Trial Report REGAIN: A Randomized Controlled Clinical Trial of Oxaloacetate for Improving the Symptoms of Long COVID, 2025, Vernon et al

forestglip

forestglip

Moderator
Staff member
REGAIN: A Randomized Controlled Clinical Trial of Oxaloacetate for Improving the Symptoms of Long COVID

Suzanne D Vernon, Candace Rond, Jennifer Bell, Brindisi Butler, Sara Isolampi, Annaleah Otteson, Pearl Phalwane, Samantha Mower, Shad Roundy, David Kaufman, Alan Brian Cash, Lucinda Bateman

Background
Long COVID is characterized by fatigue, cognitive dysfunction, and other persistent symptoms. This randomized, double-blind, controlled trial evaluated the efficacy of oral oxaloacetate (OAA) in improving fatigue and cognitive function in adults with Long COVID.

Methods
Sixty-nine participants were randomized to receive either 2,000 mg/day of OAA or control for 42 days. Primary outcome was fatigue reduction measured by the Chalder Fatigue Questionnaire (CFQ). Secondary and exploratory outcomes included the DePaul Symptom Questionnaire Short Form (DSQ-SF), health-related quality of life (RAND-36), cognitive function (DANA Brain Vital), and time upright (UP Time).

Results
No significant difference in CFQ fatigue reduction was observed between groups.

However, the OAA group showed significantly greater improvements in DSQ-SF fatigue and total symptom burden 21 days into the trial. Cognitive performance improved significantly in the OAA group, with strong correlations between symptom response and cognitive gains. OAA was well tolerated.

Conclusions
OAA may contribute to earlier improvements in symptom burden and cognitive function in individuals with Long COVID. Further studies are warranted.

Link (Frontiers in Neuroscience) [Provisionally accepted, currently only abstract]
 
Last edited:
They mentioned time upright as a measure but unlike the other measures didn’t say anything about the results. Surely time upright is going to be an important factor on cognitive performance for those affected by OI so should be considered in any judgment of improvement.
 
Abstract said:
However, the OAA group showed significantly greater improvements in DSQ-SF fatigue and total symptom burden 21 days into the trial. Cognitive performance improved significantly in the OAA group, with strong correlations between symptom response and cognitive gains. OAA was well tolerated.
Click to expand...
«Significant», but no mention of «clinical significant». Might just be a low p-value for a small change.
 
I wonder if this study is also sponsored by the supplement company. I sure hope that if they do have any responders in this trial, they do the honest thing and let them know that malic acid is cheaply available and will almost certainly do the exact same thing.
 
I had my test done through a biomedical center which included panels testing for antioxidants, iron metabolism, amino acids, fatty acids et. I think the tests are legit, but making up 'custom formula' to correct the imbalances is a scam. It did nothing for me.
 
Mij said:
https://benagene.com/pages/faqs

https://oxaloacetatecfs.com/pages/doctors
"CFS patients have low oxaloacetate levels"

True, mine were the lowest level on my OAT results. I took the supplement two decades ago and felt zero difference.
Click to expand...
The benaGene one? My understanding is they have some kind of patent for stabilizing it with vitamin C and that's why they can sell it for ridiculously expensive, but it may also work better than other oxaloacetate supplements if such have been on the market.
 
No, I didn't take the BeneGene brand. I don't remember which one I took or what dosage it was.
 
JES said:
The benaGene one? My understanding is they have some kind of patent for stabilizing it with vitamin C and that's why they can sell it for ridiculously expensive, but it may also work better than other oxaloacetate supplements if such have been on the market.
Click to expand...
Oxoaloacetate has to be converted into malate/malic acid first anyways, so stabilizing with vitamin C would not make any difference since it's going to get lobbed off before it does anything. I've been taking regular malic acid for months and can testify that it is absorbed perfectly fine, which is the only thing that vitamin C stabilization might help with. I suspect it's a situation where a drug company puts a completely irrelevant spin on an existing formula solely for the purposes of keeping the patent going for longer. You only need to show a tiny margin of benefit to qualify for the new patent.
 
Last edited:
IMG_0247.png
Figure 2. Change-from-baseline (delta) in the DSQ-SF fatigue and total symptom scores by visit and treatment group. Bar plots show the mean change from baseline (Visit 1) for the DSQ-SF fatigue and total symptom scores at Visits 2 and 3, comparing the OAA and control groups. Error bars represent standard errors. Repeated measures ANOVA revealed significant group × time interactions for both fatigue (p = 0.005) and total symptom burden (p = 0.028), indicating that the participants in the OAA group experienced earlier and greater symptom improvements. Tukey post hoc comparisons confirmed that the OAA group had significantly lower fatigue scores at Visit 2 (p = 0.012). The DSQ responder rates (defined as a ≥ 10% reduction in the total DSQ-SF score from Visit 1 to Visit 3) were 63% in the OAA group and 41% in the control group.

IMG_0248.png
Figure 3. Percent change in cognitive performance measures from Visit 1 to Visits 2 and 3 for the OAA and control groups. Boxplots display the percent change for simple reaction time, procedural reaction time, go/no-go, and total cognitive efficiency scores. The changes are normalized to each participant’s Visit 1 score. Horizontal lines within boxes indicate medians; the boxes span the interquartile range, and whiskers extend to 1.5 × IQR. Outliers are plotted as individual points. Between-group comparisons were conducted using independent t-tests for each visit. Exact p-values are annotated within each panel. Red font indicates statistically significant differences (p < 0.05).

IMG_0249.png
Figure 4. Correlation between the DANA cognitive scores and DSQ-SF symptom scores by treatment group and visit. Scatterplots with linear regression lines (black) illustrate the relationship between cognitive function (DANA brain vital scores) and symptom burden (total DSQ-SF scores) across the three study visits (Visits 1, 2, and 3) in the OAA and control groups. Each point represents an individual participant at a specific visit. In the OAA group, a statistically significant negative correlation was observed at Visit 1 (r = −0.40, p = 0.023), indicating that higher cognitive scores were associated with lower symptom severity. At Visits 2 and 3, the negative association persisted but was weaker and not statistically significant. In the control group, correlations were consistently weak and non-significant across all time points. These results suggest a stronger cognitive–symptom relationship in the OAA group, particularly at baseline.
 
The demographics shows that far more people in the control group were working (p<0.05). If that’s the case, they might have had less symptoms that could be reduced which might create some kind if ceiling effect.

They say they’ve calculated between-group differences, but those are scattered around in the text instead of being put in a table:
Fatigue, as measured by the CFQ, was the primary outcome. There were no statistically significant differences in improvement between the OAA and control groups for any individual CFQ item or for the CFQ total score (Table 2). The between-group effect size (Cohen’s d) for the change in the total CFQ score from Visit 1 to Visit 3 was −0.093, indicating a negligible and non-significant difference in fatigue reduction between the groups.
Click to expand...
The RAND-36 was used as a secondary outcome measure to assess health-related quality of life. Of the eight domains evaluated, only the energy domain showed a statistically significant between-group difference. RM-ANOVA for the energy domain revealed a significant main effect of time (F(2,126) = 28.3, p < 0.001) but no significant main effect of group (F(1,63) = 2.9, p = 0.09) or group × time interaction (F(2,126) = 0.82, p = 0.44). The Tukey post hoc tests showed that the between-group difference in energy scores was statistically significant only at Visit 1, where the OAA group reported lower energy levels (mean ± SE: 10 ± 2.3) than the control group (18 ± 2.1; p = 0.01) (Table 3). No significant differences were observed at Visits 2 or 3, although mean energy scores remained numerically higher in the control group.
Click to expand...
Tukey post hoc comparisons confirmed that fatigue scores were significantly lower in the OAA group compared to the control group at Visit 2 (p = 0.012), indicating earlier symptom improvement in the treatment arm. By Visit 3, both groups had improved, and the between-group difference was no longer statistically significant (p = 0.088), although the OAA group continued to show numerically lower fatigue scores.
Click to expand...
SRT improved in both groups, and there were no between-group differences at either time point (p = 0.054 at Visit 2; p = 0.091 at Visit 3).
Click to expand...
There were no differences in GNG at Visit 2 (p = 0.547); however, by Visit 3, the OAA group demonstrated significantly greater improvement in GNG compared to the control group (+10.0% vs. +0.9%; p = 0.017). The between-group comparisons of the percent change in total cognitive efficiency revealed significantly greater improvement in the OAA group at both Visit 2 (mean change: +8.7% vs. +0.2%; p = 0.021) and Visit 3 (+10.7% vs. –0.04%; p = 0.007), indicating a robust cognitive benefit relative to the control group.
Click to expand...
The CFQ showed similar improvements across both treatment arms, with a negligible between-group effect size. This underscores the limitations of the CFQ’s sensitivity in detecting treatment effects in long COVID, particularly given the complex, multidimensional nature of fatigue in this condition (Gladwell et al., 2024). In contrast, the DSQ-SF, a multidomain symptom instrument developed for ME/CFS and used to evaluate long COVID, detected significantly greater reductions in fatigue and total symptom burden in the OAA group by Visit 2 (McGarrigle et al., 2024). Although the between-group responder rate did not reach statistical significance, these findings underscore the importance of selecting outcome measures, such as the DSQ-SF, that are designed to detect clinically meaningful change in heterogeneous disease populations.
Click to expand...
The only reason to do this is to obfuscate the null results. Shame on them.
 
It’s very frustrating for me to see people I genuinely care about all over bluesky starting to worry about saving up to try this supplement. I don’t have the energy to argue-explain to every one of them that this study isn’t very good…
 
Utsikt said:
The demographics shows that far more people in the control group were working (p<0.05). If that’s the case, they might have had less symptoms that could be reduced which might create some kind if ceiling effect.
Click to expand...
You quoted this related part, but I want to highlight it:

Control group started off with a significantly higher energy level as measured by RAND-36:
The Tukey post hoc tests showed that the between-group difference in energy scores was statistically significant only at Visit 1, where the OAA group reported lower energy levels (mean ± SE: 10 ± 2.3) than the control group (18 ± 2.1; p = 0.01) (Table 3).
Click to expand...

And the effect on the primary fatigue outcome was negligible:
Fatigue, as measured by the CFQ, was the primary outcome. There were no statistically significant differences in improvement between the OAA and control groups for any individual CFQ item or for the CFQ total score (Table 2). The between-group effect size (Cohen’s d) for the change in the total CFQ score from Visit 1 to Visit 3 was −0.093, indicating a negligible and non-significant difference in fatigue reduction between the groups.
Click to expand...
 
forestglip said:
You quoted this related part, but I want to highlight it:

Control group started off with a significantly higher energy level as measured by RAND-36:


And the effect on the primary fatigue outcome was negligible:
Click to expand...
Thank you! I missed that finding, I just pasted all sections with «between-group» and got fed up reading them after one or two.

Makes it even worse then..
 
Yann04 said:
It’s very frustrating for me to see people I genuinely care about all over bluesky starting to worry about saving up to try this supplement. I don’t have the energy to argue-explain to every one of them that this study isn’t very good…
Click to expand...
Would linking to this thread have any effect? This is exactly what I was worried would happen too. I’ve had several pwME I know privately message me since they saw the results, but I don’t have social media myself
 
jnmaciuch said:
Would linking to this thread have any effect? This is exactly what I was worried would happen too. I’ve had several pwME I know privately message me since they saw the results, but I don’t have social media myself
Click to expand...
I strongly doubt it.

I think linking to the thread on the ME/CFS study would probably be even more meaningful (https://www.s4me.info/threads/resto...atigue-syndrome-me-cfs-2024-cash-et-al.40538/), but my impression has been that once people have believed something it doesn't matter what some weird avatars on some obscure forum that people have never seen before are writing independent of whether their arguments have scientific validity.
 
Last edited:
You must log in or register to reply here.
Back
Top Bottom