Letter: Stem Cell Transplantation for Parkinson Disease: Déjà Vu All Over Again?

To the Editor: Recently, Schweitzer et al1 published a case report in the New England Journal of Medicine (NEJM) discussing a Parkinson disease (PD) patient whom they had treated with bilateral putaminal implants of genetically engineered dopaminergic neurons. The neurons were generated in Vitro from induced pluripotent stem cells that were derived from autologous skin fibroblasts. At 2 yr, the patient's motor performance, as measured with the UPDRS (Unified Parkinson's Disease Rating Scale) motor subscale, was only modestly improved as was his 18F-fluorodopa positron emission tomography (PET), a surrogate marker for graft survival. In contrast, the patient's PDQ-39 (39-item Parkinson's Disease Questionnaire) score, representing his subjective experience, was dramatically improved. The authors claim that this single case represents a “proof of principle,” and it is likely that this technique will move on to larger clinical trials that may involve dozens more patients. Unfortunately, the history and the details of this case suggest that this novel therapy is more likely than not to represent the latest in a series of failures that have plagued the field of cellular replacement therapy for PD over the last 30+ yr. Consequently, closer scrutiny is warranted. Beginning in the 1980s, neural transplantation investigators trialed 4 unique tissue sources (adrenal medullary autografts, embryonic mesencephalic allografts, porcine mesencephalic xenografts, and retinal pigmented epithelial allografts) as cellular replacement strategies for PD.2-13 In each case, the robust clinical responses and acceptable safety profiles reported in the small initial open label trials (OLTs) could not be replicated under more rigorous scientific scrutiny. Adrenal medullary autotransplantation was abandoned when independent investigators reported only modest clinical improvements and unacceptably high rates of serious adverse events (SAEs).3,4 The other therapies were tested in ground-breaking, prospective, randomized, double-blind, sham surgery-controlled trials (randomized controlled trials), all with similar results: (1) failure to generate clinical improvement statistically greater than placebo; (2) significantly reduced clinical responses to the active therapy as compared to the preceding OLTs; and (3) the emergence of SAEs that were not observed in the initial OLTs.9,10,13 Most notable of these complications was graft-induced dyskinesiae, a serious and debilitating side effect that often required an additional neurosurgical intervention to control.9,10 Though these trials failed to achieve their primary goal of developing an effective cellular replacement therapy for PD, a great deal was learned in the process. First, we learned how strong and durable the placebo response is in patients with PD, particularly those who commit to undergoing an experimental surgical therapy.14,15 While the use of sham controls was controversial in the 1990s,16 their value in assessing the true biological impact of surgically administered cellular and gene therapies for PD has now been demonstrated convincingly.17-19 Second, we discovered the impact that observer bias in OLTs and “effect declines” have on clinical research, as time and again, therapies that performed robustly in OLTs profoundly underperformed in the subsequent blinded evaluations.20,21 Third, we learned that small OLTs are insufficient to determine the true risks of a novel invasive procedure.20 Certainly, no one would consider 6 to 12 patients an adequate cohort for a phase I safety trial of a new medication, yet, repeatedly, investigators have incorrectly claimed as safe a procedure they performed precisely that many times. Furthermore, both our understanding of PD and the therapies available to treat it have advanced significantly over the last 30 yr, raising the bar even higher for the introduction of a novel therapy that is expensive and invasive with unknown long-term effects. For example, we know that PD-related neural degeneration is not restricted to the substantia nigra (SN), but is widespread,22 so that replacing only the dopaminergic cells of the SN can never represent a cure for PD and will likely provide no more effective symptom control than is currently available with proven therapies such as deep brain stimulation and dopaminergic medications. We also know that the aging PD brain is a hostile environment to grafted cells as evidenced by the presence of Lewy-body degeneration in postmortem analyses of transplanted patients.23,24 Even when grafted tissue survives for long periods of time, and incorporates into the putamen as desired, patients are not necessarily improved.25,26 Furthermore, we now understand that alpha-synuclein functions as a lethal prion in PD, jumping from neuron to neuron as it spreads throughout the brain, leaving degenerating neurons in its wake.27,28 Most importantly, we are learning that this abnormal alpha-synuclein likely forms in the gut where it interacts with environmental toxins and/or intestinal flora and then gains access to the nervous system via termini of the vagus nerve, establishing the foundation for developing prevention or harm mitigation strategies focused on the gut microbiome and the vagus nerve.29-31 Returning to the recent NEJM paper, STAT News reported that this patient, a physician, spent $2M of his own money to develop and receive this therapy.32 While not unethical, this level of commitment is known to enhance the placebo response15 and so, given the history delineated above, this modest open-label result in 1 very committed patient must be taken with the proverbial grain of salt. Moreover, the fact that the patient did not suffer an SAE says nothing about the true safety of the intervention. Consequently, if this work proceeds, it should with utmost caution, oversight, and skepticism. Funding This study did not receive any funding or financial support. Disclosures The author has no personal, financial, or institutional interest in any of the drugs, materials, or devices described in this article.