Glial cell line-derived neurotrophic factor (GDNF) family of ligands (GDFLs) as well as other trophic factors have, in animal models of Parkinsons disease (PD), demonstrated the potential for excellent ameliorative properties. Clinical trials that have mechanically injected GDNF intracerebrally, while demonstrating relative safety, have been clinically disappointing to date. Likewise, recombinant adeno-associated virus (rAAV) delivered neurturin (cere-120) has also been demonstrated to be safe in humans, however clinical results have been negative. The failure of the major clinical trials has cast some doubt in the field about trophic factor delivery for the treatment of PD. In this review, we make the case that GDFLs are likely to function only when there are remaining dopamine neurons in the nigrostriatal pathway as opposed to other candidate modes of action. Thus, it is our view that utilizing earlier stage PD patients who have significant nigrostriatal dopamine innervation remaining would be more ideal to demonstrate the efficacy of GDFLs. This is particularly true when considering a novel delivery method such as gene transfer. However, if earlier stage patients are to be enrolled in GDFL gene transfer trials, then a much better safety profile must be demonstrated by preclinical experiments. One important safety advance might be the use of an external regulation system to control the expression level of the transgene. However, gene regulation systems pose unique safety issues and we will discuss these in detail. It is our view that GDFLs still remain as a promising therapeutic approach for PD.