LCIG infusion, also known as carbidopa/levodopa enteral suspension and by the trademark names of Duopa™ and Duodopa™, is a suspension gel containing carbidopa–levodopa in a standard 1:4 ratio that aims to deliver the drug directly to the proximal small intestine (5). The drug is delivered via a portable pump that is connected to a percutaneous endoscopic transgastric jejunostomy (PEG-J) (6). Pharmacokinetic studies have demonstrated that LCIG results in lower variability and fluctuations in levodopa plasma concentrations compared to oral levodopa (7). Several studies have evaluated the safety and efficacy of LCIG and are described below.

Apomorphine is a dopamine agonist with a mixed affinity for D1 and D2 receptors and an affinity for serotonergic and alpha-adrenergic receptors (2). While the motor efficacy of apomorphine is similar to that of levodopa, apomorphine cannot be administered orally due to its low oral bioavailability (18, 19). Subcutaneous infusions have a similar pharmacokinetic profile to the intravenous route. Compared with intermittent subcutaneous injections, CSAI has a longer apparent plasma half-life and simulates the physiological stimulation of striatal neurons (20–22).

The efficacy and safety of novel formulations of subcutaneous carbidopa/levodopa are also explored. ND01612 (NeuroDerm) is a novel formulation of levodopa/carbidopa that is administered subcutaneously via a pump. In a 28-day open-label study, 33 PD patients were randomized to treatment with either a 24-h infusion or a 14-h waking day infusion. The overall off time was reduced by 2 h from a baseline 5.3 h, with a larger reduction in the 24-h group (2.8 compared to 1.3). Eight patients in the 24-h group experienced complete resolution of the off time. The on time without TSD was increased by 3.3 h from a baseline of 8.9 h, and the on time with TSD was reduced by 1.2 h overall from 1.9 h at baseline. The 24-h infusion group experienced a −19.1 mean change in the UPDRS-III score from baseline. The most common AEs were infusion site reactions (4).

Foslevodopa/foscarbidopa (ABBV-951) is a subcutaneous levodopa/carbidopa prodrug formulation that is in development for treating the motor complications of advanced PD. A phase 1, single-blind study demonstrated that foslevodopa/foscarbidopa infusion provides stable levodopa and carbidopa exposures compared to oral dosing. The safety and tolerability of the infusion were also assessed, and the safety profile was determined to be favorable. Five of the 28 subjects had treatment-related AEs, including infusion site pain, dizziness, dyskinesia, paresthesia, and euphoric mood. Only one adverse event of anxiety was considered severe, while all other AEs were mild or moderate in severity (31).

Interstitial RF consists of creating a lesion through frictional heating of ionic oscillations of an intracranially placed electrode coupled to an RF generator. During the procedure, patients are awake, and the target is confirmed with a test stimulation. A lesion is created near the tip of the active electrode, which is where the greatest heating occurs. RF lesioning allows for the creation of distinct lesion borders with immediate results and intraoperative confirmation of symptom improvement (33, 34).

RF ablation allows for the creation of distinct lesion borders and intraoperative physiological confirmation of lesion creation. RF ablation provides immediate treatment effects, but the effects are not adjustable or reversible. Drawbacks include less predictability in the size and shape of the lesion. While adverse effects are generally mild and transient, the surgical risk of intracerebral hemorrhage must be considered (33). While used less commonly in high-income countries, ablative therapies requiring fewer resources such as RF remain a reasonable treatment strategy in emerging countries with limited resources (40).

SRS lesioning is a noninvasive procedure that uses computerized dosimetry planning and image-guided stereotaxy to deliver a single dose of ionizing radiation to an intracranial target volume. Different devices are used to deliver radiation, including GammaKnife® and linear accelerators (34).

Gamma knife (GK) thalamotomy has been shown to improve motor function in patients with PD. The radiation dose used in GK thalamotomy ranges from 120 to 180 Gy, and it appears that doses less than 120 Gy do not provide clinical benefits (41). In a prospective study of unilateral gamma knife thalamotomy for patients with PD and essential tremor (ET), Ohye et al. reported an improvement in tremor scores (UPDRS-II and III), with 81.1% of patients evaluated as having good or excellent results at 24-month follow-up (42). A more recent prospective, single-blind study of unilateral GK thalamotomy for patients PD and ET reported a 54.2% improvement in the upper limb tremor score (43). GK pallidotomy has also been performed to treat PD, although only a few reports have described the procedure (44). The GPi is considered a high-risk target for GK radiosurgery due to its proximity to the optic tract and the risk of optic neuropathy (45). Likewise, the STN is not an advised target due to the risk of neurologic complications including hemiparesis, dysarthria, and gait disturbances (46).

A major benefit of GK radiosurgery is that it allows for precise intracranial treatment without cranial opening, eliminating the surgical risk of hemorrhage as it does not require the creation of burr holes or puncturing of brain tissue. A major drawback to GK radiosurgery for treating movement disorders is that there are often significant delays before any symptom improvement is appreciated. In previous reports, a delay of 3–6 months was noted prior to the onset of tremor suppression (42, 43). Additionally, the radiation response after GK radiosurgery can be variable. Approximately 10% of patients exhibited excess reactions, demonstrated by a high signal zone in the thalamus and surrounding area on MR images, even when the administered radiation dose was uniform (42). The delayed effect, exposure to ionizing radiation, and lack of intraoperative feedback are all disadvantages that must be considered when utilizing GK radiosurgery as a treatment for advanced PD (33). GK radiosurgery remains a good option for patients with significant medical comorbidities who cannot undergo more invasive procedures like DBS or RF (36).

Early experiments investigating the neurosurgical applications of focused ultrasound demonstrated ultrasound as an effective tool for intracranial lesioning. However, craniotomies were required to avoid trajectory damage and wave distortion by the skull, and effective monitoring techniques were lacking. The subsequent development and hemispheric distribution of phased arrays have allowed for a completely transcranial procedure (47). In this procedure, an intracranial region is targeted with high-intensity focused ultrasound beams. Ablation of the target structure is achieved via ultrasonic mechanical energy that is absorbed within the focal target volume and converted into heat. Real-time monitoring of energy deposition and accurate targeting is made possible by MRI guidance and MR thermography (33). MRgFUS allows for accurate ablation of deep brain structures while eliminating the need for a craniotomy. An additional advancement of MRgFUS over that of other ablative procedures is that the patient can be examined after each sonication. During MRgFUS, a series of verification and low treatment sonications at low temperatures are performed, resulting in transient benefits and transient or minimal adverse effects. After clinically verifying the target with lower-temperature sonications, the temperature is then increased, resulting in a permanent lesion. MRgFUS of the VIM is currently approved for the treatment of tremors. Other targets are currently under investigation (48).

MRgFUS thermal ablation may be a reasonable treatment option for patients with advanced PD who are not good candidates for more invasive techniques like DBS. As with all ablative therapies, the permanent and nonadjustable nature of MRgFUS ablation must be considered. However, during treatment, patients are clinically assessed for symptomatic improvement and adverse effects, and treatment can be adjusted as necessary. The acoustic properties of the patient's skull impact the efficiency of focusing acoustic energy on a target. These properties vary from patient to patient, and it is critical to understand the relevant ultrasound physics and limits of the technique to avoid any potential adverse effects. In general, it is recommended that patients have a skull density ratio of >0.4 to be considered candidates for this treatment. Patients with artifacts from previous cranial surgical interventions or skull shape deformations may not be suitable candidates for MRgFUS ablation (53). Other considerations are that the patient's head must be shaved, and the MRI environment and increased operative time may be uncomfortable or intolerable for claustrophobic patients. Similarly, the presence of MRI-incompatible implants must be considered (33).

Laser-induced thermal therapy (LITT) is an FDA-approved therapy for ablating intracranial soft tissue and brain structures including brain tumors, radiation necrosis, and epileptic foci. A laser diffusing fiber within a cooling catheter is stereotactically placed into the tissue of interest, and the laser fiber provides high-energy laser light circumferentially. The use of LITT with MRI provides real-time thermal imaging, allowing control of the ablation. MRI thermal imaging and LITT software provide real-time temperature feedback and controls, resulting in precise tissue ablation. A real-time necrotic zone and transitional zone are provided in real time. The surgeon can contour the ablation via either advancing or pulling back the fiber or by using multiple catheters/fibers. This therapy has been used extensively in epilepsy and brain tumors. So far, only a few case reports have been reported describing its unilateral use for pallidotomy or thalamotomy for PD (54).

Randomized control studies of DBS to best medical treatment have demonstrated superior improvements with DBS. At 6-month postoperatively, motor symptoms in the on-medication/on-stimulation state and the quality of life improved by 20% compared with baseline, while the best medication group showed no clinical improvement (57). Furthermore, long-term reports of bilateral STN-DBS indicated that patients remain considerably improved compared with baseline (58). In a 13-year follow-up study, bilateral STN-DBS significantly improved motor symptoms in patients with early-onset PD. In the off-medication/on-stimulation state, patients had a 54% reduction in the total UPDRS-III score compared to that in the off-medication/off-stimulation state. Similarly, patients had a 48% reduction in the total UPDRS-III score in the on-medication/on-stimulation state compared to that in the on-medication/off-stimulation state. No hardware or surgical-related AEs were reported in the group of patients. Stimulation-induced dyskinesia occurred in four patients during the first 3 months of stimulation (59).

Another study retrospectively evaluated the clinical outcomes of bilateral GPi-DBS in 65 patients with PD. A 29% reduction of the total UPDRS-III score was appreciated in the off-medication state at 1-year follow-up, while no significant change was appreciated in the on-medication state. The most common AEs included gait and postural disorders, lead fractures, and dysarthria (60).

In a randomized controlled trial comparing DBS of the GPi and STN in patients with advanced PD, it was found that STN-DBS provides greater improvement in off-drug phase motor symptoms compared to GPi-DBS. Intention-to-treat analysis showed a greater reduction in the off-drug phase UPDRS-III score after 3 years in the STN group (32% reduction) compared to that in the GPi group (23% reduction) (61).

Adverse events in DBS can be separated into three categories: related to surgery, hardware, and stimulation. Adverse events related to surgery include but are not limited to infection and hemorrhage. The rate of DBS infections is generally less <7% and may necessitate complete removal of the DBS system. The use of intraoperative vancomycin powder has not been shown to decrease infection risk (62–66). The overall risk of intracranial hemorrhage is estimated to be <5%, and 1%–2% of intraventricular or intraparenchymal hemorrhage is symptomatic. Risk factors for intracranial hemorrhage include trans-sulcal trajectories and hypertension. Hardware-related adverse events occur in <2.2% of cases and include lead fracture, lead migration, and electrode malfunction (67). Stimulation side effects occur due to the electrical spread to unwanted regions that may be mediated with parameter adjustments, including focusing the volume of tissue activated with directional contacts (67). Directional leads also allow for increasing the therapeutic window, the onset of clinical benefits to that of stimulation-induced adverse effects, while resulting in less battery consumption (68).

The nonlesional and adjustable nature of DBS is among the advantages offered over other methods for neuromodulation (55). While many advantages exist, several patient-specific and ethical considerations must also be examined. DBS is a minimally invasive surgery with low incidences of infection and symptomatic hemorrhage. Additionally, DBS incurs large capital costs and requires a multidisciplinary expert team to provide optimal programming and troubleshooting for the patient. DBS can also be problematic for some in that it commits patients to a lifelong implant requiring subsequent battery (IPG) changes (55). However, due to advancing technology, battery life continues to increase and patients have the option of a rechargeable battery, thus decreasing the number of battery replacement surgeries. Patients who travel far to receive DBS surgery also now have the option to have DBS programmed remotely (69, 70). Ideal PD candidates for DBS are those with motor fluctuations and/or TSD without having other “red flags” for an alternative diagnosis. Patients with neuropsychiatric issues and multiple comorbidities may be poor candidates. Many centers exclude patients older than 70 years from DBS surgeries, although an age limit for DBS has not been defined (34).

Most clinicians favor DBS over lesioning procedures wherever the former is widely available. However, lesioning procedures are still performed in less developed countries due to financial constraints, lack of infrastructure and training, and limited follow-up care for DBS (40). Lesioning procedures are relatively cheaper and reduce postoperative care and hardware-related complications compared to DBS (71). However, lesioning procedures are not reversible, and postprocedure optimization is impossible without revision surgery. The surgical reversibility of DBS has recently been contested in the literature (72). Postmortem studies evaluating patients who have undergone DBS demonstrate pathological changes including glial scarring. It is unclear as to whether the contribution to these findings is solely due to lead insertion, chronic stimulation, or a combination (73, 74). For our review, we will consider DBS as a reversible procedure, given that the aforementioned irreversible features of DBS are not comparable to the significant irreversible changes that occur with brain lesioning procedures. Moreover, there is a lack of long-term data to support whether or not these irreversible features of DBS are truly permanent (72). The increased rate of side effects with bilateral lesions is a major limitation of lesioning procedures. These side effects include aphasia, dysarthria, dysphagia, and cognitive deficits (34). The question of morbidity with bilateral lesions has been revisited in the focused ultrasound era but has historically been problematic when treating PD patients (33, 75).

Many gene therapy studies have focused on increasing DA synthesis enzyme expression, namely, tyrosine hydroxylase (TH), L-amino acid decarboxylase (AADC), and GTP cyclohydroxylase (GCH1).

One study assessed the safety and efficacy of adeno-associated serotype 2 viral vector encoding human AADC (AAV-hAADC-2) intraputaminal infusion in advanced PD patients. At 6 months, patients demonstrated a significant improvement in both UPDRS total and UPDRS-III scores (85). In another study, advanced PD patients received low- or high-dose intraputaminal infusion of adeno-associated viral vector (AAV-hAADC). At 6 months, patients exhibited mean improvements of 36% and 28% in the off- and on-state UPDRS-III scores, respectively (86).

AADC gene therapy for PD has a favorable safety and tolerability profile. Five out of the 31 patients who received the drug across all studies reported procedure-related SAEs including four intracranial hemorrhages and one deep vein thrombosis (78, 85–88). Eight participants reported transient dyskinesia increase (87, 88).

ProSavin is a novel lentiviral vector that encodes GCH1, TH, and AADC. The phase ½ open-label, dose-escalation study has been completed, and a long-term safety and efficacy study evaluating the same patient cohort is ongoing (81, 89). OXB-102 (AXO-Lenti-PD) is an optimized version of ProSavin that expresses the same genes. A phase ½ safety and dose-escalation study is currently ongoing (90).

Patients treated with ProSavin experienced a significant reduction in the off-state UPDRS-III score, which was maintained at a 2-year follow-up (81, 89). A 21-point reduction in the UPDRS-III off-state score and a 2.2-h increase in the on time from baseline were reported in the first four patients treated with high-dose OXB-102 (90).

In total, 91% of drug-related AEs in patients treated with ProSavin were mild and/or occurred within 1-year follow-up. Acute psychosis, dyskinesia, and an unspecified nervous system disorder were among the three therapy-associated SAEs (89). No therapy-related SAEs were reported after OXB-102 treatment at 6-month follow-up (90).

Loss of dopaminergic neurons in the substantia nigra results in decreased GABA input to the STN, resulting in the loss of inhibition. Infusion of muscimol (GABA) agonist into the STN has been found to improve parkinsonian symptoms temporarily. Glutamic acid decarboxylase (GAD) is the rate-limiting enzyme for GABA production, and gene therapy consisting of insertion of the GAD gene into an adeno-associated viral vector (AAV2) has been used in animal and human studies with success. The hypothesis is that delivering GAD to the STN restores GABA transmission within the STN and normalizes output from the STN. A phase I trial of 12 patients receiving a unilateral infusion of AAV2-GAD into the STN demonstrated significant improvements in motor UPDRS scores at 3 and 12 months and confirmed the safety of the therapy (91). A phase II double-blinded trial, randomizing 23 patients to sham surgery and 22 patients to bilateral STN infusion of AAV2-GAD, showed a significant improvement in motor UPDRS scores at 6 and 12 months with an associated decrease in daily dyskinesias (92, 93).

PD patients demonstrate loss of endogenous neuronal growth factors in addition to dopaminergic neuron loss (76). Extensive research has been done on restoring growth factors to protect and improve the health of dopaminergic neurons. Glial cell-line derived neurotrophic factor (GDNF) and neurturin (NRTN) are two growth factors that have gained significant attention (94). Currently, only gene therapy with AAV2-GDNF is under active investigation in two clinical trials (76).

Results from the phase 1 study reported 36% and 54% increases in 18F-DOPA putaminal signal at 6- and 18-month follow-up, respectively. UPDRS scores remained stable over the course of the study, although motor symptoms in the patient cohort did not significantly worsen, as would be expected (76). AAV2-GDNF was shown to be safe and well tolerated. Scalp wound dehiscence in one patient was the only procedure-related SAE (79).

With genetic testing becoming more widely available, it is feasible to consider selecting patients for certain therapies based on their genotype. Patients with the advanced parkinsonian phenotype with motor fluctuations are expected to experience the greatest benefit from a gene therapy approach that aims to increase dopamine production. The unmodifiable nature of gene therapy must also be considered, particularly in patients with significantly advanced PD who are more susceptible to complications such as ICD and dopamine dysregulation syndrome. Studies investigating neurotrophic factors have emphasized the importance of targeting patients with earlier stages of PD when protecting and improving the health of dopaminergic neurons would be most beneficial (79, 80, 95).