Serum levels of 25-hydroxyvitamin D and survival in Norwegian patients with cancer of breast, colon, lung, and lymphoma: a population-based study.
Cancer Causes Control. 2012 Feb;23(2):363-70. doi: 10.1007/s10552-011-9885-6.
Tretli S1, Schwartz GG, Torjesen PA, Robsahm TE.
1The Cancer Registry of Norway, Institute of Population-based Cancer Research, Majorstuen, 0304 Oslo, Norway. steinar.tretli@kreftregisteret.no
PURPOSE:
We investigated the association between serum levels of 25-hydroxyvitamin D (25-OHD) and risk of death in Norwegian cancer patients.
METHODS:
The study population was 658 patients with cancers of the breast (n = 251), colon (n = 52), lung (n = 210), and lymphoma (n = 145), obtained from JANUS, a population-based serum bank in Norway. Serum samples were collected within 90 days of cancer diagnosis and were analyzed for 25-OHD. Patients were diagnosed during 1984-2004 and were followed for death throughout 2008. We used Cox regression models to assess the relationship between serum 25-OHD and risk of death.
RESULTS:
Three hundred and ninety-nine patients died during follow-up, of whom 343 (86%) died from cancer. Adjusted for sex, age at diagnosis, and season of blood sampling, patients with 25-OHD levels below 46 nmol/L at diagnosis experienced shorter survival. Compared to patients in the lowest quartile of serum 25-OHD, the risk of cancer death among patients in the highest quartile was significantly reduced (HR 0.36 95% CI 0.27, 0.51). The estimated change in risk of cancer death was most pronounced between the first and the second quartile. The associations between 25-OHD levels and survival were observed for all four cancers.
CONCLUSIONS:
Higher circulating serum levels of 25-OHD were positively associated with the survival for cancers of the breast, colon, lung, and lymphoma.
PMID: 22193397
Table 4 Hazard ratio of dying from local versions of any of the cancers
| < 46 nmol/L | <18.4 ng | 1.0 (Reference) |
| 46–61 nmol/L | 18.4 – 24.6 ng | 0.39 |
| 62–81 nmol/L | 24.7– 32.4 ng | 0.34 |
| >81 nmol/ | > 32.4 ng | 0.15 |
| Confidence | p = 0.06 |
Table 5 Hazard ratio (HR) and all cause death
- Number of fatal cases/Number of patients
- Adjusted for sex, age at diagnosis and season of blood sampling