Phenoxodiol, (Px) is a new oral flavonoid, structurally related to genistein, with significant antiproliferative activity in in vitro and in vivo cancer models, currently in use in early Clinical Trials. The exact mechanism for its antiproliferative effect is still unclear. To evaluate the antiproliferative effect of Px in Head and Neck (HN) Cancer, we tested Px against a battery of squamous (HN12, HN4, HN6 & HN30) and salivary gland (HSY, NCIH292, NCIH3118 & A253) cancer cell lines. Dose response studies using Px for 24 hr demonstrated an IC50, as determined with thymidine incorporation, for all HN cell lines tested of ≤10 mg/ml. Although there was evidence of apoptosis in some cell lines such as HN12, this effect occur only after longer exposure with the drug ( ≥ 48 hr exposure) Instead, a clear G1/S arrest occurs in most HN tested so far. To understand the mechanism of the cell cycle arrest by Px we focused on the HN12 (p53 mutant) carcinoma cell line. As soon as 12 hr after exposure with ≥ 5mg/ml Px, clear G1 arrest was observed (G1%: control: 30 vs Px: 79). In contrast, Genistein promoted an accumulation of cells in G2/M, as reported previously in other models. Similar G1/S arrest occurs when cells were synchronized with aphidocolin and then released into Px. To understand how Px modulates cell cycle, we initially tested whether the activity of G1 cdk’s (cdk6, cdk2 and cdk4) was modulated using cdk purified systems. Of interest no effects were observed up to 10mg/ml Px. Then, we tested the effects of HN12 treated with Px; as soon as 6 hr after exposure of ≥ 5mg/m there was clear loss in cdk2 activity while the effects of PX in cdk4 and/or cdk6 activity remained unchanged. Moreover, immunoblots against cdk4, cdk6 and cdk2 were unaltered, suggesting that the loss in cdk2 activity is not due to a direct effect on its activity or to loss in cdk2 mass. Time-course and dose response studies in HN12 demonstrated that cyclin A or E was not altered. In contrast, upregulation of p21waf1/cip1 was present as soon as 3 hr after Px while p27kip1 was not induced. In summary, Phenoxodiol has a dose-dependent antiproliferative effects in several HN squamous and salivary gland tumors. This antitumor effect was associated with a clear p53-independent G1/S arrest; and is due to an indirect loss in cdk 2 activity by upregulation of p21waf1/cip1. Exact mechanism(s) by which Phenoxodiol modulates p21 waf1/cip1 and evaluation for anti-tumor efficacy in HN cancer models are currently under investigation.

[Proc Amer Assoc Cancer Res, Volume 45, 2004]