Curis Hedgehog and Beyond --plus multi-targeting cancer inhibitors

RO4929097
Importantly, efficacy is maintained after dosing is terminated
In addition, comparative microarray analysis suggests tumor cell differentiation as an additional mode of action.

Presentation Title: Targeting Notch in malignant brain tumors: Crosstalk with Hedgehog as a potential mechanism of treatment resistance
Presentation Time: Sunday, Apr 18, 2010, 2:25 PM - 2:40 PM
Location: Room 144, Washington Convention Center
Author Block: Karisa C. Schreck1, Pete Taylor2, Eli E. Bar1, Vidya Gopalakrishnan2, Nicholas Gaiano1, Charles G. Eberhart1. 1Johns Hopkins School of Medicine, Baltimore, MD; 2University of Texas, M.D. Anderson Cancer Center, Houston, TX
Abstract Body: Developmental signaling cascades such as Notch, Hedgehog, and Wnt are increasingly being investigated as novel therapeutic targets in malignant brain tumors. These pathways are thought to play particularly important roles in stem-like cancer cells, which seem resistant to standard chemo- and radiation therapies. Indeed, inhibiting Notch signaling in medulloblastoma or glioblastoma (GBM) cultures decreases the proportion of stem-like cells. However, we thought that other developmental stem cell maintenance pathways might compensate for the pharmacological inhibition of Notch signaling, and therefore investigated Hedgehog and Wnt. Interestingly, we found that levels of the Hedgehog targets Gli1 and Patched1B increased almost two-fold in response to pharmacological inhibition of Notch in neurosphere lines derived from malignant gliomas. The Wnt target Axin2 was also upregulated in some lines, but in a less reproducible fashion. We hypothesized that there might be direct transcriptional regulation of Hedgehog targets by the Notch pathway, and used chromatin immunoprecipitation to evaluate whether the repressive factor Hes1 bound to the first intron of Gli1. We found that Hes1 does bind to multiple N-boxes in this intron, consistent with the concept that it acts as a transcriptional repressor for Gli1. Together, our data suggest a negative feedback loop between Notch and Hedgehog by which GBMs could escape from pharmacological Notch inhibition by up-regulating Hedgehog signaling.
We also investigated whether targeting both Notch and Hedgehog signaling simultaneously would kill brain tumor cells more effectively than as monotherapies. As we hypothesized, while pharmacological inhibition of Notch (using gamma-secretase inhibitor) or Hedgehog (using cyclopamine) signaling alone inhibited the growth and clonogenicity of GBM neurosphere cell lines in a dose-dependent fashion, co-treatment resulted in significantly more robust inhibition of cell growth and colony-forming ability. Moreover, when neurosphere lines capable of forming intracranial tumors were pretreated with one or both pathway inhibitors before injection into mice, co-treatment significantly prolonged survival. This is consistent with our model that inhibiting both pathways more effectively kills the xenograft-initiating cells within the culture. We next investigated whether these findings applied to primary human GBMs and found that freshly cultured neurospheres were highly susceptible to co-treatment but more refractory to monotherapy. Our findings suggest that targeting multiple developmental pathways may be necessary to fully ablate the stem-like population in malignant brain tumors.
LINK