Novel association analysis between HLA-DQB1 polymorphisms and rectal cancer based on a cross-validation design

Genetic risk analysis for an individual according to the theory of programmed onset, illustrated by lung and liver cancers

OBJECTIVE

To investigate cancer association in the genome and the genetic risk of death from major cancers according to the theory of programmed onset for an individual.

METHODS

Alleles of 15 randomly selected short tandem repeat (STR) loci, including D6S1043, D12S391, CSF1PO, D7S820, D2S1338, D3S1358, vWA, FGA, D8S1179, D21S11, D18S51, D5S818, D13S317, D16S539 and D19S433, were determined in 50 patients with lung cancer and 50 patients with liver cancer. The onset age of patients with and without the alleles was compared with Cox regression. Frequencies of significant alleles from Cox regression between the cancer group and control population were analysed through logistic regression for cross-validation. The death probability in an individual carrying two or one of two cancer-related alleles or not carrying two cancer-related alleles was calculated with outcomes of case-control studies translated into the results of the cohort studies.

RESULTS

It was confirmed that D18S51-20 was a lung cancer-related allele and that D21S11-30.2 and D6S1043-18 were liver cancer-related alleles. Probabilities of death from lung or liver cancers ranged from 0.115 to 0.395, respectively, for those who carry and/or do not carry D18S51-20, D21S11-30.2, D6S1043-18.

CONCLUSIONS

A more efficient method could be devised for genetic risk analysis according to the theory of programmed onset. The analysis of the CODIS-STR loci (STR loci listed in the US combined DNA indexing system) as genetic markers may provide an efficient and reliable approach to estimate an individual's genetic predisposition.