Abstract
Purpose
The increased breast cancer risk conferred by a diagnosis of lobular carcinoma in situ (LCIS) is poorly understood. Here, we review our 29-year longitudinal experience with LCIS to evaluate factors associated with breast cancer risk.
Patients and Methods
Patients participating in surveillance after an LCIS diagnosis are observed in a prospectively maintained database. Comparisons were made among women choosing surveillance, with or without chemoprevention, and those undergoing bilateral prophylactic mastectomies between 1980 and 2009.
Results
One thousand sixty patients with LCIS without concurrent breast cancer were identified. Median age at LCIS diagnosis was 50 years (range, 27 to 83 years). Fifty-six patients (5%) underwent bilateral prophylactic mastectomy; 1,004 chose surveillance with (n = 173) or without (n = 831) chemoprevention. At a median follow-up of 81 months (range, 6 to 368 months), 150 patients developed 168 breast cancers (63% ipsilateral, 25% contralateral, 12% bilateral), with no dominant histology (ductal carcinoma in situ, 35%; infiltrating ductal carcinoma, 29%; infiltrating lobular carcinoma, 27%; other, 9%). Breast cancer incidence was significantly reduced in women taking chemoprevention (10-year cumulative risk: 7% with chemoprevention; 21% with no chemoprevention; P < .001). In multivariable analysis, chemoprevention was the only clinical factor associated with breast cancer risk (hazard ratio, 0.27; 95% CI, 0.15 to 0.50). In a subgroup nested case-control analysis, volume of disease, which was defined as the ratio of slides with LCIS to total number of slides reviewed, was also associated with breast cancer development (P = .008).
Conclusion
We observed a 2% annual incidence of breast cancer among women with LCIS. Common clinical factors used for risk prediction, including age and family history, were not associated with breast cancer risk. The lower breast cancer incidence in women opting for chemoprevention highlights the potential for risk reduction in this population.
INTRODUCTION
Since the term lobular carcinoma in situ (LCIS) was coined in 1941,1 there has been controversy regarding the malignant potential and appropriate clinical management of this uncommon lesion.2–4 Although LCIS is a relatively rare finding, seen in approximately 0.5% to 4% of benign breast biopsies,5–7 an increase in LCIS incidence has been reported with increasing breast screening utilization, from 2.0 per 100,000 women in 2000 to 2.75 per 100,000 in 2009 among women reported to the Surveillance, Epidemiology, and End Results database.8 Compared with the general population, women with LCIS have a seven- to 10-fold increase in breast cancer risk,2,3,7,9,10 and studies with long-term follow-up report variable absolute risks of developing ductal carcinoma in situ (DCIS) or invasive carcinoma, ranging from 11% to 28%2,7,11,12 at 15 years, with a persistent risk over time.11 Currently, LCIS is managed with surveillance and risk reduction via chemoprevention or, rarely, bilateral prophylactic mastectomy.
With increasing numbers of women presenting with LCIS, a better understanding of the malignant potential and additional clinicopathologic factors that may alter the risk of breast cancer development are paramount to appropriately counsel women with this lesion. Previous work with atypical hyperplasia has indicated that multifocal atypia and younger age are strongly associated with invasive cancer development.13 Although small LCIS studies have suggested an association between disease volume and subsequent breast cancer development,14 no clear link between clinicopathologic factors in women with LCIS and disease progression has been established. Here, we present our single-institution experience with a large cohort of women with LCIS observed longitudinally; we report treatment decisions for the cohort, absolute risk of breast cancer development over time, and influence of additional clinicopathologic factors on subsequent breast cancer risk.
PATIENTS AND METHODS
Patients participating in surveillance after an LCIS diagnosis are entered into a prospectively maintained database; those entering surveillance between 1980 and 2009 were chosen for this analysis. Patients with prior or concurrent diagnosis (defined as within 6 months of LCIS diagnosis) of DCIS or invasive cancer, those who did not return for at least one follow-up visit after initial presentation, and BRCA mutation carriers were excluded. All outside pathology slides were reviewed to confirm LCIS diagnosis using current WHO diagnostic criteria.15 All patients diagnosed with LCIS by core biopsy during this period underwent surgical excision to rule out synchronous malignancy. Mammographic density defined by the Breast Imaging Reporting and Data System (BIRADS) classification (BIRADS 1: fatty; BIRADS 2: scattered fibroglandular density; BIRADS 3: heterogeneous/moderately dense; and BIRADS 4: extremely dense) was independently assigned by a single breast imager.
Patients participating in surveillance for LCIS at Memorial Sloan Kettering Cancer Center are offered annual or biannual clinical breast examination and annual mammography. The frequency of clinical breast examination and the use of ultrasound or magnetic resonance imaging (MRI) screening are at the discretion of the treating physician and patient. MRI for high-risk screening was available after April 1999, and our experience with this screening modality in this cohort has been published previously.16 Chemoprevention for LCIS was approved in 1998. This study was approved by the institutional review board.
Comparisons were made between patients who elected bilateral prophylactic mastectomy versus surveillance, with or without chemoprevention. The bilateral prophylactic mastectomy group includes all women undergoing risk-reducing surgery (n = 56). Women who elected bilateral prophylactic mastectomy within the first 6 months after LCIS diagnosis (n = 28) are included in the bilateral prophylactic mastectomy cohort only; women who underwent bilateral prophylactic mastectomy more than 6 months after LCIS diagnosis (n = 28) are included in the surveillance cohort and censored at time of surgery. Clinicopathologic characteristics associated with a cancer diagnosis were identified using log-rank tests and Cox regression analysis. The Kaplan-Meier method was used to estimate the annual risk of breast cancer among women in surveillance, with or without chemoprevention. Patients were censored at date of bilateral prophylactic mastectomy, cancer diagnosis, or last follow-up.
A nested case-control analysis was performed in the subset of patients whose original LCIS pathology slides (core biopsy plus excision) were available for independent review (J.C.G. and E.G.-R.). Patients with a subsequent cancer diagnosis were designated as patient cases and were matched for age at LCIS diagnosis ± 5 years and length of follow-up in a 1:5 ratio to patients who remained cancer free (controls). Conditional logistic regression analysis, accounting for the matched pairs design, was performed to examine the relationship between histopathologic features and breast cancer development.
RESULTS
Between November 1980 and December 2009, 1,060 patients were evaluated at Memorial Sloan Kettering Cancer Center after an LCIS diagnosis. One thousand forty-three patients presented after 1990 (Appendix Table A1, online only). Figure 1 shows the number of patients entering surveillance per year; the proportion lost to follow-up, defined as women in surveillance whose last contact was more than 18 months prior; and the proportion who have opted for bilateral prophylactic mastectomy. Data were maintained for this analysis through July 10, 2013. Median follow-up for the surveillance cohort was 81 months (range, 6 to 368 months).
Fig 1.
Number of patients diagnosed with lobular carcinoma in situ (LCIS) participating in surveillance at Memorial Sloan Kettering Cancer Center per year. The bar graph depicts the number of new patients entering surveillance for LCIS each year, patients who elected to undergo bilateral prophylactic mastectomy (n = 56), and patients lost to follow-up, defined as women who previously entered surveillance but whose last follow-up appointment was more than 18 months ago.
Median age at LCIS diagnosis was 50 years (range, 27 to 83 years). Surveillance alone was the most common management option, chosen by 831 patients (78%); 175 patients (17%) received chemoprevention (two patients subsequently went on to have bilateral prophylactic mastectomy), and 56 patients (5%) chose bilateral prophylactic mastectomy. Table 1 lists patient characteristics stratified by clinical management option. The majority of patients (745 of 1,060 patients; 70%) presented with LCIS after a biopsy was performed for abnormal findings on mammogram or ultrasound; 222 patients (21%) were found to have LCIS after biopsy for a palpable mass or nipple discharge; 46 patients (4%) presented after a biopsy triggered by MRI findings; 35 patients (3%) received an LCIS diagnosis after reduction mammoplasty; and for 13 patients, this information was not available.
Table 1.
Patient Characteristics Stratified by Treatment Group
| Characteristic | No. of Patients (%) |
P | ||
|---|---|---|---|---|
| Surveillance Alone (n = 831) | Surveillance With Chemoprevention (n = 173) | Bilateral Prophylactic Mastectomy (n = 56)* | ||
| Age at LCIS diagnosis, years | .0018 | |||
| Median | 50 | 51 | 47 | |
| Range | 27 to 83 | 36 to 77 | 36 to 60 | |
| < 50 | 413 (50) | 73 (42) | 39 (70) | |
| ≥ 50 | 418 (50) | 100 (58) | 17 (30) | |
| Menopausal status | < .001 | |||
| Premenopausal | 487 (59) | 90 (52) | 45 (80) | |
| Postmenopausal | 323 (39) | 81 (47) | 9 (16) | |
| Unknown | 21 (3) | 2 (1) | 2 (4) | |
| ≥ 1 FDR with breast cancer | < .001 | |||
| No | 603 (73) | 113 (65) | 19 (33) | |
| Yes | 202 (24) | 59 (34) | 37 (66) | |
| Unknown | 26 (3) | 1 (1) | 0 | |
| ≥ 2 SDRs with breast cancer | .42 | |||
| No | 708 (85) | 152 (88) | 46 (82) | |
| Yes | 97 (12) | 20 (12) | 10 (18) | |
| Unknown | 26 (3) | 1 (1) | 0 | |
| Bilateral synchronous LCIS† | 18 (2) | 5 (3) | 3 (5) | .30 |
| Concurrent ADH | 243 (29) | 43 (25) | 15 (27) | .49 |
| Concurrent ALH | 154 (19) | 29 (17) | 7 (113) | .48 |
| BIRADS breast density | .99 (BIRADS 1 v others); .0121 (BIRADS 4 v others) | |||
| BIRADS 1 | 19 (2) | 4 (2) | 1 (2) | |
| BIRADS 2 | 166 (20) | 45 (26) | 0 | |
| BIRADS 3 | 502 (60) | 91 (53) | 32 (57) | |
| BIRADS 4 | 133 (16) | 31 (18) | 16 (29) | |
| Unknown | 11 (1) | 2 (1) | 7 (12) | |
| ≥ 1 breast biopsy after initial LCIS diagnosis | 224 (27) | 57 (33) | 11 (20) | .11 |
| MRI screening | 431 (52) | 125 (72) | 23 (41) | < .001 |
Abbreviations: ADH, atypical ductal hyperplasia; ALH, atypical lobular hyperplasia; BIRADS, Breast Imaging Reporting and Data System; FDR, first-degree relative; LCIS, lobular carcinoma in situ; MRI, magnetic resonance imaging; SDR, second-degree relative.
Data missing for one bilateral prophylactic mastectomy patient.
Bilateral synchronous LCIS defined as bilateral breast biopsies demonstrating LCIS within 6 months of initial LCIS diagnosis.
Surveillance With or Without Chemoprevention for LCIS
Among the 1,032 women undergoing surveillance with or without chemoprevention, 150 (14%) developed 168 breast cancers (18 bilateral cancers). Figure 2A shows the Kaplan-Meier curve for time to any breast cancer diagnosis. For the entire surveillance population, the annual cancer development rate was 2% per year through the first 6 years after LCIS diagnosis with no evidence of a plateau out to 150 months of follow-up. Overall cumulative cancer incidence at 15 years was 26%. Median time to cancer diagnosis was 50 months (range, 6 to 194 months); median follow-up for those remaining cancer free was 83 months (range, 6 to 368 months).
Fig 2.
Kaplan-Meier curves for time to cancer events (ductal carcinoma in situ [DCIS] or invasive): (A) patients cancer free among the entire cohort with lobular carcinoma in situ (LCIS; n = 1,032); (B) patients cancer free by chemoprevention (CP) use (n = 175 with CP; n = 857 without CP); (C) patients cancer free by CP use and year of LCIS diagnosis; (D) ipsilateral versus contralateral cancer development (n = 1,032); and (E) DCIS versus invasive cancer development (n = 1,032).
Among 150 patients who developed cancer, 94 (63%) developed cancer in the ipsilateral breast, 38 (25%) developed cancer in the contralateral breast, and 18 (12%) developed bilateral breast cancer, 11 of which were diagnosed synchronously. Of 168 total cancers, 109 (65%) were invasive with an equal proportion of lobular and ductal histologies, and 59 (35%) were DCIS (Figs 2D and 2E). Of 109 invasive carcinomas, 93 (85%) were hormone receptor positive (Appendix Table A2, online only). The majority of cancers (94 of 168 cancers; 56%) were diagnosed by mammography/ultrasound, 41 (24%) were diagnosed by MRI, 20 (12%) were detected on physical examination, six DCISs (4%) were identified in a contralateral prophylactic mastectomy specimen, one tumor was found in a reduction mammoplasty specimen, and six cancers (4%) were diagnosed at outside hospitals with unknown details regarding diagnosis method.
One hundred seventy-three women took chemoprevention during surveillance without bilateral prophylactic mastectomy. Women opting for chemoprevention had higher MRI screening rates (72%) compared with women undergoing surveillance alone (52%) or bilateral prophylactic mastectomy (41%; P < .001; Table 1). Median time on chemoprevention was 55 months (range, 1 to 173 months); 84 women completed 5 years of chemoprevention at time of last analysis. The use of chemoprevention was strongly associated with a reduced cancer risk (Fig 2B). Although the use of chemoprevention differed by calendar year, with fewer women receiving chemoprevention before 1998 (n = 30) compared with during or after 1998 (n = 143), chemoprevention use in both time periods was significantly associated with a decrease in cancer development (P < .001; Fig 2C). Ten women who took chemoprevention developed 13 cancers during follow-up, with a median time to cancer diagnosis of 83 months (range, 45 to 194 months). Stage and hormone receptor status were available for five of seven invasive carcinomas; all were stage I hormone receptor–positive tumors. The cumulative cancer rates for women who took chemoprevention compared with those who did not were 3% and 7% at 5 years and 12% and 21% at 10 years, respectively.
There was no association between age at LCIS diagnosis, menopausal status, family history of breast cancer, finding of synchronous atypical hyperplasia or bilateral LCIS, BIRADS breast density, or the need for subsequent biopsies during surveillance and the likelihood of developing breast cancer on univariable analysis or after adjusting for the use of chemoprevention (Table 2). On multivariable analysis, including age, mammographic density, chemoprevention use, and family history of breast cancer, only chemoprevention use was significantly associated with breast cancer risk reduction (hazard ratio, 0.27; 95% CI, 0.15 to 0.50; P ≤ .001; Table 3). Outcomes by chemoprevention use in different age cohorts demonstrated reduced cancer development in all age cohorts, although this difference was statistically significant only among women younger than age 45 years (P = .023) and age 50 to 60 years (P = .008). For women age 45 to 50 and older than 60 years, this trend was nonsignificant (P = .12 and P = .07, respectively; Appendix Fig A1, online only).
Table 2.
Univariable and Cox Regression Analysis of Factors Associated With Development of Breast Cancer Among Women With LCIS for the Entire Cohort and After Adjustment for Use of Chemoprevention
| Characteristic | No. of Patients (%) (n = 1,032) | Univariable P for Association With Cancer* | Hazard Ratio (95% CI) | P Adjusted for Use of Chemoprevention† |
|---|---|---|---|---|
| Age at LCIS diagnosis, years | ||||
| Median | 50 | .52 | 1.008 (0.991 to 1.025) | .37 |
| Range | 27-83 | |||
| < 50 | 504 (49) | .73 | 0.883 (0.643 to 1.212) | .44 |
| ≥ 50 | 528 (51) | |||
| Age cohorts at LCIS diagnosis, years | .27 | .41 | ||
| < 45 | 230 (22) | 0.887 (0.546 to 1.441) | ||
| 45-50 | 273 (26) | 0.659 (0.396 to 1.097) | ||
| 51-60 | 371 (36) | 0.815 (0.511 to 1.300) | ||
| > 60 | 158 (15) | Reference | ||
| Menopausal status | .62 | .37 | ||
| Premenopausal/perimenopausal | 600 (58) | 0.861 (0.621 to 1.192) | ||
| Postmenopausal | 408 (40) | Reference | ||
| Unknown | 24 (2) | |||
| ≥ 1 FDR with breast cancer | .27 | .13 | ||
| No | 727 (70) | 0.767 (0.543 to 1.083) | ||
| Yes | 278 (27) | Reference | ||
| Unknown | 27 (3) | |||
| ≥ 2 SDRs with breast cancer | .62 | .62 | ||
| No | 886 (86) | 1.143 (0.671 to 1.949) | ||
| Yes | 119 (12) | Reference | ||
| Unknown | 27 (3) | |||
| Bilateral synchronous LCIS‡ | .93 | .98 | ||
| Absent | 1,008 (98) | 1.013 (0.323 to 3.177) | ||
| Present | 24 (2) | Reference | ||
| Concurrent ADH | .32 | .37 | ||
| Absent | 736 (71) | 0.851 (0.599 to 1.209) | ||
| Present | 296 (29) | Reference | ||
| Concurrent ALH | .39 | .46 | ||
| Absent | 846 (82) | 0.861 (0.581 to 1.278) | ||
| Present | 186 (18) | Reference | ||
| No. of breast biopsies after initial LCIS diagnosis | .58 | .51 | ||
| 0 | 741 (72) | 0.893 (0.638 to 1.249) | ||
| ≥ 1 | 291 (28) | Reference | ||
| BIRADS breast density | ||||
| BIRADS 1 | 24 (2) | .28 | BIRADS 1 v 2/3/4: 0.456 (0.113 to 1.841) | .27 |
| BIRADS 2 | 211 (20) | |||
| BIRADS 3 | 609 (59) | |||
| BIRADS 4 | 173 (17) | .56 | BIRADS 1/2/3 v 4: 0.870 (0.570 to 1.327) | .52 |
| Unknown | 15 (1) |
Abbreviations: ADH, atypical ductal hyperplasia; ALH, atypical lobular hyperplasia; BIRADS, Breast Imaging Reporting and Data System; FDR, first-degree relative; LCIS, lobular carcinoma in situ; SDR, second-degree relative.
Log-rank test; tests two or more Kaplan-Meier curves.
Cox regression analysis.
Bilateral synchronous LCIS was defined as bilateral breast biopsies demonstrating LCIS within 6 months of initial LCIS diagnosis.
Table 3.
Cox Multivariable Model Assessing Risk Factors and Risk of Cancer Development Among Women With LCIS
| Parameter | Hazard Ratio | 95% CI | P |
|---|---|---|---|
| Age at LCIS diagnosis | 1.006 | 0.99 to 1.02 | .47 |
| Mammographic density (BIRADS 1 v other) | 0.470 | 0.12 to 1.90 | .29 |
| Use of chemoprevention (yes v no) | 0.269 | 0.15 to 0.50 | < .001 |
| First-degree family history of breast cancer (no v yes) | 0.769 | 0.54 to 1.09 | .13 |
| Second-degree family history of breast cancer (no v yes) | 1.119 | 0.66 to 1.91 | .68 |
Abbreviations: BIRADS, Breast Imaging Reporting and Data System; LCIS, lobular carcinoma in situ.
Bilateral Prophylactic Mastectomy for LCIS
A minority of patients (n = 56; 5%) chose bilateral prophylactic mastectomy. Median time from LCIS diagnosis to bilateral prophylactic mastectomy was 6 months (range, 0 to 90 months). The earliest date of bilateral prophylactic mastectomy was 1993. The number of patients with LCIS choosing bilateral prophylactic mastectomy per year ranged from zero to nine, with no trend over time. Women opting for bilateral prophylactic mastectomy were younger and more likely to be premenopausal, have dense breasts, and have stronger first-degree family histories of breast cancer compared with women opting for surveillance, with or without chemoprevention (Table 1). Occult cancer was identified in six (11%) of 56 prophylactic mastectomy specimens (three subcentimeter invasive carcinomas and three DCISs). Thirty-two women opting for bilateral prophylactic mastectomy are still in active follow-up with a median follow-up time of 68 months (range, 22 to 237 months); all remain cancer free.
Case-Control Analyses
Seventy-two patient cases and 274 controls were identified for inclusion in the subset nested case-control analysis. Patients in the case-control analysis, compared with the entire LCIS population, were less likely to have taken chemoprevention (7% v 17%, respectively) or undergone MRI screening (46% v 55%, respectively). No significant differences were noted between patient cases and controls for any of the individual pathologic features recorded (Table 4); however, the ratio of number of slides with LCIS to total number of slides reviewed was significantly higher in women who went on to develop cancer (patient cases) compared with those who did not (median ratio, 0.5 for patient cases v 0.3 for controls; P = .003). When the ratio was stratified as less than 0.25, 0.25 to 0.50, and more than 0.50, a higher ratio was associated with increasing odds for cancer development (Table 4).
Table 4.
Pathologic Review of Original LCIS Slides Among Age- and Length of Follow-Up–Matched Patient Cases and Controls
| Pathologic Variable | No. (%) |
Odds Ratio (95% CI) | P | |
|---|---|---|---|---|
| Patient Cases (n = 72) | Controls (n = 274) | |||
| Bilateral LCIS at presentation | 42 (3) | 6 (2) | .81 | |
| Multiple unilateral LCIS lesions (multicentric) at diagnosis | 7 (10) | 25 (9) | .68 | |
| Recurrent LCIS (future excisions after initial diagnosis) | 9 (13) | 24 (9) | .30 | |
| Nuclear grade | .54 | |||
| 1 | 14 (19) | 58 (21) | ||
| 2 | 54 (75) | 189 (69) | ||
| 3 | 4 (6) | 27 (10) | ||
| Necrosis | 0 | 6 (2) | NA | |
| Calcifications | 27 (38) | 114 (42) | .57 | |
| Ductal extension | 60 (83) | 249 (91) | .092 | |
| Synchronous ALH | 12 (17) | 31 (11) | .19 | |
| Median LCIS size, cm (range) | 0.4 (0-2.5) | 0.4 (0-2.3) | .90 | |
| Median No. of slides with LCIS (range) | 4 (1-19) | 3 (1-41) | — | |
| Median No. slides reviewed (range) | 10.5 (1-37) | 13 (1-46) | — | |
| Median No. of TDLUs with LCIS (range) | 11 (1-129) | 6 (1-342) | .40 | |
| Median No. of TDLUs with ALH (range) | 0 (0-6) | 0 (0-11) | .93 | |
| Median sum of TDLUs (range) | 11 (1-129) | 7 (1-342) | .45 | |
| Median ratio (range)* | 0.5 (0.027-1) | 0.3 (0-1) | .003 | |
| Ratio stratification* | ||||
| < 0.25 | 17 (24) | 122 (45) | Reference | .0079 |
| 0.25-0.50 | 24 (33) | 75 (27) | 2.28 (1.16 to 4.46) | |
| > 0.50 | 31 (43) | 77 (28) | 2.68 (1.41 to 4.99) | |
NOTE. In patients with multiple lesions or multiple samples from the same lesion, the greatest diameter of the largest lesion and the highest-grade features were recorded, and the number of terminal ductal lobular units affected by LCIS was considered additive. Lesions were considered discrete if they were from two different breasts or separate quadrants of the same breast.
Abbreviations: ALH, atypical lobular hyperplasia; LCIS, lobular carcinoma in situ; NA, not available; TDLU, terminal ductal-lobular unit.
Ratio is defined as the number of slides with LCIS over the number of slides reviewed.
DISCUSSION
In this large population of women diagnosed with LCIS from 1980 to 2009 and observed longitudinally at a single center, the rate of cancer development was approximately 2% per year and translated into a cumulative long-term rate of 26% at 15 years, mirroring the historic high rates reported by Haagensen et al2 from the 1940s to 1960s and the updated results from the National Surgical Adjuvant Breast and Bowel Project (NSABP) B-17 study. Fisher et al17 reported an overall cancer rate of 22% at 12 years of follow-up for 180 women with LCIS treated with excision alone (14% ipsilateral, 8% contralateral). Other studies with long-term follow-up have reported lower cancer rates, ranging from 11% to 18% at 15 years.7,11,12 In a study of 4,853 patients with LCIS as reported to the Surveillance, Epidemiology, and End Results Program between 1973 and 1998, Chuba et al12 reported an 11% invasive cancer incidence at 15 years; however, this study is limited by lack of central pathology review and treatment information.
Additional variation in breast cancer risk seen in prior studies may be related to inclusion of women with a spectrum of lobular neoplasia, thus representing a spectrum of risk ranging from atypical lobular hyperplasia (ALH) to LCIS. Hartmann et al18 published the Mayo Clinic's long-term experience with atypical hyperplasia and reported a 19% breast cancer risk at 15 years, slightly lower than the risk seen in our LCIS cohort. Although younger age at LCIS diagnosis has previously been reported as an increased risk predictor,11 we found no association in this cohort between breast cancer risk and patient age or other clinicopathologic factors, including concurrent atypical ductal hyperplasia or ALH, menopausal status, bilateral LCIS, or breast density. Importantly, and similar to what has been reported for atypical hyperplasia,19–21 in our population of women with LCIS, we found no additional risk increase for women with a family history of breast cancer.
Data regarding the association between disease volume and subsequent breast cancer risk among women with high-risk lesions are variable.14,17,18 Fisher et al17 examined 19 pathologic variables as potential predictors for the development of invasive carcinoma or DCIS, and only LCIS grade was significantly associated with the risk of breast cancer (number of invasive/DCIS recurrences by grade: grade 1, 1.5%; grade 2, 12.5%; grade 3, 13.5%; P = .037); the number of involved lobules with LCIS was not predictive. In a small cohort of 88 patients with LCIS, Ottesen et al14 found that the presence of ≥ 10 lobules with LCIS was significantly associated with a higher rate of breast cancer development (24% with ≥ 10 lobules v 8% with < 10 lobules; P = .028). In a nested case-control analysis, we also found that LCIS disease volume, represented by the ratio of number of slides with LCIS to the total number of slides examined, was significantly associated with breast cancer development, with a ratio of 0.5 to 1 representing the highest risk (odds ratio, 2.68). If validated in an independent population, this factor may provide an opportunity for risk stratification among women with LCIS.
Chemoprevention with a selective estrogen receptor modulator or an aromatase inhibitor reduces breast cancer risk by 40% to 65% in women at elevated risk,22–25 and data from the NSABP P-1 trial suggest that women with LCIS derive even greater benefit.24,25 The NSABP P-1 trial randomly assigned women at increased breast cancer risk to 5 years of tamoxifen or placebo therapy. Tamoxifen reduced the risk of invasive breast cancer by 49% in all women and by 56% among 826 women with LCIS.24 Coopey et al26 evaluated the effect of chemoprevention on women with a history of atypical breast lesions (atypical ductal hyperplasia, ALH, or LCIS) in a community practice setting and reported a significant decrease in 10-year rates of breast cancer development among all women (21% with no chemoprevention v 8% with chemoprevention; P < .001), with numerically greater reductions seen among women with LCIS (32% with no chemoprevention v 10% with chemoprevention). Updated American Society of Clinical Oncology practice guidelines endorse the recommendation for tamoxifen, raloxifene, or exemestane as chemoprevention options for high-risk women, including women with LCIS.27 Although there is strong evidence for risk reduction with endocrine therapy, uptake28 and completion of chemoprevention therapy are low.29 Here, we also report a substantially significant decrease in breast cancer incidence for women with LCIS who opted for chemoprevention, with an annual cancer rate of less than 1%, further supporting the recommendation for chemoprevention in this patient population. Further, because the chemoprevention and bilateral prophylactic mastectomy cohorts had more high-risk features than the surveillance-only cohort (Table 1) and the bilateral prophylactic mastectomy patients were censored at time of surgery, one could postulate that informative censoring in this observational data set reduced the risk in the surveillance-only cohort, leading to an even more impressive effect of risk-reduction strategies.
Although a history of LCIS clearly portends an increased breast cancer risk, the mechanism for this remains an area of ongoing investigation. The laterality of breast cancer after an LCIS diagnosis provides insight into the behavior of this lesion, either as a direct precursor or a risk factor. Within this LCIS cohort, 63%, 25%, and 12% of subsequent cancers were ipsilateral to the breast with LCIS, contralateral, and bilateral, respectively. This finding and molecular evidence demonstrating that gene expression can distinguish two LCIS subclasses,30 that approximately 70% of LCIS lesions seem to be clonally related to synchronously diagnosed invasive lesions by copy number analysis,31 and the frequency of common mutations in microdissected LCIS and infiltrating lobular carcinoma32 suggest that LCIS is a heterogenous lesion representing both a nonobligate precursor lesion and a high-risk marker.
With a cumulative breast cancer risk of more than 20% at 15 years, the role of MRI screening among women with LCIS has been questioned. We did not examine the benefit of MRI screening in this overall population; however, we did previously report our findings in a cohort diagnosed after April 1999 and were unable to demonstrate a benefit for MRI screening in women with LCIS.16 Importantly, trials supporting MRI screening in women with a greater than 20% lifetime risk of breast cancer were based on elevated risk incurred by family history or suspected BRCA mutations33 and, therefore, represent a population at risk for potentially more aggressive breast cancer subtypes and interval cancers. In our experience and that of others,34 women with LCIS typically develop low- to moderate-grade estrogen receptor–positive breast cancers.
To our knowledge, this study represents the largest single-institution experience with a cohort of women with LCIS observed longitudinally over time. Single-institution pathology review of all slides for inclusion in the database and extensive pathologic review of the patients included in the case-control analysis provide the opportunity to ensure accurate diagnoses and further refine risk based on extent of LCIS at presentation. Although it seems that the cancer risk may plateau after 150 months, there are only 165 patients with more than 150 months of follow-up; therefore, longer follow-up is necessary to determine whether the annual cancer risk persists. Furthermore, although the risk reduction we report with chemoprevention use is consistent with other prospective and retrospective reports,24,26 we were unable to control for all possible confounding factors in this observational data set.
Our study provides a thorough analysis of clinicopathologic factors that may influence breast cancer development in a prospectively followed cohort of women with LCIS and demonstrates an overall 2% annual risk of breast cancer, influenced only by disease volume as defined by the ratio of slides with LCIS over the total number of slides examined. Additional clinicopathologic features commonly discussed with women when providing risk assessment, such as age and family history, were not found to affect breast cancer risk in this cohort. Notably, within the cohort of women who chose surveillance with chemoprevention, the breast cancer development rate was substantially decreased, highlighting the significant impact in risk reduction that is possible with proper education and counseling for women with LCIS.
Acknowledgment
We thank Janice Sung, MD, Department of Radiology, Memorial Sloan Kettering Cancer Center, for assistance in reviewing the breast imaging for this project.
Appendix
Table A1.
Frequency per Year (1980 to 2009) of Women Entering Surveillance for LCIS and Opting for Bilateral Prophylactic Mastectomy or Chemoprevention
| Year | No. of Women per Year Entering Surveillance | Bilateral Prophylactic Mastectomy | No Chemoprevention | Chemoprevention |
|---|---|---|---|---|
| 1980 | 1 | 0 | 1 | 0 |
| 1983 | 1 | 0 | 1 | 0 |
| 1984 | 2 | 0 | 0 | 2 |
| 1985 | 1 | 0 | 1 | 0 |
| 1986 | 1 | 0 | 0 | 1 |
| 1987 | 2 | 0 | 2 | 0 |
| 1988 | 4 | 0 | 4 | 0 |
| 1989 | 3 | 0 | 2 | 1 |
| 1990 | 2 | 0 | 1 | 1 |
| 1991 | 3 | 0 | 2 | 1 |
| 1992 | 14 | 0 | 12 | 2 |
| 1993 | 14 | 1 | 9 | 4 |
| 1994 | 10 | 2 | 7 | 1 |
| 1995 | 26 | 2 | 20 | 4 |
| 1996 | 31 | 2 | 26 | 3 |
| 1997 | 51 | 0 | 41 | 10 |
| 1998 | 57 | 0 | 44 | 13 |
| 1999 | 55 | 3 | 42 | 10 |
| 2000 | 61 | 2 | 49 | 10 |
| 2001 | 85 | 7 | 67 | 11 |
| 2002 | 58 | 2 | 46 | 10 |
| 2003 | 68 | 3 | 51 | 14 |
| 2004 | 82 | 2 | 62 | 18 |
| 2005 | 77 | 3 | 63 | 11 |
| 2006 | 79 | 9 | 62 | 8 |
| 2007 | 81 | 5 | 67 | 9 |
| 2008 | 91 | 5 | 71 | 15 |
| 2009 | 100 | 8 | 78 | 14 |
Abbreviation: LCIS, lobular carcinoma in situ.
Table A2.
Pathologic Characteristics of Breast Cancers That Developed in Patients With LCIS
| Pathologic Characteristic | No. of Patients (%) (n = 168) |
|---|---|
| Histology | |
| DCIS | 59 (35) |
| ILC | 45 (27) |
| IDC | 48 (29) |
| IDC with lobular features | 8 (5) |
| Special types | 7 (4) |
| Invasive (NOS) | 1 (1) |
| Laterality of cancer* | |
| Ipsilateral | 94 (63) |
| Contralateral | 38 (25) |
| Bilateral | 18 (12) |
| Mean tumor size, cm (range)† | 1.1 (0.1-12) |
| Nodal status‡ | |
| pN0 | 81 (74) |
| pN1-N3 | 20 (18) |
| Nodal status unknown | 8 (7) |
| Pathologic stage | |
| 0 | 59 (35) |
| I | 76 (45) |
| II | 20 (12) |
| III | 5 (3) |
| Unknown | 8 (5) |
| Receptor status‡ | |
| ER/PR positive, HER2 negative | 87 (80) |
| ER/PR positive, HER2 positive | 6 (6) |
| ER/PR negative, HER2 negative | 3 (3) |
| ER/PR negative, HER2 positive | 3 (3) |
| Unknown | 10 (9) |
Abbreviations: DCIS, ductal carcinoma in situ; ER estrogen receptor; HER2, human epidermal growth factor receptor; IDC, infiltrating ductal carcinoma; ILC, infiltrating lobular carcinoma; LCIS, lobular carcinoma in situ; NOS, not otherwise specified; PR progesterone receptor.
Laterality of breast cancer compared with laterality of initial LCIS.
Invasive tumors with known data only (n = 105); one patient presented with inflammatory breast cancer with a tumor size of 12 cm. Excluding this patient, the mean tumor size is 0.99 cm.
Invasive tumors only (n = 109).
Fig A1.
Kaplan-Meier curves for time to cancer events (ductal carcinoma in situ or invasive): (A) patients younger than age 45 years cancer free by chemoprevention (CP) use (n = 230); (B) patients age 45 to 50 years cancer free by CP use (n = 273); (C) patients age 50 to 60 years cancer free by CP use (n = 371); and (D) patients older than age 60 years cancer free by CP use (n = 158). LCIS, lobular carcinoma in situ.
Footnotes
Supported, in part, by National Institutes of Health/National Cancer Institute Cancer Center Support Grant No. P30 CA008748, the Walsh Family Fund, and the Cary Grossman Breast Research Fellowship.
Authors' disclosures of potential conflicts of interest are found in the article online at www.jco.org. Author contributions are found at the end of this article.
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Disclosures provided by the authors are available with this article at www.jco.org.
AUTHOR CONTRIBUTIONS
Conception and design: Tari A. King, Melissa Pilewskie, Shirin Muhsen, Sujata Patil, Starr K. Mautner, Anna Park, Elena Guerini-Rocco, Jessica C. Gooch, Jorge S. Reis-Filho, Mary Morrogh, Monica Morrow
Administrative support: Tari A. King, Melissa Pilewskie, Shirin Muhsen, Sujata Patil, Starr K. Mautner, Anna Park, Sabine Oskar, Camilla Boafo, Monica Morrow
Provision of study materials or patients: Tari A. King, Melissa Pilewskie, Shirin Muhsen, Sujata Patil, Starr K. Mautner, Anna Park, Sabine Oskar, Elena Guerini-Rocco, Camilla Boafo, Jessica C. Gooch, Marina De Brot, Victor P. Andrade, Rita A. Sakr
Collection and assembly of data: Tari A. King, Melissa Pilewskie, Shirin Muhsen, Sujata Patil, Starr K. Mautner, Anna Park, Sabine Oskar, Elena Guerini-Rocco, Camilla Boafo, Jessica C. Gooch, Marina De Brot, Victor P. Andrade, Rita A. Sakr, Monica Morrow
Data analysis and interpretation: Tari A. King, Melissa Pilewskie, Shirin Muhsen, Sujata Patil, Starr K. Mautner, Anna Park, Elena Guerini-Rocco, Jessica C. Gooch, Jorge S. Reis-Filho, Monica Morrow
Manuscript writing: All authors
Final approval of manuscript: All authors
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
Lobular Carcinoma in Situ: A 29-Year Longitudinal Experience Evaluating Clinicopathologic Features and Breast Cancer Risk
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or jco.ascopubs.org/site/ifc.
Tari A. King
No relationship to disclose
Melissa Pilewskie
No relationship to disclose
Shirin Muhsen
No relationship to disclose
Sujata Patil
No relationship to disclose
Starr K. Mautner
No relationship to disclose
Anna Park
No relationship to disclose
Sabine Oskar
No relationship to disclose
Elena Guerini-Rocco
No relationship to disclose
Camilla Boafo
No relationship to disclose
Jessica C. Gooch
No relationship to disclose
Marina De Brot
No relationship to disclose
Jorge S. Reis-Filho
No relationship to disclose
Mary Morrogh
No relationship to disclose
Victor P. Andrade
No relationship to disclose
Rita A. Sakr
No relationship to disclose
Monica Morrow
Honoraria: Genomic Health
REFERENCES
- 1.Foote FW, Stewart FW. Lobular carcinoma in situ: A rare form of mammary cancer. Am J Pathol. 1941;17:491–496. doi: 10.3322/canjclin.32.4.234. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Haagensen CD, Lane N, Lattes R, et al. Lobular neoplasia (so-called lobular carcinoma in situ) of the breast. Cancer. 1978;42:737–769. doi: 10.1002/1097-0142(197808)42:2<737::aid-cncr2820420247>3.0.co;2-t. [DOI] [PubMed] [Google Scholar]
- 3.Rosen PP, Kosloff C, Lieberman PH, et al. Lobular carcinoma in situ of the breast: Detailed analysis of 99 patients with average follow-up of 24 years. Am J Surg Pathol. 1978;2:225–251. doi: 10.1097/00000478-197809000-00001. [DOI] [PubMed] [Google Scholar]
- 4.Page DL, Schuyler PA, Dupont WD, et al. Atypical lobular hyperplasia as a unilateral predictor of breast cancer risk: A retrospective cohort study. Lancet. 2003;361:125–129. doi: 10.1016/S0140-6736(03)12230-1. [DOI] [PubMed] [Google Scholar]
- 5.Frykberg ER. Lobular carcinoma in situ of the breast. Breast J. 1999;5:296–303. doi: 10.1046/j.1524-4741.1999.00002.x. [DOI] [PubMed] [Google Scholar]
- 6.Hussain M, Cunnick GH. Management of lobular carcinoma in-situ and atypical lobular hyperplasia of the breast: A review. Eur J Surg Oncol. 2011;37:279–289. doi: 10.1016/j.ejso.2011.01.009. [DOI] [PubMed] [Google Scholar]
- 7.Page DL, Kidd TE, Jr, Dupont WD, et al. Lobular neoplasia of the breast: Higher risk for subsequent invasive cancer predicted by more extensive disease. Hum Pathol. 1991;22:1232–1239. doi: 10.1016/0046-8177(91)90105-x. [DOI] [PubMed] [Google Scholar]
- 8.Portschy PR, Marmor S, Nzara R, et al. Trends in incidence and management of lobular carcinoma in situ: A population-based analysis. Ann Surg Oncol. 2013;20:3240–3246. doi: 10.1245/s10434-013-3121-4. [DOI] [PubMed] [Google Scholar]
- 9.Page DL, Dupont WD, Rogers LW, et al. Atypical hyperplastic lesions of the female breast: A long-term follow-up study. Cancer. 1985;55:2698–2708. doi: 10.1002/1097-0142(19850601)55:11<2698::aid-cncr2820551127>3.0.co;2-a. [DOI] [PubMed] [Google Scholar]
- 10.Dupont WD, Page DL. Risk factors for breast cancer in women with proliferative breast disease. N Engl J Med. 1985;312:146–151. doi: 10.1056/NEJM198501173120303. [DOI] [PubMed] [Google Scholar]
- 11.Bodian CA, Perzin KH, Lattes R. Lobular neoplasia: Long term risk of breast cancer and relation to other factors. Cancer. 1996;78:1024–1034. doi: 10.1002/(SICI)1097-0142(19960901)78:5<1024::AID-CNCR12>3.0.CO;2-4. [DOI] [PubMed] [Google Scholar]
- 12.Chuba PJ, Hamre MR, Yap J, et al. Bilateral risk for subsequent breast cancer after lobular carcinoma-in-situ: Analysis of Surveillance, Epidemiology, and End Results data. J Clin Oncol. 2005;23:5534–5541. doi: 10.1200/JCO.2005.04.038. [DOI] [PubMed] [Google Scholar]
- 13.Degnim AC, Visscher DW, Berman HK, et al. Stratification of breast cancer risk in women with atypia: A Mayo cohort study. J Clin Oncol. 2007;25:2671–2677. doi: 10.1200/JCO.2006.09.0217. [DOI] [PubMed] [Google Scholar]
- 14.Ottesen GL, Graversen HP, Blichert-Toft M, et al. Lobular carcinoma in situ of the female breast: Short-term results of a prospective nationwide study—The Danish Breast Cancer Cooperative Group. Am J Surg Pathol. 1993;17:14–21. doi: 10.1097/00000478-199301000-00002. [DOI] [PubMed] [Google Scholar]
- 15.Lakhani SR, Ellis IO, Schnitt SJ, et al., editors. WHO Classification of Tumours of the Breast. Lyon, France: IARC Press; 2012. [Google Scholar]
- 16.King TA, Muhsen S, Patil S, et al. Is there a role for routine screening MRI in women with LCIS? Breast Cancer Res Treat. 2013;142:445–453. doi: 10.1007/s10549-013-2725-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Fisher ER, Land SR, Fisher B, et al. Pathologic findings from the National Surgical Adjuvant Breast and Bowel Project: Twelve-year observations concerning lobular carcinoma in situ. Cancer. 2004;100:238–244. doi: 10.1002/cncr.11883. [DOI] [PubMed] [Google Scholar]
- 18.Hartmann LC, Degnim AC, Santen RJ, et al. Atypical hyperplasia of the breast: Risk assessment and management options. N Engl J Med. 2015;372:78–89. doi: 10.1056/NEJMsr1407164. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Hartmann LC, Radisky DC, Frost MH, et al. Understanding the premalignant potential of atypical hyperplasia through its natural history: A longitudinal cohort study. Cancer Prev Res (Phila) 2014;7:211–217. doi: 10.1158/1940-6207.CAPR-13-0222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.London SJ, Connolly JL, Schnitt SJ, et al. A prospective study of benign breast disease and the risk of breast cancer. JAMA. 1992;267:941–944. [PubMed] [Google Scholar]
- 21.Kabat GC, Jones JG, Olson N, et al. A multi-center prospective cohort study of benign breast disease and risk of subsequent breast cancer. Cancer Causes Control. 2010;21:821–828. doi: 10.1007/s10552-010-9508-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Vogel VG, Costantino JP, Wickerham DL, et al. Update of the National Surgical Adjuvant Breast and Bowel Project Study of Tamoxifen and Raloxifene (STAR) P-2 Trial: Preventing breast cancer. Cancer Prev Res (Phila) 2010;3:696–706. doi: 10.1158/1940-6207.CAPR-10-0076. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Goss PE, Ingle JN, Alés-Martinez JE, et al. Exemestane for breast-cancer prevention in postmenopausal women. N Engl J Med. 2011;364:2381–2391. doi: 10.1056/NEJMoa1103507. [DOI] [PubMed] [Google Scholar]
- 24.Fisher B, Costantino JP, Wickerham DL, et al. Tamoxifen for prevention of breast cancer: Report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. J Natl Cancer Inst. 1998;90:1371–1388. doi: 10.1093/jnci/90.18.1371. [DOI] [PubMed] [Google Scholar]
- 25.Cuzick J, Sestak I, Forbes JF, et al. Anastrozole for prevention of breast cancer in high-risk postmenopausal women (IBIS-II): An international, double-blind, randomised placebo-controlled trial. Lancet. 2014;383:1041–1048. doi: 10.1016/S0140-6736(13)62292-8. [DOI] [PubMed] [Google Scholar]
- 26.Coopey SB, Mazzola E, Buckley JM, et al. The role of chemoprevention in modifying the risk of breast cancer in women with atypical breast lesions. Breast Cancer Res Treat. 2012;136:627–633. doi: 10.1007/s10549-012-2318-8. [DOI] [PubMed] [Google Scholar]
- 27.Visvanathan K, Hurley P, Bantug E, et al. Use of pharmacologic interventions for breast cancer risk reduction: American Society of Clinical Oncology clinical practice guideline. J Clin Oncol. 2013;31:2942–2962. doi: 10.1200/JCO.2013.49.3122. [DOI] [PubMed] [Google Scholar]
- 28.Ropka ME, Keim J, Philbrick JT. Patient decisions about breast cancer chemoprevention: A systematic review and meta-analysis. J Clin Oncol. 2010;28:3090–3095. doi: 10.1200/JCO.2009.27.8077. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29.Roetzheim RG, Lee JH, Fulp W, et al. Acceptance and adherence to chemoprevention among women at increased risk of breast cancer. Breast. 2015;24:51–56. doi: 10.1016/j.breast.2014.11.006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Andrade VP, Morrogh M, Qin LX, et al. Gene expression profiling of lobular carcinoma in situ reveals candidate precursor genes for invasion. Mol Oncol. 2015;9:772–782. doi: 10.1016/j.molonc.2014.12.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31.Andrade VP, Ostrovnaya I, Seshan VE, et al. Clonal relatedness between lobular carcinoma in situ and synchronous malignant lesions. Breast Cancer Res. 2012;14:R103. doi: 10.1186/bcr3222. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32.Sakr RA, Scarpa JV, Schizas M, et al. Targeted capture next generation sequencing of fresh frozen lobular carcinoma in situ and invasive lobular cancer identifies a common repertoire of mutations. Cancer Res. 2014;74:927. [Google Scholar]
- 33.Warner E, Messersmith H, Causer P, et al. Systematic review: Using magnetic resonance imaging to screen women at high risk for breast cancer. Ann Intern Med. 2008;148:671–679. doi: 10.7326/0003-4819-148-9-200805060-00007. [DOI] [PubMed] [Google Scholar]
- 34.Abdel-Fatah TM, Powe DG, Hodi Z, et al. Morphologic and molecular evolutionary pathways of low nuclear grade invasive breast cancers and their putative precursor lesions: Further evidence to support the concept of low nuclear grade breast neoplasia family. Am J Surg Pathol. 2008;32:513–523. doi: 10.1097/PAS.0b013e318161d1a5. [DOI] [PubMed] [Google Scholar]