Andre Esteva’s research while affiliated with University of California, Santa Barbara and other places

PURPOSE Artificial intelligence (AI) tools could improve clinical decision making or exacerbate inequities because of bias. African American (AA) men reportedly have a worse prognosis for prostate cancer (PCa) and are underrepresented in the development genomic biomarkers. We assess the generalizability of tools developed using a multimodal AI (MMAI) deep learning system using digital histopathology and clinical data from NRG/Radiation Therapy Oncology Group PCa trials across racial subgroups. METHODS In total, 5,708 patients from five randomized phase III trials were included. Two MMAI algorithms were evaluated: (1) the distant metastasis (DM) MMAI model optimized to predict risk of DM, and (2) the PCa-specific mortality (PCSM) MMAI model optimized to focus on prediction death in the presence of DM (DDM). The prognostic performance of the MMAI algorithms was evaluated in AA and non-AA subgroups using time to DM (primary end point) and time to DDM (secondary end point). Exploratory end points included time to biochemical failure and overall survival with Fine-Gray or Cox proportional hazards models. Cumulative incidence estimates were computed for time-to-event end points and compared using Gray's test. RESULTS There were 948 (16.6%) AA patients, 4,731 non-AA patients (82.9%), and 29 (0.5%) patients with unknown or missing race status. The DM-MMAI algorithm showed a strong prognostic signal for DM in the AA (subdistribution hazard ratio [sHR], 1.2 [95% CI, 1.0 to 1.3]; P = .007) and non-AA subgroups (sHR, 1.4 [95% CI, 1.3 to 1.5]; P < .001). Similarly, the PCSM-MMAI score showed a strong prognostic signal for DDM in both AA (sHR, 1.3 [95% CI, 1.1 to 1.5]; P = .001) and non-AA subgroups (sHR, 1.5 [95% CI, 1.4 to 1.6]; P < .001), with similar distributions of risk. CONCLUSION Using cooperative group data sets with a racially diverse population, the MMAI algorithm performed well across racial subgroups without evidence of algorithmic bias.

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.

Purpose A multimodal artificial intelligence (MMAI) biomarker was developed using clinical trial data from North American men with localized prostate cancer treated with definitive radiation, using biopsy digital pathology images and key clinical information (age, PSA, and T-stage) to generate prognostic scores. This study externally validates the biomarker in a prospective, real-world dataset of men who underwent radical prostatectomy (RP) for localized prostate cancer at a tertiary referral center in Sweden. Experimental Design Association between the MMAI scores (continuous and categorical) and endpoints of interest was assessed with Fine–Gray and cumulative incidence analyses for biochemical recurrence (BCR) and logistic regression for adverse pathology (AP) at RP. Results The analysis included 143 patients with evaluable biopsy pathology images and complete clinical data to generate MMAI scores. The median follow-up was 8.8 years. At diagnosis, the median PSA was 7.5 ng/mL, the median age was 64 years, 29% had a Gleason grade group ≥3, and 88 men were evaluable for AP at RP. MMAI was significantly associated with BCR [subdistribution HR, 2.45; 95% confidence interval (CI), 1.77–3.38; P < 0.001] and AP at RP (OR, 4.85; 95% CI, 2.54–10.78; P < 0.001). Estimated 5-year BCR rates for MMAI intermediate to high versus low were 25% (95% CI, 16%–36%) versus 4% (95% CI, 1%–11%), respectively. Conclusions The MMAI biomarker, previously shown to be prognostic for distant metastasis and prostate cancer–specific mortality in men receiving definitive radiation, was prognostic for post-RP endpoints: BCR and AP. This biomarker validation study further supports the use of MMAI biomarkers in men with prostate cancer outside North America and those treated with RP.

400 Background: The ArteraAI Prostate Test (v1.2), a digital pathology-based multimodal artificial intelligence (MMAI) biomarker, was developed and validated using clinical data (age, T stage, PSA) and whole slide images (WSI) from prostate biopsies to prognosticate risk of 10-year distant metastasis for men with localized prostate cancer (PCa). This study aimed to apply the MMAI biomarker to prostatectomy (RP) tissue microarray (TMA) samples and, for the first time, explore the impact of using RP TMA in place of RP WSI on MMAI scores. Methods: The analysis cohort included men with localized PCa who had undergone RP. Black men were matched in a 4:1 ratio to White men with similar baseline characteristics. MMAI scores were generated using digitized TMA and WSI from each patient’s RP specimen. The normality of score distribution was examined using the Shapiro-Wilk (SW) test. Wilcoxon signed rank (WSR) test and Spearman rank coefficients were used to compare TMA-derived and WSI-derived scores. Analyses were performed in the entire cohort and Black or White subgroups. Results: Paired MMAI scores were generated for 98 men with a median age of 58, PSA 5.4 ng/mL, the majority were Gleason grade groups 1-2 (84%), 82% were Black. The distribution of TMA-derived scores (SW test, P=0.04) was more skewed than that of WSI-derived scores (SW test, P=0.39). The median MMAI score was significantly higher for WSI images than for TMA images in all men and in Black men (WSR test, P<0.01). There was no notable variation in either set of scores by race (Table). The correlation coefficient was 0.496 (0.519 for Black men). Using pre-specified cutoffs in MMAI scores (high, intermediate, low), 30/98 (31%) men were classified into lower risk groups by TMA than WSI scores (23/80 [29%] Black men). Conclusions: In this application of the MMAI biomarker to RP samples within this cohort of primarily Black men, we found WSI-derived and TMA-derived MMAI scores to be significantly different but moderately correlated. Given that TMA represents only a portion of WSI and sampling location may impact the results, TMA may be insufficiently reliable for MMAI score generation. MMAI biomarkers that are robust across different specimen preparation methods have potential for clinical and research utility; these hypothesis-generating results support further research along these lines. Continuous MMAI scores and categorical MMAI risk group distribution using WSI- and TMA-derived images by subgroup. WSI-derived score TMA-derived score Patients Continuous ¹ Low ² Int ² High ² Continuous ¹ Low ² Int ² High ² All 0.47(0.44-0.53) 0(0%) 69(70%) 29(30%) 0.37(0.33-0.42) 6(6%) 87(89%) 5(5%) Black 0.47(0.45-0.51) 0(0%) 58(73%) 22(27%) 0.37(0.33-0.43) 4(5%) 73(91%) 3(4%) White 0.48(0.43-0.54) 0(0%) 11(61%) 7(39%) 0.37(0.32-0.41) 2(11%) 14(78%) 2(11%) ¹ Presented as median (IQR). ² Presented as n (%).

PURPOSE The SPARTAN trial demonstrated that the addition of apalutamide to androgen deprivation therapy improves outcomes among patients with nonmetastatic castration-resistant prostate cancer (nmCRPC). We applied a previously reported digital histopathology–based multimodal artificial intelligence (MMAI) algorithm to estimate clinical outcomes in SPARTAN. METHODS Patients with available hematoxylin and eosin-stained slides from the primary tumor were included. Histopathology slides were digitized. MMAI scores ranging from 0 to 1 were generated from digital histopathology and baseline clinical parameters. Patients were categorized into MMAI non–high-risk and high-risk groups using previously validated cutoffs. Kaplan-Meier estimates were calculated for metastasis-free survival (MFS), second progression-free survival (PFS2), and overall survival (OS); comparisons were performed using Cox proportional hazards regression for treatment arms and MMAI risk. The interaction between treatment arm and risk group was evaluated using a Cox proportional hazards model. RESULTS The study included 420 evaluable patients after excluding those with missing clinical data or inadequate histopathology images. Of these, 63% (n = 266) were MMAI high risk and 37% (n = 154) were non–high risk. MMAI risk score was associated with shorter MFS (hazard ratio [HR], 1.72; P < .005), PFS2 (HR, 1.57; P < .005), and OS (HR, 1.41; P = .02). MMAI high-risk patients receiving apalutamide demonstrated significant improvement in MFS (HR, 0.19; P < .005), PFS2 (HR, 0.47; P < .005), and OS (HR, 0.6; P = .01). The interaction between MMAI risk score and treatment for MFS ( P = .01) and PFS2 ( P = .03) was significant, indicating greater benefit from apalutamide treatment in MMAI high-risk patients. CONCLUSION MMAI is a prognostic marker in nmCRPC and may serve as a predictive biomarker with high-risk patients deriving the greatest benefit from treatment with apalutamide. These results represent the first extension of an MMAI classifier to patients with castration-resistant prostate cancer, warranting additional validation.

Purpose Oligometastatic castration-sensitive prostate cancer (omCSPC) is a clinically and biologically heterogeneous disease. A multimodal artificial intelligence (MMAI) algorithm (ArteraAI Prostate Test), which incorporates digital histopathology and clinical information, is prognostic for outcomes in localized prostate cancer. We hypothesized MMAI algorithms are also prognostic in omCSPC and correlate with tumor biology. Thus, we aimed to evaluate the association between MMAI score vector features (VF) with outcomes and genomics of omCSPC. Materials/Methods We correlated somatic pathogenic mutations and ArteraAI MMAI scores from 168 omCSPC patients. RNAseq profiling was performed on a subset of 65 metachronous patients. Somatic nonsynonymous pathogenic mutations from panel DNAseq were identified using Mutect2 and ClinVAR database. Pair-end reads of RNAseq were aligned to Human reference (hg38) using HISAT2. To identify the differentially mutated genes associated with MMAI scores, samples were binned separately based on the median or quartile MMAI scores, and Fisher’s exact test were used to evaluate differentially mutated genes between bins. Differential expression of genes were evaluated using DEseq2 and gene set enrichment analysis. VFs from the MMAI scores and correlated with mutations using Wilcoxon test. Results Median follow-up was 34.7 months. Overall survival was shorter in patients with high MMAI scores (top quartile) compared to those with low MMAI scores (bottom three quartiles, p=0.017). MMAI scores were significantly higher in synchronous compared to metachronous omCSPC (p<0.05). DNAseq demonstrated high MMAI scores were associated with higher incidence of BRCA2/ATM (p=0.008) and WNT pathway (APC/CTNNB1) mutations (p=0.13). Conversely, high MMAI scores were associated with lower incidence of SPOP mutations (p=0.03). RNAseq revealed differential gene expression based on MMAI score, with higher scores being enriched for epithelial-mesenchymal transition (EMT) pathway expression (p<0.01). Of the 128 VFs that contribute to MMAI scores, 32 were associated with somatic mutations. Specifically, 15 VFs associated with mutations in DNA damage response (p<0.01); 8 VFs associated with mutated genes of PIK3 pathway(p<0.01); 4 VFs associated with mutated genes of TP53 pathway(p<0.01); and 1 VF associated with genomic alteration in WNT pathway genes. Conclusions We demonstrate digital histopathology features using MMAI algorithms are prognostic for outcomes in omCSPC. We directly correlated MMAI scores with DNA mutations and transcriptional programs involved in metastatic propagation with higher scores associated with aggressive alterations (WNT, BRCA2/ATM, EMT) and lower scores associated with more indolent alterations (SPOP). Furthermore, VF were able to be directly correlated with DNA mutations. This suggests digital histopathology-based MMAI algorithms identify phenotypic pathologic features correlated with underlying biological genomic and transcriptomic processes and can be leveraged to better understand the heterogeneity of omCSPC. Citation Format: Matthew P. Deek, Yang Song, Amol Shetty, Philip Sutera, Adrianna A. Mendes, Kim Van der Eecken, Emmalyn Chen, Timothy Showalter, Trevor J Royce, Tamara Todorovic, Huei-Chung Huang, Scott A. Houck, Rikiya Yamashita, Ana Ponce Kiess, Daniel Y. Song, Tamara Lotan, Andre Esteva, Felix Y. Feng, Piet Ost, Phuoc T. Tran.A digital pathology multimodal artificial intelligence algorithm is associated with pro-metastatic genomic pathways in oligometastatic prostate cancer.[abstract]. In: Proceedings of the AACR Special Conference in Cancer Research: Translating Targeted Therapies in Combination with Radiotherapy; 2025 Jan 26-29; San Diego, CA. Philadelphia (PA): AACR; Clin Cancer Res 2025;31(2_Suppl):Abstract nr P007

Purpose: Clinical variables alone have limited ability to determine which patients will have recurrence after radical prostatectomy (RP). We evaluated the ability of locked multimodal artificial intelligence (MMAI) algorithms trained on prostate biopsy specimens to predict prostate cancer-specific mortality (PCSM) and overall survival (OS) among patients undergoing RP with digitized RP specimens. Materials and methods: The Prostate, Lung, Colorectal, and Ovarian Cancer Screening Randomized Controlled Trial randomized subjects from 1993 to 2001 to cancer screening or control. A subset of patients who underwent RP with available digitized histopathological images and subsequent survival data were identified. Distant metastasis (DM) and PCSM MMAIs originally trained on biopsy slides for patients undergoing radiation were evaluated for prediction of PCSM and OS. Cox proportional hazards modeling and Kaplan-Meier survival curve analysis were used. Results: In total, 1032 patients who underwent RP with median follow-up of 17 years (IQR, 14.3, 19.3 years) were identified. MMAI algorithms for PCSM and DM both predicted PCSM (HR, 2.31, 95% CI, 1.6-3.35, P < .001 and HR, 1.96, 95% CI, 1.35-2.85, P < .001, respectively). Similarly, DM and PCSM MMAI predicted OS (HR, 1.22, 95% CI, 1.01-1.47, P = .04 and HR, 1.19, 95% CI, 1.02-1.4, P = .03). Conclusions: Locked MMAI algorithms previously developed and validated on biopsy specimens from patients undergoing radiation for prostate cancer successfully predicted clinical outcomes when applied to RP specimens from patients treated with surgery. MMAI models and other biomarkers may help select patients who may benefit from postoperative treatment intensification with androgen deprivation therapy or radiation.