Abstract
Aim: The aim of the current study was to analyze the type of variations in expression profiles of matrix metalloproteinase 2 (MMP2), matrix metalloproteinase 9 (MMP9), tissue inhibitor of metalloproteinase 2 (TIMP2), and vascular endothelial growth factor A (VEGFA) before and after radiochemotherapy in patients with locally advanced FIGO stage Ib-IIb cervical cancer. We analyzed the role of these biomarkers in monitoring response to treatment. Patients and Methods: Serum from 72 patients with cervical cancer treated within a phase III trial with either simultaneous radiochemotherapy (S-RC) with cisplatin, or systemic paclitaxel and carboplatin followed by percutaneous radiation (PC-R) was analyzed by ELISA. Sera were obtained during surgery and after the end of adjuvant treatment. Results: The median age at time of diagnosis was 46 years (range=30-71 years). The most common histological types were squamous cell (73.6%) and adenocarcinoma (25%). Thirty-five (48.6%) patients underwent surgery followed by S-RC and 37 (51.4%) patients were treated with surgery followed by PC-R. Five patients developed recurrence within six months after radiochemotherapy. VEGFA levels were significantly higher before and after adjuvant treatment in patients who developed early recurrence (p=0.001). An increase of more than 500 pg/ml VEGFA and a decrease of more than 9% of the pre-therapeutic value of TIMP2 were significantly associated with a higher risk of early recurrence (RR=8.5, 95% CI=1.8-39.8 and RR=11.0, 95% CI=2.5-48.2, respectively). TIMP2 expression and risk score for early relapse (which is calculated using values of VEGFA and TIMP2) were independent prognostic factors for overall survival (p=0.043, HR=0.96, 95% CI=0.93-0.99 and p=0.002, HR=1.09, 95% CI=1.03-1.15, respectively). Conclusion: Our results indicate a predictive value of VEGFA and TIMP2 in monitoring cervical cancer patients undergoing radiochemotherapy.
Cervical cancer ranks as the third most common diagnosed cancer worldwide and fourth leading cause of cancer-related death in women. Over 85% of these cases and deaths occur in developing countries (1, 2).
Lack of screening in medically under-served populations, especially in the developing world, contributes to the disproportionately high burden of cervical cancer. Access to screening would allow for detection of pre-cancerous and early-stage cervical cancer (3-5). Previous studies revealed the presence of HPV infection in over 90% of malignant cervical lesions (6, 7).
Although novel diagnostics, and surgical and adjuvant therapy methods against cervical cancer are available, there is still an imminent need to improve clinical outcome, especially in advanced FIGO stages.
For this objective, specific biomarkers could be suitable as their predictive and prognostic values have already been investigated in previous studies (8-10). Levels of biomarkers can change abnormally in certain kinds of cancers, due to their involvement in different pathways. These changes could be used as indicators in the therapeutic management and assessment of tumor biology.
Angiogenesis is one of the most crucial pathways which play a role in cancer. Both physiological and pathological, angiogenesis is mainly regulated via VEGF isoforms. Hypoxic tumor cells near necrotic areas up-regulate VEGF mRNA (11, 12), and elevated VEGF levels were shown to be correlated with advanced cervical cancer stage and increased risk of lymph node involvement (13).
Several studies have demonstrated that inhibition with antibodies against VEGF in tumor tissue where VEGF mRNA is overexpressed reduces tumor perfusion and microvascular density, and therefore blocks colorectal tumor growth significantly (14, 15).
This anti-angiogenesis effect contributes to progression-free and overall survival, as confirmed by using a bevacizumab monoclonal antibody to VEGF in phase II study of recurrent cervical cancer, metastatic colorectal and renal cancer (16-18) and a phase III study of metastatic colorectal cancer (19).
In addition, cytokines facilitate degradation of triple-helical collagen, which is the main component of extracellular matrix (ECM), and collagen IV in the basement membrane with the aid of metalloproteinases (MMP) which promote neoangiogenesis and tumor cell invasion. As a result, tumor cells detach from the primary site, enter the bloodstream and lymphatic system, and spread locally and distantly (9, 20). MMP2 and MMP9 (gelatinase-A and -B) are the most analyzed MMPs and play a major role in extracellular proteolysis. Studies have revealed enhanced expression of MMP2 and MMP9 (21, 22), and an association with high risk for the development of metastases and poor clinical outcome in cervical cancer (23). TIMPs, the endogenous inhibitors of MMPs, regulate the functional activity of MMPs via precursor zymogens (pro-MMP) (24). Recently, studies have shown that elevated TIMP2 was associated with a better clinical outcome in patients with solid tumors (25, 26).
Regarding clinical outcome, sensitivity to radiochemotherapy is correlated with better prognosis. However, there is a considerable number of cases of resistance to chemoradiotherapy and of relapses; the underlying mechanisms of resistance in cancer therapy are multifactorial (27).
To date, there are only a few studies concerning the association of tumor markers and response to chemoradiotherapy in advanced-stage cervical cancer (28).
The aim of the present study was to analyze the role of VEGFA, TIMP2, MMP2 and MMP9 in monitoring response to radiochemotherapy in patients with primary cervical cancer.
Patients and Methods
In our prospective multicenter phase III clinical trial, two hundred and sixty-three patients were included; the results of this trial were presented in our previous publication (29). Blood samples from 72 patients with primary cervical cancer were available before and after adjuvant treatment.
The patients underwent adjuvant treatment, of either simultaneous radiochemotherapy with cisplatin S-RC (n=35), or of systemic paclitaxel and carboplatin followed by percutaneous radiation (PC-R) (n=37). Our multicenter clinical trial was performed as a companion study of a phase III clinical trial.
Informed consent was obtained from each patient prior to blood withdrawal, and approval from the local Ethics Committee was obtained (no: 1857/ Si 262).
The patient's serum was collected before and after adjuvant therapy, centrifuged at 1,895 rcf for 10 min and stored at −80°C until analysis was performed.
The concentrations of VEGFA, TIMP2, MMP2 and MMP9 were analyzed in duplicate using ELISA (Quantikine, R&D Systems Inc., Minneapolis, MN, USA). Standard curves were constructed for each measurement to minimize interassay variation. The response to adjuvant chemoradiation in this observation was identified clinically. Early recurrence was defined as relapse within six months at the end of adjuvant therapy.
The changes of biomarkers from before to after adjuvant therapy were evaluated with Wilcoxon signed-ranks tests and differences of these changes between responders and patients with early relapse were analyzed with Mann–Whitney U-tests. Cut-off values for measured biomarkers were determined both as percentage and absolute (numeric) in the research using ROC-AUC and best results were selected. Using stepwise logistic regression analysis, the risk score of early recurrence was formulated as in the following formula:
Fisher's exact test was used to analyze the association between the expression profiles of cytokines and response to chemoradiotherapy. Overall survival (OS) rates and 95% confidence intervals (95% CI) were estimated according to Kaplan–Meier method. The relative importance of variables as independent predictors of survival was analyzed with the Cox proportional hazard regression. Adjusted hazard ratios (HR) and 95% CI for prognostic factors were estimated. Statistical analysis was performed using SPSS (Version 18.0; Chicago, IL, USA). A two-tailed p-value of <0.05 was considered significant.
Results
In the current study, 72 patients with primary cervical cancer were included. The mean age at diagnosis was 46 (range=30-71) years. Five patients developed early relapse within the first six months from the end of adjuvant therapy. Most patients presented with FIGO stage IB1 (36.1%) and IIB (38.9%) disease.
Fifty-three patients (73.6%) presented with squamous histological type and another 18 patients (25.0%) with an adenomatous histological type. Moreover, lymphatic invasion (L1) was detected in 40 patients (55.6%) whereas vascular invasion was found in only four (5.6%). The vast majority of surgical margins (86.1%) were, microscopically, tumor-free. The patients' characteristics are shown in Table I.
The biomarkers were measured in most cases. The concentrations of the biomarkers VEGFA, TIMP2, MMP2 and MMP9 before and after therapy are presented in detail in Table II.
Correlation of biomarkers with response to chemoradiotherapy. We focused on serum variation of four cytokines: VEGFA, TIMP2, MMP2 and MMP9 before and after adjuvant chemoradiotherapy.
Within the whole collective of patients VEGFA and MMP9 significantly decreased after adjuvant treatment (p=0.002 and p<0.001, respectively), while TIMP2 and MMP2 increased significantly after adjuvant treatment (p<0.001 in both biomarkers) (Table II).
Furthermore we evaluated the differences in biomarker variations in patients who developed early recurrence and patients with no or late relapse. Median VEGFA levels in responders to chemoradiotherapy were 741.2 pg/ml and 530.8 pg/ml before and after chemotherapy. In patients who developed early recurrence, serum VEGFA levels were 1,244.4 pg/ml to 850.6 pg/ml before and after treatment. These differences were not statistically significant (p=0.144).
Variation in TIMP2 expression was also determined before and after therapy, even if this analysis revealed no statistical significance (p=0.112). In early relapse, median TIMP2 expression was 78.7 ng/ml and 86.5 ng/ml before and after adjuvant therapy, whereas in the group with no or late relapse, the mean TIMP2 values were 81.9 ng/ml and 98.13 ng/ml before and after chemoradiotherapy. The TIMP2 values after chemoradiotherapy were significantly higher in patients with good prognosis (p=0.005).
For MMP9, levels at the time of surgery were 886.9 ng/ml and 1,340.8 ng/ml in patients with longer progression-free survival and in patients who developed early relapse. This difference in expression was statistically significant (p=0.019). After adjuvant treatment, MMP9 expression was 504.9 and 782.6 ng/ml, in long-term survivors and patients who developed early relapse, respectively. In both subgroups, MMP9 levels decreased at the end of adjuvant treatment, without statistical significance (p=0.426).
MMP2 levels increased in both subgroups of patients from 185.6 ng/ml to 215.7 ng/ml in long-term survivors and from 178.3 ng/ml to 191.8 ng/ml in early recurrent patients, before and after chemoradiotherapy, respectively (p=0.42).
Furthermore, we analyzed the differences in cytokine development between the two groups of patients. As mentioned before, cut-off values of measured biomarkers were determined both as percentage and absolute (numeric) in the research using ROC-AUC and best results were selected. The only significant associations were found for expression of VEGFA and TIMP2 before and after chemoradiotherapy. An increase of more than 500 pg/ml VEGFA and a decrease of more than 9% of the pre-therapeutic value of TIMP2 were significantly associated with a higher risk of early relapse (p=0.039, RR=8.5, 95% Cl=1.8-39.8 and p=0.023, RR=11.0, 95% Cl=2.5-48.2, respectively).
Patients' characteristics and description of primary tumors.
We also developed a new risk score, based on the combination of VEGFA and TIMP2, as indicated above. The association of VEGFA and TIMP2 revealed a higher sensitivity and specificity than either biomarker alone. The AUC (95% CI) for VEGFA, TIMP2 and risk score were 0.609 (0.326-0.893), 0.692 (0.440-0.945) and 0.728 (0.447-1.000), respectively. The sensitivity of the biomarkers increased when combined, 60%, compared to 40% for VEGFA and to 40% for TIMP2 alone (Table III). Based on AUC-ROC analysis, we found that the 17.4% cut-off value was associated with better sensitivity and specificity. Patients presenting a risk score higher than 17.4%, were more likely to develop early relapse (p=0.004, RR=15.7, 95% CI=3.2-76.5).
Levels of serum biomarkers before and after therapy.
Results of ROC analysis of biomarkers for resistance to chemoradiotherapy.
No differences were detected between the two therapy arms.
Survival analysis. During a median follow-up period of 59.3 (range=2.5-78.8) months, 14 (19.4%) patients experienced recurrence and 15 (20.8%) died. Five (6.94%) patients developed early relapse within six months after the end of adjuvant treatment.
Overall survival. In the univariate analysis, an increase of VEGFA concentration greater than 500 pg/ml and high risk score for early relapse negatively influenced overall survival (p=0.014, HR=5.00, 95% CI=1.39-17.97 and p=0.004, HR=1.08, 95% CI=1.02-1.14, respectively). The estimated 2-year survival rates for patients where VEGFA levels increased by more than 593 pg/ml, after chemoradiotherapy were 93.3% (95% CI=80.7-100%) and for those in between 593 and 500 pg/ml was 93.8% (95% CI=87.1-100%) and for patients with an increase of less than 500 pg/ml was 60.0% (95% CI=17.1-100%). The estimated 2-year overall survival for patients having a risk score less than −17.4% was 50.0% (95% CI=10.0-90.0%), and for more than 17.4% was 95.2% (95% CI=89.9-100%).
The estimated 2-year survival rate for patients with a reduction of TIMP2 expression of more than 9% was 50.0% (95% CI=1.0-99.0%), for those between 7.5% and 9% was 93.3% (95% CI=80.7-100%) and for those less than 7.5% was 92.1% (95% CI=84.6-99.5%).
In the multivariate analysis, lower TIMP2 expression and higher risk score for early relapse had a negative impact on overall survival (p=0.043, HR=0.96, 95% CI=0.93-0.99 and p=0.002, HR=1.09, 95% CI=1.03-1.15, respectively).
Discussion
In the current study, we examined the role of specific biomarkers (VEGFA, TIMP2, MMP2 and MMP9) in monitoring the response to chemoradiotherapy in locally advanced primary cervical cancer.
Kaplan-Meier curves (A-B) for progression-free survival of TIMP2 and risk score expression and (C-E) for overall survival in relation to VEGF-A, TIMP2, new risk score based on TIMP2 and VEGF-A combination.
Nowadays, the treatment modality for patients with high-risk cervical cancer in early advanced stage consists of simultaneous adjuvant platinum-based chemoradiotherapy following radical hysterectomy (30).
Several studies have shown that tumors expressing VEGF behave more aggressively (31), and in the case of cervical carcinoma, VEGF overexpression has been associated with disease-positive lymph nodes (20). In the study of Hashimoto et al., VEGFC was implicated in the advancement of lymph node metastases, and biopsy specimens were found to be valuable predictors of pelvic lymph node metastasis. VEGF was significantly enhanced in patients with cervical cancer (31), and reflected more aggressive tumor behavior (29). Moreover, in their study VEGFC was strongly co-expressed with MMP9 which is also increased in malignant cervical tissues whereas MMP2 mRNA levels were not increased.
In a recent publication of our study group, we showed that circulatory levels of VEGFR and VEGFC were associated with poor progression-free survival in primary cervical cancer (25). In the current study, the VEGFA levels decreased in all patients with cervical cancer after chemoradiotherapy. Nevertheless, the patients who developed early relapse presented higher VEGFA concentrations before and after chemoradiotherapy (p<0.001).
MMPs are involved in the process of metastasis formation (20) and their overexpression correlated with the process of tumor cell invasion and metastasis of human cancer, including uterine neoplasms (21, 32). More than that, the up-regulation of MMPs seems to be associated with progression of cervical uterine neoplasm (21). In our analysis, MMP9 was significantly more expressed at the time of diagnosis in patients who developed early recurrence than in patients without relapse or with late relapse. The levels remained higher in patients developing early relapse than in the long-term survivors. The variation in MMP9 expression before and after adjuvant treatment failed to reach statistical significance. The low number of early events may explain this lack of statistical significance.
TIMP2 increased in all patients with primary cervical cancer, although in the early-relapse sub-group, the levels increased slowly compared to the other patients. This variation in expression failed to reach statistical significance; even here, one explanation for this could be the low number of patients who developed early relapse.
Furthermore, our study revealed that an increase of more than 500 pg/ml VEGFA and a decrease of more than 9% of the pre-therapeutic value of TIMP2 were significantly associated with a higher risk of recurrence. We developed a new score for early relapse. Combining both biomarkers, the predictive value of VEGFA and TIMP2 reached better sensitivity and specificity than those for either biomarker alone.
In the study of Zhu et al., a significantly increased sensitivity to chemoradiotherapy in cervical cancer was revealed by combining S100A9, galectin-7, NMP-238, and HSP70 as a full-predictive factor (33), while the authors suggest that none of these protein markers could be used as independent predictive factor alone. Recently, Shi et al. identified the expression of DNA polymerase ζ as predictor for chemoradiation resistance and that it correlated with poorer progression-free survival and with depth of cervical stromal invasion (28). Choi et al. showed that patients with squamous cervical cancer and VEGFA positivity in biopsy specimens, by immunohistochemistry, presented a poorer response to neoadjuvant chemotherapy and progression-free survival compared to the non-VEGF group (34). The authors point out unclear contradiction between VEGF increase and chemoradiotherapy resistance. Since increased vascularity should also increase oxygen and drug delivery levels, the resistance was predicted to decrease. The disproportionate of this relationship between increased vascularity and resistance still remains unclear (34-36). Several studies demonstrated that overexpression of VEGF is correlated with tumor progression and poor prognosis in cervical cancer (37, 38), and is an independent prognostic predictor of poor progression-free and overall survival (39) and is an indication of poorer therapy outcome (40, 41).
In the current study, lower TIMP2 expression and higher risk score for early relapse were independent prognostic factors for poorer overall survival rates. In our previous study, preoperative serum TIMP2 levels were also the only independent prognostic factors for overall survival (25). Higher TIMP2 concentration in serum was associated with a better clinical outcome.
Cases of resistance to chemoradiation in primary cervical cancer are either primary or acquired, and lead to higher rates of early recurrence, progression and mortality. Therefore, an early prediction of resistant cases using biomarkers is highly desirable in order to improve efficient personalized cancer treatment.
We demonstrated that levels of biomarkers in responders and patients with resistance differ before and after therapy.
The lack of significant impact of analyzed biomarkers in monitoring the response to chemoradiotherapy may be explained by the low number of events: only five patients developed early relapse within our study population. Nevertheless, our data showed that VEGF levels normally decrease, but they remain high or even increase at the end of chemoradiotherapy in patients who will develop early relapse. Similarly for TIMP2, levels increase in all patients but seem to remain lower in patients who will experience relapse sooner.
In conclusion, VEGFA and TIMP2 might be useful in monitoring patients with cervical cancer undergoing surgery followed by adjuvant chemoradiotherapy. Identifying patients who will benefit from this is of major importance, as in this way, other therapeutic strategies could be offered. In order to analyze the impact of these biomarkers in monitoring response to chemoradiotherapy, prospective studies with a larger cohort should be performed.
- Received October 4, 2013.
- Revision received November 29, 2013.
- Accepted December 2, 2013.
- Copyright© 2014 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
