Background

Tumor-derived components, such as dual biomarkers including circulating tumor DNA (ctDNA) and circulating tumor cells (CTCs), provide comprehensive real-time insights into the tumor microenvironment. Minimal cellular residual disease (MCRD), dynamic cell surface protein overexpression, residual tumor burden after curative-intent resection, and decisions regarding adjunct therapy or therapy de-escalation are critical aspects of patient management. Similar to the PD-L1 combined positive score (CPS) in tissue samples, PD-L1 expression on CTCs represents a promising dynamic biomarker, particularly in the context of epithelial-to-mesenchymal transition (EMT). EMT may contribute to immune evasion by deactivating T cells, thereby facilitating micro-metastatic progression, which remains challenging to detect and manage. In this study, we report PD-L1 expression as a dynamic biomarker on circulating tumor cells across early- to late-stage breast cancer patients.

Methods

In this retrospective analysis, peripheral blood samples from 1,294 breast cancer patients were evaluated for the presence of CTCs, PD-L1 expression on CTCs, and CTC clusters. CTCs were enumerated using the OncoDiscover platform, approved by CDSCO-India, from 1.5 mL of peripheral blood. The system consists of a multi-component magneto-nanosystem mediated by anti-epithelial cell adhesion molecule (EpCAM) antibodies. CTCs were identified based on EpCAM⁺, CK18⁺, DAPI⁺, and CD45⁻ markers. PD-L1 expression on CTCs was quantified using linear intensity gradients of fluorescence signals acquired through an automated Zeiss microscope. Additionally, a computational model was developed to evaluate mean CTC distribution, perform regression analysis, and assess CTC predictability.

Results

CTC counts ranged from 1 to 20 per 1.5 mL of blood. At baseline analysis, 73.20% (n = 978) of patients had ≥1 CTC. Among patients with detectable CTCs, 87.69% (n = 406 of 463) exhibited PD-L1 expression. The highest proportion of total CTCs (~21.61%, n = 1,125) was observed in the 41–50-year age group. The highest frequency of CTC clusters (~29.08%, n = 41) and PD-L1–positive CTCs (~42.91%, n = 324) was observed in the 51–60-year age group. CTC clusters were detected in 2.71% (n = 141) of total patients. The mean CTC count (including clusters) was 3.90, while the mean PD-L1–positive CTC count was 3.40. The computational model demonstrated a correlation between blood-based outcomes and normal probability scores.

Conclusions

The observed inter-patient heterogeneity suggests potential biological and pharmacodynamic relevance of both CTCs and PD-L1 expression. Larger clinical studies are warranted to further evaluate PD-L1 expression on CTCs, particularly in early-stage cancers. Patients with minimal cellular residual disease, despite the absence of radiographic evidence, may represent a higher-risk group for metastasis progression and may benefit from enhanced stratification strategies.