Metastatic tumor burden and the modified Glasgow Prognostic Score are independent prognostic factors in metastatic triple‑negative breast cancer: A multicenter real‑world study

Metastatic triple‑negative breast cancer (mTNBC) is characterized by aggressive clinical behavior, poor prognosis and limited therapeutic options. Reliable prognostic factors to guide patient stratification in this setting remain elusive. The present study compared the prognostic impact of metastatic tumor burden and a broadened panel of systemic inflammatory and nutritional indices in mTNBC. The present retrospective multicenter study included 104 patients with mTNBC treated across three centers in Izmir, Turkey, between 2014 and 2025. The median age at metastatic diagnosis was 58 years, and the median follow‑up was 27.2 months. Patients with 1‑2 metastatic sites exhibited a significantly longer progression‑free survival (PFS; 10.6 vs. 5.9 months; P<0.001) and overall survival (OS; 27.9 vs. 14.5 months; P<0.001) than those with ≥3 sites. In multivariable analysis, ≥3 metastatic sites remained independently associated with a shorter PFS [hazard ratio (HR) 1.79; 95% confidence intervals (CI) 1.10‑2.90; P=0.018] and worse OS [HR 2.76; 95% CI 1.56‑4.88; P<0.001]. Among the inflammatory and nutritional indices evaluated, the modified Glasgow Prognostic Score was independently associated with OS (score ≥1 vs. 0: HR 2.03; 95% CI 1.10‑3.73; P=0.023; median OS 15.5 vs. 26.4 months), as was the systemic inflammation response index (SIRI) at its conventional threshold (SIRI ≥1.0: HR 2.05; 95% CI 1.08‑3.90; P=0.028). By contrast, the neutrophil‑to‑lymphocyte, monocyte‑to‑lymphocyte, neutrophil‑to‑monocyte and platelet‑to‑lymphocyte ratios, the systemic immune‑inflammation index, the prognostic nutritional index, albumin‑bilirubin, hemoglobin‑albumin‑lymphocyte‑platelet index and the De Ritis ratio showed no significant association with either endpoint, and data‑derived cut‑offs for these indices did not survive permutation correction. No index was associated with the disease‑free interval. Metastatic tumor burden was the dominant prognostic determinant in this cohort, but inflammation‑based scores incorporating C‑reactive protein retained an independent prognostic value, whereas indices derived solely from the differential leucocyte count did not.