The University of Southern California/Van Nuys prognostic index for ductal carcinoma in situ of the breast

Transcription

1 The American Journal of Surgery 186 (2003) Scientific paper The University of Southern California/Van Nuys prognostic index for ductal carcinoma in situ of the breast Melvin J. Silverstein, M.D.* Keck School of Medicine, University of Southern California, Harold E. and Henrietta C. Lee Breast Center, USC/Norris Comprehensive Cancer Center, 1441 Eastlake Ave., Rm. 7415, Los Angeles, CA 90033, USA Manuscript received June 3, 2003; revised manuscript June 21, 2003 Presented at the Fourth Annual Meeting of the American Society of Breast Surgeons, Atlanta, Georgia, April 30 May 4, 2003 Abstract Background: The original Van Nuys prognostic index (VNPI) was introduced in 1996 as an aid to the complex treatment decision-making process for patients with ductal carcinoma in situ (DCIS) of the breast. This update adds patient age to the previous predictors of local recurrence in breast preservation patients. Methods: A prospective database consisting of 706 conservatively patients with DCIS was examined using multivariate analysis. Four independent predictors of local recurrence (tumor size, margin width, pathologic classification, and age) were used to derive a new formula for the University of Southern California (USC)/VNPI. Results: In all, 706 patients with pure DCIS were treated with breast preservation. There was no statistical difference in the 12-year local recurrence-free survival in patients with USC/VNPI scores of 4, 5, or 6, regardless of whether or not radiation therapy was used (P not significant). Patients with USC/VNPI scores of 7, 8, or 9 received a statistically significant average 12% to 15% local recurrence-free survival benefit when treated with radiation therapy (P 0.03). Patients with scores of 10, 11, or 12, although showing the greatest absolute benefit from radiation therapy, experienced local recurrence rates of almost 50% at 5 years. Conclusions: Ductal carcinoma in situ patients with USC/VNPI scores of 4, 5 or 6 can be considered for treatment with excision only. Patients with intermediate scores (7, 8, or 9) should be considered for treatment with radiation therapy or be reexcised if margin width is less than 10 mm and cosmetically feasible. Patients with USC/VNPI scores of 10, 11, or 12 exhibit extremely high local recurrence rates, regardless of irradiation, and should be considered for mastectomy, generally with immediate reconstruction or reexcision if technically possible Excerpta Medica, Inc. All rights reserved. Keywords: Ductal carcinoma in situ; Prognostic index; Noninvasive breast cancer; Van Nuys prognostic index; USC/Van Nuys prognostic index Ductal carcinoma in situ (DCIS) of the breast represents a broad biologic spectrum of disease with a wide range of treatment options. There is, however, a lack of clear and universally accepted treatment criteria, resulting in diverse and confusing clinical recommendations, distressing to both patients and physicians. The Van Nuys prognostic index (VNPI), as originally described in 1996 [1], is a tool that quantifies four measurable prognostic factors (tumor size, margin width, nuclear grade, and the presence or absence of comedonecrosis) that can be used in the treatment decision-making process. The VNPI is based on tumor morphology and recurrence data * Corresponding author. address: melsilver9@aol.com from a large series of DCIS patients and was developed as a numerical aid to be used in conjunction with clinical experience. The University of Southern California (USC)/ VNPI adds a fifth factor, patient age, that has been shown by numerous investigators to be of clinical importance in predicting local recurrence in conservatively treated patients with DCIS [2 4]. This paper will update our data through February 2003, using the USC/VNPI Patients and methods Through February 2003, 1,103 patients with pure DCIS were treated. No patients with invasive breast cancer are included. In all, 397 patients were treated with mastectomy and therefore did not have the ipsilateral breast at risk after /03/$ see front matter 2003 Excerpta Medica, Inc. All rights reserved. doi: /s (03)

2 338 M.J. Silverstein / The American Journal of Surgery 186 (2003) treatment and are not included in the analysis of local recurrence. The subjects of this paper are 706 patients treated with breast preservation (426 by excision alone and 280 by excision and radiation therapy). Treatment was not randomized. Patients with large lesions (4 cm and more), true multicentricity, or involved margins not amenable to reexcision were advised to undergo mastectomy (usually with immediate breast reconstruction). Patients with smaller lesions (4 cm or less) and microscopically clear surgical margins ( 1 mm) were generally treated with excision alone or excision plus radiation therapy. Some patients with larger lesions elected breast preservation, however, whereas other with lesions smaller than 4 cm elected mastectomy. Level 1 and 2 axillary dissections were done routinely until 1988; thereafter a lower axillary sampling was performed in some patents treated with mastectomy. Beginning in 1995, a sentinel lymph node biopsy was performed on patients who underwent mastectomy. Whole breast external beam irradiation (40 to 50 Gy) was performed on a4or6 MeV linear accelerator. Some patients received a boost of 10 to 20 Gy to the tumor bed by iridium-192 implant or linear accelerator. Disease-free survival rates for each group were estimated by the Kaplan-Meier method. The statistical significance between survival curves was determined by the log-rank test. The original Van Nuys prognostic index [1,5] was devised by combining these three statistically significant independent predictors of local tumor recurrence (tumor size, margin width and pathologic classification (determined by nuclear grade and the presence or absence of comedo-type necrosis). A score, ranging from 1 for lesions with the best prognosis to 3 for lesions with the worst prognosis, was given for each of the three prognostic predictors. The objective with all three predictors was to create three statistically different subgroups for each, using local recurrence as the marker of treatment failure. Cut-off points (for example, what size or margin width constitutes low, intermediate or high risk of local recurrence) were determined statistically, using the log rank test with an optimum P value approach. Size score: a score of 1 was given for a small tumors 15 mm or less, 2 was given for intermediate sized tumors 16 to 40 mm, and 3 was given for large tumors 41 mm or more in diameter. Margin score: a score of 1 was given for widely clear tumor-free margins of 10 mm or more. This was most commonly achieved by reexcision with the finding of no residual DCIS or only focal residual DCIS in the wall of the biopsy cavity. A score of 2 was given for intermediate margins of 1 to 9 mm and a score of 3 for margins less than 1 mm (involved or close margins). Pathologic classification score: the Van Nuys DCIS classification was used [6,7]. A score of 3 was given for tumors classified as group 3 (all high grade lesions), 2 for tumors classified as group 2 (non-high grade lesion with comedotype necrosis), and a score of 1 for tumors classified as group 1 (non-high grade lesion without comedo-type necrosis). The final formula for the original Van Nuys prognostic index became as follows: VNPI pathologic classification score margin score size score. The modified USC/Van Nuys prognostic index Early in 2001, multivariate analysis at the University of Southern California revealed that age was an independent prognostic factor in our database (Fig. 1) and that it should be added to the VNPI with a weight equal to that of the other factors. An analysis of our local recurrence data by age revealed that the most appropriate break points for our data were between ages 39 and 40 and between ages 60 and 61 (Fig. 2). Based on this, a score of 3 was given to all patients 39 years old or younger, a score of 2 was given to patients aged 40 to 60, and a score of 1 was given to patients aged 61 or older. The new scoring system for the USC/VNPI is shown in Table 1. The final formula for the USC/Van Nuys prognostic index became as follows: USC/VNPI pathologic classification score margin score size score age score. The patients least likely to recur after conservative therapy had a score of 4. The patients most likely to recur had a score of 12. The likelihood of recurrence increased as the USC/VNPI increased. Results Patients treated with mastectomy are not included in this analysis. 706 patients were treated with breast conservation, 426 by excision alone and 280 by excision plus radiation therapy. The patients were divided into three groups with differing probabilities for local recurrence as determined by USC/VNPI scores (4, 5, or 6 versus 7, 8 or 9 versus 10, 11, or 12). Table 2 shows the clinical parameters for each group. The average follow-up for all patients was 81 months. One hundred and nineteen patients experienced local failure; 49 of 280 (17.5%) treated with excision plus breast irradiation and 70 of 426 (16.4%) treated with excision alone. Of 119 local recurrences, 49 (43%) were invasive: 24 of 49 (49%) in patients treated with excision plus irradiation and 25 of 70 (36%) in patients treated with excision alone (P not significant). Seven patients treated with radiation therapy developed local recurrences and distant metastases, five of whom have died from breast cancer. Two patients treated with excision alone developed local invasive recurrence and metastatic disease. One has died of breast cancer. There is no statistical difference in breast cancer specific survival when patients treated with excision alone are compared with those treated with excision plus irradiation.

3 M.J. Silverstein / The American Journal of Surgery 186 (2003) Fig. 1. Cox multivariate analysis of factors affecting ductal carcinoma in situ recurrence-free survival (conservatively treated patients only). There is no statistical difference in breast cancer specific survival when patients are compared by USC/VNPI groupings. Sixty additional patients have died of other causes without evidence of recurrent breast cancer. The 12-year actuarial overall survival, including deaths from all causes, is 90% The local recurrence-free survival for all 706 patients is shown by tumor size in Fig. 3, by margin width in Fig. 4, by pathologic classification in Fig. 5, and by age in Fig. 2. The differences between every local disease-free survival curve for each of the four predictors that make up the USC/VNPI are statistically significant. Fig. 6 shows all patients by USC/VNPI score (4 to 12) while Fig. 7 groups patients with low (USC/VNPI 4, 5, or 6), intermediate (USC/VNPI 7, 8, or 9), or high (USC/ VNPI 10, 11, or 12) risks of local recurrence together. Each of these three groups is statistically different from one another. Patients with USC/VNPI scores of 4, 5 or 6 do not show a local disease-free survival benefit from breast irradiation (Fig. 8; P not significant). Patients with an intermediate rate of local recurrence, USC/VNPI 7, 8, or 9, are benefited by irradiation (Fig. 9). There is a statistically significant decrease in local recurrence rate, averaging about 12% to 15% throughout the curves, for irradiated patients with intermediate USC/VNPI scores compared with those treated by excision alone (P 0.02). Fig. 10 divides patients with a USC/VNPI of 10, 11, or 12 into those treated by excision plus irradiation and those treated by excision alone. Although, the difference between the two groups is highly significant (P 0.001), conservatively treated DCIS patients with a USC/VNPI of 10, 11, or 12 recur at an extremely high rate even with radiation therapy. Comments Fig. 2. Probability of local recurrence-free survival by age group for 706 breast conservation patients (all P 0.01). Our research [2,6 10] and the research of others [3,11 21], including the National Surgical Adjuvant Breast and Bowel Project (NSABP) [22], has shown that various combinations of nuclear grade, the presence of comedo-type necrosis, tumor size, margin width, and age are all important factors that can be used to predict local recurrence in conservatively treated patients with DCIS. Combinations of these factors can be used to select subgroups of patients whose recurrence rate is theoretically so high, even with breast irradiation, that mastectomy is preferable or to select

4 340 M.J. Silverstein / The American Journal of Surgery 186 (2003) Table 1 The USC/Van Nuys Prognostic Index scoring system. One to three points are awarded for each of four different predictors of local breast recurrence (size, margin width, pathologic classification, and age). Scores for each of the predictors are totaled to yield a VNPI score ranging from a low of 4 to a high of 12 Score Size (mm) Margin width (mm) Pathologic classification Nonhigh grade without necrosis (nuclear grades 1 or 2) Nonhigh grade with necrosis (nuclear grades 1 or 2) High grade with or without necrosis (nuclear grade 3) Age (yr) patients who do not require radiation therapy, in addition to complete excision, if breast conservation is selected. We analyzed 30 prognostic factors [9,23,24]. Only three, the Van Nuys classification [9] (which is made up by a combination of grade and necrosis), tumor size, and margin width were significant predictors of local recurrence and invasive local recurrence by multivariate analysis [5]. Ductal carcinoma in situ is a heterogeneous group of lesions and a uniform approach to treatment is not appropriate. Some patients require no treatment other than excision alone; others benefit from complete excision plus radiation therapy, and some will require mastectomy. The challenge is to use available clinical and pathologic data to select the most appropriate therapy for each individual patient. The USC/VNPI quantifies the evolving knowledge of prognostic factors in DCIS to define specific subsets of patients for whom treatment with excision alone, excision plus radiation, or mastectomy could be recommended. Although mastectomy is curative for approximately 99% of patients with DCIS [8,25 28], mastectomy represents significant overtreatment for the majority of cases detected by current methods. When breast conservation is elected rather than mastectomy, radiation therapy statistically decreases the likelihood of local recurrence when compared with excision alone [29 31]; but radiation therapy, like mastectomy, may also represent over-treatment for a significant number patients who elect breast preservation. The broad recommendation by the NSABP that radiation therapy is appropriate for all patients with DCIS who are treated with breast preservation, while clearly correct based on their data, does not take into account the heterogeneity of DCIS nor the differences in subsets demonstrated by our data [5,6,8,10] and that of others [12 21,32] including their own [22,33]. Radiation therapy is not without side effects [34]. It changes the texture of the breast, makes subsequent mammography more difficult to interpret, and its use precludes additional radiation therapy and breast conservation should a metachronous invasive breast cancer develop. Radiation therapy should only be offered to those patients with DCIS likely to obtain a substantial benefit. Subsets of patients who are not likely to receive any significant benefit from radiation therapy can be identified, e.g., those with USC/VNPI scores of 4, 5, or 6 in the series presented here, low grade lesions in the series of Lagios et al [12 14], small noncomedo lesions with uninvolved margins in the series of Schwartz et al [20] or the well-differentiated lesions of Zafrani et al [21]. Such patients may account for more than 30% of the total number of patients diagnosed with DCIS [8,10,14,15,19 21,35,36]. Patients in this series with USC/VNPI scores of 10, 11, or 12 present a different problem. While these patients show the greatest absolute benefit from postexcisional radiation therapy, their local recurrence rate continues to be extremely high and a recommendation for mastectomy should be considered. Treatment recommendations for the intermediate group (patients with scores of 7, 8, or 9) are the most difficult. For Table 2 Tumor characteristics, recurrences, and breast cancer deaths by USC/Van Nuys Prognostic Index Groups. Patients treated with mastectomy are not included in this table since they are at limited risk for local recurrence VNPI 4, 5 or 6 VNPI 7, 8, or 9 VNPI 10, 11, or 12 TOTAL No. breast conservation pts Average age (yr) Average size (mm) Average nuclear grade No. of recurrences 3 (1%) 78 (20%) 38 (50%) 119 (17%) No. invasive recurrences 0 (0%) 34 (44%) 15 (39%) 49 (41%) 5 & 10-yr local recurrence-free survival 99%/97% 84%/73% 51%/34% 85%/76% Breast cancer deaths & 10-Yr breast cancer specific survival 100%/100% 100%/98.1% 97.9%/979% 99.7%/98.6%

5 M.J. Silverstein / The American Journal of Surgery 186 (2003) Fig. 3. Probability of local recurrence-free survival by tumor size for 706 breast conservation patients (all P 0.01). patients with intermediate USC/VNPI scores and margin scores of 2 or 3, reexcision may lower their USC/VNPI score and improve local recurrence-free survival. If the score remains intermediate after reexcision, radiation therapy should be considered. However, some patients with scores of 9 may be better treated with mastectomy (eg, a 50-year-old patient with a large nuclear grade 2 lesion without necrosis with less than 1 mm margins after reexcision) while some patients with scores of 7 (eg, a 56-year-old patient with widely clear margins, small tumor size, but high nuclear grade) may elect no further treatment. These are independent judgments that must be made by the patient and her physician. Hopefully, the USC/VNPI will be a helpful adjunct as these difficult decisions are discussed. To date, no study of DCIS patients has shown a statistically significant difference in mortality when the three available treatments (mastectomy, excision alone, and excision plus radiation therapy) are compared. However, there are clear differences in local recurrence rates and they are of extreme importance. Local recurrences in patients who have Fig. 5. Probability of local recurrence-free survival for 706 breast conservation patients using Van Nuys ductal carcinoma in situ pathologic classification (all P 0.05). struggled to save their breasts are both demoralizing and theoretically, if invasive, a threat to life [17,37]. In this series (44%) and in most other reported series [12,17,20,38], approximately one half of all local recurrences are invasive. Treatment selection bias is not an important factor when using the USC/VNPI because the USC/VNPI does not compare different treatments. Rather, the USC/VNPI is based on measured parameters and compares patients who have achieved similar scores. Although the patient and her clinician control treatment selection, neither can influence final margin measurement, tumor size, pathologic classification, or age. The fact that some patients opted for suboptimal treatments that were not recommended (eg, 67 patients with USC/VNPI scores of 10, 11, or 12 who selected breast conservation therapy were all advised to undergo mastectomy) was actually helpful in developing and evaluating the USC/VNPI. Counseling patients with DCIS in a rational manner can be extremely difficult when the range of treatment options is extreme. The USC/VNPI allows a scientifically based discussion with the patient, using the parameters of the lesion Fig. 4. Probability of local recurrence-free survival by margin width for 706 breast conservation patients (all P 0.001). Fig. 6. Probability of local recurrence-free survival for 706 breast conservation patients by modified USC/Van Nuys prognostic index score 4 to 12.

6 342 M.J. Silverstein / The American Journal of Surgery 186 (2003) Fig. 7. Probability of local recurrence-free survival for 706 breast conservation patients grouped by modified USC/Van Nuys prognostic index score (4, 5, or 6 versus 7, 8, or 9 versus 10, 11, or 12; all p ). obtained after an initial excision. Thus, in some cases, a patient can choose a reexcision, in an effort to downscore her lesion. Successful downscoring of a patient with a USC/ VNPI of 10 or 11 could result in substantial reduction in the risk of local recurrence, perhaps changing a recommendation from mastectomy to radiation therapy. Similarly, patients with close or involved margins, with USC/VNPI scores of 7 or 8 after initial excision could opt for reexcision. Successful downscoring by achieving widely clear margins could result in a final USC/VNPI score sufficiently low to avoid breast irradiation. Downscoring can be achieved only by reexcising patients with margins scores of 2 or 3. Reexcision will not lower the pathologic classification score nor will it reduce the size of the tumor. In some cases, reexcision will upscore the tumor, increasing the USC/VNPI score by revealing a larger tumor size, a higher nuclear grade, the presence of previously undetected comedo necrosis, or an involved margin. The USC/VNPI may be useful to clinicians because it Fig. 9. Probability of local recurrence-free survival by treatment for 400 breast conservation patients with modified USC/Van Nuys prognostic index scores of 7, 8, or 9 (P 0.03). divides DCIS into three groups with statistically significant different risks for local recurrence after breast conservation therapy. Although there is an apparent treatment choice for each group (Table 3), excision only for patients with scores of 4, 5, or 6, excision plus radiation therapy for patients with scores of 7, 8, or 9, and mastectomy for patients with scores of 10, 11, or 12, the USC/VNPI is offered only as a guideline, a starting place in the discussions with patients. This work suggests that patients with DCIS can be stratified into specific subsets based on age, pathologic classification (using nuclear grade and necrosis), the size of the lesion, and the adequacy of surgical excision as determined by histologic margin assessment. The USC/VNPI is an attempt to quantify the known important prognostic factors in DCIS, making them clinically useful in the treatment decision-making process. The validity of the USC/VNPI must be independently and prospectively confirmed by other groups with access to large numbers of DCIS patients. In the future, other factors, Fig. 8. Probability of local recurrence-free survival by treatment for 230 breast conservation patients with modified USC/Van Nuys prognostic index scores of 4, 5, or 6; P not significant). Fig. 10. Probability of local recurrence-free survival by treatment for 76 breast conservation patients with modified USC/Van Nuys prognostic index scores of 10, 11, or 12 (P ).

7 M.J. Silverstein / The American Journal of Surgery 186 (2003) Table 3 Treatment guidelines by USC/Van Nuys Prognostic Index Score 4 to 6 Excision alone 7 to 9 Excision radiation 10 to 12 Mastectomy such as molecular markers may be integrated into the index when they are shown to be statistically important predictors of local recurrence. References [1] Silverstein MJ, Poller D, Craig P, et al. A prognostic index for ductal carcinoma in situ of the breast. Cancer 1996;77: [2] Sposto R, Epstein M, Silverstein MJ. Predicting local recurrence in patients with ductal carcinoma in situ of the breast. In: Silverstein MJ, Recht A, Lagios MD, editors. Ductal carcinoma in situ of the breast. Philadelphia: Lippincott, Williams and Wilkins, 2002, p [3] Vicini F, Kestin L, Goldstein N, et al. Impact of young age on outcome in patients with ductal carcinoma-in-situ treated with breastconserving therapy. J Clin Oncol 2000;18: [4] Goldstein N, Vicini F, Kestin L, et al. Differences in the pathologic features of ductal carcinoma in situ of the breast based on patient age. Cancer 2000;88: [5] Silverstein MJ, Lagios M, Craig P, et al. The Van Nuys prognostic index for ductal carcinoma in situ. Breast J 1996;2: [6] Silverstein MJ, Poller D, Waisman J, et al. Prognostic classification of breast ductal carcinoma-in-situ. Lancet 1995;345: [7] Poller D, Silverstein MJ. The Van Nuys ductal carcinoma in situ classification: an update. In: Silverstein MJ, Recht A, Lagios MD, editors. Ductal carcinoma in situ of the breast. Philadelphia: Lippincott, Williams and Wilkins, 2002, p [8] Silverstein MJ, Barth A, Poller D, et al. Ten-year results comparing mastectomy to excision and radiation therapy for ductal carcinoma in situ of the breast. Eur J Cancer 1995;31A: [9] Silverstein MJ, Barth A, Waisman J, et al. Predicting local recurrence in patients with intraductal breast carcinoma (DCIS). Proc Am Soc Clin Oncol 1995;14:117. [10] Silverstein MJ, Lagios M, Groshen S, et al. The influence of margin width on local control in patients with ductal carcinoma in situ (DCIS) of the breast. N Engl J Med 1999;340: [11] Goldstein NS, Kestin L, Vicini F. Intraductal carcinoma of the breast: pathologic features associated with local recurrence in patients treated with breast-conserving therapy. Am J Surg Pathol 2000;24: [12] Lagios M. Duct carcinoma in situ: pathology and treatment. Surg Clin North Am 1990;70: [13] Lagios M, Westdahl P, Margolin F, et al. Duct carcinoma in situ: relationship of extent of noninvasive disease to the frequency of occult invasion, multicentricity, lymph node metastases, and shortterm treatment failures. Cancer 1982;50: [14] Lagios M, Margolin F, Westdahl P, et al. Mammographically detected duct carcinoma in situ. Frequency of local recurrence following tylectomy and prognostic effect of nuclear grade on local recurrence. Cancer 1989;63: [15] Lagios M. Controversies in diagnosis, biology, and treatment. Breast J 1995;1: [16] Bellamy C, McDonald C, Salter D, et al. Noninvasive ductal carcinoma of the breast: the relevance of histologic categorization. Human Pathol 1993;24: [17] Solin L, Fourquet A, McCormick B, et al. Salvage treatment for local recurrence following breast-conserving surgery and definitive irradiation for ductal carcinoma in situ (intraductal carcinoma) of the breast. Int J Radiat Oncol Biol Phys 1994;30:3 9. [18] Poller D, Silverstein MJ, Galea M, et al. Ductal carcinoma in situ of the breast: a proposal for a new simplified histological classification association between cellular proliferation and c-erbb-2 protein expression. Mod Pathol 1994;7: [19] Ottesen G, Graversen H, Blichert-Toft M, et al. Ductal carcinoma in situ of the female breast. Short-term results of a prospective nationwide study. Am J Surg Pathol 1992;16: [20] Schwartz G. The role of excision and surveillance alone in subclinical DCIS of the breast. Oncology 1994;8:21 6. [21] Zafrani B, Leroyer A, Fourquet A, et al. Mammographically-detected ductal in situ carcinoma of the breast anaylyzed with a new classification. A study of 127 cases: correlation with estrogen and progesterone receptors, p53 and c-erbb-2 proteins, and proliferative activity. Semin Diagnost Pathol 1994;11: [22] Fisher E, Constantino J, Fisher B, et al. Pathologic findings from the National Surgical Adjuvant Breast Project (NSABP) protocol B-17. Cancer 1995;75: [23] Silverstein MJ. Predicting local recurrence in patients with ductal carcinoma in situ. In: Silverstein MJ, editor. Ductal carcinoma in situ of the breast. Baltimore: Williams and Wilkins, p [24] Silverstein MJ. Prognostic factors and local recurrence in patient with ductal carcinoma in situ of the breast. Breast J 1998;4: [25] Ashikari R, Hadju S, Robbins G. Intraductal carcinoma of the breast. Cancer 1971;28: [26] Bradley S, Weaver D, Bouwman D. Alternative in the surgical management of in situ breast cancer. Am Surg 1990;56: [27] Fentiman I, Fagg N, Millis R, et al. In situ ductal carcinoma of the breast: implications of disease pattern and treatment. Eur J Surg Oncol 1986;12: [28] Rosner D, Bedwani R, Vana J, et al. Noninvasive breast carcinoma. Results of a national survey of the American College of Surgeons. Ann Surg 1980;192: [29] Fisher B, Dignam J, Wolmark N, et al. Findings from National Surgical Adjuvant Breast and Bowel Project B-17. J Clin Oncol 1998;16: [30] Julien J, Bijker N, Fentiman I, et al. Radiotherapy in breast conserving treatment for ductal carcinoma in situ: first results of EORTC randomized phase III trial Lancet 2000;355: [31] Houghton J, George W. Radiotherapy and tamoxifen following complete excision of ductal carcinoma in situ of the breast. In: Silverstein MJ, Recht A, Lagios MD, editors. Ductal carcinoma in situ of the breast. Philadelphia: Lippincott, Williams and Wilkins, 2002, p [32] Solin L, Yeh I, Kurtz J, et al. Ductal carcinoma in situ (intraductal carcinoma) of the breast treated with breast-conserving surgery and definitive irradiation. Correlation of pathologic parameters with outcome of treatment. Cancer 1993;71: [33] Fisher E, Dignam J, Tan-Chiu E, et al. Pathologic findings from the National Surgical Adjuvant Breast Project (NSABP) eight-year update of protocol B-17: intraductal carcinoma. Cancer 1999;86: [34] Recht A. Side effects of radiation therapy. In: Silverstein MJ, editor. Ductal carcinoma in situ of the breast. Baltimore: Williams and Wilkins, 1997, p [35] Solin L, Recht A, Fourquet A, et al. Ten-year results of breastconserving surgery and definitive irradiation for intraductal carcinoma of the breast. Cancer 1991;68: [36] Schwartz G, Finkel G, Carcia J, et al. Subclinical ductal carcinoma in situ of the breast: treatment by local excision and surveillance alone. Cancer 1992;70: [37] Silverstein MJ, Lagios M, Martino S, et al. Outcome after local recurrence in patients with ductal carcinoma in situ of the breast. J Clin Oncol 1998;16: [38] Solin L, McCormick B, Recht A, et al. Mammographically detected, clinically occult ductal carcinoma in situ (intraductal carcinoma) treated with breast conserving surgery and definitive breast irradiation. Cancer J Sci Am 1996;2: