Table 1.
Associations of blood inflammatory markers with cancer outcomes.
| Cancer | Biomarker | N Participants/n Cases | Countries | Design | Mean of Median Age at Inclusion | Effect Size | Heterogeneity and Bias (for Meta-Analysis) |
Follow-Up | Reference |
|---|---|---|---|---|---|---|---|---|---|
| Meta-analyses focused on incidence | |||||||||
| Breast | CRP | 69,610/3522 | America (n = 6)/Europe (n = 5)/Asia (n = 1) | 12 studies | 49–73 y 9 studies restricted to post-menopausal women |
RR per doubling CRP concentration: 1.07 (95% CI: 1.02–1.12) 1.06 (1.01–1.11) for post-menopausal women |
I² = 47%, Ph = 0.04 Egger’s tests showed some evidence of publication or small study bias (p = 0.08) |
5–13 y | [35] |
| Breast | CRP | /5286 | America (n = 6) Europe (n = 6)/Asia (n = 3) | 15 cohorts + case-control studies | 45–73 y | Combined OR per natural log unit change in CRP: 1.16 (95% CI: 1.06–1.27) Only significant in persons >60 y (6 studies) |
I2 = 45.9% No evidence of publication bias, Begg’s (0.805) and Egger’s tests (0.172) |
[36] | |
| Colorectal | IL6 | 8420/1308 | USA (n = 5), UK (n = 2) | 3 cohorts and 3 nested case-control studies | summary RR per natural log unit change in IL6: 1.22 (95 % CI 1.00–1.49) Inverse association was noted for rectal cancer (RR 0.69; 95 % CI 0.54–0.88) |
Some evidence of heterogeneity (I2 = 46 %). No evidence of small study effect, Egger’s p 0.77 I2 = 0 %. Evidence for small-study effects (p 0.02). |
[37] | ||
| Colorectal | CRP | 1159 cases/37,986 controls | USA (n = 3), Europe (n = 3), Japan (n = 2) | 3 cohort and 5 nested case-control studies | 53–73 y | Summary RR per one unit change in natural log-transformed high-sensitivity CRP: 1.13 (95% CI, 1.00–1.27) for colon cancer, and 1.06 (95% CI, 0.86–1.30) for rectal cancer Only significant in persons >60 y (3 studies) |
Ph 0.16 (colon) and 0.02 (rectal) | 5.5–14 y | [38] |
| All | CRP | 194,796/11,459 | USA (n = 5), Europe (n = 6) | 11 cohort studies | 45–73 y | Pooled HR per natural log unit change in CRP: 1.105 (95% (CI): 1.033–1.178) for all cancers 1.308 (95% CI: 1.097–1.519) for lung cancer Not significant for breast, prostate and colorectal separately Only significant in persons >60 y (n = 24,000) |
Substantial heterogeneity across studies (Ph = 0.000, I2 = 70.10%) | [39] | |
| Single studies focused on cancer mortality | |||||||||
| Colorectal | haptoglobin | 325,599/1467 | Sweden | Cohort | Mean (SD) 46 (14) | Adjusted HR (>1.2 vs. <0.9 g/L): 1.19; 95% CI: 1.01–1.41 | No significant association with CRP and albumin | 18 y (mean) | [40] |
| Breast | haptoglobin | 155,179/736 | Sweden | Cohort | Mean (SD) 50 (11) | Adjusted HR (>1.4 vs. <1.4 g/L): 1.27, 95% CI: 1.02–1.59 | No significant association with CRP and albumin | 18 y (mean) | [41] |
| Breast, lung, all | leukocytes | 143,748/3062 | USA | Cohort | 63 (50–79) | Adjusted HR: (higher vs. lower quartile) 2.16 (1.33–3.50) for breast 1.65 (1.29–2.12) for lung 1.33 (1.17–1.51) for all |
Not significant for endometrial and colorectal | 8 y (mean) | [42] |
| All | CRP, IL6, TNF | 2438 | USA | Cohort | 73 (70–79) | Adjusted HR (1-unit increase on natural log scale): 1.63 (1.19–2.23) for IL-6 1.64 (1.20–2.24) for CRP 1.82 (1.14–2.92) for TNF |
5.5 y (mean) | [43] | |
| All | CRP, albumin, neutrophils | 160,481/13,173 | UK | Cohort | 35% > 65 y | Adjusted HR 1.85 for CRP > vs. <10 mg/L (p < 0.001) 2.08 for albumin < vs. >35 g/L (p < 0.001) Neutrophils > vs. < 7.5 × 109/L (p < 0.001) |
69 months (mean) | [44] | |
| Prostate | Leukocytes, neutrophils | 210,000/323 | UK | Cohort | Mean (SD) 57 (8) | Per one SD increase: HR = 1.14, 1.05–1.24 for leukocytes HR = 1.27, 1.09–1.48 for neutrophils |
6.8 y | [45] | |
| Colorectal | CRP | 16,000/92 | USA | Cohort | 50 | Adjusted HR = 3.96 (95% CI, 1.64–9.52) for levels >1.00 vs. <0.22 mg/dL | 14.2 y | [46] | |
CI—confidence interval. CRP—C reactive protein. HR—hazard ratio. IL6—interleukin-6. TNF—Tumor necrosis factor. OR—odds ratio. RR—relative risk. SD—standard deviation. UK—United Kingdom. USA—United States of America.