Investigating how variants in this gene, DNA ligase IV (involved in double-stranded break repair in antigen gene rearrangement mechanisms — which are thought to be involved in lymphoproliferative disorders), these authors found that A3V and T9I variants have a protective association with multiple myeloma. They also examined adult acute lymphoblastoma cases and lymphoma cases, in which they found no significant correlations.
Among the multiple myeloma cases, case A/A: 247, A/V: 20, V/V: 1, and in controls A/A: 189, A/V: 31, V/V: 0. Counting carriers, this is case+: 21, case-: 247, control+: 31, control-: 189. Using a two-tailed Fisher’s Exact test this is p = 0.0277.
The authors report a “significant interaction (p < 0.05)” using the chi-squared test, but do not mention correcting for multiple hypotheses — at least three different disease hypotheses and two variants were being tested here, and so this correction would imply the required p-value should be more like 0.025 (for 2 hypotheses) to 0.008 (for 6 hypotheses). Based on this, we treat this variant as having an effective p-value between 0.1 and 0.05.