Invasive breast cancer (IBC) is the most commonly diagnosed malignancy and the second leading cause of cancer-related mortality in women worldwide. Although standardized multimodal therapies have significantly improved disease-free survival (DFS) and overall survival (OS) in breast cancer patients, substantial interindividual variability in treatment response persists, driven by tumor biological heterogeneity and complex clinical factors (1). This heterogeneity underscores the critical need for robust prognostic biomarkers to enable precise risk stratification and personalized therapeutic decision-making.

Current clinical practice relies on established prognostic indicators, including hormone receptor status, human epidermal growth factor receptor 2 (HER2) expression, and Ki-67 proliferation index. The integration of comprehensive genomic profiling has further enhanced risk assessment and therapeutic selection, particularly for early-stage disease. However, despite these advancements, many proposed biomarkers exhibit inconsistent clinical utility and have not been widely adopted into routine clinical practice (2, 3).

A major clinical challenge remains the significant proportion of patients who experience disease recurrence or metastasis despite receiving optimal standard therapy, highlighting the urgent need for more accurate prognostic tools. The discovery of novel, reliable biomarkers could substantially improve risk prediction and facilitate the development of tailored treatment strategies.

In humans, thymidine kinase exists as two distinct isoenzymes: cytosolic thymidine kinase 1 (TK1) and mitochondrial thymidine kinase 2 (TK2) (4). As a cell cycle-regulated proliferation marker, TK1 is markedly upregulated during the S-phase to facilitate DNA synthesis. It plays a pivotal role in the pyrimidine salvage pathway—one of the two major routes for DNA precursor biosynthesis (alongside de novo synthesis)—thereby contributing to DNA replication and repair (5). Under normal physiological conditions, serum TK1 activity in healthy individuals is negligible or undetectable. By contrast, it is substantially elevated in malignant tumors, with strong correlations to tumor proliferation, progression, and metastatic potential (6). These findings have positioned TK1 as a promising biomarker for prognostic assessment and therapeutic decision-making in oncology.

While the utility of serial TK1 monitoring for treatment response and prognosis in breast cancer has been established, significant discrepancies persist regarding the diagnostic, prognostic, and predictive value of baseline TK1 levels (7–9). The correlation between baseline TK1 levels and distinct prognostic outcomes in breast cancer patients, as well as its potential utility in guiding diagnosis and treatment, remains controversial due to inconsistent conclusions across studies. To address this evidence gap, we conducted the first meta-analysis to systematically assess the relationship between baseline TK1 levels and prognosis in breast cancer.

This meta-analysis evaluated the prognostic significance of baseline TK1 expression in breast cancer, with strict inclusion criteria requiring TK1 assessment prior to treatment initiation. Pooled analysis of 11 high-quality studies (NOS scores 8–9) encompassing diverse clinical subtypes (early/advanced stage, various receptor statuses, and operable/inoperable cases) revealed significantly worse progression-free survival in patients with elevated baseline TK1 (HR = 1.63, 95% CI: 1.28–2.10, p < 0.001; random-effects model, I² = 87.9%).

The significant heterogeneity observed may originate from several potential sources. First, clinical heterogeneity across patient populations likely contributed substantially. The included studies involved a broad spectrum of breast cancer patients, ranging from those with early-stage operable disease (9, 11, 18) to those with advanced metastatic disease (13–17, 19), as well as patients receiving neoadjuvant therapy (9, 20). Notable differences exist between early- and advanced-stage patients regarding tumor burden and the tumor microenvironment. For example, TK1 may more directly reflect high proliferative activity and disease burden in advanced-stage patients, whereas in early-stage patients, it could be associated with micrometastatic potential. Although a subgroup analysis based on treatment-naive versus recurrent disease was performed, this stratification did not fully account for the observed heterogeneity(Adj R-squared = 34.03%, p = 0.175), suggesting that additional clinicopathological factors—such as hormone receptor status, HER-2 expression, prior lines of therapy, and metastatic sites—may also influence the results.

Second, methodological variations in detection assays and specimen types represent another key source of heterogeneity. As summarized in Table 1, most studies (nine of 11) measured serum TK1 levels using ELISA. However, considerable differences existed among studies in the commercial kits used and in the definitions of positive thresholds (e.g., median, upper quartile, or prespecified cutoff values). Moreover, one study employed immunohistochemistry (IHC) to assess TK1 protein expression in tumor tissue (18), whereas an earlier study utilized the BCA method (11). Importantly, IHC captures in situ TK1 expression within tumor tissue, whereas ELISA quantifies circulating TK1 in peripheral blood. These two approaches reflect biologically distinct dimensions of TK1, differing in quantitative units, analytical sensitivity, and clinical interpretation—differences that inevitably introduce substantial heterogeneity across studies. Due to the limited overall number of available studies, performing univariate subgroup analyses for heterogeneity factors beyond treatment−naive/recurrent status would further reduce the number of studies in each subgroup, thereby precluding a meaningful examination or reduction of heterogeneity through subgroup comparisons. For these reasons, a random−effects model was ultimately employed to conduct the meta−analysis of PFS.

Analysis of six studies reporting overall survival demonstrated consistent prognostic value across subgroups (treatment-naive: HR = 1.30; recurrence: HR = 2.10; pooled HR = 1.58, 95% CI: 1.40–1.78, p = 0.003), with heterogeneity resolved through stratification by disease status. Based on the observed disparity in HR between the two subgroups, the prognostic significance of TK1 appears to be disease stage-dependent. These findings confirm that baseline TK1 expression is consistently associated with adverse outcomes in breast cancer patients, establishing it as a reliable prognostic indicator.

As a key enzyme in the DNA salvage pathway, TK1 demonstrates peak expression during the S-phase of the cell cycle, serving as a precise indicator of tumor proliferative activity. Its biological properties have been well characterized. Current research on TK1 spans diverse solid and hematological malignancies (21–24), with the most clinically significant positive findings concentrated in breast cancer (25). Accumulating evidence has established TK1 as a key mediator of breast cancer progression. Recent mechanistic studies have further defined its specific function as a downstream effector of the cyclin-dependent kinase 4/6 (CDK4/6) pathway. In patients with hormone receptor-positive/human epidermal growth factor receptor 2-negative (HR+/HER2−) advanced breast cancer, an early elevation in TK1 activity following CDK4/6 inhibitor administration is consistently correlated with decreased endocrine therapy sensitivity or the development of acquired drug resistance (26, 27). Our findings substantiate the clinical utility of baseline TK1 levels in prognostic assessment for breast cancer. Importantly, baseline TK1 maintained stable prognostic performance across diverse clinical subtypes, suggesting its potential to overcome the limitations of conventional pathological classification and provide more broadly applicable clinical guidance.

Nevertheless, this study has several limitations that should be acknowledged. The primary limitation stems from substantial heterogeneity in the progression-free survival analysis (I² = 87.9%), which persisted despite employing disease status stratification and other statistical adjustments. Interpretation of PFS outcomes from this meta-analysis requires caution, as the high heterogeneity observed suggests that the prognostic value of TK1 may be influenced by variations in patient characteristics and detection methodologies. To consolidate the clinical utility of TK1, subsequent investigations are recommended to establish standardized inclusion criteria and uniform TK1 detection protocols, coupled with rigorous prospective validation studies, thereby yielding robust evidence to advance the development and optimization of TK1-driven therapeutic strategies. Second, our analysis may be susceptible to publication bias, in which studies demonstrating a significant association between TK1 expression and poor prognosis are more likely to be published, potentially leading to an overestimation of its true prognostic value. Although an exhaustive literature search was conducted, this limitation cannot be entirely excluded.

Building on our findings, we propose three critical directions for future investigation: (1) establishing standardized TK1 detection protocols to enhance interlaboratory reproducibility and clinical applicability; (2) conducting mechanistic studies to elucidate TK1’s prognostic role in breast cancer, with particular emphasis on its crosstalk with core signaling pathways to inform targeted therapy development; and (3) performing biomarker-driven clinical trials to evaluate TK1-guided therapeutic strategies, especially in patients exhibiting elevated TK1 expression.

An optimal prognostic biomarker should fulfill a tripartite role: outcome prediction, treatment response monitoring, and guidance for therapeutic decision-making. Our meta-analysis demonstrates a significant association between elevated baseline TK1 expression and adverse prognostic outcomes in breast cancer. Although high baseline TK1 expression indicates poor prognosis, whether adjusting treatment strategies in advance or implementing more active interventions can improve clinical benefits for these patients under the current treatment system requires further support from high-quality clinical research data.