Prognostic Impact of Marginal Adipose Tissue Invasion in Ductal Carcinoma of the Breast

Transcription

1 natomic Pathology / dipose Tissue Invasion in reast Cancer Prognostic Impact of Marginal dipose Tissue Invasion in Ductal Carcinoma of the reast Junzo Yamaguchi, MD, 1 Hiroshi Ohtani, MD, 2 Kazukuni Nakamura, MD, 1 Isao Shimokawa, MD, 2 and Takashi Kanematsu, MD 3 Key Words: reast cancer; dipose tissue invasion; Lymph node status; Prognosis; Lymphatic vessel invasion DOI: /MX6KK1UNJ1YG8VN bstract This study aimed to investigate whether adipose tissue invasion (TI) of cancer cells at the tumor margin influenced lymph node status and prognosis in patients with invasive ductal carcinoma of the breast. Data for 245 patients with breast cancer with marginal TI were clinicopathologically compared with data for 65 patients without TI. We also examined the value of the combination of TI and peritumoral lymphatic vessel invasion (LVI). The frequency of axillary lymph node metastases was 40.7% in patients with TI (99/243) and 11.3% in patients without TI (7/62; P <.0001), and TI was an independent factor influencing nodal metastasis. Patients with TI had a poorer prognosis than patients without TI (10-year diseasefree survival, 76% and 94%, respectively; P =.0323). In addition, patients without TI or LVI had neither lymph node metastasis (n = 52) nor recurrent disease (n = 53). TI is one of the biologic indicators of tumor aggressiveness. The spread of cancer cells via the lymphatic system is the first step in the dissemination of breast cancer, 1 but little is known as to how cancer cells enter the lymphatic vessels. 2 It has been speculated that lymph node metastasis is promoted by tumor cell aggressiveness, ie, invasiveness, motility, and proliferation. Tumor-stromal interactions seem to be a fundamental aspect of tumor aggressiveness. 3-5 It has also been suggested that functional lymphatics at the tumor margins are responsible for lymphatic metastasis. 6 Therefore, we propose that the ability to invade neighboring tissues is more significant than the ability to proliferate with regard to the tumorigenic-metastatic potential of breast cancer cells. 7 In contrast, these tumor characteristics indicate the need to identify reliable markers to predict patients at low risk who do not require adjuvant chemotherapy or endocrinotherapy. The International Union gainst Cancer classification has shown the significance of chest wall and skin invasions in evaluating local tumor extension in breast cancer. Some investigators noted that pathologic evidence of fat invasion, such as scattered invasion into fat tissues 5 and the invasive length of fat invasion, 8,9 was related to a poor prognosis. However, the prognostic significance of adipose tissue invasion (TI) at the tumor margin has not been fully evaluated in breast cancer, and the biologic characteristics of tumors with TI are also insufficiently known. Meanwhile, many studies have been conducted to define the significance of lymphovascular invasion. Peritumoral lymphatic vessel invasion (LVI) is associated with a higher frequency of nodal metastasis and has also been cited as a risk factor in breast carcinoma Therefore, we hypothesize that the combination of the values of marginal TI and peritumoral LVI may be an important tool for risk allocation in patients with breast cancer because 382 m J Clin Pathol 2008;130: DOI: / MX6KK1UNJ1YG8VN

2 natomic Pathology / Original rticle the invasive potential of cancer cells may be critical for the early dissemination of breast cancer. We looked at marginal TI in invasive ductal carcinoma of the breast and investigated the relationship between TI and other clinicopathologic findings to clarify the biologic behavior of breast cancer with TI. Finally, we evaluated the prognostic value of the combination of marginal TI and peritumoral LVI to confirm our hypothesis. Materials and Methods In the present study, we defined TI with strict emphasis on reproducibility and simplicity and restricted the cases to early stage breast carcinoma to assess the biologic characteristics of marginal TI in invasive ductal carcinoma. Patients retrospective analysis was conducted using the data for 310 patients with invasive ductal carcinoma of the breast who had undergone surgery (total or partial mastectomy) from 1993 to The study materials represented 58.6% of all new breast carcinomas diagnosed in the same period (310/529). Except for 5 cases, the patients had all undergone axillary lymph node dissection or a sampling procedure. Data for patients with a special histologic type such as mucinous, lobular, medullary, or squamous cell carcinoma were excluded from the study. We also excluded data for patients with bilateral breast cancer (synchronous, metachronous), clinically multifocal or multicentric tumors in the unilateral breast, skin or striated muscle invasion, inflammatory carcinoma, distant metastasis, or malignancy at another site and for patients who had received preoperative neoadjuvant chemotherapy. The patient age ranged from 27 to 87 years (median, 52 years). ody mass index (MI) was calculated (kg/m2) in all patients. In the study, 300 (96.8%) of 310 were followed up until October 31, 2006; the median follow-up period was 63 months (range, months). Tissue Preparation and Tumor Size The resected breast and lymph node tissues were fixed in 10% neutral buffered formalin, and the breast tissues were cut into 5-mm slices. Each paraffin-embedded block was cut into 4-µm sections and stained with H&E. ll but a few specimens from surgery, excision biopsy, and mammotome (vacuum-assisted) biopsy were available for histologic examination. Tumor size was measured as the largest dimension of the microscopic invasive component on the pathologic specimen (range, cm, pt1-2). Lymph node involvement was pathologically assessed in 305 patients who underwent axillary dissection. The numbers of patients with disease at each ptnm0 stage were as follows: pt1n0, 171; pt2n0, 28; pt1n+, 67; pt2n+, 38; and unknown, 6. Definitions of dipose Tissue and Marginal dipose Tissue Invasions In the present study, adipose tissue was defined as a pure aggregate consisting of more than 20 fat cells without intervening fibrous tissues in the breast zimage 1z. The adipose tissue included tissues surrounding the mammary ducts or lobules and those in the subcutaneous layers. Fibroadipose tissue (fat cells mixed with various fibrous tissues) was strictly distinguished from adipose tissues (Image 1). Marginal TI was defined as the presence of more C TDLU FT T zimage 1z, Normal histologic features of the breast. The terminal duct-lobular unit (TDLU) is surrounded by fibrous and fibroadipose tissue (FT), followed by adipose tissue (T) (H&E, 40)., dipose tissue invasion (TI) specimen. Ductal carcinoma cells infiltrate the fibroadipose tissues but not the adipose tissues (H&E, 200). C, TI+ specimen. More than 20 cancer cells are present in the adipose tissue (H&E, 200). 383 m J Clin Pathol 2008;130: DOI: /MX6KK1UNJ1YG8VN

3 Yamaguchi et al / dipose Tissue Invasion in reast Cancer than 20 cancer cells in direct contact with the adipose tissue or the location of cancer cells in the adipose tissue zimage 1z and zimage 1Cz. Only cases with unequivocal TI were considered positive (TI+). Doubtful cases were considered negative (TI ). Peritumoral LVI LVI was defined as the presence of peritumoral lymphatic emboli in endothelium-lined spaces, according to the literature zimage 2z.16 Immunohistochemical analysis using lymphatic endothelial markers such as D2-40 was not performed. Samples with obvious emboli were considered positive (LVI+), and doubtful samples were considered negative (LVI ). involvement according to multivariate analysis (P =.0079) ztable 2z. ecause patient age and pathologic tumor size were closely related to TI, these 2 parameters were used for multivariate analysis. ge (P =.0012) and tumor size (P <.0001) were independently significant factors influencing marginal TI (data not shown). Histologic Grade and Hormone Receptor Status ll tumors were graded according to a modified version of the loom-richardson histologic grading criteria.17 The status of estrogen and progesterone receptors was evaluated by immunohistochemical analysis using commercially available antibodies (DakoCytomation, Kyoto, Japan). Cases with more than 10% of the tumor cell nuclei stained were defined as positive. Statistics The χ2 analysis and Student t test were used to analyze the differences between study groups. multivariate analysis was conducted to investigate the factors influencing nodal involvement or marginal TI. Survival curves for cancer-specific survival were constructed using the method of Kaplan and Meier, and the differences between the 2 groups were assessed by using the reslow-gehan-wilcoxon test. Overall survival was calculated as the period from surgery to death of breast cancer. Probability (P) values were calculated using StatView 5 software (SS Institute, Cary, NC). P values less than.05 were considered statistically significant. Information on tumor size, lymph node status, hormone receptor status, or survival was unavailable for some patients. Results Relationship etween TI and Clinicopathologic Parameters Of 310 cases, 245 (79.0%) showed marginal TI. s shown in ztable 1z, TI+ cases had the highest frequency of axillary lymph node metastases (99/243 [40.7%]) compared with the TI cases (7/62 [11.3%]) (P <.0001). Patient age (P =.0009) and pathologic tumor size (P <.0001) were significantly related to TI. No statistical differences in MI, LVI, histologic grade, or hormone receptor status (estrogen receptor and progesterone receptor) were observed. TI was an independently significant factor influencing nodal m J Clin Pathol 2008;130: DOI: /MX6KK1UNJ1YG8VN zimage 2z, Lymphatic vessel invasion (LVI)+ specimen. tumor embolus (arrow) is present at the peritumoral area (H&E, 100)., High-power view of peritumoral LVI. The dilated lymphatic channel adjacent to the blood vessel has an endothelial lining (H&E, 400). ztable 1z Relationship etween TI and Clinicopathologic Parameters in 310 Cases of Ductal Carcinoma of the reast Parameters Lymph node metastasis Present bsent Unknown Mean ± SD age (y) Mean ± SD body mass index (kg/m2) Tumor size (cm) Lymphatic vessel invasion Positive Negative Histologic grade I II III Estrogen receptor Positive Negative Progesterone receptor Positive Negative TI+ (n = 245) TI (n = 65) P ± ± ± ± 3.9 <.0001* 1.7 ± 1.0 ± < * *.0689*.1858* TI, adipose tissue invasion. * χ2 analysis. Student t test.

4 natomic Pathology / Original rticle Prognostic Value of TI Data for 237 TI+ cases and 64 TI cases were available for the survival analysis. The 5- and 10-year disease-free survivals were 83% and 76%, respectively, in TI+ cases, and both values were 94% in TI cases zfigure 1z (P =.0323). The 10-year overall survivals were 81% and 98% in TI+ and TI cases, respectively zfigure 1z (P =.0577). ztable 2z Multivariate nalysis for Nodal Involvement in 304 Cases of Ductal Carcinoma of the reast Factor Lymphatic vessel invasion <.0001 Tumor size.0004 dipose tissue invasion.0079 P Prognostic Value of LVI Data for 80 LVI+ cases and 220 LVI cases were available for the survival analysis. The 5- and 10-year disease-free survivals were 69% and 60%, respectively, in LVI+ cases and were 92% and 90%, respectively, in LVI cases zfigure 2z (P <.0001). The 10-year overall survivals were 66% and 91% in LVI+ and LVI cases, respectively zfigure 2z (P =.0008). Significance of the Combination of TI and LVI In the next step, we investigated the clinical relevance of marginal TI when combined with peritumoral LVI. Lymph node metastasis was found in 75% of TI+/LVI+ cases (53/71) and 70% of TI /LVI+ cases (7/10). In TI+/LVI cases, the frequency was 26.7% (46/172). However, no nodal metastasis was observed in TI /LVI cases (0/52) TI TI TI TI+ zfigure 1z Disease-free () and overall () survival for 237 adipose tissue invasion (TI)+ and 64 TI cases of ductal carcinoma of the breast. The 5- and 10-year survivals (%) are shown., P =.0323;, P =.0577; Wilcoxon test for both LVI LVI LVI LVI+ zfigure 2z Disease-free () and overall () survival for 80 lymphatic vessel invasion (LVI)+ and 220 LVI cases of ductal carcinoma of the breast. The 5- and 10-year survivals (%) are shown., P <.0001;, P =.0008; Wilcoxon test for both. m J Clin Pathol 2008;130: DOI: /MX6KK1UNJ1YG8VN 385

5 Yamaguchi et al / d i p o s e Tissue In v a s i o n in reast Cancer Moreover, as shown in zfigure 3z the 5- and 10-year disease-free survivals were 69% and 59%, respectively, in TI+/LVI+ cases (n = 70); 89% and 85%, respectively, in TI+/LVI cases (n = 167); and 100% in TI /LVI cases (n = 53). Data for the TI /LVI+ cases (n = 11) were excluded from Figure 3. On the other hand, no difference was observed in disease-free survival between the LVI+ (Figure 2) and TI+/LVI+ cases (Figure 3) TI /LVI TI+/LVI TI+/LVI Discussion The present study revealed 3 novel findings. First, TI of cancer cells at the tumor margin was independently associated with nodal involvement in patients with invasive ductal carcinoma of the breast. Second, patient age and invasive tumor size were significant factors for TI. Finally, patients without TI or LVI had an excellent prognosis. The breast stroma histologically consists of fibrous tissues and adipocytes in variable proportions, and the adipocytes physiologically increase with aging in older women. Generally, when intraductal carcinoma cells infiltrate the breast stroma, the cells initially penetrate the fibrous tissues, followed by the fibroadipose tissues, and, finally, the adipose tissues in breast cancer (Image 1). We focused on the TI of cancer cells and established strict criteria for adipose tissue and TI in histologic examinations. Marginal TI of cancer cells correlated with lymph node metastasis. lthough the relationship between TI and LVI was statistically insignificant (P =.0501; Table 1), marginal TI may lead to a larger contact area between cancer cells and the peritumoral functional lymphatic endothelium. The presence of TI may reflect the infiltrating growth pattern of cancer cells at the marginal site, while the absence of TI may be associated with an expanding growth pattern or the existence of abundant fibrous tissues surrounding the ducts and lobules as a result of other underlying conditions such as fibrocystic changes and fibrous mastopathy. Furthermore, stromal reaction patterns, including edema and desmoplasia, in invasive cancer may affect TI or prognosis in patients with breast cancer. 18 In most patients with breast cancer, only fibrous or fibroadipose tissue invasion by cancer cells appears insufficient for nodal metastasis. The true mechanisms of TI-associated nodal metastasis are unknown, but distinct molecular mechanisms are likely to be concerned with TI- and peritumoral LVI-associated nodal metastasis. romatase activity and adipocytokines such as leptin in mammary adipose tissues may participate in TI-associated nodal involvement. Several investigators have reported that aromatase activity and its expression in breast adipose tissues are most prominent in regions proximal to the tumor in breast cancer zfigure 3z Disease-free survival for 70 adipose tissue invasion (TI)+ and lymphatic vessel invasion (LVI)+, 167 TI+/LVI, and 53 TI /LVI cases of ductal carcinoma of the breast. The 5- and 10-year tumor-free survivals (%) are shown. There is a possibility that tissue estrogen concentration is higher in breast carcinoma with TI. This may cause the aggressive biologic behavior of breast cancer with TI and the subsequent involvement of the lymphatic network. 22 In addition, mammary adipose tissue is an important source of various adipocytokines, including leptin. Leptin is one of the neurohormone regulators in the hypothalamus and is necessary to normal mammary gland development and lactation. It might be also involved in carcinogenesis and progression in breast cancers. 23 Such paracrine adipocytokines may contribute to tumor aggressiveness and TI-associated nodal metastasis. lkarain et al 24 recently reported that the lymphatic endothelial marker D2-40 is useful for detecting intratumoral obliterated LVI that cannot be visualized on H&Estained sections. These intratumoral LVIs in adipose tissue may be associated with nodal metastasis. Taken together, further investigations on the molecular mechanisms of vessel invasion and genetic participation in the invasive process are essential for anticancer therapy, including new molecular target therapy. The size of the invasive component, a significant factor for TI in the present study, has been shown to be closely related to nodal involvement. 25 In the current setting, not only TI but also tumor size (P =.0004) were shown to be independently significant factors influencing nodal metastasis (Table 2). It is natural to assume that enlarged tumors can extend into adipose tissues and acquire greater opportunities for cancer cell intravasation at the tumor margin. Moreover, in the present study, patients with TI+ disease were significantly older than patients with TI disease. 386 m J Clin Pathol 2008;130: DOI: /MX6KK1UNJ1YG8VN

6 natomic Pathology / Original rticle ecause the MI showed no significant association with TI, TI may be affected by an age-related increase in fat cells (fatty change) in the breast stroma rather than the quantity of body adipose tissue. However, age was not a prognostic factor in this study. Patients with LVI+ disease were significantly younger than patients with LVI disease (P <.0001), being in sharp contrast to the relationship between TI and patient age. We think that breast tissue is less fatty and cancer cells are more aggressive in younger patients, whereas breast tissue is more fatty and cancer cells are less aggressive in older patients. Peritumoral lymphovascular invasion is thought to be a useful factor for risk allocation, 12,15,16 and at the St Gallen meeting in 2005, lymphovascular invasion was newly added to the list of adverse prognostic factors in patients with nodenegative disease. 14 Schoppmann et al 12 have shown that the 10-year disease-free survival rates were 45% and 70% in patients with (n = 105) and without (n = 269) lymphovascular invasion, respectively. They also indicated the predictive value of lymphovascular invasion in the development of lymph node metastasis. These observations are more or less consistent with our findings. In the present study, no prognostic significance of TI could be found in LVI+ cases, but it was particularly interesting that patients without TI or LVI had neither nodal involvement nor recurrent disease. These patients are expected to be free of distant metastasis. The combination of these 2 microscopic features may be a useful predictive factor in identifying patients who require no axillary dissection, chemotherapy, or hormonal treatment. From this series of investigations, TI of cancer cells at the tumor margin is one of the biologic indicators of tumor aggressiveness in early stage breast cancer. We came to the conclusion that TI+ cases were associated with adverse outcomes in addition to the converse, that TI cases were associated with favorable outcomes. We think that TI should be incorporated in a standard surgical pathology report of infiltrating ductal carcinoma of the breast. The precise evaluation of marginal TI and peritumoral LVI will prove to be useful in the formulation of therapeutic strategies and the prediction of which patients with breast cancer have an excellent prognosis. From the 1 Department of Surgery, National Hospital Organization, Saga Hospital, and 2 Department of Investigative Pathology and 3 Surgery II, Nagasaki University Graduate School of iomedical Science, Nagasaki, Japan. ddress reprint requests to Dr Yamaguchi: Dept of Surgery, National Hospital Organization, Saga Hospital, Hinode, Saga , Japan. References 1. Mattila MM, Ruohola JK, Karpanen T, et al. VEGF-C induced lymphangiogenesis is associated with lymph node metastasis in orthotopic MCF-7 tumors. Int J Cancer. 2002;98: Ohtani O, Shao XJ, Saitoh M, et al. Lymphatics of the rat mammary gland during virgin, pregnant, lactating and postweaning periods. J Steroid iochem Mol iol. 1998;103: cs G, Dumoff KL, Solin LJ, et al. Extensive retraction artifact correlates with lymphatic invasion and nodal metastasis and predicts poor outcome in early stage breast carcinoma. m J Surg Pathol. 2007;31: garwal, Saxena R, Morimiya, et al. Lymphangiogenesis does not occur in breast cancer. m J Surg Pathol. 2005;29: Kimijima I, Ohtake T, Sagara H, et al. Scattered fat invasion: an indicator for poor prognosis in premenopausal, and for positive estrogen receptor in postmenopausal breast cancer patients. Oncology. 2000;59: Padera TP, Kadambi, di Tomaso E, et al. Lymphatic metastasis in the absence of functional intratumor lymphatics. Science. 2002;296: Firon M, Shaharabany M, ltstock RT, et al. Dominant negative Met reduces tumorigenicity-metastasis and increases tubule formation in mammary cells. Oncogene. 2000;19: Hasebe T, Imoto S, Sasaki S, et al. proposal for a new histological classification scheme for predicting short-term tumor recurrence and death in patients with invasive ductal carcinoma of the breast. Jpn J Cancer Res. 1998;89: Hasebe T, Mukai K, Tsuda H, et al. New prognostic histological parameter of invasive ductal carcinoma of the breast: clinicopathological significance of fibrotic focus. Pathol Int. 2000;50: Gajdos C, Tartter PI, leiweiss IJ. Lymphatic invasion, tumor size, and age independent predictors of axillary lymph node metastases in women with T1 breast cancers. nn Surg. 1999;230: Nathanson SD, Zarbo RJ, Wachna DL, et al. Microvessels that predict axillary lymph node metastases in patients with breast cancer. rch Surg. 2000;135: Schoppmann SF, ayer G, umayr K, et al. Prognostic value of lymphangiogenesis and lymphovascular invasion in invasive breast cancer. nn Surg. 2004;240: Fitzgibbons PL, Page DL, Weaver D, et al. Prognostic factors in breast cancer: College of merican Pathologists consensus statement rch Pathol Lab Med. 2000;124: Goldhirsch, Glick JH, Gelber RD, et al. Meeting highlights: international expert consensus on the primary therapy of early breast cancer nn Oncol. 2005;16: Woo CS, Silberman H, Nakamura SK, et al. Lymph node status combined with lymphovascular invasion creates a more powerful tool for predicting outcome in patients with invasive breast cancer. m J Surg. 2002;184: Mascarel I, onichon F, Durand M, et al. Obvious peritumoral emboli: an elusive prognostic factor reappraised: multivariate analysis of 1320 node-negative breast cancers. Eur J Cancer. 1998;34: Robbins P, Pinder S, Klerk N, et al. Histological grading of breast carcinomas: a study of interobserver agreement. Hum Pathol. 1995;26: Samaratunga H, Fairweather P, Purdie D. Significance of stromal reaction patterns in invasive urothelial carcinoma. m J Clin Pathol. 2005;123: ulun SE, Price TM, itken J, et al. link between breast cancer and local estrogen biosynthesis suggested by quantification of breast adipose tissue aromatase cytochrome P450 transcripts using competitive polymerase chain reaction after reverse transcription. J Clin Endocrinol Metab. 1993;77: m J Clin Pathol 2008;130: DOI: /MX6KK1UNJ1YG8VN 387

7 Yamaguchi et al / d i p o s e Tissue In v a s i o n in reast Cancer 20. O Neill JS, Elton R, Miller WR. romatase activity in adipose tissue from breast quadrants: a link with tumor site. r Med J. 1988;296: Sasano H, Ozaki M. romatase expression and its localization in human breast cancer. J Steroid iochem Mol iol. 1997;61: Sasano H, Harada N. Intratumoral aromatase in human breast, endometrial, and ovarian malignancies. Endocr Rev. 1998;19: Garofalo C, Koda M, Cascio S, et al. Increased expression of leptin and leptin receptor as a marker of breast cancer progression: possible role of obesity-related stimuli. Clin Cancer Res. 2006;12: lkarain, Kahn HJ, Narod S, et al. Significance of lymph vessel invasion identified by the endothelial lymphatic marker D2-40 in node negative breast cancer. Mod Pathol. 2007;20: Seidman JD, Schnaper L, isner SC. Relationship of the size of the invasive component of the primary breast carcinoma to axillary lymph node metastasis. Cancer. 1995;75: m J Clin Pathol 2008;130: DOI: /MX6KK1UNJ1YG8VN