Antiangiogenic therapy, as a new anticancer method, can improve the anticancer effect of traditional therapies. Different antiangiogenic drugs may have different vascular normalization time windows. Whether the antiangiogenic treatment is within the vascular normalization time window is very important in the treatment of cancers. Previous studies have indicated that recombinant human endostatin (rh-ES) can transiently normalize tumor microvessels. Yet the molecular mechanism and the time window of rh-ES remains unclear. The aim of the present study was to explore the optimal time window and molecular mechanism of rh-ES in inhibiting Lewis lung cancer (LLC). By comparatively accessing the changes in microvascular and hypoxic conditions of tumors in host mice treated with rh-ES or saline for different days, the authors aimed to investigate the best administration time of rh-ES treatment on human lung cancers and obtain a better understanding concerning the involved molecular mechanism. A total of 40 C57/BL6 mice with LLC xenografts were randomly divided into normal saline (NS) and rh-ES groups (20 mice/group). 0.2 ml NS or 5 mg/kg rh-ES were administrated via intraperitoneal injection (i.p.) into each mouse each day during the 9-day experiment. A total of 5 mice from each group were sacrificed at day 2, 4, 6 or 9. CA9 and RGS5 expression levels of both groups were compared using immunohistochemistry, reverse transcription-quantitative polymerase chain reaction and ELISA. Rh-ES caused vascular normalization and improved hypoxia at days 4 and 6. Compared with the control (NS) group, both CA9 and RGS5 expression in rh-ES group were significantly decreased at days 4 and 6 (P<0.05), while no significant change between two groups was observed at days 2 and 9. Rh-ES can induce transient tumor vascular normalization and improves tissue hypoxia in LLC tumors. The vascular normalization window is accompanied by the reduction in RGS5 and CA expression.