An erratum is being issued to the article published in March 2022 in Muscle&Nerve 65(3):291–302, to ensure that data from the trial is shown correctly per the prespecified analysis plan. Subgroup analyses reported in the original publication for the key secondary endpoint, average change from baseline to endpoint in ALSFRS‐R, used an incorrect model and incorporated interaction terms between the subgroup and treatment. These analyses have been corrected to employ the pre‐specified efficacy model for subgroup analysis. The following corrections and clarifications apply to this publication and its supplement document.
DOI: 10.1002/mus.27472
Under “Results, Secondary Endpoints” section, in paragraph 2 sentence 2, added the p‐values for both subgroups. Now the sentence reads as “For example, MSC‐NTF participants progressed on average two points less on the ALSFRS‐R compared to placebo (p=0.05), while for participants with more advanced disease the change from baseline to Week 28 was similar between treatment groups (p=0.97).”
Under “Results, Post‐hoc efficacy analyses” section, in paragraph 3 sentence 2, corrected LS means. Now the sentence reads as “LS means for placebo exhibit far less spread in the LS mean (range) change compared to MSC‐NTF across baseline thresholds: MSC‐NTF (−1.39, ‐4.82), placebo (−4.24, ‐5.98), meaning placebo‐treated participants lost 4‐5 points on average in 28 weeks across baseline thresholds while MSC‐NTF‐treated participants lost less function at higher compared to lower baseline levels.”
In the same paragraph as above, added text to the end of sentence 3. Now the sentence reads as “Comparison of the treatment groups across baseline thresholds highlights that MSC‐NTF‐treated‐participants lose on average, 2 points of function less compared to placebo‐treated‐participants, which is maintained or gets larger through the prespecified threshold of 35 (Figure S4B, Table 4, Table 3), a difference which is nominally statistically significant (p < .05) for all thresholds >26 up to 35 and an exact p‐value of 0.050 for the threshold ≥35).”
Under “Discussion” section, in paragraph 2 sentence 2, text revised to reflect corrected results. The sentence reads “In a pre‐specified subgroup with baseline ALSFRS‐R ≥35, there was a higher response rate on the primary endpoint with MSC‐NTF compared to placebo along with a larger treatment effect across all endpoints with MSC‐NTF compared to placebo including a LS mean difference of 2.09 in the average change from baseline to week 28 (p=0.05); no other endpoint had a p‐value of 0.05 or less, a finding which is perhaps not surprising given the additional power loss on the categorial response endpoints.”
In the last sentence of the same paragraph as above, corrected treatment difference estimates. Now the sentence reads as “A 2.48‐point treatment difference was observed at this threshold in the change from baseline to week 28 on the ALSFRS‐R, also with a nominally significant p‐value (p=0.020).”
Under “Discussion” section, in paragraph 4 and second to last sentence, corrected LS means and added clarification. Now the sentence reads as “However, a similar sensitivity applied to average changes on the ALSFRS‐R highlights that Placebo participants were consistently losing far more function over time than MSC‐NTF (p≤0.05 across all thresholds >26), with the range of LS means on placebo being much tighter (−4.24, −5.98) across 28 weeks compared to MSC‐NTF (−1.39, −4.82).”
Table 3 revised: Corrected MMRM estimates for the >=35 and <35 subgroup analysis of the LS Mean Change in ALSFRS‐R endpoint. The changes to Table 3 are shown below.
Table 4 revised: Corrected MMRM LS estimates in the Mean Change from Baseline in ALSFRS‐R endpoint. The changes to Table 4 are shown below.
TABLE 4.
Efficacy Analyses Across ALSFRS‐R Baseline Thresholds
| ALSFRS‐R MMRM Change from Baseline to Week 28 Secondary Endpoint | |||||
|---|---|---|---|---|---|
| ALSFRS‐R Baseline Score | n (%) total participants | MSC‐NTF LS mean (SE) | placebo LS mean (SE) | Delta | p‐value |
| ≥35 | 48 (25) | −1.56 (0.82) | −3.65 (0.70) | 2.09 | 0.050 |
| ≥34 | 57 (30) | −1.39 (0.96) | −4.24 (0.80) | 2.86 | 0.022 |
| ≥33 | 69 (37) | −2.27 (0.90) | −4.66 (0.79) | 2.39 | 0.044 |
| ≥32 | 78 (41) | −2.49 (0.78) | −4.67 (0.74) | 2.18 | 0.040 |
| ≥31 | 80 (42) | −2.56 (0.78) | −5.04 (0.75) | 2.48 | 0.020 |
| ≥30 | 92 (49) | −3.18 (0.72) | −5.57 (0.69) | 2.39 | 0.016 |
| ≥29 | 97 (51) | −3.71 (0.74) | −5.84 (0.69) | 2.13 | 0.033 |
| ≥28 | 105 (56) | −3.61 (0.71) | −5.72 (0.68) | 2.11 | 0.030 |
| ≥27 | 110 (58) | −3.93 (0.72) | −5.87 (0.68) | 1.94 | 0.046 |
| ≥26 | 116 (61) | −4.82 (0.73) | −5.98 (0.72) | 1.16 | 0.250 |
TABLE 3.
Efficacy Results: Primary and Secondary Endpoints at 28 Weeks, mITT Population
| ALSFRS‐R ≥ 35 | ALSFRS‐R < 35 | |||
|---|---|---|---|---|
| Pre‐specified Subgroup ALSFRS‐R Total Baseline | MSC‐NTF (N = 26) | Placebo (N = 32) | MSC‐NTF (N = 69) | Placebo (N = 62) |
| Secondary Endpoints through Week 28 | ||||
| ALSFRS‐R Total Score, LS Mean Change (SE) | −1.56 (0.82) | −3.65 (0.70) | −6.95 (0.84) | −7.00 (0.88) |
| LS Mean Difference (95% CI), p‐value | 2.09 (0, 4.18), p = 0.05 | 0.05 (−2.32, 2.42), p = 0.97 | ||
Supplementary material: changes associated with Table 4 have been updated online in associated graphs:
In the supplement of the publication, under “Statistical Methodology”, added one sentence after paragraph 1 which reads as “Analyses of all endpoints run by subgroup used the pre‐defined efficacy model.”
Updated Figure S2 Panel B and Panel C.
Updated Figure S3 Panel C.
Updated Figure S4 Panel B.
Updated Figure S5.