, PA; ¹⁸University of California San Diego Moores Cancer Center, La Jolla, CA; ¹⁹University of Michigan, Ann Arbor, MI; ²⁰Brigham and Women's Hospital, Boston, MA

Radiographic Progression With and Without Prostate-Specific Antigen Rise in Patients With Advanced Prostate Cancer Treated With Enzalutamide

An important post-hoc analysis from ARCHES and PROSPER that quantifies something we've suspected clinically, roughly 1 in 4 enzalutamide-treated patients who progress radiographically have no prostate-specific antigen (PSA) rise, and the majority won't meet PCWG2/3 PSA progression criteria. Liver metastases were 5-fold more common at radiographic progression on enzalutamide versus control, suggesting lineage plasticity and AR-independent clones are the culprit. Bottom line for practice: don't wait for PSA to rise before imaging your metastatic hormone-sensitive prostate cancer (mHSPC) and nonmetastatic castration-resistant prostate cancer (nmCRPC) patients on ARPIs, periodic cross-sectional imaging, especially in the first two years, is warranted regardless of PSA trend.

Alliance A222001: Oxybutynin Versus Placebo for the Treatment of Hot Flashes in Patients Receiving Androgen-Deprivation Therapy for Prostate Cancer

Solid pragmatic phase II showing oxybutynin 2.5mg and 5mg BID (twice a day) both beat placebo for hot flash reduction in men on androgen-deprivation therapy (ADT), with no serious toxicities beyond expected dry mouth. Given that venlafaxine failed in this population and megestrol/medroxyprogesterone carry real concerns, this gives us an evidence-based option we can feel good about reaching for, especially in patients who also have urinary symptoms. Neurokinin B (NKB) antagonists are coming but not yet studied here, so oxybutynin fills a real gap now.

PARP and Androgen-Signaling Inhibition plus ADT in Metastatic Prostate Cancer

Moving PARP inhibition earlier into the hormone-sensitive androgen pathway modulation–sensitive (APMS) setting, this trial shows talazoparib plus enzalutamide cuts the risk of progression or death in half (HR 0.48) in HRR-altered metastatic prostate cancer, with an impressive 3-year progression-free survival (PFS) of 77% versus 56%. The benefit extends beyond BRCA1/2 to ATM and CDK12 as well, which is clinically meaningful given that two-thirds of this population was non-BRCA.

Improved Survival with Enzalutamide in Biochemically Recurrent Prostate Cancer

Overall survival was 78.9% in the combination group compared with 69.5% in the leuprolide-alone group (HR 0.60). Enzalutamide monotherapy resulted in an overall survival of 73.1%, which was not statistically significant. We were already aware of the progression-free survival benefit; now we have overall survival data. This case was also presented in the MOC questions, where the correct answer was to add enzalutamide for a patient with a rising PSA on an LHRH agonist without imaging evidence of disease.

Metformin Active Surveillance Trial in Low-Risk Prostate Cancer

Metformin did not significantly reduce the risk of disease progression compared with active surveillance in patients with low-risk prostate cancer.