A recently published phase 2 randomized, placebo-controlled trial demonstrated that three infusions of 10 × 10⁶ allogeneic human mesenchymal stem cells (allo-hMSCs) significantly improved motor function in patients with mild to moderate Parkinson disease (PD). Additional clinical trials are needed to ensure consistency and reliability of allo-hMSC batches, as investigators observed a profound improvement in the placebo group and a less robust improvement in the two-infusion group.1
Conducted between November 2020 and July 2023, the study comprised 45 patients with PD who received either 3 allo-hMSC infusions (n = 16), 1 placebo followed by 2 allo-hMSC infusions (n = 14), or 3 placebo infusions at 18-week intervals (n = 15). Over an 88-week follow-up, the primary outcome assessed was achieving at least a 70% posterior probability (PP) of a difference in the proportion of patients with a ≥5-point improvement in OFF-medication MDS-UPDRS Part III scores at week 62.
Led by Mya C. Schiess, MD, a neurologist at UTHealth Houston Neurosciences Neurology, Texas Medical Center, a greater proportion of patents in the 3-infusion group achieved at least a 5-point improvement in MDS-UPDRS III compared with placebo at week 62 (mean difference, 5.0%; 95% credible Bayesian interval, –2.3% to 24.8%; PP = 93.7%). Using a more stringent threshold of at least 11 points, the results remained consistent, with more patient in the 3-infusion group improving relative to placebo (MD, 13.3%; 95% Crl, –6.1% to 37.8%; PP = 91.5%).
At week 62, fewer patients in the 2-infusion group met the primary end point (MD, –62.4%; 95% Crl, –85.5% to –32.1%; PP ≥99.9%) as well as the more stringent threshold (MD, –63.8%; 95% Crl, –86.0% to –32.5%; PP≥99.9%). Additional data from this time point revealed a –16.9-point (–19.5% to –14.2%) improvement in the 3-infusion group, –3.9-point (95% Crl, –6.9% to –1.1%) improvement in the 2-infusion group, and a –14.6-point (95% Crl, –17.5% to –11.6%) improvement in the placebo group. Overall, the three-infusion group differed from placebo by −2.3 points (95% CrI, −6.1 to 1.6; PP = 87.8%), while the two-infusion group showed a +20.6-point difference (95% CrI, −37.6 to 24.5; PP ≥99.9%).
READ MORE: Fully Enrolled CYPRESS Phase 3 Trial to Test Ampreloxetine in Symptomatic Neurogenic Orthostatic Hypotension of MSA
Regarding the high improvement in placebo scores, the study authors wrote, “Although this finding may seem counterintuitive, it is important to note that studies on PD progression suggest an average annual motor decline (increase) of approximately 2.4 points in MDS-UPDRS-III scores. Yet in this trial, all treatment groups showed improvement in their motor scores over the 88 weeks of participation; an outcome that may, in part, reflect a placebo effect.”
They went on to add that the residual placebo effect may have influenced the results since the primary outcome was assessed at week 62, which was 26 weeks or approximately 6 months after the last infusion. By week 88, and with a more stringent threshold (≥11-point improvement in MDS-UPDRS-III), the differences between the 3-infusion and placebo groups widened (–3.3 points; 95% Crlm –7.3% to 0.7%; PP = 94.7%), supportive of a waning placebo response. For context, the 2-infusion group remained worse than placebo by +9.0 points (95% Crl, 4.8% to 13.0%; PP ≥99.9%) by week 88.
When compared with placebo, more participants in the 3-infusion group achieved at least a 12-point improvement in MDS-UPDRS at week 62 (MD, 22.0%; 95% Crl, –21.0% to 49.4%; PP = 96.3%) and week 88 (MD, 48.4; 95% Crl, 23.9% to 74.0%; PP ≥99.9%). In line with other data, fewer participants in the 2-infusion group showed improvement relative to placebo at week 62 (MD, –43.0; 95% Crl, –71.2% to –8.3%; PP = 99.1%) and at week 88 (MD, –20.8; 95% Crl, –53.1% to 14.1%; PP = 87.8%).
In terms of safety, 10 mild and transient treatment-emergent adverse events were reported, with no severe events. In the 3-infusion group, single cases of malaise, transient hypertension not requiring medication, and vomiting occurred, while the two-infusion group experienced only constitutional symptoms (fatigue, flu-like symptoms, headache), all of which resolved. Three patients initially showed a panel reactive antibody response presumed to be donor-specific, but HLA typing confirmed two were unrelated to the donor, and the third—found in a placebo patient—was also deemed unrelated after unblinding.