Search

KR-20260065833-A - Alternative Potential Target Setting System and Method

KR20260065833AKR 20260065833 AKR20260065833 AKR 20260065833AKR-20260065833-A

Abstract

The present embodiment relates to the manufacture of an alternative potency target setting. The subject of the present embodiment provides a method, a computer system, and a computer-readable storage medium implemented by a computer for predicting the result of a method for manufacturing an alternative potency target setting.

Inventors

  • 판자니, 샴
  • 스페트시리스, 콘스탄티노스
  • 쉔, 대릭

Assignees

  • 바이엘 헬쓰케어 엘엘씨

Dates

Publication Date
20260511
Application Date
20240909
Priority Date
20230911

Claims (18)

  1. As a method implemented by a computer: (a) providing a front-end dashboard including user input for inputting user input parameters into a data acquisition module to determine alternative potency target values, and output for displaying processed results; and (b) preprocessing data from a data acquisition module, including normalization; applying statistical modeling calculations using OOS (out of specification) risk selection criteria to the preprocessed data; and providing a backend for outputting results; A method implemented by a computer that determines alternative potency target values and outputs each OOS risk.
  2. A method implemented by a computer according to claim 1, wherein user input parameters include a finished pharmaceutical product, a filling amount, a batch selection, and a manufacturing date range.
  3. A method implemented by a computer according to claim 1, wherein the data acquisition module further includes a source data system.
  4. A method implemented by a computer in which, in paragraph 3, the data acquisition module further includes a data acquisition system.
  5. A method implemented by a computer in which, in paragraph 4, the data acquisition module further includes a cloud-based data storage device.
  6. A method implemented by a computer, wherein, in paragraph 5, the data acquisition module further includes cloud-based computing.
  7. A method implemented by a computer according to claim 1, wherein the preprocessing step further includes a step of normalizing data by statistical analysis.
  8. In Paragraph 7, the data is the following mathematical formula: It is normalized by calculating P, which is normalized according to, and = Titer value, = Current bulk potency target, and = A method implemented by a computer, which is the target bulk potency value at the time of manufacturing.
  9. In paragraph 1, the OOS risk selection criteria for candidate alternative potency targets are, 1) : % OOS risk where at least one individual iteration result exceeds the upper release limit, provided that the calculated %CV is within user-defined limits; 2) : % OOS risk where at least one individual iteration result exceeds the lower release limit, provided that the calculated %CV is within user-defined limits; 3) : % OOS risk where the average result exceeds the lower mean release limit, provided that the calculated %CV is within user-defined limits; and 4) A method implemented by a computer that can be determined from a model of % OOS risk where at least one of the above scenarios occurs and the specifications are out of range.
  10. A method implemented by a computer, wherein, in claim 1, the data acquisition module further includes a cloud-based relational database.
  11. In paragraph 10, a method implemented by a computer in which data retrieval can be performed between user input and a cloud-based relational database.
  12. As a system implemented by a computer: (a) a frontend dashboard comprising user input for inputting user input parameters into a data acquisition module to determine alternative potency targets, and an output for displaying processed results; and (b) preprocess data from a data acquisition module including normalization; apply statistical modeling calculations using OOS (out of specification) risk selection criteria to the preprocessed data; and include a backend for outputting results; A system implemented by a computer, in which the system determines alternative potency target values and the output displays each OOS risk.
  13. A system implemented by a computer, wherein, in paragraph 12, user input parameters include a finished pharmaceutical product, a fill quantity, a batch selection, and a manufacturing date range.
  14. In paragraph 12, a system implemented by a computer, wherein the data acquisition module further includes a source data system.
  15. In paragraph 14, a system implemented by a computer, wherein the data acquisition module further includes a data acquisition system.
  16. In paragraph 15, a system implemented by a computer, wherein the data acquisition module further includes a cloud-based data storage device.
  17. In paragraph 16, a system implemented by a computer in which the data acquisition module further includes cloud-based computing.
  18. As a non-transient computer-readable storage medium storing software instructions, When the above software instruction is executed by the processor of a computer system, the computer system is caused to do the following: (a) Allowing the user to input user input parameters into the data acquisition module to determine alternative potency target values, and allowing an output to display the processed results; (b) preprocessing data from a data acquisition module, including normalization; and (c) Applying statistical modeling calculations using OOS (out of specification) risk selection criteria to preprocessed data; and outputting the results Make it execute; The above method determines alternative titer target values, and the output is a non-transient computer-readable storage medium that displays each OOS risk.

Description

Alternative Potential Target Setting System and Method The systems, methods, and computer programs disclosed in this specification relate to biopharmaceutical manufacturing. [Cross-reference to related applications] This application claims the benefit of U.S. Provisional Application No. 63/537,583 filed on September 11, 2023, the entire disclosure of which is incorporated herein by reference. The manufacturing of biopharmaceuticals requires a series of complex processes broadly classified into the categories of drug substance (DS) manufacturing and drug product (DP) manufacturing. DS manufacturing is characterized by core unit processes of cell culture and purification, while DP manufacturing involves key steps related to sterile fill-finish. The sterile fill-finish unit process involves combining the DS with excipients through validated processes including bulking, filling, freeze-drying, capping, inspection, and packaging. More specifically, frozen drug substance bags are thawed and pooled, and partially diluted with a formulation buffer. Next, the potency of the partially diluted bulk is tested, sterile filtered, and further diluted using the formulation buffer to a predefined target potency concentration (also known as the bulk potency target). Subsequently, the sterile bulk is filled into a glass vial, freeze-dried, and capped to produce a DP (drug product) vial. During the inspection phase, the Quality Control (QC) department tests DP vials according to Standard Operating Procedures (SOPs) to ensure that all Critical Quality Attributes (CQAs) are within specification limits, which serves as a requirement for commercial shipment. The specification limits for various DP CQAs are regulated by regulatory health agencies to ensure the safety and efficacy of the manufactured DP. Generally, individual vials are tested to verify they are within the specification limits, but in specific cases, the mean internal action limit may be implemented. A batch of DP with CQAs falling outside the validated range is considered ineligible for commercial shipment. Therefore, a Continued Process Verification (CPV) program is operated to continuously ensure that the manufacturing process maintains a state of control during commercial production, demonstrate an understanding of process variability at a commercial scale, and drive process improvement. The CQA of interest is the potency of DP vials, which is known to be monitored by the CPV program and controlled by the bulk potency target concentration set during the solution preparation phase of DP manufacturing. Intuitively, raising or lowering the bulk potency target causes the average potency distribution of DP vials to shift proportionally, which will affect the proportion of DP batches deemed out of specification (OOS) at shipment. As process optimization is required, refining the bulk potency target provides an opportunity to improve process capability (robustness) by reducing the probability (risk) of OOS at shipment. The established (current) bulk potency targets for the DP manufacturing process are set for each fill size (dosage strength) of the DP. Changes to the current bulk potency targets of DP are generally based on manufacturing experience (industry expertise) to ensure that business needs and quality requirements are met. Traditionally, prior to adjusting bulk potency targets, statistical evaluations of theoretical (alternative) potency targets have been performed, including estimating the Out-of-Ship (OOS) risk for each alternative potency target under consideration. Although commercial software exists to perform these analyses, conducting them in an ad-hoc manner is often time-consuming due to the cumbersome manual steps involved in the workflow. The overall workflow required to evaluate alternative potency targets involves a series of manual steps involving data acquisition, data preprocessing, and statistical calculations. In particular, when GMP (Good Manufacturing Practice) related decisions are required due to the occurrence of Out-of-Services (OOS), it is advisable to have an automated statistical analysis workflow that enables timely response. In addition to limiting the overall efficiency of the workflow, commercial software also entails constraints regarding customizability and accessibility. Unless there is a major update, commercial software is generally fixed in its functionality and user interface (UI), making it difficult to meet the ever-changing requirements of users. Furthermore, traditional commercial software requires installation on the user's local operating system and necessitates local computing resources, which potentially limits accessibility and scalability across the organization in the future. This stands in stark contrast to cloud platform-based software, which relies on a network of remote servers to provide computing resources (i.e., storage capacity, processing power, etc.). The need arises to automate the evaluation and metho