Worldwide, the prevalence of colorectal cancer (CRC) has been increasing (1). It is currently considered the third most commonly diagnosed cancer and the second most common cause of cancer death accounting for 9.4% of all cancer-related deaths (2). Surgical resection remains the only curative treatment option for the majority of colorectal cancers. However, a considerable proportion of CRC patients experience recurrence or metastasis within 5 years of surgery (3).

Several factors have been identified as predictors of recurrence and prognosis among CRC patients. These include tumor-specific factors and patient-related factors (4). Of these, the presence of metastasis, the volume of liver metastasis, and the nodal stage have been shown to be independent prognostic factors for survival (4). Beyond these variables, inflammation has emerged as a key factor in the development of cancers.

The link between inflammation and cancer, first identified by Rudolf Virchow in the 19th century (5), has been well established, with inflammation playing a significant role in tumor development and progression (6, 7). This response often manifests as abnormalities in blood components, particularly neutrophils, lymphocytes, and platelets. Among the various inflammatory markers, neutrophil-to-lymphocyte ratio (NLR) and platelet-to-lymphocyte ratio (PLR) have been demonstrated to influence clinical outcomes in cancers, including colorectal cancer (CRC). Studies that assessed the utilization of pre-operative NLR as a prognostic biomarker of CRC showed that high NLR correlated with poor clinical outcomes; both overall survival (OS) and cancer-specific survival (CSS) (8). Similarly, PLR is increasingly being evaluated in different diseases, with high platelet count being considered a poor prognostic indicator for several cancers, including CRC (9). PLR has been shown in numerous studies to be associated with CRC progression, patients’ outcomes, and survival (10).

The association between these inflammatory markers and age had been identified by several studies, demonstrating a correlation between higher levels, especially NLR, and advanced age (11, 12). However, the link with age in specific disease groups, especially younger patients and notably in CRC, remains limited. Therefore, the primary aim of this study is to evaluate NLR and PLR across different age groups and to assess their prognostic values regarding OS and CSS among patients with CRC.

For NLR and PLR, the optimal cutoff point was determined as 3.19 and 224.4, respectively. High NLR was defined as ≥3.19 and low NLR as <3.19. High PLR was defined as ≥224.4 and low PLR as <224.4.

The role of the immune system in the progression of various cancers, including CRC, has become increasingly recognized (13). This response can be evaluated by several serum markers, with the most robust being NLR and PLR, as they have been confirmed to have a significant role in predicting the prognosis of CRC (14). This study investigated the values of NLR and PLR among different age groups of CRC patients and assessed their prognostic value in regard to CSS and OS. Cutoff values of NLR and PLR have varied among studies, because articles presented different populations with varying tumor characteristics. However, NLR and PLR values ranged between 2 to 5 and 160 to 350, respectively, which aligns with our study (15).

When comparing NLR across age groups, two patterns emerged from our data. Older patients (≥70 years) demonstrated a statistically significant association with higher NLR values. However, median values categorized by age groups revealed that younger patients (<50 years) had the highest median NLR. Specifically, the median NLR in the <50 years group was 4.54, and the proportion of patients with high NLR (≥3.19) was also elevated (63.4%), comparable to the ≥70 years age group (61.4%). This paradox, where the youngest group has the highest median yet shares a similar proportion of high NLR with the oldest group, likely indicates different underlying mechanisms. In the younger age groups, this elevated median may be reflected by disproportionately high NLR in some individuals, which may represent a more aggressive disease course. Recent studies suggest that younger patients may have a unique subtype of CRC, with expanding literature relating elevated inflammatory markers to worse outcomes in this group (16, 17). However, further research is needed to confirm whether elevated NLR in early-onset CRC reflects a distinct immune-related phenotype with potential implications for prognosis and treatment. On the other hand, the association with older age groups reflects an overall trend when considering the entire age group distribution. These findings suggest that while older age is generally associated with higher NLR, specific outliers in younger patients may significantly influence the median, highlighting the complexity of interpreting NLR distribution across age groups. In comparison to other studies, Li et al. found that CRC patients with older age at diagnosis (>60) have significantly higher NLR (18). Other similar observations have also been published (11, 12, 19). These findings are consistent with the well-known effects of aging on increased inflammation (20), which can be confirmed by elevated NLR levels in older healthy patients (21). Importantly, although our age comparisons were descriptive in nature, the multivariate Cox models, which were adjusted for age, demonstrated that high NLR remained independently associated with cancer-specific survival, indicating that these age-related differences do not diminish its overall prognostic relevance. Regarding PLR, its associations with prognosis and age are still ambiguous. Similar to other studies, our findings showed no statistical relation between PLR levels and age (18, 22, 23). However, more research is needed to evaluate the relation between PLR and specific age groups.

When assessing other variables, depth of invasion, represented by pT stage, had a significant association with NLR. These findings are in alignment with other studies, where depth of invasion is frequently shown to have a significant association with inflammatory markers (11, 12, 18, 23–26). However, lymph nodes involvement (pN), Metastases (pM), and TNM stage did not show any significant relation with either NLR or PLR, with their association with these inflammatory markers still in debate (11, 12, 18, 23–26). In addition, the primary site of the tumor did not correlate significantly with either NLR or PLR. A discrepancy in the literature has been observed regarding the effect of tumor site on inflammatory markers, with some studies showing no influence on the levels (11, 27–29). However, PLR showed a significant association with the type of surgery performed, particularly in patients undergoing right hemicolectomy. The association between surgical approach and inflammatory markers like NLR and PLR may indicate that tumor location, rather than the surgery itself, is linked to the inflammatory response. This conveys the debate in literature regarding the relation between tumor location and inflammatory markers levels, as shown in our results, where tumor location itself did not have any significant association with markers levels.

While traditional prognostic factors such as TNM staging remain the cornerstone of CRC risk stratification, they are not always sufficient to accurately predict individual outcomes. According to that, inflammatory markers like NLR may provide prognostic values by reflecting the body’s immune response to tumor biology. Our findings support this by showing that NLR remained an independent predictor of CSS after adjusting for clinicopathological variables such as tumor stage, lymph node involvement, and metastasis. Although our dataset did not include other molecular biomarkers such as serum CEA, prior studies have demonstrated that elevated NLR correlates with poor outcomes even after controlling for these biomarkers (30, 31).

Regarding adjuvant chemotherapy, it was found to have no significant association with either NLR or PLR, which is in contrast to results published by Li et al., showing that patients who received adjuvant chemotherapy had significantly higher preoperative NLR and PLR levels compared to those who did not (18). This finding highlights that patients selected for adjuvant chemotherapy often had elevated inflammatory markers at baseline, reflecting more aggressive systemic inflammatory responses.

Our results also showed that patients with preoperative high NLR had poorer OS and CSS. Following the adjustment of confounding factors, high NLR remained significantly associated with worse CSS. In a systematic review of 71 studies, high NLR was prognostic of poor clinical outcomes regarding OS, comprising disease, recurrence and progression-free survival, which was in agreement with our current findings (8). The fact that NLR remained significantly associated with CSS, but not for OS, after multivariable adjustment reflects direct relation between systemic inflammation and tumor biology. In contrast, OS includes other non-cancer-related causes of death, which may dilute the cancer-specific prognostic value of NLR, particularly in older patients. Therefore, our findings suggest that NLR may function more accurately as a predictor of CSS rather than OS in CRC.

On the contrary, our study reported no association between PLR and OS or CSS, in contrast to previous studies where patients with high PLR possessed a significantly poorer OS (10). Additionally, PLR discriminatory capacity was limited in our cohort, which further reduces its clinical utility. These results suggest that routine use of PLR as a prognostic biomarker may not be useful in all CRC settings. Further large-scale, prospective studies are needed to evaluate whether PLR may hold predictive value in specific subgroups or treatment aspects.

Since NLR is calculated from a basic CBC, it comes at no extra cost as it is already part of routine preoperative assessments. Its availability makes it eligible for inclusion in risk stratification, especially in settings where advanced and expensive imaging, genomic studies, or tumor markers may not be available. In clinical practice, NLR could help identify patients who might need more intense follow-up, or consideration for adjuvant therapies including chemotherapy and radiotherapy. It may also be an early sign for high-risk cases. That said, major CRC guidelines, such as those from the NCCN and ESMO, have not yet included inflammatory markers like NLR or PLR in their risk models (32–34). However, growing evidence supports their prognostic value in preoperative risk stratification. Our findings add to the existing literature and propose that markers like NLR could one day become a useful part of standard prognostic models.

In conclusion, the findings of our study indicate that preoperative NLR is an independent prognostic factor for CSS, with higher NLR significantly associated with poorer survival outcome. The feasibility of NLR calculation allows for simple biomarkers to be incorporated in colorectal cancer patients’ assessment preoperatively. We also observed meaningful variation in NLR across age groups, including disproportionately elevated values in a subset of younger patients, suggesting possible biological differences in early-onset colorectal cancer. Despite being consistently demonstrated across multiple studies, the role of NLR levels across different age groups should be studied more thoroughly to aid in the identification of patients with a more aggressive disease course at a younger age.