Abstract
BACKGROUND:
There is limited prior literature comparing survival and economic outcomes between those with and without recurrent disease among older individuals aged at least 66 years with locoregionally advanced head and neck squamous cell carcinoma (LA HNSCC).
OBJECTIVE:
To estimate the burden of recurrence among older patients with LA HNSCC in the United States.
METHODS:
Older patients diagnosed with LA HNSCC (2007-2019) from the linked Surveillance, Epidemiology, and End Results (SEER)-Medicare data were included. Overall survival (OS), health care resource utilization (HCRU), and health care costs were compared between patients with or without recurrence, separately for patients receiving surgical (resected) vs nonsurgical (unresected) primary treatment.
RESULTS:
In the resected group (n = 502), 333 patients experienced recurrence and had a 63% increased risk of death vs nonrecurrent patients (adjusted hazard ratio [aHR] = 1.63; 95% confidence interval [CI] = 1.24-2.15; P < 0.001). In those unresected (n = 1,217), 791 patients experienced recurrence and had a 77% increased risk of death (aHR = 1.77; 95% CI = 1.49-2.10; P < 0.001). Recurrence was associated with higher HCRU and monthly health care costs in both resected ($3,831; 95% CI = $2,319-$5,360) and unresected ($4,730; 95% CI = $3,611-$5,870) groups (P < 0.001).
CONCLUSIONS:
Recurrence in older patients with LA HNSCC is associated with shorter OS and higher HCRU/health care costs.
Plain language summary
In this study, we compared survival and health care costs between older patients with locoregionally advanced head and neck squamous cell carcinoma (LA HNSCC), with or without recurrence. We looked at patients with and without surgery separately. Patients with recurrence were 63% (with surgery) or 77% (without surgery) more likely to die from any cause than those without recurrence. Monthly health care costs were $3,831 (with surgery) or $4,730 (without surgery) higher in patients with recurrence than those without.
Implications for managed care pharmacy
In this real-world study, recurrence in older adult patients with locoregionally advanced head and neck squamous cell carcinoma (LA HNSCC) was associated with shorter overall survival and higher health care resource utilization and costs. These findings highlight the need for more effective and cost-effective therapies that may delay disease recurrence or progression and potentially reduce the considerable economic burden associated with LA HNSCC.
Head and neck squamous cell carcinoma (HNSCC) is the most common type of head and neck cancer, with an estimated 54,000 new cases diagnosed in 2022 (approximately 3% of all cancers) in the United States.1,2 HNSCC develops from the mucosal epithelium of the oral cavity, pharynx, and larynx, resulting in site-specific symptoms like local swelling, pain, bleeding, and trouble breathing, speaking, or hearing, depending on the location of the tumor.3,4 HNSCC more commonly affects older individuals, with a median age of 64 years at diagnosis.1
Approximately 60% of patients with HNSCC are diagnosed with locoregionally advanced (LA) disease.2,5 Management of LA HNSCC typically requires an aggressive, multimodal approach with intent to cure while preserving function.3,6 Patients may receive a combination of surgical resection, radiation, and systemic therapy (eg, chemotherapy, targeted therapy, or immunotherapy) depending on the location of the primary tumor and involved nodes.3,6 For patients with unresectable HNSCC, concurrent chemotherapy and radiation is the standard of care for functional preservation.6 Given the available therapeutic options and the need for individualized care, treatment of LA HNSCC requires a multidisciplinary approach that encompasses collaborative efforts from several physicians and services, including medical oncologists, surgeons, radiation oncologists, physiatrists, speech therapy, physical therapy, and dental care.6,7
Although some patients with newly diagnosed LA HNSCC may achieve a cure with initial treatment, approximately 40%-50% of patients eventually develop locoregional recurrence and 20%-30% develop metastatic recurrence.8 Recurrent or metastatic HNSCC is associated with a poor prognosis and most cases are incurable, with a median survival of approximately 12 months. However, prognosis may vary by disease subtype; for example, patients with human papillomavirus (HPV)–positive oropharyngeal cancers often have improved outcomes compared with those with HPV-negative disease.6,8 From an economic perspective, because of the complexity of managing recurrent or metastatic disease, patients with disease progression often require more frequent hospitalizations, emergency department visits, and specialist consultations, leading to increased health care costs.
In literature, a limited number of existing studies have demonstrated the substantial health care resource utilization (HCRU) and costs associated with HNSCC but not specifically for LA HNSCC or when it recurs.9–11 Of note, a systematic literature review suggested that recurrent or metastatic head and neck cancer may be associated with elevated health care costs, but more comparable data were needed to draw any conclusions.12 As the treatment landscape for LA HNSCC continues to evolve, real-world evidence on the clinical and economic burden of disease recurrence can help highlight the unmet need to improve patient outcomes through more effective therapies and strategies to prevent or delay recurrence. Therefore, this retrospective study was conducted to evaluate the real-world impact of recurrence on overall survival (OS), HCRU, and health care costs among patients with LA HNSCC receiving primary treatments in the United States.
Methods
DATA SOURCE
This study used the Surveillance, Epidemiology, and End Results (SEER)-Medicare database to identify patients diagnosed with LA HNSCC between 2007 and 2019 who received primary treatment (defined as 1 or more guideline-recommended treatment modalities of surgery, radiation, or systemic therapy initiated within 4 months of initial diagnosis). The SEER-Medicare database integrates 2 comprehensive data sources: (1) SEER cancer registries, which provide clinical, demographic, and mortality information, and (2) Medicare insurance claims, which include health care service utilization during the time of Medicare eligibility.13 The database encompasses all Medicare-eligible individuals in SEER diagnosed with cancer through 2019, with corresponding Medicare claims data available through 2020.
The dataset is limited and complies with Health Insurance Portability and Accountability Act regulations. This study was granted an exemption from full review by the Advarra Institutional Review Board. Consent was not required given the data were de-identified.
STUDY DESIGN AND COHORTS
A sample selection flowchart outlining the eligibility criteria is presented in Supplementary Figure 1 (206.5KB, pdf) (available in online article). Key inclusion criteria were (1) newly diagnosed primary LA HNSCC of the oropharynx, larynx, hypopharynx, or oral cavity between 2007 and 2019; (2) age 66 years or older at diagnosis (to ensure at least 1 year of data before initial diagnosis); (3) receipt of primary treatment for LA HNSCC; and (4) continuous Medicare enrollment before and after the index date. Patients were excluded if they were enrolled in a Health Maintenance Organization any time during the study period.
Eligible patients were first categorized into 2 cohorts based on their primary treatment modality (ie, surgical vs nonsurgical), as studies have shown that clinical and economic outcomes may differ accordingly.12,14 The resected cohort included patients who received surgery within 4 months after the initial diagnosis. The unresected cohort included patients who, within 4 months after the initial diagnosis, did not receive surgery and were treated with radiation therapy and/or systemic therapy. Patients were then further stratified based on recurrence status. Disease recurrence (ie, exposure) was identified using a previously published algorithm that considered both diagnosis codes and treatment initiation.15 Specifically, recurrence was defined as the earlier between the following: (1) at least 2 visits with a secondary malignancy diagnosis code, spaced at least 30 days apart, with the first occurring no earlier than 30 days after primary treatment initiation, and (2) new treatment initiation after primary treatment, including additional surgery, radiation therapy, or systemic therapy.
The index date for patients with recurrence was defined as 30 days before the recurrence event. For those without recurrence, the index date was randomly assigned to match the distribution of time from primary treatment initiation to recurrence in patients with the same primary cancer site with recurrence, so that on average, patients without recurrence were assessed at a comparable time point in their disease trajectory as those who experienced recurrence; specifically, index dates were assigned by randomly sampling from the observed distribution of time from primary treatment initiation to recurrence among patients with recurrence, stratified by primary cancer site, and applying the sampled interval to each patient’s treatment initiation date. The baseline period was defined as the 12-month period before the index date. The follow-up period spanned from the index date until death, end of data availability, or end of Medicare Part A, B, or D eligibility, whichever occurred the earliest.
STUDY OUTCOMES
OS was defined as the time from the index date to all-cause death. Patients were censored at the earlier of loss to follow-up or the end of data availability.
HCRU outcomes included inpatient admissions, emergency department visits, skilled nursing facility stays, outpatient visits, and inpatient length of stay. HNSCC-related HCRU was defined as medical claims associated with an HNSCC diagnosis. HCRU outcomes were summarized as incidence rates (IRs) per patient per month (PPPM), calculated as the total number of events divided by total person-months during follow-up to account for varying follow-up durations.
Mean monthly all-cause and HNSCC-related health care costs per patient, including medical and pharmacy costs, were also summarized. HNSCC-related health care costs were defined as costs associated with HNSCC-related HCRU and drug costs for clinical guidelines–recommended and US Food and Drug Administration–approved therapies for HNSCC. All costs were adjusted to 2023 US dollars (USD) using the Personal Consumption Expenditures medical care component.
STATISTICAL ANALYSIS
All analyses were conducted separately for the resected and unresected cohorts.
Baseline characteristics, including demographic (eg, age, sex, and race), socioeconomic (eg, residential area and region), and clinical (eg, disease stage, tumor site, comorbidity profile, and health care costs) characteristics, were summarized and compared between patients with and without recurrence to describe these 2 cohorts of patients. Continuous variables were reported as means with SD and medians with IQR, whereas categorical variables were presented as frequencies and percentages. Comparisons of baseline characteristics were conducted using chi-square tests for categorical variables and 2-sample t-tests for continuous variables, except for health care costs, for which Mann-Whitney U tests were used.
OS was described using the Kaplan-Meier (KM) method. The proportion of patients experiencing an event, the number of censored patients, and the median time to event were reported. OS was compared between patients with and without recurrence using the log-rank test. Cox proportional hazards models were used to evaluate OS differences adjusting for key baseline characteristics, including primary treatment, age at index date, sex, race, region, disease stage, tumor site, and Charlson comorbidity index (CCI) (excluding cancer). Hazard ratios (HRs) with 95% CIs were reported.
For HCRU, generalized linear models (GLMs) with a negative binomial distribution and a log-link function were employed. To account for variable follow-up durations across patients, an offset term was included. IR ratios between patients with and without recurrence were estimated. Health care cost comparisons were conducted using GLMs with a Tweedie distribution and log-link function. Cost differences between patients with and without recurrence were estimated. Both HCRU and health care cost models adjusted for key baseline characteristics that can be associated with follow-up HCRU and health care costs, including age at index date, sex, race, region, CCI, disease stage, cancer site, primary treatment, and the respective baseline HCRU or health care costs.
Results
PATIENT CHARACTERISTICS
A total of 1,719 patients with LA HNSCC met the study eligibility criteria. Among them, 502 patients underwent primary surgical treatment (ie, resected cohort), with a median follow-up duration of 18.7 months, and 1,217 received nonsurgical primary treatments (ie, unresected cohort), with a median follow-up of 18.5 months. In the resected cohort, 333 patients (66%) experienced recurrence following their primary treatment, compared with 169 (34%) who remained recurrence-free throughout the follow-up period. In the unresected cohort, 791 patients (65%) developed recurrence and 426 did not (35%) (Supplementary Figure 1 (206.5KB, pdf) ).
Baseline demographic and clinical characteristics were largely comparable between patients with and without recurrence in the resected cohort (Table 1). In the unresected cohort, patients who experienced recurrence were on average slightly younger and more often male (69.9% vs 63.1%) compared with those without recurrence (Table 2). In addition, baseline annualized all-cause health care costs were significantly lower among patients with recurrence compared with those without ($51,169 vs $61,890), and similar trends ($27,989 vs $33,891) were observed for HNSCC-related costs (Table 2).
TABLE 1.
Baseline Characteristics of Patients With Resected Disease by Recurrence Statusa
| With recurrence (n = 333) | Without recurrence (n = 169) | P value | |
|---|---|---|---|
| Demographic characteristics | |||
| Age at diagnosis, years | 0.046 | ||
| Mean±SD | 73.4 ±6.4 | 74.7 ±7.0 | |
| Median (IQR) | 71.4 (68.3-77.3) | 73.2 (68.6-78.8) | |
| Age at index, years | 0.085 | ||
| Mean±SD | 74.4 ±6.4 | 75.5 ±7.0 | |
| Median (IQR) | 73.0 (69.1-78.3) | 73.9 (69.6-79.2) | |
| Male, n (%) | 199 (59.8) | 104 (61.5) | 0.700 |
| Race and ethnicity, n (%) | 0.686 | ||
| White | 273 (82.0) | 141 (83.4) | |
| Others | 60 (18.0) | 28 (16.6) | |
| Residential area at diagnosis, n (%) | 0.268 | ||
| Metropolitan areas | 261 (78.4) | 125 (74.0) | |
| Nonmetropolitan areas | 72 (21.6) | 44 (26.0) | |
| Region, n (%) | 0.160 | ||
| West | 129 (38.7) | 53 (31.4) | |
| South | 96 (28.8) | 64 (37.9) | |
| Northeast | 60 (18.0) | 32 (18.9) | |
| Midwest | 48 (14.4) | 20 (11.8) | |
| Year of initial diagnosis, n (%) | 0.223 | ||
| 2007-2010 | 86 (25.8) | 35 (20.7) | |
| 2011-2014 | 102 (30.6) | 48 (28.4) | |
| 2015-2019 | 145 (43.5) | 86 (50.9) | |
| Primary payer at initial diagnosis, n (%) | 0.092 | ||
| Medicare | 261 (78.4) | 121 (71.6) | |
| Others | 72 (21.6) | 48 (28.4) | |
| Clinical characteristics | |||
| Disease stage group, n (%) | 0.380 | ||
| Stage III | 129 (38.7) | 70 (41.4) | |
| Stage IVa | 167 (50.2) | 86 (50.9) | |
| Stage IVb | 32 (9.6) | <11 (<6.5) | |
| Stage IV unspecified | <11 (<3.3) | <11 (<6.5) | |
| Missing stage | <11 (<3.3) | <11 (<6.5) | |
| Tumor site at diagnosis, n (%) | 0.170 | ||
| Oral cavity | 139 (41.7) | 69 (40.8) | |
| Larynx | 109 (32.7) | 66 (39.1) | |
| Oropharynx | 73 (21.9) | >23 (>13.6) | |
| Hypopharynx | 12 (3.6) | <11 (<6.5) | |
| Tumor size at diagnosis, cm | 0.220 | ||
| Mean±SD | 3.8 ±5.7 | 3.5 ±1.6 | |
| Median (IQR) | 3.3 (2.4-4.4) | 3.5 (2.3-4.5) | |
| Missing observations | 33 | 25 | |
| Comorbidity profile | |||
| CCI scoreb | 0.847 | ||
| Mean±SD | 1.8 ±1.9 | 1.8 ±1.7 | |
| Median (IQR) | 1.0 (0.0-3.0) | 1.0 (1.0-3.0) | |
| Primary treatment pattern, n (%) | 0.590 | ||
| Surgery only | 86 (25.8) | 40 (23.7) | |
| Surgery+radiation therapy | 108 (32.4) | 63 (37.3) | |
| Surgery+systemic therapy | <11 (<3.3) | <11 (<6.5) | |
| Surgery+systemic and radiation therapy | 79 (23.7) | 43 (25.4) | |
| Systemic/radiation therapy+surgery | 45 (13.5) | 19 (11.2) | |
| Systemic/radiation therapy+surgery+systemic/radiation therapy | <11 (<3.3) | <11 (<6.5) | |
| Annualized all-cause health care costs per patient, mean ± SD [median], 2023 USD | 68,191 ±70,028 [47,657] | 69,817 ±76,158 [48,188] | 0.926 |
| Annualized HNSCC-related health care costs per patient, mean ± SD [median], 2023 USD | 38,173 ±50,339 [19,723] | 41,109 ±47,862 [27,376] | 0.280 |
P values less than 0.05 were considered statistically significant.
Categories with n less than 11 were suppressed.
The conditions included in the CCI were identified using International Classification of Diseases, Ninth and Tenth Revisions diagnosis codes reported by Quan et al (2005), excluding codes for cancer (Quan H, Sundararajan V, Halfon P, et al, Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130-9. doi:10.1097/01.mlr.0000182534.19832.83). The CCI weights were based on Quan et al (2011) (Quan H, Li B, Couris CM, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol. 2011;173(6):676-82. doi:10.1093/aje/kwq433).
CCI = Charlson comorbidity index; HNSCC = head and neck squamous cell carcinoma; USD = United States dollars.
TABLE 2.
Baseline Characteristics of Patients With Unresected Disease by Recurrence Statusa
| With recurrence (n = 791) | Without recurrence (n = 426) | P value | |
|---|---|---|---|
| Demographic characteristics | |||
| Age at diagnosis, years | 0.007 | ||
| Mean±SD | 73.3 ±5.9 | 74.4 ±7.2 | |
| Median (IQR) | 71.8 (68.8-76.4) | 72.4 (68.7-78.2) | |
| Age at index, years | 0.077 | ||
| Mean±SD | 74.6 ±5.9 | 75.3 ±7.1 | |
| Median (IQR) | 73.4 (70.1-77.9) | 73.6 (69.5-79.1) | |
| Male, n (%) | 553 (69.9) | 269 (63.1) | 0.016 |
| Race and ethnicity, n (%) | 0.109 | ||
| White | 669 (84.6) | 345 (81.0) | |
| Others | 122 (15.4) | 81 (19.0) | |
| Residential area at diagnosis, n (%) | 0.301 | ||
| Metropolitan areas | 654 (82.7) | 342 (80.3) | |
| Nonmetropolitan areas | 137 (17.3) | 84 (19.7) | |
| Region, n (%) | 0.010 | ||
| West | 349 (44.1) | 158 (37.1) | |
| South | 206 (26.0) | 142 (33.3) | |
| Northeast | 134 (16.9) | 83 (19.5) | |
| Midwest | 102 (12.9) | 43 (10.1) | |
| Year of initial diagnosis, n (%) | 0.206 | ||
| 2007-2010 | 166 (21.0) | 82 (19.2) | |
| 2011-2014 | 310 (39.2) | 140 (32.9) | |
| 2015-2019 | 315 (39.8) | 204 (47.9) | |
| Primary payer at initial diagnosis, n (%) | 0.539 | ||
| Medicare | 603 (76.2) | 318 (74.6) | |
| Others | 188 (23.8) | 108 (25.4) | |
| Clinical characteristics | |||
| Disease stage group, n (%) | 0.105 | ||
| Stage III | 264 (33.4) | 172 (40.4) | |
| Stage IVa | 455 (57.5) | 213 (50.0) | |
| Stage IVb | 51 (6.4) | 31 (7.3) | |
| Stage IV unspecified | >10 (>1.3) | <11 (<2.6) | |
| Missing stage | <11 (<1.4) | <11 (<2.6) | |
| Tumor site at diagnosis, n (%) | 0.694 | ||
| Oropharynx | 318 (40.2) | 163 (38.3) | |
| Larynx | 227 (28.7) | 127 (29.8) | |
| Oral cavity | 192 (24.3) | 112 (26.3) | |
| Hypopharynx | 54 (6.8) | 24 (5.6) | |
| Tumor size at diagnosis, cm | |||
| Mean±SD | 3.7 ±4.3 | 3.6 ±1.8 | |
| Median (IQR) | 3.3 (2.5-4.5) | 3.4 (2.4-4.5) | |
| Missing observations | 172 | 93 | |
| Comorbidity profile | |||
| CCI scoreb | 0.755 | ||
| Mean±SD | 1.6 ±1.6 | 1.6 ±1.8 | |
| Median (IQR) | 1.0 (0.0-3.0) | 1.0 (0.0-3.0) | |
| Primary treatment pattern, n (%) | 0.062 | ||
| Radiation therapy only | >221 (>27.9) | >132 (>31.0) | |
| Systemic therapy only | <11 (<1.4) | <11 (<1.4) | |
| Systemic therapy+radiation therapy | 559 (70.7) | 283 (66.4) | |
| Annualized all-cause health care costs per patient, mean±SD [median], 2023 USD | 51,169 53,625 [38,166] | 61,890 ±60,620 [43,321] | <0.001 |
| Annualized HNSCC-related health care costs per patient, mean±SD [median], 2023 USD | 27,989 ±39,290 [12,299] | 33,891 ±43,440 [17,735] | <0.001 |
P values less than 0.05 were considered statistically significant.
Categories with n less than 11 were suppressed.
The conditions included in the CCI were identified using International Classification of Diseases, Ninth and Tenth Revisions diagnosis codes reported by Quan et al (2005), excluding codes for cancer (Quan H, Sundararajan V, Halfon P, et al, Coding algorithms for defining comorbidities in ICD-9-CM and ICD-10 administrative data. Med Care. 2005;43(11):1130-9. doi:10.1097/01.mlr.0000182534.19832.83). The CCI weights were based on Quan et al (2011) (Quan H, Li B, Couris CM, et al. Updating and validating the Charlson comorbidity index and score for risk adjustment in hospital discharge abstracts using data from 6 countries. Am J Epidemiol. 2011;173(6):676-82. doi:10.1093/aje/kwq433).
CCI = Charlson comorbidity index; HNSCC = head and neck squamous cell carcinoma; USD = United States dollars.
OS OF PATIENTS WITH AND WITHOUT RECURRENCE
KM analyses revealed that patients with recurrence had significantly shorter OS compared with those without recurrence, regardless of treatment modality. Within the resected cohort, patients with recurrence had a median OS of 23.8 months vs 44.4 months for patients without recurrence (P = 0.008); the 5-year OS rate was 30.2% compared with 42.5%, respectively (Figure 1). Multivariable Cox proportional hazards modeling indicated a 63% increased risk of death for patients with recurrence (adjusted HR = 1.63; 95% CI = 1.24-2.15; P < 0.001; results not shown in the figure). Similarly, in the unresected cohort, the median OS was 23.7 months for patients with recurrence compared with 64.5 months for those without (P < 0.001), and the 5-year OS rate was 30.6% compared with 50.8%, respectively (Figure 2), with an adjusted HR of 1.77 (95% CI = 1.49-2.10; P < 0.001).
FIGURE 1.
Kaplan-Meier Analysis of OS Among Patients With Resected Head and Neck Squamous Cell Carcinoma, With and Without Recurrence, During the Follow-Up Period
P values less than 0.05 were considered statistically significant.
OS = overall survival.
FIGURE 2.
Kaplan-Meier Analysis of OS Among Patients With Unresected Head and Neck Squamous Cell Carcinoma, With or Without Recurrence, During the Follow-Up Period
P values less than 0.05 were considered statistically significant.
OS=overall survival.
HCRU AND COSTS ASSOCIATED WITH HNSCC RECURRENCE
HCRU (events PPPM) and mean monthly costs per patient post-index were substantially higher in patients with vs without recurrence. In the resected cohort, patients with recurrence had nearly double the rate of all-cause outpatient visits as well as significantly higher rates of all-cause inpatient admissions and all-cause emergency department visits relative to patients without recurrence (Table 3). These differences translated into an adjusted mean monthly increase of $3,831 (95% CI = $2,319-$5,360) in all-cause health care costs (Table 4). Similar patterns were observed in HNSCC-related HCRU and costs. Among patients with unresected disease, those with recurrence similarly incurred elevated HCRU, with adjusted analyses demonstrating an increase of $4,730 (95% CI = $3,611-$5,870) in all-cause costs and $2,372 (95% CI = $1,597-$3,164) in HNSCC-related costs PPPM. The full results of all-cause and HNSCC-related HCRU and costs are presented in Supplementary Tables 1-4 (206.5KB, pdf) .
TABLE 3.
Comparison of All-Cause HCRU Between Patients With and Without Recurrence
| All-cause HCRU | Resected | Unresected | ||
|---|---|---|---|---|
| Adjusted IRR (95% CI)a | P value | Adjusted IRR (95% CI)a | P value | |
| Number of visits, events/person-month | ||||
| Inpatient admissions | 1.73 (1.31-2.29) | <0.001 | 2.94 (2.46-3.53) | <0.001 |
| Emergency department visits | 2.02 (1.47-2.77) | <0.001 | 2.26 (1.85-2.75) | <0.001 |
| Outpatient visits | 2.00 (1.72-2.32) | <0.001 | 1.87 (1.71-2.04) | <0.001 |
| Skilled nursing facility stays | 0.77 (0.45-1.35) | 0.366 | 2.36 (1.55-3.59) | <0.001 |
| Inpatient length of stay, days/person-month | ||||
| Inpatient days | 1.86 (1.26-2.75) | 0.002 | 2.47 (1.93-3.16) | <0.001 |
P values less than 0.05 were considered statistically significant.
IRRs and P values were estimated using univariable generalized linear models with a negative binomial distribution and a log-link function. The adjusted models controlled for age at index date, sex, race, region, Charlson comorbidity index, disease stage at diagnosis, cancer site, primary treatment, and corresponding HCRU during the baseline period. An IRR greater than 1 indicates patients with recurrence incurred more HCRU than patients without recurrence.
HCRU = health care resource utilization; IRR = incidence rate ratio.
TABLE 4.
Comparison of All-Cause Health Care Costs Between Patients With and Without Recurrence
| All-cause health care costs | Resected | Unresected | ||
|---|---|---|---|---|
| Adjusted monthly cost differencea | 95% CI | Adjusted monthly cost differencea | 95% CI | |
| Total costs, 2023 USD | 3,831 | (2,319-5,360) | 4,730 | (3,611-5,870) |
| Medical costs | 3,732 | (2,232-5,247) | 4,649 | (3,541-5,779) |
| Part D pharmacy costs | 92 | (−11-198) | 340 | (17-636) |
Monthly cost differences and P values were estimated using univariable generalized linear models with a Tweedie distribution (a compound Poisson-gamma distribution) and log-link function. The adjusted models controlled for age at index date, sex (reference: male), race (reference: White), region (reference: west), Charlson comorbidity index, disease stage at diagnosis (reference: stage III), cancer site (reference: oropharynx), primary treatment (reference: surgery only), and corresponding costs during the baseline period. A monthly cost difference greater than 0 indicates patients with recurrence incurred more health care costs than patients without recurrence.
USD = United States dollars.
Discussion
This retrospective observational study compared clinical and economic outcomes following primary treatment between older patients with LA HNSCC, with or without recurrence, in the United States, separately for those with resected and unresected disease. Patients with recurrence had a significantly shorter OS from the index date, with a 63%-77% increased risk of all-cause death compared with those without recurrence. From an economic perspective, patients with recurrence incurred significantly higher rates of all-cause and HNSCC-related HCRU, which translated to an adjusted PPPM increase of $3,831 and $4,730 in all-cause costs and $1,820 and $2,372 in HNSCC-related costs for patients with resected and unresected disease, respectively. Importantly, these results were generally consistent across both patients with resected and unresected LA HNSCC.
In this study, 66% and 65% of patients with resected and unresected LA HNSCC, respectively, developed recurrence following their primary treatment, which is in line with recurrence rates in the literature, with 40%-50% developing locoregional recurrence and 20%-30% developing metastatic recurrence.6,8 One prior assessment of patients with LA HNSCC, primarily treated with definitive radiation therapy with or without concurrent systemic therapy and a subset treated with postoperative radiation, reported a lower recurrence rate of 24.3%. However, this population differed from that of the current study in a number of ways, including age (median of 61-63 years vs 73 years, respectively), data source (single center vs SEER-Medicare), and tumor site.16 Notably, 48% of oropharynx cancer cases in this prior study were HPV-positive disease, whereas HPV status was not available in our data. However, the proportion of oropharyngeal tumors in our study was substantially lower (13.6%-21.9%) compared with 70.3% in the prior study. As a result, the overall proportion of HPV-positive oropharyngeal cases in our cohort would be expected to be lower, which may partially explain the higher recurrence rate in our study.
There is limited prior literature comparing survival outcomes between recurrent and nonrecurrent LA HNSCC among older patients. Struckmeier et al retrospectively evaluated survival patterns in patients with oral squamous cell carcinoma who underwent surgical treatment and found that OS was significantly worse when recurrence occurred (log-rank P < 0.001), which is consistent with the current findings.17 In terms of postrecurrence OS, Hoell et al reported a 1-year OS rate of 59% in their retrospective cohort of adults with locally recurrent HNSCC (with or without distant metastasis),18 which is generally aligned with the rates of 65.3%-66.2% estimated among patients with recurrence in the present analysis at 1 year. A separate study by Beckham et al of patients with HNSCC who developed metastasis following initial definitive treatment demonstrated 1-year and 5-year OS rates of 51% and 9%, respectively.19 These rates are lower than those estimated in the current study (5-year OS: 30.2%-30.6%) potentially because Beckham et al exclusively evaluated patients who developed distant recurrence, a population with a particularly poor prognosis.20 Regardless of differences in patient populations, the existing literature aligns with the current study findings highlighting that patients with recurrent HNSCC have suboptimal survival outcomes.
To our knowledge, this is the first study to assess the real-world impact of recurrence on the HCRU and costs of older patients with LA HNSCC in the United States. A prior study by Fisher et al found that adults with recurrent or refractory locally advanced or metastatic head and neck cancers in US community oncology practices incurred mean monthly health care costs of $14,391, more than 50% of which were hospitalization costs (2016 USD).11 Although this study also emphasizes the large economic burden associated with HNSCC recurrence, the costs are not directly comparable with those of the present study given the differences in patient populations (eg, insurance coverage or inclusion of patients with stage IVc disease) and different cost measurements; their costs reflected the actual incurred expenses whereas our costs reflected payments from Medicare. Notably, a systematic literature review from 2014 suggested that health care costs may be higher for patients with recurrent or metastatic head and neck cancer relative to those without, but the large amount of heterogeneity in the published literature limited the comparison of the cost data.12 Given the scarcity of real-world evidence on this topic, the current study contributes comprehensive and up-to-date insight by providing a direct comparison of HCRU and health care costs between older patients with recurrent and nonrecurrent LA HNSCC among populations with resected and unresected disease.
The findings of this study have important long-term implications for the management of LA HNSCC. The estimated incremental all-cause total health care costs with recurrence were $3,831 and $4,730 PPPM in the resected and unresected cohorts, respectively. Under the assumption that these incremental monthly cost differences remain stable over time within the study population, delaying disease recurrence by 1 year with appropriate and effective interventions may yield all-cause health care cost savings of up to $45,972 and $56,760 per patient with resected and unresected disease, respectively. As such, there is an unmet need for more effective and cost-effective treatment options that may delay disease recurrence or progression and subsequently reduce the economic burden of LA HNSCC. It is important to note that our study did not account for certain cost components, such as the expenses associated with novel treatments or indirect costs, including productivity losses for patients and caregivers because of HNSCC. Therefore, the actual cost implications for the broader health care system may differ from our current estimates.
Novel agents and regimens comprising targeted therapy and immunotherapy are currently under investigation in clinical trials in both the resectable and unresectable settings and may broaden the treatment landscape of LA HNSCC in the near future.21,22 Indeed, the phase 3 trial KEYNOTE-689 recently demonstrated that pembrolizumab as neoadjuvant and adjuvant therapy in addition to standard of care (adjuvant radiotherapy or chemoradiotherapy) was associated with significant and clinically meaningful event-free survival benefits (HR = 0.73; 95% CI = 0.58-0.92; P = 0.008) and a positive trend in OS (HR = 0.76; 95% CI = 0.59-0.98) compared with standard of care among patients with newly diagnosed resectable LA HNSCC.23 Accordingly, as the treatment landscape continues to evolve over the next few years, further research is warranted to assess its cost-effectiveness and the resulting impact on the clinical and economic burden of LA HNSCC.
LIMITATIONS
The study findings should be interpreted in the context of some limitations. The patient cohort was restricted to older individuals, as the SEER-Medicare database includes only Medicare beneficiaries. Consequently, the results may not be generalizable to younger patients or those with different types of insurance coverage. Additionally, because of the inherent constraints of administrative claims data, HNSCC recurrence could not be directly identified. Instead, a previously published algorithm based on diagnosis codes, procedure codes, drug codes, and specific assumptions was employed.15 As a result, potential coding inaccuracies may have introduced misclassification or errors in identifying recurrence. Lastly, SEER data lack certain HNSCC-specific details, including HPV status and smoking history, which are important prognostic factors. Future research using datasets that incorporate these variables may provide more comprehensive insights.
Conclusions
In this real-world study of the older population with LA HNSCC in the SEER-Medicare dataset, patients with disease recurrence had significantly shorter OS, higher HCRU, and elevated health care costs compared with those without recurrence, regardless of resection status. These findings highlight the need for more effective and cost-effective therapies that may delay disease recurrence or progression and potentially reduce the considerable economic burden associated with LA HNSCC.
Disclosures
This study was funded by Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA. The study sponsor was involved in several aspects of the research, including the study design, interpretation of data, and writing of the manuscript.
Drs Zheng, Lerman, Tang, and Merchant are employees of Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, and are shareholders of Merck & Co., Inc., Rahway, NJ, USA. Dr Bidadi was an employee of Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, and a shareholder of Merck & Co., Inc., Rahway, NJ, USA, at the time this study was conducted. Dr Zhang, Dr Y. Song, Dr Li, Ms Zhu, and Dr Signorovitch are employees of Analysis Group, Inc., a consulting company that has provided paid consulting services to Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, which funded the development and conduct of this study and manuscript, for this work. Dr R. Song was an employee of Analysis Group, Inc., at the time this study was conducted. Dr Hanna has received consulting fees from Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA.
Acknowledgments
Medical writing assistance was provided by Christine Tam, MWC, an employee of Analysis Group, Inc., a consulting company that has provided paid consulting services to Merck Sharp & Dohme LLC, a subsidiary of Merck & Co., Inc., Rahway, NJ, USA, which funded the development and conduct of this study and manuscript.
Data Availability
This study used the SEER-Medicare Linked Database to generate the data. The SEER-Medicare Linked Database is not publicly available because of the sensitive nature of the data and risk of reidentification. Investigators are required to obtain approval from SEER-Medicare to obtain the data.
This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the National Cancer Institute; Information Management Services, Inc.; and the SEER Program tumor registries in the creation of the SEER-Medicare database. In addition, the collection of cancer incidence data used in this study was supported by the California Department of Public Health pursuant to California Health and Safety Code Section 103885; Centers for Disease Control and Prevention’s National Program of Cancer Registries, under cooperative agreement 1NU58DP007156; the National Cancer Institute’s SEER Program under contract HHSN261201800032I awarded to the University of California, San Francisco, contract HHSN261201800015I awarded to the University of Southern California, and contract HHSN261201800009I awarded to the Public Health Institute. The ideas and opinions expressed herein are those of the author(s) and do not necessarily reflect the opinions of the State of California, Department of Public Health, the National Cancer Institute, and the Centers for Disease Control and Prevention or their contractors and subcontractors.
References
- 1.Barsouk A, Aluru JS, Rawla P, Saginala K, Barsouk A. Epidemiology, risk factors, and prevention of head and neck squamous cell carcinoma. Med Sci (Basel). 2023;11(2):42. doi: 10.3390/medsci11020042 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Wang H, Zheng Z, Zhang Y, et al. Locally advanced head and neck squamous cell carcinoma treatment efficacy and safety: A systematic review and network meta-analysis. Front Pharmacol. 2023;14:1269863. doi: 10.3389/fphar.2023.1269863 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Johnson DE, Burtness B, Leemans CR, Lui VWY, Bauman JE, Grandis JR. Head and neck squamous cell carcinoma. Nat Rev Dis Primers. 2020;6(1):92. doi: 10.1038/s41572-020-00224-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.National Cancer Institute . Head and neck cancers. National Institutes of Health. Accessed February 26, 2025, https://www.cancer.gov/types/head-and-neck/head-neck-fact-sheet
- 5.Black CM, Kong A, Chiu G, et al. Treatment patterns and clinical outcomes of newly diagnosed locally advanced head and neck squamous cell carcinoma (LA HNSCC): A retrospective cohort analysis of national real-world data in England. J Clin Oncol. 2023;41(16)(suppl):e18075. doi: 10.1200/JCO.2023.41.16_suppl.e18075 [DOI] [Google Scholar]
- 6.Muzaffar J, Bari S, Kirtane K, Chung CH. Recent advances and future directions in clinical management of head and neck squamous cell carcinoma. Cancers (Basel). 2021;13(2):338. doi: 10.3390/cancers13020338 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Hansen CC, Egleston B, Leachman BK, et al. Patterns of multidisciplinary care of head and neck squamous cell carcinoma in medicare patients. JAMA Otolaryngol Head Neck Surg. 2020;146(12):1136-46. doi: 10.1001/jamaoto.2020.3496 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Ionna F, Bossi P, Guida A, et al. Recurrent/metastatic squamous cell carcinoma of the head and neck: A big and intriguing challenge which may be resolved by integrated treatments combining locoregional and systemic therapies. Cancers (Basel). 2021;13(10):2371. doi: 10.3390/cancers13102371 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Amonkar MM, Chastek B, Samant N, Teitelbaum A. Economic burden of resected squamous cell carcinoma of the head and neck in a US managed-care population. J Med Econ. 2011;14(4):421-32. doi: 10.3111/13696998.2011.584096 [DOI] [PubMed] [Google Scholar]
- 10.Lang K, Menzin J, Earle CC, Jacobson J, Hsu MA. The economic cost of squamous cell cancer of the head and neck: Findings from linked SEER-Medicare data. Arch Otolaryngol Head Neck Surg. 2004;130(11):1269-75. doi: 10.1001/archotol.130.11.1269 [DOI] [PubMed] [Google Scholar]
- 11.Fisher MD, Fernandes AW, Olufade TO, Miller PJ, Walker MS, Fenton M. Patient characteristics and costs in recurrent or refractory head and neck cancer: Retrospective analysis of a community oncology database. Clin Ther. 2018;40(4):562-73. doi: 10.1016/j.clinthera.2018.02.010 [DOI] [PubMed] [Google Scholar]
- 12.Wissinger E, Griebsch I, Lungershausen J, Foster T, Pashos CL. The economic burden of head and neck cancer: A systematic literature review. PharmacoEconomics. 2014;32(9):865-82. doi: 10.1007/s40273-014-0169-3 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.National Cancer Institute . SEER-Medicare: Brief description of the SEER-Medicare database. National Institutes of Health. Accessed March 2024. https://healthcaredelivery.cancer.gov/seermedicare/overview/
- 14.Gauduchon T, Dalban C, Philippe Z, et al. Comparison between exclusive (chemo)radiation and surgery or induction chemotherapy as first curative treatment for localized or locally advanced head and neck squamous cell carcinomas (L(A)-HNSCC): Sub-analysis of the EPOCKS study. Oral Oncol Rep. 2023;6:1-6. doi: 10.1016/j.oor.2023.100042 [DOI] [Google Scholar]
- 15.Zheng D, Zhang S, Bidadi B, et al. Real-world treatment patterns and clinical outcomes among elderly patients with locoregionally advanced head and neck squamous cell carcinoma in the United States. Front Oncol. 2025;15:1606990. doi: 10.3389/fonc.2025.1606990 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Leeman JE, Li JG, Pei X, et al. Patterns of treatment failure and postrecurrence outcomes among patients with locally advanced head and neck squamous cell carcinoma after chemoradiotherapy using modern radiation techniques. JAMA Oncol. 2017;3(11):1487-94. doi: 10.1001/jamaoncol.2017.0973 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Struckmeier AK, Buchbender M, Lutz R, Kesting M. Improved recurrence rates and progression-free survival in primarily surgically treated oral squamous cell carcinoma - Results from a German tertiary medical center. Clin Oral Investig. 2024;28(5):262. doi: 10.1007/s00784-024-05644-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Hoell PA, Elsayad K, Berssenbruegge H, et al. Efficacy of multimodality approach in patients with recurrent head and neck squamous cell carcinoma. Anticancer Res. 2023;43(3):1255-63. doi: 10.21873/anticanres.16272 [DOI] [PubMed] [Google Scholar]
- 19.Beckham TH, Leeman JE, Xie P, et al. Long-term survival in patients with metastatic head and neck squamous cell carcinoma treated with metastasis-directed therapy. Br J Cancer. 2019;121(11):897-903. doi: 10.1038/s41416-019-0601-8 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20.Wiegand S, Zimmermann A, Wilhelm T, Werner JA. Survival after distant metastasis in head and neck cancer. Anticancer Res. 2015;35(10):5499-502. [PubMed] [Google Scholar]
- 21.Kiyota N, Tahara M, Homma A. Current status and future perspective of postoperative treatment for locally advanced squamous cell carcinoma of the head and neck. Jpn J Clin Oncol. 2024;54(6):613-9. doi: 10.1093/jjco/hyae029 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Zheng S, Feng Y, Li C, Zhang J, Xie K. Induction therapy for locally advanced head and neck squamous cell carcinoma. Oncol Ther. 2023;11(2):185-98. doi: 10.1007/s40487-023-00226-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Uppaluri R, Haddad RI, Tao Y, et al. ; KEYNOTE-689 Investigators. Neoadjuvant and adjuvant pembrolizumab in locally advanced head and neck cancer. N Engl J Med. 2025;393(1):37-50. doi: 10.1056/NEJMoa2415434 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
This study used the SEER-Medicare Linked Database to generate the data. The SEER-Medicare Linked Database is not publicly available because of the sensitive nature of the data and risk of reidentification. Investigators are required to obtain approval from SEER-Medicare to obtain the data.
This study used the linked SEER-Medicare database. The interpretation and reporting of these data are the sole responsibility of the authors. The authors acknowledge the efforts of the National Cancer Institute; Information Management Services, Inc.; and the SEER Program tumor registries in the creation of the SEER-Medicare database. In addition, the collection of cancer incidence data used in this study was supported by the California Department of Public Health pursuant to California Health and Safety Code Section 103885; Centers for Disease Control and Prevention’s National Program of Cancer Registries, under cooperative agreement 1NU58DP007156; the National Cancer Institute’s SEER Program under contract HHSN261201800032I awarded to the University of California, San Francisco, contract HHSN261201800015I awarded to the University of Southern California, and contract HHSN261201800009I awarded to the Public Health Institute. The ideas and opinions expressed herein are those of the author(s) and do not necessarily reflect the opinions of the State of California, Department of Public Health, the National Cancer Institute, and the Centers for Disease Control and Prevention or their contractors and subcontractors.