The prevalence and prognostic value of systemic inflammation in good performance status patients with advanced, inoperable non‐small cell lung cancer receiving palliative radiotherapy: Comparison of composite ratios and cumulative scores

Abstract Introduction The present study sought to examine the relationships between systemic inflammatory composite ratios/cumulative scores, magnitude of systemic inflammatory response (SIR) and survival in good performance status patients (ECOG‐PS 0/1) with advanced NSCLC receiving palliative radiotherapy. Methods Systemic inflammatory composite ratios/cumulative scores included the neutrophil‐lymphocyte ratio (NLR), platelet‐lymphocyte ratio (PLR), lymphocyte‐monocyte ratio (LMR), C‐reactive protein, (CRP)‐albumin ratio (CAR), neutrophil‐ lymphocyte score (NLS), platelet‐lymphocyte score (PLS), lymphocyte‐monocyte score (LMS), neutrophil‐platelet score (NPS), modified Glasgow prognostic score (mGPS). The magnitude of SIR was determined by serum CRP concentration, with a median CRP concentration of >10 m mg/L considered to be systemically inflamed. Relationships between systemic inflammatory composite ratios/ cumulative scores and clinicopathological characteristics were examined using chi‐square analysis. Relationships between overall survival (OS) and systemic inflammatory composite ratios/ cumulative scores were examined using cox regression analysis. Results 479 patients were included. 48% (n = 231) of patients were male and 70% (n = 338) were ≥65 years of age. 29% (n = 140) patients were ECOG‐PS 0 and 71% (n = 339) were ECOG‐PS 1. 98% (n = 469) of patients died during follow‐up. The median survival was 5 months (2–11). A similar prevalence of systemic inflammation was noted across the various ratios/scores (NLR >3 68%; LMR <2.4 65%; PLR >150 70%; CAR >0.20 83%; NLS ≥1 66%; LMS ≥1 71%; NPS≥1 50%; PLS≥1 60% and mGPS≥1 75%). Despite not considered to be systemically inflamed, an NLR <3, LMR ≥2.4, PLR ≤150, NLS 0, LMS 0, NPS 0 and PLS 0 all had a median CRP concentration of >10 mg/L. When adjusted for ECOG‐PS, CAR>0.40 (p < 0.001) and mGPS 2 (p < 0.05) remained significantly associated with OS. Conclusion Liver‐based measures of systemic inflammation (CAR and mGPS) appear more reliable for the quantification of the magnitude of SIR and have prognostic value in patients with advanced NSCLC.


| INTRODUCTION
Non-small cell lung cancer (NSCLC) is responsible for approximately 75% of lung cancer cases. 1 Contemporary studies suggest that most patients with NSCLC are diagnosed with locally advanced or metastatic disease. 2 As such, survival outcomes for NSCLC remain poor in comparison to other cancer subtypes. 1While treatment options for advanced, inoperable NSCLC were historically limited to palliative chemotherapy and/or radiotherapy, 3,4 novel target therapies and immunotherapy may have the potential to improve survival outcomes. 5,6Therefore, the identification of prognostic factors that may predict likely outcome in patients with advanced NSCLC remains of interest. 7he Eastern Cooperative Oncology Group (ECOG) performance status has long been considered a cornerstone of prognosis in patients with advanced cancer. 8,9owever, this self-reported assessment is subject to limitations including bias raising doubts on reliability of observations. 10Indeed, poor performance status may not be an absolute contraindication to patients with advanced cancer receiving treatments such as immunotherapy, with recent literature reporting that it is safe and tolerable. 11,12urthermore, questioning whether it is a determinant of treatment efficacy. 13n contrast to ECOG-PS, biomarkers of inflammation are generally considered reliable determinants of both treatment effectiveness and likely outcome in patients with advanced cancer. 14,15Indeed, biomarkers of systemic inflammation have been reported to have prognostic value in both randomised clinical trials of patients with advanced NSCLC 15 and in real world clinical practice. 16owever, with a range of composite scores and cumulative ratios reported within the literature, it remains unclear which ratio/scores should be used.Specifically, if there is heterogeneity in the prevalence and magnitude of systemic inflammation across the different ratios/scores and in their prognostic value to survival.
Composite ratios such as the neutrophil-lymphocyte ratio (NLR), lymphocyte-monocyte ratio (LMR) and platelet-lymphocyte ratio (PLR) have been reported to have prognostic value to clinical outcomes in patients with advanced NSCLC receiving targeted therapies such as tyrosine kinase inhibitors 17,18 and immunotherapy. 19,20][23] Comparison with a sensitive, routinely clinically available measure of the systemic inflammatory response, such as C-reactive protein (CRP), may be useful method to determine the reliability of such measures using a rational common base. 24he aim of the present study was to examine the relationships between systemic inflammatory composite ratios/cumulative scores, the magnitude of the systemic inflammatory response and survival in good performance status patients with advanced, inoperable NSCLC.

| Patients
Consecutive patients with locally advanced or metastatic NSCLC (TNM stage III or IV disease), who received radiotherapy with palliative intent at the Beatson Oncology Centre, Glasgow, between May 2010 and 2016, were identified from a prospectively maintained database.Patients who were deemed to have good performance status (ECOG-PS 0/1) and had pre-treatment bloods facilitating assessment of the systemic inflammatory response were assessed for inclusion.Patients with histologically confirmed small cell lung cancer or those who underwent radiotherapy with curative intent were excluded.
The primary endpoint was overall survival from date of commencing palliative radiotherapy.Date of death was confirmed using hospital electronic records, until the 1st and mGPS≥1 75%).Despite not considered to be systemically inflamed, an NLR <3, LMR ≥2.4,PLR ≤150, NLS 0, LMS 0, NPS 0 and PLS 0 all had a median CRP concentration of >10 mg/L.When adjusted for ECOG-PS, CAR>0.40 (p < 0.001) and mGPS 2 (p < 0.05) remained significantly associated with OS.

Conclusion: Liver-based measures of systemic inflammation (CAR and mGPS)
appear more reliable for the quantification of the magnitude of SIR and have prognostic value in patients with advanced NSCLC.
May 2023, which served as the censor date.Ethical approval for this study was granted Greater Manchester East Research Ethics Committee (Rec number: 17/NW/0190).The present study was a retrospective, observational cohort study with no change in patient management.As a result, informed consent was not required in accordance with ethical approval.

| Clinicopathological characteristics
Routine demographic details included age, sex and BMI.Age categories were grouped into <64, 65-74 and >74 years.Histological subtype was broadly categorised as adenocarcinoma, squamous cell carcinoma and other/unknown.All tumours were retrospectively staged using the eighth edition of the tumour, node and metastases (TNM) classification and categorised into clinical AJCC stage groupings. 25The administration of chemotherapy prior to the patient undergoing radiotherapy and the radiotherapeutic regime administered was identified from hospital electronic records.

| Systemic inflammation
Haematological and biochemical results of venous blood samples, obtained within the 6 weeks prior to the patient undergoing radiotherapy, were prospectively recorded.Serum CRP concentration, albumin concentration and differential blood cell counts were used to calculate cumulative scores and composite ratios.The NLR, PLR, LMR and C-reactive protein albumin ratio (CAR) were all calculated by directly dividing the former by the latter and categorised using thresholds values from the contemporary literature. 21,22Similarly, the Neutrophil-lymphocyte score (NLS), Neutrophil-platelet score (NPS), platelet-lymphocyte score (PLS), lymphocyte-monocyte score (LMS) and modified Glasgow Prognostic score (mGPS) were constructed as previously described, using validated threshold values. 21,22An auto analyser was used to measure serum CRP (mg/L) and albumin (g/L) concentrations (Architect; Abbot Diagnostics, Maidenhead, UK).

| Statistical analysis
The corresponding CRP concentration (mg/L) of each ratio/score category was presented as median (interquartile range, IQR).Demographic data, clinicopathological variables, TNM stage, ECOG-PS, NLR, LMR, PLR, CAR, NLS, LMS, NPS, PLS, mGPS and overall survival (OS) were presented as categorical variables.Relationships between categorical variables were analysed using the chisquare test for linear-by-linear association.
Univariate and multivariate survival data were analysed using Cox's proportional-hazards model.Variables associated with OS at a significance level of p < 0.1 on univariate analysis were included in multivariate model using a backward conditional approach.OS was defined as the time (months) from the date of commencing palliative radiotherapy to the date of death due to any cause.
Missing data were excluded from analysis on a variableby-variable basis.Two-tailed p < 0.05 were considered statistically significant.Statistical analysis was performed using SPSS software version 28.0.(SPSS Inc., Chicago, IL, USA).
The correlation between CAR, mGPS and clinicopathological characteristics in good performance status patients with advanced, inoperable NSCLC receiving palliative radiotherapy is shown in Table 3.On univariate analysis, CAR was significantly associated with histological subtype only (p < 0.05).On univariate analysis, mGPS was significantly associated with chemotherapy (p < 0.05) and ECOG-PS (p < 0.05, see Table 3).
The relationship between ECOG-PS, CAR, mGPS and OS in good performance status patients with advanced, inoperable NSCLC receiving palliative radiotherapy is shown in Table 4. On univariate analysis, CAR>0.40 (p < 0.001) and mGPS 2 (p < 0.001) were significantly associated with OS.When adjusted for ECOG-PS, CAR > 0.40 (p < 0.001) and mGPS 2 (p < 0.05) remained significantly associated with OS (see Table 4).

| DISCUSSION
The results of the present study show that myeloid-based systemic inflammation measures such NLR, LMR, PLR, NLS, LMS, NPS, and PLS all capture increasing systemic inflammation as measured using CRP, the prototypical acute phase protein and marker of systemic inflammation and have prognostic value.However, the low-risk group in each of these measures was associated with CRP concentrations (medians 13-19 mg/L) above the normal range (≤10 mg/L) and therefore these myeloid based measures are not a sensitive measure of systemic inflammation.In contrast, in the CRP based prognostic measures, CAR and mGPS, the low-risk groups were associated with CRP concentrations (medians 6

T A B L E 4
The relationship between ECOG-PS, CAR, mGPS and OS in good performance status patients with advanced, inoperable non-small cell lung cancer receiving palliative radiotherapy (n = 395).
Systemic inflammatory ratios/scores comprised of circulating white blood cells or acute phase proteins represent the systemic responses the bone marrow (myeloid) and liver, respectively. 22Both NLR (myeloid) and mGPS (liver) have been reported to have prognostic value to survival in patients with advanced, inoperable NSCLC. 27,28owever, few studies to date have directly compared the prognostic value of systemic inflammatory composite ratios and cumulative scores in patients with advanced, inoperable NSCLC.As such, the present results are informative.Specifically, that the prevalence of elevated ratio and score varied, with the highest prevalence observed in the acute phase protein-based CAR and mGPS.This would suggest the greater sensitivity of liver-based measures of systemic inflammation.
It was of interest that the low-risk group of all myeloid based inflammatory ratios/scores (NLR <3, LMR ≥2.4,PLR ≤150, NLS 0, LMS 0, NPS 0 and PLS 0) had a median CRP concentration of >10 m mg/L, generally considered to be inflamed (CRP >10 mg/L).While the specificity of myeloid based composite ratios has previously been questioned in studies of primary operable colonic, 22 rectal 21 and hepatocellular cancer, 23 the present study is the first to report this in patients with advanced inoperable cancer.Furthermore, the present results also question the reliability of myeloid based cumulative scores to differentiate those who are inflamed from those who are not.If the present observation were confirmed in future studies, then it would advocate the use of liver-based measures as the basis to determine the systemic inflammatory status of patients with cancer. 26ecent clinical trials have reported that novel target therapies and immunotherapy may improve survival outcomes in patients with NSCLC. 5,6Currently, only patients with good performance status patients are generally considered eligible candidates for such treatment. 29owever, the subjective nature of performance status has implications for the external validity of clinical trials in real world clinical practice. 30Furthermore, contemporary studies have challenged the utility of ECOG-PS to predict likely outcome in patients with advanced NSCLC receiving immunotherapy. 31In contrast, liver-based biomarkers of systemic inflammation, such as mGPS, have been reported to reliably stratify survival in clinical trials of patients with NSCLC. 15,27Therefore, it is clear that the mGPS is an objective adjunct to the subjective ECOG-PS and may be useful in the future allocation of immunotherapy in patients with advanced, inoperable NSCLC.
There are several limitations to the present study.Firstly, this is a single-centre, retrospective cohort study and has limitations seen with this study design including the potential for sample bias.Secondly, in the present study, when examined in Table 2, the specificity of myeloid measures for quantification of the magnitude of SIR was questioned, with the baseline composite ratios (NLR, PLR, LMR) and baseline cumulative scores (NLS, PLS, LMS, NPS) all having a median CRP concentration >10 mg/L (generally considered systemically inflamed).Therefore, only the prognostic value of liver-based measures of systemic inflammation (CAR and mGPS) was examined in relation to overall survival, adjusted for ECOG-PS. 8Nevertheless, given their extensively reported prognostic value and provided that their baseline threshold was associated with a CRP ≤10 mg/L, comparison of the prognostic value of these composite ratios (NLR, PLR, LMR) and cumulative scores (NLS, PLS, LMS, NPS) would be of considerable interest.Thirdly, the patients included in the present study were recruited between 2010 and 2016.As such, few patients were likely to have received immunotherapy or targeted therapies which are now recognised to have a profound impact on survival in a proportion of patients. 29However, it of interest that systemic inflammation based prognostic scores are now recognised to prognostic value in such targeted therapies and immunotherapy. 32,33Therefore, the present results are likely to be relevant for current clinical practice.Lastly, in the present cohort included only good performance status patients (ECOG-PS 0/1).Further study of larger cohorts including poor performance status patients (in particular ECOG-PS 2) are required to determine whether systemic inflammatory biomarkers stratify survival of patients with advanced, inoperable NSCLC in real world clinical practice.
In summary, compared with myeloid based measures of the systemic inflammatory response, liver-based measures (CAR and mGPS) appear to be more reliable in quantifying the magnitude of the systemic inflammatory response and therefore should form the basis of systemic inflammation based prognostic scores in patients with advanced NSCLC.

1
Flow diagram of included patients.The clinicopathological characteristics of good performance status patients with advanced, inoperable NSCLC receiving palliative radiotherapy.The prevalence and prognostic value of systemic inflammation across composite ratios and cumulative scores in good performance status patients with advanced, inoperable NSCLC who received palliative radiotherapy (n = 479).The correlation between CAR and mGPS with clinicopathological characteristics in good performance status patients with advanced, inoperable NSCLC receiving palliative radiotherapy (n = 395).
26A B L E 1 and 4 mg/L respectively, within the normal range).The present results, if confirmed in other studies, have important implications for the continued use of myeloidbased measures of systemic inflammation as prognostic scores in with advanced cancer.In particular, how they may be combined with acute phase protein measures of the systemic inflammatory response to enhance prognostic value.26Note:Boldvaluesare statistically significant.ap-Valuewas determined using cox proportional hazards regression.T A B L E 2 (Continued) T A B L E 3 Note: Each cell shows p value from chi-square analysis.Bold values are statistically significant.UnivariateAdjusted

for ECOG-PS OS HR (95% CI) p-Value a OS HR (95% CI) p-Value a
a p-Value was determined using cox proportional hazards regression.
non-small-cell lung cancer: a systematic review and metaanalysis.J Oncol.2022;2022:3601942.32.Horstman IM, Vinke PC, Suazo-Zepeda E, et al.The association of nutritional and inflammatory biomarkers with overall survival in patients with non-small-cell lung cancer treated with immune checkpoint inhibitors.Thorac Cancer.2024;15(23):1764-1771.33.Saal J, Bald T, Eckstein M, et al.Integration of on-treatment modified Glasgow prognostic score (mGPS) to improve imaging-based prediction of outcomes in patients with nonsmall cell lung cancer on immune checkpoint inhibition.Lung Cancer.2024;189:107505.How to cite this article: McGovern J, O'Rourke F, Will S, et al.The prevalence and prognostic value of systemic inflammation in good performance status patients with advanced, inoperable non-small cell lung cancer receiving palliative radiotherapy: Comparison of composite ratios and cumulative scores.Cancer Med.2024;13:e70139.doi:10.1002/cam4.70139