No abstract

Background: The European Medicines Agency (EMA) aims to resolve uncertainties associated with conditionally approved drugs by imposing post-approval studies. Results from these studies may be relevant for health technology assessment (HTA) organizations. This study investigated the role of regulator-imposed post-approval studies within HTA. Methods: For all conditionally approved drugs up to December 2018, regulator-imposed post-approval studies were identified from EMA’s public assessment reports. The availability for and inclusion of study results in relative effectiveness (re)assessments were analyzed for 4 European HTA organizations: NICE (National Institute for Health and Care Excellence, England/Wales), HAS (Haute Autorité de Santé, France), ZIN (Zorginstituut Nederland, the Netherlands) and the European Network for Health Technology Assessment (EUnetHTA, Europe). When study results became available between an HTA organization’s initial assessment and reassessment, it was evaluated whether and how they affected the assessment and its outcome. Results: For 36 conditionally approved drugs, 98 post-approval studies were imposed. In total, 81 initial relative effectiveness assessments (REAs) and 13 reassessments were available, with numbers of drugs (re)assessed varying greatly between jurisdictions. Study results were available for 16 initial REAs (20%) and included in 14 (88%), and available for 10 reassessments (77%) and included in all (100%). Five reassessments had an outcome different from the initial REA, with 4 (2 positive and 2 negative changes) relating directly to the new study results. Reassessments often cited the inability of post-approval studies to resolve the concerns reported in the initial REA. Conclusion: Results from regulator-imposed post-approval studies for conditionally approved drugs were not often used in REAs by HTA organizations, because they were often not yet available at the time of initial assessment and because reassessments were scarce. When available, results from post-approval studies were almost always used within HTA, and they have led to changes in conclusions about drugs’ relative effectiveness. Post-approval studies can be relevant within HTA but the current lack of alignment between regulators and HTA organizations limits their potential.

Neuroblastomas account for approximately 15% of all childhood cancer deaths. Clinical complete remission is achieved in many stage 4 neuroblastoma patients, but the high risk of relapse and the accompanying treatment-resistant nature of these tumors is still a challenge. We have previously identified higher frequencies of mutations that affect both the p53 and the pRb pathway, such as the homozygous loss of CDKN2A or co-amplification of MDM2 and CDK4. In addition, both cyclin D1 and MDM2 overexpression is a common characteristic of primary neuroblastoma. These genes act upstream of pRb and p53 and hence play a major role in cell cycle regulation. To study whether these G1 checkpoint aberrations render cells more sensitive to CDK4 or MDM2 inhibition, we exposed a series of 11 cell lines to the most promising CDK4 (ribociclib, palbociclib, and abemaciclib) and MDM2 (SAR405838, HDM-201, and idasanutlin) inhibitors. These cell lines have well-defined mutational properties: homozygous CDKN2A deletion, co-amplification of CDK4 and MDM2, TP53 mutation, and MYCN amplification. Whereas sensitivity for the MDM2 inhibitors inversely correlated with TP53 mutation status, it did not correlate to other G1 checkpoint aberrations. Similarly, there was no clear correlation seen for the CDK4 inhibitors. To further explore this, we generated cell lines with inducible overexpression of CDK4, MDM2, or both. The overexpression did indeed not increase sensitivity to CDK4 or MDM2 inhibitors. Next, we were interested whether combined treatment with the most potent CDK4 and MDM2 inhibitors, abemaciclib and idasanutlin respectively, would be of added value. Cells were exposed to 10 different concentrations of both compounds and Bliss Independence values were calculated as a measure of synergy. Interestingly, combined treatment resulted in slight antagonism in the range of clinically relevant doses in most cell lines. This adverse effect was supported by cell cycle analyses, which showed a lower apoptotic fraction, as well as PARP and caspase 3 cleavage, after combined treatment, as opposed to MDM2 inhibition alone. As was previously suggested in the first clinical trials with CDK4 inhibitors, neither CDK4 nor CDKN2A status is a clear biomarker for CDK4 inhibitor sensitivity. Our results suggest that MDM2 and CDKN2A status also fail as biomarkers for MDM2 inhibitor sensitivity. Further testing is necessary to identify (other) biomarkers for these inhibitors. Moreover, the logical rationale to combine CDK4 and MDM2 inhibitors in neuroblastoma patients with both pRb and p53 pathway disturbances should be taken with precaution, as first results do not show a beneficial response. In vivo experiments to confirm this finding are currently ongoing. Citation Format: Nil Schubert, Linda Schild, Stijn van Oirschot, Kaylee Keller, Lindy Alles, Lindy Vernooij, Marloes Nulle, Emmy Dolman, Marlinde van den Boogaard, Jan Molenaar. Combined targeting of the p53 and pRb pathway in neuroblastoma [abstract]. In: Proceedings of the AACR Special Conference on the Advances in Pediatric Cancer Research; 2019 Sep 17-20; Montreal, QC, Canada. Philadelphia (PA): AACR; Cancer Res 2020;80(14 Suppl):Abstract nr A49.