Background

The neurotrophic peptide CDNF promotes survival of midbrain dopaminergic neurons, which degenerate in Parkinson’s disease. The peptide does not cross the blood-brain barrier. It is administered directly into the brain via a surgically implanted delivery device.

Unlike other neurotrophic factors, CDNF is not a secreted protein. It is found mainly in the lumen of the endoplasmic reticulum in cells. Studies suggest it regulates the unfolded protein response (UPR), a signaling pathway that contributes to ER stress and cell death in multiple neurodegenerative proteinopathies (reviewed in Huttunen and Saarma, 2019). CDNF knockout mice show age-dependent deficits in dopamine function in the brain and loss of enteric dopamine neurons, both of which occur in the early stages of PD (Lindahl et al., 2020).

Extensive preclinical work on CDNF has been done in the laboratory where it was discovered. In rodent models of toxin-induced dopaminergic cell loss, single intracerebral doses or chronic brain infusion of CDNF were reported to improve motor function and neuron survival and function (Lindholm et al., 2007; Voutilainen et al., 2011; Airavaara et al., 2012). CDNF also enhanced the therapeutic benefit of acute subthalamic deep-brain stimulation (DBS), a current treatment for PD, in a rat model of late-stage disease (Huotarinen et al., 2018).

The same group, as well as independent investigators, published data on a gene-therapy approach. Using viral vectors to express CDNF in the striatum protected neurons and improved motor behaviors in the 6-hydroxydopamine toxicity model of parkinsonism in rats (Bäck et al., 2013; Ren et al., 2013; Wang et al., 2017).

CDNF was reported to inhibit α-synuclein oligomer toxicity in cultured dopaminergic neurons (Latge et al., 2015). No data are published on CDNF in α-synuclein animal models.

In 6-OHDA-lesioned marmoset monkeys, CDNF treatment increased dopamine transporter (DAT) binding activity on PET scans, suggesting it protected dopaminergic neurons (Garea-Rodríguez et al., 2016).

In other disease models, CDNF improved memory in both APP/PS1 and wild-type mice after intrahippocampal injections of the peptide or gene therapy with a viral vector. Treatment caused no change in amyloid load or hippocampal neurogenesis in the AD mice (Kemppainen et al., 2015). CDNF reportedly promoted recovery in rat models of spinal cord injury and stroke (Zhao et al., 2016; Zhang et al., 2018).

Findings

In September 2017, Herantis and Renishaw began a Phase 1/2 study of CDNF and an experimental delivery device in people who have had idiopathic Parkinson’s disease for about a decade. Called TreatER, the trial enrolled 17 patients at three university hospitals in Finland and Sweden. Participants first underwent neurosurgical implantation of the infusion device in a procedure comparable to the placement of a DBS electrode in advanced PD. Then they received six monthly infusions of recombinant human CDNF, starting at 120 micrograms and escalating to 400 or 1,200 micrograms, or placebo, into their putamina. After six months, all groups were eligible to receive CNDF monthly for an additional six months, with four years of follow-up. Twenty-four endpoints cover safety and tolerability, plus multiple clinical measures of mood, cognition, motor function, non-motor symptoms and general function; device-related safety, accuracy of implantation and infusion success; formation of anti-CDNF antibodies; DAT PET imaging; and serum and CSF levels of CDNF and α-synuclein.

On February 25, 2020, Herantis announced partial results from the six-month placebo-controlled main study (press release; slide presentation). It reported no serious adverse events related to CDNF treatment, and only mild to moderate side effects that were similar in treated and placebo groups. Two patients left the study because of serious infections requiring hospitalization, probably related to the device implantation and infusion process. Both recovered, and surgical procedures were altered to prevent future infections, the company said.

In the DAT-PET study, one of the treatment groups reportedly had a 17 percent increase in DAT signal in the putamen after six months, compared with a decline in the other treatment group and placebo. All participants who completed the first stage decided to continue into the six-month extension, where they received CDNF monthly at the low or high dose.

In November 2020, the company announced it would stop developing CDNF by intracranial infusion, and instead pursue non-surgical modes of subcutaneous injection or intranasal delivery (press release). Both approaches are currently in preclinical stages (pipeline).

In March 2021 at AD/PD, 12-month TreatER trial results were presented. Compared to the first six months, fewer adverse events were reported in the extension phase, with no dose-limiting toxicities and no detection of anti-CDNF antibodies. Most drug-related side effects were mild to moderate, and all participants recovered. The most common were dyskinesias, headache, feeling cold, fever, or fatigue, impulse control disorder, nausea, and weight loss. On the exploratory efficacy endpoints, treatment did not worsen symptoms, and some patients showed signs of potential benefits. The low-dose group had a reduction in bradykinesia at six months, as measured with a movement-measuring device worn on the wrist. However, there was no significant difference between groups in the UPRDS motor score. DAT levels in the low-dose group remained stable after 12 months, while the other groups declined. A proteomics analysis of CSF found that changes in 50 previously selected CSF biomarkers were more prominent in the lower-dose group. In three patients, proteomics changes correlated with improvements in the UPRDS or DAT-PET.

For details on CDNF trials, see clinicaltrials.gov.