(B-224) Towards Optimizing Identification of First Line of Therapy (LOT) for Metastatic Lung, Breast and Colorectal Cancer in Real World Data: A Scoping Review

Background: Identifying the start of treatment in real-world data (RWD) oncology studies is crucial for avoiding immortal time bias in treatment-based cohorts. When defining cohorts with solid metastatic cancer in RWD, it is critical to distinguish lines of treatment (LOTs) intended for metastatic disease from those intended for prior non-metastatic disease. However, most RWD lack granular information on LOTs. The evidence about algorithms used to identify the first LOT for solid metastatic cancer in global RWD studies has not been systematically examined.

Objectives: This scoping review aimed to describe and synthesize existing algorithms for identifying the initiation of LOTs for metastatic lung, breast, and colorectal cancer across global RWD.

Methods: The scoping review was conducted using Joanna Briggs Institute guidelines. Searches were performed in MEDLINE (PubMed), Embase, Google Scholar, Web of Science Core Collection and grey literature, focusing on English language publications from January 1, 2014, to April 29, 2024. The review included non-interventional studies that utilized claims data, cancer registries, electronic health records (EHR) or paper medical charts. Included studies were required to report their algorithms for identifying the first LOT for metastatic lung, breast, and colorectal cancer in RWD studies of adults aged 18 years or older. Pairs of reviewers independently screened articles, resolving disagreements with a third reviewer.

Results: Our review identified 24 studies that met our inclusion criteria (9 lung cancer, 8 breast cancer, 7 colorectal cancer). Most studies were conducted in the United States (n=16) and Europe (n=5). EHR (n=16) and claims (n = 8) were the most frequently used RWD sources to develop algorithms across included studies. Fourteen algorithms (58%) considered the first systemic anti-cancer therapy (SACT) recorded within a pre-specified interval after an incident diagnosis of metastatic lung (n = 7), breast (n=5) or colorectal (n=2) cancer (index date) as the first LOT for metastatic disease. Others (n= 10) applied the presence of a treatment gap (≥3-6 months) before new LOT to indicate the first LOT for metastatic disease. Three studies overall validated their algorithms using manual chart review.

Conclusions: Observance of a new SACT after incident diagnosis of metastatic disease may indicate the first LOT for metastatic lung, breast or colorectal cancer in global RWD. Ongoing research and conceptual frameworks are needed to optimize existing RWD algorithms for enumerating oncology LOT. Validation of these approaches through medical chart review and multidisciplinary consensus is necessary.