Warning: Genome-wide association studies can be misleading. An example in hepatology

Authors

  • Pierre Galichon M.D.,

    1. Institut National de la Santé et de la Recherche Médicale, Unité 702, Paris, France
    2. Université Pierre et Marie Curie, Universite de Paris 6, Paris, France
    3. Emergency Nephrology and Renal Transplantation, Hôpital Tenon, Assistance Publique des Hôpitaux de Paris, Paris, France
    Search for more papers by this author
  • Alexandre Hertig M.D., Ph.D.,

    1. Institut National de la Santé et de la Recherche Médicale, Unité 702, Paris, France
    2. Université Pierre et Marie Curie, Universite de Paris 6, Paris, France
    3. Emergency Nephrology and Renal Transplantation, Hôpital Tenon, Assistance Publique des Hôpitaux de Paris, Paris, France
    Search for more papers by this author
  • Eric Rondeau M.D., Ph.D.,

    1. Institut National de la Santé et de la Recherche Médicale, Unité 702, Paris, France
    2. Université Pierre et Marie Curie, Universite de Paris 6, Paris, France
    3. Emergency Nephrology and Renal Transplantation, Hôpital Tenon, Assistance Publique des Hôpitaux de Paris, Paris, France
    Search for more papers by this author
  • Laurent Mesnard M.D., Ph.D.

    1. Institut National de la Santé et de la Recherche Médicale, Unité 702, Paris, France
    2. Université Pierre et Marie Curie, Universite de Paris 6, Paris, France
    3. Emergency Nephrology and Renal Transplantation, Hôpital Tenon, Assistance Publique des Hôpitaux de Paris, Paris, France
    Search for more papers by this author

  • Potential conflict of interest: Nothing to report.

Warning: Genome-wide Association Studies Can Be Misleading. An Example in Hepatology

To the Editor:

We read with interest the article by Clifford and colleagues in HEPATOLOGY.1 This well-designed study identified genetic factors associated with hepatocellular carcinoma (HCC) or liver cirrhosis (LC). Their analysis isolated a single-nucleotide polymorphism (SNP) for the TPTE2 gene, a PTEN (phosphatase and tensin) homolog encoded by chromosome 13, that differentiates HCC and LC.

As for ourselves, we identified sequences near the TPTE2 gene that are replicated in the genome, flawing the interpretation of a genome-wide association study. We have inputted the flanking sequence of the aforementioned SNP rs2880301 (CTTTGCAGCAATCCAG [C/T] CTAAAAGCCTAAAAGC) in the Basic Local Alignment Search Tool (BLAST) from the National Center for Biotechnology Information website. Strikingly, we found total homology with a nucleotide sequence located both on chromosome 13 and the Y chromosome. To rule out that the association found by Clifford et al. was due to a major bias in the sex distribution of the patients affected by HCC or LC, we reanalyzed the raw data. There was a strong disequilibrium of sex between the two disease states: thus, the male/female sex ratio was 3.4 in HCC and 1.3 in LC. We found that the value of SNP rs2880301 was associated with sex rather than with disease status (Fig. 1).

Figure 1.

Values for rs2880301 in male and female patients. Triangles are patients with HCC. Squares are patients with LC. SNP values vary with sex, not with disease state.

We thus believe that the data from this genome-wide association study does not encourage further study of TPTE2 in these diseases.

Pierre Galichon M.D.* † ‡, Alexandre Hertig M.D., Ph.D.* † ‡, Eric Rondeau M.D., Ph.D.* † ‡, Laurent Mesnard* † ‡, * Institut National de la Santé et de la Recherche Médicale, Unité 702, Paris, France, † Université Pierre et Marie Curie, Universite de Paris 6, Paris, France, ‡ Emergency Nephrology and Renal Transplantation, Hôpital Tenon, Assistance Publique des Hôpitaux de Paris, Paris, France.

Ancillary