Clinical characteristics, racial inequities, and outcomes in patients with breast cancer and COVID-19: A COVID-19 and cancer consortium (CCC19) cohort study

Background: Limited information is available for patients with breast cancer (BC) and coronavirus disease 2019 (COVID-19), especially among underrepresented racial/ethnic populations. Methods: This is a COVID-19 and Cancer Consortium (CCC19) registry-based retrospective cohort study of females with active or history of BC and laboratory-confirmed severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2) infection diagnosed between March 2020 and June 2021 in the US. Primary outcome was COVID-19 severity measured on a five-level ordinal scale, including none of the following complications, hospitalization, intensive care unit admission, mechanical ventilation, and all-cause mortality. Multivariable ordinal logistic regression model identified characteristics associated with COVID-19 severity. Results: 1383 female patient records with BC and COVID-19 were included in the analysis, the median age was 61 years, and median follow-up was 90 days. Multivariable analysis revealed higher odds of COVID-19 severity for older age (aOR per decade, 1.48 [95% CI, 1.32–1.67]); Black patients (aOR 1.74; 95 CI 1.24–2.45), Asian Americans and Pacific Islander patients (aOR 3.40; 95 CI 1.70–6.79) and Other (aOR 2.97; 95 CI 1.71–5.17) racial/ethnic groups; worse ECOG performance status (ECOG PS ≥2: aOR, 7.78 [95% CI, 4.83–12.5]); pre-existing cardiovascular (aOR, 2.26 [95% CI, 1.63–3.15])/pulmonary comorbidities (aOR, 1.65 [95% CI, 1.20–2.29]); diabetes mellitus (aOR, 2.25 [95% CI, 1.66–3.04]); and active and progressing cancer (aOR, 12.5 [95% CI, 6.89–22.6]). Hispanic ethnicity, timing, and type of anti-cancer therapy modalities were not significantly associated with worse COVID-19 outcomes. The total all-cause mortality and hospitalization rate for the entire cohort was 9% and 37%, respectively however, it varied according to the BC disease status. Conclusions: Using one of the largest registries on cancer and COVID-19, we identified patient and BC-related factors associated with worse COVID-19 outcomes. After adjusting for baseline characteristics, underrepresented racial/ethnic patients experienced worse outcomes compared to non-Hispanic White patients. Funding: This study was partly supported by National Cancer Institute grant number P30 CA068485 to Tianyi Sun, Sanjay Mishra, Benjamin French, Jeremy L Warner; P30-CA046592 to Christopher R Friese; P30 CA023100 for Rana R McKay; P30-CA054174 for Pankil K Shah and Dimpy P Shah; KL2 TR002646 for Pankil Shah and the American Cancer Society and Hope Foundation for Cancer Research (MRSG-16-152-01-CCE) and P30-CA054174 for Dimpy P Shah. REDCap is developed and supported by Vanderbilt Institute for Clinical and Translational Research grant support (UL1 TR000445 from NCATS/NIH). The funding sources had no role in the writing of the manuscript or the decision to submit it for publication. Clinical trial number: CCC19 registry is registered on ClinicalTrials.gov, NCT04354701.


Introduction
The COVID-19 pandemic has had a devastating impact worldwide and within the United States (US) (World Health Organization, 2021;CDC, 2020a).Previous studies have reported that patients with cancer are at an increased risk for SARS-CoV-2 infection and have higher rates of adverse outcomes with mortality rates ranging from 14% to 33% (Grivas et al., 2021;Garassino et al., 2020;Lee et al., 2020;Wang et al., 2021;de Azambuja et al., 2020;Albiges et al., 2020;Sharafeldin et al., 2021;Lièvre et al., 2020).COVID-19 has also highlighted the long-standing health inequities in the US, as underrepresented racial and ethnic populations have disproportionately been affected.Some studies have reported non-White race/ethnicity to be an independent risk factor for worse COVID-19 outcomes such as hospitalization and death (Grivas et al., 2021;Wang et al., 2021;CDC, 2020b;Millett et al., 2020;Muñoz-Price et al., 2020;Gross et al., 2020;Price-Haywood et al., 2020;Azar et al., 2020;Mackey et al., 2021;Garg et al., 2020;Mahajan and Larkins-Pettigrew, 2020;Kim and Bostwick, 2020).Recently published data from CCC19 also showed that Black patients with cancer experienced worse COVID-19 outcomes compared to White patients after adjusting for key risk factors including cancer status and comorbidities (Fu et al., 2022).
Breast cancer (BC) is the most common cancer diagnosed in females and affects all major racial/ ethnic groups (Siegel et al., 2021;Sung et al., 2021;SEER, 2021).There are well-described racial/ ethnic differences in BC incidence and outcomes in females in the US attributable to multiple social and biological factors (Chlebowski et al., 2005;Bigby and Holmes, 2005;Yedjou et al., 2019).Few studies have specifically evaluated the impact of COVID-19 in patients with BC; interpretation from prior studies has been limited by small sample sizes (Vuagnat et al., 2020;Kalinsky et al., 2020).Data specifically on the impact of COVID-19 among underrepresented racial/ethnic groups with BC are also lacking.Understanding the sociodemographic and clinical factors associated with higher risk for adverse COVID-19 outcomes will help guide patient care.Hence, we aimed to evaluate the prognostic factors, racial disparities, interventions, complications, and outcomes among patients with active or previous history of BC diagnosed with COVID-19.

Study population
The COVID-19 and Cancer Consortium (CCC19) consists of 129 member institutions capturing granular, detailed, and uniform data on demographic and clinical characteristics, treatment information, and outcomes of COVID-19.Details of CCC19 protocol, data collection, and quality assurance have been previously described (Kuderer et al., 2020;COVID-19 and Cancer Consortium. Electronic address: jeremy. warner@ vumc. org and COVID-19 and Cancer Consortium, 2020).This registrybased retrospective cohort study included all female adults (age ≥18 years) with an active or previous history of invasive BC and laboratory-confirmed diagnosis of SARS-CoV-2 by polymerase chain reaction (PCR) and/or serology from March 17, 2020, to June 16, 2021, in the US.Patient records with multiple invasive malignancies including history of multiple invasive BC were excluded; patients with unknown or missing race and ethnicity, inadequate data quality (quality score >4), and those not evaluable for the primary ordinal outcome were also excluded (supplementary appendix 1) (COVID-19 and Cancer Consortium. Electronic address: jeremy. warner@ vumc. org and COVID-19 and Cancer Consortium, 2020).This study was exempt from institutional review board (IRB) review (VUMC IRB#200467) and was approved by IRBs at participating sites per institutional policy.CCC19 registry is registered on ClinicalTrials.gov,NCT04354701.

Outcome definitions
The primary outcome was a five-level ordinal scale of COVID-19 severity based on each individual patient's most severe reported disease status: none of the following complications; admitted to the hospital; admitted to an intensive care unit (ICU); mechanically ventilated at any time after COVID-19 diagnosis; or death from any cause.Other COVID-19-related complications (cardiovascular; gastrointestinal; and pulmonary complications, acute kidney injury, multisystem organ failure, superimposed infection, sepsis, any bleeding); 30-day mortality; and anti-COVID-19 directed interventions (supplemental oxygen, remdesivir, systemic corticosteroids, hydroxychloroquine, and other treatments) are also reported.

Covariates
Covariates were selected a priori and included: age; sex; race/ethnicity (non-Hispanic White [NHW], Black, Hispanic, Asian Americans and Pacific Islanders [AAPI], and Other) as recorded in the EHR, based on the Center for Disease Control and Prevention Race and Ethnicity codes (CDC, 2021); US census region of reporting institution (Northeast [NE], Midwest [MW], South and West); month/ year of COVID-19 diagnosis (classified into 4-month intervals); smoking status; obesity; comorbidities (cardiovascular, pulmonary, renal, or diabetes mellitus); Eastern Cooperative Oncology Group (ECOG) performance status (PS); BC subtypes based on hormone receptor (HR) and human epidermal growth factor receptor 2 (HER2) expression (HR+/HER2-, HR+/HER2+, HR-/HER2+, HR-/HER2-[triple negative], missing/unknown); cancer status at time of COVID-19 diagnosis; timing of most recent anti-cancer therapy relative to COVID-19 diagnosis (never or after COVID-19 diagnosis, 0-4 weeks, 1-3 months, >3 months); and modality of anti-cancer therapy received within 3 months of COVID-19 diagnosis.Cancer status was defined as remission or no evidence of disease (NED) for >5 years, remission or NED for ≤5 years, and active disease, with active disease further classified as responding to therapy, stable, or progressing.Anti-cancer modalities were categorized as chemotherapy; cyclindependent kinase (CDK) 4/6 inhibitor; anti-HER2 therapy; other targeted therapy (non-CDK 4/6 inhibitor, non-anti-HER2 therapy); endocrine therapy; immunotherapy; and locoregional therapy (surgery and/or radiation).In the survey, drug classes (modalities) along with a few specific drugs (through checkboxes) were captured.Survey respondents were also encouraged to provide additional details in the free text boxes which were reviewed extensively by the Informatics Core at VUMC, and queries were sent to participating sites to clarify ambiguous reports.CDK 4/6 inhibitor, anti-HER2 therapy, and other targeted therapy information were extracted from free text in the registry survey while the others were checkboxes.In addition, baseline severity of COVID-19 at presentation, classified as mild (no hospitalization indicated), moderate (hospitalization indicated), and severe (ICU admission indicated), was collected.Other variables included location of patient residence (urban, suburban, rural) and treatment center characteristics (academic medical center, community practice, tertiary care center).The CCC19 data dictionary is available at https://github.com/covidncancer/CCC19_dictionary(Mishra and Warner, 2023).The project approved variables used for the analysis are provided in supplementary appendix 3.

Statistical methods
Covariates, outcome definitions, and statistical analysis plan were prespecified by the authors and the CCC19 Research Coordinating Center prior to analysis (supplementary appendix 2).Standard descriptive statistics were used to summarize prognostic factors, rates of clinical complications, interventions during hospitalization, and rates of outcomes such as 30-day mortality, hospitalization, oxygen requirement, ICU admission, mechanical ventilation, and overall mortality among racial and ethnic groups.The primary analysis was restricted to females with BC.
Multivariable ordinal logistic regression models for the COVID-19 severity outcome among females with BC included age, race/ethnicity, obesity, ECOG PS, comorbidities, cancer status, anti-cancer therapy and timing, month/year of COVID-19 diagnosis (classified into 4-month intervals), and US census region of reporting institution.These covariates were identified a priori as the most clinically relevant for COVID-19 severity and were included in a single model, given a sufficient number of events and corresponding degrees of freedom.Because the ordinal outcome was assessed over a given patient's total follow-up period, the model included an offset for (log) follow-up time.The results are presented as adjusted odds ratio (ORs) with 95% CIs.Model stability was assessed by comparing unadjusted and adjusted models and variance inflation factors.Graphical methods were used to verify the proportional odds assumption (Appendix 4-figure 1).We used the e value to quantify sensitivity to unmeasured confounding for the observed OR for race/ethnicity (VanderWeele and Ding, 2017;Haneuse et al., 2019).Multiple imputation (20 imputed datasets) was used to impute missing and unknown data for all variables included in the analysis, with some exceptions: unknown ECOG performance score and unknown cancer status were not imputed and treated as a separate category in analyses.Imputation was performed on the largest dataset possible (i.e., after removing test cases and other manual exclusions, but before applying specific exclusion criteria).Analyses were completed using R v4.0.4 (R Foundation for Statistical Computing, Vienna, Austria), including the rms and EValue extension packages.Descriptive statistics for males with BC and females with metastatic BC (MBC) are presented separately but multivariable modeling was not attempted due to small sample sizes.

Baseline characteristics and COVID-19 outcomes in female patients with BC
Of the total 12,034 reports on all cancers submitted to the CCC19 registry at the time of this analysis, 1383 females with BC met the eligibility criteria and were included (Figure 1).The median age for the cohort was 61 years (IQR 51-72 years) and median follow-up was 90 (IQR 30-135) days.BC subtypes by biomarker distribution in CCC19 registry included: 52% HR+/HER2-, 14% HR+/HER2+, 8% HR-/HER2+, 11% triple negative, and 14% unknown or missing.BC subtype distribution based on biomarkers in the CCC19 cohort are similar to SEER data which adds broader applicability of these findings (SEER, 2022).With regard to BC status, 27% were in remission/NED for over 5 years and 32% were in remission/NED for less than 5 years since the initial BC diagnosis and 32% had active cancer (13% had active and responding, 12% had active and stable and 7% had active and progressing cancer).57% of patients had received some form of anti-cancer therapy within 3 months of COVID-19 diagnosis.The unadjusted total all-cause mortality and hospitalization rate, included in the primary ordinal outcome, for the female cohort was 9% and 37%, respectively.However, the unadjusted rates of COVID-19 outcomes varied by their BC status; females with active and progressing cancer had the highest all-cause mortality (38%) and hospitalization rates (72%) compared to the rest of the group (Appendix 5-table 1).Other clinical outcomes for the female cohort included 30-day all-cause mortality (6%), mechanical ventilation (5%), and ICU care (8%).Additional details on patients with BC and COVID-19 by specific characteristics of interest are presented below.

BC treatment characteristics
758 (55%) out of 1383 female patients with BC received some form of systemic treatment within 3 months prior to COVID-19 diagnosis, and specific drug information was available for 679 (90%) (Table 3).Of these 679 patients, the most common systemic therapy was endocrine therapy alone (n=336, 49.5%).This was followed by chemotherapy in 163 (24%) patients who received it either as single agent (n=55, 8%) or combination chemotherapy (n=60, 9%) or combined with anti-HER2 therapy (n=48, 7%).78 (11.5%) patients received anti-HER2 therapy with or without endocrine therapy, and 63 (9%) patients received CDK4/6 inhibitors with or without endocrine therapy.6.89-22.6]).Association between Hispanic ethnicity, obesity, pre-existing renal disease, anti-cancer treatment modalities including all forms of systemic therapy and locoregional therapy, month/year, and geographic region of COVID-19 diagnosis and COVID-19 severity did not reach statistical significance (Table 4).The e value for the COVID-19 severity OR and CI for each racial group are shown in Appendix 7-table 1.This value demonstrates the impact of unknown _residual_ confounding above that adjusted for by including adjustment variables in the multivariable model.For example, an unmeasured confounder would need to be associated with both race and mortality with an OR of at least 1.97 to fully attenuate the observed association for Black females and the OR would need to be at least 1.47 for the null-hypothesized value (1.0) to be included in the CI.Similarly, e value estimates are noted for AAPI and Other groups.The unmeasured confounding for other races based on the e value is larger than most documented associations in the CCC19 cohort (Grivas et al., 2021).

Discussion
In this large, multi-institutional and racially diverse cohort of females with BC and COVID-19 from CCC19 registry, we assessed the clinical impact of COVID-19.The all-cause mortality from COVID-19 was 9% and hospitalization rate was 37%, which is numerically lower than in the entire CCC19 cohort at 14% and 58%, and other previously reported studies of COVID-19 in patients with cancer (Grivas et al., 2021;Garassino et al., 2020;Lee et al., 2020;Wang et al., 2021;de Azambuja et al., 2020;Albiges et al., 2020;Zhang et al., 2021).These differences in outcomes could indicate differences in the immunocompromised status of patients due to intensity of therapy regimens, complex comorbidities, or concomitant medications, which may affect outcomes.Females with BC, however, form a heterogenous group, and the rates of outcomes varied widely with their disease status; patients with active and progressing cancer had the highest total all-cause mortality (38%) and hospitalization rates (72%).We observed older age, pre-existing cardiovascular and pulmonary comorbidities, diabetes mellitus, worse ECOG PS, and active and progressing cancer status were associated with adverse COVID-19 outcomes in females with BC.Prior studies have reported similar factors to be associated with adverse COVID-19 outcomes in patients with all cancer types.The majority of these studies have reported older age to be an important prognostic factor for adverse outcomes from COVID-19, including mortality, which is consistent with data presented here (Grivas et al., 2021;Sharafeldin et al., 2021;Lièvre et al., 2020;Zhang et al., 2021;Chavez-MacGregor et al., 2022).Non-cancer comorbidities, contributing to poor COVID-19 outcomes, as noted in our study, have also been a consistent finding in patients with and without a cancer diagnosis (Grivas et al., 2021;Sharafeldin et al., 2021;Lièvre et al., 2020;Chavez-MacGregor et al., 2022;CDC, 2020c).Similarly, poor ECOG PS in cancer patients has been noted to be an important factor associated with worse COVID-19 severity, including our study (Grivas et al., 2021;Albiges et al., 2020;Lièvre et al., 2020).While obesity was reported in some cancer studies to have a negative impact on COVID-19 (Grivas et al., 2021;Chavez-MacGregor et al., 2022), our study did not identify this association.In this cohort of females with BC, all forms of anti-cancer therapy were thoroughly evaluated and none of the systemic therapies including chemotherapy, endocrine therapy, and targeted therapy (anti-HER2, CDK4/6 inhibitors, other non-HER2 or non-CDK4/6 inhibitors), or locoregional therapy (surgery and radiation) received within 3 months of COVID-19 diagnosis was significantly associated with adverse COVID-19 outcomes.Our finding suggests that systemic therapy for females with BC may not add excess COVID-19 risk.Multiple large cohort studies and meta-analysis of patients with cancer diagnosed with COVID-19 similarly did not identify active anti-cancer therapy, specifically chemotherapy, as a factor associated with adverse COVID-19 outcomes, which is consistent with our results (Garassino et al., 2020;Lee et al., 2020;Albiges et al., 2020;Zhang et al., 2021;Liu et al., 2021;Jee et al., 2020).However, in contrast, some studies of patients with other cancers have shown a negative impact of chemotherapy (Grivas et al., 2021;Sharafeldin et al., 2021;Lièvre et al., 2020;Chavez-MacGregor et al., 2022) and immunotherapy use (Chavez-MacGregor et al., 2022).These findings have important clinical implications while counselling and providing patient care during the pandemic.
We also report important findings related to the impact of racial/ethnic inequities in females with BC and COVID-19, which adds to the growing body of literature on COVID-19-related racial/ethnic disparities.In our study, Black females with BC had significantly worse COVID-19 outcomes compared to NHW females.Multiple studies have similarly reported Black patients in US with and without cancer diagnosis having significantly worse COVID-19 outcomes (Grivas et al., 2021;Wang et al., 2021;CDC, 2020b); however, our study is the first to show such racial/ethnic disparities in COVID-19 outcomes in females with BC.There was no statistically significant association of worse outcomes for Hispanic females compared to NHW females.This is different in comparison to our overall CCC19 cohort (Grivas et al., 2021), and may be explained by younger age and lower rates of comorbid conditions in Hispanic females compared to NHW females.We also found females belonging to AAPI, and Other racial/ethnic group to have worse COVID-19 outcomes.Notably, females belonging to Black, AAPI, and Other racial/ethnic groups presented with higher rates of moderate or severe symptoms of COVID-19 at baseline, which likely contributed to their worse outcomes.This in turn is possibly related to barriers to health care access, and other socio-cultural reasons for delay in seeking early medical care.Future studies including social determinants of health, access to health care, and lifestyle behaviors, among others, are warranted to identify barriers contributing to worse clinical presentation in racial/ethnic minority groups, and eventually impacting future health policies.
In summary, this is one of the largest cohort studies to evaluate the clinical impact of COVID-19 on females with BC.Strengths of our study include standardized data collection on the most common cancer in females in the US and large sample size to evaluate the effect of major clinical and demographic factors.The study had representative population by race and ethnicity from geographically diverse areas and variable time/period of COVID-19 diagnosis.In addition, our study has detailed manually collected information on both cancer status and treatment modalities which contrasts with other studies that have utilized either of these variables as surrogate.Limitations of this study include the retrospective nature of data and inherent potential for confounding because of its observational nature.It's possible that ascertainment bias could have led to some of the high values observed in specific groups such as females with MBC and those with active and progressing cancer.Additional information on drivers for inequity such as socio-economic status, occupation, income, residence, education, and insurance status may have provided added insights on the root causes for disparities; however, unavailability of these factors does not nullify our current findings of existing racial disparities in COVID-19 outcomes in females with BC.Vaccination status was not part of this study as vaccines were not available during the predominant time frame for this cohort.Data presented here including the risk of hospitalization and death applies to the specific COVID-19 variants prevalent during the study period.Despite these limitations, the study reports important sociodemographic and clinical factors that aid in identifying females with BC who are at increased risk for severe COVID-19 outcomes.Given the largely unknown long-term impact of this novel virus, systematic examination of the post-acute sequelae of COVID-19 in patients with breast and other cancer subtypes is warranted.
Our study addresses an important knowledge gap in patients with BC diagnosed with COVID-19 using the CCC19 registry.In addition to clinical and demographic factors associated with adverse COVID-19 outcomes, racial/ethnic disparities reported here significantly contribute to the growing literature.At this stage, it is irrefutable that one of the principal far-reaching messages the pandemic has conveyed is that any such major stressors on the health care system increases risk of detrimental outcomes to the most vulnerable patient population, including the underrepresented and the underserved.These are important considerations for future resource allocation strategies and policy interventions.We also report an important finding that cancers that are active and progressing are associated with severe COVID-19 outcomes.During the ongoing pandemic, this has significant implications for shared decision-making between patients and physicians.The funders had no role in study design, data collection and interpretation, or the decision to submit the work for publication.
We hypothesize that there will be higher rates of severe COVID-19-related outcomes in the racial and ethnic minority subgroups compared to NHW patients with active or previous history of breast cancer.
• Exploratory aims: 1.To evaluate the frequency of hospitalization, supplemental oxygen use, ICU admission, and use of mechanical ventilation in the various racial ethnic groups.2. To describe the distribution of major clinical, sociodemographic, breast cancer risk factors and outcomes in men with active or previous history of breast cancer diagnosed with COVID-19.3. Assess the rate of major clinical complications such as cardiovascular, pulmonary, gastrointestinal, superimposed infection, vascular thrombosis, and others among various racial and ethnic groups of women with active or previous history of breast cancer.

Study design
Present key elements of study design early in the paper This is a retrospective cohort study using de-identified data from the CCC19 database which is a centralized multi-institution registry of patients with current or past history of cancer diagnosed with COVID-19.Study data are collected and managed using REDCap software hosted at Vanderbilt University Medical Center.

Setting
Describe the setting, locations, and relevant dates, including periods of recruitment, exposure, follow-up, and data collection The CCC19 international registry consists of de-identified data on adult patients (18 years and older) with a current or past history of hematologic malignancy or invasive solid tumor who either have laboratory-confirmed SARS-CoV-2 infection or presumptive diagnosis of COVID-19.The CCC19 registry includes patients with either active cancer or a history of cancer and contains variables related to patient demographics, cancer history, and COVID-19 clinical course including receipt of COVID-19-related therapeutics along with follow-up data.The member institutions of the consortium report data through the online REDCap data collection survey developed by CCC19.Data collection period is ongoing, for the purpose of this analysis, the data collected from March 17, 2020, to February 9, 2021, will be used.

Participants (a) Give the eligibility criteria, and the sources and methods of selection of participants. Describe methods of follow-up
Patients with active or previous history of invasive breast cancer with evaluable self-reported race/ ethnicity data, and with laboratory-confirmed COVID-19 will be our study population.Primary analysis will be restricted to women with active or previous history of breast cancer.Descriptive data on men with active or previous history of breast cancer will be provided separately as part of the exploratory analysis given the small numbers.We will restrict our analysis to patients diagnosed in the US since the racial and ethnic disparities of interest have been previously described in the US.We will also exclude patients who have multiple malignancies including a history of bilateral breast cancer with the exception of contralateral DCIS only.Further, patients who are not evaluable for the primary ordinal outcome or with a data quality score >4 will be excluded.For this analysis, the unknown/not reported category of race and ethnicity will be excluded.

(b) For matched studies, give matching criteria and number of exposed and unexposed
Not applicable as the CCC19 registry does not carry data for cancer patients who are not exposed to COVID-19.

Effect modifiers
None.

Data sources/measurement
For each variable of interest, give sources of data and details of methods of assessment (measurement).Describe comparability of assessment methods if there is more than one group.

Bias
Describe any efforts to address potential sources of bias.
Multivariable regression models will be used to adjust for known confounding variables.

Study size Explain how the study size was arrived at
Study size is based on the number of breast cancer cases reported in the registry at the time of analysis.Breast cancer is the single largest solid tumor cohort within the CCC19 registry accounting for roughly 21% of cases.The numbers are expected to rise given the steep accrual rate.

Quantitative variables
Explain how quantitative variables will be handled in the analyses.If applicable, describe which groupings will be chosen and why.
12. Statistical methods (a) Describe all statistical methods, including those to be used to control for confounding Primary analysis among women Standard descriptive statistics will summarize major clinical, demographic, and breast cancer prognostic factors; clinical complications during hospitalization; and rates of 30-day mortality, hospitalization, oxygen requirement, ICU admission, and mechanical ventilation among racial and ethnic subgroups.Multivariable ordinal and binary logistic regression models will estimate differences in adjusted odds of COVID-19 severity and 30-day mortality, respectively, between racial and ethnic subgroups.Because the ordinal outcome is assessed over patient's total follow-up period, the model will include an offset for (log) follow-up time.Adjustment covariates will be selected first from the 'higher priority' confounders listed above, followed by those listed as 'lower priority'.Coefficients and standard errors from models with different levels of adjustment, variance inflation factors, and clinical judgement will be used to assess model stability.

Descriptive analysis among men
We will calculate standard descriptive statistics for major clinical, demographic, and breast cancer prognostic factors and clinical complications during hospitalization; rates of 30-day mortality, hospitalization, oxygen requirement, ICU admission, and mechanical ventilation among men with active or previous history of breast cancer.
(b) Describe any methods that will be used to examine subgroups and interactions None included.
(c) Explain how missing data will be addressed Multiple imputation will be used to impute missing and unknown data for all variables included in the analysis, with some exceptions: unknown ECOG performance score and unknown cancer status will not be imputed and treated as a separate category in analyses.Imputation will be performed on the largest dataset possible (i.e., after removing test cases and other manual exclusions, but before applying specific exclusion criteria).At least 10 imputed datasets will be used.
(d) If applicable, explain how loss to follow-up will be addressed All observed outcomes will be used with models adjusted for duration of follow-up.
(e) Describe any sensitivity analyses

Figure 1 .
Figure 1.Consort flow diagram.Descriptive flow chart of patients included in the study.

Total Reports Submitted N= 12034 Female Patients with Breast Cancer c and COVID-19 N=1462 Included in Analysis N= 1383 Excluded (N=10,572)
Non-melanoma skin cancer, in situ cancers, premalignant conditions are excluded in this step. a

Table 1 .
Baseline characteristics by race/ethnicity.

Table 1 continued
With regard to baseline severity of COVID-19 at presentation, 39% of Black and 38% of AAPI patients presented with moderate or higher severity of COVID-19 infection compared to 27% in both NHW and Hispanic patients.Table 2 summarizes the clinical outcomes, complications, and interventions, stratified by race/ethnicity.

Table 2 .
Outcomes, clinical complications, . The unadjusted total all-cause mortality and hospitalization rate in females with MBC was 19% and 53% respectively.Further details of baseline characteristics and unadjusted rates of COVID-19 outcomes, complications, and interventions are presented in Appendix 6-table 1 and Appendix 6table 2.

Table 4 .
Adjusted associations of baseline characteristics with COVID-19 severity outcome.the predominant comorbidity (44%).The hospitalization rate was 60% and all-cause mortality was 20%.Additional clinical characteristics, complications, interventions, and unadjusted outcomes among males with BC in the CCC19 registry are provided in Appendix 8-table 1 and Appendix 8table 2.