My thoughts about the Viraleze trial results:
First of all, a word about p value and statistical significance. This has nothing directly to do with percentage successfully treated, or anything like that. It is a statistical measure of confidence in the findings of a trial. The P stands for probability and measures how likely it is that any observed difference between groups is due to chance. P = 0.05 (95% confidence that the difference is not due to chance), or less, is generally accepted as 'statistically significant'.
If there is a real, but small, difference between two groups, then larger trial parcipitant numbers will be required in a trial to get a p value less than 0.05. This is known as a study having sufficient statistical power to have statistical confidence in the findings.
The Viraleze trial did not have enough participants to generate statistically significant findings (for or against efficacy) in the younger age groups but the results were better in the viraleze group. Just not reaching p value of less than 0.05. They don't say the p value - we will see in the future. (As an example, if the p value is 0.1, this means a 90% chance that the difference is 'real' and not due to chance but it is not considered statistically significant. 95% confidence is the standard required).
If there is a real, but large, difference between two groups, then smaller numbers in a trial can still achieve statistical significance. This trial had sufficient numbers of people to generate statistically significant findings (for efficacy) in the older cohorts due to a greater difference in viral clearance in the treated group versus placebo than in younger cohorts.
If you combine data from different groups with different types of outcomes in a study (ie. a defined population group with statistically significant differences - low p value - with a defined population group which has a higher p value) then the overall p value goes up, due to averaging. The 2nd group's results have 'diluted' the results of the first groups. This is what happened in the reporting of results of this trial, with the p value of the entire group going above 0.05 when younger group's data was combined with older group's data AT DAY 7 (see comments below).
It is not 'cherry-picking' to do sub-group analysis. It is the right thing to do to find out what is happening in the different groups. In this trial's case, different age groups, as we all know that immunity to respiratory viruses diminishes significantly with age and it is important to look at any differences in efficacy. The sub-group analysis also clearly shows a clear trend of increasing effect (difference between the viraleze-treated and placebo groups) as the age goes up (look at the decreasing p value reported as age increases). Again, AT DAY 7.
The wording of Starpharma's announcement to the market is poor, as it leaves investors having to guess as to the significance of the viral load reduction in all groups / younger groups. As most people don't have a background in statistics, I think they could have explained things a bit better to the market. People have read 'not statistically significant' for all groups (at day seven) and this has caused unnecessary worry.
Regarding the trial announcement:
The announcement makes it clear that the viral load reduction was statistically clinically significant in groups aged > 45 years old, at the 7-day primary endpoint, and that the benefit of viraleze was more pronounced in the older age groups.
Regarding the older groups, the announcement states:
Less than 24 hours after starting dosing, Viraleze™ reduced SARS-CoV-2 viral load in the nose by 80% from baseline in participants aged 65 and over (N=50, p=0.008). In contrast, viral load in the placebo arm for this age group remained unchanged from baseline to day 2.
AND:
Viraleze™ achieved a statistically significantly lower level of SARS-CoV-2 viral load in the nose over the 7-day treatment period, which was the primary endpoint of the study, than a placebo nasal spray in the cohort of participants aged 45 years and over (N=118, p=0.017).
So, with the older cohorts, there is a clear advantage demonstrated at the end of the 7-day treatment period, the pre-set endpoint of the trial. That is clear.
Statements in the announcement regarding viral load reduction in all (including younger) groups, however, initially appear to contradict each-other.
Regarding all age groups, they state:
Viraleze™ reduced viral load in the full study population including all patient age groups (N=197), although the difference vs placebo was not statistically significant.
THEY THEN SAY:
The study also demonstrated a more rapid rate of viral clearance for Viraleze™, leading to an approximately 66% greater reduction in viral load compared with placebo between days 2 and 6 of treatment in participants of all ages, and the difference was statistically significant (-0.66 vs -0.54 logs per day, respectively, p=0.035, N=197).
With the limited information given in the announcement to the market, the only way I can interpret this apparent contradiction is that, between day 2 and 6, the reduction in viral load was greater in the viraleze treated group in all age groups (AND STATISTICALLY SIGNIFICANT) but, by day 7, in the younger groups, due to their more healthy immune system than the older groups, the placebo group's viral clearance had almost 'caught up' to the viraleze-treated group before, or by, day 7. (Making the difference no longer statistically significant by day 7 - the pre-defined primary endpoint of the trial).
This would not be surprising: that younger (predominantly vaccinated - having antibodies against COVID) people are able to more rapidly substantively clear the virus, on their own (ie. without help from viraleze, the placebo group) by day 7 of the trial (which would, for example, equate to day 11 of their illness if they had enrolled on day 4 of their illness). (They do not specify the average duration of illness prior to starting the trial but this would realistic to expect that people would take 3-4 days duration to: present to outpatients in the NHS, test positive, and enrol in the trial).
If only they had provided a graph! And some clearer explanation in the announcement.
If this is correct (and I can only go on what has been released) a comparison graph of viral load on the Y axis and time on the X axis, of all age groups, would show a steeper descending curve in the viraleze treated group in the early days of treatment, flattening out around day 6, and a less steep descending curve in the placebo-treated group, approaching, almost meeting (but not touching), the viraleze-treated group curve by day 7. That is, statistically significant better viral clearance at day 6, and still better, but NO LONGER statistically significant, viral clearance by day 7, the pre-appointed end point of the trial.
Anyway, as I said above, if only they had provided a clearer explanation! And a graph/graphs.
More precise details will come to us in time. And, hopefully, some graphs!
Overall, the study is a success, as it does demonstrate clearly that Viraleze has an antiviral effect in-vivo (reduces viral load) in the nose, where respiratory viruses get started / establish their infection in the body.
It really is remarkable that a nasal spray can demonstrate some reduction in symptoms when many of the people will have already had the illness for a few days, at least, before entering the trial, WHEN THE HORSE HAS ALREADY BOLTED (the virus infection is already well-established in the respiratory tract and the body).
Logic would suggest that it is HIGHLY LIKELY that, from these results, there should be a greater benefit from this spray WHEN USED AS PREVENTION, OR WHEN USED EARLIER IN THE COURSE OF THE ILLNESS (eg. at first symptom onset, not days later).
Before I get shouted down by others for stating this, yes, it is true that EFFICACY OF VIRALEZE AS A PREVENTIVE HAS NOT YET BEEN DEMONSTRATED OR PROVEN IN A CLINICAL TRIAL but, I think that any reasonable, logical thinking person would acknowledge it is highly likely. Certainly, the health professsional experts quoted believe the trial results are highly encouraging for viraleze to be used to reduce risk of infection, reduce transmission and reduce symptoms . Please read their comments again, if you are not sure about that.
'A stitch in time, saves nine' definitely applies to infectious diseases of all types and many illnesses are more successfully prevented, or their severity much reduced, by prophylactic (preventive), or early use of antibiotics/antivirals, in contrast with starting treatment later, once an infection has already become established (the horse has bolted). This is a theme that is seen time and time again through-out clinical medicine and I am confident it should also apply to both preventive and/or earlier use of viraleze.
I do wish that the company had planned and executed a proper prevention trial ages ago but they didn't. They chose, instead, to do this study first, which is more of a statistically under-powered (in the case of younger groups) 'treatment' trial. I understand there are some good reasons they took this pathway instead but it still disappoints me. That being said, I do believe that this study should, at least, help Starpharma in some ways with their product claims of efficacy against the virus in the human nose.
Some other thoughts and comments:
It is interesting to compare this trial result to SaNOtize trial.
Both the SaNOtize and the Viraleze trials are a great success. They demonstrate that their antiviral nasal sprays successfully reduced viral load in the nose and did also alleviate symptoms.
The two trials are different enough that care needs to be taken in direct comparison regarding efficacy. On the face of it, SaNOtize was more effective at reducing viral load. This may ultimately turn out to be the case, but their numbers were smaller and a different dosing schedule was used, so we are comparing apples with oranges. Also, we don't know the average number of days that people had had symptoms between the two trials. I believe this is crucial, when looking at symptoms and viral clearance, as described above.
Only a head-to-head trial, at doses that have been chosen as 'treatment doses' (higher doses) if it is a treatment trial, or 'prevention doses' (which would likely be lower) could determine the better product for each indication (prevention, or treatment).
Starpharma's trial was not designed for such a purpose (to prove it as a treatment - ie. a medicine). That would require further research of different dosing regimens for treatment of covid and a larger, more expensive trial. This trial was done to explore/demonstrate if the product can successfully block/reduce viral load in the human nose to support its claims as a medical device to block virus in the nose. It succeeded in demonstrating this. As did SaNOtize in their trial.
SaNOtize and Viraleze have only been tested head-to-head in a pre-clinical trial. In genetically altered mice with 'humanised' respiratory tracts, Viraleze was found to be far superior to nitric oxide in that study (see RESPI-DARTT presentation).
So, we have conflicting information as to which product may be better for prevention, or treatment. I suspect they will both be effective. At this stage, we just don't know if one is better than the other.
Regards
Gumnut
