Prognostic performance of thymidine kinase 1 activity in patients with hormone receptor-positive and HER2-negative metastatic breast cancer treated with CDK4/6 and aromatase inhibitors.
Thymidine kinase 1 activity (TKa) has previously been demonstrated as a prognostic biomarker for progression-free survival (PFS) in hormone receptor-positive, HER2-negative metastatic breast cancer (MBC), but its optimal use remains undefined. This study evaluated the prognostic performance of TKa across multiple sampling time points and thresholds in patients receiving first-line CDK4/6 inhibitor plus aromatase inhibitor therapy.
TKa was measured (DiviTum® assay) at baseline (BL), cycle 1 day 15 (C1D15), and cycle 2 day 1 (C2D1) in patients enrolled in the PDM-MBC study (n = 90). Thresholds for PFS discrimination were identified using maximally selected rank statistics, with predefined cut-offs tested for comparison. Prognostic performance was assessed using the corrected Akaike information criterion (AICc) and Harrell's concordance index (C-index).
TKa was prognostic at all time points. Data-derived thresholds identified groups with differing PFS and overall survival (OS), and predefined cut-offs (≥ 50, ≥ 100, ≥ 250 DiviTum® units of Activity [DuA]) also discriminated survival outcomes. BL and C2D1 models performed better than C1D15 and comparably to multi-time-point models. Among patients with BL TKa ≥ 250 DuA, suppression to < 100 DuA at C1D15 was associated with longer median PFS (23.9 vs. 10.3 months; p = 0.034).
Baseline TKa provides prognostic information, with potential added value from repeated testing in those with high baseline levels. TKa behaves as a continuous biomarker of risk, and continuous modelling may offer more clinically informative individual risk estimation, while thresholds may retain value for specific clinical contexts. Validation in larger cohorts is warranted to support integration into routine practice.
TKa was measured (DiviTum® assay) at baseline (BL), cycle 1 day 15 (C1D15), and cycle 2 day 1 (C2D1) in patients enrolled in the PDM-MBC study (n = 90). Thresholds for PFS discrimination were identified using maximally selected rank statistics, with predefined cut-offs tested for comparison. Prognostic performance was assessed using the corrected Akaike information criterion (AICc) and Harrell's concordance index (C-index).
TKa was prognostic at all time points. Data-derived thresholds identified groups with differing PFS and overall survival (OS), and predefined cut-offs (≥ 50, ≥ 100, ≥ 250 DiviTum® units of Activity [DuA]) also discriminated survival outcomes. BL and C2D1 models performed better than C1D15 and comparably to multi-time-point models. Among patients with BL TKa ≥ 250 DuA, suppression to < 100 DuA at C1D15 was associated with longer median PFS (23.9 vs. 10.3 months; p = 0.034).
Baseline TKa provides prognostic information, with potential added value from repeated testing in those with high baseline levels. TKa behaves as a continuous biomarker of risk, and continuous modelling may offer more clinically informative individual risk estimation, while thresholds may retain value for specific clinical contexts. Validation in larger cohorts is warranted to support integration into routine practice.
Authors
Brown Brown, Howell Howell, Papantoniou Papantoniou, Eriksson Eriksson, Bergqvist Bergqvist, Williams Williams, Kavanagh Kavanagh, Backlund Backlund, Albu-Kareem Albu-Kareem, Elinder Elinder, Larsson Larsson, Uminska Uminska, Ekholm Ekholm
View on Pubmed