Share on Facebook
Share on Twitter
Share on LinkedIn
Share by Email
Print
Individual participant data (IPD) packages for cancer trials are often incomplete, according to a study published in JAMA Oncology.
Researchers found that IPD packages were missing outcome data, assessment variables, adjustment data, clinical study reports, and other information.
For this study, the researchers made IPD requests related to 91 cancer trials. The trials supported US Food and Drug Administration registrations for 40 solid tumor therapies. The requests were sent to 16 different companies.
Companies did not provide IPD packages for 21 of the trials (23%). The reasons for refusal included ongoing regulatory activities (n=4), data codevelopment and/or divestment (n=4), the meta-analysis was repetitive with publicly available information (n=4), and concerns with the planned statistical evaluations (n=7). Reasons were not given for 2 of the trials.
IPD packages were provided for 70 of the trials (77%), and the median time to receipt was 123 days (range, 117-352 days). However, information was missing or redacted from some of these packages.
Outcome data were redacted from 26% of the packages. Overall survival data were removed (9%), progression- or disease-free survival data were removed (9%), and adverse event data were partially redacted (17%).
Assessment variables were redacted in 16% of packages. This included removal of race information (7%), partially redacted race information (7%), and removal of sex information (1%).
Adjustment data were redacted in 27% of packages. This included partially redacted age information (26%) and removal of weight information (14%) and performance status information (3%).
IPD packages were also missing or had incomplete data derivation specifications (30%), were missing clinical study reports (29%), were missing anonymization or redaction guides (29%), and had incomplete or missing data dictionaries (6%).
“[W]e noted substantial variability in the completeness of key data variables and supporting documents across the provided IPD packages,” the researchers wrote. “These findings suggest that there is a substantial opportunity to enhance the data sharing ecosystem, including ensuring that clinical trials indicated as eligible for sharing are truly accessible and that IPD packages meet a standard of utility.”
Disclosures: Some study authors declared affiliations with biotech, pharmaceutical, and/or device companies. Please see the original reference for a full list of disclosures.
Reference
Hopkins AM, Modi ND, Abuhelwa AY, et al. Heterogeneity and utility of pharmaceutical company sharing of individual-participant data packages. JAMA Oncol. Published online October 5, 2023. doi:10.1001/jamaoncol.2023.3996
This article originally appeared on Cancer Therapy Advisor
