DOI: https://doi.org/10.21203/rs.3.rs-757775/v1
Background: recent experimental studies have shown a potential link between cathepsin S (CTTS) and gastric cancer progression. Herein, we aimed to evaluate the expression of CTTS in gastric adenocarcinoma.
Methods: Cross-sectional study that included two groups, gastric adenocarcinoma (n=42) and gastritis (n=50). The gastritis group was then subdivided into H. pylori positive (n=25) x negative (n=25). Gastric tissue samples were analyzed in order to determine the CTTS expression through immunohistochemistry.
Results: In patients with gastritis, the age ranged from 18 to 78 years. Among them, 34% were male, and 66% were female. In patients with gastric cancer, the age ranged from 37 to 85 years. Among them, 50% were male, and 50% were female. When comparing the expression of CTTS between the two groups, only 16% of the gastritis samples had an expression higher than 25%. On the other hand, among patients with gastric adenocarcinoma, 19% had expression between 25-50%, 14.3% between 51-75%, and 26.2% had expressions higher than 75% (p < 0.001). CTTS expression was significantly higher in patients with positive test for H. pylori: 87.5% x 38.5% (p<0.001). There was no statistically significant association between the positivity of CTTS and the clinical-pathological variables, including tumor staging, histological type, angiolymphatic invasion, recurrence, current status and death.
Conclusion: CTTS has a higher expression in samples of gastric adenocarcinoma. Patients with gastritis by H. pylori also show a higher expression of CTTS compared with patients with negative results for this bacterium.
Cathepsins are enzymes that comprise a family of 15 lysosomal proteases widely distributed in intracellular and extracellular spaces, among which five have been implicated repeatedly in the progression of solid cancers (cathepsin B, H, K, L, and S) [1, 2].
Cathepsins plays roles in a wide range of body activities based on their hydrolysis effect. In digestive cancers, the expression of cathepsin is positively regulated by tumor-promoting factors, such as C-myc, K-ras, AGR2, MAPK, p38, and the Hedgehog (Hh) signaling pathways. Activated cathepsins hydrolyze growth factors, such as EGF, VEGF and TGFβ, promoting the proliferation of cancer cells, and they appear to play a role, although still uncertain, in the regulation of apoptosis [3, 4]. In addition, analyses of expression of cathepsins in tumor microenvironments have sparked discussion about the role of these molecules in the response to anti-cancer therapy and in the phenomenon of therapeutic resistance [5].
Recently, studies have pointed to a supposed relation between gastric cancer and cathepsin expression, specifically cathepsin S (CTTS). Data are however still incipient, suggesting a therapeutic, prognostic, and diagnostic potential of this enzyme in the evolution of this disease [6]. In vitro studies have shown that increased CTTS expression is related to increased tumor invasion and metastasis, and that its inhibition is capable of preventing tumor cell invasion and migration in gastric cancer [7, 8]. Parallel to this, studies have also shown that the serum expression of CTTS may be a great ally in the early diagnosis of gastric cancer, even presenting a sensitivity superior to the usual tumor markers, such as CEA, CA 19.9, and CA 72.4 [8].
The present study aims to evaluate the expression of CTTS in gastric tissue samples of patients with gastric adenocarcinoma and compare it with the expression in gastric tissue samples of patients without cancer, only with gastritis. In addition, this study seeks to evaluate the impacts of H. pylori infection on CTTS expression in gastric tissue samples without cancer.
This is a cross-sectional study carried out at Hospital das Clínicas, Federal University of Pernambuco, Recife, Brazil, aiming to evaluate the expression of CTTS in gastric tissue samples of patients diagnosed with gastric adenocarcinoma (n = 42) and of patients diagnosed only with gastritis (n = 50). The group of patients with gastritis was subdivided into another two subgroups: one with a positive result for H. pylori (n = 25) and one with a negative result for H. pylori (n = 25). The primary result was to compare CTTS expression assessed by immunohistochemistry in gastric tissue samples from patients with adenocarcinoma and patients with gastritis (with and without H. pylori).
All procedures performed in this study involving human participants were in accordance with the ethical standards of the institutional research committee and the 1964 Helsinki declaration and its later amendments, or comparable ethical standards. This research's protocol was approved by the Ethics Committee of the Hospital das Clínicas da Universidade Federal de Pernambuco (HC/UFPE-EBSERH) under the protocol CAAE no. 38000620.9.0000.8807. Informed consent was obtained from all participants in the study.
We included patients undergoing surgical treatment of gastric adenocarcinoma with curative intent in our center from 2017 to 2019. We excluded patients at stage IV and those undergoing neoadjuvant chemotherapy. After each surgical procedure for the treatment of gastric adenocarcinoma, we always selected the first patient to present a confirmed histopathological result for gastritis in the Pathology sector of our institution, aiming to form the control group with the lowest possible risk of selection bias. This group included patients who underwent esophagogastroduodenoscopy (EGD) with biopsy of lesions, confirming that it was gastritis. The search for H. pylori was performed in all patients using the urease test and confirmed by histopathology with Giemsa stain. We excluded patients previously submitted to gastroplasty and those with reports of previous treatment for H. pylori.
We performed immunohistochemical staining (IHC) to study the expression of CTTS in 42 samples of human gastric cancer tissue (Fig. 1) and in 50 samples of gastric tissue with gastritis (Fig. 2). We made 3-µm sections in series for immunohistochemical analysis and placed them on Superfrost Plus glass slides. We performed immunostaining using the Ventana BenchMark ULTRA System automated staining system using rabbit polyclonal antibody directed against CTTS (Clone No. A13482; ABclonal, Massachusetts, United States). We used a 1:100 dilution and incubated it for 30 min to 37 ºC. We used the DAB IHC Detection Kit as the chromogen substrate. All specimens were counterstained with hematoxylin. We interpreted immunohistochemical reactions using a standard optical microscope and analyzed them according to the specific pattern of the investigated antibody. We assessed marking intensity using the following grading:
| Grading | Marking |
|---|---|
| 0 | no detectable coloring |
| 1 | weak coloring = light yellow |
| 2 | moderate coloring = brown-yellow |
| 3 | strong coloring = brown |
We graded the percentage of stained tumor cells as follows:
| Grading | Stained cells |
|---|---|
| 0 | no positive tumor cells |
| 1 | 1–25% of positive tumor cells |
| 2 | 26–50% of positive tumor cells |
| 3 | 51–75% of positive tumor cells |
| 4 | > 75% of positive tumor cells |
We calculated the staining index score as the product of the percentage of positive tumor cells and the intensity of staining. We defined CTTS expression according to the color index:
| Expression | Coloring index |
|---|---|
| - | 0: negative |
| + | 1–4: weakly positive |
| ++ | 5–8: positive |
| +++ | 9–12: strongly positive |
For analysis purposes, the CTTS expression intensity was categorically assessed: high expression or low expression. We defined a high expression as a color index score > 4, while low expression was a score ≤ 4. An index = 0 corresponds to a missing expression.
For statistical analysis, we used the software STATA/SE 12.0 and Excel 2010. Mean values and standard deviations were calculated to describe data population. We considered a 95% confidence for all tests and a p-value < 0.05 as significant in all tests. The results are presented in tables containing respective absolute and relative frequencies. We verified the existence of associations using the Chi-square test and Fisher's exact test for categorical variables. The statistical power of this sample was 72.7% according to the presence or absence of CTTS in the gastric cancer groups compared to benign stomach lesions.
From ninety-two (92) patients studied, 50 patients had gastritis and 42 patients had gastric adenocarcinoma. In patients with gastritis, the age ranged from 18 to 78 years. Among them, 35 patients (70%) were under 50 years old, 17 (34%) were male, and 33 (66%) were female. In patients with gastric cancer, the age ranged from 37 to 85 years. Among them, 34 patients (81%) were over 50 years old, 21 (50%) were male, and 21 (50%) were female.
Regarding the tumor topography, 23 patients had tumors in the antrum (54.8%) and 19 in the gastric body (45.2%). Subtotal gastrectomy was performed in 20 patients (47.6%), and total gastrectomy in 22 patients (52.4%). As for pathological staging, ten patients were at stage IA (23.8%), ten at stage IB (23.8%), 19 at stage IIIB (45.3%), and three at stage IIIC (7.1%). Regarding histological type, 18 were classified as intestinal (42.9%), 18 as diffuse (42.9%), and six as mixed tumors (14.2%). We detected lymphatic vascular embolization of neoplastic cells in 23 cases (54.8.7%), and absent in 19 cases (45.2%). Recurrence occurred in 23.8% of cases. The situation of these patients at the end of the study was alive without disease (71.4%), alive with disease (7.1%), death from other causes (4.8%), death from cancer (16.7%). Table 1 shows the frequencies of clinical variables, staging, therapy, and follow-up of 42 patients with CG in this study.
| Gastric cancer group | ||
|---|---|---|
| Variable | n | % |
| Topography | ||
| Antrum | 23 | 54.8 |
| Body | 19 | 45.2 |
| Type of surgery | ||
| Subtotal gastrectomy | 20 | 47.6 |
| Total gastrectomy | 22 | 52.4 |
| Pathological stage | ||
| IA | 10 | 23.8 |
| IB | 10 | 23.8 |
| IIIB | 19 | 45.3 |
| IIIC | 03 | 7.1 |
| Primary tumor | ||
| T1 | 10 | 23.8 |
| T2 | 10 | 23.8 |
| T3 | 19 | 45.3 |
| T4 | 03 | 7.1 |
| Lymph nodes | ||
| N0 | 20 | 47.6 |
| N3 | 22 | 52.4 |
| Histological type | ||
| Intestinal | 18 | 42.9 |
| Diffuse | 18 | 42.9 |
| Mixed | 06 | 14.2 |
| Histological grade | ||
| Well differentiated | 03 | 7.1 |
| Moderate | 08 | 19.0 |
| Poorly differentiated | 31 | 73.9 |
| Angiolymphatic invasion | ||
| Positive | 23 | 54.8 |
| Negative | 19 | 45.2 |
| Recurrence | ||
| No | 32 | 76.2 |
| Yes | 10 | 23.8 |
| Current status | ||
| Alive without disease | 30 | 71.4 |
| Alive with disease | 03 | 7.1 |
| Death without cancer | 02 | 4.8 |
| Death with cancer | 07 | 16.7 |
When comparing the expression of CTTS between the two groups, in gastritis samples 38% did not express CTTS, 46% had low expression (1–25%), and only 16% had an expression higher than 25%. On the other hand, among patients with gastric adenocarcinoma, 19% had expression between 25–50%, 14.3% between 51–75%, and 26.2% had expressions higher than 75%, with statistically significant results (p < 0.001). Analyses involving the CTTS staining index in IHC and the intensity of expression also showed a statistical significance, being higher in the group of patients with gastric adenocarcinoma (Table 2).
| Group | |||||
|---|---|---|---|---|---|
| Gastric cancer | Gastritis | ||||
| Variable | n | % | n | % | p-value* |
| Age | |||||
| Under 50 | 8 | 19.0 | 35 | 70.0 | < 0.001 |
| Over 50 | 34 | 81.0 | 15 | 30.0 | |
| Gender | |||||
| Male | 21 | 50.0 | 17 | 34.0 | 0.121 |
| Female | 21 | 50.0 | 33 | 66.0 | |
| H. pylori | |||||
| Positive | 8 | 19.0 | 24 | 48.0 | 0.004 |
| Negative | 34 | 81.0 | 26 | 52.0 | |
| Percentage of CTTS stained cells | |||||
| No positive cells | 07 | 16.7 | 19 | 38.0 | < 0.001 |
| 1–25% | 10 | 23.8 | 23 | 46.0 | |
| 25–50% | 08 | 19.0 | 04 | 8.0 | |
| 51–75% | 06 | 14.3 | 02 | 4.0 | |
| > 75% | 11 | 26.2 | 02 | 4.0 | |
| CTTS coloring index | |||||
| Negative (0) | 7 | 16.7 | 19 | 38.0 | 0.002 |
| Weakly positive (1–4) | 14 | 33.3 | 24 | 48.0 | |
| Positive (5–8) | 09 | 21.4 | 03 | 6.0 | |
| Strongly positive (9–12) | 12 | 28.6 | 04 | 8.0 | |
| Intensity of expression | |||||
| Absent − 0 | 7 | 16.7 | 19 | 38.0 | 0.001 |
| Low < 4 | 14 | 33.3 | 24 | 48.0 | |
| High ≥ 4 | 21 | 50.0 | 07 | 14.0 | |
| (*) Chi Square Test. | |||||
In the evaluation of CTTS expression in the group of patients with gastritis, CTTS expression was significantly higher in patients with positive test for H. pylori: 87.5% x 38.5% (p < 0.001) (Table 3).
| CTTS coloring score | |||||
|---|---|---|---|---|---|
| Positive | Negative | ||||
| Variable | n | % | n | % | p-value* |
| Age | |||||
| Under 50 | 22 | 62.9 | 13 | 37.1 | 0.849 |
| Over 50 | 09 | 60.0 | 06 | 40.0 | |
| Gender | |||||
| Male | 11 | 64.7 | 06 | 35.3 | 0.777 |
| Female | 20 | 60.6 | 13 | 39.4 | |
| H. pylori | |||||
| Positive | 21 | 87.5 | 03 | 12.5 | < 0.001 |
| Negative | 10 | 38.5 | 16 | 61.5 | |
| (*) Chi Square Test. | |||||
In the evaluation of CTTS expression in the group of patients with gastric adenocarcinoma, there was no statistically significant association between the positivity of the expression and the clinical-pathological variables presented in Table 4.
| CTTS | |||||
|---|---|---|---|---|---|
| Positive | Negative | ||||
| Variable | n | % | n | % | p-value* |
| Age | |||||
| Under 50 | 7 | 87.5 | 01 | 12.5 | 1.000 |
| Over 50 | 28 | 82.4 | 06 | 17.6 | |
| Gender | |||||
| Male | 17 | 81.0 | 04 | 19.0 | 1.000 |
| Female | 18 | 85.7 | 03 | 14.3 | |
| H. pylori | |||||
| Positive | 27 | 79.4 | 7 | 20.6 | 0.312 |
| Negative | 08 | 100.0 | 0 | 0.0 | |
| Topography | |||||
| Antrum | 18 | 78.3 | 05 | 21.7 | 0.428 |
| Body | 17 | 89.5 | 02 | 10.5 | |
| Type of surgery | |||||
| Subtotal gastrectomy | 15 | 75.0 | 05 | 25.0 | 0.229 |
| Total gastrectomy | 20 | 90.9 | 02 | 9.1 | |
| Staging | |||||
| IA | 08 | 80.0 | 02 | 20.0 | 0.490 |
| IB | 07 | 70.0 | 03 | 30.0 | |
| IIIB | 17 | 89.5 | 02 | 10.5 | |
| IIIC | 03 | 100.0 | 0.0 | 0.0 | |
| Primary tumor | |||||
| T1 | 08 | 80.0 | 02 | 20.0 | 0.490 |
| T2 | 07 | 70.0 | 03 | 30.0 | |
| T3 | 17 | 89.5 | 02 | 10.5 | |
| T4 | 03 | 100.0 | 0.0 | 0.0 | |
| Lymph nodes | |||||
| N0 | 15 | 75.0 | 05 | 25.0 | 0.229 |
| N3 | 20 | 90.0 | 02 | 9.1 | |
| Histological type | |||||
| Intestinal | 14 | 77.8 | 04 | 22.2 | 0.852 |
| Diffuse | 16 | 88.9 | 02 | 11.1 | |
| Mixed | 05 | 83.3 | 01 | 16.7 | |
| Histological grade | |||||
| Well differentiated | 03 | 100.0 | 0 | 0.0 | 0.177 |
| Moderate | 05 | 62.5 | 03 | 37.5 | |
| Poorly differentiated | 27 | 87.1 | 04 | 12.9 | |
| Angiolymphatic invasion | |||||
| Positive | 20 | 87.0 | 03 | 13.0 | 0.682 |
| Negative | 15 | 78.9 | 04 | 21.1 | |
| Recurrence | |||||
| No | 26 | 81.3 | 06 | 18.8 | 1.000 |
| Yes | 09 | 90.0 | 01 | 10.0 | |
| Current status | |||||
| Alive without disease | 24 | 80.0 | 06 | 20.0 | 0.475 |
| Alive with disease | 02 | 66.7 | 01 | 33.3 | |
| Death without cancer | 02 | 100.0 | 0 | 0.0 | |
| Death with cancer | 07 | 100.0 | 0 | 0.0 | |
| Death | |||||
| Yes | 09 | 100.0 | 0 | 0.0 | 0.314 |
| No | 26 | 78.8 | 07 | 21.2 | |
| (*) Fisher's exact test | |||||
The cathepsins that have already shown an increased expression in the presence of gastric cancer are B, E, K, L, S, X, and Z. To date, there are only a few studies that sought to assess the relation between CTTS and gastric cancer [7, 8]. This enzyme appears to play an important role in the tumor invasion process through the degradation of the extracellular matrix, modulation of the immune response, and regulation of several cell signaling pathways, including the activation of tyrosine kinase receptors, especially c-Met, matrix metalloproteinases, IL-11, CXCL16, and Integrin alpha-6-beta-4 [2, 7, 9]. Still, specifically for gastric adenocarcinoma, CTTS appears to have an activating effect on the MKN7 and MKN45 cancer cell lines [7].
Liu et al. [8] evaluated the serum dosage of CTTS in patients with gastric cancer by comparing the results with healthy patients and with benign gastric lesions. They observed that the serum CTTS values of patients with gastric cancer were significantly higher than those of non-tumor gastric tissue controls (P < 0.001). Still in this study, the authors investigated the diagnostic power of CTTS in their sample of 496 patients, finding sensitivity and specificity values of 60.7% and 90%, respectively. Additionally, in this study, there was a significant decrease in serum CTTS levels after surgical resection of the tumor, suggesting an intimate relation between this enzyme and the tumor microenvironment. In our study, we found similar results, with CTTS expression significantly higher in the group of patients with gastric adenocarcinoma compared to those of controls. The results of these studies suggest that CTTS may be a potential biomarker for the diagnosis of gastric cancer.
Yang et al. [7] studied the expression of cathepsins through a proteomic analysis of cultures of normal cells and gastric cancer. We observed a higher protein expression and a positive regulation of cathepsin S in gastric cancer cell secretome. There were no statistically significant differences in CTSS expression between the intestinal, diffuse, and mixed subtypes.
Researchers have shown a correlation between CTTS and disease characteristics, such as tumor size, lymph node invasion, distant metastases, and overall survival, noting that higher CTTS expressions were related to more advanced TNM stages and worse survival rates [8]. In the present study, there was no statistically significant association between CTTS expression and tumor staging or survival rates. A possible explanation for such a difference between the studies is the number of patients included, which was noticeably lower in our analysis.
Infection of the gastric mucosa by H. pylori is known to be an important risk factor for the development of gastric adenocarcinoma. However, the exact mechanisms of activation of carcinogenesis are not yet fully elucidated [10]. One of the possible mechanisms pointed out in this process is the pro-inflammatory response orchestrated by Th17 cells in the infected gastric mucosa [11, 12]. Previous studies have shown an association between infection by H. pylori and increased levels of expression of cathepsins D and X. However, there are no studies determining the behavior of CTTS in the presence of an infection by H. pylori [13, 14]. In the present study, we evaluated the expression of CTTS in samples of gastric mucosa infected by H. pylori. We observed that 87.5% of the samples in the gastritis group with H. pylori showed positive expression for CTTS, contrasting with only 12.5% of the gastritis group without H. pylori. These results reinforce the hypothesis that CTTS is involved in the process of carcinogenesis of gastric adenocarcinoma, since it also has a higher expression.
This study has some limitations that deserve attention. First is the sample size, which, as a result of the single-center character of this study, was limited, with a sample power of 72.7%. As the sample was non-probabilistic, and selected by convenience, we did not calculate the sample size since we included in the analysis all patients operated on during the study period. Another limitation that is worth mentioning is in relation to the observational and cross-sectional nature of this study. A longitudinal analysis could provide more accurate information about the relationship between CTTS expression and patient survival. However, for our primary result, the methodology applied was adequate.
In contrast to the limitations discussed above, the present study reports important data that provide robustness and authenticity to the analysis. It is one of the few studies to study the expression of CTTS in samples of gastric adenocarcinoma in humans and the first to attest a possible relationship between the expression of this enzyme and infection by H. pylori, an important risk factor for the development of gastric adenocarcinoma.
Considering the results of the present study, the authors conclude that CTTS has a higher expression in samples of gastric adenocarcinoma compared to samples of non-tumor tissue. In addition, as a secondary finding, we report that patients with gastritis by H. pylori also show a higher expression of CTTS compared with patients with gastritis with negative results for this bacterium. These results reinforce the discussion about the role of CTTS in the evolution of gastric cancer. However, further studies are needed aiming to define the relation of this enzyme in the process of gastric adenocarcinoma carcinogenesis.
The authors expressed their appreciation to the Anatomopathological Diagnostic Center (CEDAP), João Pessoa, PB, Brazil
This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.
All authors have declared no conflicting interests.
Authors have nothing to disclose.