ELSEVIER
The American Journal of Surgery
journal homepage: www.americanjournalofsurgery.com
E
AJS
The American Journal of Surgery”
+
CLATINO
Original Research Article
Demystifying delays: Factors associated with timely treatment of adrenocortical carcinoma
Check for updates
Jesse E. Passman ª,”, Julia A. Gasior a,b, Sara P. Ginzbergª, Wajid Amjad ª, Amanda Bader ª, Jasmine Hwang ª, Heather Wachtel a,b
a Department of Surgery, University of Pennsylvania Health System, Philadelphia, PA, USA
b Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA
ARTICLE INFO
Keywords:
Adrenocortical carcinoma ACC Adrenal Disparities Delays in care Outcomes Endocrine
ABSTRACT
Background: Delays in management of adrenocortical carcinoma (ACC) may lead to worse outcomes. We assessed for delays in ACC treatment according to sociodemographic factors.
Methods: We performed a retrospective cohort study of patients treated for ACC (2010-2019) utilizing the Na- tional Cancer Database. Cox proportional hazards modeling was used to evaluate the associations between sociodemographic, geographic, and clinical factors and time to intervention from diagnosis.
Results: Across 1399 subjects treated for ACC, the median time to treatment was 27 days (IQR 15-47). Non- Hispanic Black patients (HR 0.798, p = 0.033) and patients aged 40-64 years (HR 0.800, p = 0.008) were at greater risk of delays in care, whereas female patients (HR 1.169, p = 0.011) and those with metastatic disease (HR 1.176, p = 0.010) received more timely care.
Conclusions: Older age, male sex, and Black race were associated with delays in care for ACC though these delays did not translate to worsened overall survival.
1. Introduction
Adrenocortical carcinoma (ACC) is a rare and aggressive cancer, which is often diagnosed at an advanced stage.1-3 Disease-specific sur- vival is relatively high in early-stage ACC, as surgical resection may be curative for patients with localized or oligometastatic disease. 4,5 Indeed, patients with non-metastatic, Stage I-III disease have a median survival of 61 months.6 However, for those with widespread metastases, surgery plays a more limited role in care. Patients are generally managed with systemic chemotherapy with or without radiation, with a median sur- vival of only 8 months and an estimated five-year survival of only 0-28 %.7,8 Thus, timely diagnosis and treatment initiation is generally considered of utmost importance in controlling disease spread and maximizing patient survival.
Inequalities in access to timely medical care in the United States have increased in the past two decades, exacerbating existing racial and so- cioeconomic disparities in health outcomes.9,10 This phenomenon is well-documented in cancer care, with certain patient groups facing disproportionate delays in the initiation of guideline-concordant ther- apy.11 For example, uninsured, Black, and Spanish-speaking women with
breast cancer are more likely to experience delays to treatment ini- tiation.12-14 Among patients with adrenal pathology, prior studies have demonstrated racial disparities in access to and recovery from adrenalectomy.15-17 The etiology of these disparities is multifactorial, derived from differences in insurance coverage, discrimination and bias within the health care system, and communication failures between pa- tients and clinicians.18 To date, however, there have been few studies examining potential disparities in the timeliness of treatment initiation among patients with ACC.
To better understand the opportunities to improve time to treatment in ACC, the goal of this study was to assess for sociodemographic, geographic, and clinical factors that may put patients at risk for delays in care using a large, national cohort.
2. Methods
2.1. Data source
We performed a retrospective cohort study of patients treated for ACC using the National Cancer Database (NCDB). The NCDB is one of the
* Corresponding author. 3400 Spruce Street, 4th Floor, Maloney Building, Philadelphia, PA, 19104, USA. E-mail address: Jesse.Passman@pennmedicine.upenn.edu (J.E. Passman).
https://doi.org/10.1016/j.amjsurg.2024.116048
world’s largest cancer registries, capturing over 70 % of incident malig- nancies in the United States each year.19 A joint initiative of the Amer- ican College of Surgeons Commission on Cancer and the American Cancer Society, this nationwide database features contributions from 1500 accredited cancer programs. This study was deemed exempt from approval by the University of Pennsylvania Institutional Review Board.
2.2. Study cohort
All adult patients with new ACC diagnoses from 2010 to 2019 were identified using the International Classification of Diseases (ICD)-O-3 histology codes for adrenal cortical carcinoma (8370 and 8373) or the ICD-10 code for malignant neoplasm of the adrenal cortex (C74.0). Pa- tients were excluded if they did not receive any treatment (n = 372), if they did not have documentation of staging information (n = 416), or if they were documented as having an identical date of diagnosis and date of first treatment (n = 1064), leaving a final cohort of 1399 patients (Fig. 1). Of the patients excluded for an identical date of diagnosis and date of first treatment, 1019 (95.8 %) were recorded to have equivalent date of diagnosis and date of first surgery.
2.3. Study measures
Patient demographic, geographic, and clinical characteristics were abstracted. For analysis, pathologic stage was utilized if available; otherwise, clinical stage was used. All patients were converted to staging based on the American Joint Committee on Cancer Staging Manual, 8th edition.20 Within each patient’s home ZIP code, geographic variables such as the proportion of residents without a high school degree and the median income level were grouped into quartiles.21 Urbanicity was defined via typology as determined by the United States Department of Agriculture Economic Research Service.21 Distance to treating facility was defined by NCDB as distance between the zip code centroid of a patient’s residence to the address of the hospital that reported the case.21 Hospital type was defined using classifications from the Commission on Cancer: (1) Community Cancer Program (CCP; 100-500 newly diagnosed cancer cases/year), (2) Comprehensive Community Cancer Program (more than 500 cases/year), (3) Academic/Research Program (more than 500 cases/year, with participation in postgraduate medical education), or (4) Integrated Network Cancer Program (multiple facilities providing integrated cancer care, no case quantity requirement).21 However, this variable is suppressed for patients under the age of 40 (n = 262) in the NCDB and so was excluded from the primary analysis. Given the sup- pression of this variable by age, the age variable was separated into three categories: less than 40 years, 40-64 years, and greater than 65 years.
The primary outcome was time from diagnosis to treatment initiation, where treatment initiation was defined as either the first surgical inter- vention, the first course of chemotherapy, or the first course of radio- therapy. The secondary outcome was overall survival.
2.4. Statistical analysis
Descriptive statistics were reported as frequencies with percentages for categorical variables. Means with standard deviations (SD) or me- dians with interquartile ranges (IQR) were reported for normally and non-normally distributed continuous variables, respectively. x2 tests, Student’s t-tests, and Wilcoxon rank-sum tests were used for group comparisons, as appropriate.
Time-to-event analysis was performed using the Kaplan-Meier method. Index time was defined as date of diagnosis. The event was defined as the date of first medical or surgical treatment. Median time to treatment was calculated, and Mantel-Haenszel tests were used to compare unadjusted differences between cohorts. Univariable Cox pro- portional hazards models were created to determine associations be- tween demographic, geographic, and clinical characteristics and overall time to treatment. Covariates of interest and covariates with p-values less
3,251 diagnosed with ACC
372 did not receive treatment
2,879 treated for ACC
416 without staging data
1,064 missing timing data*
1,399 patients included
than 0.3 were integrated into multivariable models. For all time to treatment analyses, a higher hazard ratio was indicative of more expe- dient care delivery, whereas a lower hazard ratio was associated with delayed care.
The primary analysis evaluated time from diagnosis to any treatment across all patients with ACC. A subset analysis was performed for patients over age 40 to determine the impact of adding facility type to the main model. Several additional subset analyses were then performed to examine clinical scenarios of interest. First, given the potential differ- ences in management strategy and need for multidisciplinary consulta- tion in the setting of metastatic disease, time to any treatment was examined after stratification by the presence of distant metastases. Next, to assess whether delays in care differed between treatment modalities, the models were repeated for (a) time to surgery in patients managed with upfront surgery, (b) time between surgery and adjuvant therapy for those who received multimodal treatment, and (c) time to chemotherapy in patients managed with chemotherapy only.
Finally, overall survival was then analyzed using multivariable modeling using the same demographic and clinical characteristics as prior, with time to first treatment as an additional covariate. Statistical inference was performed utilizing Stata, version 17.0 (Stata Corp, College Station, TX).22
3. Results
3.1. Cohort characteristics
A total of 1399 patients with ACC were included. The mean age was 54.0 (SD: 15.4) years, 59.0 % (n = 825) were female, and 77.1 % (n = 1079) were non-Hispanic White. The majority were insured by private insurance [53.6 % (n = 732)] or Medicare [29.5 % (n = 402)]. Most patients [80.8 % (n = 1130)] were categorized as residing in metropol- itan areas. The median distance traveled by patients to the treating fa- cility was 1.5 (IQR: 0.6-4.3) miles.
Nearly two-thirds of patients [63.1 % (n = 882)] presented with Stage IV disease at diagnosis. Most patients received treatment with surgery alone [26.0 % (n = 363)], chemotherapy alone [27.8 % (n = 389)], or surgery and chemotherapy [22.2 % (n = 310)]. These demographic and clinical characteristics are summarized in Table 1.
3.2. Time to therapy
Time between diagnosis and a composite endpoint of any surgical or medical therapy was evaluated as the primary outcome to assess delays in care. Median time to treatment across all patients was 27 (IQR: 15-47) days. Among the patients who presented with non-metastatic disease, 95.2 % (n = 492) were treated with upfront surgery with a median time to treatment of 28 (IQR: 16-49) days. Among the patients who received only chemotherapy for metastatic disease (n = 366), the median time to treatment was 28 (IQR: 18-53) days.
On univariable analysis, age >40 years, non-Hispanic Black race/ ethnicity, Medicare insurance, and increasing Charlson-Deyo score were associated with greater time to treatment, whereas female sex, the highest income quartile, the highest education quartile, treatment at an Academic/Research facility, and presence of metastatic disease were associated with more expedient care (Table 2). These differences trans- lated to a median time to treatment of 26 days (IQR: 15-45) for non- Hispanic White patients, 35 days (IQR: 19-57) for non-Hispanic Black patients (p = 0.020 compared to White patients; Fig. 2a), 32 days (IQR: 19-50) for Hispanic White patients (p = 0.540), and 27 days (IQR: 17-42) for Hispanic non-White patients (p = 0.957). The median time to treatment was significantly shorter for female patients [26 days (IQR: 15-44)] than for male patients [29 days (IQR: 16-49), p = 0.011] (Fig. 2b). Time to treatment varied by age; patients under the age of 40 received more expedient care [20 days (IQR: 12-34)] than those between age 40-64 [28 days (IQR: 16-46)] and those >65 years [33 days (IQR: 18-52)] (p <0.001) (Fig. 2c). There was no difference in unadjusted time to treatment by urbanicity (p = 0.428) (Fig. 2d) nor by facility type [CCP: 35 days (IQR: 20-58); Comprehensive CCP: 32 days (IQR: 15-53); Aca- demic/Research: 28 days (IQR: 16-47); Integrated Network: 31 days (IQR: 18-46)] (p = 0.070).
On multivariable modeling, non-Hispanic Black patients (HR 0.798, 95 % CI 0.649-0.982, p = 0.033) and patients aged 40-64 (HR 0.800, 95 % CI 0.678-0.943, p = 0.008) were at greater risk of delays in care, whereas female patients (HR 1.169, 95 % CI 1.037-1.320, p = 0.011) and those with metastatic disease (HR 1.176, 95 % CI 1.010-1.331, p = 0.010) received more timely care (Table 2, Fig. 3).
To assess the role of the treating facility type on time to initiation of therapy, a subset analysis was performed for patients over the age of 40. When facility type was incorporated into the multivariable model, the previously observed differences in time to care according to race and sex were no longer significant. Rather, the presence of metastatic disease (HR 1.180, 95 % CI 1.030-1.352, p = 0.017), rural location (HR 1.753, 95 %
| All Patients (n =1399) | Metastatic Disease (n = 882) | Non- Metastatic Disease (n = 517) | p-value | |
|---|---|---|---|---|
| Mean Age, years (SD) | 54.0 (15.4) | 53.3 (15.3) | 55.2 (15.5) | 0.031* |
| Sex, n (%) | ||||
| Male | 574 (41.0) | 360 (40.8) | 214 (41.4) | 0.832 |
| Female | 825 (59.0) | 522 (59.2) | 303 (58.6) | |
| Race, n (%) | ||||
| White | 1188 (84.9) | 740 (83.9) | 448 (86.7) | 0.292 |
| Black | 130 (9.3) | 85 (9.6) | 45 (8.7) | |
| Other | 81 (5.8) | 57 (6.5) | 24 (4.6) | |
| Spanish/Hispanic Origin, n (%) | ||||
| Yes | 135 (9.7) | 92 (10.4) | 43 (8.3) | 0.196 |
| No | 1264 (90.3) | 790 (89.6) | 474 (91.7) | |
| Insurance Status, n (%) | ||||
| Private | 732 (52.3) | 473 (53.6) | 259 (50.1) | 0.046* |
| Medicare | 402 (28.7) | 228 (25.9) | 174 (33.7) | |
| Medicaid | 136 (9.7) | 94 (10.7) | 42 (8.1) | |
| Other Government | 26 (1.9) | 16 (1.8) | 10 (1.9) | |
| Uninsured | 69 (4.9) | 43 (4.9) | 26 (5.0) | |
| Unknown | 34 (2.4) | 28 (3.2) | 6 (1.2) | |
| Charlson-Deyo Score, n (%) | ||||
| 0 | 1026 (73.3) | 656 (74.4) | 370 (71.6) | 0.523 |
| 1 | 257 (18.4) | 154 (17.5) | 103 (19.9) | |
| 2 | 76 (5.4) | 55 (5.1) | 31 (6.0) | |
| ≥3 | 40 (2.9) | 27 (3.1) | 13 (2.5) | |
| Urbanicity | ||||
| Metropolitan | 1130 (80.8) | 718 (81.4) | 412 (79.7) | 0.203 |
| Urban | 176 (12.6) | 110 (12.5) | 66 (12.8) | |
| Rural | 24 (1.7) | 11 (1.3) | 13 (2.5) | |
| Unknown | 69 (4.9) | 43 (4.9) | 26 (5.0) | |
| Median Distance to | 1.5 | 1.4 (0.6-3.8) | 1.7 (0.6-5.0) | 0.100 |
| Facility, miles | (0.6-4.3) | |||
| (IQR) | ||||
| Median Tumor Size, | 11.3 | 11.8 | 11.0 | 0.083 |
| cm (IQR) | (8.0-15.8) | (8.5-16.0) | (7.9-15.2) | |
| Stage of Disease, n (%) | ||||
| I | 39 (2.8) | - | - | – |
| II | 256 (18.3) | – | – | |
| III | 222 (15.9) | – | – | |
| IV | 882 (63.1) | – | – | |
| Median Time to | 27 (15-47) | 26 (14-45) | 29 (16-50) | 0.019* |
| Treatment, days | ||||
| (IQR) | ||||
| Therapy Received, n (%) | ||||
| Surgery Only | 363 (26.0) | 114 (12.9) | 249 (48.2) | < |
| Chemotherapy | 389 (27.8) | 374 (42.4) | 15 (2.9) | 0.001* |
| Only | ||||
| Radiation Only | 53 (3.8) | 51 (5.8) | 2 (0.4) | |
| Surgery and | 310 (22.2) | 187 (21.2) | 124 (23.8) | |
| Chemotherapy | ||||
| Surgery and | 67 (4.8) | 25 (2.8) | 42 (8.1) | |
| Radiation | ||||
| Chemotherapy and | 79 (5.7) | 78 (8.8) | 1 (0.2) | |
| Radiation | ||||
| Surgery, | 138 (9.9) | 53 (6.0) | 86 (16.4) | |
| Chemotherapy, | ||||
| and Radiation | ||||
CI 1.076-2.854, p = 0.047), and treatment at a Comprehensive CCP (HR 1.471, 95 % CI 1.050-2.060, p = 0.025), Integrated Network (HR 1.433, 95 % CI 1.000-2.054, p = 0.050), or Academic/Research facility (HR 1.451, 95 % CI 1.045-2.015, p = 0.026) were associated with shorter time to treatment. Age, sex, race, insurance status, education, income, and distance to hospital were not significantly associated with time to treatment.
3.3. Stratification by presence of metastatic disease
Given the potential impact of the presence of metastatic disease on time to care, multivariable modeling was repeated after stratification by
| Univariable | Multivariable | |||||
|---|---|---|---|---|---|---|
| HR | 95 % CI | P-value | HR | 95 % CI | P-value | |
| Age (Ref: < 40 years) | ||||||
| 40-64 years | 0.758 | 0.659-0.873 | < 0.001* | 0.800 | 0.678-0.943 | 0.008* |
| ≥65 years | 0.662 | 0.564-0.776 | < 0.001* | 0.790 | 0.612-1.021 | 0.072 |
| Female Sex | 1.146 | 1.030-1.276 | < 0.001* | 1.169 | 1.037-1.320 | 0.011* |
| Race/Ethnicity (Ref: White, Non-Hispanic) | ||||||
| White, Hispanic | 0.943 | 0.775-1.149 | 0.563 | 0.931 | 0.738-1.175 | 0.548 |
| Black, non-Hispanic | 0.805 | 0.670-0.968 | 0.021* | 0.798 | 0.649-0.982 | 0.033* |
| Non-White, Hispanic | 1.011 | 0.685-1.492 | 0.957 | 1.029 | 0.630-1.679 | 0.909 |
| Other | 0.931 | 0.713-1.215 | 0.597 | 0.769 | 0.564-1.047 | 0.096 |
| Insurance (Ref: Private) | ||||||
| Uninsured | 0.907 | 0.709-1.162 | 0.440 | 1.003 | 0.772-1.304 | 0.981 |
| Medicaid | 0.859 | 0.715-1.032 | 0.104 | 0.924 | 0.746-1.144 | 0.469 |
| Medicare | 0.732 | 0.648-0.827 | < 0.001* | 0.815 | 0.661-1.007 | 0.058 |
| Other Government | 0.733 | 0.498-1.092 | 0.128 | 0.677 | 0.435-1.054 | 0.084 |
| Income Quartile (Ref: < $40,227)ª | ||||||
| 2nd Quartile: $40,227 - $50,353 | 1.085 | 0.909-1.296 | 0.367 | 1.070 | 0.880-1.301 | 0.498 |
| 3rd Quartile: $50,354 - $63,332 | 1.044 | 0.874-1.247 | 0.636 | 0.990 | 0.803-1.220 | 0.924 |
| 4th Quartile: >$63,333 | 1.274 | 1.080-1.502 | 0.004* | 1.110 | 0.882-1.395 | 0.373 |
| Education Quartile (Ref: ≥ 17.6 %)b | ||||||
| 2nd Quartile: 10.9-17.5 % | 0.960 | 0.815-1.132 | 0.628 | 0.945 | 0.790-1.131 | 0.538 |
| 3rd Quartile: 6.3-10.8 % | 1.152 | 0.980-1.353 | 0.086 | 1.068 | 0.874-1.304 | 0.522 |
| 4th Quartile: < 6.3 % | 1.247 | 1.060-1.467 | 0.008* | 1.098 | 0.878-1.371 | 0.413 |
| Medicaid Expansion State | 0.969 | 0.860-1.092 | 0.605 | - | - | – |
| County Urban Status (Ref: Metro) | ||||||
| Urban | 0.921 | 0.860-1.080 | 0.310 | 0.935 | 0.777-1.123 | 0.475 |
| Rural | 1.162 | 0.776-1.742 | 0.466 | 1.429 | 0.906-2.253 | 0.124 |
| Distance from Hospital > 25 mi | 1.028 | 0.756-1.399 | 0.859 | 0.945 | 0.638-1.397 | 0.775 |
| Facility Type (Ref: CCP) | ||||||
| Comprehensive CCP | 1.201 | 0.877-1.645 | 0.254 | - | - | – |
| Academic/Research | 1.362 | 1.004-1.846 | 0.047* | - | - | – |
| Integrated Network | 1.380 | 0.990-1.923 | 0.058 | - | - | – |
| Charlson-Deyo Score, per point | 0.881 | 0.818-0.949 | 0.001* | 0.928 | 0.857-1.006 | 0.068 |
| Presence of Metastatic Disease | 1.142 | 1.024-1.274 | 0.017* | 1.176 | 1.039-1.331 | 0.010* |
ª Income quartile defined by median household income within home ZIP code.
b Education quartile defined by percentage of adults over 25 years of age without high school degree within home ZIP code.
1.00
A
1.00
B
Proportion Untreated
0.75
Proportion Untreated
0.75
0.50
0.50
0.25
0.25
0.00
p=0.020
0.00
p=0.011
0
20
Days From Diagnosis to Treatment
40
60
80
100
0
15
Days From Diagnosis to Treatment
30
45
60
75
90
White Patients
Black Patients
Male Patients
Female Patients
1.00
C
1.00
D
Proportion Untreated
0.75
Proportion Untreated
0.75
0.50
0.50
0.25
0.25
p<0.001
p=0.428
0.00
0.00
0
15
Days From Diagnosis to Treatment
30
45
60
75
90
0
15
Days From Diagnosis to Treatment
30
45
60
75
90
Age 39 or Younger
Age 40 - 64
Metropolitan Area
Urban Area
Age 65 or Older
Rural Area
the presence of distant metastases. For patients with non-metastatic disease, no factors were significantly associated with time to care on multivariable analysis. In patients with metastatic disease, more
advanced age (age 40-64: HR 0.755, 95 % CI 0.611-0.932, p = 0.009; age ≥65: HR 0.711, 95 % CI 0.511-0.990, p = 0.043) was associated with relative delays in care, whereas female sex (HR 1.175, 95 % CI
Age 40 - 64 Age 65 and Older Female Sex
White Hispanic Black Non-Hispanic
Non-White, Hispanic Other Race Uninsured Medicaid Insurance Medicare Insurance Other Governmental Insurance 2nd Income Quartile 3rd Income Quartile 4th Income Quartile 2nd Education Quartile 3rd Education Quartile 4th Education Quartile Urban Location Rural Location Home >25 Miles from the Hospital Charlson-Deyo Score, per Point Presence of Metastatic Disease
0.5
1.0
Hazard Ratio
1.5
2.0
Age 40 - 64
Age 65 and Older
Female Sex White Hispanic
Black Non-Hispanic
Non-White, Hispanic
Other Race
Uninsured
Medicaid Insurance
Medicare Insurance
Other Governmental Insurance
Disease State
2nd Income Quartile
Non-Metastatic
3rd Income Quartile
Metastatic
4th Income Quartile 2nd Education Quartile
3rd Education Quartile
4th Education Quartile
Urban Location
Rural Location
Home >25 Miles from the Hospital
Charlson-Deyo Score, per Point
Stage II Disease
Stage III Disease
1
2
3
4
5
6
Hazard Ratio
1.009-1.369, p = 0.038) and rural environment (HR 3.039, 95 % CI 1.570-5.883, p = 0.001) were associated with more timely care (Fig. 4).
3.4. Timeliness by type of treatment
For patients with non-metastatic disease who received upfront sur- gery, time to surgery and time to adjuvant therapy (if given) were eval- uated. In these multivariable analyses, no clear relationships emerged between patient characteristics and time to treatment (Table 3).
For patients who were managed with chemotherapy only, time to therapy was evaluated as well. Only rural environment was associated with more timely initiation of care on multivariable analysis (HR 2.635, 95 % CI 1.081-6.422, p = 0.033).
3.5. Survival
In patients with non-metastatic disease, multivariable analysis revealed that time to treatment had no association with survival (HR 1.005, 95 % CI 0.987-1.023, p = 0.611). While age, sex, and race were also not significantly associated with survival, Hispanic ethnicity (HR 2.649, 95 % CI 1.537-4.566, p < 0.001) was associated with increased risk of death.
In patients with metastatic disease, longer time to treatment (HR 0.962, 95 % CI 0.943-0.983, p < 0.001) and Black race (HR 0.645, 95 % CI 0.478-0.870, p = 0.004) were associated with improved survival, while higher comorbidity score was associated with worse survival (HR 1.355, 95 % CI 1.200-1.530, p < 0.001). Age, sex, Hispanic ethnicity, insurance and urbanicity had no association. In patients over age 40, receiving treatment at an Academic/Research Facility compared to a low- volume facility was associated with a survival advantage (HR 0.693; 95 % CI 0.483-0.994; p = 0.046); though this was not significant on multivariable analysis.
4. Discussion
ACC is an aggressive malignancy, with advanced disease portending a significantly worse prognosis compared to tumors that are caught and treated early. Given the importance of expedient treatment initiation for ACC and the previously described disparities in the timely management of other malignancies in the United States, the goal of this study was to identify patient groups at greater risk of delays in care for ACC. No prior research has identified a delay in treatment deemed “safe” or “accept- able; ” surgeons and oncologists generally strive for the fastest treatment initiation possible, given known rapid disease progression. Here, while most patients received treatment within one month of diagnosis, non- Hispanic Black patients, patients older than 40 years of age, and pa- tients with non-metastatic disease were at greater risk for delays in treatment. Additionally, in patients with metastatic disease, advanced age and male sex were associated with less timely care. While these findings were statistically significant, the clinical significance of these delays is unclear. Interestingly, time to treatment was not associated with survival in patients with non-metastatic disease, and in patients with metastatic disease, longer time to treatment was associated with improved survival.
It is unclear why we did not detect an association between time to treatment and survival in patients with non-metastatic disease. It may be that patients with advanced disease are being prioritized for expedient initiation of treatment, that longer wait times are associated with high- quality multidisciplinary care, and/or that time to diagnosis is the more important driver of outcomes. It also may be that there exist mul- tiple different progression phenotypes for ACC. Growth rates for ACC are not well-characterized, given its rarity and the fact that most are resected immediately upon recognition rather than observed. One small retro- spective analysis of adrenal masses followed with observation which were subsequently identified as ACC found significant variability in lesion growth over time; some masses demonstrated size stability for up to 8 years.23 Early-stage ACC is a different disease process than its met- astatic form and the impact of small delays may be ultimately negligible in disease outcome. Regardless, the observed disparities in time to treatment according to patient demographic characteristics add to the existing evidence of disparities in high-quality care in ACC. For instance, prior work has demonstrated that Black patients with ACC are less likely to receive surgical resection and have decreased access to high-volume surgeons. 15-17
Interestingly, distance to the treating facility did not have a
| Surgery | Adjuvant Therapy | |||||
|---|---|---|---|---|---|---|
| HR | 95 % CI | P-value | HR | 95 % CI | P-value | |
| Age (Ref: < 40 years) | ||||||
| 40-64 years | 0.803 | 0.601-1.074 | 0.139 | 1.057 | 0.749-1.492 | 0.751 |
| ≥65 years | 0.920 | 0.599-1.412 | 0.702 | 1.398 | 0.803-2.433 | 0.236 |
| Female Sex | 1.167 | 0.945-1.442 | 0.152 | 0.925 | 0.705-1.212 | 0.570 |
| Race/Ethnicity (Ref: White, Non-Hispanic) | ||||||
| White, Hispanic | 0.776 | 0.513-1.174 | 0.230 | 0.722 | 0.416-1.253 | 0.246 |
| Black, non-Hispanic | 0.764 | 0.519-1.125 | 0.173 | 1.001 | 0.667-1.501 | 0.998 |
| Non-White, Hispanic | 1.026 | 0.415-2.535 | 0.956 | 0.668 | 0.192-2.328 | 0.526 |
| Other | 0.814 | 0.457-1.448 | 0.483 | 0.630 | 0.327-1.214 | 0.167 |
| Insurance (Ref: Private) | ||||||
| Uninsured | 0.979 | 0.622-1.539 | 0.925 | 1.138 | 0.666-1.944 | 0.637 |
| Medicaid | 0.947 | 0.636-1.410 | 0.787 | 1.072 | 0.670-1.714 | 0.772 |
| Medicare | 0.722 | 0.507-1.028 | 0.071 | 0.773 | 0.480-1.244 | 0.289 |
| Other Government | 0.908 | 0.429-1.920 | 0.800 | 1.052 | 0.455-2.429 | 0.906 |
| Income Quartile (Ref: < $40,227)ª | ||||||
| 2nd Quartile: $40,227 - $50,353 | 0.836 | 0.581-1.204 | 0.336 | 0.776 | 0.472-1.277 | 0.319 |
| 3rd Quartile: $50,354 - $63,332 | 0.822 | 0.561-1.206 | 0.317 | 1.065 | 0.607-1.868 | 0.826 |
| 4th Quartile: ≥$63,333 | 0.748 | 0.487-1.148 | 0.184 | 0.835 | 0.460-1.517 | 0.555 |
| Education Quartile (Ref: ≥ 17.6 %)b | ||||||
| 2nd Quartile: 10.9-17.5 % | 1.140 | 0.820-1.585 | 0.435 | 1.626 | 1.010-2.616 | 0.045* |
| 3rd Quartile: 6.3-10.8 % | 1.052 | 0.728-1.521 | 0.785 | 1.231 | 0.753-2.015 | 0.407 |
| 4th Quartile: < 6.3 % | 1.541 | 1.009-2.353 | 0.045* | 1.664 | 0.955-2.899 | 0.072 |
| County Urban Status (Ref: Metro) | ||||||
| Urban | 1.075 | 0.781-1.480 | 0.658 | 1.127 | 0.754-1.683 | 0.560 |
| Rural | 0.746 | 0.384-1.448 | 0.386 | 1.452 | 0.529-3.987 | 0.470 |
| Distance from Hospital > 25 mi | 0.859 | 0.419-1.761 | 0.678 | 2.049 | 0.827-5.076 | 0.121 |
| Charlson-Deyo Score, per point | 0.934 | 0.808-1.079 | 0.355 | 1.066 | 0.894-1.271 | 0.479 |
| Disease Stage (Ref: Stage I) | ||||||
| Stage II | 1.263 | 0.828-1.926 | 0.279 | 0.830 | 0.393-1.754 | 0.626 |
| Stage III | 1.403 | 0.916-2.147 | 0.119 | 0.709 | 0.336-1.496 | 0.367 |
ª Income quartile defined by median household income within home ZIP code.
b Education quartile defined by percentage of adults over 25 years of age without high school degree within home ZIP code.
relationship with overall delays in care in this study; in other cancers this factor has demonstrated an inconsistent and complicated relationship with disease presentation and outcomes.24-27 This finding may indicate that patients receiving treatment for ACC are more willing or more able to overcome such barriers to receive expedient care, as we excluded patients who did not receive treatment. However, we did find that pa- tients living in rural areas had a shorter interval to chemotherapeutic care compared with patients living in metropolitan areas. While our study population included a small number of rural patients and therefore may represent a statistical anomaly, this may also represent differences in facility volume leading to delays or delays attributable to multidisci- plinary evaluation. It is also important to note that the median distance to treating facility was only 1.5 miles, which is likely the result of the combination of the definition for distance to treating facility in the NCDB and the finding that 81 % of included patients lived within a metropol- itan area. The rarity of ACC and the clustering of cases within metro- politan areas makes interpretation of distance to treating facility as a prognostic factor difficult.
Across disease processes, disparities in time to treatment have been shown to be related to a combination of patient, provider, and systemic- level factors.28 At the patient level, for example, health insurance type can limit provider selection and approval of diagnostic studies; low in- come can limit the ability to take time off work and travel for appoint- ments and treatment; and comorbid disease burden can limit transportation and treatment options. Our finding that non-Hispanic Black patients and older patients were at greater risk for delayed initia- tion of ACC treatment is consistent with this prior literature, and is likely driven by a combination of these underlying factors. Systemic-level fac- tors, such as hospital case volume, have also been shown to contribute; facilities that manage a higher volume of complex and/or rare diseases are generally more likely to provide high-quality, guideline-concordant care, and previous research has shown that Black patients are less likely
to receive care at high-volume hospitals.28-30 In this study, patients managed at a low-volume CCP facility were also found to have longer wait times between diagnosis and treatment initiation; one potential explanation for this finding is that, given the rarity of ACC, providers at lower volume centers may be less familiar with management and the time-sensitive nature of this particularly aggressive disease.
Addressing these observed disparities in delays to care in ACC will require further research and multilevel intervention. The pathway from diagnosis to treatment in ACC requires complex, multidisciplinary care, from time of radiographic recognition to diagnosis to ultimate treatment. Most delays to surgery for other adrenal pathologies, such as aldoster- onoma, can be attributed to delays to in hormonal work-up, which may be secondary to differential access to specialist care and lack of famil- iarity amongst non-specialists regarding appropriate next steps in man- agement.31 In other malignancies, patient navigation services have shown promise in improving access to timely care for disadvantaged patient groups.32-35 Our finding that differential delays exist in time to initial rather than adjuvant therapy for ACC implies that most delays may be due to discrepancies in navigation and access to the specialist care required for treatment. Once patients are connected to the appropriate providers for ACC management, differential delays in subsequent treat- ment dissipate. Thus, navigation services provided by higher volume centers could serve to facilitate more expedient care for these patients with an uncommon disease where treatment pathways are not as streamlined.
This investigation is subject to limitations, primarily related to the data source. First, date of diagnosis is documented as equivalent to date of first treatment at high rates in the NCDB, precluding time to treatment analysis for nearly half of patients diagnosed with ACC; this preferen- tially affects patients with non-metastatic disease managed with upfront surgery, presumably because date of diagnosis is marked as the date of resection in 96 % of these patients. Unfortunately, data on first date of
adrenal mass recognition is not available within the NCDB, and so these patients must be excluded. Our cohort thus over-represents patients presenting with metastatic disease and there is potential for bias regarding those with non-metastatic disease. While subset analysis by presence of metastatic disease helps negate this, it important to recognize this in interpreting our results. However, despite this limitation, the strength of the NCDB is in its inclusion of the largest available cross- section of incident ACC diagnoses. This improves external validity of our findings in comparison to more granular, yet less expansive institu- tional datasets. Second, the suppression of facility type for patients under age 40 in the dataset significantly limits our ability to assess the role of facility for patients with ACC. That the addition of facility type to the primary model nullified the previous associations between sex, race and time to treatment suggests potential confounding between these vari- ables, which warrants future investigation in other datasets. Further, though we are able to assess disease stage at presentation, the NCDB does not feature any data on functional activity. Finally, there is no way to determine the order of medical therapies administered within this version of the NCDB database, limiting our subanalyses to composite endpoints of time to adjuvant therapy (chemotherapy or radiotherapy) or time to therapy in patients managed non-operatively with only one non- surgical modality (chemotherapy only). Based on available data, is also not feasible to evaluate treatment access in patients who did not receive a treatment modality, though access to palliative care services is an important tenet in the management of ACC.
Future research should be performed to further elucidate and reme- diate the mechanisms responsible for disparities in timely care for ACC. While this may be a product of differential access to care and ability to navigate a specialized healthcare system, it is also important to recognize the inherent delays in radiographic diagnosis of adrenal nodules. These well-documented diagnostic delays in which nodules are missed on scans or qualified as “benign” may arguably be more important than delays between ACC recognition and its subsequent treatment, but are not the focus of this investigation.36
5. Conclusion
While sociodemographic disparities in timely care for ACC exist in relation to age, sex, and race, these delays do not appear to impact overall survival, noting that we were not able to fully assess this in patients diagnosed surgically. Nevertheless, these delays may be emblematic of other disparities in access to high-quality care for ACC which require further investigation. Multi-level interventions will likely be necessary to improve disparities within the care of this rare disease.
CRediT authorship contribution statement
Jesse E. Passman: Writing - review & editing, Writing - original draft, Visualization, Validation, Supervision, Software, Resources, Proj- ect administration, Methodology, Investigation, Formal analysis, Data curation, Conceptualization. Julia A. Gasior: Writing - review & editing, Writing - original draft, Methodology, Investigation, Formal analysis, Conceptualization. Sara P. Ginzberg: Writing - review & editing, Writing - original draft, Visualization, Methodology, Investigation, Formal analysis, Conceptualization. Wajid Amjad: Writing - review & editing, Writing - original draft, Validation, Investigation, Formal anal- ysis, Conceptualization. Amanda Bader: Writing - review & editing, Writing - original draft, Validation, Methodology, Formal analysis, Data curation, Conceptualization. Jasmine Hwang: Writing - review & edit- ing, Validation, Methodology, Investigation, Data curation, Conceptual- ization. Heather Wachtel: Writing - review & editing, Writing - original draft, Visualization, Validation, Supervision, Software, Resources, Proj- ect administration, Methodology, Investigation, Funding acquisition, Formal analysis, Data curation, Conceptualization.
Funding
HW received funding from the National Institutes of Health, NCI grant #K08 CA270385.
Declaration of competing interest
Disclosures The authors have nothing to disclose.
References
1. Sharma E, Dahal S, Sharma P, et al. The characteristics and trends in adrenocortical carcinoma: a United States population based study. J Clin Med Res. 2018;10(8):636.
2. Kebebew E, Reiff E, Duh QY, Clark OH, McMillan A. Extent of disease at presentation and outcome for adrenocortical carcinoma: have we made progress? World J Surg. 2006;30(5):872-878.
3. Golden SH, Robinson KA, Saldanha I, Anton B, Ladenson PW. Prevalence and incidence of endocrine and metabolic disorders in the United States: a comprehensive review. J Clin Endocrinol Metabol. 2009;94(6):1853-1878.
4. Bilimoria KY, Shen WT, Elaraj D, et al. Adrenocortical carcinoma in the United States: treatment utilization and prognostic factors. Cancer: Interdiscipl Int J Am Cancer Soc. 2008;113(11):3130-3136.
5. Else T, Kim AC, Sabolch A, et al. Adrenocortical carcinoma. Endocr Rev. 2014;35(2): 282-326.
6. Tella SH, Kommalapati A, Yaturu S, Kebebew E. Predictors of survival in adrenocortical carcinoma: an analysis from the national cancer database. J Clin Endocrinol Metabol. 2018;103(9):3566-3573.
7. Fassnacht M, Dekkers OM, Else T, et al. European society of endocrinology clinical practice guidelines on the management of adrenocortical carcinoma in adults, in collaboration with the European Network for the study of adrenal tumors. Eur J Endocrinol. 2018;179(4):G1-G46.
8. Tella SH, Kommalapati A, Yaturu S, Kebebew E. Predictors of survival in adrenocortical carcinoma: an analysis from the national cancer database. J Clin Endocrinol Metab. Sep 1 2018;103(9):3566-3573. https://doi.org/10.1210/jc.2018- 00918.
9. Caraballo C, Ndumele CD, Roy B, et al. Trends in Racial and Ethnic Disparities in Barriers to Timely Medical Care Among Adults in the US, 1999 to 2018. American Medical Association; 2022:e223856.
10. Gurewich D, Beilstein-Wedel E, Shwartz M, Davila H, Rosen AK. Disparities in wait times for care among US veterans by race and ethnicity. JAMA Netw Open. 2023;6(1): e2252061.
11. Clair K, Chang J, Ziogas A, et al. Disparities by Race, Socioeconomic Status, and Insurance Type in the Receipt of NCCN Guideline Concordant Care for Select Cancer Types in California. American Society of Clinical Oncology; 2020.
12. Navarro S, Yang Y, Ochoa C, et al. Breast Cancer Surgical Delays in a Racially and Ethnically Diverse California Cancer Registry Cohort. Wolters Kluwer Health; 2021.
13. Reeder-Hayes KE, Mayer SE, Olshan AF, et al. Race and delays in breast cancer treatment across the care continuum in the Carolina Breast Cancer Study. Cancer. 2019;125(22):3985-3992.
14. Sheni R, Qin J, Viswanathan S, Castellucci E, Kalnicki S, Mehta V. Predictive factors for cancer treatment delay in a racially diverse and socioeconomically disadvantaged urban population. JCO Oncol Pract. 2023;19(6):e904-e915.
15. Hammad AY, Yen TW, Carr AA, Evans DB, Wang TS. Disparities in access to care and outcomes in patients with adrenocortical carcinoma. J Surg Res. 2017;213:138-146.
16. Holoubek SA, Maxwell J, Fingeret AL. Racial disparities of adrenalectomy. J Endocr Soc. 2020;4(9):bvaa110.
17. Holoubek SA, MacKinney EC, Khokar AM, et al. Treatment differences for adrenocortical carcinoma by race and insurance status. J Surg Res. 2022;280: 169-178.
18. Tong M, Hill L, Artiga S. Racial Disparities in Cancer Outcomes, Screening, and Treatment. Kaiser Family Foundation; 2022.
19. Mallin K, Browner A, Palis B, et al. Incident cases captured in the National Cancer Database compared with those in US population based central cancer registries in 2012-2014. Ann Surg Oncol. 2019;26(6):1604-1612.
20. Edition S, Edge S, Byrd D. AJCC Cancer Staging Manual. Eighth Edition 2017: 919-926.
21. National Cancer Database Participant User File 2019 Data Dictionary. American College of Surgeons Commission on Cancer; November 2021.
22. Stata Statistical Software: Release 17. StataCorp LLC; 2021.
23. Nogueira T, Lirov R, Caoili E, et al. Radiographic characteristics of adrenal masses preceding the diagnosis of adrenocortical cancer. Horm Cancer. 2015;6:176-181.
24. Ambroggi M, Biasini C, Del Giovane C, Fornari F, Cavanna L. Distance as a barrier to cancer diagnosis and treatment: review of the literature. Oncol. 2015;20(12): 1378-1385.
25. Ryan S, Serrell EC, Karabon P, et al. The association between mortality and distance to treatment facility in patients with muscle invasive bladder cancer. J Urol. 2018; 199(2):424-429.
26. Steele EM, Robertson SE, Holmes JA. The effect of distance from cancer facility on advanced clinical stage at diagnosis in patients with cervical cancer. Cancer Treat Res Commun. 2020;25:100226.
27. Vetterlein MW, Löppenberg B, Karabon P, et al. Impact of travel distance to the treatment facility on overall mortality in US patients with prostate cancer. Cancer. 2017;123(17):3241-3252.
28. Haider AH, Scott VK, Rehman KA, et al. Racial disparities in surgical care and outcomes in the United States: a comprehensive review of patient, provider, and systemic factors. J Am Coll Surg. 2013;216(3):482-492e12.
29. Holoubek SA, Maxwell J, Fingeret AL. Racial disparities of adrenalectomy. J Endocr Soc. Sep 1 2020;4(9):bvaa110. https://doi.org/10.1210/jendso/bvaa110.
30. Megwalu UC, Ma Y. Racial/ethnic disparities in use of high-quality hospitals among thyroid cancer patients. Cancer Invest. Jul-Aug 2021;39(6-7):482-488. https:// doi.org/10.1080/07357907.2021.1938108.
31. Makris KI, Clark DL, Buffie AW, Steen EH, Ramsey DJ, Singh H. Missed opportunities to promptly diagnose and treat adrenal tumors. J Surg Res. 2022;276:174-181.
32. Syvyk S, Roberts SE, Finn CB, Wirtalla C, Kelz R. Colorectal cancer disparities across the continuum of cancer care: a systematic review and meta-analysis. Am J Surg. 2022;224(1):323-331.
33. Baik SH, Gallo LC, Wells KJ. Patient navigation in breast cancer treatment and survivorship: a systematic review. J Clin Oncol. 2016;34(30):3686.
34. Robinson-White S, Conroy B, Slavish KH, Rosenzweig M. Patient navigation in breast cancer: a systematic review. Cancer Nurs. 2010;33(2):127-140.
35. Freund KM, Battaglia TA, Calhoun E, et al. Impact of patient navigation on timely cancer care: the Patient Navigation Research Program. J Natl Cancer Inst. 2014; 106(6):dju115.
36. Ozsari L, Kutahyalioglu M, Elsayes KM, et al. Preexisting adrenal masses in patients with adrenocortical carcinoma: clinical and radiological factors contributing to delayed diagnosis. Endocrine. 2016;51:351-359.