The relationship between systemic inflammatory response, screen detection and outcome in colorectal cancer

A raised systemic inflammatory response correlates with poorer colorectal cancer (CRC) outcomes. Faecal immunochemical test bowel screening aims to detect early‐stage disease. We assessed the relationship between systemic inflammatory response, screen detection and CRC survival.


INTRODUC TI ON
Colorectal cancer (CRC) is the fourth most common cancer in the UK, with approximately 43 000 new cases and 16 500 deaths each year [1].The Scottish Bowel Screening Programme invites patients aged 50-74 years to undertake a quantitative faecal immunochemical test (FIT) followed by colonoscopy for those testing positive (threshold 80 μg Hb/g faeces) [2].This approach to screening increases the number of early-stage cancers diagnosed and reduces cancer-specific mortality [3][4][5][6].Those who undergo resection have lower T staging and less evidence of adverse pathological features such as venous invasion, peritoneal and margin involvement [7][8][9].
Additionally, there are some data that the incidence of CRC may be reduced within a screened population through the removal of premalignant polyps [6].
As well as demonstrating improved outcomes with screening, it is also important to understand the inherent host-factor differences that exist between the screen-detected and non-screen-detected populations.Previous studies have shown that screen-detected patients are more likely to be men, younger, less socioeconomically deprived and have a lower systemic inflammatory response [7][8][9].The presence of an elevated systemic inflammatory response is known to be associated with an adverse outcome after a diagnosis of CRC.
Further work is required to determine the impact of an elevated systemic inflammatory response on outcomes within the Scottish Bowel Screening Programme.The aim of this study was to assess the relationship between systemic inflammatory response, screen detection and overall survival (OS) and cancer-specific survival (CSS) in patients with CRC.

Study design, setting and participants
A retrospective observational cohort study was conducted.The cohort was formed from all patients invited to the first complete round of the Scottish Bowel Screening Programme in National Health Service Greater Glasgow and Clyde (NHS GG&C) between April 2009 and March 2011, whether they participated in screening or not.Patients were only included if they were diagnosed with a CRC and underwent resection with curative intent within 2 years of their screening invitation.Patients were classified as those diagnosed with CRC directly through Scottish Bowel Screening Programme participation (screen-detected patients) or via symptomatic pathways (non-screen-detected patients).In Scotland, colonoscopy is only routinely performed in asymptomatic individuals within the Scottish Bowel Screening Programme and so all non-screen-detected patients were scoped via symptomatic referral pathways.Approval for this study was given by the Caldicott Guardian of the screening dataset and by the West of Scotland CRC Managed Clinical Network Management Group.Ethical approval and individual patient consent were waived as the study was entirely retrospective, observational and anonymized and the study was performed in accordance with the Declaration of Helsinki.The results have been reported according to STROBE guidelines [10].

Variables and data sources
The formation of this cohort has been previously described [8].
Briefly, details of all individuals invited to the first complete round of the Scottish Bowel Screening Programme in NHS GG&C between April 2009 and March 2011 were extracted from a prospectively maintained database.To ensure identification of both screendetected patients and patients with non-screen-detected CRC diagnosed during the same period, the West of Scotland CRC Managed Clinical Network and Scottish Cancer Registry (SMR06) datasets were cross-referenced.Baseline demographics, preoperative blood results and survival were obtained on a case-by-case basis from NHS electronic patient records.
The presence of a systemic inflammatory response was quantified using three previously validated scores, the lymphocyte/monocyte ratio (LMR), neutrophil/lymphocyte ratio (NLR) and platelet/ lymphocyte ratio (PLR).These scores are derived from circulating lymphocyte, monocyte, neutrophil and platelet counts, taken from Conclusion: Screen-detected CRC patients have a lower systemic inflammatory response.
Non-screen detection and systemic inflammatory response (measured by LMR and PLR respectively) were independent predictors of poorer OS and CSS.

What does this paper add to the literature?
We have established that screen-detected colorectal cancer patients have lower systemic inflammatory response compared to non-screen-detected patients.This is the first paper to measure systemic inflammatory response in such patients using lymphocyte/monocyte, neutrophil/ lymphocyte and platelet/lymphocyte ratios.Additionally, systemic inflammatory response was shown to predict overall survival and cancer-specific survival, independent of screening status.a preoperative full blood count.In each case the ratios were calculated by dividing the former by the latter.A greater systemic inflammatory response is associated with a lower LMR and a higher NLR or PLR.Thresholds were derived from previously published data [11]: low LMR <2.4,high LMR ≥2.4; low NLR <3, moderate NLR 3-5, high NLR >5; low PLR ≤150, high PLR >150.Deprivation was quantified using the Scottish Index of Multiple Deprivation, derived from each patient's post code.The Scottish Index of Multiple Deprivation is a measure of an area's deprivation based on income, employment, education, health, access to services, crime and housing [12].Comorbidity was quantified using the American Society of Anesthesiologists (ASA) score and the Lee index [13].Patients were excluded from the final analysis if their records were absent from the NHS electronic patient record system or if preoperative blood results were unavailable.

Data analysis and statistical methods
Covariables were compared using crosstabulation and the χ 2 test for linear trend.A value of P < 0.05 was considered statistically significant.OS and CSS were analysed using Cox regression.All covariables found to be statistically significant (P < 0.05) predictors of survival on univariate analysis were carried forward to a multivariate survival analysis.To reduce the impact of collinearity between explanatory variables, a stepwise backward method was used to produce a final model of variables with a significant independent impact on survival, where variables were removed from the model when the corresponding P value was >0.05.This statistical analysis was performed using SPSS software (SPSS Inc.).

Participants
Of all 395 097 patients invited to participate in the first complete round of screening in NHS GG&C, 204 535 (51.7%) responded of whom 6159 (3.0%) tested positive.Of those testing positive, 4797 (77.9%) proceeded to colonoscopy and 421 (8.8%) of those patients were found to have CRC.There were 708 patients with non-screendetected CRCs diagnosed in NHS GG&C during the same time period of whom 468 (66.1%) were non-responders to screening, 182 (25.7%) were interval cancers (within 2 years of a negative screening test), 43 (6.1%) were individuals who chose not to attend colonoscopy following a positive screening FIT test and 15 (2.1%) had no malignancy detected at index screening colonoscopy.Of the 1129 total (421 screen-detected and 708 non-screen-detected), 761 patients underwent a surgical resection with curative intent, had complete NHS electronic portal records including preoperative blood results and were included in the final analysis.326 (42.8%) of these patients had screen-detected and 435 (57.2%) had non-screen-detected disease (Figure 1).Of the 435 non-screen-detected patients, 269 (61.8%) were non-responders, 125 (28.7%) were interval cancers, 29 (6.7%) refused colonoscopy following a positive FIT test and 12 (2.8%) had a normal index screening colonoscopy.
A comparison of demographics between screen-detected and non-screen-detected patients can be seen in Table 1.Patients with screen-detected disease were significantly more likely to be TA B L E 1 Baseline demographics and comparison of patients with screen-detected and non-screen-detected colorectal cancer.

DISCUSS ION
In the current study we have established that patients with screendetected CRC have a significantly lower systemic inflammatory response compared to their non-screen-detected counterparts, as quantified by LMR, NLR and PLR.This is the first study to compare the systemic inflammatory response between screen-detected and non-screen-detected CRC patients, using all three of these validated markers.Additionally, we have shown that a raised systemic inflammatory response as measured by LMR is associated with poorer OS, and a raised systemic inflammatory response as measured by PLR is associated with poorer CSS, independent of screening status.
It has been well established that patients with screen-detected CRC have improved outcomes compared to their non-screendetected counterparts [3][4][5][6][7][8][9].Earlier stage of presentation is certainly a key determinant of these improved outcomes.For example, in the current study we have shown that patients with screendetected disease have significantly lower TNM staging and fewer emergency operations than those with non-screen-detected disease.Additionally, previous work has shown that screen-detected patients undergoing resection have less venous invasion and less peritoneal and margin involvement [7,8].However, as can be seen in Figure S1A,B, screen-detected patients in this study had improved OS and CSS regardless of stage at diagnosis.There are several inherent differences between screen-detected and non-screen-detected patients which may also contribute to improved outcome.In the current study, screen-detected patients were more likely to be men, less deprived, less comorbid and have fewer rectal cancers.The systemic inflammatory response is one host factor that to date has not been studied in detail in relation to screen-detected versus nonscreen-detected disease.Previous studies on the current cohort of patients have revealed a higher systemic inflammatory response as measured by NLR [8,34].In the current study we decided to expand our investigation by using three validated markers of systemic inflammatory response (LMR, NLR and PLR) and by performing multivariate survival analysis.All three markers showed significantly less systemic inflammation amongst screen-detected patients.
Additionally, for the first time, on multivariate survival analysis LMR was able to independently predict OS and PLR was able to predict CSS.Simultaneously, screen detection retained significance as an independent predictor of both OS and CSS.We can therefore conclude that screen-detected patients have less systemic inflammation and that, along with other screen-detected benefits including earlier staging at diagnosis, less deprivation and lower comorbidity, this may be one factor which contributes to the improved outcomes seen within this group.However, while there is a relationship between screen detection and a lower systemic inflammatory response, it is important to note that both represent independent and valuable

F I G U R E 2 F I G U R E 3 F I G U R E 5
Relationship between screen detection and OS and CSS.Relationship between TNM stage and OS and CSS. the inherent differences between screen-detected and non-screendetected patients, in terms of both host and tumour factors.The present study has several strengths.We have been able to form a comprehensive cohort of both screen-detected and non-screen-detected CRC patients diagnosed during the study period, within our health board.Access to Scottish Bowel Screening Programme data allowed the identification of all screen-detected patients, while the use of cancer registries ensured capture of F I G U R E 4 Relationship between LMR and OS and CSS.Relationship between NLR and OS and CSS.non-screen-detected patients diagnosed via symptomatic pathways at the same time.This is the first study to compare the systemic inflammatory response between screen-detected and non-screen-detected patients using a broad panel of markers (LMR, NLR and PLR).By performing multivariate survival analysis with a long median follow-up of 63 months and with an extensive F I G U R E 6 Relationship between PLR and OS and CSS.

Non-screen-detected n (%) P value
Abbreviations: ASA, American Society of Anesthesiologists; LMR, lymphocyte/monocyte ratio; NLR, neutrophil/lymphocyte ratio; PLR, platelet/ lymphocyte ratio; SIMD, Scottish Index of Multiple Deprivation.a Two (0.3%) patients were not included in this comparison as they had synchronous colonic and rectal tumours.b Data missing for 111 (14.6%) patients.