USE OF REAL-TIME AI-ASSISTED LIVING SYSTEMATIC LITERATURE REVIEW (REAL-SLR) IN PROSTATE CANCER (PC) ENABLES ON-DEMAND ACCESS TO PIVOTAL CLINICAL TRIAL EVIDENCE
Author(s)
Rhiannon Campden, PhD1, Xinyue Jiang, PhD1, Saro Sarkisian, MD, MHA2, Jessicca Rege, PhD1, Rozee Liu, MSc1, Anna Forsythe, MBA, MSc, PharmD1;
1Oncoscope-AI, Miami, FL, USA, 2Frederick Health, Frederick, MD, USA
1Oncoscope-AI, Miami, FL, USA, 2Frederick Health, Frederick, MD, USA
OBJECTIVES: Systemic literature reviews (SLRs) are essential to health economics and outcomes research (HEOR), but are commonly conducted through manual, ad hoc processes that are time- and resource-intensive and quickly become outdated. This study describes and evaluates the efficiency of a REAL-SLR in PC.
METHODS: A REAL-SLR was used to build a PC library incorporating standardized search strategies and daily updates with AI-assisted screening and extraction of relevant interventional and surveillance clinical trials. Evidence was reviewed in compliance with PRISMA/Cochrane guidelines using the Population, Intervention/Comparators, Outcomes, and Study design (PI/COS) framework. Operational efficiency was evaluated by comparing the workflows, staffing requirements, and timelines between the REAL-SLR and a conventional ad hoc manual SLR, focusing on abstracts added in the 2025 calendar year.
RESULTS: In 2025, the REAL-SLR approach was applied to screen 15,048 records and added 1,193 new abstracts, with 57 from ASCO GU, 50 from ASCO, 43 from ESMO. Conference abstracts were included within 5-10 days of publication. The REAL-SLR approach provided immediate, on-demand access to structured data through an online platform with lower staffing requirements, saving >90% of total project time. In comparison, a conventional SLR of the same size would require 12-14 weeks of full-time commitment from 2 junior and 1 senior researcher. The REAL-SLR identified potentially practice-changing treatments for metastatic castration-resistant PC (mCSPC) treatments previously not available in earlier settings, including first line use of [177Lu]Lu-PSMA-617, niraparib+abiraterone for BRCA-mutated mCSPC, and capivasertib for PTEN-mutated mCSPC. The on-demand availability of these publications prior to approval or guideline adaptation through the online platform can provide visibility to treatment options for patients and inform HEOR and market access activities.
CONCLUSIONS: REAL-SLR enables timely, resource-efficient evidence synthesis for oncology HEOR and market access. Transitioning from episodic manual reviews to a daily-updated model lowers costs and shortens timelines, while maintaining access to up-to-date, decision-ready evidence.
METHODS: A REAL-SLR was used to build a PC library incorporating standardized search strategies and daily updates with AI-assisted screening and extraction of relevant interventional and surveillance clinical trials. Evidence was reviewed in compliance with PRISMA/Cochrane guidelines using the Population, Intervention/Comparators, Outcomes, and Study design (PI/COS) framework. Operational efficiency was evaluated by comparing the workflows, staffing requirements, and timelines between the REAL-SLR and a conventional ad hoc manual SLR, focusing on abstracts added in the 2025 calendar year.
RESULTS: In 2025, the REAL-SLR approach was applied to screen 15,048 records and added 1,193 new abstracts, with 57 from ASCO GU, 50 from ASCO, 43 from ESMO. Conference abstracts were included within 5-10 days of publication. The REAL-SLR approach provided immediate, on-demand access to structured data through an online platform with lower staffing requirements, saving >90% of total project time. In comparison, a conventional SLR of the same size would require 12-14 weeks of full-time commitment from 2 junior and 1 senior researcher. The REAL-SLR identified potentially practice-changing treatments for metastatic castration-resistant PC (mCSPC) treatments previously not available in earlier settings, including first line use of [177Lu]Lu-PSMA-617, niraparib+abiraterone for BRCA-mutated mCSPC, and capivasertib for PTEN-mutated mCSPC. The on-demand availability of these publications prior to approval or guideline adaptation through the online platform can provide visibility to treatment options for patients and inform HEOR and market access activities.
CONCLUSIONS: REAL-SLR enables timely, resource-efficient evidence synthesis for oncology HEOR and market access. Transitioning from episodic manual reviews to a daily-updated model lowers costs and shortens timelines, while maintaining access to up-to-date, decision-ready evidence.
Conference/Value in Health Info
2026-05, ISPOR 2026, Philadelphia, PA, USA
Value in Health, Volume 29, Issue S6
Code
HTA63
Topic
Health Technology Assessment
Disease
SDC: Oncology, SDC: Reproductive & Sexual Health