En Cheng, MD, PhD

Jeffrey ­Meyerhardt, MD, MPH, FASCO

In a study reported in JAMA Oncology, En Cheng, MD, PhD, of Kaiser Permanente Northern California, Jeffrey ­Meyerhardt, MD, MPH, FASCO, of Dana-Farber Cancer Institute, and colleagues found that higher levels of inflammatory biomarkers measured after surgery but before starting chemotherapy were associated with poorer survival in patients with stage III colon cancer from the CALGB/SWOG 80702 adjuvant chemotherapy trial.¹ Dr. ­Meyerhardt is the corresponding author of the JAMA Oncology article.

Study Details

In the trial, patients were randomly assigned in a 2 × 2 design to receive adjuvant (1) fluorouracil, leucovorin, and oxaliplatin for 3 months vs 6 months, and (2) daily celecoxib (COX-2 inhibitor, a selective nonsteroidal anti-inflammatory drug [NSAID]) vs placebo for 3 years between June 2010 and November 2015; follow-up was through August 2020. No significant difference in survival among groups was observed, with patients thus being combined for the current analysis.

A total of 1,494 patients with plasma samples available for inflammatory biomarker assays were included in the analysis. Levels of interleukin 6 (IL-6), soluble tumor necrosis factor α receptor 2 (TNF-αR2), and high-sensitivity C-reactive protein (CRP) were measured 3 to 8 weeks after surgery but prior to chemotherapy. The primary outcome was disease-free survival, a composite outcome of colon cancer recurrence or death from any cause. Analyses were adjusted for clinical and pathologic characteristics, including trial randomization (celecoxib vs placebo).

Inflammatory Biomarker Levels

Plasma samples were collected at a median of 6.9 weeks (interquartile range [IQR] = 5.6–8.1 weeks) after surgery and 6 days (IQR = 3–12 days) before chemotherapy randomization. Median plasma concentrations were 3.8 pg/mL (IQR = 2.3–6.2 pg/mL) for IL-6, 2.9 × 10³ pg/mL (IQR = 2.3–3.6 × 10³ pg/mL) for soluble TNF-αR2, and 2.6 mg/L (IQR = 1.2–5.6 mg/L) for high-sensitivity CRP. Correlations (r) were 0.40 (P < .001) between IL-6 and soluble TNF-αR2, 0.55 (P < .001) between IL-6 and high-sensitivity CRP, and 0.30 (P < .001) between soluble TNF-αR2 and high-sensitivity CRP. For each biomarker, compared with patients in the lowest quintile for each biomarker, those in the highest quintile of IL-6 levels were more likely to be older, non-Hispanic, and to have worse Eastern Cooperative Oncology Group (ECOG) performance status, more right-sided tumors, and higher body mass index (BMI); those in the highest quintile of soluble TNF-αR2 levels were more likely to be older, White and non-Hispanic, and to have worse ECOG performance status, higher T category, more right-sided tumors, and higher BMI; those in the highest quintile of high-sensitivity CRP levels were more likely to be Black and to have worse ECOG performance status, more right-sided tumors, and higher BMI.

Association With Disease-Free Survival

Median follow-up was 5.9 years (IQR = 4.7–6.1 years). Compared with patients with biomarker levels in the lowest quintile, those with levels in the highest quintile had significantly poorer disease-free survival for IL-6 (adjusted hazard ratio [HR] = 1.52, 95% confidence interval [CI] = 1.07–2.14,  P = .01 for trend), soluble TNF-αR2 (adjusted HR = 1.77, 95% CI = 1.23–2.55, P < .001 for trend), and high-sensitivity CRP (adjusted HR = 1.65, 95% CI = 1.17–2.34, P = .006 for trend). The estimated differences in 5-year rates between lowest and highest quintiles ranged from 12.9% for high-sensitivity CRP to 17.6% for soluble TNF-αR2.

Association With Recurrence-Free and Overall Survival

For recurrence-free survival, adjusted hazard ratios for the highest vs lowest quintile were 1.68 (95% CI = 1.07–2.65, P = .08 for trend) for IL-6, 1.54 (95% CI = 1.04–2.28, P = .001 for trend) for soluble TNF-αR2, and 1.73 (95% CI = 1.11–2.68, P = .03 for trend) for high-sensitivity CRP. For overall survival, adjusted hazard ratios for the highest vs lowest quintile were 1.54 (95% CI = 1.05–2.25, P = .03 for trend) for IL-6, 2.33 (95% CI = 1.40–3.89, P = .02 for trend) for soluble TNF-αR2, and 1.56 (95% CI = 1.07–2.26, P = .003 for trend) for high-sensitivity CRP.

Subgroup Analyses

No significant interactions for disease-free survival were observed between any inflammatory marker and celecoxib vs placebo randomization (significance level of P < .006 to account for multiple comparisons). However, a significant interaction between IL-6 and baseline low-dose aspirin use was observed (P = .002). Adjusted hazard ratios for high vs low IL-6 were 1.19 (95% CI = 1.09–1.29) among those not taking low-dose aspirin and 0.91 (95% CI = 0.78–1.05) among those taking low-dose aspirin.

The investigators concluded: “This cohort study found that higher inflammation after diagnosis was significantly associated with worse survival outcomes among patients with stage III colon cancer. This finding warrants further investigation to evaluate whether anti-inflammatory interventions may improve colon cancer outcomes.” 

DISCLOSURE: The study was supported by the National Cancer Institute, Pfizer, and others. Dr. Cheng reported no conflicts of interest. Dr. Meyerhardt has received personal fees from Merck outside the submitted work and institutional research funding from Boston Biomedical; and has served as an advisor or consultant to Cota Healthcare and Merck Pharmaceutical.

REFERENCE

1. Cheng E, Shi Q, Shields AF, et al: Association of inflammatory biomarkers with survival among patients with stage III colon cancer. JAMA Oncol 9:404-413, 2023.