Prognostic factors for response to chemotherapy in advanced tumors of the uterine cervix : the role of neoangiogenesis

Objective: to analyze the expression of Vascular Endothelial Growth Factor (VEGF), its receptor (VEGFR-2), age and histological type of advanced cervical carcinomas with respect to the clinical response to neoadjuvant chemotherapy. Methods: we studied 40 patients with cervical carcinoma (IB2 and IVA) diagnosed by biopsies prior to treatment. All patients underwent neoadjuvant chemotherapy and evaluation for clinical response and expression of VEGF. We considered a tumor regression greater than 50% as a good clinical response. Results: eighteen patients (45%) had good response to chemotherapy, and 22 (55%), poor response. VEGF expression was positive in 16 patients and negative in 24. When analyzed separately for response to chemotherapy, only the positive expression of VEGF was associated with good clinical response (p=0.0157). Conclusion: VEGF expression alone was an important marker of good response to neoadjuvant chemotherapy in patients with advanced carcinoma of the cervix.


C
ervical cancer is the third most common in the female population, surpassed only by breast and bowel cancers.According to absolute data on cancer incidence and mortality from the Brazilian National Cancer Institute (INCA), cervical cancer was responsible for the death of 5430 women in Brazil in 2015 1 .In the world, it is the second cause of cancer death in women 2 .The most effective approach for the control of cervical carcinoma is the screening and treatment of pre-neoplastic lesions.The introduction of the Pap smear as a screening test significantly reduced the incidence of advanced cases and their mortality 3 .However, diagnosis is still common in advanced stages.
The five-year survival rate for patients with initial disease (stages 0-IIA) remains around 80%, and for patients with more advanced disease (stages IIIB-IVA) it is only 20% to 42% 3 .
For the treatment of uterine cervix carcinoma in stages IB, IIA and some selected cases of stage IIB, surgery or radiotherapy are used 4 .In the United States, patients with stage IIB tumors were treated primarily with radiotherapy, while, in Europe, radical hysterectomy was the preferred approach 5 .
For stages III and IVA, the treatment of choice was radical radiotherapy comprising external pelvic irradiation associated with brachytherapy, when possible 4 .With radiotherapy, survival rates remain around 65% for patients with tumors in stage IIB and 40% for those in IIIB, but with increased morbidity 6 .

Original Article
Prognostic factors for response to chemotherapy in advanced tumors of the uterine cervix: the role of neoangiogenesis.
The search for better results has led several centers throughout the world to introduce, for advanced cervical cancer, treatments employed for solid tumors from other locations, i.e., the combination of chemotherapy and radiotherapy with or without surgery 7 .Cisplatin is reported as the most active chemotherapy drug in cervical cancer and has been included as the basic component of its primary treatment 3 .Duration of treatment is critical and the ideal would be to anticipate which patients would present good response to chemotherapy, avoiding indicating it to those without prospects of good response, gaining time on the prescription of another therapeutic modality: radiotherapy alone or chemosensitization.
On the other hand, patients who presented good response could undergo surgery, dispensing with radiotherapy and thus avoiding complications due to the association of treatments 3,4 .
Angiogenesis is a critical factor in the progression of solid tumors, including cancer of he cervix.Several studies have shown correlation of angiogenesis with worse prognosis in solid tumors such as breast, prostate, colon 8,9 .The mechanisms responsible for angiogenesis in the cervical neoplasia, however, are not well defined 10 .

The increased expression of Vascular Endothelial
Growth Factor (VEGF) in primary tumors correlates with a more aggressive biological behavior (increased vascular invasion index, metastatic lymph nodes and liver metastases) and with a worse prognosis compared with cases in which its expression is decreased 11 .The presence of VEGF and VEGFR-2 in a uterine cervix tumor can be determined by immunohistochemistry, and its subtypes, by Western blot, ELISA and PCR 12 .
There are few studies that correlate the expression of VEGF, or its VEGFR-2 receptor, with the response to chemotherapy.The objective of this study was to evaluate the expression of tumor VEGF in patients submitted to neoadjuvant chemotherapy for locally advanced uterine cervix cancer.

METHODS
We conducted a case-series, prospective study, The same observers examined the patients three or four weeks after the end of each cycle of chemotherapy.According to WHO criteria, we considered a complete response the disappearance of the entire lesion; a partial response, when there was an estimated decrease in tumor size of 50% or more; stable disease, when there was an estimated decrease of less than 50% or an estimated increase of less than 25%; and progression of the disease, when a new unidentified lesion appeared or there was an estimated increase of 25% or more of the existing lesion 13 .We considered a good response to chemotherapy when there was a complete and/or partial response, and a poor response when there was stable disease and/or tumor progression.
The samples collected were fixed in formalin and included in paraffin.Histological preparations were submitted to immunohistochemical reaction using mouse monoclonal primary antibodies produced by clone C1, anti-VEGF (reference SC-7269, Santa Cruz, CA, USA).For its VEGFR-2 receptor, we used as the primary antibodies the We performed the statistical analysis with the GraphPad Prisma software 2.01.For the analysis of the contingency tables (2x2) and comparison between VEGF and VEGFR-2 expression and clinical response to chemotherapy alone, we used the Fisher's exact test (univariate analysis).
To verify independence (age less than or equal to 60 years versus greater than 60, histological type, expression of VEGF and VEGFR-2) and how it would affect the dependent variable (response to chemotherapy), we used logistic regression (multivariate analysis) 14 .

RESULTS
Regarding the epidemiological aspects, the 40 patients included in this study were aged between 28 and 76 years, the mean age being 53.2 years.As for clinical staging, five patients had stage IB2, two IIA, 23 IIB, seven IIIB, and three had IVA.We observed that the mean age showed a tendency to increase with the more advanced stages (Table 1).

DISCUSSION
Neoadjuvant chemotherapy for uterine cervix tumors has been incorporated into the conventional treatment in order to reduce tumor volume and extension, so that radiotherapy can be instituted with better local conditions or transform clinically inoperable cases in operable ones 15 .
Another beneficial effect would be the possibility of treating micrometastases followed by surgery or irradiation, depending on the primary tumor response.Chemotherapy is also used simultaneously with radiotherapy, being called chemosensitization.
In theory, they would have a synergistic effect, since chemotherapy can increase the tumor's sensitivity to radiation by synchronizing the cells to a radio-sensitive phase of the cell cycle 16 .However, its side effects also add up, increasing morbidity and mortality 8 .
Tumor size and presence of metastasis are risk factors that affect the final result and are associated with worse prognosis and greater resistance to therapy 16,17 .Thus, it would be of great importance to determine the predictive factors of response to chemotherapy through biological markers.
The role of angiogenesis in tumor growth has been clearly established.For progression of solid tumors, a sufficient supply of blood vessels is necessary 18 .The tumor induces angiogenic mechanisms, directly and indirectly leading to the growth of microvessels, providing access for tumor cells to the vascular system for its metastatic spread.Among the various angiogenic factors, VEGF and its receptors play an important role in tumor neoangiogenesis.VEGF is widely distributed in squamous cell carcinomas of the head and neck, regardless of stage or histological grade 19 .
Several clinical studies have shown that VEGF expression and angiogenesis play an important prognostic role in advanced squamous cell carcinoma, being associated in most cases with poor prognosis and decreased survival [20][21][22][23][24] .In cervical neoplasms, the correlation between VEGF expression and dysplasia progression has been demonstrated.However, the association between VEGF expression and the clinical-pathological features of the disease needs further studies 9,25 21 .We did not observe significant differences between these variables.However, one study showed a significant association only between VEGF expression and tumor the histological type.
This finding supports the idea that the mechanisms of angiogenesis may be different, depending on the organ involved and the histological type of the tumor.
Regarding the value of VEGF in predicting the response to chemotherapy, little is known so far.In this study, VEGF proved to be a good marker to predict response to chemotherapy.
Due to the role of VEGF in tumor angiogenesis and vascular permeability, the greater distribution and release of drugs through the neoformed vessels could explain the relationship between the response to chemotherapy and the presence of VEGF.However, Foekens et al. 26  Therefore, positive VEGF cases were considered a favorable phenotype for response to chemotherapy.
Takiuchi et al. 14  In this way, a larger number of papers, with more cases and with similar therapeutic schemes are necessary to better understand the relationship between the clinical response to anti-neoplastic agents and the expression of biological markers.
We found no studies correlating the expression of VEGFR-2 with response to chemotherapy in solid tumors.In our study, the cases considered positive for VEGFR-2 showed no association with the clinical response to chemotherapy, nor when combined with positive VEGF expression.
The VEGF expression was the sole indicator of good response to neoadjuvant chemotherapy in patients with advanced cervical cancer.Age and tumor histological type showed no association with response to neoadjuvant chemotherapy.
scheme.Patients who did not respond with a significant decrease in tumor volume after the second cycle of this regimen and those who had not enough tumor reduction to allow radical surgery were considered resistant to chemotherapy and classified as cases with poor response.

FLK- 1
mouse monoclonal, produced by clone A3 (reference SC-6251, Santa Cruz, Ca, USA).For the immunohistochemical study for VEGF and VEGFR-2 we used the Quik Novostain Universal Kit (reference NCL-RTU-Qu, Novocastra Laboratories, Newcastle, UK), which includes normal horse serum to block non-specific antigens; pan-specific secondary biotinylated antibody to bind to the primary antibody; and the extreptavidin-peroxidase complex for detection of the reaction.The preparations were developed with 3'3' diaminobenzidine tetrahydrochloride, using Liquid DAB (3'3' Diaminobenzidine Tetrahydrocloride) Substrate Kit (reference NCL-L-DAB; Novocastra, UK) for tissues embedded in paraffin and counterstained with Mayer's hematoxylin.
studied the immunohistochemical expression of VEGF in 30 patients with gastric adenocarcinoma and its relation with the response to 5-fluoracil and cisplatin.The response rate of positive VEGF and negative VEGF cases was 75% (12/16) and 16.7% (2/14), respectively, and positive VEGF cases had a significant treatment impact (p=0.0031).In the present study, VEGF-positive patients also had a better response to the use of neoadjuvant chemotherapy with 5-fluoracil and cisplatin.The correlation between the response to chemotherapy and the expression of VEGF might be explained by the greater release of drugs within tumor neoangiogenesis regions, greater vascular permeability and, consequently, better local performance.The differences found between the studies in relation to VEGF expression and response to chemotherapy could be explained by the use of different therapeutic schemes, the different tumor types involved, the different drug associations, the different mechanisms of action and varied dosages.

Table 2 .
Distribution of patients according to histological type.

Table 3
). Twenty-six patients became operable after initial treatment with neoadjuvant chemotherapy, and underwent surgery.

Table 3 .
Correlation between histological type and response to chemotherapy.

Table 4 .
Distribution of VEGF expression in relation to histological type.

Table 5 .
Distribution of VEGF expression, positive or negative, regarding the clinical response to chemotherapy.

Table 6 .
Multivariate analysis of positive VEGF, positive VEGFR-2, age less than or equal to 60 years and histological type versus clinical response to neoadjuvant chemotherapy (good or poor).