Friday, 29 July 2016

CrowdSpeak: Why am I a bottleneck?

Thank you we now have our power calculations for the Charcot 2 project? #MSResearch #CrowdSpeak #Crowdacure #MSBlog

"David Holden successfully completed all the lab work on the saliva samples and sent me the spreadsheet several weeks ago. I only managed to analyse them two days ago. We have achieved what we wanted to with the samples and now know how many people with MS shed virus in their saliva. This has allowed me to do two sets of power calculations for the proposed trial of our antiviral drug in pwMS."

"This is the summary, or headline, results. Before reading them I need to explain what some of the jargon or stats speak is. In summary sample size determination is the process of defining the number of observations, or replicates, to include in a clinical trial. The sample size is an important feature of any clinical study in which the goal is to make inferences about the population from which the study subjects are recruited. In practice, the sample size is critical to make sure we have sufficient statistical power to answer the question in hand; for example in our proposed study, do we have enough subjects to confident that our drug works in suppressing lytic EBV infection in the salivary glands of pwMS. Although the sample size calculation uses real data we have to make some basic assumptions, for example how effective we expect the drug to be when tested in pwMS. We also have to set other variables, i.e. the actual power of the study and the so called alpha. Traditionally in medical trials the power of studies is set at 80% or 90%, i.e. we have an 80% or 90% chance based on the number of subjects studied to get a positive result. In comparison, the alpha sets the level of significance we are prepared to accept; we typically set this at 5% using a one- or two-sided comparison. This means that if we get a significant result there is a 5% chance of the result being a false positive result. False positive results are not that uncommon in research, which is why we have to reproduce results in second, or follow-on, studies. Please also note the power calculations, i.e. the size of the study, determines its cost."


CrowdaCure EBV Viral Shedding Trial Power Calculation Output


Proposed Power Calculations for a 6-month study
In the INSPIRE samples, which collected monthly saliva for 6 months, 9 out of 20 subjects with MS (45%) shed EBV, i.e. had a detectable EBV virus in their saliva, on at least one occasion.  Based on this data we are now able to do a power calculation of a trial to assess the ability of an antiviral drug to suppress EBV lytic reactivation in the salivary glands.


Power calculation 1: Based on the conservative assumption that the antiviral drug will only be 75% effective in suppressing EBV reactivation over 6 months with monthly salivary sampling we will need at least 29 trial subjects per arm. This calculation is based on having a power of 90% and an alpha of 5%.


Power calculation 2: Based on the conservative assumption that the antiviral drug will only be 80% effective in suppressing EBV reactivation over 6 months with monthly salivary sampling we will need at least 18 trial subjects per arm. This calculation is based on having a power of 80% and an alpha of 5%.


Proposed Power Calculations for a 3-month study
In the Sheffield cohort, which collected monthly samples for 3 months, 44 out of 119 subjects with MS (37%) shed EBV, i.e. had a detectable EBV virus in their saliva, on at least one occasion.  Based on this data we are now able to do a power calculation of a trial to assess the ability of an antiviral drug to suppress EBV lytic reactivation in the salivary glands.


Power calculation 1: Based on the conservative assumption that the antiviral drug will only be 75% effective in suppressing EBV reactivation over 3 months, with monthly salivary sampling, we will need at least 38 trial subjects per arm. This calculation is based on having a power of 90% and an alpha of 5%.


Power calculation 2: Based on the conservative assumption that the antiviral drug will only be 80% effective in suppressing EBV reactivation over 3 months, with monthly salivary sampling, we will need at least 24 trial subjects per arm. This calculation is based on having a power of 80% and an alpha of 5%.

"In the Charcot Project 2 we are using viral shedding of EBV as a readout for antiviral drugs targeting EBV. If our lead drug Famciclovir suppresses virus in saliva then we will be confident of it also being suitable as a treatment for infectious mononucleosis. This hasn't escaped our attention, nor has the case report below. This is why IM is one of our major focuses going forward and forms part of Charcot 3."

"Some of you may ask why are you behind schedule in getting these results out? It is my fault I have too many balls in the air and not enough bandwidth or time to do everything I am involved in. One of my team has called me the bottleneck in the system. When we have our annual research strategy meeting in a few weeks time I am going to hopefully address this problem. May be I should be called Prof. Bottleneck rather than Prof G?"


"Once more, thank you for your support. The money raised has give us the necessary results that we can now use to develop a protocol for our next study. I sincerely hope we will now be able to leverage these results to get funding for our study."



Click here to find out more!


Goldani. Treatment of severe infectious mononucleosis with famciclovir. J Infect. 2002 Feb;44(2):92-3.

We report a patient with severe acute infectious mononucleosis who was successfully treated with famciclovir. A 15-year-old male was admitted with a 6-week history of fever, malaise, generalized lymphadenopathy, and hepatosplenomegaly, the patient was acutely ill with a temperature of 39.0 degrees C. Oropharyngeal examination revealed enlarged tonsils partially obstructing the airways. EBV serology obtained during admission showed a positive Monospot test, virus capsid antigen IgM, 1:320, Epstein-Barr nuclear and early antigen, negative. After 72 hours of treatment with famciclovir(500 mg t.i.d.), the patient was afebrile with important regression of the lymphadenopathy, enlarged tonsils and hepatosplenomegaly. Because acute infectious mononucleosis may be associated with extensive and prolonged disease, the potential therapeutic role of famciclovir in the treatment of severe forms of the disease deserves further studies.

CoI: Team G are the recipients of a grant from Crowdacure that was used to perform this research.

19 comments:

  1. Prof G why don't you advertise for volunteers to help you with your workload? I am sure medical students would be keen to help and there are a lot of untapped skills in pwMS who are unable to work because of of premature medical retirement. I would be willing to help.

    ReplyDelete
    Replies
    1. Thanks for offering! Using volunteers to help is a good idea.

      Delete
  2. I sincerely hope that you are considering funding Charcot 2 via the same route. I is great to be part of something real. Thank you for your efforts and commitment. You have convinced me that EBV is the cause of MS. Finally, don't worry about being a bottleneck; you bottle seems to have a very wide neck.

    ReplyDelete
    Replies
    1. Can't agree more!
      P.S.
      I'm donor too

      Delete
  3. Hi Prof. G,

    I understand bottlenecks. My neck extends upwards about 75 ft :)

    I seem always have a, "This is how much I need accomplish today" followed at the end of the day with a, "Well I managed get that and that done... But..."

    I am happy (I think LOL) to hear the study produced results.

    I'll get the update online. As moves ahead towards the costing and such we are of course available as a resource that is at your disposal as always.

    ReplyDelete
  4. I am the fourth in my familiy to get MS. My daughter has had a severe case of infectious mononucleosis. She doesn't have MS. Please hurry.

    ReplyDelete
    Replies
    1. Yes. Agree. I have 3 kids. There isn't a day go by that I'm fearful about their future. You have my full support (financial and othereise).

      Delete
  5. Better a bottle of pure water than a barrel of contaminated stuff!

    ReplyDelete
  6. The 3 or 4 winters leading up to my first relapses were dominated by repeated episodes of swollen glands/chills/sore throats/flu like symptoms. Month long episodes of viral chills in the afternoons and again in the evenings without any other symptoms. All most peculiar, it would start as soon as the temperature dropped in the autumn. It all stopped with dx though. Vit D was 19. Haven't had these feelings since and wonder if it was the Vit D issue or something else.

    ReplyDelete
  7. Do you think that treatment of EBV/IM will reduce MS activity in those with progressive disease, also where "relapse" activity is low?

    ReplyDelete
    Replies
    1. yes..is any use to people who already got it ???...is this a No both in theory and in practice so far..?

      Delete
  8. prof g can you post a bit more on how you came about the patient numbers? it seems a bit low given repeated measures.

    ReplyDelete
    Replies
    1. It is not repeated measures; study subjects were only counted once. In other words if you shed once, twice or more in 3 or 6 months you were classed as a shedder. For the drug to work it has to convert a large proportion of these subject to non-shedders, i.e. every saliva sample has to be below the detection threshold of the assay.

      Delete
    2. As less than 50% of study subjects are non-shedders they will be non-informative and not affect the results. Only the EBV shedders' will be informative.

      Delete
    3. But you can't compare shedders at 1 month vs shedders at 3 months. You can only compare like with like so you should use the numbers by time period

      Delete
    4. We could use a survival analysis, which will deal with this issue. But as shedding is intermittent this is not really appropriate. Our design is based on a biological question;'does FamV inhibit lytic EBV reactivation'? If you have a better trial design we would be interested to hear about it; particularly if it reduces the sample size and cost. Thanks.

      Delete
  9. Can you plot out everyone's profile- x axis time, y shedding or not, line for each individual? Thanks

    ReplyDelete
  10. Honestly if EBV isn't the sole cause of MS, at least can be one of the causes. There are cases of MS in families, but as far as I read come to 20% of cases. I am, for exemple, just 01 cases in the whole history of my family, my family in general have cardiovascular problems.

    Sometimes I wonder if cases of MS among families also would fall in cases of subnotification, or misdiagnosis involving other diseases, for exemple ALD, or NMO, or Human T-cell lymphotropic virus type 1 (HTLV-1) -associated myelopathy, etc.

    In my city of almost 1.000 cases followed only know of 01 among um family, an aunt and a niece.
    The rest of the cases are isolated, at the moment.


    Thoughts...

    ReplyDelete
  11. Why are less than 50% of study subjects non-shedders?
    Have they stopped being shedders?

    ReplyDelete

Please note that all comments are moderated and any personal or marketing-related submissions will not be shown.