1. Problem Definition

Clinical Data Management (CDM) is a crucial aspect of clinical trials, ensuring data accuracy, consistency, and regulatory compliance. This project focuses on oncology clinical trials, where data management plays a pivotal role in assessing treatment efficacy and patient safety. The project will involve collecting, cleaning, analyzing, and reporting oncology data using SAS, following industry standards such as CDISC SDTM and ADaM.

Project Objectives:

• Establish an industry-standard Clinical Data Management System (CDMS) for oncology trials.
• Collect and validate patient demographic, adverse event, laboratory, and treatment datasets.
• Perform rigorous data cleaning, transformation, and standardization.
• Conduct statistical analysis and generate regulatory-compliant reports.
• Implement automated reporting using Base SAS, Advanced SAS, and Macros.
• Demonstrate end-to-end project implementation as in a real-world pharmaceutical industry setting.

2. Scope of Work

Phase 1: Data Collection & Importing

• Define clinical trial protocol and CRF (Case Report Form) design.
• Import oncology clinical trial data from various sources (CSV, Excel, databases, EDC).
• Establish libraries and manage data storage in SAS.

Phase 2: Data Cleaning & Transformation

• Identify and handle missing values, outliers, and inconsistencies.
• Apply data validation rules using PROC FREQ, PROC MEANS, and PROC UNIVARIATE.
• Standardize datasets following CDISC SDTM format.
• Implement SAS Macros for automation and efficiency.

Phase 3: Data Integration & Preparation

• Merge and integrate multiple clinical datasets (Demographics, Adverse Events, Laboratory Results, Treatment Data).
• Ensure consistency between raw and cleaned datasets.
• Prepare datasets for statistical analysis (ADaM datasets).

Phase 4: Statistical Analysis & Reporting

• Generate summary statistics using PROC MEANS and PROC UNIVARIATE.
• Perform survival analysis using PROC LIFETEST and PROC PHREG.
• Generate safety and efficacy reports using PROC REPORT and PROC TABULATE.
• Create customized visualizations using PROC SGPLOT and PROC GPLOT.

Phase 5: Compliance & Submission

• Structure datasets per regulatory guidelines (CDISC, FDA, EMA).
• Generate Submission-Ready Analysis Data Model (ADaM) datasets.
• Validate datasets using Pinnacle21.
• Create define.xml for submission to regulatory authorities.

Phase 6: Final Delivery & Documentation

• Submit clinical trial datasets, SAS programs, and logs.
• Deliver statistical summaries, reports, and visualizations.
• Provide detailed project documentation and user manuals.

3. Gantt Chart for Project Management

4. Creating Oncology Datasets in SAS (Minimum 10-20 Observations)

Demographics Dataset

/* Creating Demographics Dataset */

data work.demographics;

   input Patient_ID $ Age Gender $ Race $ Enrollment_Date MMDDYY10. Study_Arm $;

   format Enrollment_Date MMDDYY10.;

datalines;

   P001 45 M Asian 01/15/2023 Treatment

   P002 60 F White 02/20/2023 Control

   P003 52 M Black 03/10/2023 Treatment

   P004 67 F Hispanic 04/05/2023 Control

   P005 59 M White 05/12/2023 Treatment

   P006 41 F Asian 06/18/2023 Control

   P007 55 M Black 07/22/2023 Treatment

   P008 62 F Hispanic 08/30/2023 Control

   P009 47 M White 09/15/2023 Treatment

   P010 50 F Asian 10/08/2023 Control

   ;

run;

Adverse Events Dataset

/* Creating Adverse Events Dataset */

data work.adverse_events;

   input Patient_ID $ Event_Type $ Severity $ Outcome $ Treatment_Relationship $;

datalines;

   P001 Nausea Mild Resolved Related

   P002 Fatigue Moderate Ongoing Unrelated

   P003 Headache Severe Resolved Related

   P004 Rash Mild Resolved Related

   P005 Dizziness Moderate Ongoing Related

   P006 Vomiting Severe Resolved Unrelated

   P007 Fever Mild Resolved Related

   P008 Diarrhea Moderate Ongoing Unrelated

   P009 Fatigue Severe Ongoing Related

   P010 Headache Mild Resolved Related

   ;

run;

Treatment Dataset

/* Creating Treatment Dataset */

data work.treatment;

   input Patient_ID $ Drug_Name $ Dose Duration Response $ Discontinuation_Reason $;

datalines;

   P001 DrugA 50mg 6months Positive None

   P002 DrugB 75mg 4months Negative Side Effects

   P003 DrugC 100mg 8months Positive None

   P004 DrugA 50mg 6months Neutral None

   P005 DrugB 75mg 4months Negative Side Effects

   P006 DrugC 100mg 8months Positive None

   P007 DrugA 50mg 6months Neutral None

   P008 DrugB 75mg 4months Negative Side Effects

   P009 DrugC 100mg 8months Positive None

   P010 DrugA 50mg 6months Neutral None

   ;

run;

5. Implementing the Project Using Base SAS & Advanced SAS

/* Merging Demographics and Treatment Data */

data work.final_data;

   merge work.demographicswork.treatment;

   by Patient_ID;

run;

6. SAS Code for Industry Implementation

/* Generating Adverse Events Summary */

proc freq data=work.adverse_events;

   tables Severity Treatment_Relationship;

run;

7. Industry Implementation Clarifications

• Base SAS: Used for importing, cleaning, merging, and transforming data.
• Advanced SAS: Includes Macro programming, SQL, and statistical analysis.
• SAS Graphs & Reports: Used to generate listings, summaries, and visualizations.
• Regulatory Compliance: Ensuring CDISC standards (SDTM &ADaM) for FDA submission.

8. Delivery Procedure

• Submission of Clinical Datasets: Final datasets formatted per CDISC standards.
• Delivery of SAS Programs: Well-documented SAS scripts and logs.
• Statistical Reports: Detailed reports covering safety, efficacy, and survival analysis.
• Regulatory Compliance Files: Define.xml and Pinnacle21 validation reports.
• User Documentation: Guidelines for dataset usage and data flow explanations.

This project demonstrates a complete, real-world Clinical Data Management implementation using SAS. Covering all major aspects of CDM, from data collection to regulatory submission, this project serves as a reference for industry professionals and students alike, providing a hands-on experience in oncology clinical trial data management.