Abstract
Aim
To study the prognostic value of clinical and biological features of rectal cancer and potential gender differences in patients’ overall survival (OS), local recurrence-free survival (RFS) and metastasis-free survival (MFS) after short-course preoperative radiotherapy (SCRT) with short or long interval between RT and surgery (break).
Background
The length of the interval between RT and surgery in SCRT is debatable and gender-related differences in patients survival are not established yet.
Materials and methods
126 patients received SCRT with 5 Gy dose per fraction during 5 days, followed by radical surgery after short break ≤17 days, and a long break >17 days. Pretreatment tumor proliferation (bromodeoxyuridine labeling index, BrdUrdLI and S-phase fraction) was evaluated by flow cytometry and proteins: CD34, Ki-67, GLUT-1, Ku70, BCL-2, P53 expression was studied immunohistochemically.
Results
The studied group included 84 men and 42 women. There were 33, 76, and 17 cTNM (AJCC) tumor stages I, II, III, respectively. The median follow-up time was 53.3 months (range 2–142 months). For the whole group Cox multivariate analysis revealed that tumor grade (G > 1), interval between RT and surgery >17 days, pTNM stage >1 and P53 positivity + BrdUrdLI > 7.9% were negative prognostic factors for OS. Tumor aneuploidy and MVD > 140.8 vessels/mm2 were important for RFS. pTNM stage > 1 and P53 positivity combined with BrdUrdLI > 7.9% were risk predictors for MFS. Based on tumor biological features, gender-related difference in OS, RFS, and MFS were observed. In multivariate analysis, male patients age > 62 years and break >17 days only appeared to be significant for OS.
Conclusions
In male rectal patients treated with SCRT, breaks between RT and surgery >17 days should be avoided because they negatively influence patients’ survival.
Keywords: BrdUrdLI, P53, Ku70, Rectal cancer, Short-course preoperative radiotherapy, Break in treatment
1. Background
Treatment of locally advanced but resectable rectal carcinoma mainly utilizes two modalities of preoperative radiotherapy (RT): either(1) long course radio-chemotherapy with 50.4 Gy in 25–28 fractions and surgery after 4 to 8 weeks break, or (2) short course RT (SCRT) with 5 × 5 Gy followed by immediate surgery.1, 2, 3 In Europe, SCRT has become the standard treatment for patients with resectable cancer, and seems to offer the same reduction of local recurrence rate.2, 4 However, in some centers, apart from treatment with 5 × 5 Gy with a short (5–7 days) break, also 5 × 5 Gy and a long (4–5 weeks) break schedule before surgery is applied, and both were used in our study. Some authors1, 5, 6, 7 suggest that SCRT with delayed surgery is suitable for early cancer.
Personalized treatment based on predictions for patients’ outcome requires early characterization of patients’ sensitivity to treatment. It is important to verify if some tumor biological factors may be predictive for early and late tumor response (recurrence and metastasis rate) and patients survival.
Therefore, prognostic value of selected proteins was assessed: the P53 – tumor suppressor protein, BCL2 – antiapoptotic protein, Ki-67 (MIB1), a marker of cell proliferation, GLUT-1 – a hypoxia-regulated membrane protein glucose transporter-1, allowing the energy-independent transport of glucose across the cell membrane, CD-34 – a protein that indicates endothelial cells of blood vessels and, therefore, determines tumor vascularization and Ku70 – a protein involved in the DNA repair process. Earlier, we studied the pretreatment tumor proliferation rate (bromodeoxyuridine labeling index; BrdUrdLI, S-phase fraction; SPF)8, 9 and expression of six proteins to assess prognostic significance for early tumor response.10 We showed that higher Ki-67 and lower BCL2 expression were correlated with pathological tumor responses.
2. Aim
Currently, we would like to assess (1) prognostic significance of tumor proliferation (BrdUrdLI, S-phase fraction (SPF), Ki-67 LI) and expression of six proteins for patients’ overall survival (OS), local recurrence-free survival (RFS) and metastasis-free survival (MFS) after SCRT with a short (≤17 days) or long (>17 days) interval between RT and surgery (break), (2) to check if potential gender differences exist in patients’ long term characteristics.
3. Materials and methods
3.1. Patients
A total of 126 patients with rectal carcinoma were included in the study in which a tumor biopsy was taken before preoperative RT performed between November 2003 and January 2006. There were 84 males and 42 females with a mean age for the entire group of 62.0 (range 30–82) years.
3.2. Treatment
This is a retrospective analysis of two patient cohorts treated with SCRT (5 Gy/5 days) and surgery: one with an interval ≤17 days (median) and the other >17 days before surgery. In the first cohort there were 64 and in the second 62 patients. Inclusion and exclusion criteria and detailed information on irradiation and surgery were presented earlier.8 Tumors were classified according to the WHO classification of intestinal carcinoma11 and clinical (cTNM) and pathological (pTNM) stages according to the AJCC TNM 2010 classification.12
3.3. Flow cytometric analysis and immunohistochemistry
Tumor samples from each of the 126 patients were taken before RT. One fragment was used for flow cytometric analysis (BrdUrdLI, SPF) and the other for immunohistochemical evaluation of proteins expression. The details of flow cytometric analysis were described earlier.9 BrdUrdLI was calculated as a percentage of cells which incorporated BrdUrd. DNA ploidy and SPF were calculated from the DNA profile with ModFit software.8 The presence of aneuploidy was estimated by evaluating the DNA index, i.e. the ratio of the modal DNA fluorescence of abnormal to normal G1/G0 cells.
Expression of examined proteins was visualized by immunohistochemical staining. Detailed information on the staining procedure was presented earlier.10 Immunoreactivity of Ki-67, GLUT-1 and Ku 70 was scored as the number of positive tumor cells to the total number of counted tumor cells – labeling index (LI). P53 protein expression was considered positive if >25% of tumor cells showed immunopositivity and BCL-2 if expression was noted in any cancer cell. For CD34 expression, 7–10 high power (400×) tumor fields were counted for each patient. The mean vessel count per 1 mm2 of tumor area (microvessel density; MVD) was used in the analysis. Immunohistochemical assessments were performed by researchers blinded to clinical outcome.
3.4. Statistical analysis
Statistical analysis was performed with STATISTICA vs.9. For determination of mean values for variables and standard errors of means (SE), the descriptive statistics were used. Intergroup differences in the mean values were tested with one-way ANOVA test or Student's t-test. Associations between investigated categorical parameters and clinicopathological variables were evaluated by Pearson's Chi2 test. Differences were considered significant at P value of <0.05. Survival was estimated using the Kaplan–Meier method.13 The difference in survival rates between groups was assessed by the log-rank test. In univariate analysis, for the break between RT and surgery, that is the time interval between the last day of RT and surgery, and overall treatment time (OTT), that is the time from the beginning of RT to surgery, median values were used as cut-off points. For biological variables, these were mean or median values. Survival time was calculated from the date of surgery to the date of death, local relapse, metastasis or last follow-up. Multivariate analysis to determine prognostic factors for OS, local RFS and MFS was carried out by means of Cox's regression model in which all significant prognostic factors from the univariate analysis were entered. The Cox analysis was performed for all patients and separately for the male and female subgroups.
4. Results
4.1. Patients
A total of 126 consecutive patients were included in the study. Clinical and pathomorphological characteristics of the analyzed patients are presented in Table 1.
Table 1.
Gender-related clinicopathological and treatment characteristics of rectal cancer patients.
| Characteristics | All | Sex |
Break in treatment |
||||
|---|---|---|---|---|---|---|---|
| Men | Women | P-value | ≤17 days | >17 days | P-value | ||
| Age (years) | |||||||
| Median | 126a, 62.0 | 84, 62.0 | 42, 61.5 | 0.844 | 64, 61.5 | 62, 63.5 | 0.370 |
| (mean ± SD) | (61.3 ± 9.8) | (61.5 ± 9.2) | (61.1 ± 11.1) | (60.6 ± 10.3) | (62.1 ± 9.4) | ||
| Clinical stage (AJCC) | |||||||
| I | 33a (26.2%)b | 24 (28.6%) | 9 (21.4%) | 0.369 | 13 (20.3%) | 20 (32.3%) | 0.308 |
| II | 76 (60.3%) | 51 (60.7%) | 25 (59.5%) | 42 (65.6%) | 34 (54.8%) | ||
| III | 17 (13.5%) | 9 (10.7%) | 8 (19.1%) | 9 (14.1%) | 8 (12.9%) | ||
| Clinical tumor category | |||||||
| T2 | 34a (27.0%)b | 25 (29.7%) | 9 (21.4%) | 0.282 | 14 (21.9%) | 20 (32.3%) | 0.417 |
| T3 | 85 (67.5%) | 56 (66.7%) | 29 (69.1%) | 46 (71.9%) | 39 (62.9%) | ||
| T4 | 7 (5.5%) | 3 (3.6%) | 4 (9.5%) | 4 (6.2%) | 3 (4.8%) | ||
| Clinical node category | |||||||
| N0 | 109a (86.5%)b | 75 (89.3%) | 34 (80.9%) | 0.197 | 55 (85.9%) | 54 (87.1%) | 0.849 |
| N1 | 17 (13.5%) | 9 (10.7%) | 8 (19.1%) | 9 (14.1%) | 8 (12.9%) | ||
| Pathological stage (AJCC) | |||||||
| 0 | 6a (4.8%)b | 4 (4.8%) | 2 (4.8%) | 0.567 | 0 (0.0%) | 6 (9.7%) | 0.031 |
| 1 | 56 (44.4%) | 40 (47.6%) | 16 (38.1%) | 28 (43.8%) | 28 (45.1%) | ||
| 2 | 22 (17.5%) | 12 (14.3%) | 10 (23.8%) | 15 (23.4%) | 7 (11.3%) | ||
| 3 | 42 (33.3%) | 28 (33.3%) | 14 (33.3%) | 21 (32.8%) | 21 (33.9%) | ||
| Pathological tumor category | |||||||
| pT0 | 6a (4.8%)b | 4 (4.4%) | 2 (4.8%) | 0.045 | 0 (0.0%) | 6 (9.7%) | 0.086 |
| pT1 | 10 (7.9%) | 6 (7.2%) | 4 (9.5%) | 4 (6.3%) | 6 (9.7%) | ||
| pT2 | 51 (40.5%) | 39 (46.4%) | 12 (28.6%) | 27 (42.2%) | 24 (38.7%) | ||
| pT3 | 52 (41.3%) | 28 (33.3%) | 24 (57.1%) | 28 (43.7%) | 24 (38.7%) | ||
| pT4 | 7 (5.5%) | 7 (8.3%) | 0 (0.0%) | 5 (7.8%) | 2 (3.2%) | ||
| Pathological node category | |||||||
| pN0 | 84a (66.7%)b | 56 (66.7%) | 28 (66.7%) | 0.551 | 43 (67.2%) | 41 (66.1%) | 0.462 |
| pN1 | 20 (15.9%) | 15 (17.8%) | 5 (11.9%) | 8 (12.5%) | 12 (19.4%) | ||
| pN2 | 22 (17.4%) | 13 (15.5%) | 9 (21.4%) | 13 (20.3%) | 9 (14.5%) | ||
| Histological grade | |||||||
| G1 | 32a (25.8%)b | 19 (22.6%) | 13 (32.5%) | 0.495 | 11 (17.2%) | 21 (35.0%) | 0.025 |
| G2 | 89 (71.8%) | 63 (75.0%) | 26 (65.0%) | 50 (78.1%) | 39 (65.0%) | ||
| G3 | 3 (2.4%) | 2 (2.4%) | 1 (2.5%) | 3 (4.7%) | 0 (0.0%) | ||
| OTT (days) | |||||||
| Median | 126a, 22.0 | 84, 19.0 | 42, 29.5 | 0.261 | 64, 15.0 | 62, 37.5 | – |
| (mean ± SD) | (26.3 ± 12.7) | (25.4 ± 12.3) | (28.1 ± 13.3) | (14.8 ± 3.4) | (38.1 ± 6.0) | ||
| Interval between RT and surgery (days) | |||||||
| Median | 126a, 17.0 | 84, 14.0 | 42, 29.5 | 0.256 | 64, 10.0 | 62, 32.5 | – |
| (mean ± SD) | (21.3 ± 12.7) | (20.4 ± 12.4) | (28.1 ± 13.3) | (9.8 ± 3.4) | (33.1 ± 6.0) | ||
| Type of surgery | |||||||
| Abdominoperineal excision | 53a (42.1%)b | 40 (47.6%) | 13 (30.9%) | 0.074 | 35 (54.7%) | 38 (61.3%) | 0.453 |
| Anterior resection | 73 (57.9%) | 44 (52.4%) | 29 (69.1%) | 29 (45.3%) | 24 (38.7%) | ||
| Overall survival (months) | |||||||
| Median | 126a, 53.5 | 84, 53.5 | 42, 53.3 | 0.403 | 64, 54.3 | 62, 50.8 | 0.182 |
| (mean ± SD) | (53.3 ± 24.0) | (54.6 ± 24.2) | (50.8 ± 23.6) | (56.2 ± 25.1) | (50.4 ± 22.7) | ||
| Time to recurrence (months) | |||||||
| Median | 18a, 17.7 | 10, 23.5 | 8, 15.2 | 0.234 | 6, 13.0 | 12, 20.5 | 0.947 |
| (mean ± SD) | (19.7 ± 10.8) | (22.4 ± 12.6) | (16.2 ± 7.5) | (19.4 ± 15.8) | (19.8 ± 8.2) | ||
| Time to metastasis (months) | |||||||
| Median | 23a, 17.5 | 14, 14.0 | 9, 23.2 | 0.093 | 12, 22.8 | 11, 16.0 | 0.583 |
| (mean ± SD) | (19.3 ± 11.7) | (16.0 ± 11.0) | (24.4 ± 11.5) | (20.6 ± 13.2) | (17.8 ± 10.4) | ||
Number of cases.
Percentage of cases within the subgroups.
The time interval between the end of irradiation and surgery averaged 22.0 days (range 3–53) and did not differ significantly for males and females (Table 1). In the whole group, OTT ranged from 7 to 58 days (Table 1).
At the time of recruitment no significant differences between the male and female subgroups were found for all clinical parameters apart from the pT category (Table 1). In addition, the mean time to recurrence and metastasis did not differ between male and female patients (23.5 vs. 15.2 months, P = 0.234; 14.0 vs. 23.2 months, respectively, P = 0.092) (Table 1).
There was also no significant difference in the assessed prognostic parameters for patients treated with short or long break in the treatment apart from pTNM, and tumor grade (Table 1). In addition, no significant difference in prognostic parameters (listed in Table 1) was indicated between a short/long break within the male and female cohorts.
4.2. Biological markers
BrdUrdLI, SPF and DNA ploidy was assessed in all tumors. Immunostaining for CD34, Ki-67, GLUT-1, Ku70, P53 and BCL-2 protein was performed on 88, 108, 106, 102, 106 and 90 tumors, respectively, as in some cases there was insufficient material left in the paraffin block. Expression of CD34, Ki-67 and Ku70 was observed in all cases. Immunonegativity was noted in 21 (19.8%) cases of GLUT-1, 44 (41.5%) of P53 (LI ≤ 25%), and 63 (70.0%) of BCL-2. The mean MVD was 140.8 vessels/mm2, mean MIB-1LI, GLUT-1LI, Ku70LI and P53LI was 52.8%, 14.7%, 74.5% and 43.6%, respectively. These values did not differ between the male and female subgroups. In the whole group, median BrdUrdLI was 7.9% ± 0.5 and mean SPF was equal to 21.0 ± 0.9. Sixty-six (50.8%) aneuploid tumors were identified. There was no significant correlation between analyzed markers and cTNM and pTNM stage or grade.
4.3. Overall survival analysis
Thirty (23.8%) not cancer-specific deaths (20 males and 10 females) were indicated. The median follow-up time was 53.5 months (Table 1). Overall, 10-year survival was 69% in females and 59% in males (P = 0.832). Younger patients (≤62.0 years) with well differentiated tumors (P = 0.013) and those with less advanced pTNM (P = 0.040) survived significantly better (Table 2). However, in the male subgroup, only patients’ age and tumor grade were significant, and in the female subgroup none of these parameters were important (data not shown). In the whole patient cohort a break (≤17 days) in the treatment, OTT (≤22 days), and P53 negativity were significant prognostic factors for overall survival (Table 2). Patients treated with SCRT with a break longer than 17 days survived significantly worse than those treated with a shorter break (P = 0.030). This was particularly important for male patients (Fig. 1a) and not significant in the female subgroup (Fig. 1b). Only when the population was separated by gender, was favorable association of low Ku70 expression with OS observed in the female patients (Fig. 1c, d), and it was the only parameter significant for OS in the female subgroup. P53 negativity was a predictive factor for long OS but significant only in the male subgroup (Fig. 1e).
Table 2.
Univariate analysis for overall survival (OS), local recurrence-free survival (RFS), and metastasis-free survival (MFS) for 126 rectal cancer patients.
| Variable | OS |
RFS |
MFS |
||||||
|---|---|---|---|---|---|---|---|---|---|
| RR | 95% CI | P-value* | RR | 95% CI | P-value* | RR | 95% CI | P-value* | |
| Age (years) | |||||||||
| ≤62.0 | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| >62.0 | 4.61 | 1.98–10.75 | <0.001 | 1.33 | 0.53–3.35 | 0.551 | 0.67 | 0.28–1.58 | 0.355 |
| Gender | |||||||||
| Male | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| Female | 1.09 | 0.51–2.33 | 0.832 | 1.76 | 0.69–4.45 | 0.236 | 1.28 | 0.55–2.96 | 0.561 |
| Grade | |||||||||
| 1 | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| 2–3 | 5.08 | 1.21–21.39 | 0.013 | 1.74 | 0.50–6.02 | 0.372 | 2.33 | 0.69–7.86 | 0.157 |
| Clinical stage (AJCC) | |||||||||
| I | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| II–III | 2.04 | 0.78–5.34 | 0.125 | 1.03 | 0.37–2.90 | 0.950 | 1.78 | 0.61–5.24 | 0.279 |
| Pathological stage (AJCC) | |||||||||
| 0–1 | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| 2–3 | 2.17 | 1.02–4.65 | 0.040 | 2.88 | 1.03–8.08 | 0.036 | 11.92 | 2.79–50.86 | <0.001 |
| Break in the treatment | |||||||||
| ≤17 days | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| >17 days | 2.26 | 1.06–4.84 | 0.030 | 2.58 | 0.97–6.89 | 0.050 | 0.99 | 0.44–2.26 | 0.991 |
| Cut off point 21.5 days | |||||||||
| OTT | |||||||||
| ≤22 days | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| >22 days | 2.32 | 1.09–4.97 | 0.030 | 2.32 | 0.87–6.18 | 0.100 | 1.03 | 0.46–2.34 | 0.991 |
| Ki-67LI | |||||||||
| ≤ 52.8% | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| >52.8% | 0.74 | 0.36–1.54 | 0.422 | 0.57 | 0.22–1.48 | 0.246 | 0.41 | 0.16–1.09 | 0.065 |
| Ku70 | |||||||||
| ≤74.9% | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| >74.9% | 1.53 | 0.72–3.24 | 0.264 | 1.21 | 0.48–3.06 | 0.690 | 0.98 | 0.39–2.48 | 0.964 |
| GLUT-1 | |||||||||
| <14.7% | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| ≥14.7% | 0.84 | 0.38–1.87 | 0.675 | 0.49 | 0.16–1.49 | 0.193 | 0.46 | 0.15–1.39 | 0.155 |
| CD34 (MVD) | |||||||||
| <140.8 vessels/mm2 | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| ≥140.8 vessels/mm2 | 1.23 | 0.52–2.84 | 0.634 | 3.17 | 0.97–10.30 | 0.043 | 1.85 | 0.69–4.96 | 0.216 |
| P53 | |||||||||
| ≤25.0% | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| >25.0% | 3.29 | 1.24–8.79 | 0.010 | 1.12 | 0.43–2.94 | 0.820 | 2.94 | 0.97–8.95 | 0.043 |
| BrdUrdLI | |||||||||
| ≤7.9% | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| >7.9% | 2.03 | 0.96–4.29 | 0.058 | 2.99 | 1.06–8.40 | 0.029 | 3.95 | 1.47–10.64 | 0.003 |
| BCL-2 | |||||||||
| Positivity | 0.62 | 0.27–1.42 | 0.073 | 0.38 | 0.12–1.23 | 0.404 | 0.39 | 0.12–1.23 | 0.145 |
| Negativity | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| SPF | |||||||||
| ≤21.0% | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| >21.0% | 1.57 | 0.76–3.25 | 0.215 | 1.47 | 0.58–3.72 | 0.414 | 2.23 | 0.95–5.27 | 0.059 |
| DNA ploidy | |||||||||
| Diploid | 1.00 | Reference | 1.00 | Reference | 1.00 | Reference | |||
| Aneuploid | 0.97 | 0.47–1.99 | 0.936 | 2.60 | 0.93–7.29 | 0.059 | 2.14 | 0.88–5.20 | 0.085 |
P values from log-rank test.
Fig. 1.
Gender-related differences in the influence of break in the treatment, Ku70, and P53 expression on overall survival of male (a, c, e) and female (b, d, f) rectal cancer patients.
4.4. Recurrences
In 18 (14.3%) patients, local recurrences were indicated: 10 (11.9%) in male and 8 (18.2%) in female patients. In a break >17 days, recurrences appeared in 6 (60%) males and in 6 (75%) females. The univariate analysis revealed that higher pTNM, a break ≥ 21.5 days (significant cut-off point), MVD > 140.8/mm2, BrdUrdLI > 7.9% and tumor aneuploidy (border significance) were unfavorable prognostic factors for RFS (Table 2). The different influence of tumor biological factors on RFS was noticed for males and females (Fig. 2a–d). Male patients with low tumor proliferation (Fig. 2a) and low MVD (Fig. 2c) had higher probability of RFS. In females, these differences were not significant (Fig. 2b, d).
Fig. 2.
The difference in recurrence-free survival between male and female rectal cancer patients based on BrdUrdLI (a, b) and microvessel density (MVD) (c, d).
4.5. Metastases
Metastases appeared in 23 (18.2%) patients: 14 (16.7%) males and 9 (21.4%) female patients. In treatment with a break ≤17 days, metastases were made in 7 males (50%) and in 5 females (55%). While in treatment with a long break, cancer spread was observed in 7 (50%) males and 4 (44.4%) female patients. There was no statistical difference in MFS between male and female patients (P = 0.561) (Table 2). In univariate analysis, higher pTNM, P53 immunopositivity and BrdUrdLI >7.9% were unfavorable prognostic factors for MFS (Table 2). In addition, for MFS the different biological factors were significant for male and female patients (Fig. 3a–f). Female patients with slowly proliferating (Fig. 3b) and diploid tumors (Fig. 3d) had significantly higher probability of MFS. In male patients, P53 negative tumors less frequently created metastasis (Fig. 3e) than in females with the same status (Fig. 3f).
Fig. 3.
Influence of patient's gender and BrdUrdLI (a, b), tumor DNA ploidy (c, d) and P53 expression (e, f) on metastasis-free survival of rectal cancer patients.
In our series, no female patient with slowly proliferating and only one with a diploid tumor developed metastases, whereas, for aneuploid and fast proliferating tumors, MFS in females was lower than in males (Fig. 3a–d).
4.6. Multivariate Cox analysis
Univariate analysis revealed a different influence of biological parameters between the patient genders. Multivariate Cox analysis demonstrated that for the whole patients’ cohort higher tumor grade, pTNM stage, a break > 17 days and P53 positivity + BrdUrdLI > 7.9% appeared to be independent negative factors for OS. In the male patients subgroup, however, only age > 62 years and a long break were significant (Table 3). In the female subgroup multivariable analysis was not performed because only one variable (Ku70) was significant in univariate analysis.
Table 3.
Multivariate Cox analysis performed to determine prognostic factors for overall survival (OS), recurrence-free survival (RFS) and metastasis-free survival (MFS) for rectal cancer patients.
| All patients |
Male patients |
||||||
|---|---|---|---|---|---|---|---|
| Variable | RR | 95% CI | P-value | Variable | RR | 95% CI | P-value |
| OS | |||||||
| Grade | Age | ||||||
| 1 | 1.00 | Reference | ≤62 years | 1.00 | Reference | ||
| 2–3 | 6.48 | 1.43–29.35 | 0.015 | >62 years | 4.68 | 1.55–14.08 | 0.006 |
| Break in the treatment | Break in the treatment | ||||||
| ≤17 days | 1.00 | Reference | ≤17 days | 1.00 | Reference | ||
| >17 days | 4.69 | 1.89–11.60 | 0.001 | >17 days | 2.78 | 1.06–7.28 | 0.037 |
| Pathological stage (AJCC) | |||||||
| 0–1 | 1.00 | Reference | |||||
| 2–3 | 2.84 | 1.19–6.78 | 0.019 | ||||
| P53 and BrdUrdLI | |||||||
| P53 ≤ 25.0% or BrdUrdLI ≤7.9% | 1.00 | Reference | |||||
| P53 > 25.0% and BrdUrdLI > 7.9% | 2.82 | 1.26–6.30 | 0.011 | ||||
| RFS | |||||||
| CD34 (MVD) | BrdUrdLI | ||||||
| <140.8 vessels/mm2 | 1.00 | Reference | ≤7.9% | 1.00 | Reference | ||
| ≥140.8 vessels/mm2 | 3.82 | 1.16–12.52 | 0.027 | >7.9% | 4.97 | 1.05–23.45 | 0.043 |
| DNA ploidy | |||||||
| Diploid | 1.00 | Reference | |||||
| Aneuploid | 4.18 | 1.14–15.31 | 0.031 | ||||
| MFS | |||||||
| Pathological stage (AJCC) | Pathological stage (AJCC) | ||||||
| 0–1 | 1.00 | Reference | 0–1 | 1.00 | Reference | ||
| 2–3 | 8.31 | 1.90–36.37 | 0.005 | 2–3 | 7.94 | 1.77–35.50 | 0.007 |
| P53 and BrdUrdLI | |||||||
| P53 ≤ 25.0% or BrdUrdLI ≤ 7.9% | 1.00 | Reference | |||||
| P53 > 25.0% and BrdUrdLI > 7.9% | 2.99 | 1.17–7.61 | 0.022 | ||||
In the whole group high MVD and tumor aneuploidy were negative prognosticators for RFS. Higher tumor proliferation (BrdUrdLI > 7.9%) was a negative prognostic factor for RFS in the male subgroup. In the female subgroup the multivariate analysis could not be performed because only one variable (pTNM stage) was significant. Higher pTNM stage, and P53 positivity combined with higher proliferation (BrdUrdLI > 7.9%) were important for MFS in the whole cohort (Table 3). In female patients the analysis for MFS could not be performed because there are a low number of uncensored cases in some variables (100% MFS for females with lower pTNM stage and slowly proliferating tumors).
5. Discussion
We analyzed patients with locally advanced rectal cancer after short-course preoperative RT. The difference in overall survival between male and female patients was not found, although, gender-related difference in the influence of clinical and tumor biological factors for OS, RFS and MFS was shown.
Multivariate Cox analysis revealed, for the first time, that the interval between RT and surgery ≤17 days, lower tumor grade and pTNM stage, P53 negativity, or low tumor proliferation (BrdUrdLI ≤ 7.9%) are related to higher OS in patients with locally advanced rectal cancer. Tumor diploidy and lower MVD were identified as independent prognostic factors for higher RFS. Higher pTNM stage and P53 overexpression together with faster tumor proliferation (BrdUrdLI > 7.9%) led to lower MFS.
A gender-related difference in the influence of biological parameters on OS, RFS and MFS was observed. We believe that these differences were not probably influenced by clinical factors as differences in clinicopathological factors (apart from a higher pT in the male subgroup), and treatment characteristics between the male and female subgroups were not statistically important. These parameters were also not different in patients’ cohorts treated with a short and long break, apart from pTNM stage and grade. Stage pTNM = 0 observed in 6 patients treated with a long break was expected, as about two weeks after RT (short break) downstaging is difficult to indicate. Significantly higher grade in male tumors treated with SCRT with a short break if important should give adverse clinical effect than observed.
Sex-specific differences in biological factors and relevant underlying mechanisms may be caused, for example, by hormonal signaling (estrogen, progesterone) which can act concurrently, either stimulating or suppressing immune response,14 cell proliferation and radiosensitivity. It was previously shown that gender differences in tumor response may be caused by: epigenetic changes (higher frequency of DNA hypermethylation in males rather than females), which could be down-regulated by estrogen,15, 16 or estrogen receptor α expression which can protect p53 from deactivation.17 This could confirm our results showing slightly higher OS in female rather than male patients with negative P53 (Fig. 1e, f).
In addition, as shown in our study, higher tumor cell radiosensitivity (less effective DNA damage repair in Ku70 ≤74.9% subgroup) in female rather than male tumors was probably the cause of higher female survival. This finding is in agreement with recent results of Alsbeih et al.,18 where based on non-transformed fibroblast cell strains, difference in radiosensitivity between males and females was shown.
Tumor cell proliferation, efficiency of DNA repair damage and P53 activity are known important factors in RT response. These parameters may be responsible for tumor repopulation within the interval between RT and surgery. In our study time from RT to surgery >17 days appeared to be an unfavorable prognostic factor for OS, especially in males. Other authors who used SCRT or different adjuvant treatments did not show the significant influence of break length on OS (but they did not analyze tumor biological factors).2, 3, 4 We believe that the lack of break influence on OS may be true only for patients treated more intensively: CHT/RT with dose of 50.4 Gy,19 SCRT with immediate surgery, or two-week course of preoperative chemoradiotherapy20 but not for those treated with SCRT with a long break and BED10 of 37.5 Gy (NTD = 31.2 Gy), like in the case of our series. Therefore, our data are in strong contradiction with the results of Rega et al.6 who after SCRT recommend a break before the surgery longer than 8 weeks. Our opinion may support the results of Lefevre and colleagues21 who reported that surgical morbidity was higher when surgery was delayed for 11 weeks rather than for 7 weeks after radiochemotherapy.
In the entire group Ku70 expression and MVD were not important prognostic factors but appeared significant in at least one subgroup when analyzed separately for males and females. This shows a potential importance of analyzing rectal cancer data both with and without gender as a stratifying factor.
In our study, P53 expression was a strong predictor in multivariate analysis of distal metastases and OS, which is in agreement with other studies analyzing patients after surgery22 or neoadjuvant CHT and/or RT.23 It has been shown that in rectal cancer, the loss of P53 function induces mitochondrial genomic instability, and mitochondrial mutations promote aggressive tumor behavior and apoptotic resistance.24 In our study patients with P53 positivity and higher tumor proliferation (BrdUrdLI > 7.9%) indicated higher tumor cell survival after RT, which led to lower patients survival, and which may support an earlier study.25 In female patients cohort, low (≤74.9%) Ku70 expression was the strongest parameter for high OS, which was not true for the male subgroup (Fig. 1c, d). This may suggest high effectiveness of the DNA repair process in male tumors and greater cell radioresistance. After RT, male tumor cells with Ku70 ≤ 74.9% expression might have more efficient DNA damage repair than female cells and in P53 positive patients they may cause lower tumor cell apoptosis which, in consequence, causes lower OS, particularly when the break between RT and surgery is longer than 17 days.
Multivariate analysis pointed out that aneuploid and better oxygenated tumors had higher potential to create recurrences, which caused lower probability of patients’ RFS, and has been confirmed by other studies.26, 27 Moreover, only in male patients was lower proliferation a positive prognostic factor and indicated lower incidence of local recurrences. For MFS, in the whole patient cohort, negative biological factors were – BrdUrdLI > 7.9% and P53 immunopositivity, which led to higher incidence of metastasis and lower MFS, particularly in male patients with a higher pTNM stage. In the female cohort, tumor diploidy and low proliferation indicated highest MFS. Therefore, in clinical practice negative biomarkers, especially in male patients, should be used for more aggressive treatment.
Our data also showed that tumor biological factors, like DNA repair, hypoxia, and tumor proliferation, are important players in tumor response after RT. These findings emphasize the clinical utility of tumor biological markers assessment in rectal cancer for personalized therapy. We also think that the impact of gender difference on the effect of cancer treatment should be considered in clinical trials and the results should be reported sex-specifically. This strategy can facilitate individualization of treatment.
6. Conclusions
Our study revealed that in male rectal patients treated with SCRT, intervals between RT and surgery >17 days should be avoided because they negatively influence patients’ survival. Negative biological factors may be helpful in personalized therapy.
Conflict of interest
None declared.
Financial disclosure
This work was supported by the Polish Government as a grant from the State Committee for Scientific Research No: PBZ-KBN-091/P05/2003.
Acknowledgments
The authors are grateful to Prof. Andrzej Sokolowski for help in statistical analysis.
References
- 1.Valentini V., Glimelius B., Haustermans K. URECCA consensus conference highlights about rectal cancer clinical management: the radiation oncologist's expert review. Radiother Oncol. 2014;110:195–198. doi: 10.1016/j.radonc.2013.10.024. [DOI] [PubMed] [Google Scholar]
- 2.Zhou Z.R., Liu S.-X., Zhang T.-S. Short-course preoperative radiotherapy with immediate surgery versus long-course chemoradiation with delayed surgery in the treatment of rectal cancer: a systematic review and meta-analysis. Surg Oncol. 2014;23:211–221. doi: 10.1016/j.suronc.2014.10.003. [DOI] [PubMed] [Google Scholar]
- 3.Van der Broek C.B.M., Vermeer T.A., Bastiaannet E., Rutten H.J., van de Velde C.J., Marijnen C.A. Impact of the interval between short-course radiotherapy and surgery on outcomes of rectal cancer patients. Eur J Cancer. 2013;49:3131–3139. doi: 10.1016/j.ejca.2013.05.025. [DOI] [PubMed] [Google Scholar]
- 4.Van Gijn W., van Stiphout R.G.P.N., van de Velde C.J.H. Nomograms to predict survival and the risk for developing local or distant recurrence in patients with rectal cancer treated with optional short-term radiotherapy. Ann Oncol. 2015;26:928–935. doi: 10.1093/annonc/mdv023. [DOI] [PubMed] [Google Scholar]
- 5.Pettersson D., Holm T., Iversen H., Blomqvist L., Glimelius B., Martling A. Preoperative short-course radiotherapy with delayed surgery in primary rectal cancer. Br J Surg. 2012;99:577–583. doi: 10.1002/bjs.7796. [DOI] [PubMed] [Google Scholar]
- 6.Rega D., Pecori B., Scala D. Evaluation of tumor response after short-course radiotherapy and delayed surgery for rectal cancer. PLOS ONE. 2016;11:1–10. doi: 10.1371/journal.pone.0160732. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Erlandsson J., Holm T., Pettersson D. Optimal fractionation of preoperative radiotherapy and timing to surgery for rectal cancer (Stockholm III) a multicentre, randomized, non-blinded, phase 3, non-inferiority trial. Lancet Oncol. 2017;18:336–346. doi: 10.1016/S1470-2045(17)30086-4. [DOI] [PubMed] [Google Scholar]
- 8.Gasinska A., Skolyszewski J., Popiela T. Bromodeoxyuridine labeling index as an indicator of early tumor response to preoperative radiotherapy in patients with rectal cancer. J Gastrointest Surg. 2007;11:520–528. doi: 10.1007/s11605-007-0127-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Gasinska A., Richter P., Darasz Z. Gender-related differences in repopulation and early tumor response to preoperative radiotherapy in rectal cancer patients. J Gastrointest Surg. 2011;15:1568–1576. doi: 10.1007/s11605-011-1589-4. [DOI] [PubMed] [Google Scholar]
- 10.Gasinska A., Adamczyk A., Niemiec J., Biesaga B., Darasz Z., Skolyszewski J. Gender-related differences in pathological and clinical tumor response based on immunohistochemical proteins expression in rectal cancer patients treated with short course of preoperative radiotherapy. J Gastrointest Surg. 2014;18:1306–1318. doi: 10.1007/s11605-014-2526-0. [DOI] [PubMed] [Google Scholar]
- 11.Hamilton S.R., Bosman F.T., Boffetta P. Carcinoma of the colon and rectum. In: Bosman F.T., Carneiro F., Hruban R.H., Theise N.D., editors. WHO classification of tumours of the digestive system. 4th ed. International Agency for Research on Cancer Lyon; 2010. pp. 132–146. [Google Scholar]
- 12.Edge S.B., Byrd D.R., Compton C.C., Fritz A.G., Greene F.L., Trotti A., editors. AJCC cancer staging manual. 7th ed. Springer; New York, NY: 2010. [Google Scholar]
- 13.Kaplan E.L., Meier P. Non-parametric estimation for incomplete observations. J Am Stat Assoc. 1956;53:457–481. [Google Scholar]
- 14.Wichmann M.W., Muller C., Hornung H.M., Lau-Werner U., Schildberg F.W., Colorectal Cancer Study Group Gender differences in long-term survival of patients with colorectal cancer. Br J Surg. 2001;88:1092–1098. doi: 10.1046/j.0007-1323.2001.01819.x. [DOI] [PubMed] [Google Scholar]
- 15.Lai J.C., Cheng Y.W., Chiou H.L., Wu M.-F., Chen C.-Y., Lee H. Gender difference in estrogen receptor alpha promoter hypermethylation and its prognostic value in non-small cell lung cancer. Int J Cancer. 2005;117:974–980. doi: 10.1002/ijc.21278. [DOI] [PubMed] [Google Scholar]
- 16.Issa J.-P.J., Ottaviano Y.L., Celano P. Methylation of the oestrogen receptor CpG island links ageing and neoplasia in human colon. Nat Genet. 1994;7:536–540. doi: 10.1038/ng0894-536. [DOI] [PubMed] [Google Scholar]
- 17.Liu G., Schwartz J.A., Brooks S.C. Estrogen receptor protects p53 from deactivation by Human Double Minute-21. Cancer Res. 2000;60:1810–1814. [PubMed] [Google Scholar]
- 18.Alsbeih G., Al-Meer R.S., Al-Harabi N. Gender bias in individual radiosensitivity and the association with genetic polymorphic variations. Radither Oncol. 2016;119:236–243. doi: 10.1016/j.radonc.2016.02.034. [DOI] [PubMed] [Google Scholar]
- 19.Calvo F.A., Morillo V., Santos M. Interval between neoadjuvant treatment and definitive surgery in locally advanced rectal cancer: impact on response and oncologic outcomes. J Cancer Res Clin Oncol. 2014;140:1651–1660. doi: 10.1007/s00432-014-1718-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Lee J.H., Kim J.G., Oh S.T. Two-week course of preoperative chemoradiotherapy followed by delayed surgery for rectal cancer: a phase II multi-institutional clinical trial (KROG 11-02) Radiother Oncol. 2014;110:150–154. doi: 10.1016/j.radonc.2013.11.013. [DOI] [PubMed] [Google Scholar]
- 21.Lefevre J.H., Mineur L., Kotti S. Effect of interval (7 or 11 weeks) between neoadjuvant radiochemotherapy and surgery on complete pathologic response in rectal cancer: a multicenter, randomized, controlled trial (GRECCAR-6) J Clin Oncol. 2016;34:3773–3780. doi: 10.1200/JCO.2016.67.6049. [DOI] [PubMed] [Google Scholar]
- 22.Resnick M.B., Routhier J., Konkin T., Sabo E., Pricolo V.E. Epidermal growth factor receptor, c-MET, β-catenin, and p53 expression as prognostic indicators in stage II colon cancer: a tissue microarray study. Clin Cancer Res. 2004;10:3069–3075. doi: 10.1158/1078-0432.ccr-03-0462. [DOI] [PubMed] [Google Scholar]
- 23.Torsello A., Garufi C., Cosimelli M. P53 and bcl-2 in colorectal cancer arising in patients under 40 years of age: distribution and prognostic relevance. Eur J Cancer. 2008;44:1217–1222. doi: 10.1016/j.ejca.2008.03.002. [DOI] [PubMed] [Google Scholar]
- 24.Chen D., Yu Z., Zhu Z., Lopez Ch D. The p53 pathway promotes efficient mitochondrial DNA base excision repair in colorectal cancer cells. Cancer Res. 2006;66:3485–3494. doi: 10.1158/0008-5472.CAN-05-4103. [DOI] [PubMed] [Google Scholar]
- 25.Komuro Y., Watanabe T., Hosoi Prediction of tumor radiosensitivity in rectal carcinoma based on p53 and Ku 70 expression. J Exp Clin Cancer Res. 2003;22:223–228. [PubMed] [Google Scholar]
- 26.Kimura O., Kijma T., Moriwaki S. DNA index as a significant indicator of lymph node metastasis and local recurrence of rectal cancer. Dis Colon Rectum. 1992;35:1130–1134. doi: 10.1007/BF02251963. [DOI] [PubMed] [Google Scholar]
- 27.Noda E., Maeda K., Inoue T. Predictive value of vascular endothelial growth factor-C expression for local recurrence of rectal carcinoma. Oncol Rep. 2007;17:1327–1331. [PubMed] [Google Scholar]