Bubbles of released mucus were visualized within the oil layer by oblique illumination and electronically imaged

At the moment, there are no open pediatric clinical trials with this agent for pediatric high-grade gliomas and in particular DIPG. Therefore, the rationale for this study was to use our genetically engineered mouse modeling system to determine the efficacy of PD in pediatric BSGs. Here we noted that both of our BSG mouse models overexpress CDK4/6 and all three D-type cyclins. As PD can inhibit CDK4/cyclinD1, CDK4/cyclinD3, and CDK6/cyclinD2 at low nanomolar concentrations, we were intrigued to determine if PD would be effective in our BSG models. Our results SAR131675 demonstrate that PD-0332991 is significantly more efficacious at inhibiting cell growth of PDGF-B; Ink4a-ARF deficient BSG cells through inhibition of pRb. In vitro cell cycle analysis suggests this is due to a cytostatic effect with cells halting in G0/G1, which is consistent with in vivo immunohistochemistry for phospho-H3 and cleaved caspase 3. However, in vitro assays of caspase 3/7 activities indicate a very small but significant increase in apoptosis at 2μM and 5μM, which we attribute to the high sensitivity of the assay. Our data indicate that PD is more efficacious against Ink4a- ARF deficient BSG cells than p53 deficient cells. Importantly, Rb phosphorylation of serine 780, a CDK4/6 phosphorylation site, was inhibited by PD only in the Ink4a-ARF deficient cells but not in the p53 deficient cells. The exact mechanism for the differential response is not clear. Previous studies have shown that hyperphosphorylation of Rb requires both cyclin complexes D1-CDK4/6 and E-CDK2. Both cyclin D1- CDK4/6 and cyclin E-CDK 2 are capable of phosphorylating Rb, however neither is sufficient to hyperphosphorylate and in turn inactivate Rb. The process of Rb hyperphosphorylation first requires partial phosphorylation by cyclin D1-CDK4/6. This then enables phosphorylation by cyclin E-CDK2 and inactivation of Rb. As Ink4a is an endogenous inhibitor of CDK4/6 and because it is absent in PDGF-B; Ink4a-ARF deficient cells, it is likely that these cells do require CDK4/6 as described above to proliferate and thus PD induces cell cycle arrest. On the contrary, PDGF-B; p53 deficient cells already harbor an endogenous CDK4/6 inhibitor, Ink4a, and therefore it is likely these cells have already found a way to circumvent the requirement for CDK4/6 to proliferate, and as a result PD is less effective. In summary, our observations suggest that Ink4a-ARF deficient BSG cells require CDK4/6 to cycle while p53 deficient cells do not. In vivo testing with PD in PDGF-B; Ink4a-ARF-/- tumor bearing mice caused a significant cell cycle arrest after only two doses. Surprisingly, we did not observe a correlation between pRb and cell cycle arrest in vivo. This suggests that cell cycle arrest may be induced by some other mechanism in vivo as CDK4/6 is known to phosphorylate other target proteins besides Rb such as FoxM1, or more likely, a change in pRb levels is best detected sometime between 4 hours and 24 hours. Since short-term treatment with PD provided evidence that PD can successfully reach the tumor in the brainstem and inhibit cell growth we were interested to determine if PD would prolong the survival benefit of PDGF-B driven, Ink4a-ARF deficient BSGs. Our results demonstrate a significant increase in survival of PD-treated versus vehicle-treated mice. Furthermore, the survival benefit with PD was also present when PD was administered in conjunction with RT. Our results demonstrate the first therapeutic regimen that provides an improved survival benefit relative to RT alone in preclinical trials for BSG. As our BSG model is a pediatric model, our results suggest that PD-0332991 should be evaluated in clinical trials for children with Ink4a-ARF deficient gliomas including DIPGs. While Ink4a-ARF deletion at the genomic level is rare in DIPGs, there is evidence that Ink4a protein expression is lost through other mechanisms. DIPGs that lose Ink4a protein expression through other mechanisms may also respond well to PD. In addition, as secondary gliomas in children have a high frequency of Ink4a-ARF loss and PDGFRα amplifications, PD-0332991 may be particularly suitable for this population. Our results are consistent with recent publications in several preclinical glioma models and other tumor types such as malignant rhabdoid tumors noting that Ink4a-ARF loss is a biomarker for response to PD-0332991. Interestingly, potential synergy between PD-0332991 and RT has been noted in adult glioma xenograft studies as well, although in those studies PD-0332991 was either given concurrently with RT or prior to RT. In summary, our results in genetically engineered mouse models of BSG suggest that PD-0332991 may be efficacious in the treatment of pediatric gliomas that have Ink4a-ARF loss.

Latest comments

No comments