Re: The natural history of small-duct PSC - genetics

[Guest guest]

Hi Don,

Here an attempt to explain the statistical methods used in genetic

studies to estimate heritability of PSC.

The prevalence of PSC ranges from 8 to 14 per 100,000 persons in

Northern Europe and the United States (1). So this prevalance is

equal to about 0.008% to 0.014% of the general population in these

countries.

When the Swedish group reports that " First-degree relatives of

patients with PSC have a PSC prevalence of 0.7% (5/717). In siblings

the prevalence was 1.5% (4/269) " (2), this represents a nearly 100-

fold increased risk of developing PSC compared with the general

population. Stated simply, 0.7 to 1.5 approximately equals 0.008 x

100 to 0.014 x 100.

One way of expressing this data is the " heritability coefficient " ,

which equals the prevalence among siblings divided by the general

population prevalence. In the example above, this " heritability

coefficient " is about 100 (3).

For simple genetic disorders caused by single gene

defects, " heritability coefficients " range from several hundreds to

several thousands, whereas " heritability coefficients " in complex

diseases (those determined by several genes) are usually below 100

(3).

An example of a simple genetic disorder caused by a single gene

defect would be cystic fibrosis, caused by a mutation in the cystic

fibrosis transmembrane conductance regulator (CFTR) gene). However,

even this genetic disorder is proving to be a bit more complex ....

recently several additional genes have been identified which can

alter the course of the disease, or affect the severity of disease in

specific organs. These are termed " modifier " genes (4).

Inflammatory bowel disease is considered to be a " complex " disease,

determined by several genes. For comparison with PSC, consider that

for Crohn's disease the " heritability coefficient " is 15-35, and for

typical UC the " heritability coefficient " is about 6-9 (but please

bear in mind that UC in PSC may be different from typical UC, and may

actually be a third form of IBD) (3).

If you believe that Crohn's disease and UC have a strong genetic

component with several genes involved, and this is quite clear from

many recent publications; see e.g.

http://www.psc-literature.org/IBDSG.htm

then you would have to conclude that PSC is likely also caused by

several genes, but perhaps FEWER than Crohn's or typical UC.

It's this kind of analysis that supports the notion that genetic

factors are of importance for development of PSC ... and Dr.

Karlsen is hot on their trail:

http://www.psc-literature.org/KarlsenT.htm

and will update us on recent progress at the upcoming meeting in

ville, FL in early May.

(1) LaRusso NF, Shneider BL, Black D, Gores GJ, SP, Doo E,

Hoofnagle JH 2006 Primary sclerosing cholangitis: summary of a

workshop. Hepatology 44: 746-764.

(2) Bergquist A, Lindberg G, Saarinen S, Broome U 2005 Increased

prevalence of primary sclerosing cholangitis among first-degree

relatives. J. Hepatol. 42: 252-256.

(3) Karlsen TH, Schrumpf E, Boberg KM 2007 Genetic epidemiology of

primary sclerosing cholangitis. World J. Gastroenterol. 13: 5421-

5431.

(4) Accurso FJ, Sontag MK 2008 Gene modifiers in cystic fibrosis. J.

Clin. Invest. 118: 839-841.

Best regards,

Dave

(father of (22); PSC 07/03; UC 08/03)

> An understanding of statistical methods would probably help us.

When

studies that include statistics are cited, perhaps a brief

explanation could be given, in layman's term, if that is possible.