Real estate appraisal is a crucial issue for urban applications, which aims to value the properties on the market. Traditional methods perform appraisal based on the domain knowledge, but suffer from the efforts of hand-crafted design. Recently, several methods have been developed to automatize the valuation process by taking the property trading transaction into account when estimating the property value. However, existing methods only consider the real estate itself, ignoring the relation between the properties. Moreover, naively aggregating the information of neighbors fails to model the relationships between the transactions. To tackle these limitations, we propose a novel Neighbor Relation Graph Learning Framework (ReGram) by incorporating the relation between target transaction and surrounding neighbors with the attention mechanism. To model the influence between communities, we integrate the environmental information and the past price of each transaction from other communities. Moreover, since the target transactions in different regions share some similarities and differences of characteristics, we introduce a dynamic adapter to model the different distributions of the target transactions based on the input-related kernel weights. Extensive experiments on the real-world dataset with various scenarios demonstrate that ReGram robustly outperforms the state-of-the-art methods. Furthermore, comprehensive ablation studies were conducted to examine the effectiveness of each component in ReGram.