<p>Traditional Optical Character Recognition (OCR) algorithms often struggle with documents that exhibit complex layouts, as they typically fail to model spatial relationships between textual elements. To address this limitation, we propose a novel architecture, termed LayoutGAT-OCR, which explicitly captures spatial dependencies between textual units using a Graph Attention Network (GAT) within a handwritten text recognition framework. In our approach, document elements (e.g., words or text regions) are represented as nodes in a graph, where each node is associated with visual features and positional encodings. These node representations are iteratively refined through graph attention mechanisms, enabling the model to incorporate contextual information from neighboring nodes and better understand layout-aware relationships. Experimental results demonstrate that LayoutGAT-OCR outperforms strong sequence-based baselines, including PARSeq, achieving a Character Error Rate (CER) of 9.85 on the challenging Qiaopi-HTR dataset, corresponding to a relative error reduction of over 49%. Furthermore, the proposed model generalizes well to the public HWDB2.0 dataset, achieving a CER of 7.91, indicating its effectiveness on more structured handwritten documents. Ablation studies further validate the effectiveness of graph-based spatial modeling, highlighting its critical role in performance improvement. Compared to methods relying on extensive pre-training, our approach offers a more efficient and data-effective solution by explicitly modeling structural relationships between textual regions.</p>