INTRODUCTION
Breast cancer was reported to be the second most frequently diagnosed cancer in 2018, and it is the most common cause of cancer-related death in women1. Neoadjuvant therapy, which is an important modality in the treatment of breast cancer, is defined as all systemic treatments applied to the breast tumor before surgical operations2. Although discussions continue, it has been found that patients with a complete response to neoadjuvant chemotherapy, especially in HER2-positive and triple-negative breast cancers, experience better results, have a longer disease-free survival, and have a higher overall survival compared to those without any response3,4.
Calprotectin is a heterodimeric calcium-binding protein consisting of S100A8 and S100A9 subunits from the family of S100 proteins5. Calprotectins are expressed in a wide variety of cell types, but are particularly abundant in myeloid cells such as macrophages in the early differentiation stage, neutrophils, monocytes, and keratinocytes. After being released from activated granulocytes, S100A8/S100A9 binds to cell surface receptors that trigger signaling pathways associated with inflammatory processes, with a cytokine-like behavior pattern, and plays a critical role in numerous cellular processes such as cell cycle progression, cell survival, proliferation, differentiation and cell migration6,7.
In this study, it was aimed to compare the serum calprotectin levels of patients with breast cancer planned to receive various neoadjuvant treatments and the serum calprotectin levels of patients without breast cancer, to try to determine the cut-off value for calprotectin if found to be associated with the diagnosis, to investigate the relationship between the calprotectin level and the response to the treatment in the breast cancer group, and to evaluate the relationship between calprotectin and other blood parameters, already known to be related to the inflammatory processes.
MATERIALS AND METHODS
In this study, which is planned to be a specialty thesis in medicine, the screening model, one of the quantitative research models, and the relational screening model, one of the sub-types of survey models, were used.
Selection and Definition of Cases
Our study included 89 patients who were admitted to Tekirdağ Namık Kemal University Hospital, Department of Internal Medicine Diseases, Unit of Medical Oncology and Department of General Surgery, Breast Outpatient Clinic between 15.03.2019 and 19.10.2020. While the “patient arm” of our study was formed with 69 patients diagnosed with breast cancer as a result of their application to the relevant polyclinic and planned to receive neoadjuvant treatment in the multidisciplinary council of our hospital, the “control group” of our study was also formed with 20 patients who applied to the same polyclinics with the suspicion of breast cancer but were not found to have malignant pathology. Patients over the age of 18 years, who gave the consent form prepared for participation in the study, were included in the study. Those who were pregnant, who did not sign the voluntary consent form, who had additional malignancies, who had a diagnosis or sign of infection during sampling and evaluation, and who had hematological or rheumatological disorders were excluded from the study. Patient data were obtained from the hospital archive and the hospital electronic recording system.
Calprotectin Measurement
After the decision was made to include the patients in the study and their written and verbal consents were obtained, 5 mL of venous blood sample was taken to study the calprotectin levels of the patient in addition to the routine blood tests. These tubes were centrifuged at 2000 G for 20 minutes and at room temperature. After centrifugation, the supernatant was aliquoted into two microtubes as it helped to partition the separator gel in the tube. These serum samples, which were taken into microtubes, were labeled and stored at -20 °C to be stored until the time of batch analysis. Analysis of serum calprotectin levels from samples was performed with a commercially produced kit based on the enzyme-linked immunosorbent analysis method (Bioassay Technology Laboratory, Cat. No: E4010Hu). It was ensured that the person performing the experiments was blinded in terms of the knowledge of the groups in which the studied blood samples were included. After all samples were collected, they were analyzed with the same kit. Evaluation was made by following all the instructions and experimental processes specified in the commercial kit.
Statistical Analysis
In our study, statistical evaluations were made using a package program called Statistical Package for the Social Sciences (SPSS) (IBM SPSS Statistics 24). Categorical measurements were summarized as numbers and percentages, and continuous measurements as median and standard deviation. Comparison between two qualitative clinical variables was analyzed with the “Mann-Whitney U test”, “independent sample-t test” and “Kruskal-Wallis H test”. The “Fisher’s exact” and “Pearson-c2” crosstabs were used according to expected value levels in the examination of the relationships between two qualitative variables. The relationship of calprotectin level with treatment response was evaluated with ROC curve and ROC-AUC analysis. Logistic regression model was used in predictive factor analysis for chemotherapy response.
RESULTS
Eighty-nine patients (69 cancer, 20 control) were included in the study. The median age was 48 [minimum (min): 24-maximum (max): 73] years in patients with breast cancer and 44.5 (min: 19-max: 68) years in the control group, and the ages of the two groups were similar (p=0.599). Of the cancer patients, 33% were ER negative, 46% were PR negative, and 54% were HER2 negative (Table 1).
In breast cancer patients, the mean calprotectin level was found to be 28.63±30.5, the median level was 16.5 (min: 6.7-max: 160.7). In the control group, the mean calprotectin level was detected to be 16.09±6.1 (min: 8.7-max: 27.4) and there was no statistical difference between the two groups (p=0.072). A statistically significant difference was found in terms of calprotectin values according to Ki67 categories (Z=-20.043; p=0.041). Calprotectin values of those with Ki67 level >20 were statistically significantly higher than those with ≤20.
Calprotectin level, which predicted complete response, was not diagnostically predictive of complete response when analyzed with ROC-curve (p=0.587, AUC: 0.453) (Figure 1).
Parameters that could predict complete chemotherapy response were evaluated with logistic regression analysis. There was no correlation between calprotectin level and complete response [Odds ratio (OR): 1.049 95% confidence interval (CI): 0.982-1.120, p=0.153]. There was a positive correlation between increasing age and complete response (OR: 1.092, 95% CI: 1.022-1.168, p=0.010). Estrogen negative (OR: 0.284 95% CI: 0.082-0.984, p=0.047) and progesterone negative (OR: 0.238 95% CI: 0.067-0.841, p=0.026) were more likely to have complete responses (Table 2).
DISCUSSION
In our study, the value of serum calprotectin level in the diagnosis of breast cancer and response to treatment was investigated. In our analyses, we found that the calprotectin level was higher in those with high Ki67. Although there was a numerical difference in calprotectin levels between the included patient and control groups, no statistically significant difference was found, whereas pre-treatment calprotectin level could not predict treatment response. Age, estrogen receptor and progesterone receptor were predictive for complete response to neoadjuvant chemotherapy.
In the study of Khorrami et al.8 investigating the value of calprotectin for the diagnosis of breast cancer, it was reported that serum calprotectin level could be a valuable marker in the diagnosis of cancer patients. According to the study of Gunaldi et al.9, which was published in 2015 and examined the diagnostic importance of S100A9 and S100A12 proteins for breast cancer, the S100A9 protein, one of the 2 sub-units of the calprotectin protein, could not distinguish breast cancer from the control group. In our study, calprotectin levels were similar in cancer and control groups, and there was no statistical difference between the two groups. These differences between the studies may be related to the histological types of the patients included. While Gunaldi et al.’s9 study included different histological types as in our study, Khorrami et al.’s8 study consisted of only invasive ductal types. Future studies involving homogeneous patients will clarify this situation.
There was no significant difference between the serum calprotectin levels of the patient and control groups. However, while the mean serum calprotectin value was 28.63±30.46 in the patient group, it was found as 16.09±6.12 in the control group. The mean serum calprotectin value was found to be 40.03±1.54 for the control group consisting of 30 healthy volunteers in the study of Zaki et al.10, who used exactly the same serum calprotectin kit we used in our study with all procedures. Although statistical comparison analysis was not performed for the control group in our study, we think that it was relatively high. However, unlike our study, the rate of male volunteers was 66.7% and the mean age was 32.30±11.43 years in the control group of this study, which was quite different in design from our study in terms of sample distribution10. In the current publication of Shaik et al.11 in 2021, it was reported that even though benign breast disease was diagnosed, there was a significant increase in inflammatory markers in the benign diagnosis periods of patients with a high risk of breast cancer afterwards. It also comes to mind that this may be the reason why the inflammatory markers, including the calprotectin value, of the patients in the control group of our study were similar to the patient group because in the design of our study, clinical follow-up of the patients diagnosed as benign was not carried out to determine whether these pathologies converted to malignancy within a specified period.
Analyses were made in order to evaluate the patient and control groups in our study in terms of our main variable, and in these analyses, a difference was determined in terms of age. The age of our patient group was higher than that of the control group. Due to this difference, when the change in serum calprotectin value with age variable is examined from the literature, it is seen that many studies indicate that age is not an effective parameter. In the study of Oosterwijk et al.12 in 2020, in which they examined variables that might be determinants of serum calprotectin levels in patients with diabetes mellitus, the age variable was determined as an insignificant parameter in this respect (r=-0.035 for age, p=0.051). In addition, in a study by Zaki et al.10 in 2019, evaluating the relationship between clinical severity and serum calprotectin levels in psoriasis patients, the age variable was not found to be associated with serum calprotectin levels (r=0.214; p=0.136).
In our study, when two groups were formed according to the 20 cut-off value of Ki67 level, serum calprotectin value was found to be significantly higher in those with Ki67 above 20. Ki67 is one of the most well-known predictors of chemotherapy response13,14. Ki67 is a parameter that is generally thought to be expressed less than 3% in healthy breasts and has been used in routine pathology reports in the last 10 years15. At the same time, the number of studies reporting that Ki67 is prognostic has been increasing recently16,17. However, in our treatment response analysis in this study, Ki67 did not show predictive properties for treatment response. Nevertheless, the results of our study are promising for serum calprotectin values in response to treatments or in follow-up, since it is known that there is a significant correlation between the increased percent expression of Ki67 value and the clinical aggressiveness of the tumor.
Current oncology guidelines report that luminal breast cancer subtypes have a worse response to chemotherapy18. Chou et al.’s19 study in 2019 and Verdial et al.’s20 study in 2022 evaluated the predictors of response to neoadjuvant chemotherapy in locally advanced breast cancers, and young age was determined to predict complete response. In our study, age and hormone receptors were found to be predictive for treatment response. Hormone receptor positive patients and older patients gave worse response to the treatment.
Study Limitations
There were some limitations regarding our study. The most important limitation was the small number of patients included in the study and the single center design. In addition, the heterogeneity of the included breast cancer patient groups was our other limitation. However, it is important that this study is the first to our knowledge to evaluate calprotectin for breast cancer treatment response.
CONCLUSION
In conclusion, the level of calprotectin did not differ significantly between the normal population and breast cancer patients, and there was no relationship between calprotectin and chemotherapy response. Further studies should investigate the relationship between calprotectin and breast cancer treatment response, with a larger number of patients and a more homogeneous patient population.
Ethics
Ethics Committee Approval: The study was approved by the Tekirdağ Namık Kemal University of Local Ethics Committee (protocol no: 2019.206.11.03, date: 26.11.2019).
Informed Consent: Consent form was filled out by all participants.
Peer-review: Externally peer-reviewed.
Authorship Contributions
Surgical and Medical Practices: E.B., S.Ö.G., Concept: E.B., S.Ş., Design: E.B., Data Collection or Processing: E.B., A.Ç., S.Ö.G., Analysis or Interpretation: E.B., A.Ç., Literature Search: E.B., S.Ş., Writing: E.B.
Conflict of Interest: No conflict of interest was declared by the authors.
Financial Disclosure: The authors declared that this study received no financial support.