N. A. Borja1, J. Franco2, E. Knudsen2,3, J. Mansour1,3, M. Choti1,3, A. Witkiewicz2,3 1UT Southwestern,Department Of Surgery,Dallas, TX, USA 2UT Southwestern,Department Of Pathology,Dallas, TX, USA 3Simmons Cancer Center,UT Southwestern,Dallas, TX, USA
Introduction: Pancreatic ductal adenocarcinoma (PDA) carries a dismal prognosis, even in patients who undergo successful surgical interventions. Unlike other cancers, an approach tailored to its specific genetic features has not yet been developed. Loss of the CDKN2A tumor suppressor is an exceedingly frequent occurrence in PDA. The CDKN2A gene encodes the p16ink4a protein, which is a potent inhibitor of cyclin-dependent kinases 4 and 6 (CDK4/6). Prior studies have shown that p16ink4a is dominant to KRAS and can serve to constrain oncogenic proliferation in many contexts. Specific CDK4/6 pharmacological inhibitors have been developed and could represent a means to restore the physiological loss of CDKN2A and treat PDA. Here we utilized multiple models to investigate the impact of CDK4/6 inhibition on PDA.
Methods: Three specific model systems were utilized to determine the therapeutic efficacy of CDK4/6 inhibition in the treatment of PDA. First, a panel of established PDA cell lines was employed to delineate overall features of the response to CDK4/6 inhibition, mechanisms of resistance, and define novel combination treatments. Second, in order to recapitulate the complex microenvironment of PDA, we utilized a primary tumor explant model where slices from 13 clinically resected PDA specimens were cultured on semi-solid support. These explants maintained the histologic architecture, biomarker profile, and proliferative index of the primary surgical specimen. Third, a panel of patient-derived xenografts representative of resected disease were developed and utilized to examine the response to CDK4/6 inhibition in vivo.
Results: The established cell lines had variable response to PD-0332991 (CDK4/6 inhibitor) as a single agent; however, in combination with mTOR or MEK inhibitors provided substantial efficacy across all models studied. Surprisingly, this diversity of response was not recapitulated in tumor explants. In this context, treatment with PD-0332991 led to profound suppression of proliferation in all models, with the exception of a single resistant case harboring a loss of the RB tumor suppressor. Similarly, patient-derived xenografts exhibited profound inhibition of Ki-67 proliferation marker and growth suppression. These data indicate that CDK4/6 inhibition is effective in suppressing the growth of PDA under diverse physiological contexts.
Conclusion: Our data demonstrates that PDA growth is profoundly inhibited by selective targeting of the CDK4/6 in primary tumor explants and patient-derived xenografts. The variable response to PD-0332991 in established cell lines likely reflects a more aggressive phenotype, and potentially the altered biology of models propagated in long term culture. Importantly, the data suggest that CDK4/6 inhibition could be particularly effective in the control of resectable disease given the potent suppression of tumorigenic growth in both explant and patient-derived xenograft models.