Background/Aims: Anaplastic thyroid cancer (ATC), with 25% BRAFV600E mutation, is one of the most lethal human malignancies that currently has no effective therapy. Vemurafenib, a BRAFV600E inhibitor, has shown promise in clinical trials, including ATC patients, but is being hampered by the acquisition of drug resistance. Therefore, combination therapy that includes BRAFV600E inhibition and avoids resistance is a clinical need. Methods: ATC cell lines 8505C (BRAFV600E/mt), SW1736 (BRAFV600E/mt), KAT18 (BRAFV600E/wt) and Cal-62(BRAFV600E/wt) cells were used in the study. The ability of S100A knockout or /and in combination with the BRAF inhibitor vemurafenib on growth, apoptosis, invasion and apoptosis in ATC cells in vitro was demonstrated by MTT and BrdUrd incorporation assay, Annexin-V-FITC staining analyzed by flow cytometry, Transwell migration and Matrigel invasion assay. S100A4,pERK1/2, pAKT and pROCK1/2 protein was detected by western blot assay; Small molecule inhibitors of Y27632, U0126, MK-2206 and constitutively active forms of pCDNA-Myc-pERK, pCMV6-HA-Akt, pCMV-RhoA were employed, and the mechanistic studies were performed. We assessed the efficiency of in vivo combination treatment with S100A4 knockout and Vemurafenib on tumors. Results: S100A4 knockout induced apoptosis and reduced proliferation by inactivation of pAKT and pERK signals, and inhibited invasion and migration by inactivation of pAKT and RhoA/ROCK1/2 signals in 8505C or Cal-62 cells in vitro, and vice versa in SW1736 and KAT18 cells. Vemurafenib did not affect apoptosis of both 8505C and SW1736 cells, but reduced proliferation via arresting cell cycle, and promoted cell migration and invasion in vitro. Combination treatment with S100A4 knockdown and vemurafenib reduced cell proliferation, migration and invasion in vitro compared to the S100A4 knockdown or Vemurafenib alone. Vemurafenib treatment resulted in a transient inhibition of pERK expression and gradually activation of pAKT expression, but quickly recovery from ERK1/2 activation inhibition by vemurafenib treatment in 4 h for SW1736 and 8505C cells. Combined treatment completely inhibited ERK1/2 and AKT activation during 48 h. In an in vivo mouse model of SW1736 and 8505C, vemurafenib treatment alone did not significantly inhibit tumor growth in both of the tumors, but inhibited tumor growth in combined groups. Conclusion: Our results show S100A4 knockout alone inhibits ATC cells (rich endogenous S100A4) survival and invasion, regardless of the BRAFV600E status, and potentiates the effect of vemurafenib on tumor regression in vitro and in vivo. In addition, S100A4 knockout potently inhibits the recovery from ERK1/2 activation inhibition and the AKT activation following vemurafenib treatment and reversed the vemurafenib resistance. This therapeutic combination may be of benefit in patients with ATC.