Multicenter Study of Controlling Nutritional Status (CONUT) Score as a Prognostic Factor in Patients With HIV-Related Renal Cell Carcinoma

Objective In recent years, the controlled nutritional status (CONUT) score has been widely recognized as a new indicator for assessing survival in patients with urological neoplasms, including renal, ureteral, and bladder cancer. However, the CONUT score has not been analyzed in patients with HIV-related urological neoplasms. Therefore, we aimed to evaluate the prognostic significance of the CONUT score in patients with HIV-related renal cell carcinoma (RCC). Methods A total of 106 patients with HIV-related RCC were recruited from four hospitals between 2012 and 2021, and all included patients received radical nephrectomy or partial nephrectomy. The CONUT score was calculated by serum albumin, total lymphocyte counts, and total cholesterol concentrations. Patients with RCC were divided into two groups according to the optimal cutoff value of the CONUT score. Survival analysis of different CONUT groups was performed by the Kaplan–Meier method and a log rank test. A Cox proportional risk model was used to test for correlations between clinical variables and cancer-specific survival (CSS), overall survival (OS), and disease-free survival (DFS). Clinical variables included age, sex, hypertension, diabetes, tumor grade, Fuhrman grade, histology, surgery, and CD4+ T lymphocyte count. Result The median age was 51 years, with 93 males and 13 females. At a median follow-up of 41 months, 25 patients (23.6%) had died or had tumor recurrence and metastasis. The optimal cutoff value for the CONUT score was 3, and a lower CONUT score was associated with the Fuhrman grade (P=0.024). Patients with lower CONUT scores had better CSS (HR 0.197, 95% CI 0.077-0.502, P=0.001), OS (HR 0.177, 95% CI 0.070-0.446, P<0.001) and DFS (HR 0.176, 95% CI 0.070-0.444, P<0.001). Multivariate Cox regression analysis indicated that a low CONUT score was an independent predictor of CSS, OS and DFS (CSS: HR=0.225, 95% CI 0.067-0.749, P=0.015; OS: HR=0.201, 95% CI 0.061-0.661, P=0.008; DFS: HR=0.227, 95% CI 0.078-0.664, P=0.007). In addition, a low Fuhrman grade was an independent predictor of CSS (HR 0.192, 95% CI 0.045-0.810, P=0.025), OS (HR 0.203, 95% CI 0.049-0.842, P=0.028), and DFS (HR 0.180, 95% CI 0.048-0.669, P=0.010), while other factors, such as age, sex, hypertension, diabetes, tumor grade, histology, surgery, and CD4+ T lymphocyte count, were not associated with survival outcome. Conclusion The CONUT score, an easily measurable immune-nutritional biomarker, may provide useful prognostic information in HIV-related RCC.


INTRODUCTION
RCC is the most common pathologic type of renal cancer and the seventh most common tumor, accounting for 2% to 3% of all cancers (1). Twenty percent of newly diagnosed RCC patients have advanced disease, and approximately 30% experience local or distant disease recurrence after surgery for localized RCC (2). In recent years, relatively few infection cases of HIV-related RCC have been reported worldwide. Patients with such RCC have concurrent immune infection. Human immunodeficiency virus (HIV) infects human dendritic cells and macrophages and activates CD4+ T lymphocytes, resulting in disruption of the immune system, so the tumor incidence and mortality differ from those in ordinary RCC patients (3). RCC is more common in HIV-infected individuals than in age-matched non-HIVinfected individuals and is a common cause of morbidity and mortality. Possible mechanisms for this increased risk include reduced immune surveillance, direct effects of viral proteins, or cytokine dysregulation (4,5). With the widespread application of and tremendous progress in early activation of highly active antiretroviral therapy (HAART), both virological suppression and immune recovery in patients with HIV-related RCC have been maintained at a good level (6). However, non-AIDSdefining cancers (non-ADCs), including urinary cancers, anal cancers, lung cancers, breast cancers and skin cancers, are still three times more frequent (7). If patients with HIV-related RCC can be assessed early, their survival could be significantly improved. Therefore, it is important to develop off-the-shelf biomarkers that can predict and even modify tumor outcomes based on risk stratification (8).
Immunological status comprising inflammatory and nutritional status, remains an important predictor of prognosis in patients with malignant tumors (9). Several biomarkers, such as the prognostic nutritional index (PNI) and the neutrophil to lymphocyte ratio (NLR), have been reported to be independent prognostic factors (10)(11)(12). Recently, the CONUT score, which is calculated from serum albumin, total lymphocyte counts, and total cholesterol concentration, has gained attention as a biomarker for predicting survival in patients with multiple cancers. A high CONUT score means lower levels of albumin, lymphocytes, and cholesterol, which are often associated with poorer nutritional and immune status in patients and may lead to poorer survival (13). Maintaining optimal nutritional status can greatly improve quality of life while reducing comorbidities, progression of HIV infection, and HIV-related mortality (14,15). In addition, good nutrition also helps HIV-infected patients absorb HIV drugs (16). The effects of poor nutritional status and HIV are synergistic and interrelated, thus amplifying their respective harmful effects (15,17). Increasing evidence suggests that, in addition to the genetic basis, host nutritional status and inflammatory responses also play an important role in cancer development and progression (18). At present, some articles suggest that the CSS, OS and DFS of patients with 5-year ordinary RCC (non-HIV related) in the low-CONUT group are significantly higher than those in the high-CONUT group (8,(19)(20)(21)(22)(23)(24), but some articles suggest that a high CONUT score is not related to the prognosis of patients with ordinary RCC (25). However, there are no articles about the relationship between CONUT score and HIV-related RCC. To the best of our knowledge, this is the first multicenter study to evaluate the prognostic value of the CONUT score in HIV-related RCC.

Patients
We performed an open-label, retrospective, multicenter, cohort study. A total of 106 patients with HIV-related RCC who underwent radical nephrectomy or partial nephrectomy were included. All participants underwent preoperative urological CT examination showing a renal carcinoma volume ≤7 cm between 2012 and 2021. We excluded patients with ordinary RCC without HIV infection, patients with lymph node metastasis or distant metastasis, and patients with no follow-up results. All enrolled patients had provided blood samples with results for serum albumin, total lymphocyte counts, and total cholesterol concentration one week before surgery and were treated with HAART and monitored for associated CD4+ T lymphocyte count.
Pathological stage was determined according to the 2010 TNM grade and tumor grade according to the Fuhrman grading system. This study was in accordance with the Helsinki Declaration and approved by the Ethics Review Committee of all included hospitals. During follow-up, patients or their next of kin were informed of the study in detail, and verbal consent was obtained. All data are kept confidential. and metastasis is equal to the first discovery of lymph node or distant organ metastases (lung metastasis, brain metastasis, liver metastasis, etc.). Death was confirmed by relevant information from the hospitals or notification by the patient's family during telephone follow-up.

Study Endpoints
We considered CSS, OS, and DFS as the end points of the study (in months). CSS was defined as the time from the date of surgery to cancer-related death. OS was defined as the time from the date of surgery to the death of the individual from any cause. DFS was defined as the time from the date of surgery to radiologically or histologically confirmed recurrence or metastasis.

CONUT Score, PNI and NLR
The CONUT score was calculated by serum albumin, total lymphocyte counts, and total cholesterol concentration ( Table 1). The optimal cutoff value of the CONUT score was determined using the receiver operating characteristic (ROC) curve and the maximum Youden index value. The PNI was calculated as 10 × serum albumin (g/dl) + 0.005 × total lymphocyte count (per mm3). The NLR was calculated as the ratio of the number of neutrophils to the number of lymphocytes.

Statistics
A chi-square test was used to analyze the correlations between the CONUT score and variables including age, sex, hypertension, diabetes, tumor grade, Fuhrman grade, histology, surgery, and CD4+ T lymphocyte count. Kaplan-Meier survival curves were plotted to estimate CSS, OS, and DFS. The predictors of CSS, OS and DFS were determined by univariate analysis, a Cox proportional risk model was used for multivariate analysis evaluation, and variables with P<0.05 in univariate analysis were included in subsequent multivariate analysis. GraphPad Prism Version 9 (GraphPad Software, La Jolla California USA, www.graphpad.com) was used to generate survival curves. Statistical analysis and ROC curves mapping were performed using SPSS version 23 (SPSS Inc., Chicago, IL, USA).

CONUT Score and Its Cutoff Value
According to ROC analysis, the Youden index was used to determine the optimal cutoff value of the CONUT score as 3 (AUC: 0.746, 95% CI: 0.638-0.855, P<0.001, Figure 1. The area under the receiver operating characteristics curve, AUC). The CONUT score was assessed by dichotomous variables (low: <3, high: ≥3).

Clinicopathological Features
Of the 106 patients enrolled, with a median age of 51 (IQR 27-75) years at the time of surgery, 93 were males, 13 were females, 80 underwent radical nephrectomy, and 26 underwent partial nephrectomy. Among them, 104 cases were clear cell carcinoma, 87 cases were T1N0M0, and 19 cases were T3N0M0. For the Fuhrman classification, 65 were grades I-II, and 41 were grades III-IV. The median CD4+ T lymphocyte count value was 435 (IQR 48-1536) cells/µl. The highest preoperative viral load was 1,018,049 copies/mL, and the lowest was undetectable ( Table 2). The CONUT score was high in 45 cases (42%) and low in 61 cases (58%), and was closely correlated with the Fuhrman grade. A low CONUT score was significantly associated with lower Fuhrman grade (I-II vs III-IV, 66.2% vs 43.9%, respectively, p=0.024). The CONUT score had no significant correlation with age, sex, hypertension, diabetes, tumor grade, histology, surgery, or CD4+ T lymphocyte count (P>0.05) ( Table 3).

Survival Outcome
The median postoperative follow-up time of CSS and OS was 41 (IQR 6-105) months, and that of DFS was 41 (IQR 4-105) months. In the high-and low-CONUT groups, the 5-year CSS rates were 47.79% and 85.54% (P<0.001) (Figure 2A As shown in Table 4, patients with lower CONUT scores had better CSS ( Table 4), while other factors, such as age, sex, hypertension, diabetes, tumor grade, histology, surgery, and CD4+ T lymphocyte count, were not associated with survival outcome.

Compare CONUT Score With Other Biomarkers in Patients With HIV-Related RCC for Survival Prediction
The PNI and NLR values of 106 patients are also shown in Table 2. The ROC curve with the most sensitive and specific cutoff values of PNI and NLR is also shown in Figure 1. We compared the AUCs for predicting 5-year OS by CONUT, PNI and NLR. Among the prognostic factors, the CONUT has the highest AUC (0.746). The AUC score of PNI and NLR in relation to 5-year OS was 0.682 (95% CI 0.553-0.811) and 0.674 (95% CI 0.545-0.803), respectively.

DISCUSSION
In this study, patients with surgically treated HIV-related RCC with high CONUT scores had significantly shorter CSS, OS, and DFS than patients with low CONUT scores. Multivariate analysis further showed that the CONUT score was an independent factor influencing these survival outcomes. In addition, a low Fuhrman grade was significantly associated with survival outcomes. This study is the first to study the prognostic factors of patients with HIV-related RCC, which has certain reference significance for the surgical treatment selection of patients with HIV infection and the prognosis of patients with HIV-related RCC. Among people living with HIV, non-HIV-related morbidity and mortality are becoming more common, and non-HIV hypotoxicity is becoming an important source of mortality (26).
In recent years, there has been a significant increase in the incidence of malignant tumors in HIV-infected people because of the increased life expectancy associated with HAART, and urologists are increasingly likely to encounter HIV-infected patients with the same urinary problems as the general population (4). Strictly evaluating the surgical indications of patients with HIV-related RCC and early surgical treatment are crucial for patient prognosis. HIV mainly invades human CD4+ T lymphocytes, causing a reduction in their number and functional defects, thereby resulting in low immune function and an increasing incidence of various opportunistic infections. Although early literature on surgical outcomes in HIV-positive patients suggested an increased risk of perioperative complications (27), recent studies have shown that most procedures can be performed safely in HIV-positive patients with appropriate preoperative evaluation of CD4+ T lymphocyte count and viral load (28). Therefore, preoperative routine examination of CD4+ T lymphocytes in patients with HIV-related RCC is an aspect of evaluating whether patients can tolerate surgery, but CD4+ T lymphocytes do not serve as a good predictor of the survival prognosis of these patients; therefore, it is necessary to find prognostic predictors for patients with HIV-related RCC.
The CONUT score was determined by serum albumin, total lymphocyte counts, and total cholesterol concentration. Serum albumin is an indicator that can reflect patients' nutritional status and is closely related to patients' surgical tolerance and postoperative recovery. A lower level of serum albumin means the loss of immunity (29). HIV RNA levels and CD4+ T lymphocyte counts provide some prognostic information about HIV disease progression, but data published by Shruti H Mehta suggest that serum albumin levels provide more prognostic  information than RNA levels and CD4 counts. Low levels of serum albumin not only reflect the general health status of HIV-infected patients but also reflect the effects of HIV on the host (30). Several studies of HIV seroepidemic cohorts have shown that low serum albumin is associated with all-cause mortality, even among individuals receiving HAART (31,32). Sabin et al. demonstrated an independent role of serum albumin detected shortly after HIV serotransformation in all-cause mortality and a smaller but still significant role in AIDS progression. These associations were independent of CD4+ T lymphocyte count and HIV viral load (33).
Lymphocytes are believed to have antitumor ability by affecting the growth, migration, and apoptosis and inducing the cytotoxicity of tumor cells. The high density of lymphocytes reflects the immune response of tumors (34). F A Post et al. found that total lymphocyte count and CD4+ T lymphocyte count were equally good predictors of HIV infection disease progression, and severe lymphocytopenia (total lymphocyte counts <750/µl) predicted low survival and may reflect high susceptibility to opportunistic infections (35,36). Moses R Kamya's results showed a strong correlation between total lymphocyte counts and CD4+ T lymphocytes. Similar correlations between total lymphocyte counts and CD4+ T lymphocytes have been reported in North America, England and India (37).
Low cholesterol levels are associated with cancer outcomes. Cholesterol affects the structure and function of the membrane, such as membrane protein activity and membrane fluidity, thus affecting the ability of immunoactive cells to fight cancer cells (38). Dyslipidemia has also been observed in untreated HIVinfected patients, suggesting that HIV infection itself has deleterious metabolic effects (39). In the Swiss HIV cohort study, the use of HIV protease inhibitors was found to be associated with an increase in plasma total cholesterol (40). In the SMART study, discontinuation of HAART led to a reduction in total cholesterol concentration (41). After the initiation of HAART, lipid abnormalities in HIV patients become more obvious, and hypercholesterolemia is the most related disease (42)(43)(44). Serum total cholesterol concentration was a correlated and independent predictor of HIV RNA load, CD4+ T lymphocyte count and WHO clinical stage. In this era of testing and treatment, it is possible to use low serum total cholesterol concentration as a marker to predict the efficacy of HAART (45).
CONUT is mainly associated with malignant tumors and survival prognosis through the nutritional immune pathway. For HIV-infected patients, nutritional and immune functions are already low, and these patients have an increased risk of RCC, which is largely due to the loss of control of the oncogenic genome and the high prevalence of exposure to other carcinogens (46). Adam B Murphy et al. found that the frequency of metastatic disease in an HIV-related cohort was 2.5 times that observed in a non-HIV-related cohort, although this difference did not reach statistical significance (47). Wee Loon ONG's study showed that a total of five HIV-related RCC patients in Australia's statewide HIV centers underwent surgery without any perioperative complications (48). In metastatic clear cell RCC, targeted therapy or immunotherapy may interact with antiretroviral drugs to some extent (49). Annah B Layman et al. found a similar incidence, clinical presentation and outcome of  RCC in HIV-infected and non-HIV-infected populations and no association between CD4+ T lymphocyte count and RCC risk at the onset of AIDS (50). Our study has some limitations. First, the follow-up time of some patients was short, at only half a year, so it would be more meaningful to extend the follow-up time to ensure the accuracy of the results. Furthermore, the duration of treatment with antiviral drugs may also be a prognostic factor, but since most of the included patients were uncertain about the duration of HAART, no reliable data were obtained.

CONCLUSION
HIV-infected patients eligible for HAART have a potentially normal life expectancy. Therefore, diseases such as RCC and other malignancies should be treated in the same way as those in non-HIV-infected patients. As it becomes increasingly possible to operate on HIV-infected patients undergoing HAART, CONUT's role in predicting survival for HIV-related RCC is becoming increasingly important. The preoperative CONUT score not only objectively reflects the nutritional and immune Bold values indicate statistical significance in univariate and multivariate analysis which had been detailed in the "Results" section. RCC, renal cell carcinoma; OS, overall survival; CSS, cancer-specific survival; DFS, disease-free survival; HR, hazard ratio; CONUT, controlling nutritional status; RN, radical nephrectomy; PN, partial nephrectomy.
statuses of the host but also is an independent predictor of CSS, OS and DFS in patients with HIV-related RCC.

DATA AVAILABILITY STATEMENT
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by Beijing You'an Hospital Affiliated to Capital Medical University. The patients/participants provided their written informed consent to participate in this study.

AUTHOR CONTRIBUTIONS
WX participated in manuscript preparation and writing. XH provided suggestion and edits. YZ (4th author) conceptualized, wrote, and revised manuscript. YZ (2nd author) and HW provided relevant patients data of their hospitals and offered suggestions for revising the article. All authors contributed to the article and approved the submitted version.