Treatment patterns and clinical outcomes according to PD-L1 status in >2000 patients with early-stage or metastatic triple-negative breast cancer treated in the real-world setting: VANESSA study results
Popovic L.; Barroso-Sousa R.; El Saghir N.S.; Dent R.; Tuzlali S.; Akhtar S.; Juozaityté E.; Eglitis J.; Doval D.C.; Castaneda C.A.; Zirtiloglu A.; Hartleben G.; Deurloo R.; Toro P.; Estaytieh I.; Weber E.; Mouta J.; D'Arrigo C. · Breast (2026) · DOI: 10.1016/j.breast.2026.104720
Resumen
Background: The prognostic effect of PD-L1 status in triple-negative breast cancer (TNBC) is uncertain and little is known about PD-L1-positive prevalence and outcomes in the real-world setting. Patients and methods: The multicentre retrospective observational VANESSA study evaluated the prevalence and impact of PD-L1-positive status in 2054 patients receiving systemic therapy for early-stage or metastatic (e/m)TNBC between 2014 and 2017. PD-L1 expression was assessed locally and centrally on archival samples. Descriptive analyses of demographic and clinicopathological characteristics, treatment patterns and clinical outcomes (extracted from patients’ medical records) according to PD-L1 status were prespecified. Results: Among 1902 patients with eTNBC, 681 (36%) received neoadjuvant chemotherapy and 1261 (66%) adjuvant chemotherapy. Demographic characteristics were generally similar regardless of PD-L1 status, but lower-risk tumour characteristics were more common in the PD-L1-positive subgroup. Invasive disease-free and overall survival were more favourable in PD-L1-positive eTNBC. In the mTNBC cohort, 120/145 (83%) patients had de novo mTNBC. Median progression-free survival on first-line treatment was 7.6 months (95% CI: 4.1–15.0) in PD-L1-positive mTNBC (n = 30) and 4.9 months (95% CI: 3.6–6.1) in PD-L1-negative mTNBC (n = 83). Conclusion: In eTNBC and mTNBC, PD-L1-positive status was associated with more favourable long-term outcomes, possibly due to tumour-intrinsic characteristics and/or the host immune response. The high proportion with de novo mTNBC may suggest enrolment bias and/or geographic variations in stage at diagnosis.
Instituciones: Universidad Científica del Sur