Development and Validation of a Claims-Based Approach to Identify Patients with Metastatic Triple Negative Breast Cancer (mTNBC)
Moderator
Nazneen Fatima Shaikh, BS, PhD, IQVIA, Inc., Harrison, NJ, United States
Speakers
Jenny Tse, MS, IQVIA, Cambridge, MA, United States; Dajun Tian; Simon Collin; Aimee Near
OBJECTIVES: This retrospective study used US claims data to develop a treatment-based proxy to identify patients with metastatic triple negative breast cancer (mTNBC) and evaluated the performance of the proxy using electronic medical records (EMR).
METHODS: Between 03/2017-9/2023, patients with ≥1 ICD-10 diagnosis code for BC and metastasis were identified in IQVIA PharMetrics® Plus. Patients with continuous enrollment 6 months prior to (baseline) and ≥3 months after (follow-up) the index date (earliest metastasis code), and no diagnosis codes for other cancers or metastasis within 15 months pre-index were included. A treatment-based proxy was used to ascertain TNBC (i.e., HR- and HER2-) where patients with no claims for treatments indicated for HR+ HER2- (CDK4/6, mTOR, PIK3CA, AKT inhibitors), HR+ (endocrine therapy), or HER2+ mBC during the study period were included. Patients were further linked to IQVIA Oncology EMR for HR and HER2 status based on biomarker testing.
RESULTS: Out of the 32,396 mBC patients identified in claims, 3,651 (11.3%) patients had ≥1 line of therapy and were defined as TNBC using the treatment-based proxy. Of these, 117 (3.2%) patients were linkable to Oncology EMR and 54 linked patients had non-missing HR and HER2 biomarker data. The positive predictive value of the treatment-based proxy was 68.5% for HR- status and 98.1% for HER2- status; 37 (68.5%) patients were classified as having TNBC. In OncEMR, 16 patients (29.6%) had evidence of HR+ HER2- disease and were misclassified by the treatment-based proxy. These patients had claims for chemotherapy and no evidence of targeted therapy during the study period.
CONCLUSIONS: Due to the lack of biomarker data in claims databases, patients with mTNBC are often identified using treatment-based proxies. Although patients with mTNBC can be identified based on the absence of treatments indicated for HR+ or HER2+ mBC, additional criteria are needed to classify patients treated with chemotherapy only.
METHODS: Between 03/2017-9/2023, patients with ≥1 ICD-10 diagnosis code for BC and metastasis were identified in IQVIA PharMetrics® Plus. Patients with continuous enrollment 6 months prior to (baseline) and ≥3 months after (follow-up) the index date (earliest metastasis code), and no diagnosis codes for other cancers or metastasis within 15 months pre-index were included. A treatment-based proxy was used to ascertain TNBC (i.e., HR- and HER2-) where patients with no claims for treatments indicated for HR+ HER2- (CDK4/6, mTOR, PIK3CA, AKT inhibitors), HR+ (endocrine therapy), or HER2+ mBC during the study period were included. Patients were further linked to IQVIA Oncology EMR for HR and HER2 status based on biomarker testing.
RESULTS: Out of the 32,396 mBC patients identified in claims, 3,651 (11.3%) patients had ≥1 line of therapy and were defined as TNBC using the treatment-based proxy. Of these, 117 (3.2%) patients were linkable to Oncology EMR and 54 linked patients had non-missing HR and HER2 biomarker data. The positive predictive value of the treatment-based proxy was 68.5% for HR- status and 98.1% for HER2- status; 37 (68.5%) patients were classified as having TNBC. In OncEMR, 16 patients (29.6%) had evidence of HR+ HER2- disease and were misclassified by the treatment-based proxy. These patients had claims for chemotherapy and no evidence of targeted therapy during the study period.
CONCLUSIONS: Due to the lack of biomarker data in claims databases, patients with mTNBC are often identified using treatment-based proxies. Although patients with mTNBC can be identified based on the absence of treatments indicated for HR+ or HER2+ mBC, additional criteria are needed to classify patients treated with chemotherapy only.
Conference/Value in Health Info
2025-05, ISPOR 2025, Montréal, Quebec, CA
Value in Health, Volume 28, Issue S1
Code
RWD21
Topic
Real World Data & Information Systems
Topic Subcategory
Reproducibility & Replicability
Disease
No Additional Disease & Conditions/Specialized Treatment Areas, SDC: Oncology