Fri, Dec 26, 2014, 2:22 PM EST - U.S. Markets close in 1 hr 38 mins

Recent

% | $
Quotes you view appear here for quick access.

Exact Sciences Corporation Message Board

  • techtrader79 techtrader79 Apr 8, 2013 12:32 PM Flag

    Can someone explain the concerns about Sensitivity in the trial?

    I see this being mentioned in a couple posts. Is this new? I thought the sensitivity and specificity were confirmed in a verification trial last year. Thanks.

    SortNewest  |  Oldest  |  Most Replied Expand all replies
    • There is no data out on the trial. Therfore any concerns you see are idle speculation from posters who don't present rational arguments as to why there will be a drop in sensitivity. Could it happen ? Yes. Is it likely, no. If you compare the last three validation studies the sensitivity numbers were very consistent and averaged around 58% for adenomas. If you look at the 95% confidence intervals then you would predict a lower limit of 45% (although that is unlikley). As far as the overall sensitivity to all cancers that only has to hit 70% - a target that should be easy to reach given that 85% was attained in the previous trial. At this point, without additional information it is just idle speculation. We will know soon enough since the CEO guided to March-April for results.

      There are arguments that could be presented that would send the sensitivity higher or lower in the trial. But at this point, what is the use of coming up with hypotheticals since the real data will be out any day.

    • Specificity is "locked-down" for the Deep-C trial at 90%.
      Corporate guidance for Sensitivity of Cancer is 85%.

      This is ALL very well known by anyone that has listened to a presentation by the CEO or CFO, or anyone who has listened to the recent quarterly CC late last month.

      You sound like a SHORT.

      • 2 Replies to sailingaway82
      • specificity was actually 87%; not 90% from Deep-C results; this means that on average 13% of positives from cologuard will actually be negative by colonoscopy. I would love to know how many of the false positives from cologuard actually eventually test positive; implication that colonoscopy has false positives too.

      • specificity = true negatives / (true negatives + false positives), making it a measured quantity, so I don't understand how it can be locked-down.

        The cut-off study locks down the threshold fluorescent intensity level above which a marker is considered positive and any interrelations between markers, etc.

        Once approved, subsequent improvements to the test will not require FDA approval, so it would be interesting to see if Exact uses the DeeP-C samples to find more optimal cut-offs.