Higher Neutrophil-to-Lymphocyte Ratio (NLR) Is a Preoperative Inflammation Biomarker of Poor Prognosis in HIV-Infected Patients with Colorectal Cancer: A Retrospective Study

Background The serum systemic inflammation biomarkers are known predictors of colorectal cancer (CRC) patient prognosis. However, their significance in human immunodeficiency virus (HIV)-infected patients with CRC has not been studied. To address this gap, we conducted a retrospective study to evaluate the prognostic value of preoperative systemic inflammation biomarkers in HIV-infected patients with CRC. Methods The study enrolled 57 patients with colorectal cancer (CRC) and HIV who underwent surgery at the Shanghai Public Health Clinical Center between January 2015 and December 2021. Preoperative tests were conducted, and systemic inflammation biomarkers were measured. The patients were categorized into two groups using the optimal cut-off value. The Kaplan–Meier method and the log-rank test were used to determine overall survival (OS) and progression-free survival (PFS). Multivariate analysis was performed using the Cox proportional regression model. A time-dependent receiver operating characteristic (t-ROC) was used to compare the prognostic abilities of the biomarkers. Results The study included 57 HIV-infected CRC patients, with a median age of 60 and a follow-up time ranging from 3 to 86 months. Of the patients, 49 were male and 8 were female. The cumulative three-year OS and PFS rates were 55.0% and 45.0%, respectively. The optimal cut-off value for preoperative NLR was found to be 2.8, which was significantly correlated with lower CD8+ T and CD3+ T lymphocyte counts. Multivariate Cox regression analysis revealed that a low NLR was an independent predictor of better OS and PFS (OS: HR = 0.094, 95% CI: 0.02–0.45, P=0.003; PFS: HR = 0.265, 95% CI: 0.088–0.8, P=0.019). The time-dependent receiver operating characteristic (t-ROC) analysis showed that NLR was a superior systemic inflammation biomarker for predicting the prognosis of HIV-infected CRC patients throughout the observation period. Conclusion The preoperative neutrophil-to-lymphocyte ratio (NLR), an easily measurable immune biomarker, may provide useful prognostic information in HIV-infected colorectal cancer (CRC) patients.


Introduction
Colorectal cancer (CRC) is a prevalent form of cancer and the leading cause of cancer-related deaths worldwide [1]. Te use of combination antiretroviral therapy (cART) has resulted in an increase in CRC incidence in patients infected with the human immunodefciency virus (HIV) [2]. Furthermore, HIV-positive patients with CRC have a poorer prognosis than HIV-negative patients [2,3]. While known traditional risk factors for CRC include obesity, excessive red meat consumption, family history, and hereditary polyposis, HIV-induced immunodefciency is also believed to increase CRC risk [4]. HIV activates CD4 + T lymphocytes, dendritic cells, and macrophages, resulting in immune system dysfunction [5]. In HIVinfected CRC patients, a reduction in immune surveillance, increased expression of the immune checkpoint, direct viral protein efects, or cytokine dysregulation may lead to poorer outcomes [6]. Few studies to date have explored the relationship between HIV infection and colorectal cancer (CRC). Moreover, in China, HIV-infected CRC patients often experience poorer outcomes due to insufcient understanding and discriminatory attitudes towards HIV.
Although molecular subtypes had the potential to identify CRC patients with a worse prognosis, the results based on somatic copy number, CpG island hypermethylation, and gene expression made them difcult to use in clinical practice [7,8]. Another important predictor of prognosis in CRC patients is the cancer-associated systemic infammation status [9]. Te tumor microenvironment includes monocytes, neutrophils, platelets, and lymphocytes. Terefore, serum systemic infammation biomarkers, such as the neutrophil-to-lymphocyte ratio (NLR), lymphocyteto-monocyte ratio (LMR), platelet-to-lymphocyte ratio (PLR), systemic infammation score (SIS), and prognostic nutritional index (PNI), are useful in numerous studies [9]. New biomarkers, including the fbrinogen to prealbumin ratio and the albumin to fbrinogen ratio, have emerged as potential predictors of CRC outcome [10]. HIV infection functionally impairs the HIV-specifc CD8 + T and CD4 + T lymphocytes, resulting in the inability of the host immune system to control HIV [11]. Although cARTs do not completely eradicate HIV, continued treatment can restore the immune system function [12]. A previous study also found no diference in tumor-infltrating lymphocytes between HIV-positive and HIV-negative CRC patients [13]. Assessing systemic infammation biomarkers in CRC may be aided by studying immune reconstitution in HIV-infected patients. Early identifcation of potential predictors in these patients could improve their chances of survival. However, to our knowledge, no research has been conducted to investigate the relationship between serum systemic infammation biomarkers and HIV-related CRC prognosis. Terefore, the objective of this study is to evaluate the prognostic value of systemic infammation biomarkers in patients with HIV-related CRC.

Patients.
Tis study retrospectively collected data from all human immunodefciency virus (HIV)-infected patients with colorectal cancer (CRC) who underwent surgery between January 2015 and December 2021 at the Shanghai Public Health Clinical Center (SHPHC). A total of 57 HIVinfected patients were recruited. Te inclusion criteria were as follows: preoperatively diagnosed as primary colorectal adenocarcinoma and anal squamous cell carcinoma. Te exclusion criteria were as follows: evidence of a severe infammatory condition; chronic infammation diseases except HIV infection; incomplete clinicopathological data. Te following characteristics were included: gender, age, body mass index (BMI), disease history (hypertension, diabetes, tuberculosis infection, smoking, and alcohol abuse), presence of intestinal obstruction on admission (diagnosis was based on the symptoms of no bowel movements combined with radiographic or colonoscopic fndings), duration of HIV infection, and treatment (antiviral drugs), preoperative tests (routine blood tests, liver function, tumor markers, and plasma lipids), tumor location (if the primary tumor was located from cecum to transverse colon, it was defned as right colon cancer, and from splenic fexure of colon to upper rectum, it was defned as left colon cancer, and from middle or lower rectum to anal, it was defned as rectal cancer), postoperative pathological stage (using the eighth AJCC edition), mismatch repair (MMR) status (using immunohistochemical results of four MMR proteins) and surgical resection with/without tumor residuals. Tis study was conducted according to the Declaration of Helsinki, and the Shanghai Public Health Clinical Center approved it. Te patients' data were obtained from the hospital database and used for research purposes.

Follow-Up.
Regular follow-up assessments were conducted every three months during the frst two years postsurgery and every six months during the following two years. Te patients were admitted to the inpatient unit for routine blood tests and enhanced chest and abdominal CT scans. After three years, follow-up assessments were performed annually. Local recurrence, enlargement of unresected tumor lesions, and distant organ metastases were all considered indicators of progression. In the event of patient death, confrmation was obtained either from relevant hospital records or from notifcation provided by the patient's family during a telephone follow-up. Te latest censoring date for survival time evaluation was March 2022. Overall survival (OS) was defned as the time from surgery to death from any cause. Progression-free survival (PFS) was defned as the interval between surgery and either progression or death.

Defnition of Infammation-Related Biomarkers.
Te laboratory tests from each patient were obtained within 1-week before surgical resection of the primary tumor. Te interval between the end of last neoadjuvant treatment and surgery is at least over six weeks. According to a recent review [9], the neutrophil-to-lymphocyte ratio (NLR) was calculated by dividing the absolute neutrophil count by the absolute lymphocyte count. Te lymphocyte-to-monocyte ratio (LMR) was defned as the absolute lymphocyte count divided by the monocyte count. Te platelet-to-lymphocyte ratio (PLR) was the platelet count divided by the lymphocyte count. Te prognostic nutritional index (PNI) was defned as the albumin level (g/L) + 5 × lymphocyte count per liter [14]. Te systemic infammation score (SIS) defnition was based on the combination of the preoperative albumin level and lymphocytemonocyte ratio (LMR) [15], and modifed SIS (mSIS) was an improved scoring based on the LMR optimal cut-of value [16]. Te detailed scoring of SIS and mSIS is shown in Table S1.

Statistics.
Categorical variables were presented as proportions and integers, while continuous variables were reported as medians, means (standard deviations), and maximum ranges. Te Wilcoxon rank-sum test was used for continuous variables, and Fisher's exact test was used for categorical variables. To analyze the survival diferences, patients were categorized into two groups based on the continuous variables best cut-of value using the "MaxStat" R package (maximally selected rank statistics) [17]. Kaplan-Meier survival curves were constructed according to the group diferences, which were analyzed using the log-rank test. Univariate Cox regression analysis was performed to identify variables that signifcantly impacted survival. Variables with a P value <0.05 in the univariate analysis were included in the multivariate analysis. Te multivariate analysis estimated the adjusted hazard ratio (HR) of each variable, which represents the independent impact of each variable on survival. Te prognostic abilities of the biomarkers were compared by time-dependent receiver operating characteristic (t-ROC) curves using the "time-ROC" R package [18]. Te results of stratifed survival analyzes were discussed based on clinicopathological characteristics. A two-tailed P < 0.05 was considered statistically signifcant. All statistical analyzes were determined using R software (version 3.6.3, https://www.r-project.org).

Relationship between NLR and Clinicopathological Variables in Patients with HIV-Infected CRC.
Te relationship between clinicopathological variables and NLR is given in Table 1. Tere was no signifcant correlation between NLR and the following variables, but a higher NLR indicated a lower BMI (19.9 vs. 21.3, P � 0.061) and increased tendency for lymph node metastasis (52.9% vs. 27.5%, P � 0.078) and advanced stage (III-IV: 58.8% vs. 33.5%, P � 0.082). Besides the neutrophil and lymphocyte counts, the mean value of CD8 + T lymphocytes was signifcantly higher in the low NLR group (758.8 cells per µL) than in the high NLR group (457.6 cells per µL). Tere was no signifcant diference in CD4 + T lymphocyte counts, resulting in higher CD3 + T lymphocyte counts in the low NLR group (1102.9 cells per µL) than in the high NLR group (759.8 cells per µL).

Prognosis Evaluation of Infammation Biomarkers and Clinicopathological Features in HIV-Infected CRC Patients.
Te follow-up time of overall survival (OS) and progression-free survival (PFS) were 3 to 86 months. Te cumulative three-year OS rate was 55.0%, and the cumulative three-year PFS rate was 45.0%. Te univariate analysis presented the following clinicopathological indicators associated with OS: intestinal obstruction, TNM stage, CA125 level, CD8 + T lymphocyte count, NLR, LMR, and mSIS (Alb level <40 g/L and LMR < 3.0) (P < 0.05, Table 2). In multivariate analyzes, NLR ≤ 2.8 (HR: 0.094, 95% CI: 0.02-0.45), CD8 + T lymphocyte counts ≤912 per µL (HR: 0.198, 95% CI: 0.044-0.9), III-IV stage (HR: 13.633, 95% CI: 3.824-48.603), and intestinal obstruction (HR: 5.872, 95% CI: 1.369-25.18) were independently related with OS (both P < 0.05, Table 2). Te prognostic efect of NLR, LMR, PNI, SIS, mSIS, intestinal obstruction, tuberculosis infection, TNM stage, and CA125 level was also signifcantly related to PFS in univariate analysis (  Table 3). In the multivariate analysis, NLR was found to be the most signifcant biomarker compared to the other fve biomarkers. All covariates satisfed the assumption of hazard proportionality (Figures S1 and S2). Tis study also used the time-ROC method to measure the prognostic value of each infammation marker that was preoperatively available. In both OS and PFS, the time-ROC curve of NLR was consistently better than other indexes throughout the observation period ( Figure 1). Comparing the OS after one, two, and three years of follow-up, the AUC values of NLR (0.81, 0.64, and 0.61) were higher than those of mSIS (0.75, 0.61, and 0.59), LMR (0.75, 0.61, and 0.59), and PNI (0.56, 0.53, and 0.52). Due to the small number of cases, we were unable to obtain statistical signifcance. However, NLR demonstrated a tendency to have better predictability in HIVpositive CRC patients.

Assessment of NLR Subgroups on OS and PFS Based on
Clinicopathological Features. According to the cut-of value of NLR, the OS and PFS subgroups of HIV-infected CRC patients were assessed. In CRC subgroup evaluation, the Kaplan-Meier curve exposed that gender (male and female), age (younger ≤58.2 years), BMI ≤ 20.9, stage III-IV, CD3 + T lymphocyte count >998.8 cells per µL, CD8 + T lymphocyte count >667.3 cells per µL, CD4 + T lymphocyte count >291.3 cells per µL, and CD4/CD8 ≤ 0.6 in the high NLR group (>2.8) were closely associated with poor OS (Figure 2). In addition, gender (male and female), age (younger ≤ 58.2 years), BMI ≤ 20.9, stage III-IV, CD8 + T lymphocytes count >667.3 cells per µL, CD4 + T lymphocytes count >291.3 cells per µL, and CD4/CD8 (≤0.6 and >0.6) in the high NLR group (>2.8) were also associated with a poor PFS rate ( Figure 3).   Canadian Journal of Gastroenterology and Hepatology

Discussion
Tis study investigated the prognostic value of systemic infammation biomarkers in HIV-infected colorectal cancer (CRC) patients. We found that a high neutrophil-tolymphocyte ratio (NLR) was signifcantly associated with shorter overall survival (OS) and progression-free survival (PFS) in surgically treated HIV-infected CRC patients compared to those with a low NLR. Multivariate analysis showed that NLR was an independent factor infuencing survival outcomes. Our fndings suggest that NLR can serve as a useful biomarker for surgical selection and prognosis of CRC patients infected with HIV.
In the past, individuals living with HIV (PLWH) had a higher incidence of certain malignancies such as non-Hodgkin lymphoma, Kaposi's sarcoma, anal cancer, and lung cancer, which were typically caused by opportunistic pathogens [19]. However, with the advancement of combination antiretroviral therapy (cART), non-HIV-related malignancies, including primary cancers in the liver, colorectal, breast, and prostate, have become a signifcant proportion of risk factors for life expectancy in HIV-infected people [20]. Studies conducted in Europe [21], North America [22], and Africa [23] have shown that HIV infection does not increase the incidence of colorectal cancer (CRC). However, there has been an increased expression of  [24]. Continuous HIV antigen stimulation persists even in patients receiving antiretroviral therapy, resulting in the tumor-specifc CD8+ T cells becoming positive cells of CTLA-4 through the infuence of continuously activated HIV-specifc CD8+ T cells [25]. Although our previous study found no diference in immune checkpoint expression between HIV-infected CRC and normal patients [13], these controversial clinical fndings suggest a need for more interest in the relationship between HIV-associated immunodefciency and CRC development. However, some systematic reviews have discovered that PLWH with colorectal cancer are less likely to receive cancer screening or regular treatment, resulting in higher mortality rates than HIV-negative patients [26,27]. Te three-year overall survival (OS) rate in this cohort was only 55%, which is far below the average [1]. Additionally, our previous study found that colorectal cancer patients with HIV were more likely to be diagnosed at an advanced stage than CRC patients without HIV [13]. Terefore, simple and easily achievable prognostic predictors for patients with HIVinfected CRC are required. Te earlier the subgroup of patients with a poorer prognosis can be identifed, the sooner intervention and follow-up can be provided. Te preoperative systemic infammation biomarker is reliable in various cancers, making it worth investigating in HIVpositive CRC patients. Te neutrophil-to-lymphocyte ratio (NLR) is a common biomarker used to predict colorectal cancer prognosis [9]. Neutrophils secrete cytokines and chemokines that play a crucial role in cancer progression, while lymphocytes are involved in tumor antigen presentation and tumor cell killing. In non-HIV-infected CRC patients, low circulating lymphocytes are associated with poor prognosis [28,29]. Several large-scale studies involving over 1000 CRC patients have found that a low preoperative NLR is an independent predictor of better survival outcomes [30][31][32][33][34]. Te cut-of value of NLR reported in previous studies ranged from 2 to 3.75, with a mean value of 3, which is consistent with the current study. A preoperative NLR lower than 2.8 independently predicted better OS and PFS in a multivariate analysis of HIV-infected CRC patients. Several studies have shown that circulating monocytes in cancer patients' exhibit functional changes and that increased circulating monocytes are related to poorer OS in colorectal cancer patients. In previous CRC studies, lower lymphocyte counts and higher monocyte counts were linked to worse outcomes [30,33,35]. Te current study found that an LMR below 3.0 was associated with poor OS and PFS but was not signifcant in multivariate analysis. Te serum albumin indicates cancer patients' nutritional status and infammation, impacting tumor progression. Te prognostic nutritional index (PNI) combining serum albumin and lymphocyte at a low level (<45) negatively afected the OS and PFS in CRC patients [14,36]. PNI < 45.7 was linked with worse PFS of HIVinfected CRC in univariate analysis but had no diference in OS. In recent years, a new prognostic biomarker, SIS, combining serum albumin and LMR, was explored in CRC patients. Suzuki et al. reported that increased SIS was independently related to poor prognosis [15]. However, the optimal LMR cut-of value in most studies was lower than the SIS defnition value. Some studies modifed SIS according to the optimal LMR cut-of value to obtain mSIS scoring [16]. Consequently, the scores in the current study were adjusted based on the best LMR. Te increased mSIS could predict better than the SIS in HIV-infected CRC, but they were only correlated to OS and PFS in univariate analysis (Tables 2 and 3). Platelets promote cancer progression in diverse ways in the tumor microenvironment    Figure 2: Analysis of OS based on gender, age, BMI, AJCC stage, CD3+ T lymphocytes counts, CD4+ T lymphocytes counts, CD8 T lymphocytes counts, and CD4/CD8 ratio in NLR subgroups.
[37], but PLR was not signifcantly associated with OS or PFS in HIV-infected CRC patients. On the other side, comprehensive molecular stratifcation of CRC has shown that patients with the mesenchymal subtype have worse relapsefree and overall survival [7]. However, there is still a defciency of genomic data on HIV-positive CRC patients. Terefore, obtaining this data in further clinical research is critical to provide subtype-based targeted interventions. In this study, six systemic infammation biomarkers were investigated, including NLR, in HIV-infected CRC patients. Te signifcance of the survival data was visualized using the popular time-dependent ROC analysis (t-ROC) method [38,39]. Te results of the multivariate COX and t-ROC analyzes suggest that NLR is the most efective systemic infammation biomarker for predicting outcomes in CRC patients with HIV. Tis study investigated the relationship between NLR and clinicopathological features. Te results exposed that NLR was not interrelated with several clinicopathological features, including duration of HIV infection, tumor location, histology, and mismatch repair (MMR) status. Although there was no statistical diference, a high percentage of postoperative lymph node metastasis (52.9% vs. 27.5%) and stage III-IV (58.8% vs. 32.5%) were found in the high NLR subgroup (Table 1). In the stratifed analysis, NLR had better prognostic performance in the subgroups of younger, low BMI, and stages III-IV (Figures 2 and 3). It may depict that NLR was negatively allied with disease severity in HIVinfected CRC patients. Alternatively, tumors with high infltrate of CD8 + T and CD4 + T lymphocytes had a better prognosis [40]. However, in HIV-infected patients, serum CD4 + T lymphocytes counts mainly refected a probability of various opportunistic infections, which is not a good predictor of cancer survival prognosis [5]. Hence, low NLR was associated with high CD8 + T lymphocytes counts but not with CD4 + T lymphocytes counts and CD4/CD8 ratio (Table 1) Figure 3: Analysis of PFS based on gender, age, BMI, AJCC stage, CD3+ T lymphocytes counts, CD4+ T lymphocytes counts, CD8+ T lymphocytes counts, and CD4/CD8 ratio in NLR subgroups. counts, implying immune system recovery (Figures 2 and 3). We found that adjuvant therapy did not improve patient outcomes (Tables 2 and 3). While chemotherapy agents appear to downregulate the expression of immune inhibitors in CRC [24], it is possible that some patients in this cohort did not receive the full range of adjuvant therapy. Due to potential drug interactions between chemotherapeutic agents and cARTs drugs, chemotherapy should be used with caution in CRC patients with HIV infection. Further clinical studies are necessary to explore the efcacy of adjuvant therapy.
Tis study has some limitations that should be noted. First, due to the lack of preoperative C-reactive protein (CRP) and fbrinogen data, these markers were not analyzed in this study. Second, the small number of patients and short follow-up time may have afected the precision of the results. Terefore, well-designed clinical trials are needed to validate the role of infammation biomarkers in HIV-infected CRC patients. It is also important to note that this study included nine cases of anal squamous cell carcinoma, which has a diferent pathogenesis than colorectal adenocarcinoma and is associated with immunosuppressive opportunistic infection. While previous studies have shown that low CD4 + T lymphocytes promote anal squamous carcinoma progression but not colorectal adenocarcinoma progression [41], the cases included in this study were unable to receive curative chemoradiotherapy for various reasons and were treated with surgery as a salvage option. Although these cases were rare, they were included in the study to explore the relationship between infammation markers, surgical treatment, and prognosis. Terefore, further studies with more cases are needed to better understand the diferences in the future.

Conclusion
With the availability of combination antiretroviral therapy (cART), people living with HIV can now have a similar life expectancy to that of the general population. Our fndings demonstrate that the preoperative neutrophil-tolymphocyte ratio (NLR) refects the immune status and is an independent predictor of both overall survival (OS) and progression-free survival (PFS) in HIV-infected colorectal cancer patients. Tis simple prognostic biomarker has the potential to be used as a preoperative risk stratifcation tool, which could enable the development of individualized treatment plans for patients with HIV and colorectal cancer.

Data Availability
Te data used to support the fndings of this study are available from the corresponding author upon request. Li Deng, Yanhui Si, and Qian Wu share frst authorship.

Conflicts of Interest
Te authors declare that there are no conficts of interest. Table S1: Te defnition of SIS and modifed SIS (mSIS). Table S2: Baseline clinicopathologic features and preoperative laboratory tests of CRC patients living with HIV. Figure S1: Te proportional hazard test for each covariate included in the multivariate Cox analysis of OS. Figure