Abstract
Purpose
To examine the association between race and clinical outcomes (pathological complete response [pCR]; recurrence free survival [RFS], and overall survival [OS]) in patients diagnosed with triple-negative (TNBC) or HER2-positive breast cancer treated with neoadjuvant chemotherapy (NAC).
Methods
Patients who self-identified as non-Hispanic white (NHW) or non-Hispanic Black (NHB) and were diagnosed with Stage I-III TNBC (n=171 including 124 NHW and 47 NHB) and HER2-positive (n=161 including 136 NHW and 25 NHB) breast cancer who received NAC from 2000–2018 at Roswell Park Comprehensive Cancer Center were included. Associations of race with pCR and survival outcomes were evaluated using logistic and Cox regression models, respectively.
Results
There was no statistically significant difference in pCR between NHB and NHW patients with TNBC (31.9 vs 29.8%; OR: 1.11, 95% CI 0.54–2.29) or HER2-positive breast cancer (36.0 vs 39.7%; OR: 0.87, 95% CI 0.36–3.11). After controlling for potential confounders, including age, stage, treatment regimens, insurance status, and comorbidities, no statistically significant difference in OS or RFS was observed between NHB and NHW patients within either subtype.
Conclusion
TNBC or HER2-positive breast cancer patients treated at a single academic center in Buffalo, NY, showed similar outcomes independent of patients’ race. Given the known genetic diversity of African American ancestry in the US, further studies investigating the interplay between race, geography and clinical outcomes are warranted.
Introduction
Breast cancer survival rates in the United States have significantly improved over the last several decades but racial disparities in survival outcomes persist. Data from Surveillance, Epidemiology and End Results Program (SEER) show that the 5-year survival rate for African American/Black women with triple negative breast cancer (TNBC) is 72% compared to 78% for white women. Along similar lines, Black women with human epidermal growth factor receptor 2 (HER2) enriched breast cancer have a 5-year survival rate of 77% in contrast to 85% for white women [1]. These differences have been attributed to multiple causes including advanced stage at presentation, associated co-morbidities, higher body mass index (BMI) and socio-economic factors such as inadequate access to healthcare which may affect initial diagnosis, follow up and possibly exposure to lower doses of systemic therapy [2].
Pathological complete response (pCR) post neoadjuvant chemotherapy (NAC) is considered a surrogate marker of long-term survival in patients with TNBC and HER2-positive breast cancer [3]. Population-based and single center studies have attempted to investigate relationships between race and pCR or survival outcomes with conflicting results [4–6]. It is possible that the lack of clarity could be due, in part, to geographic variations in the genetic admixture in Black women [7].
The purpose of this study was to determine if racial disparities existed in pCR and survival in patients with non-metastatic (Stage I-III) breast cancer treated with NAC at Roswell Park Comprehensive Cancer Center, located in the Western New York region.
Methods
Data source:
Patient information was obtained from the breast cancer database at Roswell Park Department of Breast Medicine, with individual patient records verified to maintain accuracy. We included female patients above 18 years of age, diagnosed between the years 2000 and 2018, with stage I-III TNBC or HER2-positive breast cancer including inflammatory breast cancer. Patients diagnosed with metastatic disease at presentation and those who did not undergo definitive surgery were excluded. Variables collected included age, race, co-morbidities, insurance status, tumor subtype, clinical stage, pathologic stage, residual disease at surgery, evidence of recurrence, date of last follow up or death. The study was approved by Roswell Park Institutional Review Board. We included a total of 171 patients with TNBC and 161 patients with HER2-positive breast cancer, including 124 patients who were treated post the FDA approval of trastuzumab in the non-metastatic setting on November 16th, 2006.
Staging and Pathology:
We defined initial clinical and final pathological staging according to the standardized staging criteria proposed by American Joint Committee on Cancer (eighth edition). All pathologic specimens were reviewed by board-certified breast pathologists at Roswell Park. Invasive carcinoma was confirmed on initial core biopsy specimens. Histologic type was defined according to the WHO classification system and grade according to Scarff-Bloom-Richardson (SBR) system and the Black method or the Elston-Ellis modification of the SBR grading system (Nottingham grading system)[8, 9]. pCR was defined as the absence of residual invasive disease in both the breast and axillary lymph nodes.
Estrogen receptor (ER) negative and progesterone receptor (PR) negative were defined as immunohistochemical (IHC) staining Allred score <3[10].HER2-negative status was defined as either 1+ or no staining by IHC or absence of gene amplification by fluorescence in situ hybridization[11].
Treatment:
Information was collected regarding the regimens and dosages of NAC, with relative dose intensity (RDI) used as a measure of the chemotherapy patients received. RDI was calculated as (100* Delivered dose intensity/Summation dose intensity) with the final RDI being the summation of individual RDI for each drug in that regimen. Delivered dose intensity was measured as total dose given/duration of chemotherapy and Summation dose intensity was calculated as total dose planned/planned duration of chemotherapy. A patient’s duration of chemotherapy was calculated from the day of the first cycle to the last day of the final cycle that the patient received in weeks. If the plan was to give the patient a certain drug and it wasn’t ultimately given to the patient, RDI for that drug was considered to be 0.
After the completion of NAC, all patients underwent definitive surgery.
Outcome Measures and Statistical Analysis:
We divided the total population into two cohorts, TNBC and HER2 positive breast cancer, where the analyses were performed within each cohort separately. Race was patient self-identified and only non-Hispanic Black (NHB) and non-Hispanic White (NHW) patients were included. Demographic and clinical characteristics were summarized separately for NHB and NHW patients using the median, and range for continuous variables, and frequencies and relative frequencies for categorical variables. Comparisons were made using the Mann Whitney-U and Pearson chi-square tests, as appropriate. The survival outcomes were summarized by racial group using standard Kaplan-Meier methods, with comparisons made using the log-rank test.
Continuous variables in the study included age, BMI, time interval between the original diagnosis to chemotherapy and surgery in months, and RDI. Categorical variables were age categories, race, comorbidities (diabetes mellitus, hypertension, chronic kidney disease, asthma, hypothyroidism, cardiac disease), clinical tumor stage, clinical nodal stage, grade, pCR and final pathologic staging.
Associations between pCR and patient demographic/clinical characteristics were evaluated using logistic regression models. Associations with survival outcomes were evaluated using Cox regression models. All models were fit using Firth’s method and odds ratios (ORs) and hazard ratios (HRs) were obtained with 95% confidence intervals (95% CIs). Multivariate models were also considered to evaluate the association between race and response or survival while adjusting for relevant confounders (other factors associated with response or survival, where p<0.1 on univariate analyses). Analyses were performed separately in the TNBC and HER2 breast cancer subsets.
All analyses were conducted in SAS v9.4 (Cary, NC) at a significance level of 0.05.
Results
A total of 171 TNBC and 161 HER2 positive breast cancer patients diagnosed between 2000 and 2018 met the eligibility criteria and were included in the analysis. Additionally, a HER2 subset was created using patients (124 patients) treated post November 16, 2006, which corresponds to the date FDA approved trastuzumab in the non-metastatic setting.
Baseline demographic and clinical characteristics with tumor and treatment details are depicted in Table 1. Among patients with TNBC, NHB women (n=47) had higher incidence of hypertension (42.6% vs 23.4%, p < 0.05), diabetes mellitus (21.3% vs 7.3%, p <0.05), and more advanced clinical nodal involvement/stage cN3 (14.9% vs 3.2%, p < 0.05) compared to their NHW counterparts (n=124). In addition, NHB women were more likely to have government insurance such as Medicare and Medicaid (55.3% vs 27.4%, p < 0.05). There was no statistically significant difference between the two groups in terms of the time interval between diagnosis to initiation of chemotherapy and surgery.
Table 1:
Baseline demographic and clinical characteristics of TNBC and HER2-positive breast cancer patients.
| TNBC | HER2-positive | ||||||
|---|---|---|---|---|---|---|---|
| NHW n (%) | NHB n (%) | P-value | NHW n (%) | NHB n (%) | P-value | ||
| Overall | N | 124 (72.5) | 47 (27.5) | 136 (84.5) | 25 (15.5) | ||
| Age (years) | Median | 50.50 | 50.00 | 0.549 | 53.00 | 53.00 | 0.905 |
| < 40 | 20 (16.1) | 8 (17.0) | 0.984 | 22 (16.2) | 4 (16.0) | 0.764 | |
| 40–50 | 39 (31.5) | 15 (31.9) | 36 (26.5) | 6 (24.0) | |||
| 50–60 | 34 (27.4) | 14 (29.8) | 35 (25.7) | 9 (36.0) | |||
| 60–70 | 20 (16.1) | 7 (14.9) | 29 (21.3) | 3 (12.0) | |||
| 70+ | 11 (8.9) | 3 (6.4) | 14 (10.3) | 3 (12.0) | |||
| BMI (kg/m2) | Median | 27.13 | 31.59 | 0.596 | 28.79 | 29.67 | 0.120 |
| Insurance | Private | 89 (71.8) | 21 (44.7) | 0.003 | 92 (67.6) | 13 (52.0) | 0.131 |
| Gov | 34 (27.4) | 26 (55.3) | 44 (32.4) | 12 (48.0) | |||
| None | 1 (0.8) | 125 (92.6) | 24 (96.0) | ||||
| Clinical Stage | 1 | 5 (4.0) | 3 (6.5) | 0.216 | 84 (61.8) | 12 (48.0) | 0.536 |
| 2 | 75 (60.5) | 21 (45.7) | 39 (28.7) | 10 (40.0) | |||
| 3 | 44 (35.5) | 22 (47.8) | 3 (2.3) | 1 (4.3) | |||
| Grade | I/II | 17 (14.0) | 6 (13.3) | 0.905 | 43 (33.1) | 7 (30.4) | 0.838 |
| III | 104 (86.0) | 39 (86.7) | 84 (64.6) | 15 (65.2) | |||
| Diagnosis to Chemotherapy (months) | Median | 1.31 | 1.20 | 0.900 | 1.22 | 1.43 | 0.028 |
| Diagnosis to Surgery (months) | Median | 6.14 | 6.21 | 0.665 | 6.23 | 6.80 | 0.032 |
| HTN | No | 95 (76.6) | 27 (57.4) | 0.013 | 106 (77.9) | 12 (48.0) | 0.002 |
| Yes | 29 (23.4) | 20 (42.6) | 30 (22.1) | 13 (52.0) | |||
| DM | No | 115 (92.7) | 37 (78.7) | 0.009 | 131 (96.3) | 22 (88.0) | 0.078 |
| Yes | 9 (7.3) | 10 (21.3) | 5 (3.7) | 3 (12.0) | |||
| CKD | No | 122 (98.4) | 46 (97.9) | 0.819 | 134 (98.5) | 25 (100.0) | 0.542 |
| Yes | 2 (1.6) | 1 (2.1) | 2 (1.5) | ||||
| Asthma | No | 118 (95.2) | 44 (93.6) | 0.686 | 124 (91.2) | 21 (84.0) | 0.270 |
| Yes | 6 (4.8) | 3 (6.4) | 12 (8.8) | 4 (16.0) | |||
| Cardiac disease | No | 122 (98.4) | 45 (95.7) | 0.307 | 131 (96.3) | 23 (92.0) | 0.330 |
| Yes | 2 (1.6) | 2 (4.3) | 5 (3.7) | 2 (8.0) | |||
| Hypothyroidism | No | 108 (87.1) | 42 (89.4) | 0.687 | 118 (86.8) | 24 (96.0) | 0.188 |
| Yes | 16 (12.9) | 5 (10.6) | 18 (13.2) | 1 (4.0) | |||
| RDI | Mean/SD | 92.04/16.23 | 87.97/18.02 | 0.063 | 91.87/13.52 | 81.18/27.19 | 0.184 |
| Chemotherapy | Anthracycline based | 9 (7.3) | 5 (10.6) | 0.233 | 10 (7.4) | 1 (4.0) | 0.188 |
| Taxane based | 8 (6.5) | 5 (10.6) | 66 (48.5) | 14 (56.0) | |||
| Both | 107 (86.3) | 36 (76.6) | 58 (42.6) | 8 (32.0) | |||
| Other | 1 (2.1) | 2 (1.5) | 2 (8.0) | ||||
| cT | cT1 | 11 (8.9) | 4 (8.9) | 0.363 | 14 (10.4) | 5 (20.0) | 0.594 |
| cT2 | 73 (58.9) | 20 (44.4) | 70 (51.9) | 12 (48.0) | |||
| cT3 | 18 (14.5) | 10 (22.2) | 32 (23.7) | 5 (20.0) | |||
| cT4 | 22 (17.7) | 11 (24.4) | 19 (14.1) | 3 (12.0) | |||
| cN | cN0 | 58 (46.8) | 16 (34.0) | 0.037 | 72 (53.3) | 9 (36.0) | 0.100 |
| cN1 | 46 (37.1) | 18 (38.3) | 51 (37.8) | 13 (52.0) | |||
| cN2 | 16 (12.9) | 6 (12.8) | 10 (7.4) | 1 (4.0) | |||
| cN3 | 4 (3.2) | 7 (14.9) | 2 (1.5) | 2 (8.0) | |||
| Pathologic Stage | pCR | 37 (29.8) | 15 (32.6) | 0.877 | 54 (39.7) | 8 (33.3) | 0.633 |
| yp1 | 36 (29.0) | 11 (23.9) | 27 (19.9) | 3 (12.5) | |||
| yp2 | 21 (16.9) | 7 (15.2) | 33 (24.3) | 8 (33.3) | |||
| yp3 | 30 (24.2) | 13 (28.3) | 22 (16.2) | 5 (20.8) | |||
Abbreviations: BMI-Body mass index, HTN-Hypertension, DM-Diabetes Mellitus, CKD-Chronic Kidney Disease, RDI-Relative Dose Intensity, cT-AJCC (eighth edition) Clinical category of primary tumor, cN- AJCC (eighth edition) Clinical category of regional node involvement, pCR-Complete Pathological Response, yp1-AJCC Pathologic Stage I, yp2- AJCC Pathologic Stage II, yp3- AJCC Pathologic Stage III, SD-standard deviation
Among patients with HER2-positive disease, there were significantly higher rates of hypertension (52.0 vs 22.1%, p < 0.05) among NHB women (n=25) in comparison to NHW women (n=136). Also, there was a significant difference in the time between the diagnosis to the initiation of chemotherapy (median 1.43 vs 1.22 months, p < 0.05) and to surgery (6.8 months vs 6.23 months, p < 0.05). There was no statistically significant difference observed in the type of insurance. There was also no significant difference between the chemotherapy RDI that either group received. We did not observe a statistically significant difference in pCR between NHB and NHW patients in either TNBC or HER2-positive group.
Table 2 describes the univariate associations of pCR with demographic and clinical variables. In patients with TNBC, pCR was associated with younger age at diagnosis and higher grade.
Table 2:
Univariate association of pCR with demographic and clinical variables in TNBC and HER2-positive breast cancer patients.
| Variable | TNBC | HER2-positive | |||||||
|---|---|---|---|---|---|---|---|---|---|
| pCR | Univariate | pCR | Univariate | ||||||
| No (%) | Yes (%) | OR (95% CI) | P-value | No (%) | Yes (%) | OR (95% CI) | P-value | ||
| Age (years) | Mean/SD | 53.05/12.46 | 47.06/10.14 | 0.96 (0.93–0.99) | 0.004 | 53.02/12.22 | 51.67/14.90 | 0.99 (0.97–1.02) | 0.538 |
| < 40 | 18 (64.3) | 10 (35.7) | 1.00 | 0.064 | 12 (46.2) | 14 (53.8) | 1.00 | 0.500 | |
| 40–50 | 30 (55.6) | 24 (44.4) | 1.42 (0.55–3.62) | 29 (69.0) | 13 (31.0) | 0.39 (0.14–1.08) | |||
| 50–60 | 37 (77.1) | 11 (22.9) | 0.54 (0.19–1.50) | 27 (61.4) | 17 (38.6) | 0.55 (0.21–1.46) | |||
| 60–70 | 20 (74.1) | 7 (25.9) | 0.64 (0.20–2.03) | 20 (62.5) | 12 (37.5) | 0.53 (0.18–1.50) | |||
| 70+ | 14 (100.0) | 0.06 (0.00–1.24) | 10 (58.8) | 7 (41.2) | 0.62 (0.18–2.12) | ||||
| BMI (kg/m2) | Mean/SD | 30.19/7.40 | 29.44/7.89 | 0.99 (0.94–1.03) | 0.596 | 30.68/7.93 | 28.73/6.88 | 0.97 (0.92–1.01) | 0.158 |
| Insurance | Private | 76 (69.1) | 34 (30.9) | 1.00 | 0.987 | 62 (59.0) | 43 (41.0) | 1.00 | 0.530 |
| Gov | 42 (70.0) | 18 (30.0) | 0.97 (0.49–1.91) | 36 (64.3) | 20 (35.7) | 0.81 (0.41–1.58) | |||
| None | 1 (100.0) | 0.73 (0.01–69.62) | 94 (63.1) | 55 (36.9) | 1.00 | ||||
| Clinical Stage | 1 | 5 (62.5) | 3 (37.5) | 1.00 | 0.537 | 53 (55.2) | 43 (44.8) | 0.64 (0.22–1.86) | 0.106 |
| 2 | 64 (66.7) | 32 (33.3) | 0.79 (0.18–3.49) | 38 (77.6) | 11 (22.4) | 0.24 (0.07–0.78) | |||
| 3 | 49 (74.2) | 17 (25.8) | 0.56 (0.12–2.55) | 4 (100.0) | - | 1.00 | |||
| Grade | I/II | 21 (91.3) | 2 (8.7) | 1.00 | 0.037 | 33 (66.0) | 17 (34.0) | 4.70 (0.17–129.67) | 0.269 |
| III | 95 (66.4) | 48 (33.6) | 4.37 (1.10–17.40) | 55 (55.6) | 44 (44.4) | 7.21 (0.27–193.57) | |||
| Diagnosis to Chemotherapy (months) | Median | 1.31 | 1.22 | 0.96 (0.81–1.13) | 0.583 | 1.35 | 1.25 | 0.99 (0.96–1.02) | 0.545 |
| Diagnosis to Surgery (months) | Median | 6.06 | 6.44 | 1.00 (0.97–1.03) | 0.986 | 6.34 | 6.24 | 0.99 (0.95–1.04) | 0.728 |
| Race | NHW | 87 (70.2) | 37 (29.8) | 1.00 | 0.772 | 82 (60.3) | 54 (39.7) | 1.00 | 0.761 |
| NHB | 32 (68.1) | 15 (31.9) | 1.11 (0.54–2.29) | 16 (64.0) | 9 (36.0) | 0.87 (0.36–2.11) | |||
| HTN | No | 82 (67.2) | 40 (32.8) | 1.00 | 0.311 | 74 (62.7) | 44 (37.3) | 1.00 | 0.427 |
| Yes | 37 (75.5) | 12 (24.5) | 0.68 (0.32–1.44) | 24 (55.8) | 19 (44.2) | 1.33 (0.66–2.70) | |||
| DM | No | 105 (69.1) | 47 (30.9) | 1.00 | 0.752 | 92 (60.1) | 61 (39.9) | 1.00 | 0.497 |
| Yes | 14 (73.7) | 5 (26.3) | 0.84 (0.29–2.44) | 6 (75.0) | 2 (25.0) | 0.58 (0.12–2.81) | |||
| CKD | No | 116 (69.0) | 52 (31.0) | 1.00 | 0.512 | 98 (61.6) | 61 (38.4) | 1.00 | 0.275 |
| Yes | 3 (100.0) | 0.32 (0.01–9.86) | 2 (100.0) | 8.01 (0.19–335.19) | |||||
| Asthma | No | 114 (70.4) | 48 (29.6) | 1.00 | 0.341 | 87 (60.0) | 58 (40.0) | 1.00 | 0.551 |
| Yes | 5 (55.6) | 4 (44.4) | 1.93 (0.50–7.49) | 11 (68.8) | 5 (31.3) | 0.72 (0.24–2.15) | |||
| Cardiac disease | No | 116 (69.5) | 51 (30.5) | 1.00 | 0.977 | 92 (59.7) | 62 (40.3) | 1.00 | 0.274 |
| Yes | 3 (75.0) | 1 (25.0) | 0.97 (0.11–8.44) | 6 (85.7) | 1 (14.3) | 0.34 (0.05–2.34) | |||
| Hypothyroidism | No | 100 (66.7) | 50 (33.3) | 1.00 | 0.054 | 88 (62.0) | 54 (38.0) | 1.00 | 0.433 |
| Yes | 19 (90.5) | 2 (9.5) | 0.25 (0.06–1.02) | 10 (52.6) | 9 (47.4) | 1.47 (0.56–3.85) | |||
| RDI (OR for 10% change) | Mean/SD | 90.91/17.72 | 91.15/14.54 | 1.00 (0.79–1.25) | 1.000 | 91.05/14.97 | 89.72/17.8 | 0.95 (0.75–1.19) | 0.676 |
| Chemotherapy | Anthracycline based | 13 (92.9) | 1 (7.1) | 1.00 | 0.373 | 10 (90.9) | 1 (9.1) | 1.00 | 0.136 |
| Taxane based | 10 (76.9) | 3 (23.1) | 3.00 (0.35–25.77) | 42 (52.5) | 38 (47.5) | 6.34 (1.01–39.93) | |||
| Both | 95 (66.4) | 48 (33.6) | 4.57 (0.77–27.11) | 43 (65.2) | 23 (34.8) | 3.78 (0.59–24.22) | |||
| Other | 1 (100.0) | 2.94 (0.02–384.30) | 3 (75.0) | 1 (25.0) | 3.00 (0.18–48.68) | ||||
| cT | cT1 | 9 (60.0) | 6 (40.0) | 1.00 | 0.132 | 10 (52.6) | 9 (47.4) | 1.00 | 0.074 |
| cT2 | 61 (65.6) | 32 (34.4) | 0.77 (0.25–2.36) | 43 (52.4) | 39 (47.6) | 1.00 (0.37–2.73) | |||
| cT3 | 19 (67.9) | 9 (32.1) | 0.71 (0.19–2.61) | 27 (73.0) | 10 (27.0) | 0.42 (0.13–1.34) | |||
| cT4 | 29 (87.9) | 4 (12.1) | 0.22 (0.05–0.94) | 17 (77.3) | 5 (22.7) | 0.35 (0.09–1.31) | |||
| cN | cN0 | 52 (70.3) | 22 (29.7) | 1.00 | 0.968 | 48 (59.3) | 33 (40.7) | 1.00 | 0.529 |
| cN1 | 43 (67.2) | 21 (32.8) | 1.15 (0.56–2.37) | 40 (62.5) | 24 (37.5) | 0.88 (0.45–1.72) | |||
| cN2 | 16 (72.7) | 6 (27.3) | 0.92 (0.32–2.63) | 5 (45.5) | 6 (54.5) | 1.71 (0.48–6.07) | |||
| cN3 | 8 (72.7) | 3 (27.3) | 0.96 (0.24–3.87) | 4 (100.0) | 0.16 (0.01–4.35) | ||||
Abbreviations: BMI-Body mass index, HTN-Hypertension, DM-Diabetes Mellitus, CKD-Chronic Kidney Disease, RDI-Relative Dose Intensity, cT-AJCC (eighth edition) Clinical category of primary tumor, cN- AJCC (eighth edition) Clinical category of regional node involvement, pCR-Complete Pathological Response, yp1-AJCC Pathologic Stage I, yp2- AJCC Pathologic Stage II, yp3- AJCC Pathologic Stage III, SD-standard deviation
Next, we examined the association with factors influencing overall survival (OS) and recurrence free survival (RFS) among NHB and NHW patients and results are depicted in Table 3. In patients with TNBC, older age, higher clinical or pathological stage, and increased time interval between diagnosis to surgery were significantly associated with worse OS. The factors associated with worse RFS included advanced clinical/pathological stage, tumor size or nodal involvement and increased time interval between diagnosis to surgery. There was no observed association between race and survival (OS HR: 1.13 95%CI 0.66–1.93, RFS HR: 1.43 95%CI 0.89–2.31).
Table 3:
Association of demographic and clinical factors with overall survival and recurrence free survival in TNBC and HER-2 positive breast cancer patients.
| TNBC | HER2-positive | ||||||||
|---|---|---|---|---|---|---|---|---|---|
| Overall Survival | Recurrence free Survival | Overall Survival | Recurrence free survival | ||||||
| Variable | HR (95% CI) | P-value | HR (95% CI) | P-value | HR (95% CI) | P-value | HR (95% CI) | P-value | |
| Age (years) | Unit Inc. | 1.03 (1.00, 1.05) | 0.023 | 1.01 (0.99, 1.03) | 0.165 | 1.02 (1.00, 1.05) | 0.087 | 1.00 (0.98, 1.02) | 0.923 |
| <40 | 1.00 | 0.033 | 1.00 | 0.076 | 1.00 | 0.015 | 1.00 | 0.063 | |
| 40–50 | 0.48 (0.22, 1.05) | 0.48 (0.24, 0.96) | 2.17 (0.79, 5.93) | 1.61 (0.70, 3.67) | |||||
| 50–60 | 0.68 (0.33, 1.42) | 0.64 (0.33, 1.25) | 0.79 (0.26, 2.44) | 0.59 (0.23, 1.49) | |||||
| 60–70 | 1.04 (0.47, 2.28) | 0.97 (0.47, 1.98) | 1.36 (0.41, 4.46) | 0.81 (0.29, 2.25) | |||||
| 70+ | 1.74 (0.73, 4.15) | 1.28 (0.56, 2.94) | 3.65 (1.21, 11.05) | 1.68 (0.64, 4.40) | |||||
| BMI (kg/m2) | Unit Inc. | 1.01 (0.98, 1.05) | 0.405 | 1.02 (0.99, 1.05) | 0.133 | 1.02 (0.97, 1.07) | 0.488 | 1.01 (0.97, 1.06) | 0.487 |
| Insurance | Private | 1.00 | 0.755 | 1.00 | 0.866 | 1.00 | 0.174 | 1.00 | 0.043 |
| Gov | 0.82 (0.48, 1.38) | 0.88 (0.54, 1.41) | 1.51 (0.83, 2.75) | 1.75 (1.02, 3.02) | |||||
| Clinical Stage | 1 | 1.00 | 0.003 | 1.00 | 0.005 | 1.00 | <.001 | 1.00 | <.001 |
| 2 | 0.93 (0.22, 3.94) | 0.57 (0.20, 1.61) | 0.59 (0.17, 2.09) | 0.94 (0.28, 3.16) | |||||
| 3 | 2.21 (0.53, 9.17) | 1.24 (0.44, 3.47) | 2.49 (0.75, 8.24) | 3.28 (0.99, 10.81) | |||||
| Grade | I | - | 0.275 | - | 0.194 | 1.00 | 0.597 | 1.00 | 0.729 |
| II | 1.00 | 1.00 | 0.53 (0.15, 1.86) | 0.61 (0.18, 2.08) | |||||
| III | 0.70 (0.37, 1.32) | 0.68 (0.38, 1.22) | 0.55 (0.16, 1.85) | 0.66 (0.20, 2.17) | |||||
| Diagnosis to Chemotherapy (months) | Unit Inc. | 0.98 (0.88, 1.09) | 0.709 | 0.98 (0.88, 1.09) | 0.654 | 1.01 (0.99, 1.03) | 0.283 | 1.01 (0.99, 1.03) | 0.252 |
| Diagnosis to Surgery (months) | Unit Inc. | 1.04 (1.02, 1.06) | <.001 | 1.04 (1.02, 1.06) | 0.001 | 1.01 (1.00, 1.03) | 0.130 | 1.02 (1.00, 1.04) | 0.028 |
| Race | NHW | 1.00 | 0.655 | 1.00 | 0.139 | 1.00 | 0.083 | 1.00 | 0.392 |
| NHB | 1.13 (0.66, 1.93) | 1.43 (0.89, 2.31) | 1.87 (0.92, 3.81) | 1.35 (0.68, 2.70) | |||||
| HTN | No | 1.00 | 0.579 | 1.00 | 0.700 | 1.00 | 0.059 | 1.00 | 0.441 |
| Yes | 1.16 (0.69, 1.95) | 1.10 (0.68, 1.77) | 1.88 (0.98, 3.62) | 1.27 (0.69, 2.36) | |||||
| DM | No | 1.00 | 0.428 | 1.00 | 0.309 | 1.00 | 0.075 | 1.00 | 0.287 |
| Yes | 1.35 (0.64, 2.84) | 1.42 (0.73, 2.76) | 2.92 (0.90, 9.51) | 1.89 (0.59, 6.08) | |||||
| CKD | No | 1.00 | 0.053 | 1.00 | 0.093 | 1.00 | 0.991 | 1.00 | 0.412 |
| Yes | 4.04 (0.98, 16.62) | 3.34 (0.82, 13.68) | 0.00 (0.00, .) | 2.30 (0.31, 16.88) | |||||
| Asthma | No | 1.00 | 0.348 | 1.00 | 0.845 | 1.00 | 0.099 | 1.00 | 0.325 |
| Yes | 0.51 (0.12, 2.08) | 0.90 (0.33, 2.48) | 2.07 (0.87, 4.91) | 1.54 (0.65, 3.60) | |||||
| Cardiac disease | No | 1.00 | 0.611 | 1.00 | 0.924 | 1.00 | 0.595 | 1.00 | 0.419 |
| Yes | 0.60 (0.08, 4.32) | 1.07 (0.26, 4.38) | 0.58 (0.08, 4.25) | 0.44 (0.06, 3.20) | |||||
| Hypothyroidism | No | 1.00 | 0.641 | 1.00 | 0.984 | 1.00 | 0.629 | 1.00 | 0.458 |
| Yes | 1.18 (0.58, 2.39) | 0.99 (0.51, 1.94) | 0.75 (0.23, 2.42) | 0.68 (0.24, 1.89) | |||||
| RDI | 10% change | 0.97 (0.81, 1.15) | 0.718 | 0.97 (0.83, 1.15) | 0.747 | 0.86 (0.69, 1.07) | 0.177 | 0.92 (0.74, 1.15) | 0.471 |
| Chemotherapy | Anthracycline based | 1.00 | 0.065 | 1.00 | 0.177 | 1.00 | 0.253 | 1.00 | 0.370 |
| Taxane based | 1.22 (0.34, 4.38) | 1.23 (0.38, 3.95) | 0.60 (0.22, 1.67) | 0.55 (0.20, 1.51) | |||||
| Both | 1.05 (0.45, 2.45) | 1.16 (0.52, 2.56) | 0.58 (0.22, 1.55) | 0.70 (0.27, 1.84) | |||||
| Other | 17.45 (1.93, 157.47) | 10.94 (1.28, 93.60) | 2.16 (0.41, 11.42) | 1.65 (0.31, 8.69) | |||||
| cT | cT1 | 1.00 | 0.002 | 1.00 | 0.003 | 1.00 | 0.002 | 1.00 | 0.001 |
| cT2 | 0.99 (0.34, 2.85) | 0.90 (0.38, 2.16) | 1.70 (0.38, 7.57) | 2.37 (0.55, 10.21) | |||||
| cT3 | 1.58 (0.51, 4.93) | 1.03 (0.38, 2.77) | 4.17 (0.95, 18.23) | 5.86 (1.36, 25.27) | |||||
| cT4 | 2.90 (1.00, 8.44) | 2.38 (0.97, 5.83) | 6.57 (1.47, 29.34) | 7.38 (1.67, 32.68) | |||||
| cN | cN0 | 1.00 | 0.006 | 1.00 | 0.003 | 1.00 | 0.059 | 1.00 | 0.013 |
| cN1 | 2.20 (1.22, 3.97) | 1.94 (1.15, 3.27) | 1.81 (0.89, 3.68) | 1.46 (0.79, 2.72) | |||||
| cN2 | 1.76 (0.78, 3.95) | 1.55 (0.74, 3.26) | 3.43 (1.35, 8.73) | 4.02 (1.74, 9.25) | |||||
| cN3 | 4.39 (1.81, 10.65) | 2.93 (0.65, 13.17) | 1.95 (0.45, 8.45) | ||||||
| Pathologic Stage | pCR | 1.00 | <.001 | 1.00 | <.001 | 1.00 | 0.135 | 1.00 | 0.268 |
| yp1 | 1.64 (0.66, 4.09) | 1.45 (0.70, 3.01) | 2.30 (0.89, 5.98) | 1.63 (0.70, 3.82) | |||||
| yp2 | 3.17 (1.30, 7.78) | 1.84 (0.84, 4.04) | 2.18 (0.91, 5.21) | 1.81 (0.88, 3.75) | |||||
| yp3 | 7.46 (3.44, 16.18) | 5.29 (2.79, 10.02) | 2.86 (1.16, 7.06) | 2.08 (0.95, 4.53) | |||||
Abbreviations: BMI-Body mass index, HTN-Hypertension, DM-Diabetes Mellitus, CKD-Chronic Kidney Disease, RDI-Relative Dose Intensity, cT-AJCC (eighth edition) Clinical category of primary tumor, cN- AJCC (eighth edition) Clinical category of regional node involvement, pCR-Complete Pathological Response, yp1-AJCC Pathologic Stage I, yp2- AJCC Pathologic Stage II, yp3- AJCC Pathologic Stage III, SD-standard deviation
Among patients with HER2-positive disease, older age and advanced clinical stage were found to be significantly associated with worse OS. Government insurance, advanced clinical stage, and increased time from original diagnosis to surgery were found to be significantly associated with worse RFS. We did not observe any significant association between race and survival (OS:1.87 95%CI: 0.92–3.81, RFS:1.35 95% CI: 0.68–2.70).
Kaplan Meier curves with log-rank test were used to compare OS (Figures 1A and 1B) and RFS (Figure 2A and 2B) between NHW and NHB patients, with no significant difference observed in either TNBC or HER2-positive patients.
Figure 1: Overall survival in (A) TNBC and (B) HER2-positive breast cancer patients by race.
Figure 2: Recurrence free survival in (A) TNBC and (B) HER2-positive breast cancer patients by race.
After controlling for potential confounders, including, age, insurance status, comorbidities, stage, treatment regimens, race was not significantly associated with pCR, OS, or RFS (Table 4). However, there was no statistically significant difference in the pCR between NHB and NHW patients (32.6 vs 29.8% OR: 1.05 95% CI 0.41–2.70). Among patients with HER2-positive breast cancer, there was no statistical difference in pCR between NHB and NHW women (33.3 vs 39.7% OR: 0.73 95% CI 0.25–2.15).
Table 4:
Multivariate analysis for pathologic complete response (pCR), recurrence free and overall survival in TNBC and Her-2 positive breast cancer patients by race.
| TNBC | HER2-positive | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| pCR | Overall Survival | Recurrence-free Survival | pCR | Overall Survival | Recurrence-free Survival | |||||||
| Group | OR 95% CI) | P-value | HR 95% CI) | P-value | HR 95% CI) | P-value | OR 95% CI) | P-value | HR 95% CI) | P-value | HR 95% CI) | P-value |
| NHW | 1.00 | 0.92 | 1.00 | 0.94 | 1.00 | 0.45 | 1.00 | 0.57 | 1.00 | 0.55 | 1.00 | 0.78 |
| NHB | 1.05 (0.41, 2.70) | 1.03 (0.53, 1.99) | 1.25 (0.70, 2.23) | 0.73 (0.25, 2.15) | 1.34 (0.51, 3.50) | 1.13 (0.47, 2.72) | ||||||
Furthermore, we examined the trends in the rates of pCR among NHW and NHB women in both cohorts across different time periods. Figure 3A shows pCR trends over time in patients with TNBC. X axis depicts various time periods and Y axis depicts the rates of pCR corresponding to the period. In patients with TNBC, there is increase in the rates of pCR in both NHB (23% to 30%) and NHW (21% to 38%) cohorts, without any statistically significant difference between the two groups.
Figure 3: Trends in pathologic complete response (pCR) over time by race in (A) TNBC and (B) HER2-positive breast cancer patients.
In patients with HER2-positive breast cancer after trastuzumab was approved for non-metastatic disease in 2006, there was a time effect in pCR in both cohorts with pCR increasing over the observed time period (NHW 38% to 51% and NHB 0% to 80%), yet we did not observe any statistically significant difference between races (Figure 3B).
Discussion
This retrospective study at our institution aimed to analyze potential differences in pCR and survival outcomes between NHB and NHW women in those with TNBC and HER2-positive breast cancer. Our study found that NHB women with TNBC were more likely to have higher BMI, Medicare/Medicaid usage, more nodal involvement at diagnosis, accompanied by higher rates of hypertension and diabetes mellitus in this population. NHB women with HER2 positive breast cancer had increased time from diagnosis to both surgery/chemotherapy and higher rates of hypertension. Although many of these factors could affect the type and dosage of chemotherapy and survival outcomes, RDI and survival outcomes were not significantly different between the two racial groups. We showed that there was no statistically significant difference in pCR by race in patients of either breast cancer subtypes treated at our center. Interestingly, we also noted that in both the TNBC and HER2-positive cohorts, NHB women have a consistent uptrend in pCR rates over time.
pCR has shown to be associated with improved survival outcomes, especially in patients with TNBC and HER2-positive breast cancer [3]. Given this association, several studies have attempted to investigate the relationship between race and pCR in different breast cancer subtypes. Dawood et al in their retrospective study of 471 patients with TNBC treated at University of Texas M. D. Anderson Cancer Center observed that there was no association between race and pCR [6]. This was also mirrored in a single center study at University of North Carolina that did not find any evidence that pCR was affected by racial characteristics across different breast cancer subtypes [5]. In contrast to the previous two single center studies, Balmanoukian, A., et al observed lower pCR rates in Black women with TNBC treated at John Hopkins University Hospital [12]. It must be noted that the study had a small sample size of 23 patients and could have been swayed by the predominance of basal like tumors in Black patients, as mentioned by the authors. An analysis of the National Cancer Database (NCDB) done by Killelea and colleagues, observed that pCR was lower in Black women with TNBC and HER2 positive breast cancer [4].This study evaluated treatment delays but could not capture information related to optimal chemotherapy dosing due to limitations of the NCDB. It should also be noted that it reviewed patients during a shorter time frame (treated in 2010 and 2011 only). Nonetheless, this begs the question if such racial differences in pCR would exist if it were adjusted based on the “adequacy” of chemotherapy with respect to dosing and duration.
Given the above-mentioned discrepancies, it is vital to understand if pCR differences exist despite controlling for confounding factors, especially in retrospective studies. It must be noted that our study is unique because it is the first retrospective study that examined optimal treatment as a function of RDI rather than categorizing it as the type of chemotherapy received. RDI is an effective metric of optimal treatment as it takes both treatment delays and dosages into account with RDI ≥ 75% being associated with better survival in TNBC patients [13, 14]. However, irrespective of the metric used to determine optimal treatment, all these studies when taken together highlight the fact that racial disparities in pCR are not as evident when patients are treated optimally, specifically referring to NAC in our patient population. Our results are in concordance with the meta-analysis of Alliance clinical trials where Black patients with breast cancer receiving similar treatment did not have inferior pCR compared to White patients [15].
We also noted improving pCR trends over time in our NHB patients. It must be placed in context of socio-economic factors that affect access to care, including the enactment of the Affordable Care Act which has enabled benefits in coverage for Black patients[16].
Our study has several limitations. Although we treated a sizable number of NHB patients both in the TNBC and HER2-positive subtypes, our patient population was predominantly NHW patients. By virtue of being a retrospective analysis of our database, we could not have accounted for missing information related to type and dosage of chemotherapy, especially for some patients diagnosed closer to 2000 when our institute was transitioning to its current electronic health record system. It should also be noted that the period of our retrospective analysis is mostly prior to the approval of adjuvant capecitabine [17] and TDM-1 [18] for residual TNBC and HER2-positive breast cancer respectively which could affect the OS and RFS. However, this would have no bearing on the pCR rates. Additionally, checkpoint inhibitors are now approved for neoadjuvant therapy in patients with TNBC and it would be of importance to determine if there are racial differences in pCR rates with their incorporation[19]. We also acknowledge that the sample size could have restricted the desired power to evaluate the possibility of finding disparities in survival outcomes between races.
To summarize, our study argues against the existence of intrinsic race-dependent differences in response to neoadjuvant treatment (pCR and survival outcomes) in patients with TNBC and HER2-positive breast cancer treated in the same cancer center, which was proposed by large NCDB studies [4] but questioned by some institutional studies. In addition, we observed a positive trend in pCR rates over time in our NHB women. This finding which contrasts with those noted in the larger NCDB analysis[4] raises the possible role of differences in access to specialized cancer centers, treatment intensity (accounting for therapeutic agents used, dose given, treatment delays), co-morbidities including higher BMI/obesity, geographic disparities affecting outcomes as well as the known genetic heterogeneity amongst African Americans across the United States contributing to our results. Our study addresses an important knowledge gap as it analyzed clinical outcomes while adjusting for RDI, especially since prior studies have shown that RDI is lower in NHB breast cancer patients[20]. Larger prospective studies on these factors and the intersection of race, genetic heterogeneity and geography are warranted.
Acknowledgment
Dr. Takabe was supported by United States National Cancer Institute (NCI) grants R01CA251545, R01CA250412, and R37CA248018, and United States Department of Defense Breast Cancer Research Program grants W81XWH-19-1-0674 and W81XWH-19-1-0111. Dr. Ambrosone is a recipient of funding from the Breast Cancer Research Foundation.
Funding
Research reported in this publication was supported by the National Center for Advancing Translational Sciences of the National Institutes of Health under award Number KL2TR001413 and UL1TR001412. The content is solely the responsibility of the authors and does not necessarily represent the official views of the NIH. This research was also supported by the National Cancer Institute Cancer Center Support Grant (P30CA016056).
Footnotes
Statement and Declarations
Ethics Approval
This research study was conducted retrospectively from data obtained for clinical purposes. The study was approved by Roswell Park Institutional Review Board
Conflict of interest
The authors declare that they have no conflict of interest.
References:
- 1.Breast SEER 5 year survival rates 2011–2017. 2021. [Google Scholar]
- 2.Siddharth S and Sharma D, Racial Disparity and Triple-Negative Breast Cancer in African-American Women: A Multifaceted Affair between Obesity, Biology, and Socioeconomic Determinants. Cancers (Basel), 2018. 10(12). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Cortazar P, et al. , Pathological complete response and long-term clinical benefit in breast cancer: the CTNeoBC pooled analysis. Lancet, 2014. 384(9938): p. 164–72. [DOI] [PubMed] [Google Scholar]
- 4.Killelea BK, et al. , Racial Differences in the Use and Outcome of Neoadjuvant Chemotherapy for Breast Cancer: Results From the National Cancer Data Base. J Clin Oncol, 2015. 33(36): p. 4267–76. [DOI] [PubMed] [Google Scholar]
- 5.Tichy JR, et al. , Race, response to chemotherapy, and outcome within clinical breast cancer subtypes. Breast Cancer Res Treat, 2015. 150(3): p. 667–74. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Dawood S, et al. , Triple receptor-negative breast cancer: the effect of race on response to primary systemic treatment and survival outcomes. J Clin Oncol, 2009. 27(2): p. 220–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Bryc K, et al. , The genetic ancestry of African Americans, Latinos, and European Americans across the United States. Am J Hum Genet, 2015. 96(1): p. 37–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Elston CW and Ellis IO, Pathological prognostic factors in breast cancer. I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology, 1991. 19(5): p. 403–10. [DOI] [PubMed] [Google Scholar]
- 9.Cserni G, Histological type and typing of breast carcinomas and the WHO classification changes over time. Pathologica, 2020. 112(1): p. 25–41. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Harvey JM, et al. , Estrogen receptor status by immunohistochemistry is superior to the ligand-binding assay for predicting response to adjuvant endocrine therapy in breast cancer. J Clin Oncol, 1999. 17(5): p. 1474–81. [DOI] [PubMed] [Google Scholar]
- 11.Wolff AC, et al. , Recommendations for human epidermal growth factor receptor 2 testing in breast cancer: American Society of Clinical Oncology/College of American Pathologists clinical practice guideline update. J Clin Oncol, 2013. 31(31): p. 3997–4013. [DOI] [PubMed] [Google Scholar]
- 12.Balmanoukian A, et al. , African American women who receive primary anthracycline- and taxane-based chemotherapy for triple-negative breast cancer suffer worse outcomes compared with white women. J Clin Oncol, 2009. 27(22): p. e35–7; author reply e38–9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Usiskin I, et al. , Association of relative dose intensity with BMI and pathologic complete response in patients treated with neoadjuvant chemotherapy for breast cancer. Breast Cancer Res Treat, 2021. 186(1): p. 191–197. [DOI] [PubMed] [Google Scholar]
- 14.Zhang L, et al. , Impact of chemotherapy relative dose intensity on cause-specific and overall survival for stage I-III breast cancer: ER+/PR+, HER2- vs. triple-negative. Breast Cancer Res Treat, 2018. 169(1): p. 175–187. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Warner ET, et al. , Impact of race, ethnicity, and BMI on achievement of pathologic complete response following neoadjuvant chemotherapy for breast cancer: a pooled analysis of four prospective Alliance clinical trials (A151426). Breast Cancer Res Treat, 2016. 159(1): p. 109–18. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Artiga S, Orgera K, and Damico A. Changes in Health Coverage by Race and Ethnicity since the ACA, 2010–2018. 2020. March 5th]; Available from: https://www.kff.org/racial-equity-and-health-policy/issue-brief/changes-in-health-coverage-by-race-and-ethnicity-since-the-aca-2010-2018/.
- 17.Masuda N, et al. , Adjuvant Capecitabine for Breast Cancer after Preoperative Chemotherapy. N Engl J Med, 2017. 376(22): p. 2147–2159. [DOI] [PubMed] [Google Scholar]
- 18.von Minckwitz G, et al. , Trastuzumab Emtansine for Residual Invasive HER2-Positive Breast Cancer. N Engl J Med, 2019. 380(7): p. 617–628. [DOI] [PubMed] [Google Scholar]
- 19.Schmid P, et al. , Pembrolizumab for Early Triple-Negative Breast Cancer. N Engl J Med, 2020. 382(9): p. 810–821. [DOI] [PubMed] [Google Scholar]
- 20.Griggs JJ, et al. , Racial disparity in the dose and dose intensity of breast cancer adjuvant chemotherapy. Breast Cancer Res Treat, 2003. 81(1): p. 21–31. [DOI] [PubMed] [Google Scholar]