Background: The association between inflammatory disease states (e.g., hepatitis B, inflammatory bowel disease) and related carcinomas, and the potential protective effects of nonsteroidal anti-inflammatory drugs and aspirin against certain malignancies, has supported the idea that chronic inflammation may play a role in cancer development. However, studies examining various biomarkers such as C-reactive protein and fibrinogen as predictors of developing cancer have yielded conflicting results. Because the incidence of new cancer in the general population is relatively small, Margolis and colleagues used the large-scale, prospective Women's Health Initiative (WHI) to evaluate the potential association of an inflammatory marker with specific types of cancer that are prominent in this population.

The Study: The authors reviewed data from all WHI participants in the clinical trials and observational study arms. White blood cell (WBC) count and history of cancer was assessed before being enrolled in the WHI study. Participants were excluded from the current analysis if they had an initial history of any cancer except non-melanoma skin cancer, or if they had an abnormal baseline WBC count. Participants were contacted semiannually (clinical trial participants) or annually (observational study) to determine whether they had recently been diagnosed with a new cancer. Relevant clinical reports, discharge summaries, and death certificates were cross-checked by blinded evaluators to verify the patient self-reported information. The study focused on cancers with an annualized incidence of 0.1 percent or higher, namely invasive breast, colorectal, endometrial, and lung cancers.

Results: The analysis included 143,748 post-menopausal women 50 to 79 years of age, with 55 percent of participants belonging to the observational study arm. There were 4,639 cases of invasive breast cancer, 1,341 cases of colorectal cancer, 766 cases of endometrial cancer, and 1,237 cases of lung cancer developed during the study. Higher and lower WBC counts were significantly associated with different health or background characteristics (see accompanying table). After adjusting for age and potential confounders, women in the highest quartile of WBC count (6,800 to 15,000 per mm³ [6.80 to 15.00 ×10⁹ per L]) had significantly greater risks of developing invasive breast cancer (15 percent higher risk), colorectal cancer (19 percent), endometrial cancer (42 percent), and lung cancer (63 percent) than those in the lowest WBC quartile (2,500 to 4,790 per mm³ [2.50 to 4.79 × 10⁹ per L]).

When cancers diagnosed within the first two years of the study were excluded (to remove cancers that may have been present but undiagnosed at the beginning of the WHI), similar trends were observed. Overall cancer mortality was also positively correlated with increasing WBC quartile. Breast cancer among those in the highest WBC quartile was associated with a greater than twofold higher mortality rate, and lung cancer had a threefold greater mortality rate in this group. No significant associations between WBC quartiles and mortality were noted when colorectal or endometrial cancers were examined individually.

Conclusion: Higher WBC counts are associated with higher risks of invasive breast, colorectal, endometrial, and lung cancers among postmenopausal women, as well as a greater risk of lung, breast, and overall cancer mortality. Although the WHI study is a large-scale prospective study, the authors caution that further studies of this association using multiple WBC measurements might be useful before this can be applied clinically.