Abstract
Background:
Multiple studies have demonstrated a link between obesity and breast cancer. The potential association between obesity and atypical high-risk breast lesions, however, has not been well characterized. We sought to evaluate the characteristics and clinical outcomes of patients with breast atypia based on a woman’s body mass index (BMI).
Methods:
We retrospectively identified adult women diagnosed with atypical ductal hyperplasia (ADH), atypical lobular hyperplasia (ALH), and/or lobular carcinoma in situ (LCIS) at a single institution from 2008–2017. BMI groups were defined as a BMI 18.5 to <30 or BMI ≥30 (obese). Adjusted logistic regression was used to estimate the association of BMI group with (1) the odds of upstage to cancer after atypia on needle biopsy and (2) odds of subsequent diagnosis of breast cancer.
Results:
Breast atypia was identified in 503 patients (most advanced atypia: 74.8% ADH, 4.6% ALH, 20.7% LCIS), and 41% of these patients were classified as obese. After adjustment, BMI group was not associated with upstage to breast cancer at surgical excision following needle biopsy (p=0.16), nor development of a subsequent breast cancer (p=0.08). For those upstaged to breast cancer at the time of surgical excision, or those who developed a subsequent malignancy, tumor subtype, grade and stage were not associated with BMI group (p>0.05).
Conclusion:
In a large cohort of patients diagnosed with atypical breast histology, the risk of upstaging and/or subsequent progression to a breast malignancy was not associated with BMI. Factors other than obesity may influence breast cancer risk.
INTRODUCTION
Obesity is an epidemic, with over 13% of the adult population worldwide and 42% of Americans considered obese.1,2 The National Institutes of Health (NIH), has defined four body mass index (BMI) categories: underweight <18.5, normal weight 18.5–24.9, overweight 25–29.9, and obese ≥30 kg/m2.3 It is well known that obesity is associated with many disease states, and some studies have shown that obesity is associated with higher risks of adenocarcinoma, including breast cancer, in post-menopausal women.4 Since breast cancer continues to be the most common non-cutaneous malignancy amongst American women, with the average risk being 1 in 8 or 13% in the woman’s lifetime, understanding how obesity potentially impacts this disease is highly relevant.5
Studies have shown that high-risk breast lesions increased the risk of breast cancer. Benign, high-risk breast lesions include atypical ductal hyperplasia (ADH), atypical lobular hyperplasia (ALH), and lobular carcinoma in situ (LCIS).6–8 Atypical hyperplasia (AH) is found in approximately 10% of benign breast biopsies 9 with equivalent incidence of ductal and lobular histology. Patient age, number of foci and degree of involution are factors associated with risk for the development of malignancy.10 AH confers an absolute risk of 30% at 25 years of follow up. Patients diagnosed with lobular neoplasia (ALH or LCIS) are at an increased lifetime risk of developing breast cancer, with a 1% per year for ALH and approximately 2% per year for LCIS, with both breasts at increased risk. Depending on the type of lesion and concordance with imaging, surgical excision is often recommended.11 In particular, the pathologic upgrade of ADH at the time of surgical excision to ductal carcinoma in situ (DCIS) or invasive carcinoma, may be as high as 20% or more.12 For women with LCIS, the lifetime risk of breast cancer may approach 40%,7,13 whereas women with ADH or ALH have an estimated lifetime risk of breast cancer approaching 20–30%.6,14,15 Although several studies have looked at the potential association between high-risk lesions and certain factors such as age, race, and ethnicity,8,16 none, to our knowledge have reviewed the possible association with BMI. Given the correlation between breast cancer and obesity, we hypothesize that obesity may also influence outcomes for patients with atypical breast lesions. Therefore, we sought to evaluate the characteristics and clinical outcomes of women with breast atypia based on BMI.
METHODS
Adult women diagnosed with breast atypia on pathology at a major academic institution from 2008–2017 were included in this study. Patients were categorized as having ADH, ALH, and/or classic LCIS. Patients could have several atypia diagnoses, and the first and most advanced diagnoses were included for analysis. Patients with a history of breast cancer (DCIS or invasive carcinoma) prior to their atypia diagnosis were excluded. BMI was recorded from the initial consult note, which typically occurred within weeks of the biopsy diagnosis. BMI groups were defined as BMI 18.5 to <30 (non-obese) and BMI ≥30 (obese). Patients with BMI<18.5 or unknown BMI were excluded.
Patient characteristics were summarized with N (%) for categorical variables and median (interquartile range, IQR) for continuous variables. Differences between groups were tested using the Chi-square test or Fisher’s exact test for categorical variables, and the analysis of variance (ANOVA) or Kruskal-Wallis test for continuous variables, as appropriate. Adjusted logistic regression was used to estimate the association of BMI group with the odds of upstaging to cancer after atypia on needle biopsy and odds of subsequent diagnosis of DCIS or invasive breast cancer, after adjustment for select covariates. Odds ratios (ORs) and 95% confidence intervals (CIs) were reported. Adjusted Poisson regression was used to estimate the association of the BMI group with the likelihood of subsequent diagnosis of DCIS or invasive breast cancer, after adjustment for available covariates. This modeling approach was selected to allow for follow-up time to be accounted for in the model as an offset. Due to the skewed nature of follow-up, a log-transformation was applied for inclusion in the model. Rate ratios (RRs) and 95% CIs were reported.
Time to subsequent diagnosis of DCIS or invasive breast cancer was calculated as the time, in months, from date of first atypia pathologic diagnosis to date of subsequent progression or date of last follow-up for patients who did not progress. Patients who were upstaged to DCIS or invasive breast cancer after needle biopsy were excluded from the time to subsequent diagnosis model. Median time to subsequent progression and 5-year progression-free rates were estimated using the Kaplan-Meier method, and differences between BMI groups were tested using the log-rank test. Due to lack of events, Hispanic patients were excluded from all adjusted and time-to-event analyses.
Only those patients with available data for all covariates were included in each model, and effective sample sizes are reported for each table/figure. No adjustments were made for multiple comparisons. All statistical analyses were conducted using SAS, version 9.4 (SAS Institute, Cary NC). This study was deemed exempt by our Institutional Review Board.
RESULTS
Demographics and Clinicopathologic characteristics by BMI
We identified 503 patients with breast atypia at a single institution from 2008–2017 who met the selection criteria (Table 1). Of these patients, 41% were classified as obese (BMI ≥30) [median BMI of 34.9 (IQR 32.1–39.4)], and 59% were classified as non-obese with a BMI of 18.5–29.9 [median BMI of 25.0 (IQR 22.6–27.1)]. The median age of all patients was 54 (IQR 48–64) with most women being post-menopausal, (69.2%). Patients with a BMI of 18.5–29.9 were more likely than the obese population to be non-Hispanic White (83.5% vs 59.7%, p<0.001). With the exception of race/ethnicity, there were no other statistically significant differences in patient demographics between BMI groups (all p>0.05).
Table 1.
Patient Characteristics for women diagnosed with breast atypia (ADH, ALH, LCIS) at a single academic institution from 2008–2017
| All Patients (N=503) N(%) |
BMI 18.5 to <30 (N=297) N(%) |
BMI ≥ 30 (N=206) N(%) |
P-Value | |
|---|---|---|---|---|
| Age (Years) – Median (IQR) | 54 (48–64) | 53 (47–64) | 55 (49–64) | 0.51 |
| BMI – Median (IQR) | 28.1 (24.3–33.4) | 25.0 (22.6–27.1) | 34.9 (32.1–39.4) | - |
| Age at First Live Birth (Years) – Median (IQR) | 25 (20–30) | 26 (20–30) | 24 (20–29) | 0.14 |
| Race/Ethnicity | <0.001 | |||
| Hispanic | 15 (3) | 5 (1.7) | 10 (4.9) | |
| Non-Hispanic Black | 117 (23.3) | 44 (14.8) | 73 (35.4 | |
| Non-Hispanic White | 371 (73.8) | 248 (83.5 | 123 (59.7) | |
| Menopausal Status | 0.18 | |||
| Pre | 89 (17.7) | 56 (18.9) | 33 (16) | |
| Peri | 31 (6.2) | 22 (7.4) | 9 (4.4 | |
| Post | 348 (69.2) | 196 (66) | 152 (73.8) | |
| Use of Hormone Replacement Therapy | 0.05 | |||
| Yes; Currently | 19 (3.8) | 14 (4.7) | 5 (2.4) | |
| Yes; In the Past | 100 (19.9) | 58 (19.5) | 42 (20.4) | |
| Never | 292 (58.1) | 180 (60.6) | 112 (54.4) | |
| Not Sure | 87 (17.3) | 41 (13.8) | 46 (22.3) | |
| Family History of Cancer | 0.68 | |||
| No | 82 (16.3) | 50 (16.8) | 32 (15.5 | |
| Yes | 415 (82.5) | 243 (81.8) | 172 (83.5) | |
| Gravidity – Median (IQR) | 2 (1–3) | 2 (1–3) | 2 (1–3) | 0.55 |
| Parity – Median (IQR) | 2 (1–2) | 2 (1–2) | 2 (1–3) | 0.95 |
| History of Breastfeeding* | 0.12 | |||
| No | 133 (35.1) | 74 (32.2) | 59 (39.6) | |
| Yes | 150 (39.6) | 97 (42.2) | 53 (35.6) | |
| Use of Oral Contraceptives | 0.31 | |||
| No | 129 (25.6) | 72 (24.2) | 57 (27.7) | |
| Yes | 215 (42.7) | 132 (44.4) | 83 (40.3) | |
| Diagnosis Method | 0.13 | |||
| Surgical Excision | 57 (11.3) | 39 (13.1) | 18 (8.7) | |
| Needle Biopsy | 446 (88.7) | 258 (86.9) | 188 (91.3) | |
| First Atypia Diagnosis | 0.07 | |||
| ADH | 372 (74) | 209 (70.4) | 163 (79.1) | |
| ALH | 58 (11.5) | 41 (13.8) | 17 (8.3) | |
| LCIS | 65 (12.9) | 42 (14.1) | 23 (11.2) | |
| Most Advanced Atypia Diagnosis | 0.16 | |||
| ADH | 376 (74.8) | 214 (72.1) | 162 (78.6) | |
| ALH | 23 (4.6) | 17 (5.7) | 6 (2.9) | |
| LCIS | 104 (20.7) | 66 (22.2) | 38 (18.4) | |
| Atypia on Needle Biopsy Upstaged to Cancer** | 0.30 | |||
| No | 380 (85.2) | 216 (83.7) | 164 (87.2) | |
| Yes | 66 (14.8) | 42 (16.3) | 24 (12.8) | |
| Upstaged after Needle Biopsy Diagnosis*** | 0.12 | |||
| DCIS | 41 (62.1) | 23 (54.8) | 18 (75) | |
| Invasive Carcinoma | 25 (37.9) | 19 (45.2) | 6 (25) | |
| Subsequent Progression to DCIS/Invasive Cancer | 0.13 | |||
| No | 353 (81.1) | 200 (78.7) | 153 (84.5) | |
| Yes | 82 (18.9) | 54 (21.3) | 28 (15.5) | |
| Subsequent Diagnosis**** | 0.53 | |||
| DCIS | 37 (45.1) | 23 (42.6) | 14 (50) | |
| Invasive Carcinoma | 45 (54.9) | 31 (57.4) | 14 (50) |
ADH=atypical ductal hyperplasia, ALH=atypical lobular hyperplasia, LCIS=lobular carcinoma in situ, DCIS=ductal carcinoma in situ, BMI=body mass index
Data presented as N (%) unless otherwise specified. Percentages may not add up to 100 due to rounding or missing value.
Among women who have had at least one pregnancy.
Among women who were diagnosed with atypia by needle biopsy.
Among women who were upstaged after needle biopsy.
Among women who were diagnosed with subsequent disease
Of the initial histopathologic diagnoses, ADH was the most identified atypical lesion (74.0%, n=372). ALH was identified in 58 patients (11.5%), and LCIS was diagnosed in 65 patients (12.9%). Most advanced atypia diagnosis followed a similar distribution for ADH (74.8%), but with a higher proportion of LCIS (20.7%) and a lower proportion of ALH (4.6%). The distribution of first atypia and most advanced diagnosis were similar between obese and non-obese patients (p=0.07 and 0.16, respectively, Table 1).
BMI and risk of upstaging at time of surgical excision
The majority of initial atypia diagnoses were made via core needle biopsy (88.7%) with the remainder of diagnoses made via surgical excision (11.3%). Of those diagnosed with breast atypia first by needle biopsy (n=446), 92.5% underwent surgical excision for the atypia. At the time of surgical excision (n=431), 15.3% were upstaged to either DCIS (n=41, 9.5%) or invasive cancer (n=25, 5.8%); Table 1. For those patients upstaged to DCIS (n=41), the majority were estrogen receptor (ER)-positive (90.2%) and grade 1 or 2 (87.8%). For those upstaged to invasive cancer (n=25), the majority of tumors were pathologic stage T1 (88%), N0 (84%), grade 1 or 2 (88%), ER-positive (92%), progesterone receptor (PR)-positive (88%), and human-epidermal-growth-factor-receptor-2 (HER2)-negative (76%). BMI groups did not differ on tumor type (DCIS vs invasive), hormone sensitivity, grade, and/or stage at diagnosis (p>0.05; Table 2). In addition, after being upstaged to DCIS or invasive cancer, the majority of women received radiation therapy (56.1%) and endocrine therapy (57.6%); and only two patients (8%) received chemotherapy. Administration of these adjunctive therapies did not differ by BMI (all p>0.05; Table 2).
Table 2.
Summary of those with atypia (ADH, ALH, LCIS) on needle biopsy and found to have an underlying malignancy at surgical excision (N=66) by BMI, including DCIS (N=41) or invasive disease (N=25) at a single academic institution from 2008–2017.
| All Patients (N=66) N(%) |
BMI 18.5 to <30 (N=42) N(%) |
BMI ≥ 30 (N=24) N(%) |
P-Value | |
|---|---|---|---|---|
| DCIS, ER status | 1.00 | |||
| Positive | 37 (90.2) | 20 (87) | 17 (94.4) | |
| Negative | 3 (7.3) | 2 (8.7) | 1 (5.6) | |
| DCIS, PR status | 0.61 | |||
| Positive | 36 (87.8) | 19 (82.6) | 17 (94.4) | |
| Negative | 4 (9.8) | 3 (13) | 1 (5.6) | |
| DCIS, Grade | 1.00 | |||
| 1 | 7 (17.1) | 4 (17.4) | 3 (16.7) | |
| 2 | 29 (70.7) | 15 (65.2) | 14 (77.8) | |
| 3 | 2 (4.9) | 1 (4.3) | 1 (5.6) | |
| Invasive Cancer, T stage | 0.58 | |||
| T0 | 1 (4) | 1 (5.3) | 0 (0) | |
| T1 | 22 (88) | 17 (89.5) | 5 (83.3) | |
| T2 | 2 (8) | 1 (5.3) | 1 (16.7) | |
| Invasive Cancer, N stage | 1.00 | |||
| N0 | 21 (84) | 16 (84.2) | 5 (83.3) | |
| N1 | 1 (4) | 1 (5.3) | 0 (0) | |
| Invasive Cancer, ER Status | - | |||
| Positive | 23 (92) | 17 (89.5) | 6 (100) | |
| Negative | 0 (0) | 0 (0) | 0 (0) | |
| Invasive Cancer, PR Status | 1.00 | |||
| Positive | 22 (88) | 16 (84.2) | 6 (100) | |
| Negative | 1 (4) | 1 (5.3) | 0 (0) | |
| Invasive Cancer, HER2 Status | 1.00 | |||
| HER2+ | 4 (16) | 3 (15.8) | 1 (16.7) | |
| HER2− | 19 (76) | 14 (73.7) | 5 (83.3) | |
| Invasive Cancer, Grade | 1.00 | |||
| 1 | 11 (44) | 8 (42.1) | 3 (50) | |
| 2 | 11 (44) | 9 (47.4) | 2 (33.3) | |
| 3 | 3 (12) | 2 (10.5) | 1 (16.7) | |
| Treatment with Radiation Therapy* | 0.32 | |||
| Yes | 37 (56.1) | 22 (52.4) | 15 (62.5) | |
| No | 28 (42.4) | 20 (47.6) | 8 (33.3) | |
| Treatment with Chemotherapy** | 0.45 | |||
| Yes | 2 (8) | 1 (5.3) | 1 (16.7) | |
| No | 22 (88) | 17 (89.5) | 5 (83.3) | |
| Treatment with Endocrine Therapy* | 0.60 | |||
| Yes | 45 (68.2) | 30 (71.4) | 15 (62.5) | |
| No | 20 (30.3) | 12 (28.6) | 8 (33.3) | |
| Treatment with Endocrine Therapy, if ER+ and/or PR+* | 0.28 | |||
| Yes | 38 (57.6) | 25 (59.5) | 13 (54.2) | |
| No | 16 (24.2) | 8 (19) | 8 (33.3) |
ADH=atypical ductal hyperplasia, ALH=atypical lobular hyperplasia, LCIS=lobular carcinoma in situ, DCIS=ductal carcinoma in situ, BMI=body mass index, ER=estrogen receptor, PR=progesterone receptor, HER2=human epidermal growth factor receptor 2
Data presented as N (%) unless otherwise specified. Percentages may not add up to 100 due to rounding or missing value.
Percentages may not add up to 100 due to rounding for missing values.
Among women who have either DCIS and/or invasive disease.
Among women who have invasive disease only.
After adjustment for age, race/ethnicity, and first atypia diagnosis, BMI was not associated with upstaging to an underlying malignancy at the time of the initial surgical excision following needle biopsy [reference: BMI 18.5 to <30; vs BMI ≥ 30, OR 0.66 (95% CI 0.37–1.18), p=0.16; Table 3].
Table 3.
Adjusted breast cancer risk by BMI, following core needle biopsy of atypia. Analysis includes patients diagnosed with underlying malignancy at surgical excision of breast atypia (upstaging) and those who developed a subsequent diagnosis of DCIS/invasive cancer (progression). Analysis adjusted for age, race/ethnicity, and histology of first atypia diagnosis. 65/431 patients included in this model were found to have an underlying malignancy at surgical excision after atypia on needle biopsy and 81/394 patients had subsequent diagnoses of DCIS or invasive breast cancer. Hispanic patients were excluded due to lack of events.
| Underlying malignancy at surgical excision after needle biopsy with atypia (N=431) * | |||
| Odds Ratio (95% CI) | P-Value | Overall P-Value | |
| BMI | 0.16 | ||
| 18.5 to <30 | REF | ||
| ≥ 30 | 0.663 (0.373–1.179) | 0.16 | |
| Subsequent Diagnosis of DCIS/invasive Cancer (N=394) ** | |||
| Odds Ratio (95% CI) | P-Value | Overall P-Value | |
| BMI | 0.08 | ||
| 18.5 to <30 | REF | ||
| ≥ 30 | 0.618 (0.358–1.067) | 0.08 | |
Adjusted for age, race/ethnicity, and first atypia diagnosis
Adjusted for age, race/ethnicity, most advanced atypia diagnosis, treatment with chemoprevention, and family history of cancer
ADH=atypical ductal hyperplasia, ALH=atypical lobular hyperplasia, LCIS=lobular carcinoma in situ, DCIS=ductal carcinoma in situ, BMI=body mass index
BMI and long-term breast cancer risk
At a median follow up of 89.9 months from the first atypia diagnosis (95% CI 86.6–94.7), 18.9% (82 of 435) patients were subsequently diagnosed with breast cancer (Table 1). Of these, 45.1% were diagnosed with DCIS and 54.9% were diagnosed with invasive carcinoma. For those subsequently diagnosed with DCIS (n=37), the majority of lesions were ER-positive (91.9%) and grades 1 or 2 (64.9%). For those diagnosed with invasive cancer (n=45), the majority of the tumors were ipsilateral (62.2%), stage T1 (68.9%) and N0 (82.2%). Interestingly, 13 patients had T2 (26.7%) or T3 (2.2%) disease at time of subsequent diagnosis. Regardless, most tumors were grade 1 or 2 (82.2%), ER-positive (93.3%), PR-positive (84.4%), and HER2-negative (82.2%). For those patients subsequently diagnosed with DCIS or invasive carcinoma, laterality, hormone sensitivity, grade, and/or stage (both tumor and nodal) did not differ based on BMI (p>0.05; Table 4). In addition, after having a subsequent malignant diagnosis, most women received endocrine therapy (43.9%) but did not have subsequent treatment with radiation therapy (64.6%) or chemotherapy (80%). Use of these adjunctive therapies did not differ between BMI groups (p>0.05; Table 4).
Table 4.
Summary of those with a subsequent malignant diagnosis after atypia (ADH, ALH, LCIS, (N=82)) by BMI, including DCIS (N=37) or invasive disease (N=45) at a single academic institution from 2008–2017.
| All Patients (N=82) N(%) |
BMI 18.5 to <30 (N=54) N(%) |
BMI ≥ 30 (N=28) N(%) |
P-Value | |
|---|---|---|---|---|
| DCIS, ER status | 1.00 | |||
| ER+ | 34 (91.9) | 21 (91.3) | 13 (92.9) | |
| ER− | 3 (8.1) | 2 (8.7) | 1 (7.1) | |
| DCIS, PR status | 0.63 | |||
| PR+ | 30 (81.1) | 17 (73.9) | 13 (92.9) | |
| PR− | 4 (10.8) | 3 (13) | 1 (7.1) | |
| DCIS, Grade | 0.91 | |||
| 1 | 7 (18.9) | 5 (21.7) | 2 (14.3) | |
| 2 | 17 (45.9) | 10 (43.5) | 7 (50) | |
| 3 | 13 (35.1) | 8 (34.8) | 5 (35.7) | |
| Invasive Cancer, T stage | 0.16 | |||
| T0 | 1 (2.2) | 0 (0) | 1 (7.1) | |
| T1 | 31 (68.9) | 23 (74.2) | 8 (57.1) | |
| T2 | 12 (26.7) | 8 (25.8) | 4 (28.6) | |
| T3 | 1 (2.2) | 0 (0) | 1 (7.1) | |
| Invasive Cancer, N stage | 0.52 | |||
| N0 | 37 (82.2) | 26 (83.9) | 11 (78.6) | |
| N1 | 5 (11.1) | 3 (9.7) | 2 (14.3) | |
| N2 | 1 (2.2) | 0 (0) | 1 (7.1) | |
| N3 | 1 (2.2) | 1 (3.2) | 0 (0) | |
| Invasive Cancer, ER Status | 1.00 | |||
| ER+ | 42 (93.3) | 29 (93.5) | 13 (92.9) | |
| ER− | 3 (6.7) | 2 (6.5 | 1 (7.1) | |
| Invasive Cancer, PR Status | 1.00 | |||
| PR+ | 38 (84.4) | 26 (83.9) | 12 (85.7) | |
| PR− | 7 (15.6) | 5 (16.1) | 2 (14.3) | |
| Invasive Cancer, HER2 Status | 0.65 | |||
| HER2+ | 7 (15.6) | 6 (19.4) | 1 (7.1) | |
| HER2− | 37 (82.2) | 25 (80.6) | 12 (85.7) | |
| Invasive Cancer, Grade | 0.24 | |||
| 1 | 13 (28.9) | 7 (22.6) | 6 (42.9) | |
| 2 | 24 (53.3) | 19 (61.3) | 5 (35.7) | |
| 3 | 8 (17.8) | 5 (16.1) | 3 (21.4) | |
| Laterality of Subsequent Malignancy compared to Initial Atypia | 0.80 | |||
| Ipsilateral | 51 (62.2) | 32 (59.3) | 19 (67.9) | |
| Contralateral | 15 (18.3) | 11 (20.4 | 4 (14.3) | |
| Unknown | 16 (19.5) | 11 (20.4) | 5 (17.9) | |
| Treatment with Radiation Therapy* | 0.09 | |||
| Yes | 25 (30.5) | 13 (24.1) | 12 (42.9) | |
| No | 53 (64.6) | 38 (70.4) | 15 (53.6) | |
| Treatment with Chemotherapy** | 1.00 | |||
| Yes | 9 (20) | 6 (19.4) | 3 (21.4) | |
| No | 36 (80) | 25 (80.6) | 11 (78.6) | |
| Treatment with Endocrine Therapy* | 0.20 | |||
| Yes | 42 (51.2) | 30 (55.6) | 12 (42.9) | |
| No | 38 (46.3) | 22 (40.7) | 16 (57.1) | |
| Treatment with Endocrine Therapy, if ER+ and/or PR+* | 0.18 | |||
| Yes | 36 (43.9) | 25 (46.3) | 11 (39.3) | |
| No | 23 (28) | 12 (22.2) | 11 (39.3) |
ADH=atypical ductal hyperplasia, ALH=atypical lobular hyperplasia, LCIS=lobular carcinoma in situ, DCIS=ductal carcinoma in situ, BMI=body mass index, ER=estrogen receptor, PR=progesterone receptor, HER2=human epidermal growth factor receptor 2
Percentages may not add up to 100 due to rounding for missing values.
Among women who have either DCIS and/or invasive disease.
Among women who have invasive disease only.
After adjustment for age, race/ethnicity, and first atypia diagnosis, BMI was not associated with the risk of a subsequent malignant diagnosis [reference: BMI 18.5 to <30; vs BMI ≥ 30, OR 0.62 (95% CI 0.36–1.07), p=0.08; Table 3]. Among those with available chemoprevention data, the utilization of chemoprevention after an atypia diagnosis was low at 11.2%(n=46) and did not differ on BMI (p=0.86; Supplemental Table 1).
The unadjusted time to a subsequent malignant diagnosis (DCIS or invasive breast cancer) was similar between BMI groups (log-rank p=0.39; Figure 1). When accounting for differing lengths of follow-up time, and after adjustment for age, race/ethnicity, most advanced atypia diagnosis, treatment with chemoprevention, and family history, there was no association between BMI group and subsequent diagnosis of DCIS or invasive breast cancer, [reference: BMI 18.5 to <30; vs BMI ≥ 30 RR 0.76 (95% CI 0.47–1.24), p=0.28; Table 5].
Figure 1.
The unadjusted time to a subsequent diagnosis of DCIS or invasive cancer after atypia (ADH, ALH, or LCIS) by BMI (N=424) at a single academic institution from 2008–2017.
ADH=atypical ductal hyperplasia, ALH=atypical lobular hyperplasia, LCIS=lobular carcinoma in situ, DCIS=ductal carcinoma in situ, BMI=body mass index
Table 5.
Adjusted Poisson regression for a subsequent diagnosis of DCIS or invasive breast cancer (N=393) after atypia (ADH, ALH, LCIS). Analysis adjusted for log-transformed follow up time. 81/393 patients included in this model had a subsequent diagnosis of DCIS or invasive breast cancer. Hispanic patients were excluded due to lack of events.
| Risk Ratio (95% CI) | P-Value | Overall P-Value | |
|---|---|---|---|
| Age (Years) | 1.01 (0.99–1.03) | 0.36 | 0.36 |
| BMI | 0.28 | ||
| 18.5 to <30 | REF | ||
| ≥ 30 | 0.76 (0.47–1.24) | 0.28 | |
| Race/Ethnicity | 0.66 | ||
| Non-Hispanic White | REF | ||
| Non-Hispanic Black | 1.13 (0.66–1.91) | 0.66 | |
| Most Advanced Atypia Diagnosis* | 0.17 | ||
| ADH | REF | ||
| ALH | 0.35 (0.05–2.58) | 0.31 | |
| LCIS | 1.45 (0.88–2.38) | 0.14 | |
| Treatment with Chemoprevention | 0.86 | ||
| No | REF | ||
| Yes | 0.94 (0.49–1.83) | 0.86 | |
| Family History of Cancer | 0.13 | ||
| No | REF | ||
| Yes | 1.78 (0.85–3.73) | 0.13 |
DCIS=ductal carcinoma in situ, ADH=atypical ductal hyperplasia, ALH=atypical lobular hyperplasia, LCIS=lobular carcinoma in situ, DCIS=ductal carcinoma in situ, BMI=body mass index
Patients with multiple types of atypia were coded according to their most advanced atypia as follows: (1) LCIS, (2) ADH, (3) ALH.
DISCUSSION
Though atypical breast lesions such as ADH, ALH, and LCIS are considered benign, women with these high-risk lesions are considered to be at increased risk for developing a breast malignancy in their lifetime.6–8 The potential association between these high-risk breast lesions and certain socio-demographic factors, however, remains unclear. Since obesity is associated with higher rates of hormone sensitive breast cancers, the relationship between obesity and an increased risk of developing high-risk breast lesions can also be hypothesized. In this large retrospective study, however, we found no statistically significant association between BMI group and the risk of upstage or subsequent diagnosis of DCIS or invasive cancer.
Various retrospective studies have demonstrated that certain clinical and histopathologic risk factors are associated with breast atypia and underlying malignancy. In 2007, Degnim et al. published their findings in 331 women in the Mayo Benign Breast Disease Cohort and found that the extent and multiplicity of the atypia can predict carcinoma after needle biopsy.6 The increasing risk of developing a malignancy was associated with the number of foci of atypia, with a relative risk of 2.33 with a single focus, 5.26 for two foci and 7.97 for three or more foci, p<0.001. Stratification based on the number of foci with atypia showed a cumulative incidence of breast cancer reaching 48% at 25 years for those with three or more foci. In addition, they also found that not only did multiple foci increase the risk of a malignancy, but calcifications along with an increased number of foci dramatically increased the risk of developing cancer (relative risk 10.4; 95% CI 6.13–16.4). When they looked at family history of breast cancer, there were no significant differences in risk among those subgroups with a strong family history (risk ratio, 3.59; 95% CI 1.96–6.03) vs those with a negative family history (relative risk, 3.81; 95% CI 2.60–5.37).6 In a more recent 2016 study also by Degnim et al., a total of 1174 women with atypical hyperplasia were included (both from a Mayo independent cohort (708) and a Nashville independent cohort (466)), and demonstrated that the extent of atypia stratifies the long-term cancer risk for both ADH and ALH and the risk was highest with ≥3 foci of ADH.17
In their retrospective chart review of 151 patients, Linsk et al., also focused on histologic and clinical risk factors associated with a primary atypia upstage and created a predictive model with a 78% sensitivity and 80% specificity to help determine which patients were at highest risk. They found that the maximum lesion size (>16.3mm) at the time of initial imaging and radiographic presence of a residual lesion after core needle biopsy were associated with an upstage to malignancy (p ≤0.05). Family history, personal history of cancer, and race were not associated with upstaging (p>0.05).18
Although we did not look at the number of atypia foci, associated calcifications, or residual radiographic abnormality, we evaluated other histopathological features such as the tumor type (DCIS vs invasive), grade, stage (including lymph node status) and hormone receptors, and how these factors were associated with patient BMI. In general, among those patients with atypia on needle biopsy found to have an underlying malignancy at the time of excision, or those who subsequently developed a breast malignancy, we found no statistically significant association with obesity.
Other research in this area has sought to focus on demographic characteristics to see how these might impact high-risk lesions and the rate of upstage and/or development of a subsequent breast malignancy. For instance, Sergesketter et al. first focused on age and how this relates to atypia and associated malignancy. They found that older age was associated with a greater risk of underlying carcinoma (OR 1.028, 95% CI 1.003–1.053, p=0.03) but that the long-term risk of malignancy associated with atypia was not affected by age (p=0.48) nor did it affect a subsequent diagnosis of carcinoma (log-rank p=0.41). 8 In their second study, they looked at race/ethnicity and high-risk breast lesions, and found no association between long-term risk for carcinoma or difference in time to subsequent carcinoma (log-rank p=0.52). 16 Notably, in the current study, nearly all patient characteristics were similarly distributed amongst the two BMI groups, except for race/ethnicity.
The association between obesity and breast cancer, in particular post-menopausal breast cancer is evident; yet the molecular reasoning is not completely understood19. It has been suggested that the inflammatory cytokines and mediators produced from adipose tissues, which are increased in obese individuals, creates a chronic inflammatory microenvironment, promoting cancer invasion and metastases.20 In addition to a chronic inflammatory state other factors such actual body weight, adiposity, and physical activity have also been studied and have been shown to be associated with breast cancer.19 Increased physical activity decreases the risk of breast cancer, and more recent literature by Kehm et al. found that physical activity is beneficial for risk reduction among women with high-risk factors such as BRCA1 and BRCA2 mutations.21 Even though in our study, we did not find an association between BMI and the risk of upstage or subsequent breast cancers in high-risk breast lesions, perhaps other underlying variables influencing BMI were more of a factor in these atypical breast lesions. For example, Kabat et al. have previously demonstrated that metabolic phenotypes may have a stronger association with breast cancer risk than BMI alone,22 and this likely merits further research in future studies.
According to Coopey et al., the 10-year cancer risk with atypical lesions may be as high as 26% for those patients diagnosed with ADH. If excised and treated with chemoprevention, this risk of malignancy can decrease to 7.5%.13 In our study, 15.3% of the high-risk lesions diagnosed on needle biopsy were upstaged to either DCIS or invasive cancer at the time of initial excision, and 18.9% of patients were subsequently diagnosed with breast cancer at a median of 90 months of follow up. Unfortunately, we noted only a small percentage of patients on chemoprevention after their atypia diagnosis (11.2%), which is similar to other studies.23 However, it is possible that chemoprevention may be more or less effective in select populations based on BMI, as estrogen levels have been shown to be associated with BMI.24 More specifically, postmenopausal obese women have been shown to have higher levels of circulating estrogen,25 which may suggest that chemoprevention could be more effective in obese patients.
In this large contemporary cohort of patients diagnosed with high-risk breast histology, the risk of upstaging to malignancy at the time of surgical excision, or of subsequent progression to a future breast cancer, was not associated with BMI. Factors other than obesity may have a more significant influence on progression after a diagnosis of one of these high-risk lesions. We invite future studies with larger cohorts of patients to investigate this further.
Study Limitations
Limitations to our study are inherent to the retrospective nature of the research design. Data collected for retrospective studies rely on data entry which can occur after a long time following the event, rather than in real time. Inevitably, data can be missing and certain variables that have the potential to impact the outcome may not be recorded or may be recorded inaccurately. Differences in baseline characteristics, chart selections, missing data in charts, assumptions to complete missing data and loss of follow-up are all examples of potential confounders and sources of error or bias in retrospective studies. 26 Another limitation of this study is the limited variability in BMI in this study group. It may be that more granular divisions based on BMI category (such as 18.5–25, 25.1–29.9, 30–34.9, ≥35+) may yield different results. Lastly, studies have demonstrated a significant rate of interobserver variability in the pathologic diagnosis of atypical breast lesions;27 thus without central pathology review, the consistency of the atypia diagnoses (or lack thereof) could also be biasing our results.
CONCLUSION
Although high-risk histologies are known to confer an increased future risk of breast cancer, we found that in this large cohort of patients diagnosed with high-risk breast histology (ADH, ALH and LCIS), the risk of upstaging and/or subsequent progression to a breast malignancy was not associated with BMI. Factors other than obesity may have a more significant influence on progression after an atypia diagnosis and should be examined in future studies.
Supplementary Material
Synopsis:
For patients with atypical high-risk breast lesions, select factors are associated with the risk of upstage to and/or subsequent development of breast cancer. In this large cohort of women diagnosed with high-risk breast histology, there was no association with BMI.
ACKNOWLEDGEMENTS
We are grateful for the contributions of the Duke Breast Database (DB2) team, including but not limited to Chandra Almond, Keenan Caddell, Oluwatomi Ladipo, Anuyuga Sampathkumar, and Madeline Thornton.
FUNDING SOURCES
This work was in part supported by Duke Cancer Institute through NIH grant P30CA014236 (PI: Kastan) for the Biostatistics Core.
This publication was made possible (in part) by philanthropic funds through the generosity of Sara and Bruce Brandaleone.
Footnotes
Disclosures: We have no commercial interests to disclose
- The authors report no proprietary or commercial interest in any product mentioned or concept.
- The content of this manuscript is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health (NIH).
- Dr. J. Plichta is a recipient of research funding by the Color Foundation (PI: Plichta). She serves on the National Comprehensive Cancer Network (NCCN) Breast Cancer Screening Committee.
- Dr. E.S. Hwang serves on the National Cancer Institute (NCI) Breast Cancer Steering Committee and the NCCN Breast Cancer Prevention Committee.
REFERENCES
- 1.Prevention CfDCa. NCHS Health E-Stats. Updated February 8, 2021. Accessed January 29, 2022, www.cdc.gov/nchs/data/hestat/obesity-adult-17-18/obesity-adult.htm
- 2.Roser HRaM. Obesity. Accessed 3/8/22, http://ourworldindata.org/obesity
- 3.Serviceds USDoHaH. Calculate your body mass index. Accessed 9/23/21, www.nhlbi.nih.gov/health/educational/lose_wt/BMI/bmicalc.htm
- 4.Jiralerspong S, Goodwin PJ. Obesity and Breast Cancer Prognosis: Evidence, Challenges, and Opportunities. J Clin Oncol. Dec 10 2016;34(35):4203–4216. doi: 10.1200/jco.2016.68.4480 [DOI] [PubMed] [Google Scholar]
- 5.Society AC. How common is breast cancer. Accessed 3/9/22, http://www.cancer.org/cancer/breeast-cancer/about/how-common-is-breast-cancer.html.
- 6.Degnim AC, Visscher DW, Berman HK, et al. Stratification of breast cancer risk in women with atypia: a Mayo cohort study. J Clin Oncol. Jul 01 2007;25(19):2671–7. doi: 10.1200/JCO.2006.09.0217 [DOI] [PubMed] [Google Scholar]
- 7.King TA, Pilewskie M, Muhsen S, et al. Lobular Carcinoma in Situ: A 29-Year Longitudinal Experience Evaluating Clinicopathologic Features and Breast Cancer Risk. J Clin Oncol. Nov 20 2015;33(33):3945–52. doi: 10.1200/JCO.2015.61.4743 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Sergesketter AR, Thomas SM, Fayanju OM, et al. The Influence of Age on the Histopathology and Prognosis of Atypical Breast Lesions. J Surg Res. Sep 2019;241:188–198. doi: 10.1016/j.jss.2019.03.047 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Simpson JF. Update on atypical epithelial hyperplasia and ductal carcinoma in situ. Pathology. 2009;41(1):36–39. doi: 10.1080/00313020802568097 [DOI] [PubMed] [Google Scholar]
- 10.Hartmann LC, Degnim AC, Dupont WD. Atypical hyperplasia of the breast. N Engl J Med. Mar 26 2015;372(13):1271–2. doi: 10.1056/NEJMc1501046 [DOI] [PubMed] [Google Scholar]
- 11.Bevers TB HM, Bonaccio E, et al. NCCN Guidelines: Breast Cancer Screening and Diagnosis (version 1.2021). Accessed 12/13/2021, https://www.nccn.org/professionals/physician_gls/pdf/breast-screening.pdf
- 12.Committee ASsR. ASBrS Consensus Guideline on Concordance Assessment of Image-Guided Breast Biopsies and Management of Borderline or High-Risk Lesion. . Approved on 11/2/2016.
- 13.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. Dec 2012;136(3):627–33. doi: 10.1007/s10549-012-2318-8 [DOI] [PubMed] [Google Scholar]
- 14.Degnim AC, Visscher DW, Radisky DC, et al. Breast cancer risk by the extent and type of atypical hyperplasia. Cancer. Oct 2016;122(19):3087–8. doi: 10.1002/cncr.30151 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Collins LC, Baer HJ, Tamimi RM, Connolly JL, Colditz GA, Schnitt SJ. Magnitude and laterality of breast cancer risk according to histologic type of atypical hyperplasia: results from the Nurses’ Health Study. Cancer. 2007;109(2):180–187. [DOI] [PubMed] [Google Scholar]
- 16.Sergesketter AR, Thomas SM, Parrilla Castellar ER, et al. Do Histopathology and Clinical Outcomes of Breast Atypia Vary by Race/Ethnicity? J Surg Res. Nov 2020;255:205–215. doi: 10.1016/j.jss.2020.05.066 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Degnim AC, Dupont WD, Radisky DC, et al. Extent of atypical hyperplasia stratifies breast cancer risk in 2 independent cohorts of women. Cancer. Oct 2016;122(19):2971–8. doi: 10.1002/cncr.30153 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Linsk A, Mehta TS, Dialani V, et al. Surgical upgrade rate of breast atypia to malignancy: An academic center’s experience and validation of a predictive model. Breast J. Mar 2018;24(2):115–119. doi: 10.1111/tbj.12885 [DOI] [PubMed] [Google Scholar]
- 19.Kolb R, Zhang W. Obesity and Breast Cancer: A Case of Inflamed Adipose Tissue. Cancers (Basel). Jun 25 2020;12(6)doi: 10.3390/cancers12061686 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Gilbert CA, Slingerland JM. Cytokines, obesity, and cancer: new insights on mechanisms linking obesity to cancer risk and progression. Annu Rev Med. 2013;64:45–57. doi: 10.1146/annurev-med-121211-091527 [DOI] [PubMed] [Google Scholar]
- 21.Kehm RD, Genkinger JM, MacInnis RJ, et al. Recreational Physical Activity Is Associated with Reduced Breast Cancer Risk in Adult Women at High Risk for Breast Cancer: A Cohort Study of Women Selected for Familial and Genetic Risk. Cancer Res. Jan 1 2020;80(1):116–125. doi: 10.1158/0008-5472.Can-19-1847 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Kabat GC, Kim MY, Lee JS, et al. Metabolic Obesity Phenotypes and Risk of Breast Cancer in Postmenopausal Women. Cancer Epidemiol Biomarkers Prev. Dec 2017;26(12):1730–1735. doi: 10.1158/1055-9965.Epi-17-0495 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Ropka ME, Keim J, Philbrick JT. Patient decisions about breast cancer chemoprevention: a systematic review and meta-analysis. J Clin Oncol. Jun 20 2010;28(18):3090–5. doi: 10.1200/jco.2009.27.8077 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Karim R, Mack WJ, Hodis HN, Roy S, Stanczyk FZ. Influence of age and obesity on serum estradiol, estrone, and sex hormone binding globulin concentrations following oral estrogen administration in postmenopausal women. J Clin Endocrinol Metab. Nov 2009;94(11):4136–43. doi: 10.1210/jc.2009-0643 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Mahabir S, Baer DJ, Johnson LL, et al. Usefulness of body mass index as a sufficient adiposity measurement for sex hormone concentration associations in postmenopausal women. Cancer Epidemiol Biomarkers Prev. Dec 2006;15(12):2502–7. doi: 10.1158/1055-9965.Epi-06-0499 [DOI] [PubMed] [Google Scholar]
- 26.Talari K, Goyal M. Retrospective studies - utility and caveats. J R Coll Physicians Edinb. Dec 2020;50(4):398–402. doi: 10.4997/jrcpe.2020.409 [DOI] [PubMed] [Google Scholar]
- 27.Darvishian F, Singh B, Simsir A, Ye W, Cangiarella JF. Atypia on breast core needle biopsies: reproducibility and significance. Ann Clin Lab Sci. Summer 2009;39(3):270–6. [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.