NETWORK META-ANALYSIS (NMA) OF NEOADJUVANT-ADJUVANT TREATMENT PATHWAYS IN HER2+ EARLY BREAST CANCER (EBC) TO INFORM COST-EFFECTIVENESS
Author(s)
Giorgia Rapesta1, Carin Uyl-De Groot, Sr., PhD2, Isaac Corro Ramos, MSc, PhD3, Valesca P. Retel, MSc, PhD4.
1Erasmus School of Health Policy and Management, Rotterdam, Netherlands, 2ESHPM/iMTA Erasmus University Rotterdam, Rotterdam, Netherlands, 3Erasmus University, Rotterdam, Netherlands, 4Netherlands Cancer Institute (NKI-AVL), Amsterdam, Netherlands.
1Erasmus School of Health Policy and Management, Rotterdam, Netherlands, 2ESHPM/iMTA Erasmus University Rotterdam, Rotterdam, Netherlands, 3Erasmus University, Rotterdam, Netherlands, 4Netherlands Cancer Institute (NKI-AVL), Amsterdam, Netherlands.
OBJECTIVES: Network meta-analysis (NMA) is increasingly used in Health Technology Assessment (HTA) to compare multiple interventions and provide comparative-effectiveness inputs for Cost-Effectiveness Analyses (CEA). We conducted Bayesian NMAs of pathological Complete Response (pCR) and Overall Survival (OS) across neoadjuvant-adjuvant treatment pathways in HER2+ early (stage II-III) breast cancer to synthesize current evidence.
METHODS: We conducted a systematic literature review following a PROSPERO protocol (CRD420261319726). Eligible studies enrolled patients with HER2+ early (stage II-III) breast cancer and all reported in detail neoadjuvant treatment, pCR, adjuvant treatment, and survival follow-up. Bayesian random-effects NMAs were conducted in R (version 4.4.0) using the ‘multinma’ package. Regimens were grouped into ten nodes including both chemotherapy-based and chemotherapy-free treatments, capturing escalation and de-escalation strategies. Data validation, model-fit, consistency, and node-splitting checks were performed.
RESULTS: Concerning the pCR network, relative to chemotherapy plus dual HER2 blockade, ADC[1] plus chemotherapy plus dual HER2 blockade had the highest pCR odds ratio point estimate, followed by chemotherapy plus dual HER2 blockade plus immunotherapy. Rank-probability analysis indicated that ADC plus chemotherapy plus dual HER2 blockade had the highest Probability of Being Best for pCR. In the OS NMA, relative to chemotherapy alone, chemotherapy plus dual HER2 blockade showed the most favorable rate ratio. NMA diagnostics did not suggest major concerns.
[1] Antibody-Drug Conjugate
CONCLUSIONS: This Bayesian NMA provides comparative pCR and OS estimates across relevant treatment pathways for HER2+ early breast cancer. Cross-outcome comparison suggested that dual HER2 blockade consistently improved both pCR and OS, whereas they were not fully concordant across all strategies. Its pathway-based design supports future integration of pCR, survival in a cost-effectiveness model. Additionally, the research provides interesting insights into the relationship between surrogate end-point and long-term outcomes in HER2+ early breast cancer.
METHODS: We conducted a systematic literature review following a PROSPERO protocol (CRD420261319726). Eligible studies enrolled patients with HER2+ early (stage II-III) breast cancer and all reported in detail neoadjuvant treatment, pCR, adjuvant treatment, and survival follow-up. Bayesian random-effects NMAs were conducted in R (version 4.4.0) using the ‘multinma’ package. Regimens were grouped into ten nodes including both chemotherapy-based and chemotherapy-free treatments, capturing escalation and de-escalation strategies. Data validation, model-fit, consistency, and node-splitting checks were performed.
RESULTS: Concerning the pCR network, relative to chemotherapy plus dual HER2 blockade, ADC[1] plus chemotherapy plus dual HER2 blockade had the highest pCR odds ratio point estimate, followed by chemotherapy plus dual HER2 blockade plus immunotherapy. Rank-probability analysis indicated that ADC plus chemotherapy plus dual HER2 blockade had the highest Probability of Being Best for pCR. In the OS NMA, relative to chemotherapy alone, chemotherapy plus dual HER2 blockade showed the most favorable rate ratio. NMA diagnostics did not suggest major concerns.
[1] Antibody-Drug Conjugate
CONCLUSIONS: This Bayesian NMA provides comparative pCR and OS estimates across relevant treatment pathways for HER2+ early breast cancer. Cross-outcome comparison suggested that dual HER2 blockade consistently improved both pCR and OS, whereas they were not fully concordant across all strategies. Its pathway-based design supports future integration of pCR, survival in a cost-effectiveness model. Additionally, the research provides interesting insights into the relationship between surrogate end-point and long-term outcomes in HER2+ early breast cancer.
Conference/Value in Health Info
2026-11, ISPOR Europe 2026, Vienna, Austria
Value in Health, Volume 29, Issue 12S
Code
CO21
Topic
Clinical Outcomes, Methodological & Statistical Research, Real World Data & Information Systems
Topic Subcategory
Comparative Effectiveness or Efficacy
Disease
Oncology