Increased neutrophil counts are associated with poor overall survival in patients with colorectal cancer: a five-year retrospective analysis

Abstract

Background: Colorectal cancer (CRC) continues to be a major health concern in today's world. Despite conflictive findings, evidence supports systemic inflammation's impact on CRC patients' survival rates. Therefore, this study aims to assess the prognostic role of the innate immune system in patients with CRC.

Method: A total of 449 patients were included, with a 5-year follow-up period, and absolute neutrophil counts and their related ratios were measured.

Results: The non-survival group had increased levels of white blood cells, neutrophils (both p<0.001), and monocytes (p=0.038), compared to the survival group, along with other neutrophil-related ratios. We observed increased mortality risk in patients in the highest tertile of white blood cells [HR=1.85 (1.09-3.13), p<0.05], neutrophils [HR=1.78 (95% CI: 1.07-2.96), p<0.05], and monocytes [HR=2.11 (95% CI: 1.22-3.63)], compared to the lowest tertile, after adjusting for all clinicopathological variables. Random forest analysis identified neutrophils as the most crucial variable in predicting survival rates, having an AUC of 0.712, considering all clinicopathological variables. A positive relationship between neutrophil counts and metastasis was observed when neutrophil counts are considered continuous (β=0.92 (0.41), p<0.05) and tumor size (width) when neutrophils were considered as logistic variable (T1 vs T3) [OR=1.42, (95% CI: 1.05-1.98), p<0.05].

Conclusion: This study offers comprehensive insights into the immune factors that impact the prognosis of CRC, emphasizing the need for personalized prognostic tools.

Keywords: colorectal cancer; inflammation; neutrophils; overall survival; prognosis.

Figure 1

Flow diagram of patient selection for inclusion in our study. A total of 623 participants were initially identified as potential candidates. Exclusions were made for the following reasons: (1) sixteen participants from 1997 to 2011 were excluded due to incomplete histopathologic diagnostic records; (2) fifty patients were excluded due to missing clinicopathological data; (3) eighty-eight rectal cancer patients and (4) twenty-eight colon cancer patients were excluded because they had received neoadjuvant treatment (radiotherapy, chemotherapy, or both) before surgery, or their vital status was unconfirmed. Ultimately, 449 patients were included in the final analysis.

Figure 2

The probability of mortality according to absolute neutrophil counts was analyzed using Cox proportional hazards regression. (A) The analysis was performed using Cox regression curves with tertiles of absolute neutrophil counts at baseline, both unadjusted and adjusted for BMI, sex, age, chemo- and radiotherapy, cancer stage (I+II vs III+IV), tumor site (colon vs rectum) + Histological grade (low vs high grade). T1 represents low neutrophil levels, T2 represents intermediate levels, and T3 represents high neutrophil levels. (B) Random forest of all significant variables included in our study to predict mortality. The symbols *, **, and *** indicate statistical significance, where * represents p < 0.05, ** represents p < 0.01, and *** represents p < 0.001.

Figure 3

Prognostic Value of Hematological Parameters and Clinical-pathologic Variables in Predicting Mortality in CRC. (A) Cox regression analysis by comparing T1 and T2 vs T3, to predict the risk of mortality using neutrophil-related ratios. (B) Three Receiver Operating Characteristic (ROC) curves with corresponding Areas Under the Curve (AUC).