We have already addressed the issue of static point clouds geometry compression (G-PCC) by using Tree-Structured Point-Lattice Vector Quantization (TSPLVQ) [6]. Dynamic 3D point clouds are characterized by up millions of moving 3D positions and color attributes. Efficient point cloud (PC) compression is then fundamental. Temporally successive PC frames are very close, it however remains a challenging problem for coding as the PCs have varying numbers of points without explicit correspondence information. This paper improves and extends the prior work, which provided a new hierarchical geometry representation based on adaptive TSPLVQ. Firstly, a more robust Rate-Distortion optimization process is introduced in order to perform efficient and accurate rate-aware splitting decisions when building and coding the tree structure. Secondly, we focus on the compression of the geometry of dynamic point clouds (G-DPCC) and, the model enables to exploit the temporal dependencies of the 3D content. Exactly, TSPLVQ is a topdown method and permits to represent the PC geometry by using a scalable tree, so when quantizing the dynamic geometry of a given PCs sequence, the successive trees are represented as a trunk common for the 3D sequence, and branches added for each frame. Next, the trunk is first coded, followed by the branches that are differentially coded. Experimental results demonstrate that our method is able to bring significant improvement in terms of the overall compression performance compared to the state-of-the-art MPEG standard.