This study compared the use of physical (Dose Volume, DV) objectives alone and in combination with biological (generalized Equivalent Uniform Dose, gEUD) objectives, along with various NTO settings, to improve the quality of treatment planning. Ten Nasopharyngeal Cancer (NPC) cases treated with VMAT were selected in this study. The original plans that met standard constraints, were generated using DV objectives and automatic NTO. Two additional plan types were created: (i) DV plans combined with a single upper gEUD for spinal cord, brainstem, and parotids; (ii) DV and DV+gEUD plans with manual Normal Tissue Objective (NTO). Dose calculations were performed using Anisotropic Analytical Algorithm (AAA) and Acuros XB Dose-to-medium (AXBDm) algorithm. Plans were evaluated for dosimetry, Monitor Unit (MU), and optimization time, with optimal plans were assessed by using Gamma Passing Rate (GPR). Differences between datasets were analyzed using paired t-test. The gEUD objective with α = 20 reduced Dmax by 10% for spinal cord and by 1.21% for brainstem, while Dmean to parotids decreased by 25-26% for α = 10. However, when α = 10 was applied to parotids in combination with α = 20 for the spinal cord and brainstem, an increase in Dmax for the spinal cord and brainstem was observed. The combination of DV and gEUD objectives improved OAR sparing without compromising PTV coverage, with faster optimization and a modest increase in MU. Manual NTO dose fall-off of 0.5 have enhanced conformity and dose fall-off but slightly reduced uniformity in both DV and DV+gEUD plans. The AXBDm algorithm produced significantly higher gamma passing rates, up to 5% compared to AAA, without in creasing plan complexity. The integration of DV and gEUD objectives improves OAR sparing without compromising PTV coverage or plan quality when appropriate optimization strategies are applied. The selection of NTO settings plays an important role in balancing target dose uniformity and OAR protection. In addition, differences between dose calculation algorithms highlight the importance of algorithm selection, particularly in anatomically heterogeneous regions. Together, careful optimization strategy and algorithm selection contribute to more accurate and clinically robust VMAT planning for NPC cases.