Abstract
The accurate assessment of the risk of first locoregional recurrence is very important for improving the survival of patients with invasive ductal carcinoma of the breast. The present study investigated which histological factors (both well‐known histological factors and factors that we have proposed) were the most capable of accurately predicting first locoregional recurrence among 1042 patients with invasive ductal carcinoma and various tumor statuses (overall, nodal status, Union Internationale Contre le Cancer pathological TNM stage, adjuvant therapy status, and adjuvant radiotherapy status) using multivariate analyses by the Cox proportional hazard regression model. The present study clearly demonstrated that the best factor for accurately predicting locoregional recurrence was grade 3 lymph vessel tumor embolus (>4 mitotic figures and >6 apoptotic figures in tumor embolus), followed by type 2 invasive ductal carcinoma (negative for fibrotic foci but positive for atypical tumor‐stromal fibroblast), grade 2 lymph vessel tumor embolus (1–4 mitotic figures and >0 apoptotic figures in tumor embolus; >0 mitotic figures and 1–6 apoptotic figures in tumor embolus), primary invasive tumor cell‐related factors (>19 mitotic figures, presence of tumor necrosis, presence of skin invasion) and >5 mitotic figures in metastatic carcinomas to the lymph node. Our proposed factors were superior to well‐known histological factors of primary invasive tumors or clinicopathological factors for the accurate prediction of first locoregional recurrence in patients with invasive ductal carcinoma of the breast.
Locoregiocal recurrence is an important prognostic factor for patients with invasive ductal carcinoma of the breast,1 and several studies have been performed to clarify factors that are significantly associated with locoregional recurrence.2, 3 These studies demonstrated that lymph vessel invasion, histological grade, tumor size, hormone receptor status, and HER2 status are very important predictors of locoregional recurrence in patients with invasive ductal carcinoma. We have already reported histological factors that are significantly associated with distant‐organ metastasis or the tumor‐related death of patients with invasive ductal carcinoma of the breast.4 Since the publication of our previous study,4 we have performed additional studies that identified the following new histological factors as predictors of the outcome of patients with invasive ductal carcinoma of the breast5: (i) type of invasive ductal carcinoma;6 (ii) grading system for lymph vessel tumor emboli;7 (iii) number of apoptotic figures in blood vessel tumor emboli;8 (iv) number of mitotic figures in metastatic carcinomas to the lymph node;9 and (v) maximum dimension of metastatic carcinomas to the lymph node.8 Although our studies clearly demonstrated that the factors we previously reported were very useful for accurately predicting tumor recurrence, distant‐organ metastasis or tumor‐related death,4, 5, 6, 7, 8, 9 we have not yet investigated whether these factors are significantly associated with the locoregional recurrence of invasive ductal carcinoma of the breast. We are confident that clarification of the recurrent or metastatic patterns of invasive ductal carcinomas based on their histological features will provide clinicians, pathologists, and scientists with very important clues for accurately evaluating the true biological characteristics of invasive ductal carcinomas. Such a result would likely contribute to the establishment of targeted therapies for patients with invasive ductal carcinoma of the breast.
The purpose of the present study was to investigate which histological factors were most capable of accurately predicting first locoregional recurrence in patients with invasive ductal carcinoma of the breast.
Materials and Methods
Patients and histological examinations
The subjects of this study were 1042 consecutive patients with invasive ductal carcinoma of the breast who did not receive neoadjuvant therapy and were surgically treated at the National Cancer Center Hospital between January 2000 and December 2005 (the same case series as that used in our previous study).5 The invasive ductal carcinomas were diagnosed preoperatively using needle biopsy, aspiration cytology, mammography, or ultrasonography. All the patients were Japanese women, ranging in age from 23 to 72 years old (median, 55 years). All the tumors were classified according to the pathological UICC‐TNM (pTNM) classification.10 The protocol (20–112) for this study was reviewed by the institutional review board of the National Cancer Center.
The clinicopathological factors, well‐known histological factors and the eight factors that we previously proposed were evaluated and we arranged the above mentioned factors into five groups (Table 1). The eight factors that we previously proposed are as follows (Tables 1 and 2): (i) fibrotic focus;11, 12 (ii) type of invasive ductal carcinoma (Fig. 1a–c);6 (iii) grading system for lymph vessel tumor emboli (Fig. 1d–f);7 (iv) number of apoptotic figures in blood vessel tumor emboli;8 (v) grade of stromal fibrosis in metastatic carcinomas to the lymph node;8 (vi) maximum dimension of metastatic carcinomas to the lymph node;8 (vii) number of extranodal blood vessel tumor emboli;8 and (viii) number of mitotic figures in metastatic carcinomas to the lymph node.9 In the present study, seven of the 598 pN0 cases had isolated tumor cell clusters (ITC)10 (Table 3). We excluded these seven cases from the pN0 cases and these cases showed no stromal fibrosis in metastatic carcinomas to the lymph node, showed a ≤20 mm maximum dimension in metastatic carcinomas to the lymph node, showed ≤2 extranodal blood vessel tumor emboli, or showed ≤5 mitotic figures in metastatic carcinomas to the lymph node (Table 3). Thus, we classified these seven cases as cases with no grade of stromal fibrosis, those with a ≤20 mm maximum dimension, those with ≤2 extranodal blood vessel tumor emboli, or those with ≤5 mitotic figures in metastatic carcinomas to the lymph node (Table 3).
Table 1.
Groups and factors
Table 2.
Histological features, criteria or assessing methods of the five factors that we have proposed
| 1 | Histological features of atypical tumor‐stromal fibroblasts and the type of invasive ductal carcinoma | |||||
| (1) | The presence of atypical tumor‐stromal fibroblasts was defined based on the presence of one or more atypical tumor‐stromal fibroblasts in the tumor stroma inside and outside of the fibrotic foci in invasive ductal carcinoma. Although atypical tumor‐stromal fibroblasts are occasionally distributed at random locations in the tumor stroma, they tend to exist within the cellular area of the tumor‐stromal fibroblasts | |||||
| (2) | The number of nuclei in an atypical tumor‐stromal fibroblast is one or more. The nuclear size of an atypical tumor‐stromal fibroblast is two or more times larger than that of an ordinary tumor‐stromal fibroblast. The nuclear features of an atypical tumor‐stromal fibroblast include an irregular or convoluted shape, and also include various bizarre shapes | |||||
| (3) | An obvious small to large size nucleolus or nucleoli are seen in the nucleus or nucleoli of atypical tumor‐stromal fibroblasts and some atypical tumor‐stromal fibroblasts show a coarsely granulated nuclear chromatin pattern | |||||
| Type | Fibrotic focus | Atypical tumor‐stromal fibroblast not forming a fibrotic focus | Atypical tumor‐stromal fibroblast forming a fibrotic focus | |||
| 1 | Absent | Absent | Not applicable | |||
| 2 | Absent | Present | Not applicable | |||
| 3 | Present | Not assessed | Absent | |||
| 4 | Present | Not assessed | Present | |||
| 2 | Grading system for lymph vessel tumor embolus | |||||
| Grade 0 | Invasive ductal carcinomas with no lymph vessel tumor embolus | |||||
| Grades 1–3 | Invasive ductal carcinomas with lymph vessel tumor embolus or emboli | |||||
| No. of mitotic figures | No. of apoptotic figures | |||||
| Grade 1 | 0 | 0 | ||||
| 0 | Any | |||||
| Any | 0 | |||||
| Grade 2 | 1–4 | >0 | ||||
| >0 | 1–6 | |||||
| Grade 3 | >4 | >6 | ||||
| (1) | The numbers of tumor cell mitotic figures and apoptotic figures in lymph vessels are counted in 20 high‐power fields. In carcinomas containing a small number of lymph vessel tumor emboli, the mitotic figures and apoptotic figures are counted in fewer than 20 high‐power fields | |||||
| (2) | A large lymph vessel tumor emboli located far from the stroma‐invasive tumor margin is selected and the mitotic figures and apoptotic figures in the lymph vessel tumor emboli or embolus are counted | |||||
| (3) | The numbers of mitotic figures and apoptotic figures in tumor cells composing the lymph vessel tumor embolus or emboli in the high‐power field containing the largest number of mitotic figures, and/or the largest number of apoptotic figures are recorded as the number of mitotic figures and apoptotic figure in the lymph vessel tumor emboli or embolus. The cumulative numbers of tumor cell mitotic figures and apoptotic figures in the lymph vessel tumor emboli in all 20 high‐power fields are not used | |||||
| 3 | Grade of stromal fibrosis in metastatic carcinomas to the lymph node | |||||
| None | Metastatic carcinoma with no tumor‐stromal fibrosis | |||||
| Mild | Metastatic carcinoma occupied by ≤30% tumor‐stromal fibrosis | |||||
| Moderate | Metastatic carcinoma occupied by >30 to ≤80% tumor‐stromal fibrosis | |||||
| Severe | Metastatic carcinoma occupied by >80% tumor‐stromal fibrosis | |||||
| 4 | Extranodal blood vessel tumor embolus or emboli | |||||
| Tumor embolus or emboli in blood vessel or vessels with a smooth muscle‐supported endothelial lining in perinodal adipose tissues was/were assessed as extranodal blood vessel tumor embolus or emboli | ||||||
| 5 | Mitotic figures in metastatic carcinomas to the lymph node | |||||
| (1) | A random search for mitotic figures in metastatic mammary carcinoma to the lymph nodes is performed using high‐power magnification fields (×10 or ×20) of the tumor area | |||||
| (2) | Next, one high‐power magnification field (×40) of the tumor area containing the highest number of mitotic figures is selected to determine the largest number of metastatic mammary carcinoma to the lymph nodes exhibiting mitotic figures | |||||
Figure 1.
(a–c) Type 2 invasive ductal carcinoma. One atypical tumor‐stromal fibroblast with a large spindle nucleus is visible in the tumor stroma (arrow). The fibroblast was stained negative for AE1/3 (arrow, b) and positive for smooth muscle actin (arrow, c). The invasive tumor cells were stained positive for AE1/3 (b). (d–f) Grade 3 lymph vessel tumor emboli. Three large lymph vessel tumor emboli are present, and the wall of one of the tumor lymph vessels containing the embolus was positive for D2–40 (arrowheads, e). Five mitotic tumor cells (arrows) and eight apoptotic tumor cells (arrowheads) are visible within the tumor embolus (f).
Table 3.
Frequencies of first locoregional recurrence of the eight histological factors that we have proposed and UICC pN category
| Cases (%) 1042 | No. patients (%) | |||
|---|---|---|---|---|
| First locoregional recurrence | ||||
| Present 47 | Absent 995 | P‐value | ||
| Primary tumor‐stromal fibroblast‐related group | ||||
| Fibrotic focus, dimension (mm) | ||||
| Absent | 667 | 30 (5) | 637 (95) | 0.624 |
| ≤8 | 221 | 9 (4) | 212 (96) | |
| >8 | 154 | 8 (5) | 146 95) | |
| Types of invasive ductal carcinoma | ||||
| Type 1 | 627 | 23 (4) | 604 (96) | <0.001 |
| Type 2 | 40 | 7 (18) | 33 (82) | |
| Type 3 | 346 | 15 (4) | 331 (96) | |
| Type 4 | 29 | 2 (7) | 27 (93) | |
| Tumor embolus‐related group | ||||
| Grading system for lymph vessel tumor embolus | ||||
| Grade 0 | 666 | 20 (3) | 646 (97) | <0.001 |
| Grade 1 | 250 | 6 (2) | 244 (98) | |
| Grade 2 | 97 | 12 (12) | 85 (88) | |
| Grade 3 | 29 | 9 (31) | 20 (69) | |
| Number of apoptotic figures in blood vessel tumor emboli | ||||
| Absent | 890 | 36 (4) | 854 (96) | 0.071 |
| ≤2 | 78 | 6 (8) | 72 (92) | |
| >2 | 74 | 5 (7) | 5 (93) | |
| Metastatic carcinomas to the lymph node‐related group | ||||
| UICC pN category | ||||
| pN0 | 598 | 17 (3) | 581 (97) | <0.001 |
| pN1mi | 20 | 0 | 20 (100) | |
| pN1 | 291 | 16 (6) | 275 (94) | |
| pN2 | 85 | 6 (7) | 79 (93) | |
| pN3 | 48 | 8 (17) | 40 (83) | |
| Grade of stromal fibrosis in metastatic carcinomas to the lymph node | ||||
| No nodal metastasis | 591 | 17 (3) | 574 (97) | <0.001 |
| None, mild and moderate | 415 | 25 (6) | 390 (94) | |
| Severe | 36 | 5 (14) | 31 (86) | |
| Maximum dimension of metastatic carcinomas to the lymph node (mm) | ||||
| No nodal metastasis | 591 | 17 (3) | 574 (97) | <0.001 |
| ≤20 | 396 | 26 (7) | 370 (93) | |
| >20 | 55 | 4 (7) | 51 (93) | |
| Number of extranodal blood vessel tumor emboli | ||||
| No nodal metastasis | 591 | 17 (3) | 574 (97) | <0.001 |
| ≤2 | 423 | 25 (6) | 398 (94) | |
| >2 | 28 | 5 (18) | 23 (82) | |
| Number of mitotic figures in metastatic carcinomas to the lymph node | ||||
| No nodal metastasis | 591 | 17 (3) | 574 (97) | <0.001 |
| ≤5 | 286 | 12 (4) | 274 (96) | |
| >5 | 165 | 18 (11) | 147 (89) | |
NA, not available; pN0, no nodal metastasis, but including lymph node with isolated tumor cell clusters (single tumor cells or small clusters of cells not more than 0.2 mm in greatest dimension); pN1mi, cases with micrometastasis (larger than 0.2 mm, but none larger than 2.0 mm in greatest dimension); pN1, 1–3 nodal metastases; pN2, 4–9 nodal metastases; pN3, 10 or more nodal metastases; no nodal metastasis, pN0 cases excluding the seven cases with lymph nodes containing isolated tumor cell clusters; Grade 0 of grading system for lymph vessel tumor embolus, no lymph vessel invasion.
The following antibodies were used for immunohistochemistry: anti‐estrogen receptor mouse monoclonal antibody ER88 (BioGenex, Fremont, CA, USA), anti‐progesterone receptor mouse monoclonal antibody PR88 (BioGenex), and anti‐HER2 mouse monoclonal antibody CB11 (BioGenex). Allred scores for estrogen receptor or progesterone receptor were assessed according to our previously study.13 We defined an Allred score of 0 or 2 for ER or PR as being negative for ER or PR and Allred scores of 3 or more for ER or PR as being positive for ER or PR. HER2 expression in tumor cells was categorized according to the definition of Wolf.14 All types 2 and 4 invasive ductal carcinomas were immunohistochemically studied using monoclonal antibodies to keratins (AE1/3) to confirm that the atypical tumor‐stromal fibroblasts were not modified invasive tumor cells, and fibroblasts that were negative for keratins were considered as atypical tumor‐stromal fibroblasts (Fig. 1b). We also performed immunohistochemical staining for alpha‐smooth muscle actin for types 2 and 4 invasive ductal carcinomas to investigate whether atypical tumor‐stromal fibroblasts are myofibroblasts (Fig. 1c). Some invasive ductal carcinomas contained large lymph vessel tumor emboli, especially in invasive ductal carcinomas containing a grade 2 or grade 3 lymph vessel tumor emboli, and it was difficult to determine whether they were true lymph vessel tumor emboli or a non‐invasive ductal carcinoma component by hematoxylin and eosin staining alone. We therefore performed immunohistochemical staining with D2‐40 antibody (monoclonal mouse antibody, Signet, Dedham, MA, USA, 1:200) to confirm that the lymph vessel tumor emboli identified by hematoxylin and eosin staining were true tumor emboli in some invasive ductal carcinomas with grade 2 or grade 3 lymph vessel tumor emboli (Fig. 1e). Histologic grade, nuclear feature of primary invasive tumors, and mitotic activity index in primary invasive tumors were evaluated according to the criteria of Elston and Ellis.15 Tumor necrosis in primary invasive tumors was evaluated according to the definition of Gilchrist.16
Patient outcome and statistical analysis
Survival was evaluated using a median follow‐up period of 98 months (range: 63–134 months) until March 2011. Of the 1042 invasive ductal carcinoma patients, first locoregional recurrence was observed in 47 out of 1042 patients with invasive ductal carcinoma. The first locoregional recurrence‐free survival period was calculated using the time of surgery as the starting point. The factors that were significantly associated with first locoregional recurrence in the univariate analyses were then entered together into multivariate analyses using the Cox proportional hazard regression model. In addition, we conducted to compare the power of grading system for lymph vessel tumor emboli with that of the following three lymphatic parameters for accurately predicting the first locoregional recurrence in multivariate analysis using the Cox proportional hazard regression model: (i) the presence or absence of lymph vessel invasion; (ii) real numbers of lymph vessel invasion;17 and (iii) location of lymph vessel tumor emboli18 (inside area of the tumor, advanced area within the tumor and outside area of the tumor). In this study, we were unable to perform multivariate analyses for first locoregional recurrence because of a small sample size (fewer than 10 patients) in patients who did not receive adjuvant therapy. The case‐wise and step‐down method was applied until all the remaining factors were significant at a P‐value of <0.05. First locoregional recurrence‐free survival curves were drawn by the Kaplan–Meier method. All the analyses were performed using Statistica/Windows software (StatSoft, Tulsa, OK, USA).
Results
Patients
All of the patients had a solitary lesion; 498 patients were premenopausal, and 544 were postmenopausal. A partial mastectomy had been performed in 458 patients, and a modified radical mastectomy had been performed in 584. The surgical margins of all the partial mastectomy materials were histologically examined to confirm whether tumor cells were absent or present at the surgical margins of the materials; we confirmed that all the materials had been completely resected because the outermost edges of the tumors were 5 mm or further from the surgical margin of the materials. A Level I and II axillary lymph node dissection had been performed in all the patients, and a Level III axillary lymph node dissection had been performed in some of the patients. Of the 1042 patients, 873 received adjuvant therapy, consisting of chemotherapy in 217 patients, endocrine therapy in 281 patients, and chemoendocrine therapy in 375 patients. The chemotherapy regimens used were anthracycline‐based with or without taxane and non‐anthracycline‐based. The endocrine therapy regimens consisted of tamoxifen with or without a gonadotropin‐releasing‐hormone agonist, tamoxifen, with or without an aromatase inhibitor, an aromatase inhibitor alone, or a gonadotropin‐releasing‐hormone agonist alone. Of the 1042 patients, 466 patients received adjuvant radiotherapy.
Univariate analyses for first locoregional recurrence
Overall, age (P = 0.026), the Allred score for estrogen receptors in the tumor cells (P = 0.017), the histologic grade (P = 0.009), the invasive tumor size (P < 0.001), nuclear features of the primary‐invasive tumor cells (P < 0.001), the number of mitotic figures in the primary‐invasive tumor cells (P = 0.002), tumor necrosis (P = 0.019), the type of invasive ductal carcinoma, the grading system for lymph vessel tumor emboli, the UICC pN category, the grade of stromal fibrosis in metastatic carcinomas to the lymph node, the maximum dimension of metastatic carcinomas to the lymph node, the number of extranodal blood vessel tumor emboli, and the number of mitotic figures in metastatic carcinomas to the lymph node were significantly associated with first locoregional recurrence in the univariate analyses (Table 3). The fibrotic focus dimension (Table 3), the number of apoptotic figures in blood vessel tumor emboli (Table 3), adjuvant therapy, adjuvant radiotherapy, the Allred score for progesterone receptors in the tumor cells, the HER2 category, and the skin invasion were not significantly associated with first locoregional recurrence in the univariate analyses (data not shown). Atypical tumor‐stromal fibroblast was observed in 69 (7%) cases (type 2 and 4 invasive ductal carcinoma cases) among 1042 cases (Table 3). The presence of atypical tumor‐stromal fibroblasts stained positive for alpha‐smooth muscle actin was observed in 60 (87%) out of 69 types 2 and 4 invasive ductal carcinomas (type 2: 35/40 cases, 88%; type 4: 25/29 cases, 86%).6
Multivariate analysis for clarifying the best lymphatic factor for accurately predicting first locoregional recurrence
Number of lymph vessel invasion ranged from 0 to 494 (median number and standard error: 0 and 1.1) in the present study. Only the grading system for lymph vessel tumor emboli significantly increased the hazard ratio for first locoregional recurrence in the multivariate analysis (P = 0.002). The presence of lymph vessel invasion (P = 0.158), real number of lymph vessel tumor emboli (P = 0.144), or location of lymph vessel tumor emboli (inside area of the tumor: P = 0.227; advanced area within the tumor: P = 0.512; outside area of the tumor: P = 0.425) failed to significantly increase the hazard ratio for first locoregional recurrence in the multivariate analysis.
Multivariate analyses for first locoregional recurrence
Overall (n = 1042), lymph vessel tumor embolus grade 2 (P < 0.001, Fig. 2a) and 3 (P < 0.001, Fig. 2a), and type 2 invasive ductal carcinoma (P < 0.001, Fig. 2b) significantly increased the hazard ratios for first locoregional recurrence in the multivariate analyses (Table 8). Lymph vessel tumor embolus grade 3 was significantly associated with first locoregional recurrence in a manner that was independent of almost all the tumor statuses, except for adjuvant radiotherapy status (received adjuvant radiotherapy) (Tables 4, 5, 6, 7, 8). Type 2 invasive ductal carcinoma was significantly associated with first locoregional recurrence among the overall patients who had received adjuvant therapy (P < 0.001), the UICC pN0 patients (Tables 4 and 8), the UICC pN1‐3 patients (Tables 4 and 8), the UICC pTNM stages I and II patients (Tables 5 and 8), the patients who had received endocrine therapy (Tables 6 and 8), the patients who had received chemotherapy (Tables 6 and 8), the patients who had not received adjuvant radiotherapy (Tables 7 and 8) and the patients who had received adjuvant radiotherapy (Tables 7 and 8). Lymph vessel tumor embolus grade 2 was significantly associated with first locoregional recurrence among the overall patients who had received adjuvant therapy (P < 0.001), the UICC pN0 patients (Tables 4 and 8), the UICC pTNM stages I and II patients (Tables 5 and 8) and the patients who had received chemoendocrine therapy (Tables 6 and 8). Twenty or more mitotic figures in primary invasive tumors, the presence of tumor necrosis, and the presence of skin invasion were significantly associated with first locoregional recurrence among the UICC pN0 patients (Tables 4 and 8), among the UICC pTNM stages I and II patients (Tables 5 and 8) and among the patients who had received chemotherapy (Tables 6 and 8), respectively. Six or more mitotic figures in metastatic carcinomas to the lymph node were significantly associated with the first locoregional recurrence among the patients who had received endocrine therapy (Tables 6 and 8), and the patients who had received adjuvant radiotherapy (Tables 7 and 8).
Figure 2.
First locoregional recurrence‐free survival curves for overall patients with invasive ductal carcinoma (a and b). (a) Patients with grade 3 lymph vessel tumor emboli had the shortest locoregional recurrence‐free survival curve. Patients with grade 2 lymph vessel tumor emboli also had a significantly shorter locoregional recurrence‐free survival curve than patients with grade 1 lymph vessel tumor emboli or patients with grade 0 lymph vessel tumor emboli. (b) Patients with type 2 invasive ductal carcinoma had a significantly shorter first locoregional recurrence‐free survival curve than patients with type 1 invasive ductal carcinoma, patients with type 3 invasive ductal carcinoma and patients with type 4 invasive ductal carcinoma.
Table 4.
Multivariate analyses for first locoregional recurrence in invasive ductal carcinoma patients who received adjuvant therapy according to UICC pN category
| First locoregional recurrence | ||||
|---|---|---|---|---|
| pN0 (n = 453) | pN1‐3 (n = 420) | |||
| Hazard ratio (95% CI) | P‐value | Hazard ratio (95% CI) | P‐value | |
| Grading system for lymph vessel tumor embolus | ||||
| Grade 0 | 1.0 | 1.0 | ||
| Grade 1 | 1.0 | 1.5 (0.5–4.5) | 0.503 | |
| Grade 2 | 11.9 (3.0–46.6) | <0.001 | 2.2 (0.7–6.7) | 0.163 |
| Grade 3 | 11.9 (3.0–46.6) | <0.001 | 11.7 (3.4–39.9) | <0.001 |
| Types of invasive ductal carcinoma | ||||
| Type 1 | 1.0 | 1.0 | ||
| Type 2 | 6.1 (1.2–29.9) | 0.025 | 6.3 (2.0–20.0) | 0.002 |
| Type 3 | 2.0 (0.5–8.6) | 0.362 | 0.9 (0.4–2.2) | 0.810 |
| Type 4 | 9.8 (0.9–105.8) | 0.059 | NA | |
| Number of mitotic figures in the primary invasive tumors | ||||
| ≤9 | 1.0 | 1.0 | ||
| >9 to ≤19 | 3.4 (0.3–40.1) | 0.323 | – | |
| >19 | 4.7 (1.2–18.4) | 0.023 | – | |
–, not significant; NA, not available; no nodal metastasis, pN0 cases excluding the seven cases with lymph nodes containing isolated tumor cell clusters.
Table 5.
Multivariate analyses for first locoregional recurrence in invasive ductal carcinoma patients who received adjuvant therapy according to UICC pTNM stage
| First locoregional recurrence | ||||
|---|---|---|---|---|
| Stages I and II (n = 692) | Stage III (n = 181) | |||
| Hazard ratio (95% CI) | P‐value | Hazard ratio (95% CI) | P‐value | |
| Grading system for lymph vessel tumor embolus | ||||
| Grade 0 | 1.0 | 1.0 | ||
| Grade 1 | 1.8 (0.6–5.4) | 0.319 | 0.3 (0.03–2.10) | 0.210 |
| Grade 2 | 7.9 (2.9–20.9) | <0.001 | 0.7 (0.1–3.2) | 0.596 |
| Grade 3 | 15.8 (3.2–77.3) | <0.001 | 8.1 (2.4–28.1) | <0.001 |
| Types of invasive ductal carcinoma | ||||
| Type 1 | 1.0 | 1.0 | ||
| Type 2 | 6.4 (2.3–18.2) | <0.001 | – | |
| Type 3 | 1.2 (0.5–3.3) | 0.685 | – | |
| Type 4 | NA | – | ||
| Tumor necrosis | ||||
| Absent | 1.0 | 1.0 | ||
| Present | 2.4 (1.0–5.8) | 0.045 | – | |
–, not significant; NA, not available.
Table 6.
Multivariate analyses for first locoregional recurrence in invasive ductal carcinoma patients who received adjuvant therapy according to adjuvant therapy status
| First locoregional recurrence | ||||||
|---|---|---|---|---|---|---|
| Endocrine (n = 281) | Chemoendocrine (n = 375) | Chemotherapy (n = 217) | ||||
| Grading system for lymph vessel tumor embolus | ||||||
| Grade 0 | 1.0 | 1.0 | 1.0 | |||
| Grade 1 | 0.6 (0.1–3.4) | 0.602 | 1.7 (0.3–9.1) | 0.545 | 0.8 (0.09–6.40) | 0.795 |
| Grade 2 | 1.6 (0.2–12.0) | 0.667 | 6.8 (1.3–36.8) | 0.026 | 0.8 (0.09–7.30) | 0.866 |
| Grade 3 | 25.8 (1.2–560.0) | 0.038 | 9.8 (1.4–70.8) | 0.024 | 27.5 (6.3–119.1) | <0.001 |
| Types of invasive ductal carcinoma | ||||||
| Type 1 | 1.0 | 1.0 | 1.0 | |||
| Type 2 | 37.2 (3.6–369.7) | 0.002 | – | 18.6 (3.6–90.7) | <0.001 | |
| Type 3 | 7.4 (0.9–59.0) | 0.058 | – | 1.5 (0.4–6.1) | 0.579 | |
| Type 4 | NA | – | 5.4 (0.6–52.6) | 0.145 | ||
| Number of mitotic figures in metastatic carcinomas to the lymph node | ||||||
| No nodal metastasis | 1.0 | 1.0 | 1.0 | |||
| ≤5 | 3.1 (0.7–12.9) | 0.120 | – | – | ||
| >5 | 20.1 (1.3–312.3) | 0.032 | – | – | ||
| Skin invasion | ||||||
| Absent | 1.0 | 1.0 | 1.0 | |||
| Present | – | – | 5.4 (1.4–21.6) | 0.014 | ||
–, not significant; NA, not available.
Table 7.
Multivariate analyses for first locoregional recurrence in invasive ductal carcinoma patients who received adjuvant therapy according to adjuvant radiotherapy status
| First locoregional recurrence | ||||
|---|---|---|---|---|
| No adjuvant radiotherapy (n = 576) | Adjuvant radiotherapy (n = 466) | |||
| Hazard ratio (95% CI) | P‐value | Hazard ratio (95% CI) | P‐value | |
| Types of invasive ductal carcinoma | ||||
| Type 1 | 1.0 | 1.0 | ||
| Type 2 | 3.2 (1.1–9.5) | 0.041 | 6.0 (1.2–29.3) | 0.026 |
| Type 3 | 1.3 (0.5–3.4) | 0.563 | 0.9 (0.2–3.3) | 0.843 |
| Type 4 | 3.1 (0.3–29.8) | 0.334 | NA | |
| Grading system for lymph vessel tumor embolus | ||||
| Grade 0 | 1.0 | 1.0 | ||
| Grade 1 | 1.7 (0.5–5.1) | 0.366 | – | |
| Grade 2 | 3.5 (0.9–16.7) | 0.051 | – | |
| Grade 3 | 129.8 (29.1–578.0) | <0.001 | – | |
| Number of mitotic figures in metastatic carcinomas to the lymph node | ||||
| No nodal metastasis | 1.0 | 1.0 | ||
| ≤5 | – | 1.6 (0.3–8.4) | 0.603 | |
| >5 | – | 5.6 (1.8–17.4) | 0.003 | |
–, not significant; NA, not available.
Table 8.
Groups and factors significantly associated with first locoregional recurrence in patients with invasive ductal carcinoma
| First locoregional recurrence | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|
| A | B: Patients who received adjuvant therapy (n = 873) | ||||||||||
| Total | All | All | UICC pN category | UICC pTNM stage | Adjuvant therapy status | Adjuvant radiotherapy status | |||||
| 11 | pN0 | pN1–3 | I and II | III | Endocrine therapy | Chemoendocrine therapy | Chemotherapy | None | Yes | ||
| Tumor embolus‐related group | |||||||||||
| 10 |
G3 G2 |
G3 G2 |
G3 G2 |
G3 |
G3 G2 |
G3 | G3 |
G3 G2 |
G3 | G3 | ● |
| Primary tumor‐stromal fibroblast‐related group | |||||||||||
| 9 | T2 | T2 | T2 | T2 | T2 | ● | T2 | ● | T2 | T2 | T2 |
| Primary invasive tumor cell‐related group | |||||||||||
| 3 | ● | ● | MF19 | ● | Tumor necrosis | ● | ● | ● | Skin invasion | ● | ● |
| Metastatic carcinomas to the lymph node‐related group | |||||||||||
| 2 | ● | ● | ● | ● | ● | ● | MF5 | ● | ● | ● | MF5 |
| Clinicopathological group | |||||||||||
| 0 | ● | ● | ● | ● | ● | ● | ● | ● | ● | ● | ● |
●, not significant; A, overall patients; G3, grade 3; G2, grade 2; T2, type 2 invasive ductal carcinoma; MF19, number of mitotic figure, >19; MF5, number of mitotic figures, >5.
Discussion
The results of the present study clearly exhibited an excellent power for the tumor embolus‐related group for the accurate prediction of first locoregional recurrence in patients with invasive ductal carcinoma since this group was significantly associated with the first locoregional recurrence independent of the tumor‐status categories except among patients who had received adjuvant radiotherapy (Table 8). Especially, the results of the present study clearly exhibited an excellent power for lymph vessel tumor embolus grade 3 for the accurate prediction of first locoregional recurrence in patients with invasive ductal carcinoma independent of the tumor statuses (Table 8). In contrast, a grade 1 lymph vessel tumor embolus was not a significant predictor for first locoregional recurrence and had a similar predictive power to grade 0 lymph vessel tumor embolus (Fig. 2a); more than half of the 376 patients with lymph vessel invasion were classified as having lymph vessel tumor embolus grade 1 (Table 3). These results suggest that the lymph vessel tumor embolus grade was capable of selecting not only patients with the worst prognosis, but also patients with a good prognosis among patients with lymph vessel invasion. Although many studies have already reported that the presence or absence of lymph vessel invasion or the number of invaded lymph vessels is an important factor for accurately predicting the locoregional recurrence of invasive ductal carcinoma,19, 20 we confirmed that the grading system for lymph vessel tumor emboli is superior to the presence or absence of lymph vessel invasion, the number of invaded lymph vessels or the location of lymph vessels invaded for accurately predicting first locoregional recurrence in this study. Thus, we can conclude that the lymph vessel tumor embolus grade is the only lymph vessel assessment parameter that can accurately divide patients with lymph vessel invasion into a good prognosis group and a poor prognosis group. However, the locoregional predictive power of the lymph vessel tumor embolus grade was inferior to type 2 invasive ductal carcinoma or >5 mitotic figures in metastatic carcinomas to the lymph node in patients who had received adjuvant radiotherapy; this finding strongly suggests that adjuvant radiotherapy prevents locoregional recurrence in patients with lymph vessel tumor embolus grades 3 or 2.21 Since the lymph vessel tumor embolus grade is assessed based on the numbers of mitotic figures and apoptotic figures in tumor cells in the lymph vessel,7 adjuvant radiotherapy probably inhibits the acceleration of the cell cycle in tumor cells in the lymph vessel. Thus, adjuvant radiotherapy may contribute to improving the outcome of patients with lymph vessel tumor embolus grade 3 or those with lymph vessel tumor embolus grade 2. From these, we can conclude that the lymph vessel tumor embolus grade in the tumor embolus‐related group was the best grade for accurately predicting first locoregional recurrence among patients with invasive ductal carcinoma of a low‐risk, intermediate‐risk or high‐risk class. In addition, the results of the study also exhibited no predictive power for number of apoptotic figures in blood vessel tumor emboli for the accurate prediction of first locoregional recurrence in patients with invasive ductal carcinoma.
The next most‐important group was the primary tumor‐stromal fibroblast‐related group, because this group accurately predicted first locoregional recurrence in nine of the 13 tumor statuses (Table 8). Especially, the results of the present study clearly exhibited a useful power for type 2 invasive ductal carcinoma for the accurate prediction of first locoregional recurrence in patients with invasive ductal carcinoma independent of the tumor statuses (Table 8). Type 2 invasive ductal carcinoma and type 4 invasive ductal carcinoma have atypical tumor‐stromal fibroblasts, and the former does not have a fibrotic focus within them but the latter has a fibrotic focus with atypical tumor‐stromal fibroblasts.7 Thus, the presence of atypical tumor‐stromal fibroblasts alone probably plays an important role in the establishment of first locoregional recurrence under the condition of the absence of fibrotic foci in invasive ductal carcinomas. We have previously reported that atypical tumor‐stromal fibroblasts exhibit a significantly higher frequency of p53 protein expression than ordinary tumor‐stromal fibroblasts;6, 22 this finding clearly indicates that the presence of atypical nuclear features is closely associated with p53 expression in tumor‐stromal fibroblasts. p53 mutations in tumor‐stromal fibroblasts are relatively common among primary breast cancers and have been reported to exert a positive effect on cancer growth.23, 24 p53 gene abnormalities or specific reactive changes in p53 immunoreactivity in tumor‐stromal fibroblasts produced by tumor cell‐stromal cell interactions inside and outside of the fibrotic foci probably lead to the expression of p53 in tumor‐stromal fibroblasts. Consequently, some tumor‐stromal fibroblasts expressing p53 inside and outside of fibrotic foci probably transform into atypical tumor‐stromal fibroblasts. Furthermore, since many atypical tumor‐stromal fibroblasts were also stained for smooth muscle actin,6 one can conclude that many of the atypical tumor‐stromal fibroblasts have the biological characteristics of myofibroblasts.25, 26 Thus, these atypical tumor‐stromal fibroblasts likely play important roles in the first locoregional recurrence of invasive ductal carcinomas of the breast.
In conclusion, the present study clearly demonstrated that the following factors that we have proposed play very important roles in the establishment of first locoregional recurrence: (i) lymph vessel tumor embolus grade; and (ii) atypical tumor‐stromal fibroblast outside a fibrotic focus, and also clearly demonstrated that the primary invasive tumor cell‐related group, the metastatic carcinomas to the lymph node‐related group, and the clinicopathological group were strikingly inferior to the above two factors for the prediction of first locoregional recurrence (Table 8). Thus, we can conclude that the above two factors are very useful surrogate markers for accurately predicting first locoregional recurrence of patients with invasive ductal carcinoma of the breast. Clinicians usually plan the follow‐up care of patients after the initial operation has been completed, deciding whether patients should be treated with adjuvant therapy and which type of adjuvant therapy should be performed based on pathological reports of the clinicopathological findings for the invasive ductal carcinomas. Thus, pathology reports of invasive ductal carcinomas that are based on the assessment of our proposed factors would probably provide clinicians with more important clues for the selection of patients with a high likelihood of locoregional recurrence among patients with invasive ductal carcinoma, compared with ordinary pathology reports of invasive ductal carcinomas, throughout the follow‐up period after the initial operation. Since it has recently been reported that the gene expression profile and protein expression profile of the tumor stroma play a very important role in tumor progression in carcinoma,27, 28 key proteins that are expressed in tumor cells with highly‐accelerating cell cycle in the lymph vessels, but also by atypical tumor‐stromal fibroblasts should be carefully investigated to develop targeted therapies that eradicate tumor cells with highly‐accelerating cell cycle or atypical tumor‐stromal fibroblast expressing key proteins, resulting in the improved outcome of patients with invasive ductal carcinoma of the breast.
Disclosure Statement
The authors have no conflict of interest.
Acknowledgments
This study was supported in part by a Grant‐in‐Aid for Scientific Research (KAKENHI) (C) (21590393, 24590439) from the Japan Society for the Promotion of Science.
(Cancer Sci, doi: 10.1111/cas.12217, 2013)
References
- 1. Tanis E, van de Velde CJ, Bartelink H et al Locoregional recurrence after breast‐conserving therapy remains an independent prognostic factor even after an event free interval of 10 years in early stage breast cancer. Eur J Cancer 2012; 48: 1751–6. [DOI] [PubMed] [Google Scholar]
- 2. Truong PT, Lesperance M, Culhaci A et al Patient subsets with T1–T2, node‐negative breast cancer at high locoregional recurrence risk after mastectomy. Int J Radiat Oncol Biol Phys 2005; 62: 175–82. [DOI] [PubMed] [Google Scholar]
- 3. Albert JM, Gonzalez‐Angulo AM, Guray M et al Estrogen/progesterone receptor negativity and HER2 positivity predict locoregional recurrence in patients with T1a,bN0 breast cancer. Int J Radiat Oncol Biol Phys 2010; 77: 1296–302. [DOI] [PubMed] [Google Scholar]
- 4. Hasebe T, Imoto S, Yokose T et al Histopathologic factors significantly associated with initial organ‐specific metastasis by invasive ductal carcinoma of the breast: a prospective study. Hum Pathol 2008; 39: 681–93. [DOI] [PubMed] [Google Scholar]
- 5. Hasebe T, Iwasaki M, Akashi‐Tanaka S et al Important histologic outcome predictors for patients with invasive ductal carcinoma of the breast. Am J Surg Pathol 2011; 35: 1484–97. [DOI] [PubMed] [Google Scholar]
- 6. Hasebe T, Iwasaki M, Akashi‐Tanaka S et al Atypical tumor‐stromal fibroblasts in invasive ductal carcinoma of the breast. Am J Surg Pathol 2011; 35: 325–36. [DOI] [PubMed] [Google Scholar]
- 7. Hasebe T, Okada N, Iwasaki M et al Grading system for lymph vessel tumor emboli: significant outcome predictor for invasive ductal carcinoma of the breast. Hum Pathol 2010; 41: 706–15. [DOI] [PubMed] [Google Scholar]
- 8. Hasebe T, Iwasaki M, Akashi‐Tanaka S et al Modified primary tumour/vessel tumour/nodal tumour classification for patients with invasive ductal carcinoma of the breast. Br J Cancer 2011; 105: 698–708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Hasebe T, Iwasaki M, Akashi‐Tanaka S et al Prognostic significance of mitotic figures in metastatic mammary ductal carcinoma to the lymph nodes. Hum Pathol 2011; 42: 1823–32. [DOI] [PubMed] [Google Scholar]
- 10. Sobin LH, Gospodarowicz MK, Wittekind CH, eds. International Union Against Cancer TNM Classification of Malignant Tumours, 7th edn Geneva: Wiley‐Liss, 2009; 181–93. [Google Scholar]
- 11. Hasebe T, Tsuda H, Hirohashi S et al Fibrotic focus in invasive ductal carcinoma: an indicator of high tumor aggressiveness. Jpn J Cancer Res 1996; 87: 385–94. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Hasebe T, Sasaki S, Imoto S et al Prognostic significance of fibrotic focus in invasive ductal carcinoma of the breast: a prospective observational study. Mod Pathol 2002; 15: 502–16. [DOI] [PubMed] [Google Scholar]
- 13. Hasebe T, Okada N, Tamura N et al p53 expression in tumor‐stromal fibroblasts is closely associated with the outcome of patients with invasive ductal carcinoma. Cancer Sci 2009; 100: 2101–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Wolff AC, Hammond ME, Schwartz JN et al American Society of Clinical Oncology/College of American Pathologists guideline recommendations for human epidermal growth factor receptor 2 testing in breast cancer. Arch Pathol Lab Med 2007; 131: 18–43. [DOI] [PubMed] [Google Scholar]
- 15. Elston CW, Ellis IO. Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long‐term follow‐up. Histopathology 1991; 19: 403–l0. [DOI] [PubMed] [Google Scholar]
- 16. Gilchrist KW, Gray R, Fowble B et al Tumor necrosis is a prognostic predictor for early recurrence and death in lymph node‐positive breast cancer: a 10‐year follow‐up study of 728 Eastern Cooperative Oncology Group patients. J Clin Oncol 1993; 11: 1929–35. [DOI] [PubMed] [Google Scholar]
- 17. Hasebe T, Sasaki S, Imoto S et al Characteristics of tumors in lymph vessels play an important role in the tumor progression of invasive ductal carcinoma of the breast: a prospective study. Mod Pathol 2001; 15: 904–13. [DOI] [PubMed] [Google Scholar]
- 18. Yamauchi C, Hasebe T, Iwasaki M et al Accurate assessment of lymph vessel tumor emboli in invasive ductal carcinoma of the breast according to tumor areas, and their prognostic significance. Hum Pathol 2007; 38: 247–59. [DOI] [PubMed] [Google Scholar]
- 19. Abdulkarim BS, Cuartero J, Hanson J et al Increased risk of locoregional recurrence for women with T1‐2N0 triple‐negative breast cancer treated with modified radical mastectomy without adjuvant radiation therapy compared with breast‐conserving therapy. J Clin Oncol 2011; 29: 2852–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Matsunuma R, Oguchi M, Fujikane T et al Influence of lymphatic invasion on locoregional recurrence following mastectomy: indication for postmastectomy radiotherapy for breast cancer patients with one to three positive nodes. Int J Radiat Oncol Biol Phys 2012; 83: 845–52. [DOI] [PubMed] [Google Scholar]
- 21. Yang PS, Chen CM, Liu MC et al Radiotherapy can decrease locoregional recurrence and increase survival in mastectomy patients with T1 to T2 breast cancer and one to three positive nodes with negative estrogen receptor and positive lymphovascular invasion status. Int J Radiat Oncol Biol Phys 2010; 77: 516–22. [DOI] [PubMed] [Google Scholar]
- 22. Hasebe T, Iwasaki M, Akashi‐Tanaka S et al Atypical tumor‐stromal fibroblasts in invasive ductal carcinoma of the breast treated with neoadjuvant therapy. Hum Pathol 2011; 42: 998–1006. [DOI] [PubMed] [Google Scholar]
- 23. Kurose K, Gilley S, Matsumoto PH et al Frequent somatic mutations in PTEN and TP53 are mutually exclusive in the stroma of breast carcinoma. Nat Genet 2002; 32: 355–7. [DOI] [PubMed] [Google Scholar]
- 24. Patocs A, Zhang L, Xu Y et al Breast‐cancer stromal cells with TP53 mutations and nodal metastases. N Engl J Med 2007; 357: 2543–51. [DOI] [PubMed] [Google Scholar]
- 25. Hu B, Gharaee‐Kermani M, Wu Z, Phan SH. Epigenetic regulation of myofibroblast differentiation by DNA methylation. Am J Pathol 2010; 177: 21–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Vered M, Dobriyan A, Dayan D et al Tumor‐host histopathologic variables, stromal myofibroblasts and risk score, are significantly associated with recurrent disease in tongue cancer. Cancer Sci 2009; 101: 274–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Fiank G, Bertos N, Pepin F et al Stromal gene expression predicts clinical outcome in breast cancer. Nat Med 2008; 14: 518–27. [DOI] [PubMed] [Google Scholar]
- 28. Singer CF, Gschwantler‐Kaulich D, Fink‐Retter A et al Differential gene expression profile in breast cancer‐derived stromal fibroblasts. Breast Cancer Res Treat 2008; 110: 273–81. [DOI] [PubMed] [Google Scholar]