Summary of the 2018 Merck Manual Update

My analysis of the linguistic manipulation is summed up with the following mathematical certainty:

  • Where mutations LDLR, APOB, PCSK9, and p.Arg3558Cys are each greater than zero, LDLR alone cannot equal LDLR + APOB + PCSK9 + p.Arg3558Cys.
Familial hypercholesterolemia, LDLR, APOB and PCSK9, p.Arg3558Cys, FH patients

But the bad math, where LDLR does equal LDLR + APOB +PCSK9 + p.Arg3558Cys, has prevailed. Here is the logic leading up to the recent June 2018 “update” found in the Merck Manual.

The Chronology of the Mathematical Fracture in the Merck Manual’s FH Prevalence:

  1. In 2003 Daniel Rader explicitly defines FH, FDB, and FH3 according to the scientific record thus far, that is, separately, and with their respective prevalence rates.
  2. In 2007 Daniel Rader receives over 250,000 shares from Aegerion Pharmaceuticals, Inc.
  3. In 2011 Dr. Anne Goldberg and Dr. Daniel Rader co-author documents which explicitly combine APOB and PCSK9 with LDLR and declare a higher prevalence for a new version of “FH”:  1/300. No citations are provided that justify this number. However, with casual addition, LDLR + APOB + PCSK9 = 1:300 (rounded from 1/294). The names “FDB” and “FH3” are dropped from future, pharma-dominating FH papers.
  4. In 2012 the 1st Danish report furthers the conflation scheme and, after several shenanigans, FH is 1/200.
  5.  In 2016 the 2nd Danish report conflates names, adding data which exposes a patient swap. (I will cover this swap in a separate presentation.) FH is again said to be 1/200. Through my analysis and the removal of an inflated denominator, FH-as-LDLR is confirmed to be 1/500.
  6. In 2016 Regeneron adds in the controversial p.Arg3558Cys mutation. Now “FH” equals LDLR + APOB + PCSK9 + p.Arg3558Cys, without explanation. FH prevalence is now said to be 1 in ~250. Using the study’s own data, the prevalence of the originally defined FH – FH-as-LDLR – is confirmed to be 1/500 after linguistic integrity is restored.
  7. In 2018 Dr. Goldberg updates the Merck Manual. This is a disease manual and not an “FH” manual and so Dr. Goldberg must list all of the related diseases. The new version of “FH” must be linguistically “unpacked” back into the old categories once occupied by the APOB (“FDB”) and PCSK9 (“FH3”). Without the linguistic manipulation, the Danish and Regeneron reports confirm FH-as-LDLR to be 1/500. Without the 2011 conflation by Dr. Goldberg, FH prevalence is 1/500. So why is 1/200 used for FH-as-LDLR in the Merck Manual? By “unpacking” FH back into the pre-2011 categories, linguistically, FH is here once again exclusively defined by the LDLR, while FDB is listed immediately below – defined and counted separatelyby the presence of the APOB mutation. Nonetheless, FH-as-LDLR keeps the FDB-as-APOB and other disease quantities combined into the total.
  8. Thus, the equation behind FH prevalence in the Merck Manual, exposed by the historical record, is mathematically impossible LDLR = LDLR + APOB + PCSK9 + p.Arg3558Cys.