Searching for prognostic markers for Stage I epithelial ovarian cancer: A role for systemic inflammatory markers

To determine the prognostic role of systemic inflammatory markers for Stage I epithelial ovarian cancer (EOC).


| INTRODUC TI ON
Epithelial ovarian cancer (EOC) represents the most lethal gynecologic cancer.Over 70% are diagnosed at an advanced stage (FIGO [International Federation of Gynecology and Obstetrics] Stage III-IV) and, despite the promising results of new targeted therapies (e.g., inhibitors of the enzyme poly-ADP ribose polymerase-PARPi), the overall survival (OS) remains low. 1 However, about 30% of patients are diagnosed with early-stage EOC (FIGO Stage I-II) with a risk of relapse ranging between 10% and 50% and a 5-year OS greater than 70% in most studies. 2,3e standard treatment for FIGO Stage I EOC is total abdominal hysterectomy, bilateral salpingo-oophorectomy, pelvic and aortic lymph node dissection, omentectomy, peritoneal biopsies, and washing. 4For adjuvant treatment, platinum-based chemotherapy improves both recurrence-free survival and OS and it is usually recommended, especially in high-risk patients (Stage IA grade 3, IB or IC grade 2 or 3, clear cell histology); 2 however, a Cochrane systematic review found that the survival benefits in women with high-risk tumors are based on low-quality evidence and there is still uncertainty for lower/intermediate-risk early-stage disease. 5Several efforts have been made to stratify patients based on different prognostic factors to achieve more personalized risk estimations. 3,6,7Stage I EOC is characterized by subtype-specific molecular alterations that affect tumor aggressiveness, [8][9][10] so tumor molecular profiling is an option, but there is also a need for inexpensive, easy, and reproducible markers to help predict the long-term behavior of these tumors.
In recent years, the prognostic role of several inflammatory markers from the patient's blood count before starting treatment has been evaluated in different cancers, including EOC. 11 However, the prognostic role of systemic inflammatory markers has not been clarified in the specific subgroup of Stage I EOC.Based on this background, this study aims to evaluate the potential prognostic role of systemic inflammatory markers for Stage I EOC.

| MATERIAL S AND ME THODS
We performed a retrospective study of all patients surgically treated for FIGO Stage I EOC at the Department of Surgical Sciences, S.
Anna Hospital, University of Turin, from January 1993 to December

2016.
We excluded patients with borderline tumors, non-epithelial ovarian cancer, age younger than 18 years, and incomplete clinical data and/or follow up.We also excluded patients with concurrent infection, immunosuppressive therapy, or hematologic disorders at the time of the surgery.All cases were revised by a dedicated pathologist (LB) according to the WHO 12 Classification of Tumors of Female Reproductive Organs, while the stage of the disease was determined with the FIGO staging system. 13 calculated inflammatory markers as follows: neutrophil-tolymphocyte ratio (NLR) by dividing the absolute neutrophil count by the absolute lymphocyte count, platelet-to-lymphocyte ratio (PLR) by dividing absolute platelet count by absolute lymphocyte count, and systemic immune inflammation index (SII) was estimated as (platelet count × neutrophil count)/ lymphocyte count.
We chose the following cut-off values for serum biomarkers based on the literature data: NLR: 3, 14,15 PLR: 169, 15 SII: 730, 14,15 and CA125: 30 U/mL. 16e study was submitted to and approved by the Ethics Institutional Review Board for "Biobanking and use of human tissues for experimental studies" of the Department of Medical Sciences of the University of Turin, protocol n.DSM-ChBU no.6/2020.
Due to the retrospective nature of this study, no written informed consent from the patients was necessary, as stated by our Ethics Institutional Review Board.All the cases were recorded in a dedicated database and pseudonymized.Differences among different survival groups were tested using Pearson's χ 2 test or Fisher exact test, as appropriate.For the variable "age", the Shapiro-Wilk test was used to test the normality of distribution, and the Mann-Whitney U test was used for the comparison.Survival outcomes (CSS and DFS) were analyzed by the Kaplan Meier method and by univariate and Cox proportional hazards models.Significant variables (P values less than 0.05) were included in the multivariate analysis.Analyses were conducted with a 95% confidence interval (CI), and a two-sided P value of 0.05 was considered statistically significant.

| RE SULTS
One hundred and seventy-six women treated for apparent Stage I EOC in our center met the inclusion criteria.The mean age of the whole cohort was 57 years (20-85 years).Most of the patients had an FIGO Stage IC EOC (n = 99, 56%).

Clinical characteristics
Total The OS was 68%, but only 17 patients died of disease (CSS 90.4%).
Table 1 shows the characteristics of the whole cohort and the comparisons between patients who experienced recurrence and those who remained disease-free and between patients who died of disease and patients who did not.
The following variables had a significantly different distribu-  The prognostic role of NLR, PLR, and SII was confirmed by Kaplan-Meier curves for DFS (Figure 1) and CSS (Figure 2).predictors of CSS (Table 3).

| DISCUSS ION
The aim of this study was to search for risk factors that affect recurrence and CSS in a large number of consecutive patients with Stage I EOC who received surgical treatment in a referral center.The main finding of our study was that serum biomarkers such as NLR, PLR, and SII are associated with survival outcomes in this setting.It is known that immune system cells can affect all stages of tumor development by releasing pro inflammatory cytokines. 17In particular, neutrophils may promote genomic instability by releasing reactive oxygen species and favor extracellular matrix remodeling and cancer cell invasion by the production of proteases, growth factors, and oncostatins. 18Moreover, neutrophils may stimulate tumor angiogenesis and suppress anti-tumor adaptive immunity. 18Similarly, platelets may support cancer progression in different ways: secreting pro-inflammatory factors (CXCL1, CXCL4, CXCL5, CXCL7, CXCL12, and interleukin-8), contributing to thrombosis and vascular inflammation, and promoting the recruitment of neutrophils and monocytes. 19On the other hand, tumor-infiltrating T lymphocytes may have antitumor activities in immunogenic tumors, including EOC. 20 Because of the role of inflammatory cells in cancer development, the prognostic significance of these biomarkers has also been evaluated in EOC.Nie et al. 21 better survival.Moreover, the authors observed that NLR could be a predictive factor of bevacizumab efficacy and bevacizumab seems to be detrimental in patients with a high SII. 14The same study group also showed that NLR ≥3 and SII ≥730 are significantly associated with worse OS in platinum-sensitive EOC. 15 A high NLR value was related to worse survival in another retrospective analysis on 397 EOC regardless of BRCA-mutation status. 22More recently, the role of preoperative systemic inflammatory markers was investigated in early-stage EOC.In a study including 359 patients ≥3 and an SII ≥1000 were associated with worse 3-year DFS, and an SII ≥1000 was associated with worse 3-year OS. 23 The unfavorable prognostic role in terms of PFS and OS for NLR and PLR has also been confirmed in a meta-analysis including 2919 patients with EOC. 11r results agree with previous studies suggesting a significant prognostic role for neutrophils, lymphocytes, and platelet counts also for Stage I EOC: at multivariate analysis, SII was found to be associated with shorter PFS, while NLR and PLR were associated with CSS.Moreover, our study is the first to evaluate the association between systemic inflammatory markers and CSS instead of OS.CSS was analyzed because we considered it a more accurate outcome for evaluating a potential prognostic variable as it removes competing causes of death 24 and this is especially important considering the long median follow up of our study and the low mortality of Stage I EOC.Therefore, analysis of OS in this specific setting could be affected by significant biases preventing a correct assessment of the analyzed prognostic factors.
Systemic inflammatory markers may change in the presence of infections, hematologic disorders, or some immunosuppressive or immunomodulatory drugs, so they may not be reliable in these situations.Regarding other gynecologic inflammatory conditions such as endometriosis, one study measured the NLR in these patients and found it higher than in controls, but the patients were younger (mean age 33 years) and the NLR values were lower than in our study (mean 2.66, range 2.43-2.89),so this condition should not affect the prognostic role of these biomarkers in EOC. 25 The main limitation of this study is related to its retrospective design and to the potential differences in terms of patient management over the years due to the time range used to collect the study cohort.Abbreviations: CI, confidence interval; CSS, cancer-specific survival; DFS, disease-free survival; HG-S, high grade serous; HR, hazard ratio; NLR, neutrophil-to-lymphocyte ratio; NS, not significant; PLR, platelet-lymphocyte ratio; SII, systemic inflammatory index.

For
the selected patients, we collected the following clinical and histopathologic data: (1) age at diagnosis; (2) histopathologic features of ovarian cancer including tumor histotype, grade, and FIGO Stage; (3) type of surgical treatment, (4) preoperative total count of neutrophil, lymphocyte, platelets, and CA125 U/mL, and (5) date of death or last follow up.All patients had preoperative complete physical and gynecologic examinations, gynecologic ultrasound examination, chest X-ray, computed tomography scan, and routine blood and urine analysis.Surgical treatment consisted of total hysterectomy with bilateral salpingo-oophorectomy, careful abdominal and pelvic palpation and exploration, random peritoneal sampling biopsies, omentectomy, and peritoneal washing.Para-aortic and pelvic lymph node dissection were performed according to the patient and tumor characteristics.Preservation of the uterus and of one ovary was performed in young patients who wanted to preserve fertility.Adjuvant treatment was platinum-based chemotherapy according to the tumor and patient's clinical characteristics.
Time for OS was stopped at the time of death or the last follow up with a cut-off date in December 2021.We obtained disease status or cause of death from clinical charts and/or cancer registry data of our region (Piedmont Cancer Registry, Centre for Epidemiology and Prevention in Oncology in Piedmont).For cancer-specific survival (CSS), we counted only cancer-associated deaths, whereas other deaths unrelated to Stage I EOC were noted.Disease-free survival (DFS) was defined as the time interval from the date of the Stage I EOC diagnosis to the date of first recurrence or last follow up.Statistical analyses were performed using IBM SPSS version 25 (IBM, Armonk, NY, USA) software.Continuous variables were reported as mean and range, and categorical variables as frequency and percentage.

(
only for recurrence), adjuvant chemotherapy, CA125, NLR, PLR, and SII.Table 2 shows the univariate Cox regression model comparing the DFS and CSS for each analyzed prognostic variable: FIGO Stages IB and IC, high-grade serous histology, tumor grade 3, adjuvant chemotherapy, CA125, NLR, PLR, and SII were associated with DFS.All variables associated with DFS, except for high-grade serous histology, also correlated with CSS.
investigated the role of several prognostic factors related to progression-free survival (PFS) and OS in a retrospective cohort of 553 EOC patients and found that preoperative high-values of NLR and SII were independent factors related to poor survival in both groups, whereas PLR did not show F I G U R E 1 Kaplan-Meier curves for DFS according to (a) NLR, (b) PLR, and (c) SSI.DFS, disease-free survival; NLR, neutrophil-tolymphocyte ratio; PLR, platelet-lymphocyte ratio; SII, systemic inflammatory index.prognostic value.Also, a multicenter, retrospective analysis by the MITO Group on 375 patients with FIGO Stage III-IV EOC showed a correlation between low inflammatory markers (NLR, PLR, SII) and

F I G U R E 2
Kaplan-Meier curves for CSS according to (a) NLR, (b) PLR, and (c) SSI.CSS, cancer-specific survival; NLR, neutrophil-tolymphocyte ratio; PLR, platelet-lymphocyte ratio; SII, systemic inflammatory index.The main strength of this study is that, as far as we know, this is the first study that evaluates the prognostic role of inflammatory markers in Stage I EOC.Overall, Stage I EOC is characterized by a relatively good prognosis, but as a subset of patients shows an unfavorable outcome, there is a strong need to define novel and effective prognostic markers to promptly identify them and tailor their adjuvant treatments.Recently, a prognostic score based on the genome distribution of somatic copy number variations in a retrospective cohort of Stage I EOC was proposed; 9 however, systemic inflammatory indices are easily obtainable from patients' routine blood samples and may represent an inexpensive and reproducible marker for prognostic stratification and treatment tailoring.For example, systemic inflammatory indices could help to identify patients who are more likely to benefit from adjuvant chemotherapy or targeted therapies, such as bevacizumab or PARPi, also for high-risk Stage I EOC.Moreover, systemic inflammatory indices could be used to monitor the response to treatment and the risk of recurrence, as well as to guide the frequency and duration of follow up.Systemic inflammatory markers should be interpreted with caution in the presence of concurrent events that can alter the immune response, as they could lead to false results and hence invalidate their prognostic value.Even though increasing evidence is emerging on the prognostic role of these biomarkers in EOC, their validation in large prospective studies is needed for their implementation into clinical practice.These studies should also define the optimal cut-off values and time points for measuring these biomarkers, as well as their interaction with other prognostic factors.AUTH O R CO NTR I B UTI O N S Fulvio Borella, Luca Bertero, and Giorgio Valabrega conceived and designed the study; Fulvio Borella and Luca Bertero wrote the original draft and performed the statistical analyses; Luca Bertero revised the pathologic slides; Stefano Fucina collected the data; Paola Cassoni and Chiara Benedetto performed the project administration and supervision; writing-review and editing were performed by Fulvio Borella, Luca Bertero, Giorgio Valabrega, and Paola Cassoni.All authors have read and agreed to the published version of the manuscript.TA B L E 3 Multivariate analysis of variables associated with DFS and CSS.
Univariate analysis of variables associated with DFS and CSS.