Common Criticism of Patient-Reported Outcomes (PRO) in Health Technology Assessments (HTA) by the Institute for Quality and Efficiency in Healthcare (IQWiG) in Breast Cancer and Non-Small Cell Lung Cancer (NSCLC)
Author(s)
Jenya Antonova, MS, PhD1, Jillian Lusk, MPH2, Niclas Kuerschner, MPH3, Sarah Boehme, MS3, Magdalena Harrington, PhD4.
1Compass Strategy and Research Inc., San Francisco, CA, USA, 2Pfizer, Pittsubrgh, PA, USA, 3Pfizer Pharma GmbH, Berlin, Germany, 4Pfizer, Milford, NH, USA.
1Compass Strategy and Research Inc., San Francisco, CA, USA, 2Pfizer, Pittsubrgh, PA, USA, 3Pfizer Pharma GmbH, Berlin, Germany, 4Pfizer, Milford, NH, USA.
OBJECTIVES: German HTA submissions require PRO data, which must be of high quality to be accepted for review. We summarized IQWiG critique of PROs in metastatic HER2- breast cancer (H2NBC) and non-small cell lung cancer (NSCLC submissions to identify key issues that may limit the utility of PRO data in HTA in Germany.
METHODS: From the IQWiG website, we obtained and reviewed assessments of treatments for H2NBC (January 2019—August 2024) and NSCLC (January 2019—February 2025) and summarized and categorized IQWiG critique.
RESULTS: Among H2NBC assessments (n=17; 11 therapeutic agents), most occurred in 2019 (n=5) or 2020 (n=5), were for HER2- HR+ subtype (n=10), in the first (n=4) or first and second line (n=8).
Among NSCLC assessments (n=17; 12 therapeutic agents), most occurred in 2019 (n=6), were in the first line (n=15), and targeted epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) negative NSCLC (n=9).
IQWiG rejected PRO data because of inadequate or unsuitable endpoint definition, high proportion of missing data, imbalance between treatment arms (in missing data, subject censoring, assessment frequency or duration), absence of relevant analyses, high proportion of subject censoring, or absent justification for selecting PRO-CTCAE items. Data quality was downgraded because of open-label trial design, insufficiently long PRO data collection, moderate subject censoring, or moderate missing data. Lacking data on symptom seriousness, IQWiG classified the symptoms as non-serious, non-severe. Notably, IQWiG rejected progression-free survival as not patient-relevant endpoint, requesting surrogate-endpoint validation and long-term health-related quality of life data.
CONCLUSIONS: To improve the utility of PRO data in IQWiG assessments, sponsors should enhance PRO data quality via appropriate clinical trial design (e.g., double-blinded design, balanced data collection between treatment arms, long-term PRO assessment post-progression) and execution (e.g., minimal missing data), generate evidence to support endpoints as patient-relevant, justify symptoms selected for assessment, define symptom seriousness, and adequately define endpoints.
METHODS: From the IQWiG website, we obtained and reviewed assessments of treatments for H2NBC (January 2019—August 2024) and NSCLC (January 2019—February 2025) and summarized and categorized IQWiG critique.
RESULTS: Among H2NBC assessments (n=17; 11 therapeutic agents), most occurred in 2019 (n=5) or 2020 (n=5), were for HER2- HR+ subtype (n=10), in the first (n=4) or first and second line (n=8).
Among NSCLC assessments (n=17; 12 therapeutic agents), most occurred in 2019 (n=6), were in the first line (n=15), and targeted epidermal growth factor receptor (EGFR) and anaplastic lymphoma kinase (ALK) negative NSCLC (n=9).
IQWiG rejected PRO data because of inadequate or unsuitable endpoint definition, high proportion of missing data, imbalance between treatment arms (in missing data, subject censoring, assessment frequency or duration), absence of relevant analyses, high proportion of subject censoring, or absent justification for selecting PRO-CTCAE items. Data quality was downgraded because of open-label trial design, insufficiently long PRO data collection, moderate subject censoring, or moderate missing data. Lacking data on symptom seriousness, IQWiG classified the symptoms as non-serious, non-severe. Notably, IQWiG rejected progression-free survival as not patient-relevant endpoint, requesting surrogate-endpoint validation and long-term health-related quality of life data.
CONCLUSIONS: To improve the utility of PRO data in IQWiG assessments, sponsors should enhance PRO data quality via appropriate clinical trial design (e.g., double-blinded design, balanced data collection between treatment arms, long-term PRO assessment post-progression) and execution (e.g., minimal missing data), generate evidence to support endpoints as patient-relevant, justify symptoms selected for assessment, define symptom seriousness, and adequately define endpoints.
Conference/Value in Health Info
2025-11, ISPOR Europe 2025, Glasgow, Scotland
Value in Health, Volume 28, Issue S2
Code
PCR48
Topic
Clinical Outcomes, Health Technology Assessment, Patient-Centered Research
Topic Subcategory
Patient-reported Outcomes & Quality of Life Outcomes
Disease
Oncology