Abstract
Obesity is associated with adverse biologic features and poor outcome in patients with invasive breast cancer, yet this relationship has not been evaluated in patients with ductal carcinoma in situ (DCIS). From 1996 to 2009, body mass index (BMI) was recorded at initial diagnosis for 1,885 patients with DCIS treated at our institution. Patients were categorized as obese (BMI ≥ 30 kg/m2), overweight (BMI 25 to<30 kg/m2), or of normal weight or underweight (BMI <25 kg/m2). Logistic regression was used to examine associations between BMI and patient, clinical, and pathologic features and treatment. Local–regional recurrence was calculated using the Kaplan–Meier method. All statistical tests were two-sided. Of the 1,885 patients, 514 (27.7%) were obese, 510 (27.5%) were overweight, and 831 (44.8%) were normal/underweight. In multivariate analysis, overweight and obese patients were significantly more likely to be African American (odds ratio [OR], 3.93; 95% confidence interval [CI], 2.66–5.80) or Hispanic (OR, 1.44; CI, 1.02–2.04), be postmenopausal (OR, 1.63; CI, 1.28–2.07), have diabetes (OR, 4.60; CI, 2.60–8.12), have estrogen-receptor-positive DCIS (OR, 1.39; CI, 1.00–192), and present with a radiologic abnormality rather than clinical symptoms (OR, 1.35; CI, 1.01–1.80). At a median follow-up time of 4.96 years (range, 1.0–14.34 years), no significant differences in local recurrence rates were detected based on patients’ initial BMI category. Furthermore, there was no significant difference in risk of recurrence between diabetic patients receiving metformin or not. In conclusion, higher BMI is not associated with adverse biologic features or prognosis in patients with DCIS.
Keywords: DCIS, Obesity, Breast cancer, Body mass index, Metformin
Introduction
Obesity and breast cancer both represent major public health problems in developed countries. In the United States, 64.1% of women over the age of 20 years are overweight or obese [1]. Obesity is a risk factor for multiple chronic medical conditions, including diabetes, hypertension, hypercholesterolemia, cardiovascular disease, and arthritis, as well as for cancers of the colon, endometrium, and kidney and invasive breast cancer [2]. In the case of invasive breast cancer, there is an increasing body of literature showing that obese patients are more likely to present with advanced-stage disease, derive less benefit from adjuvant systemic therapy, are more likely to develop distant metastases, and die from breast cancer more often than normal weight or underweight patients [3–8].
Although there are increasing numbers of reports supporting a relationship between high body mass index (BMI) and worse prognosis in patients treated for invasive breast cancer, we are aware of no specific studies related to the effect of obesity on prognosis in patients with ductal carcinoma in situ (DCIS). Given that DCIS now accounts for approximately 20–25% of breast cancer diagnoses and given the high prevalence of obesity in the United States, understanding the effect of obesity on the presentation, treatment, and outcome of DCIS is essential. To elucidate these issues, we studied 1,855 patients with pure DCIS recently treated at The University of Texas MD Anderson Cancer Center. As diabetes is known to be associated with obesity and the antidiabetic metformin has been identified as a potential anticancer agent in preclinical, epidemiologic, and clinical breast cancer studies, the relationship between metformin use and clinical and pathologic presenting features of DCIS and outcome was also explored [9–11]. This is the first report to comprehensively describe the impact of initial BMI on the clinical presentation, clinicopathologic features, treatment received, and outcome in patients with DCIS.
Methods
Following approval from the Institutional Review Board of MD Anderson Cancer Center, we used the MD Anderson Breast Cancer Management System (BCMS) database to identify 1,855 patients with a diagnosis of pure DCIS who were treated with surgery and/or radiotherapy between January 1996 and July 2009, had a minimum follow-up time of 1 year, and had documentation of height and weight at initial diagnosis of DCIS. The BCMS database contains detailed information on demographic, diagnostic, clinical, pathologic, treatment, and follow-up data. The aforementioned variables were analyzed with respect to the patient’s recorded BMI at initial diagnosis of DCIS, which was calculated as the prospectively recorded patient weight (kg) divided by the square of the patient height (m2). Patients were categorized into three following groups as classified by the Centers of Disease Control and Prevention: obese (BMI ≥ 30 kg/ m2); overweight (BMI 25 to ≤ 30 kg/m2); and normal weight or underweight (BMI ≤ 25 kg/m2) [5].
Although this weight-for-height measure is widely accepted in clinical settings, it is important to note some limitations of BMI in the assessment of patients. It does not calculate the distribution of fat in the body or differentiate between fat mass and lean body mass [12]. Patient self-reported diagnosis of diabetes and use of diabetic drugs was also recorded [13].
Statistical analyses
The Chi-square test was used to compare BMI groups with respect to categorical variables. The Kruskal–Wallis test was used to compare BMI groups with respect to continuous variables. Associations between clinical factors at diagnosis and BMI were analyzed using multivariate logistic regression, with controlling for the potential confounding effect of patient factors, tumor factors, and treatment-related variables. Odds ratios (ORs) and 95% CIs were generated for demographic and clinical characteristics and treatment variables. Significant associations identified by univariate analyses were utilized in the multivariate logistic regression model. The Hosmer–Lemeshow goodness-of-fit statistic was assessed to insure the validity of the model.
Two primary outcomes of interest—local–regional recurrence and development of contralateral breast cancer (in both outcomes, the recurrence and contralateral rates included DCIS and invasive events)—were evaluated using the Kaplan–Meier product-limit method, and differences in these outcomes were compared between BMI groups with the log-rank test. Local–regional recurrence was defined as ipsilateral local or regional recurrence, and if the patient did not experience recurrence, cases were censored at the time of last follow-up or death from any cause. For development of contralateral breast cancer, if the patient did not experience contralateral breast cancer, cases were censored at the time of last follow-up or death from any cause. Time to local or regional recurrence or development of contralateral breast cancer was defined from the date of surgery. All reported P values are two-sided, and P <0.05 was considered statistically significant. Analyses were performed using STATA/IC (release 11.1; StataCorp, College Station, TX) and STATISTICA (release 9.0; StatSoft, Inc., Tulsa, OK).
Results
Relationship between BMI and clinical and pathologic characteristics
Clinical, pathologic, and treatment characteristics of the 1,855 patients with DCIS are summarized in Table 1. Overall, 831 patients (44.8%) were of normal weight or underweight at diagnosis, and 1,024 patients (55.2%) were either overweight (N = 510, 27.5%) or obese (N = 514, 27.7%). African American patients were significantly more likely to be obese than were patients of other races (P <0.001). Of the 203 African American patients, 55.2% (N = 112) were obese. Postmenopausal patients were significantly more likely to be overweight or obese than were premenopausal patients (59.4 vs. 44.2%; P <0.001). In line with the findings regarding menopausal status, patients who were of normal weight or underweight were significantly younger (median age, 52 years) than overweight (median age, 55 years) and obese patients (median age, 57 years; P <0.001). First-degree family history of breast and/or ovarian cancer, use of hormone replacement therapy, and the presence of bilateral breast cancer at diagnosis were not significantly correlated with BMI (all P >0.05). Initial presentation was more likely to be a clinical symptom (mass or nipple discharge) rather than an imaging abnormality in patients who were of normal weight or underweight than in patients who were overweight or obese (17.3 vs. 11.9 vs. 13.6%, respectively, P = 0.022). The largest recorded mammographic DCIS dimension was similar among patients in the different BMI groups, but the largest recorded pathologic DCIS dimension was marginally higher among obese patients (median, 1.5 cm) than among normal/underweight (median, 1.2 cm) and overweight patients (median, 1.1 cm; P = 0.05). Patients who were of normal weight or underweight were significantly less likely than the combined group of overweight and obese patients to have grade I DCIS lesions (10.9 vs. 12.3%; P = 0.043). Obese patients were significantly more likely than normal/underweight patients to have necrosis (41.6 vs. 38.4%; P = 0.035). Normal/underweight patients were more likely than obese patients to have estrogen receptor (ER)-negative DCIS lesions (22.9 vs. 16.8%; P = 0.035).
Table 1.
Clinical, pathologic, and treatment characteristics by BMI group at DCIS diagnosis (N = 1,855; univariate analyses)
| Normal or underweight (BMI <25 kg/m2) Group 1 | Overweight (BMI 25 to <30 kg/m2) Group 2 | Obese (BMI ≥ 30 kg/m2) Group 3 | P, all groups | P, group 1 versus groups 2 and 3 | P, group 1 versus group 3 | |
|---|---|---|---|---|---|---|
| Number of patients | 831 (44.8) | 510 (27.5) | 514 (27.7) | |||
| Race | ||||||
| White | 661 (79.5) | 382 (74.9) | 329 (64.0) | |||
| African American | 36 (4.3) | 55 (10.1) | 112 (21.8) | |||
| Hispanic | 62 (7.5) | 41 (8.0) | 59 (11.5) | |||
| Asian/Pacific Islander | 65 (7.8) | 27 (5.3) | 7 (1.4) | |||
| Other | 7 (0.8) | 5 (1.0) | 7 (1.4) | 0.001 | 0.001 | 0.001 |
| Menopausal statusa | ||||||
| Premenopausal | 293 (35.3) | 117 (23.0) | 115 (22.5) | |||
| Postmenopausal | 538 (64.7) | 392 (77.0) | 397 (77.5) | <0.001 | <0.001 | 0.001 |
| Hormone replacement therapy | 259 (31.2) | 164 (32.2) | 153 (29.8) | 0.707 | 0.922 | 0.588 |
| Age | ||||||
| Median (range) | 52 years (18–89) | 55 years (29–88) | 57 years (31–82) | <0.001 | <0.001 | <0.001b |
| ≤40 years | 94 (11.3) | 32 (6.3) | 32 (6.2) | |||
| 41–50 years | 270 (32.5) | 138 (27.1) | 119 (23.2) | |||
| 51–60 years | 255 (30.7) | 166 (32.6) | 176 (34.2) | |||
| 61–70 years | 133 (16.0) | 112 (22.0) | 132 (25.7) | |||
| >70 years | 79 (9.5) | 62 (12.2) | 55 (10.7) | <0.001 | <0.001 | <0.001 |
| First-degree family history of breast and/or ovarian cancer | 172 (20.7) | 102 (20.0) | 125 (24.3) | 0.181 | 0.444 | 0.120 |
| Diabetic at diagnosis | 15 (1.8) | 27 (5.3) | 76 (14.8) | 0.001 | <0.001 | <0.001 |
| Receiving metformin | 5 (33.3) | 12 (44.4) | 45 (59.2) | 0.117 | 0.111 | 0.066 |
| Presence of bilateral breast cancer at diagnosis | 61 (7.3) | 42 (8.2) | 36 (7.0) | 0.737 | 0.822 | 0.817 |
| Initial presenting signsc | ||||||
| Clinical | 139 (17.3) | 59 (11.9) | 68 (13.6) | |||
| Radiologic | 667 (82.8) | 436 (88.1) | 431 (86.4) | 0.022 | 0.008 | 0.082 |
| Largest recorded mammographic dimension,d median (range) | 2.0 cm (0.05–14) | 2.0 cm (0.01–18) | 2.5 cm (0.3–10) | 0.641 | 0.591 | 0.380b |
| Largest recorded pathologic dimension,e median (range) | 1.2 cm (0.03–19) | 1.1 cm (0.01–12) | 1.5 cm (0.05–18) | 0.05 | 0.724 | 0.099b |
| Nuclear gradef | ||||||
| I | 66 (8.1) | 57 (11.3) | 53 (10.6) | |||
| II or III | 747 (91.9) | 448 (88.7) | 447 (89.4) | 0.120 | 0.043 | 0.128 |
| Necrosis present | 319 (38.4) | 190 (37.3) | 214 (41.6) | 0.319 | 0.640 | 0.035 |
| ER statusg | ||||||
| Positive | 356 (77.1) | 250 (81.7) | 272 (83.2) | |||
| Negative | 106 (22.9) | 56 (18.3) | 55 (16.8) | 0.077 | 0.027 | 0.035 |
| Surgery | ||||||
| BCS | 465 (56.0) | 307 (60.2) | 314 (61.1) | |||
| Mastectomy | 366 (44.0) | 203 (39.8) | 200 (38.9) | 0.120 | 0.042 | 0.064 |
| Immediate breast reconstruction (in patients who underwent mastectomy) | 267 (73.0) | 127 (62.6) | 105 (52.5) | <0.001 | <0.001 | <0.001 |
| Contralateral prophylactic mastectomy | 77 (9.3) | 36 (7.1) | 31 (6.0) | 0.077 | 0.029 | 0.034 |
| Adjuvant tamoxifen (in all patients) | 298 (35.9) | 189 (37.1) | 192 (37.4) | 0.832 | 0.549 | 0.580 |
| Adjuvant tamoxifen (in patients who underwent BCS) | 208 (44.7) | 134 (43.6) | 138 (44.0) | 0.968 | 0.808 | 0.871 |
| Adjuvant tamoxifen (in patients who underwent mastectomy) | 90 (24.6) | 55 (27.1) | 54 (27.0) | 0.764 | 0.464 | 0.553 |
| Adjuvant radiotherapy (in patients who underwent BCS) | 353 (75.9) | 247 (80.5) | 258 (82.2) | 0.063 | 0.022 | 0.029 |
| Adjuvant tamoxifen and radiotherapy (in patients who underwent BCS) | 173 (68.7) | 118 (72.8) | 125 (74.4) | 0.398 | 0.187 | 0.203 |
P values are from the Chi-square test unless otherwise indicated
ER estrogen receptor, BCS breast-conserving surgery
Two patients were perimenopausal; menopausal status was not available for one patient
Calculated using the Kruskal–Wallis test
Initial presenting signs were not determined in 55 patients; clinical refers to presentation with a mass or nipple discharge
Largest mammographic dimension recorded in 870 patients
Largest pathologic size recorded in 1,442 patients
Nuclear grade not available for 37 patients
ER status determined in 1,095 patients
Clinical and pathologic characteristics that were associated (significant or non-significant) of being overweight or obese on multivariate logistic regression are shown in Table 2. Of these characteristics, race [African American (OR = 3.93; CI = 2.66–5.80) or Hispanic (OR = 1.44, CI = 1.02–2.04)], post-menopausal status (OR = 1.63, CI = 1.28–2.07), diagnosis of diabetes (OR = 4.60, CI = 2.60–8.12), presentation with a radiographic abnormality versus a clinical symptom (OR = 1.35, CI = 1.01–1.80), and ER-positive DCIS (OR = 1.39, CI = 1.00–1.92) were independent predictors of being overweight or obese.
Table 2.
Multivariate logistic regression model of clinical, pathologic, and treatment characteristics associated with being overweight or obese at DCIS diagnosis (N = 1,855)
| Variable | OR | 95% CI | P |
|---|---|---|---|
| Race | |||
| White | 1.00 (reference) | ||
| African American | 3.93 | 2.66–5.80 | <0.001 |
| Hispanic | 1.44 | 1.02–2.04 | 0.039 |
| Asian/Pacific islander | 0.54 | 0.36–0.80 | 0.003 |
| Menopausal status | |||
| Premenopausal | 1.00 (reference) | ||
| Postmenopausal | 1.63 | 1.28–2.07 | <0.001 |
| Age, years | |||
| <40 | 1.00 (reference) | ||
| 40–70 | 1.27 | 0.83–1.95 | 0.273 |
| >70 | 1.16 | 0.67–1.98 | 0.600 |
| Diabetic at diagnosis | |||
| No | 1.00 (reference) | ||
| Yes | 4.60 | 2.60–8.12 | <0.001 |
| Initial presenting signs | |||
| Clinical | 1.00 (reference) | ||
| Radiologic | 1.35 | 1.01–1.80 | 0.040 |
| Largest recorded pathologic dimension (cm) | |||
| <1 | 1.00 (reference) | ||
| ≥1 | 0.95 | 0.74–1.21 | 0.657 |
| Nuclear grade | |||
| I | 1.00 (reference) | ||
| II or III | 0.72 | 0.51–1.02 | 0.067 |
| Necrosis | |||
| Present | 1.00 (reference) | ||
| Absent | 0.88 | 0.71–1.09 | 0.257 |
| ER status | |||
| Negative | 1.00 (reference) | ||
| Positive | 1.39 | 1.00–1.92 | 0.042 |
| Immediate breast reconstruction (in patients with mastectomy) | |||
| Yes | 1.00 (reference) | ||
| No | 1.58 | 1.13–2.20 | 0.007 |
| Contralateral prophylactic mastectomy | |||
| Yes | 1.00 (reference) | ||
| No | 1.06 | 0.73–1.55 | 0.764 |
| Adjuvant radiotherapy (in patients who underwent BCS) | |||
| Yes | 1.00 (reference) | ||
| No | 0.78 | 0.58–1.05 | 0.102 |
OR odds ratio, BMI body mass index, ER estrogen receptor
Relationship between BMI and Treatments for DCIS
The combined groups of obese and overweight patients were more likely than normal/underweight patients to undergo breast-conserving surgery (BCS) (60.6 vs. 56.0%; P = 0.042, Table 1). Among patients who underwent BCS, the use of adjuvant radiotherapy was significantly more common among obese patients than normal/underweight patients (82.2 vs. 75.6%; P = 0.029) and the combined group of overweight and obese patients than normal/underweight patients (81.3 vs. 75.6%; P = 0.022). Among patients who underwent mastectomy, the use of immediate breast reconstruction was significantly more common in normal/underweight patients (73.0%) than in overweight (62.6%) and obese patients (52.5%; P <0.001). Contralateral prophylactic mastectomy was more likely in normal/underweight patients (9.3%) than in obese patients (6.0%) and the combination of overweight and obese patients (6.5%, P = 0.029). There were no significant differences in the use of adjuvant tamoxifen by BMI group. On multivariate logistic regression, the only treatment characteristic that was an independent predictor of being overweight or obese was immediate breast reconstruction (OR = 1.58; P = 0.007; Table 2).
Relationship between BMI and risk of local–regional recurrence or development of contralateral breast cancer
At a median follow-up of 4.96 years (range, 1.0–14.34), 45 patients (4.1%) had had a local–regional recurrence following BCS, for an overall 5-year local–regional recurrence rate of 4.14%. Among the 45 patients with local–regional recurrence, the recurrence was DCIS in 27 patients (60%) and invasive disease in 13 patients (40%) (Table 3). The 5-year rates of local–regional recurrence among patients in the 3 BMI categories by type of adjuvant therapy received are shown in Table 3. Within subgroups of patients treated with identical adjuvant therapies, no significant differences were observed in local–regional recurrence rates by BMI category (Table 3). Logistic regression analysis was also conducted to identify risks of recurrence for being overweight/obese compared to normal weight/underweight, and there were no significant differences found between BMI categories after adjusting for race, menopausal status, age, pathologic size, grade, necrosis, surgical procedure, and use of adjuvant therapy (OR, 1.87; 95% CI, 0.43–8.18; P = 0.47).
Table 3.
Local–regional recurrence and development of contralateral breast cancer by BMI group at DCIS diagnosis
| Normal or Underweight (BMI <25 kg/m2); Group 1 | Overweight (BMI 25 to <30 kg/m2); Group 2 | Obese (BMI ≥ 30 kg/ m2); Group 3 | P, All Groups | P, Group 1 versus Groups 2 and 3 | P, Group 1 versus Group 3 | |
|---|---|---|---|---|---|---|
| LRR, no. of patients (%) | ||||||
| Total | 18 (3.9) | 12 (3.9) | 15 (4.8) | |||
| DCIS | 12 (66.7) | 8 (66.7) | 7 (46.7) | |||
| Invasive | 6 (33.3) | 4 (33.3) | 8 (53.3) | 0.435 | 0.456 | 0.247 |
| 5-year LRR rate | ||||||
| BCS with radiotherapy (n = 858) | 2.2% | 2.1% | 3.2% | 0.395 | 0.417 | 0.218 |
| BCS without radiotherapy (n = 230) | 6.1% | 11.6% | 9.7% | 0.648 | 0.377 | 0.601 |
| BCS without radiotherapy without tamoxifen (n = 166) | 8.7% | 12.9% | 2.9% | 0.658 | 0.857 | 0.626 |
| BCS with radiotherapy without tamoxifen (n = 442) | 3.7% | 3.2% | 3.7% | 0.376 | 0.388 | 0.205 |
| BCS with radiotherapy with tamoxifen (n = 416) | 0.6% | 1.0% | 2.8% | 0.997 | 0.964 | 0.954 |
| Development of contralateral breast cancer, no. of patients (%)a | ||||||
| Total | 25 (3.7) | 16 (3.9) | 23 (5.4) | |||
| DCIS | 15 (57.7) | 7 (43.8) | 10 (43.5) | |||
| Invasive | 11 (42.3) | 9 (56.2) | 13 (56.5) | 0.538 | 0.265 | 0.321 |
| 5-year rate of development of contralateral breast cancera | ||||||
| Overall | 3.2% | 4.0% | 5.0% | 0.213 | 0.186 | 0.084 |
| In patients taking tamoxifen (n = 566) | 3.8% | 1.2% | 5.0% | 0.111 | 0.588 | 0.545 |
| In patients not taking tamoxifen (n = 945) | 2.8% | 5.7% | 5.1% | 0.162 | 0.057 | 0.086 |
BCS breast-conserving surgery
Excludes patients who had contralateral prophylactic mastectomy, bilateral breast cancer at diagnosis, and history of contralateral breast cancer
A total of 64 patients (3.5%) developed contralateral breast cancer during the study period, for an overall 5-year rate of contralateral breast cancer development of 3.9% (Table 3). Among the 64 patients who developed contra-lateral breast cancer, the contralateral cancer was DCIS in 32 patients (50%) and invasive cancer in 32 patients (50%). No significant differences were detected in the rate of contralateral breast cancer development based on BMI category or if the patient received adjuvant tamoxifen (Table 3). There was, however, a trend toward a higher 5-year rate of contralateral breast cancer development among overweight (5.7%) and obese patients (5.1%) not taking tamoxifen compared with normal/underweight patients not taking tamoxifen (2.8%; P = 0.057). This trend was not seen among patients taking tamoxifen (P = 0.588).
DCIS among diabetic patients with and without metformin use
As type II diabetes is known to be strongly associated with obesity, clinical and pathologic characteristics and outcome were evaluated with respect to the presence or absence of diabetes at diagnosis of DCIS. There were 118 patients (6%) in the study who were diabetic. Compared to the patients without diabetes, the diabetic patients were significantly older; were more likely to be obese, of African American descent, and postmenopausal; had larger pathologic tumor size; and were more likely to be treated with adjuvant radiotherapy (all P <0.05; Table 4). No differences were noted in patients with and without diabetes with respect to nuclear grade, presence of necrosis, ER status, or use of adjuvant tamoxifen therapy (all P >0.05). Four diabetic patients developed a local–regional recurrence or contralateral breast cancer during follow-up. No significant differences were detected in local–regional recurrences or contralateral breast cancer development between patients with and without diabetes (P >0.05).
Table 4.
Characteristics and outcomes of patients with DCIS by diabetes status and metformin use status at DCIS diagnosis
| Without diabetes | With diabetes, receiving metformin | With diabetes, Not receiving metformin | P, without versus with diabetes | P, receiving metformin versus not | |
|---|---|---|---|---|---|
| Number of patients | 1,737 (93.6) | 62 (3.3) | 56 (3.0) | ||
| BMI | |||||
| Normal or underweight (BMI <25 kg/m2) | 816 (47.0) | 5 (8.1) | 10 (17.9) | ||
| Overweight (BMI 25 to <30 kg/m2) | 483 (27.8) | 12 (19.4) | 15 (26.8) | ||
| Obese (BMI ≥ 30 kg/m2) | 438 (25.2) | 45 (72.6) | 31 (55.4) | <0.001 | 0.117 |
| Race | |||||
| White | 1,310 (75.4) | 30 (48.4) | 32 (57.1) | ||
| African American | 171 (9.8) | 19 (30.7) | 13 (23.2) | ||
| Hispanic | 147 (8.5) | 9 (14.5) | 6 (10.7) | ||
| Asian/Pacific Islander | 90 (5.2) | 4 (6.5) | 5 (8.9) | ||
| Other | 19 (1.0) | 0 (0) | 0 (0) | <0.001 | 0.659 |
| Menopausal status | |||||
| Premenopausal | 510 (29.4) | 8 (12.9) | 7 (12.5) | ||
| Postmenopausal | 1,224 (70.6) | 54 (87.1) | 49 (87.5) | <0.001 | 0.948 |
| Hormonal replacement therapy use | 539 (31.0) | 18 (29.0) | 19 (33.9) | 0.941 | 0.567 |
| Age, median (range) | 54 years (18–89) | 61 years (37–79) | 58.5 years (11–82) | <0.001 | 0.736 |
| Largest recorded mammographic dimension, median (range) | 2.5 cm (0.05–18) | 2.4 cm (0.3–11) | 2.1 cm (0.07–10) | 0.804 | 0.901 |
| Largest recorded pathologic size, median (range) | 1.2 cm (0.01–19) | 1.5 cm (0.09–17) | 1.7 cm (0.05–12) | 0.012 | 0.342 |
| Nuclear grade | |||||
| I | 163 (9.6) | 6 (10.0) | 7 (13.0) | ||
| II or III | 1,541 (90.4) | 54 (90.0) | 47 (87.0) | 0.521 | 0.619 |
| Necrosis present | 672 (38.7) | 30 (48.4) | 21 (37.5) | 0.329 | 0.233 |
| ER status | |||||
| Positive | 814 (80.0) | 35 (85.4) | 29 (80.6) | ||
| Negative | 204 (20.0) | 6 (14.6) | 7 (19.4) | 0.503 | 0.574 |
| Adjuvant tamoxifen (in all patients) | 631 (36.3) | 28 (45.2) | 20 (35.7) | 0.342 | 0.297 |
| Adjuvant radiotherapy (in patients who underwent BCS) | 799 (78.2) | 33 (89.2) | 26 (89.7) | 0.031 | 0.951 |
| Local–regional recurrence (in patients who underwent BCS) | |||||
| Total | 44 (4.3) | 1 (2.7) | 0 (0) | 0.605 | 0.376 |
| DCIS | 26 (59.1) | 1 (100.0) | 0 (0) | ||
| Invasive | 18 (40.9) | 0 (0) | 0 (0) | ||
| Development of contralateral breast cancera | |||||
| Total | 61 (4.3) | 0 (0) | 3 (6.0) | 0.856 | 0.145 |
| DCIS | 30 (48.4) | 0 (0) | 2 (66.7) | ||
| Invasive | 32 (51.6) | 0 (0) | 1 (33.3) | ||
NA not applicable (a test was not possible because no failures occurred), BCS breast-conserving surgery
Excludes patients who had contralateral prophylactic mastectomy, bilateral breast cancer at diagnosis, and history of contralateral breast cancer
The relationship between metformin use and clinical and pathologic characteristics and outcome was also explored [9–11]. Of the 118 patients with diabetes, 62 (53%) were taking metformin at diagnosis of DCIS, and 56 (48%) were not (Table 4). No significant differences were noted with respect to clinical and pathologic features, local–regional recurrence, and development of contralateral breast cancer between patients with DCIS taking and not taking metformin (P >0.05).
Discussion
Multiple published studies have documented overall worse biologic features, prognosis, and response to therapy for women with large body size (those who are overweight or obese) than for women with lean body size (those who are of normal weight or underweight) among patients with invasive breast cancer. Our study was undertaken to explore whether large body size was associated with similar effects in women with DCIS, a nonobligate precursor of invasive breast cancer and/or a marker for increased risk of development of invasive disease. In this cohort study of 1,855 women diagnosed with pure DCIS, 55.2% of patients had large body size, and large body size was independently associated with African American or Hispanic race, post-menopausal status, a diagnosis of diabetes, presentation of disease by radiologic abnormality as opposed to clinical findings, and having ER-positive disease. Large body size was not, however, associated with known adverse features of DCIS, including larger tumor size, higher nuclear grade, or the presence of necrosis. The 5-year local–regional recurrence rate was not significantly higher among women with large body size after stratification for type of therapy received. However, among women not taking tamoxifen, there was a trend toward increased rates of development of contralateral breast cancer among women with large body size compared with women of lean body size, an effect that has previously been demonstrated among patients with invasive breast cancer [14].
Our finding of an increased hazard of developing ER-positive DCIS among women with large body size is likely attributable to increased circulating estrogens related to an increased mass of adipose tissue and upregulation of aromatase seen in obese women [15, 16]. The same mechanism may also underlie the trend toward an increased risk of development of contralateral breast cancer among patients not taking tamoxifen in this study. Why then did we not see an increased rate of local–regional recurrence among patients with large body size compared with patients with lean body size? This may be because the overwhelming majority of patients presented with early, mammographically detected disease that was completely excised. Furthermore, tamoxifen and radiotherapy were utilized in a large proportion of patients. The use of adjuvant tamoxifen in patients with DCIS has a small protective effect against ipsilateral breast cancer recurrence and contralateral breast cancer [17]. However, the potential therapeutic benefit of tamoxifen use in women with large body size and DCIS must be carefully weighed against the potential risks, including thromboembolic events and endometrial carcinoma, which occur more commonly in obese women [14]. In the current study, rates of use of tamoxifen following surgery for DCIS were nearly identical among the different BMI categories. Although aromatase inhibitors are not currently being utilized in the management of DCIS as trials evaluating the efficacy of this intervention have yet to be analyzed and published, there is now evidence that anastrozole is less effective than tamoxifen in preventing reoccurrences in obese women with invasive breast cancer [18].
In addition to being the first report to assess the relationship between BMI and clinical and pathologic factors in patients with DCIS, this is also to our knowledge the first report to evaluate the relationship between BMI and type of surgery utilized for DCIS. Patients with large body size were significantly less likely to undergo mastectomy, immediate breast reconstruction following mastectomy, and contralateral prophylactic mastectomy. Although surgical decision-making and postoperative complications were not evaluated in the present study, obesity has been shown to be an independent predictor of wound and flap complications after breast surgery in general and specifically after immediate breast reconstruction and therefore may have played a role in procedure selection [19–21]. Nevertheless, even though the rate of immediate reconstruction was significantly lower in women of large body size (57.6%) than in women of lean body size, the rate was substantially higher than the overall rates of immediate reconstruction in the United States, which have been reported to be as low as 15% [22].
Approximately 10% of large-size women in this study also had diabetes. There is a complex relationship between obesity, insulin and insulin resistance, IGF-1, and potential adverse effects on cellular proliferation and angiogenesis that may impact carcinogenesis and progression of disease through direct pathways and cross-talk with estrogenic and altered adipokine and cytokine pathways [23]. The relationship between metformin use and clinical and pathologic characteristics of DCIS and outcome of DCIS was explored as metformin has been shown to be a potential anticancer agent in preclinical, epidemiologic, and clinical breast cancer studies [9–11]. No differences in adverse features of DCIS or risk of local–regional recurrence were noted between diabetic patients taking and not taking metformin. The lack of association between metformin and adverse features and prognosis of DCIS may be due to the relatively small sample size, or the drug may have no detectable efficacy in DCIS. In this regard, it is important to note that different associations of metformin with cancer may exist in nondiabetic subjects who have not had similar long-term exposure to hyperinsulinemia [10]. Metformin as a potential chemopreventive agent or as an adjuvant treatment for breast cancer may prove to be of more value in nondiabetic women, although this is yet to be proven [9, 10].
The use of BMI in clinical settings is ubiquitous; however, there are several limitations that clinicians need to be aware of. BMI is commonly used to measure adiposity; however, it does not accurately calculate the distribution of fat in the body or differentiate between fat mass and lean body mass. There are other established methods used by investigators to address this concern. Waist-to-height ratio (WHtR) and waist-to-hip ratio (WHR) are commonly used to assess the body fat distribution and distinguish between both fat mass and muscle mass [24]. However, despite the many different methods available, there has been no consensus in the field that measure adiposity accurately [25]. Because of the nature of our study, we were unable to collect WHtR and WHR at the initial diagnosis of DCIS. A number of recent studies in the field have elucidated the role of weight loss after diagnosis of invasive breast cancer, suggesting an advantage of losing weight and leading to being disease-free and global survival [26–28]. The potential impact of weight loss on local recurrence following a diagnosis of DCIS is yet to be studied.
There may be fundamental differences in DCIS and invasive breast cancer that account for the finding that large body size does not adversely affect biologic features or prognosis of patients with DCIS. In this regard, it is interesting that large body size does not appear to be related to an increased risk of development of in situ cancers, regardless of menopausal status, but shows an unambiguous relationship with increased risk of development of postmenopausal invasive breast cancer [29–31]. It is possible that a larger sample size or much longer follow-up times may be needed to demonstrate that large body size is an independent adverse prognostic feature in DCIS. The present study has several strengths, including prospective recording of BMI at diagnosis of DCIS, the large dataset, and treatment of all patients at a single institution using standardized diagnostic imaging, pathology, and multidisciplinary clinical practice protocols. Considered in light of these strengths, the findings of this study suggest that large body size is unlikely to have a significant independent prognostic impact on present-day patients with imaging-detected DCIS.
Acknowledgments
Stephanie P. Deming of MD Anderson’s Department of Scientific Publications provided editorial assistance. This research is supported by National Institutes of Health through MD Anderson’s Cancer Center Support Grant, CA016672 and Randalls Food Markets.
Footnotes
Conflict of interest The authors indicate no potential conflict of interest.
Contributor Information
Henry M. Kuerer, Email: hkuerer@mdanderson.org, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA
Sara A. Lari, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA
Banu K. Arun, Department of Breast Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
Chung-Yuan Hu, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA.
Abenaa Brewster, Department of Clinical Cancer Prevention, The University of Texas MD Anderson Cancer Center, Houston, TX, USA.
Elizabeth A. Mittendorf, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA
Constance T. Albarracin, Department of Pathology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
Gildy V. Babiera, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA
Abigail S. Caudle, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA
Jamie L. Wagner, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA
Jennifer K. Litton, Department of Breast Medical Oncology, The University of Texas MD Anderson Cancer Center, Houston, TX, USA
Isabelle Bedrosian, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA.
Funda Meric-Bernstam, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA.
Anthony Lucci, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA.
Kelly K. Hunt, Department of Surgical Oncology, DCIS Discovery Enterprise, Unit 1484, The University of Texas MD Anderson Cancer Center, 1400 Pressler Street, Houston, TX 77230-1402, USA
References
- 1.Flegal KM, Carroll MD, Ogden CL, Curtin LR. Prevalence and trends in obesity among US adults, 1999–2008. JAMA. 2010;303:235–241. doi: 10.1001/jama.2009.2014. [DOI] [PubMed] [Google Scholar]
- 2.Basen-Engquist K, Chang M. Obesity and cancer risk: recent review and evidence. Curr Oncol Rep. 2011;13:71–76. doi: 10.1007/s11912-010-0139-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Calle EE, Rodriguez C, Walker-Thurmond K, Thun MJ. Overweight, obesity, and mortality from cancer in a prospectively studied cohort of U.S. adults. N Engl J Med. 2003;348:1625–1638. doi: 10.1056/NEJMoa021423. [DOI] [PubMed] [Google Scholar]
- 4.Chen X, Lu W, Zheng W, Gu K, Chen Z, Zheng Y, Shu XO. Obesity and weight change in relation to breast cancer survival. Breast Cancer Res Treat. 2010;122:823–833. doi: 10.1007/s10549-009-0708-3. 10.1007/s 10549-009-0708-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Dawood S, Broglio K, Gonzalez-Angulo AM, Kau SW, Islam R, Hortobagyi GN, Cristofanilli M. Prognostic value of body mass index in locally advanced breast cancer. Clin Cancer Res. 2008;14:1718–1725. doi: 10.1158/1078-0432.CCR-07-1479. [DOI] [PubMed] [Google Scholar]
- 6.de Azambuja E, McCaskill-Stevens W, Francis P, Quinaux E, Crown JP, Vicente M, Giuliani R, Nordenskjold B, Gutierez J, Andersson M, Vila MM, Jakesz R, Demol J, Dewar J, Santoro A, Lluch A, Olsen S, Gelber RD, Di Leo A, Piccart-Gebhart M. The effect of body mass index on overall and disease-free survival in node-positive breast cancer patients treated with docetaxel and doxorubicin-containing adjuvant chemotherapy: the experience of the BIG 02–98 trial. Breast Cancer Res Treat. 2010;119:145–153. doi: 10.1007/s10549-009-0512-0. [DOI] [PubMed] [Google Scholar]
- 7.Ewertz M, Jensen MB, Gunnarsdottir KA, Hojris I, Jakobsen EH, Nielsen D, Stenbygaard LE, Tange UB, Cold S. Effect of obesity on prognosis after early-stage breast cancer. J Clin Oncol. 2011;29:25–31. doi: 10.1200/JCO.2010.29.7614. [DOI] [PubMed] [Google Scholar]
- 8.Loi S, Milne RL, Friedlander ML, McCredie MR, Giles GG, Hopper JL, Phillips KA. Obesity and outcomes in pre-menopausal and postmenopausal breast cancer. Cancer Epidemiol Biomarkers Prev. 2005;14:1686–1691. doi: 10.1158/1055-9965.EPI-05-0042. [DOI] [PubMed] [Google Scholar]
- 9.Cazzaniga M, Bonanni B, Guerrieri-Gonzaga A, Decensi A. Is it time to test metformin in breast cancer clinical trials? Cancer Epidemiol Biomarkers Prev. 2009;18:701–705. doi: 10.1158/1055-9965.EPI-08-0871. 10.1158/ 1055-9965.EPI-08-0871. [DOI] [PubMed] [Google Scholar]
- 10.Goodwin PJ, Stambolic V, Lemieux J, Chen BE, Parulekar WR, Gelmon KA, Hershman DL, Hobday TJ, Ligibel JA, Mayer IA, Pritchard KI, Whelan TJ, Rastogi P, Shepherd LE. Evaluation of metformin in early breast cancer: a modification of the traditional paradigm for clinical testing of anti-cancer agents. Breast Cancer Res Treat. 2011;126:215–220. doi: 10.1007/s10549-010-1224-1. [DOI] [PubMed] [Google Scholar]
- 11.Hadad S, Iwamoto T, Jordan L, Purdie C, Bray S, Baker L, Jellema G, Deharo S, Hardie DG, Pusztai L, Moulder-Thompson S, Dewar JA, Thompson AM. Evidence for biological effects of metformin in operable breast cancer: a pre-operative, window-of-opportunity, randomized trial. Breast Cancer Res Treat. 2011;128:783–794. doi: 10.1007/s10549-011-1612-1. [DOI] [PubMed] [Google Scholar]
- 12.Daniels SR. The use of BMI in the clinical setting. Pediatrics. 2009;124(Suppl 1):S35–S41. doi: 10.1542/peds.2008-3586F. [DOI] [PubMed] [Google Scholar]
- 13.Margolis KL, Lihong Q, Brzyski R, Bonds DE, Howard BV, Kempainen S, Simin L, Robinson JG, Safford MM, Tinker LT, Phillips LS. Validity of diabetes self-reports in the Women’s Health Initiative: comparison with medication inventories and fasting glucose measurements. Clin Trials. 2008;5:240–247. doi: 10.1177/1740774508091749. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14.Dignam JJ, Wieand K, Johnson KA, Fisher B, Xu L, Mamounas EP. Obesity, tamoxifen use, and outcomes in women with estrogen receptor-positive early-stage breast cancer. J Natl Cancer Inst. 2003;95:1467–1476. doi: 10.1093/jnci/djg060. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Key TJ, Appleby PN, Reeves GK, Roddam A, Dorgan JF, Longcope C, Stanczyk FZ, Stephenson HE, Jr, Falk RT, Miller R, Schatzkin A, Allen DS, Fentiman IS, Wang DY, Dowsett M, Thomas HV, Hankinson SE, Toniolo P, Akhmedkhanov A, Koenig K, Shore RE, Zeleniuch-Jacquotte A, Berrino F, Muti P, Micheli A, Krogh V, Sieri S, Pala V, Venturelli E, Secreto G, Barrett-Connor E, Laughlin GA, Kabuto M, Akiba S, Stevens RG, Neriishi K, Land CE, Cauley JA, Kuller LH, Cummings SR, Helzlsouer KJ, Alberg AJ, Bush TL, Comstock GW, Gordon GB, Miller SR. Body mass index, serum sex hormones, and breast cancer risk in postmenopausal women. J Natl Cancer Inst. 2003;95:1218–1226. doi: 10.1093/jnci/djg022. [DOI] [PubMed] [Google Scholar]
- 16.Rose DP, Vona-Davis L. Influence of obesity on breast cancer receptor status and prognosis. Expert Rev Anticancer Ther. 2009;9:1091–1101. doi: 10.1586/era.09.71. [DOI] [PubMed] [Google Scholar]
- 17.Kuerer HM. Rational individualised selection of adjuvant therapy for ductal carcinoma in situ. Lancet Oncol. 2011;12:2–3. doi: 10.1016/S1470-2045(10)70277-1. [DOI] [PubMed] [Google Scholar]
- 18.Sestak I, Distler W, Forbes JF, Dowsett M, Howell A, Cuzick J. Effect of body mass index on recurrences in tamoxifen and anastrozole treated women: an exploratory analysis from the ATAC trial. J Clin Oncol. 2010;28:3411–3415. doi: 10.1200/JCO.2009.27.2021. 10.1200/JCO. 2009.27.2021. [DOI] [PubMed] [Google Scholar]
- 19.El-Tamer MB, Ward BM, Schifftner T, Neumayer L, Khuri S, Henderson W. Morbidity and mortality following breast cancer surgery in women: national benchmarks for standards of care. Ann Surg. 2007;245:665–671. doi: 10.1097/01.sla.0000245833.48399.9a. 10.1097/01.sla.0000245833. 48399.9a. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Chang DW, Wang B, Robb GL, Reece GP, Miller MJ, Evans GR, Langstein HN, Kroll SS. Effect of obesity on flap and donor-site complications in free transverse rectus abdominis myocutaneous flap breast reconstruction. Plast Reconstr Surg. 2000;105:1640–1648. doi: 10.1097/00006534-200004050-00007. [DOI] [PubMed] [Google Scholar]
- 21.McCarthy CM, Mehrara BJ, Riedel E, Davidge K, Hinson A, Disa JJ, Cordeiro PG, Pusic AL. Predicting complications following expander/implant breast reconstruction: an outcomes analysis based on preoperative clinical risk. Plast Reconstr Surg. 2008;121:1886–1892. doi: 10.1097/PRS.0b013e31817151c4. [DOI] [PubMed] [Google Scholar]
- 22.Alderman AK, McMahon L, Wilkins EG. The national utilization of immediate and early delayed breast reconstruction and the effect of sociodemographic factors. Plast Reconstr Surg. 2003;111:695–703. doi: 10.1097/01.Prs.0000041438.50018.02. [DOI] [PubMed] [Google Scholar]
- 23.Sinicrope FA, Dannenberg AJ. Obesity and breast cancer prognosis: weight of the evidence. J Clin Oncol. 2011;29:4–7. doi: 10.1200/JCO.2010.32.1752. [DOI] [PubMed] [Google Scholar]
- 24.Canchola AJ, Anton-Culver H, Bernstein L, Clarke CA, Henderson K, Ma H, Ursin G, Horn-Ross PL. Body size and the risk of postmenopausal breast cancer subtypes in the California Teachers Study cohort. Cancer Causes Control. 2012 doi: 10.1007/s10552-012-9897-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Hartz A, He T, Rimm A. Comparison of adiposity measures as risk factors in postmenopausal women. J Clin Endocrinol Metab. 2012;97:227–233. doi: 10.1210/jc.2011-1151. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.McTiernan A, Irwin M, Vongruenigen V. Weight, physical activity, diet, and prognosis in breast and gynecologic cancers. J Clin Oncol. 2010;28:4074–4080. doi: 10.1200/JCO.2010.27.9752. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Harvie M, Howell A, Vierkant RA, Kumar N, Cerhan JR, Kelemen LE, Folsom AR, Sellers TA. Association of gain and loss of weight before and after menopause with risk of post-menopausal breast cancer in the Iowa women’s health study. Cancer Epidemiol Biomarkers Prev. 2005;14:656–661. doi: 10.1158/1055-9965.EPI-04-0001. 10.1158/ 1055-9965.EPI-04-0001. [DOI] [PubMed] [Google Scholar]
- 28.Eliassen AH, Colditz GA, Rosner B, Willett WC, Hankinson SE. Adult weight change and risk of postmenopausal breast cancer. JAMA. 2006;296:193–201. doi: 10.1001/jama.296.2.193. [DOI] [PubMed] [Google Scholar]
- 29.Kerlikowske K, Barclay J, Grady D, Sickles EA, Ernster V. Comparison of risk factors for ductal carcinoma in situ and invasive breast cancer. J Natl Cancer Inst. 1997;89:76–82. doi: 10.1093/jnci/89.1.76. [DOI] [PubMed] [Google Scholar]
- 30.Reinier KS, Vacek PM, Geller BM. Risk factors for breast carcinoma in situ versus invasive breast cancer in a prospective study of pre- and post-menopausal women. Breast Cancer Res Treat. 2007;103:343–348. doi: 10.1007/s10549-006-9375-9. [DOI] [PubMed] [Google Scholar]
- 31.Trentham-Dietz A, Newcomb PA, Storer BE, Remington PL. Risk factors for carcinoma in situ of the breast. Cancer Epidemiol Biomarkers Prev. 2000;9:697–703. [PubMed] [Google Scholar]