Longitudinal Measurements of Inflammatory Indices During Treatment for Locally Advanced Rectal Cancer and Associations with Smoking, Ethnicity and Pathological Response

Simple Summary

Inflammatory markers are an area of interest in patients with locally advanced rectal cancer (LARC). This is a hypothesis-generating study that examines the trend of common inflammatory markers before, during, and after treatment for LARC. The trends observed may guide future research into the utility of such markers in the treatment intensification or de-escalation of LARC.

Abstract

This study explores the change in inflammatory markers over the course of neoadjuvant chemoradiation and adjuvant chemotherapy for LARC and assesses the association with clinicopathological factors at pre-specified time-points. We examined the trends of neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), C-reactive protein (CRP), carcinoembryonic antigen (CEA), fibrinogen, and albumin through multilevel modelling of 29 prospective LARC patients across six time-points: before neoadjuvant chemoradiation (T1), week 3 of chemoradiation (T2), post-chemoradiation (T3), post-surgery (T4), midpoint of adjuvant chemotherapy (T5), and chemotherapy completion (T6). Variables collected included ethnic background, body mass index (BMI), smoking status, and pathological responses graded by Ryan tumour regression grade and pathological tumour and nodal status. NLR and PLR demonstrated an increasing trend during chemoradiation. Median CEA was highest at baseline and lowest at T4. The highest median values for NLR, PLR, CRP, and fibrinogen were at T4. Smokers demonstrated a trend towards a higher NLR compared to non-smokers. NLR was significantly higher in Caucasians compared to Asians at T2. Patients with pathological node-negative status had a higher NLR at T5 and T6 and a higher PLR at T1, T3, T5 and T6. Overall, inflammatory indices change dynamically throughout treatment and vary with clinicopathological factors.

1. Introduction

The systemic inflammatory response is involved in the development of locally advanced rectal cancers (LARC) [1]. Inflammation contributes to an environment that is conducive to carcinogenesis but, on the other hand, is also a reflection of the body’s response to tumour growth [2]. Chronic inflammation can lead to the accumulation of genetic variants, epigenetic changes in epithelial cells and the release of cytokines and growth factors that enhance cancer growth and alter treatment sensitivity [3]. Conversely, inflammation can be an attempt to restore homeostasis. Research shows that key inflammatory cells such as macrophages, dendritic cells, and lymphocytes have both anti-tumour and pro-tumour properties [4]. In LARC, the commonly studied inflammatory haematological parameters include, but are not limited to, the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), monocyte-to-lymphocyte ratio, C-reactive protein (CRP), fibrinogen, and leucocyte count. The easy accessibility of testing for these markers has resulted in research focussing on the association between these markers of inflammation and outcomes in LARC.

Neoadjuvant therapy, consisting of pre-operative short-course radiation or long course chemoradiation, with or without further chemotherapy, followed by total mesorectal excision, is the current standard of care for LARC [5]. Despite recent advances, systemic recurrences are still the main cause of treatment failure and cancer-related deaths [6]. Certain trends have been observed in the prediction of tumour response and survival outcomes in LARC. Meta-analyses have found that a higher baseline NLR and PLR correlated with poorer oncological outcomes [7,8,9,10]. Specifically, a higher pre-treatment NLR was associated with poorer disease-free and overall survival [11]. Notably, most studies have focussed on inflammatory markers obtained prior to the start of neoadjuvant chemotherapy or radiation or prior to surgical resection. There is a paucity of data looking at these markers of inflammation serially during treatment in both neoadjuvant and adjuvant phases of treatment in LARC.

Serum carcinoembryonic antigen (CEA) is a widely used tumour marker. In LARC, a low pre-treatment CEA level was associated with better pathological outcomes [12,13,14]. A decrease in CEA following neoadjuvant therapy also correlated with better pathological responses and disease-free survival [15]. Conversely, those with persistent positive CEA levels following neoadjuvant chemoradiation and surgery are more likely to have poorer disease-free survival [16,17]. The literature to date highlights the importance of obtaining both pre- and post-treatment CEA levels, although the exact role that CEA plays in the individualisation of treatment is not well-defined.

Albumin, as an inflammatory and nutrition biomarker, has been investigated as part of an index such as the prognostic nutritional index [18] or in combination with markers such as fibrinogen [19], neutrophils [20], globulin [21] or CRP [22] in LARC. Its association alone with treatment response in LARC is unknown.

The impacts of demographic and socio-economic factors such as smoking, ethnicity or CALD (culturally and linguistically diverse) status, and body mass index (BMI) on cancer-related inflammation and treatment response are a growing area of interest. Population research suggests an association between Black people and an elevated inflammatory status; however, the confounding influences of socio-economic and behavioural factors are difficult to measure [23,24]. One study has found that Asian people are more likely to have inferior physical and cognitive functioning compared to Caucasian people during treatment for LARC [25]. Smokers are known to have a higher CRP than non-smokers [26,27]. Rectal cancer patients who smoke at diagnosis have significantly increased cancer-related mortality compared to non-smokers [28]. Obesity is known to be associated with mild chronic inflammation [29]. One study reported obesity to be associated with elevated levels of CRP, and the association was stronger in women than men and in North American compared to Asian populations [30]. Another study demonstrated a correlation between high levels of body fat and elevated CRP and interleukin-6 [31]. Overall, there is a complex association between ethnicity and inflammation that warrants further research.

Therefore, the aim of this exploratory prospective study was to examine the trends of commonly used and easily measurable inflammatory markers over the course of neoadjuvant chemoradiation and adjuvant chemotherapy for LARC and assess the impacts of smoking, CALD ethnicity, and BMI on a panel of inflammatory markers at pre-specified time-points. We hypothesised that those with a history of smoking and higher BMI would demonstrate a higher NLR, PLR, CEA, and CRP throughout their treatment course. We did not make any assumptions about CALD status but hypothesised that there may be differences based on population studies in ethnic minority groups. We hypothesised that a sustained pro-inflammatory environment would correlate with poorer responses to treatment and poorer survival outcomes.

2. Materials and Methods

We prospectively recruited patients from three cancer centres in the South Western Sydney Local Health District, NSW, Australia. Southwest Sydney is one of the most ethnically diverse regions in the state [32]. The inclusion criteria were as follows: age > 18; tissue confirmation of LARC (rectal adenocarcinoma stage T3/T4 or node-positive disease and no evidence of metastatic disease on computed tomography imaging of chest/abdomen/pelvis); and planned treatment with neoadjuvant chemoradiation. Neoadjuvant chemoradiation consisted of infusional 5-fluorouracil (5-FU) therapy or oral capecitabine concurrent with radiotherapy for 5 to 6 weeks. This was followed by total mesorectal excision approximately 6 to 12 weeks after completion of chemoradiation. Choice of adjuvant chemotherapy was at the physician’s discretion, consisting of either FOLFOX (intravenous oxaliplatin 85 mg/m², leucovorin 50 mg, and a fluorouracil 400 mg/m² bolus followed by a fluorouracil infusion at a dose of 2400 mg/m² over 46 h every 14 days for six cycles) or CAPOX (intravenous oxaliplatin 130 mg/m² and capecitabine 1000 mg/m² BD for 6 cycles). Ethics approval was obtained from the Sydney South West Area Health Service Ethics Review Committee (reference number HREC/13/LPPL/158). Data collection commenced in 2016 and was updated in 2023.

Inflammatory markers and biomarkers collected included NLR, PLR, CEA, CRP, albumin, and fibrinogen. NLR was calculated by dividing the absolute neutrophil count by the total lymphocyte count. PLR was calculated by dividing the absolute platelet count by the total lymphocyte count.

The blood tests were obtained across six time-points, as follows: start of neoadjuvant treatment (T1), week 3 of neoadjuvant chemoradiation (T2), post chemoradiation and pre-surgery (T3), post-surgery (T4), midpoint of adjuvant chemotherapy (T5), and post adjuvant chemotherapy (T6). We excluded patients who did not have all consecutive results at all time-points due to treatment refusal, withdrawal of consent, or missing data.

Demographic data collected included age, sex, CALD status, BMI, and smoking history (current or former smoker). CALD status was divided into Caucasians and Asians. Tumour Regression Grade (TRG) and pTNM staging (American Joint Committee on Cancer AJCC 7th edition/modification of Ryan et al.) were scored by two specialist pathologists on the surgical specimens [33,34]. All discrepancies between pathologists were resolved through consensus. The system used for TRG scoring was the AJCC 4-point scale adapted from Ryan et al. (2005), with 0 being complete tumour response with no viable tumour cells, 1 being small groups of cancer cells, 2 being residual cancer outgrown by fibrosis, and 3 being poor response with extensive residual tumour [34]. Recurrence was defined as radiological evidence of local tumour growth or evidence of metastatic disease.

Statistical Analysis

Frequencies and proportions were used to summarise the variables of interest. Separate piecewise multilevel models were used to analyse NLR, PLR, CEA, CRP, albumin, and fibrinogen over the six time-points. Maximum likelihood estimation with Satterthwaite degrees of freedom was employed. Combinations of random intercepts, random slopes, and residual correlation structures were examined for each outcome, with the optimal model selected using likelihood ratio tests. For the piecewise component of the model, 1 piece was specified from each time-point to the next, i.e., T1 to T2, T2 to T3, T3 to T4, T4 to T5, and T5 to T6. Clinical and demographic predictors of interest were TRG, pathologic tumour stage (ypT), pathologic nodal stage (ypN), smoking status, ethnic background, and BMI. Interactions between the piecewise time slopes and the clinical/demographic factors were included in the model and used to estimate differences in the outcomes at each time-point by these factors. For each outcome, natural log transformations were applied and improved normality of residuals. p < 0.05 was considered statistically significant. Analysis was conducted in the SAS statistical software (SAS Enterprise Guide version 8.2).

3. Results

3.1. Patient Characteristics

Of the 52 patients recruited, 29 patients had results at all six time-points and were included in this analysis. Patient characteristics are detailed in

Table 1. Characteristics of 29 patients with locally advanced rectal cancer treated with neoadjuvant chemoradiation.

	N	%
Total patients	29	100
Male	21	72
Female	8	28
Age at diagnosis (years)
Median (range)	58 (41–82)
Culturally and linguistically diverse background
Asian	15	52
Caucasian	14	48
Smoking status
Never smoker	16	55
Current or ex-smoker	13	45
Tumour regression grade
0–1	16	55
2–3	13	45
Pathologic tumour stage (ypT)
0–2	15	52
3	14	48
Pathologic nodal stage (ypN)
0	24	83
1–2	5	17
Recurrent disease	7	24
Adjuvant therapy
Capecitabine	24	83
FOLFOX (5-fluorouracil and oxaliplatin)	2	7
CAPOX (capecitabine and oxaliplatin)	1	3

. Of these, 21 (72%) were male. The median age was 58 years. Asian and Caucasian people comprised 52% and 48%, respectively. Thirteen (45%) patients had a history of smoking, 8 (28%) patients had a BMI between 18.5, and 24.9 and 21 (73%) patients had a BMI ≥ 25.

For pathological responses, 16 patients (55%) had a TRG of 1, 11 (28%) had a TRG of 2, and 2 (7%) had a TRG of 3. Fifteen patients (52%) had ypT 0–2 and fourteen (48%) a ypT of 3. Twenty-four patients (83%) had no involved lymph nodes on resection (pN0). Five patients (17%) were classed as pN1–2. Median duration of follow up was 6 years. Seven (24%) patients had recurrent disease as of last follow up.

3.2. Adjuvant Therapy Received

Twenty-four patients (83%) received capecitabine as adjuvant therapy, two patients (7%) received FOLFOX (Table 1). One patient (3%) received CAPOX.

3.3. Overall Trends of Inflammatory Markers During Neoadjuvant and Adjuvant Treatment

Median NLR increased during chemoradiation (T2) before decreasing following completion of chemoradiation (

Table 2. The trend of inflammatory markers across the six time-points throughout neoadjuvant and adjuvant treatment for locally advanced rectal cancer patients (median values in brackets).

	T1	T2	T3	T4	T5	T6
Neutrophil-to-lymphocyte ratio	2.4 (0.9–5.4)	3.9 (1.4–10.6)	3.0 (1.6–6.3)	4.6 (1.6–9.6)	3.7 (1.1–7.3)	3.4 (1.3–6.2)
Platelet-to-lymphocyte ratio	150 (76–264)	277 (88–556)	255 (123–426)	349 (166–624)	273 (112–495)	243 (124–407)
Carcinoembryonic antigen (μg/L)	4.5 (1.1–105)	3.7 (1.1–79.4)	2.7 (0.6–15.1)	1.8 (0.6–15.1)	2.3 (0.6–7)	2.4 (0.6–7.5)
Albumin (g/L)	44 (40–50)	43 (38–47)	44 (39–50)	43 (33–52)	45 (37–49)	43 (36–51)
C-reactive protein (mg/L)	2.3 (0.4–16)	1.5 (0.2–62.2)	1.4 (0.4–15.3)	4.1 (0.4–19.8)	1.6 (0.3–39.5)	1.3 (0.4–8.5)
Fibrinogen (g/L)	3.8 (2.8–6.3)	3.7 (2.6–5.5)	3.8 (2.6–4.8)	4.5 (2.7–7.9)	3.6 (2.6–6)	3.5 (2.2–5.8)

and Figure 1). The value peaked in the post-surgical phase (T4) before demonstrating a downward decline throughout adjuvant chemotherapy. Median PLR increased during chemo-radiation (T2), decreased pre-surgery, and then peaked in the post-surgical period (T4). The value decreased over the course of adjuvant chemotherapy (T5 and T6). Median CEA was highest at baseline (T1) and lowest following surgery (T4). There were small increments in the CEA during adjuvant treatment; however, it remained within normal limits. There was notable variation in CRP values for all patients across the six time-points. The highest median values for CRP, NLR, PLR, and fibrinogen were observed in the post-surgical phase (T4). Albumin was the most stable inflammatory marker in this group of patients, with little deviation throughout the six time-points.

3.4. Inflammatory Markers by Smoking, CALD, BMI Status and Pathologic Response

Twenty-four patients (83%) received capecitabine as adjuvant therapy (Table 1) and two patients (7%) received FOLFOX. One patient (3%) received CAPOX.

Natural logarithm transformations were applied to each inflammatory marker then modelled using piecewise multilevel models.

3.4.1. NLR

Associations of NLR with patient characteristics are shown in Figure 2. NLR was higher in the Caucasian population compared to the Asian population at week 3 of neoadjuvant chemoradiation (T2) (p = 0.0133). There was no statistical difference in NLR between BMI, TRG, and ypT subgroups at all time-points. Patients with pN0 status had a higher NLR at the midpoint of adjuvant chemotherapy (T5) (p = 0.0404) and post-adjuvant chemotherapy (T6) (p = 0.0019). Smokers demonstrated a trend for higher NLR at all time-points except T1.

3.4.2. PLR

Associations of PLR with patient characteristics are shown in Figure 3. Patients with a BMI < 25 had a higher PLR at week 3 of neoadjuvant chemoradiation (T2) (p = 0.029). pN0 patients had a higher PLR at the start of neoadjuvant treatment (T1) (p = 0.0279), post-chemoradiation and pre-surgery (T3) (p = 0.0403), at the midpoint of adjuvant chemotherapy (T5) (p = 0.0081), and post-adjuvant chemotherapy (T6) (p = 0.0018). There was a non-significant trend for a higher PLR in the smokers in the adjuvant phase of treatment.

3.4.3. CEA

For CEA levels patients with pN0 status had higher CEA at week 3 of neoadjuvant chemo-radiation (T2) (p = 0.0263) (Figure 4). There were no significant differences in CEA amongst other subgroups at all time-points.

Non-significant trends were observed. Non-smokers had a trend towards a higher CEA. Asian people demonstrated higher CEA levels than Caucasian people in the post-surgical phase. Patients with a BMI < 25 had a higher CEA prior and during chemo-radiation (T1 and T2). Patients with a poor response (TRG of 2–3) had a higher CEA throughout neoadjuvant, surgical, and adjuvant time-points.

3.4.4. CRP

Caucasian people had a higher CRP than Asian people at week 3 of neoadjuvant chemo-radiation (T2) (p = 0.0217) (Figure 5). There were no significant differences in CRP amongst other subgroups at all time-points.

3.4.5. Albumin and Fibrinogen

No significant trends were observed with albumin and fibrinogen.

4. Discussion

Persistent inflammation in the body has been linked to poor overall mortality in patients with cancer as well as patients with cardiac, respiratory, and renal diseases [35]. This study adds to the existing literature on serological inflammatory biomarkers in LARC and seeks to understand the dynamic changes in inflammatory biomarkers in LARC patients undergoing chemoradiation, surgery, and adjuvant treatment. Thus, this study is unique in tracking longitudinal changes in inflammatory markers during the neoadjuvant and adjuvant treatment phases.

There were notable trends observed in the adjuvant phase. NLR, PLR, and fibrinogen and CRP all declined during adjuvant chemotherapy, reflecting the myelosuppressive effect of systemic treatment. Certainly, lymphocyte depletion has been well documented in the post-chemotherapy phase [36]. Our results showed that NLR and PLR did not decrease to pre-treatment levels, suggesting ongoing inflammation or persistent lymphopenia. This would be consistent with literature demonstrating the presence of circulating cytokines that persists for months following adjuvant chemotherapy [36].

The relationship between inflammatory markers and survival outcomes in LARC remains controversial, with conflicting results from multiple studies. For example, Shen et al. (2012) found that a higher baseline NLR was associated with worse overall survival, but there was no correlation to pathological response [37]. Carruthurs et al. (2012) reported in their retrospective analysis of 150 patients that a higher NLR was associated with worse survival but not tumour response [38]. Morias et al. (2024) also found that an elevated pre-treatment and post-treatment NLR were associated with worse prognosis [39]. Notably, the largest retrospective study of 1500 patients showed that baseline NLR and PLR were not prognostic for disease-free survival or overall survival and did not predict pathological complete responses in LARC patients [40]. The heterogeneity in these results makes the prognostic value of these inflammatory markers uncertain and hinders the applicability of these markers to real-world clinical settings.

In our study, we observed that patients with a pN0 status reflecting good responses had a higher NLR and PLR at all time-points. Additionally, patients with a TRG of 0–1 (good responders) demonstrated a trend towards a higher CRP. A large body of evidence suggests that neutrophilia is generally associated with cancer progression through a multitude of mechanisms, including the release of factors that promote angiogenesis and recruitment of macrophages and T regulatory cells (T_reg) [41]. There are fewer studies reporting the inhibitory effect of neutrophils on cancer growth. One preclinical study demonstrated that neutrophils exert anti-tumoral effects through chemokines, T cells, and other leukocytes [42]. In another study that examined mouse models of breast cancer, neutrophils inhibited tumour growth in the presence of NK cells, but in their absence, the neutrophils lost their tumoricidal activity [43]. We hypothesise that the characteristics of circulating neutrophils (and other inflammatory markers) vary according to the microenvironment and are impacted by therapies such as chemotherapy and radiation. Additionally, the neutrophils described in our study are circulating neutrophils, which may differ from intra-tumoral neutrophils [44]. Recruitment of neutrophils to the site of tumour may contribute to a decrease in circulating neutrophils. These hypotheses and other factors may explain the conflicting results from the studies of inflammatory markers in LARC.

It is also interesting that our study showed patients with poorer responses, as reflected by TRG of 2–3, had a higher CEA throughout treatment, and this was found to be significant at the completion of treatment (T6) when compared to the group with TRG of 0–1. In a study of 1800 patients, an elevated baseline CEA (>5 ng/mL) was found to be a negative predictor of pathologic response and associated with shorter recurrence-free survival [45]. Combined, this suggests that serial measurements of CEAs are required in the surveillance of LARC in order to detect early recurrences, particularly in patients with an elevated baseline and post-treatment CEA. Further evaluation in prospective studies is required.

Our study is unique in that it is the first to examine the impact of smoking, ethnicity, and BMI on the inflammatory markers in LARC patients over the entire course of treatment. We observed that smokers had a trend towards a higher NLR, PLR, and CRP. This is consistent with literature showing that smoking increases leukocytes, neutrophils, lymphocytes, and monocytes [46,47,48]. The trend towards a higher CEA in the non-smoking patient group is unexpected, as previous studies have shown that smoking is more likely to be associated with an elevated CEA [49,50]. This likely reflects the complexity of other factors in smokers that potentially have an impact on CEA levels.

We also observed that NLR and CRP were statistically higher in the Caucasian population than in the Asian population during chemo-radiation. To date, there is no literature on this. Large population studies have been primarily performed in the United States and demonstrated that those of non-Hispanic Black ethnicity typically had lower NLR [51] and higher CRP [52], while those with a Caucasian background had the lowest platelet levels [53]. It is difficult to compare our findings to studies that have not selected an Asian population, which comprises a large proportion of patients in our local health district. The diversity seen in this small cohort highlights the need for further exploration of potential biomarkers in diverse ethnic groups, which may have implications for treatment responses and outcomes.

BMI in our study was only found to be statistically significant for increased PLR for patients with a BMI < 25. This is a novel finding. There are data that NLR is associated with obesity [54]; however, this was not evident in our study. The trend towards a higher CRP for those with a BMI ≥ 25 is consistent with literature showing increased inflammation in those with a higher BMI [30]. Prior research investigating BMI suggests a higher value is associated with longer overall survival in LARC [55,56]. In our study, an increased CEA was observed in the neoadjuvant phase for patients with a BMI < 25. The association between CEA and BMI warrants further examination, as well as the relationship to longer-term survival outcomes in LARC.

Albumin is a well-known marker of inflammation and nutritional status [57]. Our study showed that despite variations in other inflammatory markers over the course of adjuvant and neoadjuvant treatment, albumin remained extremely stable. One possible explanation is the robustness of our select cohort of patients and the involvement of dieticians in providing nutritional support. It may also show that, contrary to being part of prognostic scoring in other tumour types such as renal cell carcinoma [58], this may not be as useful in LARC.

There are limitations to our study. Firstly, the small sample size means that our findings, though interesting and intriguing, are hypothesis generating, and we recognise that no firm conclusions can be drawn regarding the predictive and prognostic role of these inflammatory markers, and this was never the intention of this study. Additionally, this study was designed and initiated in a period predating the changing paradigm in LARC with the OPRA [59], RAPIDO [60] and PRODIGE 23 [61] studies that shifted treatment to the neoadjuvant setting. Additionally, there is a move towards the “wait and watch” approach for patients who achieve a complete pathological response [59], as well as recent data on the use of immune checkpoint inhibitors in mismatch repair deficient LARC [62]. Finally, the effects of radiation alone on the immune system are unable to be addressed by our study, given that chemotherapy is given as a radiosensitizer and the chemoradiation is delivered together. Although we sought to analyse the influences of ethnicity, smoking, and BMI, it is important to remember that there are a multitude of other complex patient or environmental factors that could explain the effects we are seeing in these groups.

To strengthen our findings, a validation cohort with larger patient numbers exploring these markers further, as well as the investigation of these markers within the changing paradigm of care for LARC, is warranted.

5. Conclusions

In LARC patients undergoing trimodality therapy, inflammatory indices fluctuate throughout neoadjuvant and adjuvant treatment, and important trends are apparent. Smokers and Caucasian patients exhibited a trend for higher NLR, consistent with the association between smoking and systemic inflammation and potential ethnic variations in blood cell indices. Post-treatment CEA may be associated with pathological responses. Patients with a BMI ≥ 25 showed a trend towards a higher CRP. The impact of clinicopathological factors on inflammatory indices and their potential predictive utility for neoadjuvant and adjuvant therapy warrants ongoing evaluation.