In locally advanced rectal cancer, preoperative chemoradiotherapy offers improved local control, reduced toxicity, and higher rates of sphincter preservation without compromising oncological outcome compared with postoperative treatment . However, neoadjuvant treatment protocols may be inefficient in some patients. It is important to identify which patients will respond to neoadjuvant CRT.
Several molecular markers have been proposed as predictors of therapeutic response to CRT. However, to date, no molecular marker has been definitively proven to be predictive of response to CRT [5, 23–31]. Garcia et al.  studied 132 patients with locally advanced rectal cancer treated with CRT followed by surgery. DNA from pretreatment tumor biopsies and control DNA from paired normal surgical specimens was screened for mutations and polymorphisms that are individually associated with failure to achieve a PCR after CRT. The authors found that the combination of a KRAS mutation with cyclin D1G870A (AA) and methylenetetrahydrofolate reductase (NAD[P]) C677T polymorphisms synergistically identify, with a high degree of accuracy, a subset of rectal cancer patients who do not develop a PCR in response to CRT. In light of previous studies, we evaluated whether apoptotic factors (Bcl-2, Bax expression, caspase-3 activity, and cytochrome-c) and angiogenic parameters (MMP-9 levels and VEGF expression) were correlated with the response to CRT and could predict OS and DFS.
Apoptosis is a morphologically distinct, gene-directed form of cell death that is characterized by cytoplasmic fragmentation and nuclear condensation, and it contributes to both physiological and pathological processes . Some Bcl-2 family members induce apoptosis (Bax, Bad, Bid, and Bcl-X1), whereas others inhibit apoptosis (Bcl-2 and Bcl-X1). Cell homeostasis depends on a balance between these apoptotic and anti-apoptotic factors. Chang et al.  reported that Bax expression was significantly higher in a complete response group than in a partial response group (54% versus 29%, respectively; p = 0.017). In addition, CRT response was independent of other clinicopathologic parameters, including age, sex, pretreatment tumor size, distance from the anal verge, pretreatment stage (cT or cN), and type of operation. Tsamandas et al.  noted that in rectal adenocarcinoma, bax and bcl-2 proteins are frequently co-expressed with p53. Co-expression of bax with p53 protein is associated with poor clinical outcome, especially in cases without concomitant expression of bcl-2 . Nehls et al.  investigated 92 rectal cancer patients treated with preoperative radiotherapy and found that Bax protein expression may help to predict disease recurrence in preoperatively irradiated rectal cancers, whereas expression of p53, the proposed upstream regulator of bax-induced apoptosis, did not provide additional prognostic information. In our study, bax expression was found in 86% of patients in the complete response group and in 18% of patients in the partial response group. This difference is statistically significant (p = 0.001). The results of our study, together with those of previous studies, suggest that Bax expression is important for determining the response to CRT.
The biochemical detection of caspase-3 activity is a simple quantitative method for measuring a marker of apoptosis in tissue samples and preoperative rectal cancer biopsies. The pro-apoptotic enzyme caspase-3 acts at a convergence point of the intrinsic and extrinsic apoptosis induction pathways; therefore, its activity should yield a reliable measure of ongoing levels of apoptosis in tumor samples . Heer et al.  determined the feasibility of measuring caspase-3 activity using 47 archived fresh-frozen preoperative biopsy samples and corresponding resected rectal tumor specimens. Caspase-3 activity levels were strongly correlated in preoperative biopsies and tumor samples. The levels were significantly higher in the tumor specimens than in the preoperative biopsies. Heer et al. concluded that caspase-3 activity is an important indicator of local recurrence in rectal cancer. However, in our study, caspase-3 levels did not differ between the PCR and noPCR groups.
In the present study, we investigated another apoptotic marker, Bcl-2 expression. The role of Bcl-2 expression remains controversial. Contu et al.  did not find any correlation between Bcl-2 expression and age, histological grade, depth of invasion, lymphatic involvement, distant metastasis or tumor stage. Similarly, in our study, we found no significant relationship between Bcl-2 expression and pathologic response rate.
Matrix metalloproteinase expression and VEGF expression are the most investigated indicators in terms of the role of angiogenesis in the pathological response. MMP-9 is a substantial matrix metalloproteinase that plays an important role in metastasis and cancer invasion, leading to the degradation of matrix components [40, 41]. In patients with colorectal cancer, MMP-9 expression in tumor tissue was found to be higher than that in healthy mucosa. Unsal et al.  reported that MMP-9 expression in patients with rectal cancer was correlated with poor tumor response to preoperative CRT in their study of 44 cases. In our study, the mean MMP-9 level was 488.60 ± 405.169 ng/ml in complete responders and 1166 ± 745.912 ng/ml in partial responders. This difference was statistically significant (p = 0.04).
Various regulatory angiogenic factors, such as VEGF, acidic or basic fibroblast growth factor and interleukin-8 (IL-8), work together to organize the complex process of angiogenesis. Angiogenesis is related to VEGF over-expression in tumor cells, especially in solid tumors .
Although VEGF was not detected in normal rectal mucosa in recent immunohistochemical studies, significant immunoreactivity was noted in rectal cancer tissues . Wong et al.  studied the relationship between VEGF expression and progression from adenoma to carcinoma, and they emphasized that VEGF activation is an early occurrence that may promote the initiation of angiogenesis. Nozue et al.  investigated VEGF expression in patients with locally advanced rectal cancer before and after neoadjuvant RT and detected intense VEGF immunoreactivity and numerous VEGF-positive tumors. In the study of Zlobec et al. , a strong association was found between complete tumor response and the absence (or low levels) of VEGF expression in biopsy samples of rectal tumors prior to RT. VEGF expression was elevated and had notable immunoreactivity in non-responder tumors compared with complete responder tumors. In our study, the mean VEGF level was 516.51 ± 649.634 pg/ml in the partial responder group and 129.67 ± 67.726 pg/ml in the complete responder group (p = 0.05). In other words, our results suggest there may be an increase in VEGF levels in the noPCR group, similar to the findings of Zlobec et al .
In the present study, we also evaluated the oncological outcomes of patients with stage III (cT3 N+) low or mid rectal cancer. Pathological examination of the surgical specimens revealed PCR in 7 patients (24.1%) and noPCR in 22 patients (75.8%). Our prospective survival analysis indicates that patients in the PCR group experienced better oncological outcomes than patients in the noPCR group. In the PCR patients, the 50-month OS and DFS were both 100%. In the noPCR group, the 50-month OS and DFS were 70.1% and 64%, respectively. This difference is notable but statistically not significant (OS, p = 0.58; DFS, p = 0.08).
Our results suggest that oncological and functional outcomes after preoperative CRT are likely to be worse in patients with low bax expression and high levels of VEGF and MMP-9 upon histopathological examination of endoscopic biopsy tissues. Survival rates are better in PCR, on the other hand, low VEGF, MMP-9 levels and high Bax expression is found in patients with PCR. As a result of these findings, These parameters may be useful in prediction of PCR and different treatment options may be considered with the hope of increasing response rates in this group of patients. New chemotherapy protocols may be administered to increase the likelihood of a complete response in this group.