The Challenge
The client is one of the world’s leaders in the development of drugs for cancer treatment. The client’s holding company operates under two divisions (Pharmaceuticals and Diagnostics) and is listed among the Fortune 100. The firm employed a team of 70 medical writers, who would manually draft clinical study protocol documents necessary for submission to regulatory authorities and ethics committees. Typically, the writers would leverage multiple data sources in this effort, including historical protocols, online information, and internal repositories.
The company asked Persistent to automate the process of drafting clinical study protocol documents, ideally saving both time and money while producing more accurate and persuasive content. These requirements called for harnessing the cloud, specifically Google Cloud Platform (GCP), and Generative AI (GenAI) technologies to drive team efficiency and document quality.
The Solution
Persistent recommended building the solution on GCP Vertex AI – a fully-managed, unified GenAI development platform. We also recommended leveraging Google Foundation Large Language Models (LLMs), a NextGen large language model with improved multilingual, reasoning, and coding capabilities, which excel at advanced reasoning (including code and math), as well as natural language generation with translation and multilingual proficiency. Other solution components included Google Cloud Run (for front-end and back-end services, batch jobs, deploying websites and applications, etc.), Vertex Matching Engine (a high-scale low-latency vector database), and BigQuery (a serverless multi-cloud data warehouse). Persistent also employed its NOVA accelerator to build the solution faster and provide a multi-agent framework to interact with multiple source systems, including multiple data types.
The Persistent solution, powered by GCP technologies, enabled advanced search capabilities for the firm’s medical writing team, with GenAI-led drafting of complex documents. The app-based solution also accessed highly regulated documentation and intuitive access to summarized content on request via Google Docs and Microsoft plugins. The solution’s GenAI interface served up referenceable citations from the World Health Organization (WHO) and PubMed to buttress specific claims and clinical results discussed in protocol documents. In addition, the Google Cloud-based solution was globally scalable and available, and literally almost twice as fast as formerly manual processes.
The Outcome
The advantages of leveraging a solution utilizing GCP, Google Foundation Large Language Models, and other Google technologies soon yielded measurable results. Medical writers experienced 80% faster composition of clinical study protocols, along with a simultaneous 40% quality improvement thanks in part to GenAI providing relevant citations.
From our client’s perspective, the solution also paid off economically, with a 60% cost savings on the expense of creating clinical study protocols. As with all technology solutions, the real proof is in the practical benefits. In this case, it means a document creation process that’s faster and better, and at lower costs, for one of the world’s leading pharmaceutical houses – which in turn leads to accelerated drug discoveries that could benefits untold number of cancer patients.
Persistent recently announced a Strategic Partnership Agreement with Google Cloud to jointly deliver market-leading solutions that help enterprises maximize the ROI from their cloud investments while modernizing their infrastructure and data stack. Contact us today for more details.
