Multiplex immunohistochemistry (IHC) enables the simultaneous visualization of multiple biomarkers in formalin-fixed paraffin-embedded (FFPE) tis-sue sections, offering a powerful tool for characterizing immune responses. However, image artifacts such as misalignments and elastic deformations be-tween serial sections limit the accurate co-localization of signals, which is essential for phenotyping and tracking individual cells across biomarkers. Most current segmentation methods in histopathology are tailored to specific staining protocols, limiting their applicability across different histological preparations. In this work, we propose a generalizable, stain-agnostic framework that combines spatial alignment with cell-level tracking to enable robust analysis across consecutively stained tissue sections. The workflow includes initial feature-based alignment using SIFT and SimpleITK, with optional user-guided correction and BSpline deformation when needed. Following alignment, cell segmentation was conducted using a Cell-pose model. A combined score based on embedding similarity and spatial proximity was used to match cells across sections. This two-step strategy allows the transfer of cellular information across slides with different markers, effectively extending the usability of single-stain segmentation models to multi-stain datasets. Cell tracking performance was evaluated on two manually labeled biomarker pairs (Lang, CD1a), achieving up to 100% accuracy and at least 70% in the most challenging case. By integrating a deformable registration step, followed by cell-level matching based on a composite score that integrates spatial proximity and similarity in feature embedding space, this approach enables robust correspondence across sections and aims to mitigate the limitations of traditional pixel-based overlays.