STA-8666, a novel HSP90 inhibitor/SN-38 drug conjugate, causes complete tumor regression in preclinical mouse models of pediatric sarcoma

Abstract

Long-term survival in patients with metastatic, relapsed, or recurrent Ewing sarcoma and rhabdomyosarcoma is dismal. Irinotecan, a topoisomerase 1 inhibitor, has activity in these sarcomas, but due to poor bioavailability of its active metabolite (SN-38) has had limited clinical efficacy. In this study we have evaluated the efficacy and toxicity of STA-8666, a novel drug conjugate which uses an HSP90 inhibitor to facilitate intracellular, tumor-targeted delivery of the topoisomerase 1 inhibitor SN-38, thus preferentially delivering and concentrating SN-38 within tumor tissue. We present in vivo evidence from mouse xenograft models that STA-8666 results in more persistent inhibition of topoisomerase 1 and prolonged DNA damage compared to irinotecan. This translates into superior antitumor efficacy and survival in multiple aggressive models of both diseases in mouse xenografts, as well as in an irinotecan-resistant model of pediatric osteosarcoma, demonstrated by dramatic tumor shrinkage, durable remission and prolonged complete regressions following short-term treatment, compared to conventional irinotecan. Gene expression analysis performed on xenograft tumors treated with either irinotecan or STA-8666 showed that STA-8666 affected expression of DNA damage and repair genes more robustly than irinotecan. These results suggest that STA-8666 may be a promising new agent for patients with pediatric-type sarcoma.

Keywords: SN-38; drug conjugate; pediatric; sarcoma; topoisomerase 1 inhibitor.

Figure 1. ES and RMS tumor volumes in pilot experiments with STA-8666

Tumor volumes of RH30 (RMS) (A) and TC32 (ES) (B) in mice treated with vehicle (blue), ganetespib at 50 mg/kg IV (red) or STA-8666 at 150 mg/kg IV (orange). Arrows indicate weekly treatments. Volume differences were statistically significant between the ganetespib and STA-8666 groups for days 15 to 43 in the RMS cohort (p ≤ 0.0002), and for days 15 to 22 in the ES cohort (p ≤ 0.0012).

Figure 2. ES and RMS tumor volumes and Kaplan-Meier curves for mice treated with vehicle, ganetespib, high-dose irinotecan, or STA-8666 at varying doses

Upper panel: Tumor volumes of RH30 (RMS) (A) and TC32 (ES) (B) in mice treated weekly with vehicle (dark blue), ganetespib at 150 mg/kg IV (red), irinotecan at 50 mg/kg IP (purple), STA-8666 at 50 mg/kg IV (orange), 100 mg/kg IV (light blue) and 150 mg/kg IV (gray). Arrows indicate weekly treatments for each experimental condition. Volume differences were statistically significant between the irinotecan group and the STA-8666 50 mg/kg group (p ≤ 0.007), the STA-8666 100 mg/kg group (p ≤ 0.0002), and the STA-8666 150 mg/kg group (p ≤ 0.0003) for days 64 through 85 in the RMS experiment. In the ES experiment, volume differences were statistically significant between the irinotecan group and the STA-8666 groups for all doses (p ≤ 0.0002) between days 36 and 54. Lower panel: Kaplan-Meier analysis of overall survival for mice with RH30 (C) and TC32 (D) tumors in the same experiment. Survival comparisons with irinotecan reached statistical significance in all groups (in RMS, p = 0.0004 for irinotecan v. STA-8666 at 50 mg/kg, p < 0.0001 for irinotecan v. STA-8666 at 100 mg/kg and irinotecan v. STA-8666 150 mg/kg; in ES, p < 0.0001 for irinotecan v. STA-8666 at all three doses). Dose responses on survival in both RMS (p = 0.0003) and ES (p < 0.0001) were statistically significant (Figure 2C and 2D).

Figure 3. ES tumor volumes and Kaplan-Meier curves for mice treated with vehicle, protracted-dose irinotecan, or STA-8666

(A) Tumor volumes of EW8 xenografts in mice treated with vehicle (blue), irinotecan at 10 mg/kg IP daily for 5 days per week (purple), and STA-8666 at 150 mg/kg IV once per week (orange). The beginning of each treatment week is indicated by the arrows. Treatments in all groups lasted for two weeks. Volume differences were statistically significant between the irinotecan group and STA-8666 group (p ≤ 0.0379) for day 107 and beyond. (B) Kaplan-Meier analysis of overall survival for mice with EW8 tumors in the same experiment. Overall survival was significantly superior for the mice receiving STA-8666 (p = 0.0023).

Figure 4. PDX tumor volumes and Kaplan-Meier curves for mice treated with vehicle, protracted-dose irinotecan, STA-8666, or ganetespib plus protracted-dose irinotecan

Upper panel: Tumor volumes of ES PDX (A) and RMS PDX (B) in mice treated with vehicle (blue), irinotecan at 7.5 mg/kg IP daily for 5 days per week (purple), STA-8666 at 100 mg/kg IV once per week (orange), and irinotecan at the above schedule plus ganetespib at 100 mg/kg IV once per week (red). The beginning of each treatment week is indicated by the arrows. Treatments in all groups lasted for two weeks. For the EWS PDX, volume differences were statistically significant between the irinotecan group and STA-8666 group (p ≤ 0.0043) for days 87 through 146 and between the STA-8666 group and the combination group (p ≤ 0.0087) for days 94 through 143. P-values > 0.05 were obtained for comparisons of the irinotecan and combination groups during this same time period. For the RMS PDX, p-values > 0.05 were obtained for comparisons between each of the groups from days 118 through 167. Lower panel: Kaplan-Meier analysis of overall survival for mice with ES PDX (C) and RMS PDX (D) tumors in the same experiment. At end of experiment, the remainder of living mice had no evidence of tumor. Differences in overall survival between the STA-8666 group and other groups in ES were statistically significant (p = 0.0005 for irinotecan v. STA-8666; p = 0.0008 for STA-8666 v. combination), but no significance was seen for RMS (p = 0.3675 for irinotecan v. STA-8666; p = 0.1922 for STA-8666 v. combination).

Figure 5. Osteosarcoma (OS) tumor volumes for mice treated with vehicle (blue), weekly irinotecan at 50 mg/kg IP (purple), and weekly STA-8666 at 150 mg/kg IV (orange)

Arrows indicate weekly treatments for each experimental condition. Tumor volume differences approached but did not reach statistical significance (p = 0.0571), likely due to the smaller number of mice used in this experiment.

Figure 6. γH2AX expression in TC32 tumor samples at serial time points

(A) Western blot showing that γH2AX is detectable longer in tumors from mice that received STA-8666, compared to irinotecan. Mice bearing TC32 tumors with an average size of 800 mm³ were treated with a single dose of either vehicle, irinotecan at 50 mg/kg IP or STA-8666 at 150 mg/kg IV. One mouse per group was sacrificed on day 3 and day 7. (B) Quantification of Western blot data normalized to actin. Signal in day 3 vehicle tumor was set to 1, as no vehicle mice remained alive on day 7.

Figure 7. Fold-change comparison of gene expression in tumors five days after a single treatment with STA-8666 or irinotecan

(A) Comparison of top 10 upregulated genes by fold-change versus vehicle in TC32 xenograft tumors treated with a single dose of either STA-8666 (100 mg/kg) or irinotecan (50 mg/kg) five days before tumor harvest. (B) Comparison of top 10 downregulated genes by fold-change versus vehicle in TC32 xenograft tumors treated with a single dose of either STA-8666 (100 mg/kg) or irinotecan (50 mg/kg) five days before tumor harvest.