Purpose:

The incidence of triple-negative breast cancer (TNBC) is higher among Black or African American (AA) women, yet they are underrepresented in clinical trials. To evaluate safety and efficacy of durvalumab concurrent with neoadjuvant chemotherapy for stage I–III TNBC by race, we enrolled additional AA patients to a Phase I/II clinical trial.

Patients and Methods:

Our study population included 67 patients. The primary efficacy endpoint was pathologic complete response (pCR; ypT0/is, N0) rate. χ2 tests were used to evaluate associations between race and baseline characteristics. Cox proportional hazards models were used to assess association between race and overall survival (OS) and event-free survival (EFS). Multivariate logistic regression analyses were used to evaluate associations between race and pCR, immune-related adverse events (irAE) and recurrence.

Results:

Twenty-one patients (31%) self-identified as AA. No significant associations between race and baseline tumor stage (P = 0.40), PD-L1 status (0.92), and stromal tumor–infiltrating lymphocyte (sTIL) count (P = 0.57) were observed. pCR rates were similar between AA (43%) and non-AA patients (48%; P = 0.71). Three-year EFS rates were 78.3% and 71.4% in non-AA and AA patients, respectively [HR, 1.451; 95% confidence interval (CI), 0.524–4.017; P = 0.474]; 3-year OS was 87% and 81%, respectively (HR, 1.72; 95% CI, 0.481–6.136; P = 0.405). The incidence of irAEs was similar between AA and non-AA patients and no significant associations were found between irAEs and pathologic response.

Conclusions:

pCR rates, 3-year OS and EFS after neoadjuvant immunotherapy and chemotherapy were similar in AA and non-AA patients. Toxicities, including the frequency of irAEs, were also similar.

Translational Relevance

Despite higher prevalence of triple-negative breast cancer (TNBC) among Black or African American (AA) women compared with white women, AA women are underrepresented in TNBC clinical trials, particularly those involving novel immunotherapy agents. We previously showed that the combination of durvalumab and neoadjuvant chemotherapy was well tolerated and resulted in high pathologic complete response (pCR) rate. In this study, we recruited additional AA patients to our trial for a final AA cohort comprising 31% of the total population. We show that pCR rates, 3-year overall survival and event-free survival rates as well as immune-related toxicities are similar in AA and non-AA women. These results suggest that clinical outcomes are similar regardless of race when patients receive identical treatment and follow-up in a highly structured study environment.

The addition of checkpoint inhibitors to neoadjuvant chemotherapy significantly improved pathologic complete response (pCR) rates and event-free survival (EFS) in early-stage triple-negative breast cancer (TNBC). Pembrolizumab concurrent with neoadjuvant chemotherapy (followed by single agent adjuvant pembrolizumab) is approved by the FDA for the treatment of stage II/III TNBC based on results of the KEYNOTE-522 trial. This phase III study demonstrated an increase in pCR rate from 55.6% with placebo plus chemotherapy to 63.0% with the addition of pembrolizumab to chemotherapy and an improvement in 3-year EFS rate of 84.5% in the pembrolizumab group versus 76.8% in the placebo group (1, 2). In the randomized phase II GeparNuevo trial (3), the addition of durvalumab versus placebo to nab-paclitaxel followed by dose-dense epirubicin/cyclophosphamide (EC) in non-metastatic TNBC also resulted in a numerically improved pCR rate of 53.4% versus 44.2% (P = 0.287) and an improvement of 3-year invasive disease-free survival (iDFS) of 84.9% versus 76.9% (P = 0.0559), distant DFS (dDFS) of 91.4% versus 79.5% (P = 0.0148) and overall survival (OS) of 95.1% versus 83.1% (P = 0076) with durvalumab versus placebo, respectively.

We previously reported results of a phase I/II single arm clinical trial of durvalumab, a monoclonal human immunoglobulin G1к antibody that binds to programmed death-ligand 1 (PD-L1) and inhibits its interaction with PD-1 and CD80 (B7.1), administered concurrently with sequential weekly nab-paclitaxel and dose-dense doxorubicin/cyclophosphamide (ddAC) neoadjuvant chemotherapy in stage I-III TNBC (4). The primary endpoint of the study, pCR rate in the phase II population (n = 55), was 44% [95% confidence interval (CI), 30%–57%]. We also showed that achieving pCR was statistically significantly associated with higher stromal tumor–infiltrating lymphocyte (sTIL) counts but not with PD-L1 positivity. Immune-related adverse events (irAE) were consistent with the known toxicities of immune checkpoint inhibitors (4).

Many studies have shown that Black or African American (AA) women have a higher prevalence of TNBC than women of European descent (5). AA women also have higher overall breast cancer–related mortality compared with women of European descent (6–8). Studies of the impact of race on survival in women with TNBC have had mixed results. Some showed similar pCR rates by ethnicity or race (9–11); whereas others showed lower pCR rates in AA patients and worse survival (12–15). More recent reports suggest that the significant improvements that we have seen over the past several decades in breast cancer diagnosis and treatment have not resulted in equally improved outcomes for AA women (16–19). It is also evident that AA women are underrepresented in breast cancer clinical trials. Even more worrisome is the fact that the proportion of trial participants who are AA has declined in recent years (20). Lower representation of AA patients in breast cancer trials may impede the generalizability of safety and efficacy of novel therapies to this population. The goal of this study was to increase the accrual of AA patients to our trial to assess for potential differences in safety, efficacy, and immunologic biomarker characteristics by race in patients with TNBC receiving neoadjuvant durvalumab concurrent with chemotherapy.

Study design

The full study design was reported in the original publication (4). In summary, patients received durvalumab concurrent with weekly nab-paclitaxel (100 mg/m2) × 12 treatments followed by doxorubicin (60 mg/m2) and cyclophosphamide (600 mg/m2) every two weeks (AC) × 4 treatments. The Phase II dose of durvalumab was 10 mg/kg, administered every two weeks throughout the entire neoadjuvant treatment phase (total of 10 doses). We now report updated results, including an AA extension cohort added to the trial. The primary efficacy objective was to assess pCR rate in patients who received the recommended Phase II dose. This was an investigator-initiated trial, and ethical approval was obtained from the Yale Human Investigations Committee (Yale University, HIC# 1409014537) and its Clinicaltrials.gov identifier was NCT02489448. The study was conducted in accordance with the Declaration of Helsinki and adhered to Good Clinical Practice guidelines. AstraZeneca provided study drug and funding for the trial but played no role in the study design, collection/analysis of data, or article preparation.

Patients and assessments

All patients signed written informed consent before participation in the study. Patients with clinically stage I–III, triple negative breast cancer, defined as ER and PR<1% positive and HER2 negative (IHC 0, 1+ or 2+, or FISH negative), for whom systemic chemotherapy was indicated according to NCCN treatment guidelines were eligible (21). Exclusion criteria included contraindications for anthracycline, paclitaxel or anti–PD-L1 therapies (e.g., active autoimmune disease, live vaccines within 30 days, prior transplants, immune deficiency, active immunosuppressive medications). Adverse events (AE) were graded according to NCI CTCAE v4.03. All patients who received at least one dose of durvalumab were included in the toxicity analysis. Surgery was performed within 4 weeks of completion of neoadjuvant chemotherapy and extent of residual cancer assessed by local pathologist as part of routine care. Residual cancer burden (RCB) was assessed centrally by a breast pathologist (E. Reisenbichler). pCR was defined as no invasive cancer in the breast and lymph nodes (ypN0/is,N0). EFS was defined as the absence of disease progression that precluded definitive surgery; local or distant recurrence or a second primary tumor; or death from any cause. OS was defined by death from any cause from the date of diagnosis. Cox proportional hazards model was used to assess association between race and survival. Kaplan–Meier methods were used to estimate survival medians. Multivariate logistic regression analyses were performed to assess for independent predictors of pCR, breast cancer recurrence and irAEs.

Biomarker analysis

PD-L1 expression on formalin-fixed paraffin-embedded pretreatment biopsies was assessed with chromogenic IHC using the VENTANA PD-L1 (SP263; Roche Cat# 790–4905, RRID:AB_2819099) and E1L3N (Cell Signaling Technology Cat# 51296, RRID:AB_2799389) assays performed according to the product labels. PD-L1 positivity was determined by consensus review of 2 pathologists and ≥1% staining on immune or tumor cells was considered positive. We initially selected SP263 and the combined score of 1% or higher for our cutoff value to be consistent with GeparNuevo that used the same immune checkpoint inhibitor in a similar patient population and because the Ventana SP263 assay was being developed at the time to predict response to durvalumab in other disease settings (22–24). For the last seven patients enrolled in our study, we used the E1L3N antibody clone for assessment of PD-L1 positivity, due to lack of availability of the SP263 antibody and the fact that it had not been approved for PD-L1 quantification in breast cancer. Multiple studies have shown high concordance in PD-L1 staining of these two antibodies, including in TNBC (22–24). The percentage of sTILs was assessed on hematoxylin and eosin–stained slides and calculated as the area occupied by mononuclear inflammatory cells that are located within the stroma between carcinoma cells without directly contacting the carcinoma cell nests, over the total stromal area within the boundary of the tumor. Uni- and multivariate logistic regression analyses were used to identify predictors of recurrence, pCR and irAEs. χ2 tests were used to assess the association between PD-L1 positivity and pCR rate and median sTILs percentage between cases with pCR and RD were compared using the Mann–Whitney U test.

Data availability

The data generated in this study are not publicly available due to protected patient information included in but deidentified data are available upon reasonable request from the corresponding author.

Patient population

Sixty-seven patients consented for the trial between December 18, 2015, and December 29, 2020. Baseline characteristics are shown in Table 1. Three patients did not proceed to surgery: One developed irreversible altered mental status attributed to Miller-Fischer variant of Guillain-Barre syndrome (GBS) and family opted for comfort care, the second patient completed treatment but died suddenly in her home before undergoing surgery, and the third patient died of a stroke before surgery.

Table 1.

Patient and tumor characteristics.

AA (n = 21)non-AA (n = 46)All patients (n = 67)P
Age (median/range) 52 (33–79) 50.5 (27–69) 51 (27–79) 0.434 
BMI (median/range) 29.2 (24.8–49) 27.4 (19.6–42.1) 27.5 (19.6–49) 0.075 
Charlson Comorbidity Index (median/range) 3 (2–7) 3 (2–8) 3 (2–8) 0.317 
Tumor characteristics (pre-treatment)—n (%) 
cT stage    0.403 
   T1 6 (29) 18 (39) 24 (36)  
   T2 11 (52) 23 (50) 34 (51)  
   T3 3 (14) 5 (11) 8 (12)  
   T4 1 (5) 0 (0) 1 (1)  
cN stage    0.457 
   N0 13 (62) 24 (52) 37 (55)  
   N1 6 (28) 19 (41) 25 (38)  
   N2 1 (5) 0 (0) 1 (1)  
   N3 1 (5) 3 (7) 4 (6)  
Tumor grade    0.538 
 Grade 1 0 (0) 1 (2) 1 (1)  
 Grade 2 5 (24) 7 (15) 12 (18)  
 Grade 3 16 (76) 37 (81) 53 (80)  
 Unknown 0 (0) 1 (2) 1 (1)  
Pathologic response    0.705 
 pCR 9 (43) 22 (48) 31 (46)  
 RD 10 (48) 23 (50) 33 (49)  
 No surgery 2 (9) 1 (2) 3 (5)  
RCB score    0.746 
 RCB 0 9 (43) 22 (48) 31 (46)  
 RCB I 2 (9) 6 (13) 8 (12)  
 RCB II 8 (39) 11 (24) 19 (28)  
 RCB III 0 (0) 6 (13) 6 (9)  
 No surgery 2 (9) 1 (2) 3 (5)  
PD-L1 IHC status    0.786 
 Negative 6 (29) 11 (24) 17 (25)  
 Positive 12 (57) 26 (56) 36 (54)  
 Unknown 3 (24) 9 (20) 14 (21)  
Stromal TIL count    0.704 
 0%–10% 7 (33) 22 (48) 29 (43)  
 11%–29% 4 (19) 8 (17) 12 (18)  
 ≥30 6 (29) 13 (28) 19 (28)  
 Unknown 4 (19) 3 (7) 7 (11)  
AA (n = 21)non-AA (n = 46)All patients (n = 67)P
Age (median/range) 52 (33–79) 50.5 (27–69) 51 (27–79) 0.434 
BMI (median/range) 29.2 (24.8–49) 27.4 (19.6–42.1) 27.5 (19.6–49) 0.075 
Charlson Comorbidity Index (median/range) 3 (2–7) 3 (2–8) 3 (2–8) 0.317 
Tumor characteristics (pre-treatment)—n (%) 
cT stage    0.403 
   T1 6 (29) 18 (39) 24 (36)  
   T2 11 (52) 23 (50) 34 (51)  
   T3 3 (14) 5 (11) 8 (12)  
   T4 1 (5) 0 (0) 1 (1)  
cN stage    0.457 
   N0 13 (62) 24 (52) 37 (55)  
   N1 6 (28) 19 (41) 25 (38)  
   N2 1 (5) 0 (0) 1 (1)  
   N3 1 (5) 3 (7) 4 (6)  
Tumor grade    0.538 
 Grade 1 0 (0) 1 (2) 1 (1)  
 Grade 2 5 (24) 7 (15) 12 (18)  
 Grade 3 16 (76) 37 (81) 53 (80)  
 Unknown 0 (0) 1 (2) 1 (1)  
Pathologic response    0.705 
 pCR 9 (43) 22 (48) 31 (46)  
 RD 10 (48) 23 (50) 33 (49)  
 No surgery 2 (9) 1 (2) 3 (5)  
RCB score    0.746 
 RCB 0 9 (43) 22 (48) 31 (46)  
 RCB I 2 (9) 6 (13) 8 (12)  
 RCB II 8 (39) 11 (24) 19 (28)  
 RCB III 0 (0) 6 (13) 6 (9)  
 No surgery 2 (9) 1 (2) 3 (5)  
PD-L1 IHC status    0.786 
 Negative 6 (29) 11 (24) 17 (25)  
 Positive 12 (57) 26 (56) 36 (54)  
 Unknown 3 (24) 9 (20) 14 (21)  
Stromal TIL count    0.704 
 0%–10% 7 (33) 22 (48) 29 (43)  
 11%–29% 4 (19) 8 (17) 12 (18)  
 ≥30 6 (29) 13 (28) 19 (28)  
 Unknown 4 (19) 3 (7) 7 (11)  

Abbreviations: BMI, body mass index; RCB, residual cancer burden; PD-L1, programmed death ligand 1; IHC, immunohistochemistry; TIL, tumor-infiltrating lymphocyte.

All 67 patients in the trial were female. Twenty-one patients (31%) self-identified as AA. Of the remaining 46 patients, 40 (87%) self-identified as non-Hispanic/Latino White, 3 (6.5%) as Hispanic/Latino, and 3 (6.5%) as Asian. Baseline characteristics of median age, body mass index, and Charlson comorbidity index (25) were similar between AA and non-AA patients (Table 1). Clinical characteristics did not differ by race (Table 1).

Efficacy

Four out of 67 patients were treated in the Phase I portion of the trial and received 3 mg/kg dose of durvalumab. The remaining patients received the Phase II dose of 10 mg/kg. In the total intention-to-treat population of both the Phase I and II portions of the trial (N = 67), 31 patients (46%; 95% CI, 34%–58%) achieved pCR. There was no difference in pCR rates by race: 9 out 21 patients in the AA cohort (43%) and 22 out of 46 (48%) in the non-AA cohort (P = 0.71) had pCR (Table 1). Among the patients who underwent surgery (N = 64), RCB index distributions were similar between AA and non-AA patients (P = 0.746; Table 1; Supplementary Fig. S1). Breast cancer recurrences and deaths by race are listed in Supplementary Table S1. At a median follow-up of 35 months (25–70 months), 11 patients (16%) had distant metastatic recurrence: 3 among AA (14%), and 8 among non-AA patients (17%). Only one patient with pCR had a metastatic recurrence—she progressed during the taxane phase of the neoadjuvant therapy and treatment was changed to ddAC followed by carboplatin that resulted in pCR. There were three (4%) local breast cancer recurrences: One (5%) in the AA and 2 (4%) in the non-AA cohort. Overall, seven patients died from metastatic disease, including 2 AA (10%) and 5 non-AA patients (11%).

At a median follow up of 35 months, median EFS and OS have not been reached. Because the AA extension cohort continued to enroll after accrual to the non-AA cohort stopped, median follow-up for the AA patients is shorter (30 months) than for non-AA cohort (38 months). Estimated 3-year OS rate for the total population is 85% (95% CI, 74%–93%); it was 87% (95% CI, 74%–95%) for the non-AA, and 81% (95% CI, 58%–95%) for the AA groups (HR, 1.72; 95% CI, 0.481–6.136; P = 0.405; Fig. 1A). Three-year EFS rate for the total population was 76% (95% CI, 64%–86%); 78.3% (95% CI, 64%–89%) for non-AA, and 71.4% (95% CI, 48%–89%) for AA patients (HR for event or death 1.451; 95% CI, 0.524–4.017; P = 0.474; Fig. 1B). In both the AA and non-AA cohorts, patients who achieved pCR had longer EFS and OS compared with those with residual disease (HR for pCR vs. RD, adjusted for race: 0.234; 95% CI, 0.066–0.829; P = 0.024; Fig. 1C). Three-year OS was 96.8% in all patients with pCR versus 81.8% in those with residual disease (Adjusted HR, 0.151; 95% CI, 0.018–1.284; P = 0.083, Supplementary Fig. S2A). Three-year EFS was 90.3% for those who achieved pCR versus 66.7% for those who did not (Adjusted HR, 0.230; 95% CI, 0.063–0.836; P = 0.026, Supplementary Fig. S2B).

Figure 1.

Event-free survival (EFS) and overall survival (OS) by race and by pathologic response in African American (AA) and non-AA patients. Kaplan–Meier estimates of OS (A) and EFS (B–C) are shown. Tick marks represent data censored at the last time that the patient was known to be alive (OS) or alive and without an event (EFS). Events were defined as disease progression that precluded definitive surgery, local or distant recurrence, occurrence of a second primary cancer, or death from any cause. The hazard ratio and confidence interval were analyzed with the use of a Cox proportional-hazards model, with race (A and B) or pathologic response (pCR versus RD; C) within AA (blue lines) and non-AA patients (green lines) as covariates, adjusting for age and Charlson comorbidity index (A and B) or for race (C).

Figure 1.

Event-free survival (EFS) and overall survival (OS) by race and by pathologic response in African American (AA) and non-AA patients. Kaplan–Meier estimates of OS (A) and EFS (B–C) are shown. Tick marks represent data censored at the last time that the patient was known to be alive (OS) or alive and without an event (EFS). Events were defined as disease progression that precluded definitive surgery, local or distant recurrence, occurrence of a second primary cancer, or death from any cause. The hazard ratio and confidence interval were analyzed with the use of a Cox proportional-hazards model, with race (A and B) or pathologic response (pCR versus RD; C) within AA (blue lines) and non-AA patients (green lines) as covariates, adjusting for age and Charlson comorbidity index (A and B) or for race (C).

Close modal

Biomarker results

PD-L1 IHC results and sTIL counts were available on 59 and 60 patients, respectively. In the entire cohort, 36 (54%) patients had PD-L1-positive tumors; 12 (57%) AA and 26 (56%) non-AA. The sTIL count distributions in the entire cohort and in AA and non-AA patients are shown in Table 1. No statistically significant associations between race and PD-L1 status (P = 0.92) or sTIL count (P = 0.57) were observed (Fig. 2). Among the 64 patients who completed surgery, those who achieved pCR had higher PD-L1 positivity rate compared with those with residual disease (71% vs. 45%; P = 0.05, Supplementary Fig. S3A). Patients with pCR also had higher sTIL counts compared with those with residual disease (P = 0.0032, Supplementary Fig. S3B).

Figure 2.

PD-L1 positivity and stromal TILs based on race. A, PD-L1–positivity rate by SP263 antibody in the African American (AA; n = 21), and non-AA (n = 41) groups, respectively, P = 0.92 (χ2 test). B, The percentage of manual stromal tumor infiltrating lymphocytes (sTIL) in the AA (median, 20%) and non-AA groups (median, 10%), error bars represent 95% confidence intervals, P = 0.57 (Mann–Whitney U test).

Figure 2.

PD-L1 positivity and stromal TILs based on race. A, PD-L1–positivity rate by SP263 antibody in the African American (AA; n = 21), and non-AA (n = 41) groups, respectively, P = 0.92 (χ2 test). B, The percentage of manual stromal tumor infiltrating lymphocytes (sTIL) in the AA (median, 20%) and non-AA groups (median, 10%), error bars represent 95% confidence intervals, P = 0.57 (Mann–Whitney U test).

Close modal

Safety and toxicity

All patients who received at least one dose of study-assigned therapy were evaluable for safety and toxicity as described earlier (4). Table 2 summarizes the updated AEs after the inclusion of the additional AA cohort, with a breakdown of safety events by race for visual comparison. Immune related AEs are summarized in Table 3. The most common irAEs were endocrinopathies in 13 patients (28%; including hypothyroidism n = 10, hyperthyroidism n = 4, diabetes n = 2, adrenal insufficiency n = 1), and dermatitis in 13 patients. We also observed colitis in 5 patients (7%) and pneumonitis in 2 patients (3%). Grade 3 or higher irAEs affected 6 patients (14%), including 2 autoimmune diabetes, 3 colitis, and 1 patient, developed Miller–Fischer variant of GBS leading to death. The frequencies of irAEs, including grade ≥3 events were similar between AA and non-AA patients (Table 3). There were 7 (33%) AA and 18 (39%) non-AA patients who did not complete all planned doses of durvalumab (Supplementary Table S1). Six of these cases in the AA cohort and 13 in the non-AA cohort were due to AEs; most commonly immune-related AEs (n = 9 patients total), followed by peripheral sensory neuropathy during the nab-paclitaxel portion of treatment, which led to skipped doses of durvalumab. The remaining six patients—one in the AA and five in the non-AA cohort—experienced disease progression during neoadjuvant treatment and trial drugs were discontinued.

Table 2.

Treatment-related adverse events occurring in ≥20% of patients, or grade ≥3 occurring in ≥4% of patients.

AA Patients (n = 21)Non-AA Patients (n = 46)All Patients (n = 67)
Adverse EventAll Grades n (%)Grade ≥3 n (%)All Grades n (%)Grade ≥3 n (%)All Grades n (%)Grade ≥3 n (%)
Fatigue 18 (86) 1 (5) 36 (78) 1 (2) 54 (81) 2 (3) 
Peripheral sensory neuropathy 19 (90) 2 (9) 33 (72) 1 (2) 52 (78) 3 (4) 
Nausea 17 (81) 33 (72) 50 (75) 
Alopecia 16 (76) 29 (63) 45 (67) 
Rash 8 (38) 30 (65) 1 (2) 38 (57) 1 (1) 
Diarrhea 9 (43) 2 (9) 21 (46) 30 (45) 3 (4) 
Constipation 8 (38) 22 (48) 30 (45) 
Headaches 9 (43) 19 (41) 28 (42) 
Mucositis 11 (52) 16 (35) 1 (2) 27 (40) 1 (1) 
Anemia 8 (38) 3 (14) 18 (39) 1 (2) 26 (39) 4 (6) 
Dysgeusia 10 (48) 14 (30) 24 (36) 
Dizziness 8 (38) 15 (33) 1 (2) 23 (34) 1 (1) 
Dyspnea 10 (50) 1 (5) 11 (24) 1 (2) 21 (31) 
Cough 6 (29) 15 (33) 21 (31) 
Insomnia 7 (33) 14 (30) 21 (31) 
Anorexia 9 (43) 11 (24) 20 (30) 
Myalgias 3 (14) 14 (30) 17 (25) 
Vomiting 7 (33) 7 (15) 14 (21) 
Neutropenia 5 (24) 4 (19) 8 (17) 6 (13) 13 (19) 10 (15) 
Lymphocyte count decreased 3 (14) 1 (5) 10 (22) 2 (4) 13 (19) 3 (4) 
Hyperglycemia 4 (19) 3 (14) 5 (11) 1 (2) 9 (13) 4 (6) 
Nail pain 2 (9) 7 (15) 9 (13) 
WBC decreased 3 (14) 2 (9) 5 (11) 1 (2) 8 (12) 3 (4) 
Dehydration 2 (9) 3 (7) 3 (7) 5 (7) 3 (4) 
Colitis 1 (5) 1 (5) 4 (9) 2 (4) 5 (7) 3 (4) 
Febrile neutropenia 1 (5) 1 (5) 2 (4) 2 (4) 3 (4) 3 (4) 
Syncope 3 (7) 3 (7) 3 (4) 3 (4) 
Serious Adverse Events (SAEs) 7 (33) 11 (24) 18 (27) 
AA Patients (n = 21)Non-AA Patients (n = 46)All Patients (n = 67)
Adverse EventAll Grades n (%)Grade ≥3 n (%)All Grades n (%)Grade ≥3 n (%)All Grades n (%)Grade ≥3 n (%)
Fatigue 18 (86) 1 (5) 36 (78) 1 (2) 54 (81) 2 (3) 
Peripheral sensory neuropathy 19 (90) 2 (9) 33 (72) 1 (2) 52 (78) 3 (4) 
Nausea 17 (81) 33 (72) 50 (75) 
Alopecia 16 (76) 29 (63) 45 (67) 
Rash 8 (38) 30 (65) 1 (2) 38 (57) 1 (1) 
Diarrhea 9 (43) 2 (9) 21 (46) 30 (45) 3 (4) 
Constipation 8 (38) 22 (48) 30 (45) 
Headaches 9 (43) 19 (41) 28 (42) 
Mucositis 11 (52) 16 (35) 1 (2) 27 (40) 1 (1) 
Anemia 8 (38) 3 (14) 18 (39) 1 (2) 26 (39) 4 (6) 
Dysgeusia 10 (48) 14 (30) 24 (36) 
Dizziness 8 (38) 15 (33) 1 (2) 23 (34) 1 (1) 
Dyspnea 10 (50) 1 (5) 11 (24) 1 (2) 21 (31) 
Cough 6 (29) 15 (33) 21 (31) 
Insomnia 7 (33) 14 (30) 21 (31) 
Anorexia 9 (43) 11 (24) 20 (30) 
Myalgias 3 (14) 14 (30) 17 (25) 
Vomiting 7 (33) 7 (15) 14 (21) 
Neutropenia 5 (24) 4 (19) 8 (17) 6 (13) 13 (19) 10 (15) 
Lymphocyte count decreased 3 (14) 1 (5) 10 (22) 2 (4) 13 (19) 3 (4) 
Hyperglycemia 4 (19) 3 (14) 5 (11) 1 (2) 9 (13) 4 (6) 
Nail pain 2 (9) 7 (15) 9 (13) 
WBC decreased 3 (14) 2 (9) 5 (11) 1 (2) 8 (12) 3 (4) 
Dehydration 2 (9) 3 (7) 3 (7) 5 (7) 3 (4) 
Colitis 1 (5) 1 (5) 4 (9) 2 (4) 5 (7) 3 (4) 
Febrile neutropenia 1 (5) 1 (5) 2 (4) 2 (4) 3 (4) 3 (4) 
Syncope 3 (7) 3 (7) 3 (4) 3 (4) 
Serious Adverse Events (SAEs) 7 (33) 11 (24) 18 (27) 

Abbreviation: WBC, white blood cells.

Table 3.

Treatment-related immune adverse events (irAE).

AA (n = 21)non-AA (n = 46)All patients (n = 67)
Any irAE 
 # of patients, n (%) 8 (38) 18 (39) 26 (39) 
 # of episodesa 15 26 44 
Serious irAE (grade ≥3 or requiring treatment interruption and/or systemic corticosteroids) 5 (24) 9 (20) 14 (21) 
irAEs All grades grades ≥3 All grades grades ≥3 Total 
Dermatitis 11 13 
Hypothyroidism 10 
Hyperthyroidism 
Diabetes 
Adrenal insufficiency 
Colitis 
Pneumonitis 
Guillain-Barre Syndrome (Miller–Fischer) 
Inflammatory arthritis 
Optic neuritis 
Parotitisb 
AA (n = 21)non-AA (n = 46)All patients (n = 67)
Any irAE 
 # of patients, n (%) 8 (38) 18 (39) 26 (39) 
 # of episodesa 15 26 44 
Serious irAE (grade ≥3 or requiring treatment interruption and/or systemic corticosteroids) 5 (24) 9 (20) 14 (21) 
irAEs All grades grades ≥3 All grades grades ≥3 Total 
Dermatitis 11 13 
Hypothyroidism 10 
Hyperthyroidism 
Diabetes 
Adrenal insufficiency 
Colitis 
Pneumonitis 
Guillain-Barre Syndrome (Miller–Fischer) 
Inflammatory arthritis 
Optic neuritis 
Parotitisb 

a10 patients experienced more than one irAE.

birAE without a grade.

Multivariate analyses

We performed exploratory multivariate analyses to evaluate for predictors of pCR, as well as disease recurrence, and the development of irAEs (Table 4). Pathologic response (OR for pCR 0.17; 95% CI, 0.03–0.7; P = 0.02) as well as node positive status (OR, 4.13; 95% CI, 1.05–19.88; P = 0.05) were significantly associated with disease recurrence. We did not identify any independent predictors of pCR or irAEs (Table 4). However, the small sample size limits the power of this analysis.

Table 4.

Univariate and multivariate logistic regression analyses to identify predictors of recurrence, pathologic complete response (pCR), and immune-related adverse events (irAE), respectively.

RecurrenceUnivariate analysisMultivariate analysis
VariablesOR (95% CI)POR (95% CI)P
 pCR 0.21 (0.04–0.78) 0.03 0.17 (0.03–0.7) 0.02 
 Age ≥60 1.24 (0.25–5.04) 0.77 0.67 (0.09–4.19) 0.68 
 T stage (T2/3 vs. T1) 2.25 (0.61–10.87) 0.26 1.42 (0.3–8.03) 0.67 
 Node positive status 3.2 (0.96–11.82) 0.07 4.13 (1.05–18.99) 0.05 
 Charlson comorbidity index (2 vs. >2) 0.82 (0.25–2.82) 0.74 0.92 (0.19–4.5) 0.92 
 AA Race 0.93 (0.23–3.3) 0.92 1.08 (0.22–4.75) 0.92 
pCR   
Variables OR (95% CI) P OR (95% CI) P 
 sTIL ≥30% 2.86 (0.91–9.8) 0.079 1.42 (0.35–5.92) 0.623 
 PD-L1+ 3.18 (1.02–10.84) 0.053 2.74 (0.69–12.03) 0.160 
 Age ≥60 0.29 (0.06–1.08) 0.083 0.34 (0.06–1.73) 0.206 
 Charlson comorbidity index 1.03 (0.38–2.83) 0.955 1.44 (0.39–5.38) 0.582 
 AA Race 0.94 (0.32–2.77) 0.911 0.97 (0.25–3.69) 0.959 
irAEs   
Variables OR (95% CI) P OR (95% CI) P 
 pCR 1.44 (0.52–4.05) 0.478 1.78 (0.61–5.44) 0.297 
 Charlson comorbidity index 1.59 (0.57–4.66) 0.379 1.02 (0.3–3.48) 0.973 
 AA race 0.96 (0.32–2.74) 0.936 0.94 (0.29–2.92) 0.917 
 Age ≥60 2.59 (0.79–9) 0.120 2.16 (0.5–9.83) 0.303 
 BMI ≥30 1.25 (0.43–3.59) 0.684 1.36 (0.44–4.21) 0.588 
RecurrenceUnivariate analysisMultivariate analysis
VariablesOR (95% CI)POR (95% CI)P
 pCR 0.21 (0.04–0.78) 0.03 0.17 (0.03–0.7) 0.02 
 Age ≥60 1.24 (0.25–5.04) 0.77 0.67 (0.09–4.19) 0.68 
 T stage (T2/3 vs. T1) 2.25 (0.61–10.87) 0.26 1.42 (0.3–8.03) 0.67 
 Node positive status 3.2 (0.96–11.82) 0.07 4.13 (1.05–18.99) 0.05 
 Charlson comorbidity index (2 vs. >2) 0.82 (0.25–2.82) 0.74 0.92 (0.19–4.5) 0.92 
 AA Race 0.93 (0.23–3.3) 0.92 1.08 (0.22–4.75) 0.92 
pCR   
Variables OR (95% CI) P OR (95% CI) P 
 sTIL ≥30% 2.86 (0.91–9.8) 0.079 1.42 (0.35–5.92) 0.623 
 PD-L1+ 3.18 (1.02–10.84) 0.053 2.74 (0.69–12.03) 0.160 
 Age ≥60 0.29 (0.06–1.08) 0.083 0.34 (0.06–1.73) 0.206 
 Charlson comorbidity index 1.03 (0.38–2.83) 0.955 1.44 (0.39–5.38) 0.582 
 AA Race 0.94 (0.32–2.77) 0.911 0.97 (0.25–3.69) 0.959 
irAEs   
Variables OR (95% CI) P OR (95% CI) P 
 pCR 1.44 (0.52–4.05) 0.478 1.78 (0.61–5.44) 0.297 
 Charlson comorbidity index 1.59 (0.57–4.66) 0.379 1.02 (0.3–3.48) 0.973 
 AA race 0.96 (0.32–2.74) 0.936 0.94 (0.29–2.92) 0.917 
 Age ≥60 2.59 (0.79–9) 0.120 2.16 (0.5–9.83) 0.303 
 BMI ≥30 1.25 (0.43–3.59) 0.684 1.36 (0.44–4.21) 0.588 

Abbreviations: AA, African American; sTIL, stromal tumor–infiltrating lymphocytes; BMI, body mass index.

The addition of 10 cycles of durvalumab (10 mg/kg every two weeks) to weekly nab-paclitaxel (100 mg/m2) and ddAC resulted in a pCR rate of 46% in patients with early-stage TNBC, after extending accrual for AA patients so that they comprised 31% of the final study population. At a median follow-up of 35 months, three-year OS and EFS rates were 85% and 76%, respectively, for the total population. Having residual disease and being node-positive at diagnosis, but not race, were predictors of disease recurrence in multivariate analysis. The pCR rates were similar between AA and non-AA patients; stage at presentation, comorbidities, PD-L1 status, sTIL counts, 3-year OS and EFS, and the frequency of irAEs were also similar. The updated pCR rate of 46% is similar to our previously reported rate of 44% (4) in the original cohort of 59 patients. Although this pCR rate is lower than reported in the Geparnuevo trial (pCR 53%) that used a very similar treatment regimen (in Geparnuevo, patients received durvalumab 1.5 g every 4 weeks plus nab-paclitaxel 125 mg/m2 weekly x12, followed by durvalumab plus dose dense EC x 4), the 95% CIs of the pCR point estimates between the two trials broadly overlap and therefore the results are statistically similar. Three-year EFS rate of 76% is lower than that reported in the KEYNOTE-522 (ref. 2; 84.5% in the pembrolizumab group versus 76.8% in the placebo group) or the GeparNuevo trial (3) iDFS rate of 84.9% versus 76.9%, dDFS rate of 91.4% versus 79.5% and OS rate of 95.1% versus 83.1% with durvalumab versus placebo, respectively as well as the I-SPY2 trial (26), which reported an estimated 3-year EFS rate of 86% in the pembrolizumab arm among patients with TNB. The 95% CI of the EFS point estimate in our trial (64%–86%) includes the EFS/dDFS rates reported in these trials and therefore our results are statistically compatible with the larger randomized trials.

The toxicity profile of our study was also similar to the GeparNuevo, KEYNOTE-522 and I-SPY2 trials. The most common AEs seen in each of these trials were anemia, neutropenia, thrombocytopenia, increased transaminases, fatigue, alopecia, nausea, diarrhea, mucositis, peripheral neuropathy, arthralgias and myalgias, which were overall similar to the placebo plus chemotherapy arms, and numerically similar to our study findings, suggesting that these AEs are related to the chemotherapy administration. Regarding the irAEs, all aforementioned studies and our study showed increased incidence of endocrinopathies in patients receiving ICIs. Thyroid dysfunction is the most commonly seen irAE. Durvalumab use was associated with additional cases of hypophysitis in the GeparNuevo trial and dermatitis, diabetes, adrenal insufficiency, colitis, pneumonitis, GBS, inflammatory arthritis, optic neuritis, and parotitis in our trial whereas pembrolizumab was associated with adrenal insufficiency, hepatitis, pneumonitis, colitis and pruritus in I-SPY2 and adrenal insufficiency, infusion reactions and skin reactions in KEYNOTE-522. Although all three of these randomized trials targeted almost identical patient populations, they enrolled very few, if any, Black or AA patients. The pembrolizumab arm of I-SPY2 included 6 (9%) Black patients, whereas KEYNOTE-522 and GeparNuevo did not report on race. Our study, although smaller and non-randomized, adds to the literature by, including 30% AA women with TNBC, higher rate than any other studies.

Our results indicate no statistically significant differences in pCR, EFS or toxicity by race. This is similar to several, but not all, previous reports. Dean-Colomb and colleagues (9) compared pCR rates and transcriptional profiles of tumors from 98 women who received neoadjuvant chemotherapy for early-stage TNBC and found no differences in pCR rates or in gene expression by ethnicity or race. Dawood and colleagues (10) also showed similar pCR rates in 471 women with TNBC receiving anthracycline-based neoadjuvant chemotherapy regimens. Most recently, clinical outcomes by race for the randomized neoadjuvant trial, I-SPY2, were reported at the 2021 San Antonio Breast Cancer Symposium (11) and showed similar pCR rates by race among 907 women, including 120 (12%) AA patients; patients with TNBC with residual disease after neoadjuvant chemotherapy also had similar EFS at a median follow-up of 4.4 years regardless of race. These findings are in contrast with several retrospective studies. Woodward and colleagues (15) evaluated the effect of race on survival in patients treated with adjuvant or neoadjuvant chemotherapy at the University of Texas MD Anderson Cancer Center and found that despite similar distribution of treatment, AA race was independently associated with poorer long-term survival, although they also found similar pCR rates between AA and non-AA patients. Bauer and colleagues (13) and Lund and colleagues (12) focused on AA women with TNBC and found worse survival compared with European American women after controlling for socioeconomic factors and treatment delays. The contradictory results illustrate the difficulty in comparing outcome results across trials. The composition of patient populations vary and many important socioeconomic variables and the number and severity of comorbidities that collectively influence clinical outcomes are not optimally captured in study documentations. It is increasingly recognized that inherited biological contributions to health outcome disparities are small, if any (6, 27–32). Disparities traditionally attributed to race are likely a result of a complex interplay between social and economic factors and biological consequences of experiencing discrimination or poverty (33). It is reassuring that multiple studies have been able to demonstrate that under highly controlled and structured care delivery, as in our clinical trial, race-associated outcome disparities diminish or disappear.

An increasingly recognized shortcoming of oncology trials is the exceedingly low rate of clinical trial participation by minority patients (34). A recent review of the racial composition of major clinical trials investigating immune checkpoint inhibitors for the treatment of patients with advanced solid tumors showed that the percentage of AA patients included ranged from 0% to 4% across nine clinical trials in six different cancer types, which did not include breast cancer (20). In breast immunotherapy trials in which race/ethnicity are reported, the percentage of AA population ranges from 4% to 12% (11, 35–37). Underrepresentation of minority participants in immunotherapy clinical trials prevents evaluation of the impact of racial/ethnic- or ancestry-based differences in efficacy and toxicity. By adding an accrual extension phase to our study, open for AA patients only, we increased the percentage of AA patients to 31%, which is proportional to the population size in New Haven (representativeness of our study population is summarized in Supplementary Table S2). The city of New Haven has 33% AA population according to the July 2020 US census (https://www.census.gov/quickfacts/newhavencityconnecticut), which is significantly higher than the 13.4% US population average (38). A deliberate effort to increase participation by underrepresented minority patients in clinical trials will not only improve our understanding of the efficacy and safety of these drugs in a diverse patient populations but also ensures more equal access to novel therapies. This could be accomplished by adding minority population specific accrual extension cohorts to a trial, or by a priory specifying a targeted accrual proportion for minority patients.

Limitations of this study include lack of randomization that prevents definitive conclusion about the added benefit from durvalumab over chemotherapy alone. The small sample size also limits the power of statistical comparisons between subgroups. Some of the statistically not significant numerical differences in outcome by racial groups in our study could have become statistically significant if we had a larger sample size. However, the lack of statistical significance indicates that these differences are modest and could be due to chance. Much larger studies will be needed to more precisely assess if true differences exist. A retrospective study examining treatment quality and outcomes of AA versus White patients with breast cancer using the Southwest Oncology studies S8814 and S8897, which included patients receiving chemotherapy and endocrine therapy, showed that AA women were more likely to experience early discontinuation or treatment delays, and worse survival (39). Retrospective data analyses are prone to intrinsic biases and therefore often lead to conflicting results. More recently, in the ASCENT trial, which compared the novel antibody–drug conjugate sacituzumab–govitecan with treatment of physician's choice in patients with heavily pretreated metastatic TNBC, the 12% of patients who self-identified as AA derived a similar clinical benefit as the full trial population and experienced similar toxicities (40). Prospective inclusion of AA patients and other minorities, proportional to their national and neighborhood representation, in clinical trials should be our goal, particularly in disease setting where AA patients experience poorer outcomes compared with non-AA patients. Strategies that can facilitate minority accrual include assistance with out of pocket costs of trial participants (41); increasing workforce diversity that can foster trust (42); more patient-centered inclusion and exclusion criteria (43); and partnerships between academic and community centers (44). Furthermore, it is important to start obtaining more granular and higher-quality information on social determinants of health, in addition to race, in clinical trials, to learn how these interact with outcomes disparities despite seemingly equal treatment.

In summary, durvalumab added to neoadjuvant chemotherapy in early-stage TNBC showed a pCR rate of 46%, which was similar between AA and non-AA patients. There was no association between race and stage at diagnosis, comorbidities, sTIL counts, PD-L1 status, 3-year EFS and OS, and there were no marked differences in AEs. These results suggest that when patients receive identical treatment and are monitored closely, disparities in outcomes can be mitigated or even abolished.

A. Silber reports personal fees from AstraZeneca outside the submitted work. K. Adelson reports other support from Genentech and Carrum Health, and personal fees from AbbVie outside the submitted work. A. Chagpar reports personal fees from Protean Biodiagnostics, Sanofi Aventis, Guardant Health, Puma Diagnostics, Athenex, Novartis, and Merck outside the submitted work. K. Blenman reports other support from CDI Laboratories outside the submitted work. D.L. Rimm reports grants and personal fees from Amgen, Cepheid, Konica Minolta, and NextCure, grants from Lilly, and personal fees from Astra Zeneca, Cell Signaling Technology, Danaher, Fluidigm, Immunogen, and Merck, Monopteros, NanoString, Odonate, Paige.ai, Regeneron, Roche, Sanofi, Ventana, and Verily outside the submitted work. L. Pusztai reports personal fees from Novartis, personal fees and other support from Pfizer, and grants and personal fees from Bristol Myers Squibb, AstraZeneca, Seagen, and Merck outside the submitted work. No disclosures were reported by the other authors.

J. Foldi: Conceptualization, formal analysis, methodology, writing–original draft, writing–review and editing. A. Kahn: Formal analysis, writing–original draft, writing–review and editing. A. Silber: Conceptualization, resources, investigation, writing–review and editing. T. Qing: Software, formal analysis, writing–review and editing. E. Reisenbichler: Software, formal analysis, visualization, writing–review and editing. N. Fischbach: Investigation, writing–review and editing. J. Persico: Investigation, writing–review and editing. K. Adelson: Investigation, writing–review and editing. A. Katoch: Investigation, writing–review and editing. A. Chagpar: Investigation, writing–review and editing. T. Park: Investigation, writing–review and editing. A. Blanchard: Data curation, project administration, writing–review and editing. K. Blenman: Conceptualization, formal analysis, investigation, methodology, writing–review and editing. D.L. Rimm: Formal analysis, visualization, writing–review and editing. L. Pusztai: Conceptualization, resources, supervision, funding acquisition, investigation, methodology, writing–original draft, project administration, writing–review and editing.

The conduct of the clinical trial was supported by AstraZeneca, but it played no role in the design of the trial or in the interpretation of the results. The biomarker studies were supported by a Komen Leadership Grant (SAC160076) and an NCI R01 grant (5R01CA219647–03; to L. Pusztai).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/).

1.
Schmid
P
,
Cortes
J
,
Pusztai
L
,
Mcarthur
H
,
Kümmel
S
,
Bergh
J
, et al
.
Pembrolizumab for early triple-negative breast cancer
.
N Engl J Med
2020
;
382
:
810
21
.
2.
Schmid
P
,
Cortes
J
,
Dent
R
,
Pusztai
L
,
Mcarthur
H
,
Kümmel
S
, et al
.
Event-free survival with pembrolizumab in early triple-negative breast cancer
.
N Engl J Med
2022
;
386
:
556
67
.
3.
Loibl
S
,
Untch
M
,
Burchardi
N
,
Huober
J
,
Sinn
BV
,
Blohmer
J-U
, et al
.
A randomised phase II study investigating durvalumab in addition to an anthracycline taxane-based neoadjuvant therapy in early triple-negative breast cancer: clinical results and biomarker analysis of GeparNuevo study
.
Ann Oncol Off J Eur Soc Med Oncol
2019
;
30
:
1279
88
.
4.
Foldi
J
,
Silber
A
,
Reisenbichler
E
,
Singh
K
,
Fischbach
N
,
Persico
J
, et al
.
Neoadjuvant durvalumab plus weekly nab-paclitaxel and dose-dense doxorubicin/cyclophosphamide in triple-negative breast cancer
.
NPJ Breast Cancer
2021
;
7
:
9
.
5.
Dietze
EC
,
Sistrunk
C
,
Miranda-Carboni
G
,
O'regan
R
,
Seewaldt
VL
.
Triple-negative breast cancer in African-American women: disparities versus biology
.
Nat Rev Cancer
2015
;
15
:
248
54
.
6.
Carey
LA
,
Perou
CM
,
Livasy
CA
,
Dressler
LG
,
Cowan
D
,
Conway
K
, et al
.
Race, breast cancer subtypes, and survival in the Carolina Breast Cancer Study
.
JAMA
2006
;
295
:
2492
502
.
7.
Shen
Y
,
Dong
W
,
Esteva
FJ
,
Kau
S-W
,
Theriault
RL
,
Bevers
TB
.
Are there racial differences in breast cancer treatments and clinical outcomes for women treated at M.D. Anderson Cancer Center?
Breast Cancer Res Treat
2007
;
102
:
347
56
.
8.
Chlebowski
RT
,
Chen
Z
,
Anderson
GL
,
Rohan
T
,
Aragaki
A
,
Lane
D
, et al
.
Ethnicity and breast cancer: factors influencing differences in incidence and outcome
.
J Natl Cancer Inst
2005
;
97
:
439
48
.
9.
Dean-Colomb
W
,
Yan
K
,
Liedtke
C
,
Symmans
WF
,
Holmes
FA
,
O'shaughnessy
J
, et al
.
Transcriptional profiles of triple receptor-negative breast cancer: are Caucasian, Hispanic, and African-American women different?
J Clin Oncol
2008
;
26
:
22014
.
10.
Dawood
S
,
Broglio
K
,
Kau
S-W
,
Green
MC
,
Giordano
SH
,
Meric-Bernstam
F
, 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
:
220
6
.
11.
Kyalwazi
B
,
Yau
C
,
Olopade
O
,
Chien
AJ
,
Wallace
A
,
Forero-Torres
A
, et al
.
Abstract GS4–02: analysis of clinical outcomes and expression-based immune signatures by race in the I-SPY 2 trial
.
Cancer Res
2022
;
82
:
GS4
02
.
12.
Lund
MJ
,
Trivers
KF
,
Porter
PL
,
Coates
RJ
,
Leyland-Jones
B
,
Brawley
OW
, et al
.
Race and triple negative threats to breast cancer survival: a population-based study in Atlanta, GA
.
Breast Cancer Res Treat
2009
;
113
:
357
70
.
13.
Bauer
KR
,
Brown
M
,
Cress
RD
,
Parise
CA
,
Caggiano
V
.
Descriptive analysis of estrogen receptor (ER)-negative, progesterone receptor (PR)-negative, and HER2-negative invasive breast cancer, the so-called triple-negative phenotype: a population-based study from the California cancer Registry
.
Cancer
2007
;
109
:
1721
8
.
14.
Albain
KS
,
Unger
JM
,
Crowley
JJ
,
Coltman
CA
,
Hershman
DL
.
Racial disparities in cancer survival among randomized clinical trials patients of the Southwest Oncology Group
.
J Natl Cancer Inst
2009
;
101
:
984
92
.
15.
Woodward
WA
,
Huang
EH
,
Mcneese
MD
,
Perkins
GH
,
Tucker
SL
,
Strom
EA
, et al
.
African-American race is associated with a poorer overall survival rate for breast cancer patients treated with mastectomy and doxorubicin-based chemotherapy
.
Cancer
2006
;
107
:
2662
8
.
16.
Yedjou
CG
,
Sims
JN
,
Miele
L
,
Noubissi
F
,
Lowe
L
,
Fonseca
DD
, et al
.
Health and racial disparity in breast cancer
.
Adv Exp Med Biol
2019
;
1152
:
31
49
.
17.
Yedjou
C
,
Tchounwou
P
,
Payton
M
,
Miele
L
,
Fonseca
D
,
Lowe
L
, et al
.
Assessing the racial and ethnic disparities in breast cancer mortality in the United States
.
Int. J. Environ. Res. Public. Health
2017
;
14
:
E486
.
18.
Babatunde
OA
,
Eberth
JM
,
Felder
T
,
Moran
R
,
Truman
S
,
Hebert
JR
, et al
.
Social determinants of racial disparities in breast cancer mortality among black and white women
.
J. Racial Ethn. Health Disparities
2021
;
8
:
147
56
.
19.
Newman
LA
,
Kaljee
LM
.
Health disparities and triple-negative breast cancer in African American women: a review
.
JAMA Surg
2017
;
152
:
485
93
.
20.
Nazha
B
,
Mishra
M
,
Pentz
R
,
Owonikoko
TK
.
Enrollment of racial minorities in clinical trials: old problem assumes new urgency in the age of immunotherapy
.
Am Soc Clin Oncol Educ Book Am Soc Clin Oncol Annu Meet
2019
;
39
:
3
10
.
21.
Telli
ML
,
Gradishar
WJ
,
Ward
JH
.
NCCN guidelines updates: breast cancer
.
J Natl Compr Cancer Netw JNCCN
2019
;
17
:
552
5
.
22.
Ahn
S
,
Woo
JW
,
Kim
H
,
Cho
EY
,
Kim
A
,
Kim
JY
, et al
.
Programmed death ligand 1 immunohistochemistry in triple-negative breast cancer: evaluation of inter-pathologist concordance and inter-assay variability
.
J Breast Cancer
2021
;
24
:
266
79
.
23.
Smith
J
,
Robida
MD
,
Acosta
K
,
Vennapusa
B
,
Mistry
A
,
Martin
G
, et al
.
Quantitative and qualitative characterization of Two PD-L1 clones: SP263 and E1L3N
.
Diagn Pathol
2016
;
11
:
44
.
24.
Martinez-Morilla
S
,
Mcguire
J
,
Gaule
P
,
Moore
L
,
Acs
B
,
Cougot
D
, et al
.
Quantitative assessment of PD-L1 as an analyte in immunohistochemistry diagnostic assays using a standardized cell line tissue microarray
.
Lab Investig J Tech Methods Pathol
2020
;
100
:
4
15
.
25.
Charlson
ME
,
Pompei
P
,
Ales
KL
,
Mackenzie
CR
.
A new method of classifying prognostic comorbidity in longitudinal studies: development and validation
.
J Chronic Dis
1987
;
40
:
373
83
.
26.
Nanda
R
,
Liu
MC
,
Yau
C
,
Shatsky
R
,
Pusztai
L
,
Wallace
A
, et al
.
Effect of pembrolizumab plus neoadjuvant chemotherapy on pathologic complete response in women with early-stage breast cancer: an analysis of the ongoing phase 2 adaptively randomized I-SPY2 trial
.
JAMA Oncol
2020
;
6
:
676
84
.
27.
Hayanga
AJ
,
Newman
LA
.
Investigating the phenotypes and genotypes of breast cancer in women with African ancestry: the need for more genetic epidemiology
.
Surg Clin North Am
2007
;
87
:
551
68
.
28.
Martin
DN
,
Boersma
BJ
,
Yi
M
,
Reimers
M
,
Howe
TM
,
Yfantis
HG
, et al
.
Differences in the tumor microenvironment between African-American and European-American breast cancer patients
.
PLoS ONE
2009
;
4
:
e4531
.
29.
Porter
PL
,
Lund
MJ
,
Lin
MG
,
Yuan
X
,
Liff
JM
,
Flagg
EW
, et al
.
Racial differences in the expression of cell-cycle–regulatory proteins in breast carcinoma
.
Cancer
2004
;
100
:
2533
42
.
30.
O'Meara
T
,
Safonov
A
,
Casadevall
D
,
Qing
T
,
Silber
A
,
Killelea
B
, et al
.
Immune microenvironment of triple-negative breast cancer in African-American and Caucasian women
.
Breast Cancer Res Treat
2019
;
175
:
247
59
.
31.
Cole
SW
.
Social regulation of human gene expression: mechanisms and implications for public health
.
Am J Public Health
2013
;
103
:
S84
92
.
32.
Cole
SW
.
The conserved transcriptional response to adversity
.
Curr Opin Behav Sci
2019
;
28
:
31
37
.
33.
Carlos
RC
,
Obeng-Gyasi
S
,
Cole
SW
,
Zebrack
BJ
,
Pisano
ED
,
Troester
MA
, et al
.
Linking structural racism and discrimination and breast cancer outcomes: a social genomics approach
.
J Clin Oncol
2022
;
40
:
1407
13
.
34.
Oyer
RA
,
Hurley
P
,
Boehmer
L
,
Bruinooge
SS
,
Levit
K
,
Barrett
N
, et al
.
Increasing racial and ethnic diversity in cancer clinical trials: an American Society of Clinical Oncology and Association of Community Cancer Centers Joint Research Statement
.
J Clin Oncol
2022
;
40
:
2163
71
.
35.
Cortes
J
,
Cescon
DW
,
Rugo
HS
,
Nowecki
Z
,
Im
S-A
,
Yusof
MM
, et al
.
Pembrolizumab plus chemotherapy versus placebo plus chemotherapy for previously untreated locally recurrent inoperable or metastatic triple-negative breast cancer (KEYNOTE-355): a randomised, placebo-controlled, double-blind, phase 3 clinical trial
.
Lancet North Am Ed
2020
;
396
:
1817
28
.
36.
Schmid
P
,
Adams
S
,
Rugo
HS
,
Schneeweiss
A
,
Barrios
CH
,
Iwata
H
, et al
.
Atezolizumab and nab-paclitaxel in advanced triple-negative breast cancer
.
N Engl J Med
2018
;
379
:
2108
21
.
37.
Mittendorf
EA
,
Zhang
H
,
Barrios
CH
,
Saji
S
,
Jung
KH
,
Hegg
R
, et al
.
Neoadjuvant atezolizumab in combination with sequential nab-paclitaxel and anthracycline-based chemotherapy versus placebo and chemotherapy in patients with early-stage triple-negative breast cancer (IMpassion031): a randomised, double-blind, phase 3 trial
.
Lancet North Am Ed
2020
;
396
:
1090
100
.
38.
U.S. Census Bureau. (August 12, 2021)
.
Supplementary tables on Race and Hispanic origin: 2020 Census Redistricting Data (P.L. 94–171)
.
Available from
: https://www.census.gov/data/tables/2020/dec/2020-redistricting-supplementary-tables.html.
Retrieved on: June 27, 2022
39.
Hershman
DL
,
Unger
JM
,
Barlow
WE
,
Hutchins
LF
,
Martino
S
,
Osborne
CK
, et al
.
Treatment quality and outcomes of African American versus white breast cancer patients: retrospective analysis of Southwest Oncology studies S8814/S8897
.
J Clin Oncol
2009
;
27
:
2157
62
.
40.
Carey
LA
, et al
.
Assessment of sacituzumab govitecan (SG) in Black patients (pts) from the phase 3 ASCENT study in metastatic triple-negative breast cancer (mTNBC), in San Antonio Breast Cancer Symposium
.
2021
(
2021
).
41.
Winkfield
KM
,
Phillips
JK
,
Joffe
S
,
Halpern
MT
,
Wollins
DS
,
Moy
B
.
Addressing financial barriers to patient participation in clinical trials: ASCO policy statement
.
J Clin Oncol
2018
Sep 13. [Epub ahead of print].
42.
Wallington
SF
,
Dash
C
,
Sheppard
VB
,
Goode
TD
,
Oppong
BA
,
Dodson
EE
, et al
.
Enrolling minority and underserved populations in cancer clinical research
.
Am J Prev Med
2016
;
50
:
111
7
.
43.
Ibraheem
A
,
Polite
B
.
Improving the accrual of racial and ethnic minority patients in clinical trials: time to raise the stakes
.
Cancer
2017
;
123
:
4752
6
.
44.
Vickers
SM
,
Fouad
MN
.
An overview of EMPaCT and fundamental issues affecting minority participation in cancer clinical trials
.
Cancer
2014
;
120
:
1087
90
.

Supplementary data