Effective treatment of HER2-amplified breast cancer by targeting HER3 and β1 integrin

Abstract

The central role of HER2 as the disease driver and HER3 as its essential partner has made them rational targets for the treatment of HER2-amplifed breast cancers, and there is considerable interest in developing highly effective treatment regimens for this disease that consist of targeted therapies alone. Much of these efforts are focused on dual targeting approaches, particularly dual targeting of the HER2-HER3 tumor driver complex itself, or vertical combinations that target downstream PI3K or Akt in addition to HER2. There is also potential in lateral combinations based on evidence implicating cross-talk with other membrane receptor systems, particularly integrins, and such lateral combinations can potentially involve either HER2 or HER3. We established a preclinical model of targeting HER3 using doxycycline-inducible shRNA and determined the efficacy of a β1 integrin inhibitor in combination with targeting HER3. We report that targeting HER3 and β1 integrin provides a particularly effective combination therapy approach for HER2-amplified cancers, surpassing the combination of HER2 and β1 integrin targeting, and evading some of the safety concerns associated with direct HER2-targeting. This further validates HER3 as a major hub mediating the tumorigenic functions of HER2 and identifies it as a high value target for lateral combination therapy strategies.

Keywords: AIIB2; Combination therapy; HER2; HER3; OS2966; β1 integrin.

Fig. 1

Effect of targeting HER3 or β1 integrin on growth of SKBR3 cells in 3D lrECM culture. a SKBR3 cells were grown in 3D lrECM culture according to the indicated treatment arms for 7 days. Phase contrast microscopy images were taken at 10× magnification. b Apoptotic and proliferative activity of the cells were determined on day 7 using the Tunnel (left panel) and Ki67 (right panel) assays. Results shown are the average of three experiments with the indicated standard error of the mean. Significance values were calculated using the unpaired T test. c SKBR3 cells grown in 3D lrECM culture were harvested on day 2 of treatment and cell lysates used for immunoblotting as indicated. This figure depicts the fusion of image sections from the same film with removal of extraneous lanes from the center

Fig. 2

Effect of targeting HER3 or β1 integrin on in vivo growth of HCC1569 tumors. Nude mice bearing HCC1569TR-shHER3 tumor xenografts were randomly assigned to one of four treatment arms consisting of control (n = 11), oral doxycycline (200 ug/ml) in drinking water (n = 12), AIIB2 administered twice weekly IP (1 mg/kg) (n = 12), or both (n = 12). Treatments were terminated after three weeks and the mice were continuously monitored thereafter in the post-treatment phase. Error bars indicated the standard error of the mean

Fig. 3

Effect of targeting HER3 or β1 integrin on tumor signaling in vivo. Two representative mice from each treatment arm of Fig. 2 were sacrificed and tumor lysates were assayed by Western blotting as indicated. The relative intensities of the HER3 and β1 integrin bands are quantitatively shown below as an average of the two samples in each arm

Fig. 4

Effect of targeting β1 integrin in combination with HER2 or HER3. Nude mice bearing HCC1569TR-shHER3 tumor xenografts were randomly assigned to control (n = 7) or one of five treatment arms consisting of oral doxycycline (200 ug/ml) in drinking water (n = 8), AIIB2 administered twice weekly IP (1 mg/kg) (n = 8), daily lapatinib by oral gavage (50 mg/kg) (n = 8), or the indicated combinations (n = 8, n = 8). Treatments were terminated after 12 days and the mice were continuously monitored thereafter in the post-treatment phase. Error bars indicated the standard error of the mean. The increased efficacy of HER3 and β1 integrin targeting compared with HER2 and β1 integrin targeting is highly significant (AIIB2 + dox vs AIIB2 + lapatinib; comparison at day 24, two-tailed t test p < 0.0001)