Changes of brain phosphodiesterase type IV measured with ¹¹C-(R)-rolipram PET in major depressive disorder

Objectives Phosphodiesterase type 4 (PDE4) selectively metabolizes cAMP in the brain. Basic studies have indicated that PDE4 and the cAMP cascade are downregulated in major depressive disorder (MDD) and that cAMP cascade mediates the effects of several antidepressants. The objectives of this study were to compare the binding of ¹¹C-(R)-rolipram, a PDE4 inhibitor, between unmedicated MDD patients and healthy controls and also to compare rolipram binding in MDD before and after selective serotonin reuptake inhibitor (SSRI) therapy.

Methods ¹¹ C-(R)-Rolipram PET scans were performed in 28 unmedicated patients and 25 healthy controls. Thirteen patients repeated a scan 4 - 8 weeks after SSRI treatment. Six controls had two scans to study reproducibility. Total distribution volume normalized to plasma free fraction, V_T/f_P, was measured in 10 regions throughout the brain using unconstrained two-comp. model. To study the involvement of P-gp, brain uptake of ¹¹C-(R)-rolipram was compared between wild type and P-gp KO mice.

Results Unmedicated patients showed lower V_T/f_P than controls in all regions with an average decrease of 22% (w/o partial volume correction (PVC): p=0.002, with PVC: p=0.007). Although six healthy subjects showed similar V_T/f_P in two scans with a difference of -7.7±9.2% ((2nd - 1st scan)/1st scan), 13 patients showed markedly larger variability of +15±37%. There was no correlation between baseline depression or anxiety ratings or the changes after SSRI and either baseline V_T/f_P or its changes after SSRI. The mouse scans indicated that (R)-rolipram is not a substrate of P-gp.

Conclusions Significantly lower rolipram binding in unmedicated depressives is in line with the downregulation of the cAMP cascade indicating that stimulation of the cAMP cascade may induce antidepressant effects. The large variability of rolipram binding between the two scans in the patients may indicate instability of the cAMP cascade.

Research Support NIM