Journal of Clinical Oncology, Vol 8, No 1 (January), 1990: pp

Transcription

1 The Presence of an Extensive Intraductal Component Following a Limited Excision Correlates With Prominent Residual Disease in the Remainder of the Breast By Roland Holland, James L. Connolly, Rebecca Gelman, Marcel Mravunac, Jan H.C.L. Hendriks, Andre L.M. Verbeek, Stuart J. Schnitt, Barbara Silver, John Boyages, and Jay R. Harris Previous studies of patients with infiltrating ductal breast cancer treated with conservative surgery (ie, limited excision) and radiotherapy have indicated that the presence of an extensive intraductal component (EIC) in the excision specimen is highly associated with subsequent breast recurrence. The reason for this association is not clear, but possible explanations include the presence of more extensive disease in the breast or increased radiation resistance among tumors with an EIC (EIC+) compared with those without (EIC-) tumors. To investigate this association further, we related the presence or absence of an EIC in the primary tumors of 214 women who underwent mastectomy to the likelihood of finding additional foci of cancer in their mastectomy specimens using a correlated pathologic-radiologic mapping technique. Primary tumors that were EIC+ were significantly more likely to have carcinoma in the remainder of the IN RECENT years, breast-conserving surgery and radiation therapy has become an important treatment option along with mastectomy for patients with stage I and II breast cancer. The results of long-term retrospective studies have demonstrated that this approach can provide a high level of local tumor control with satisfactory cosmetic results.'" 9 The results of more recent randomized prospective trials have demonstrated that the survival rates at 5 and 10 years achieved with breast-conserving surgery and radiation therapy are comparable to those seen with mastectomy. - 3 In previous studies from the Joint Center for Radiation Therapy (JCRT), most patients with infiltrating ductal carcinoma treated with limited excision and radiation therapy had a low rate of tumor recurrence in the breast.' 4 ' 15 (Limited excision in these studies typically consisted of a simple gross excision of the primary tumor without regard to the microscopic margins of resection.) The only subgroup of patients that had a high rate of local tumor recurrence consisted of those patients whose excision specimens contained an extensive intraductal component (EIC). breast than those which were EIC-(74% v 42%; P =.00001). This difference was primarily due to the presence of residual intraductal carcinoma. Seventyone percent of EIC+ patients had residual intraductal carcinoma compared with 28% of EIC-patients (P <.00001). In particular, 44% of EIC+ patients had "prominent" residual intraductal carcinoma compared with 3% of EIC-patients (P<.00001). We conclude that patients whose tumors contain an EIC more frequently have a large subclinical tumor burden in the remainder of the breast compared with patients whose tumors do not contain an EIC. This observation may explain the association between EIC and subsequent breast recurrence when patients are treated with a limited excision before radiotherapy. J Clin Oncol 1990 by American Society of Clinical Oncology. In a recent update of the JCRT experience on 584 stage I-II breast cancer patients with infiltrating duct tumors, the 5-year breast recurrence rate was 23% for the 143 patients whose tumors contained an EIC (EIC+) and only 5% for the 307 patients whose tumors were without (EIC-) From the Department of Pathology, University of Nijmegen, The Netherlands; Departments of Pathology, Beth Israel Hospital and Harvard Medical School, Boston, MA; Departments of Biostatistics and Epidemiology, Dana- Farber Cancer Institute and Harvard School of Public Health, Boston, MA; Department of Pathology, The Canisius Wilhelmina Hospital in Nijmegen, The Netherlands; Department of Radiology, University of Nijmegen, The Netherlands; Department of Social Medicine, University of Nijmegen, The Netherlands; and Joint Center for Radiation Therapy and Department of Radiation Therapy, Harvard Medical School, Boston, MA. Submitted April 10, 1989; accepted August 11, Supported by "Praeventiefonds," The Netherlands, Grant No Address reprint requests to Roland Holland, MD, Department of Pathology, St Radboud University Hospital of Nijmegen, Geert Grooteplein Zuid 24, PO Box 9101, 6500 HB, Nijmegen, The Netherlands. o 1990 by American Society of Clinical Oncology X/90/ $3.00/0 Journal of Clinical Oncology, Vol 8, No 1 (January), 1990: pp

2 114 (P =.0001).16 The association between an EIC + tumor and subsequent breast recurrence has been noted by some,179 but not all20,21 investigators. The results from the National Surgical Breast and Bowel Project (NSABP) do not support this association, although it is likely that this group used a different definition of EIC and, in addition, required tumor-free resection margins. 20 The reason for the association between an EIC+ tumor and breast recurrence following a limited excision and radiation therapy is not certain. One possible explanation is quantitative, ie, that EIC + tumors are associated with more extensive disease in the breast than are EIC -tumors and are less likely to be controlled by doses of radiotherapy consistent with preserving a good cosmetic result. Another possible explanation is qualitative, ie, that EIC + tumors are less sensitive to radiation killing than are EIC -tumors. The extent and distribution of additional tumor foci in the breast has been previously evaluated in a mastectomy series from Nijmegen, The Netherlands.22 Serially sectioned mastectomy specimens containing breast tumors 5 cm or less in diameter were reviewed. The relationship between the primary tumor and all additional tumor foci found in the breast was evaluated. This study demonstrated that in approximately 40% of cases, additional foci of cancer are present 2 or more cm from the primary tumor and, in most cases, this cancer is intraductal. In an attempt to investigate further the association between the presence of an EIC in a limited-excision specimen and subsequent tumor recurrence in the breast, we related the presence or absence of an EIC to the presence, location, extent, and type of cancer in the Nijmegen mastectomy specimens. The results indicate that tumors with an EIC are commonly associated with a large quantity of intraductal carcinoma extending for a considerable distance beyond the gross margins of the primary tumor, while tumors without an EIC are rarely associated with a large quantity of carcinoma beyond the primary tumor. These findings are consistent with the view that the association between an EIC+ tumor and subsequent recurrence in the breast following limited excision and radiotherapy is quantitative. MATERIALS AND METHODS HOLLAND ET AL The study population consisted of 214 women with 217 infiltrating ductal breast carcinomas measuring 5 cm or smaller on pathologic review. This group represents a subset of 282 consecutive patients treated with mastectomy for invasive breast carcinoma of all histologic types and 5 cm or less in size in the two hospitals of Nijmegen from 1980 to Sixty-eight patients having breast carcinomas with other than infiltrating ductal histology were not included in this study. In an earlier study, the presence, type, and location of additional foci of carcinoma in the mastectomy specimens were assessed in relation to the primary tumor using a combined radiologic-pathologic technique previously described by Egan and others. 22, 23 Chilled breast tissue specimens were sectioned at 5 mm intervals and a radiograph was taken of each slice. Tissue blocks for paraffin sections were obtained from all grossly and radiologically suspicious areas in addition to randomly selected areas from each quadrant and the nipple. The precise site of the blocks taken as well as the microscopically verified extension of each lesion was indicated on the radiographs, as were the histologic type and size of the tumor foci. The distance between the edge of the primary tumor (ie, the tumor that had directed the biopsy) and each of the other foci was measured on the radiographs. Lobular carcinoma in situ was noted but was not included in the results. For the present study, two additional examinations were performed (by RH and JLC). First, sections of the primary tumor and the immediate adjacent tissue were evaluated for the presence or absence of an EIC. These sections were selected to represent those that would have been obtained had a limited gross excision of the lesion been performed. Tumors were categorized as EIC+ when intraductal carcinoma was both (a) prominently present within the infiltrating tumor, and (b) clearly extending beyond the infiltrating margin of the tumor. Tumors were characterized as EIC-when only one or neither of the above features were present. In addition, tumors that were predominantly intraductal carcinomas but that showed foci of stromal invasion were also included in the EIC+ group. Second, all of the additional tumor foci (invasive, intralymphatic, and intraductal) previously identified 22 were reevaluated to assess the quantity of tumor in each focus. A total of 4,080 slides were reviewed with an average of 19 slides for each case. The intraductal carcinoma in each of these separate foci was quantified by determining the number of microscopic low power fields (LPFs) needed to encompass the lesion, using a 2.5 x objective. A LPF measured 6 mm in diameter. Prominent intraductal carcinoma was defined as the amount of intraductal carcinoma present in the upper 30% of patients with any additional foci of intraductal carcinoma. This corresponded to a total of six or more LPFs of intraductal carcinoma. The number of LPFs was compared with the Wilcoxon rank-sum test. Plots and tables of numbers of patients with additional tumor foci at various distances from the primary tumor are cumulative; that is, the figures represent numbers of patients who have any carcinoma at least the specified distance from the primary tumor. All P values are two-sided.

3 INTRADUCTAL COMPONENT AND RESIDUAL CANCER 115 Table 1. Probability of Finding Cancer Remaining in the Breast After Simulated Local Excision Related to the Distance From the Edge of the Primary Tumor Exclusive of LCIS Distance From Edge of Primary Tumor > 0.5 cm > 2cm > 4cm > 6cm > 8cm Any residual carcinoma* EIC + 74% 59% 32% 21% 9% EIC - 42% 29% 12% 8% 3% P = Invasive carcinoma EIC + 36% 20% 12% 2% 2% EIC - 19% 12% 7% 4% 1% P = Intralymphatic carcinoma EIC + 18% 11% 3% 2% 2% EIC - 11% 7% 4% 3% 1% P = Intraductal carcinoma EIC + 71% 58% 32% 21% 8% EIC - 28% 19% 5% 4% 1% P = < < < Abbreviation: LCIS, lobular carcinoma in situ. Since 20 significance levels are calculated in Table 1, an adjustment was made for the multiple comparison problem. Holm's sequentially rejective Bonferroni method 24 was used to ensure that overall type I error probability is less than RESULTS There were 66 EIC+ cases (30% of the total population) and 151 EIC-cases (70%). Table 1 shows the probability of finding any additional foci of carcinoma in the breast in the EIC + and EIC-cases expressed as a function of the distance from the edge of the primary tumor. Using Holm's sequentially rejective Bonferroni procedure, 24 all the Table 1 P values less than would be considered statistically significant. Patients whose primary tumors were EIC+ were more likely to have carcinoma of any type in the remainder of the breast than were patients with EIC- tumors (74% v 42%, P =.00001). Furthermore, the likelihood of finding additional carcinoma in the breast at any distance from the primary tumor was greater for EIC+ tumors than EIC-tumors (Table 1). The likelihood of finding each of the various forms of cancer in the breast at increasing distances from the primary tumor is also given in Table 1 for both the EIC+ and EIC-groups. EIC+ patients were more likely to have additional foci of invasive cancer than EIC- patients. However, this difference was most apparent within the immediate vicinity of the primary tumor. The likelihood of finding intralymphatic tumor foci in the remainder of the breast was similar in the EIC+ and EIC- groups. The major difference between the EIC+ and EICgroups related to the presence of foci of intraductal carcinoma. The likelihood of finding additional foci of intraductal carcinoma in the mastectomy specimen was considerably greater in the patients whose primary tumors were EIC + than in patients whose tumors were EIC-. Of note, this difference was apparent at all distances from the primary tumor. We also compared the quantity of carcinoma beyond the primary tumor in the mastectomy specimens for patients with EIC+ tumors and EIC- tumors. A large difference between the two groups was observed in the quantity of intraductal carcinoma. Forty-four percent of EIC+ cases had prominent intraductal carcinoma (defined as six or more LPFs of intraductal carcinoma) in the mastectomy specimens as compared with only 3% of patients in the EIC - group (P < 1 x 10-8). Figure 1 shows the percentage of patients with prominent intraductal carcinoma in relation to the distance from the edge of the primary tumor. Thirty-three percent of patients whose primary tumors were EIC+ had prominent intraductal carcinoma 2 or more cm from the edge of the primary tumor compared with 2% of patients whose tumors were EIC- (P = 1 x 10-8). Fourteen percent of patients whose tumors were EIC+ had prominent intraductal carcinoma 4 or more cm from the edge of the primary tumor compared with only 1% of patients whose tumors were EIC- (P = 5 x 10-8).

4 116 LJ k-k 0 LaJ 0 K DISTANCE FROM EDGE residual intraductal carcinoma on reexcision com- pared with only 2% of patients with EIC- tumors. The present study represents a larger group of unselected patients in which the precise distribution, type, and extent of additional tumor foci in the breast beyond the grossly palpable primary tumor have been identified. The findings presented here have implications Fig 1. The percentage of mastectaomy cases with prominent (a 6 LPF) intraductal carcinoma at or beyond certain distances (D) from the edge of the pri mary tumor (at 0.5 cm intervals). Wilcoxon test of amount of introductal carcinoma in EIC + cases v EIC- cases at or beyoind 2 cm from the edge of the tumor, P= 1 x O1-; at or beyo nd 4 cm, P= 5 x DISCUSSION OF TUMOR (D), cm HOLLAND ET AL "' '' ' ' ' '' Sne results presented in mis stuay are in nearly complete agreement with those of a previous study performed at the JCRT on the histologic findings in reexcision specimens among 71 patients with infiltrating duct carcinoma who initially underwent a limited excision of the tumor. 25 Residual carcinoma was seen in 62% of all patients, but was more frequent among the patients with EIC+ tumors than those with EIC- tumors (88% v 48%; P =.002). In particular, that study found that 44% of patients with EIC' -...,A :A M l..+:r r f anll 1- - LUIIor a a icons era eii quailltlly 0U The results of this collabora tive study of mastectomy specimens from patients with infiltrating regarding the optimal extent of breast surgery required before radiation therapy. The theoretit EIC+ primary cal approach to the use of breast-conserving duct carcinoma indicate tha tumors commonly have a large amount of cancer surgery and radiation therapy is to resect the (predominantly intraductal ccarcinoma) found beyond the edge of the primiary tumor, while EIC- breast cancers are rarely associated with a large amount of additional c; cancer foci beyond the primary tumor. These find ings are consistent with the hypothesis that EIC + tumors are associated with a higher rate of subsequent recurrence in the breast following limited excision and radiotherapy because they hayve a greater residtumor, leaving behind, in some cases, micro- scopic tumor foci, and then to irradiate the breast using a moderate radiation dose to eradicate these residual foci while still preserving the cosmetic appearance. Of note, a major determi- nant of the cosmetic outcome is the extent of breast surgery. Patients who undergo extensive breast surgery are more likely to have an unsatis- factory cosmetic outcome than are patients who ual tumor burden in the EIC - tumors. breast than do undergo a limited excision of the tumor. Therefore, it is preferable to limit the extent of breast Clinical-pathological studies from the JCRT have consistently shown a significant association between the presence of an ElC and subsequent recurrence in the breast amon.g patients treated with limited excision before radliotherapy. Confirmation of this finding requiress adequate followup on a group of patients tre:ated with limited excision where the pathologic material has been systematically reviewed and a similar definition of EIC applied. Such confirm; atory studies have surgery as much as possible consistent with obtaining local tumor control with radiation therapy. It should also be stressed that the dose of radiation therapy required to eradicate a given cancer is directly related to the tumor burden to be treated. Radiation doses in the range of 4,500 to 5,000 cgy given in 5 weeks to the entire breast, with a boost dose to the primary site such that the tumor area receives 6,000 cgy, provide good-to- excellent cosmetic results, but can eradicate recently been reported from Amsterdam, The relatively small amounts of residual breast can- Netherlands," and Mar- London, England,'" cer. Higher doses of radiation can be used to seilles, France.19 Studies findiing no association between EIC and breast recur*rence usually employed larger resection21 and/i or required tumorfree margins." treat a more extensive tumor burden, as in the treatment of locally advanced breast cancers, but this is achieved at the expense of the cosmetic appearance. One of the major findings of this

5 INTRADUCTAL COMPONENT AND RESIDUAL CANCER 117 study is that even after an excision of the clinically apparent tumor in the breast, the residual "subclinical" tumor burden can be considerable in some cases. The results presented here provide morphologic evidence from mastectomy specimens that infiltrating ductal carcinomas without an EIC rarely have a large tumor burden remaining in the breast after a limited excision. This finding is consistent with the observation from clinical studies that such tumors are wellmanaged with limited excision and radiation therapy. It should be emphasized that approximately three-fourths of all patients with infiltrating ductal carcinomas fall into this favorable category. It could be argued that identification of the cases with a large subclinical tumor burden following excision of a breast mass can be achieved by careful attention to the margins of resection, using ink on the surface of the specimen to facilitate recognition of margins on histologic sections. However, it is important to note that the evaluation of excision margins has many limitations. Resected breast specimens are usually soft, pliable, and highly irregular and it is often difficult to evaluate the margins of resection accurately. In addition, the duct system of the breast is highly complex and three-dimensional. As a result, for a given section, tumor extension away from the primary tumor may not be continuous. Furthermore, assessment of margin involvement is subject to sampling error as it is often not feasible to assess the entire margin completely. For all these reasons, "negative margins" do not fully ensure a small residual tumor burden and "positive margins" do not imply a large residual tumor burden. In the reexcision study from the JCRT, the presence or absence of an EIC in the limited-excision specimen was of greater value in predicting the extent of residual carcinoma in the vicinity of the primary tumor than was the presence of positive margins. 25 In current practice, we advocate both careful examination of inked excision margins and evaluation for the presence or absence of an EIC. For patients whose tumors are EIC- it is sufficient that the surgeon excise the tumor grossly and that the inked surfaces not show more than focal microscopic involvement. For patients whose tumors are EIC +, we recommend not only that the tumor be grossly excised, but also that the margins of resection be clearly negative. While interpretation of margins may be difficult in EIC + tumors, we attempt to demonstrate uninvolved breast parenchyma (not just fibroadipose tissue) between the intraductal carcinoma and the inked margin to consider the margin adequate. If the margins are positive and the patient prefers breast-conserving treatment, a reexcision of the primary site is generally performed. Should the reexcision specimen show extensive tumor, especially near the margins, we generally recommend mastectomy. However, the need for mastectomy in this situation has not been established and others have advocated the use of higher doses of radiotherapy as part of breast conserving treatment. Thus, the adequacy of the resection should be interpreted in the context of the presence or absence of an EIC, rather than in terms of one specific type of operation (ie, limited or wide excision for all patients). There is now considerable justification for the use of conservative surgery and radiotherapy in most patients with early breast cancer. In some institutions it has become the most commonly used method of treatment. This study indicates that the surgeon, in alliance with the pathologist, can help to minimize the extent of breast surgery performed before radiotherapy and ensure that both local control rates and cosmetic outcome will be satisfactory. REFERENCES 1. Bedwinek JM, Brody L, Perez CA, et al: Irradiation as the primary management of stage I and II adenocarcinoma of the breast: Analysis of the RTOG breast registry. Cancer Clin Trials 3:11-18, Chu A, Cope O, Russo R, et al: Treatment of early stage breast cancer by limited surgery and radical irradiation. Int J Radiat Oncol Biol Phys 6:25-30, Harris JR, Botnick LE, Bloomer WD, et al: Primary radiation therapy for early breast cancer: The experience at the Joint Center for Radiation Therapy. Int J Radiat Oncol Biol Phys 7: , Montague ED, Gutierrez AE, Barker JL, et al: Conservative surgery and irradiation for the treatment of favorable breast cancer. Cancer 43: , Pierquin B, Maylin C, Owen R, et al: Radical radiation therapy of breast cancer. Int J Radiat Oncol Biol Phys 6:17-24, Prosnitz LR, Goldenberg IS, Packard RA, et al: Radia-

6 118 tion therapy as initial treatment for early stage cancer of the breast without mastectomy. Cancer 39: , Amalric R, Santamaria F, Robert F, et al: Radiation therapy with or without primary limited surgery for operable breast cancers: A 20-year experience at the Marseilles Cancer Institute. Cancer 49:30-34, Calle R, Vilcoq JR, Pilleron JR, et al: Conservative treatment of operable breast carcinoma by irradiation with or without limited surgery: Ten-year results, in Harris JR, Hellman S, Silen W (eds): Conservative Management of Breast Cancer. Philadelphia, PA, Lippincott, 1983, pp Clark RM: Breast cancer: 20 years of conservative treatment, in Lewison EF, Montague ACW (eds): Diagnosis and Treatment of Breast Carcinoma. Baltimore, MD, Williams and Wilkins, 1981, pp Sarrazin D, Le M, Fontaine F, et al: Conservative treatment versus mastectomy in Tl or small T2 breast cancer: A randomized clinical trial, in Harris JR, Hellman S, Silen W (eds): Conservative Management of Breast Cancer. Philadelphia, PA, Lippincott, 1983, pp Veronesi U, DelVecchio M, Greco M, et al: Results of quadrantectomy, axillary dissection, and radiotherapy (QUART) in TINO patients, in Harris JR, Hellman S, Silen W (eds): Conservative Management of Breast Cancer. Philadelphia, PA, Lippincott, 1983, pp Sarrazin D, Le MG, Arriagada R, et al: Ten-year results of a randomized trial comparing a conservative treatment to mastectomy in early breast cancer. Radiother Oncol 14: , Fisher B, Redmond C, Poisson R, et al: Eight-year results of a randomized clinical trial comparing total mastectomy and lumpectomy with or without irradiation in the treatment of breast cancer. N Engl J Med 320: , Schnitt SJ, Connolly JL, Harris JR, et al: Pathologic predictors of early local recurrence in stage I and II breast cancer treated by primary radiation therapy. Cancer 53: , Harris JR, Connolly JL, Schnitt SJ, et al: Clinicalpathologic study of early breast cancer treated by primary radiation therapy. J Clin Oncol 1: , 1983 HOLLAND ET AL 16. Boyages J, Recht A, Connolly JL, et al: Predictors of breast recurrence for breast cancer patients treated with conservative surgery and radiation therapy. Radiother Oncol (in press) 17. Bartelink JH, Borger JH, van Dongen JA, et al: The impact of tumor size and histology on local control after breast-conserving therapy. Radiother Oncol 11: , Lindley R, Bulman A, Parsons P, et al: Histologic features predictive of an increased risk of early local recurrence after treatment of breast cancer by local tumor excision and radical radiotherapy. Surgery 105:13-20, Kurtz J: Extensive intraductal component (EIC) as a predictor of local recurrence in the breast. Presented at European Organization for the Research and Treatment of Cancer meeting, Amsterdam, The Netherlands, November, Fisher E, Sass R, Fisher B, et al: Pathologic findings from the National Surgical Adjuvant Breast Project (protocol 6) II: Relation of local breast recurrence to multicentricity. Cancer 57: , van Limbergen E, van den Bogaert W, van der Schueren E, et al: Tumor excision and radiotherapy as primary treatment of breast cancer: Analysis of patient and treatment parameters and local control. Radiother Oncol 8:1-9, Holland R, Veling SH, Mravunac M, et al: Histologic multifocality of Tis, T1-2 breast carcinomas: Implications for clinical trials of breast-conserving surgery. Cancer 56: , Egan RL: Multicentric breast carcinomas: Clinicalradiographic-pathologic whole organ studies and 10-year survival. Cancer 49: , Holm S: A simple sequentially rejective multiple test procedure. Scand J Stat 6:65-70, Schnitt SJ, Connolly JL, Khettry U, et al: Pathologic findings on re-excision of the primary site in breast cancer patients considered for treatment by primary radiation therapy. Cancer 59: , 1987