Abstract
BACKGROUND:
The number of elderly patients with oral squamous cell carcinoma (OCSCC) is increasing as the elderly population increases. Unfortunately, evidence to guide the management of these patients is lacking.
METHODS:
Patients with OCSCC identified from the National Cancer Database (NCDB) were stratified into age-based cohorts. Demographics, comorbidities, and treatment patterns were analyzed. Patients were stratified into early stage (Stage I/II) and advanced stage (Stage III/IV) disease. The likelihood of receiving multimodality therapy by age was calculated using multinomial logistic regression for each stratum while controlling for potential confounders. Cox proportional hazard regression was used to calculate 5-year mortality risk while controlling for potential confounders.
RESULTS:
Surgery alone or palliative options were offered to older patients more frequently. After controlling for confounders, older patients were less likely to receive multimodality therapy for both early stage and advanced stage disease. Patients across all age cohorts had improved 5-year survival with surgery and adjuvant therapy.
CONCLUSION:
Our analyses suggest that elderly patients have unique demographic and pathologic features. They frequently receive less treatment than similarly staged younger patients, yet they benefit from multimodality therapy when feasible. These data suggest an urgent need to critically appraise the care of elderly OCSCC patients within the broader context of their individual comorbidity burden, functional status, and treatment goals.
Keywords: elderly care, head and neck cancer, oral cavity cancer, squamous cell carcinoma, treatment decision making, age-dependent care
BACKGROUND:
In 2019, there were over 53,000 new cases of oral cavity squamous cell carcinoma (OCSCC) in the United States, with more than 13,000 deaths [1]. Patients over the age of 65 are currently the fastest growing demographic in the United States, and within this age group, the incidence of newly diagnosed OCSCC is expected to increase by 60% within the next decade [2]. There is increasing evidence suggesting a unique risk profile for the development of OCSCC in this demographic given their different exposures and prognostic factors, in part due to their medical comorbidities, but also from a lifetime of cumulative carcinogenic insults [3,4].
Because this cohort has frequently been excluded from landmark clinical trials in OCSCC, there is a paucity of evidence to guide the management of elderly patients, particularly those over the age of 80 [5–8]. Multiple studies have demonstrated that oncologic outcomes and treatment complication rates are largely independent of age when other factors are accounted for [9,10], such as preoperative weight loss or BMI <25 [11–13], frailty and co-morbidities [12,14], and surgical time [15]. Additionally, the average 90 year old American woman and man had a life expentancy of 4.78 and 4.05 years in 2017, respectively [16]. Therefore, comprehensive oncologic management may be warranted in appropriate and carefully selected elderly patients. However, clear guidelines and quality metrics for this age group remain poorly defined [17]. Further complicating the challenges in medical decision-making, expanding data and research suggests that medical practitioners believe older patients will intrinsically fare worse due to their advanced age, independent of other prognostic factors [18]. Consequently, many patients are offered less than the standard of care treatment, though practice patterns among elderly patients with OCSCC have not been well characterized [19–21].
Given the increasing incidence of OCSCC in elderly patients, along with heterogeneity in medical decision-making and counseling, a deeper understanding of treatment strategies for this particular population is essential. This study evaluates national practice patterns for the management of OCSCC in elderly patients and characterizes how patient age influences treatment decisions.
METHODS:
We constructed a retrospective cohort of patients diagnosed with OCSCC from 2004 to 2016 from the National Cancer Database (NCDB). The NCDB is an outcomes database for more than 1,500 hospitals accredited by the Commission on Cancer (CoC) in the United States. As a result, roughly 70% of all new cancer diagnoses in the United States are captured. The data is submitted to the NCDB from CoC-accredited cancer registries using standardized data and coding definitions. Patient characteristics, cancer staging, tumor pathologic characteristics, types of therapies, and outcomes are reported in this dataset.
Patient Selection
Patients were included in the study if they met the following criteria: 1) have an ICD-O-3 code for oral cavity cancer (C001-C009, C020-C023, C028-C031, C039-C041, C048-C052, C058-C062, C068, C069); 2) have squamous cell histology (8070–8078) and 3) are greater than 40 years of age.
Patients were divided into cohorts based on age as follows: 40 to 59 years, 60 to 69 years, 70 to 79 years, 80 to 89 years, and 90+ years. Data for patients less than 40 was limited, and these patients were excluded. Patients 40–49 years were similar to patients aged 50–59 years in terms of T and N stage, comorbidity, insurance status, and treatment received. To limit the number of statistical analysis levels, these groups were combined into one cohort.
Demographic information was obtained for each cohort, including gender, race, insurance status, income quartile of their home zip code, Charlson/Deyo Comorbidity Index scores, TNM stage, primary tumor site, and treatment type.
Treatment Type
Treatment types included surgery alone, radiation alone, surgery with radiation, triple modality therapy (e.g. surgery, chemotherapy, and radiation therapy), and palliative treatment. We considered patients who received at least 60 Gy of radiation to have been treated with therapeutic intent. We defined palliative treatment as patients who were documented as having received palliative care in the NCDB, underwent no treatment, received single agent therapy with chemotherapy, or received radiation less than 60 Gy.
Comorbidity Score
The Charlson/Deyo Comorbidity Index score is captured on a 0 through 3+ scale based on ICD-9 or ICD-10 codes captured in the NCDB. A score is provided for each study subject, though individual underlying conditions are not captured. For the purposes of this study, we only included scores 0 through 3+, as very few patients had a score of 4 or higher.
Analysis
Demographics, disease stage, primary tumor site, and treatment were compared across age cohorts with a chi-square test. To assess the likelihood of receiving standard of care therapy, patients were stratified into early stage (i.e. overall Stage I or II) and advanced stage (i.e. Stage III or IV). We used multinomial logistic regression to calculate odds ratios (OR) comparing the odds of each age cohort receiving each treatment type against (a) surgery alone for early-stage disease, or (b) surgery with adjuvant therapy (radiation ± chemotherapy) for advanced stage disease. Other levels of this model included insurance status, comorbidity, race, T stage, and N stage. Standard of care was considered surgery only for early-stage disease and surgery with adjuvant therapy for advanced stage disease. We also calculated hazard ratios (HR) for 5-year overall survival with Cox Proportional Hazard Regression within each age group.
RESULTS
Demographics
Table 1 summarizes the demographics and disease characteristics for patients included in this study. In total, 54,741 patients with oral cavity primary cancers were included in this study. The percentage of patients who were female increased with each increasing age decile for patients (e.g. 57.1% of 80–89 year old patients compared to 32% of 40–59 year old patients). Additionally, the percentage of Black patients decreased with increasing age, with a countervailing increase in the percentage of White patients. Comorbidity scores were comparable across age groups above 60 years of age.
Table 1.
Included patient demographics, comorbidity status, and tumor staging.
| Age Cohorts | |||||||
|---|---|---|---|---|---|---|---|
| 40 – 59 (n=20,383) |
60 – 69 (n=15,227) |
70 – 79 (n=11,123) |
80 – 89 (n=6574) |
90+ (n=1434) |
p-value | ||
| Sex | Male | 13854 (68.0) | 9890 (65.0) | 6302 (56.7) | 2818 (42.9) | 468 (32.6) | <0.001 |
| Female | 6529 (32.0) | 5337 (35.0) | 4821 (43.3) | 3756 (57.1) | 966 (67.4) | ||
| Race/Ethnicity | American Indian | 60 (0.3) | 39 (0.3) | 26 (0.2) | 4 (0.1) | 2 (0.1) | <0.001 |
| Asian/Pacific Islander | 686 (3.4) | 433 (2.9) | 337 (3.1) | 153 (2.4) | 29 (2.1) | ||
| Black | 1698 (8.5) | 977 (6.5) | 505 (4.6) | 227 (3.5) | 63 (4.5) | ||
| Hispanic | 842 (4.2) | 554 (3.7) | 434 (4.0) | 197 (3.1) | 38 (2.7) | ||
| White | 16700 (83.6) | 12922 (86.6) | 9592 (88.0) | 5870 (91.0) | 1269 (90.6) | ||
| Insurance | Uninsured | 1944 (9.9) | 655 (4.4) | 87 (0.8) | 48 (0.7) | 10 (0.7) | <0.001 |
| Private | 12511 (63.4) | 6003 (40.5) | 1145 (10.6) | 590 (9.2) | 109 (7.8) | ||
| Medicaid | 3101 (15.7) | 1054 (7.1) | 191 (1.8) | 93 (1.4) | 23 (1.6) | ||
| Medicare | 1880 (9.5) | 6788 (45.8) | 9329 (86.0) | 5645 (87.9) | 1253 (89.2) | ||
| Other Government | 294 (1.5) | 307 (2.1) | 96 (0.9) | 48 (0.7) | 9 (0.6) | ||
| Median Income (Zip code) | Less than $38,000 | 3868 (19.1) | 2772 (18.3) | 1923 (17.4) | 1089 (16.7) | 265 (18.5) | <0.001 |
| $38,000 – $47,999 | 4998 (24.7) | 3828 (25.3) | 2854 (25.8) | 1627 (24.9) | 329 (23.0) | ||
| $48,000 – $62,999 | 5366 (26.5) | 4068 (26.8) | 3017 (27.3) | 1816 (27.8) | 379 (26.5) | ||
| $63,000 + | 6040 (29.8) | 4488 (29.6) | 3266 (29.5) | 2006 (30.7) | 456 (31.9) | ||
| Charlson/Deyo | 0 | 16747 (82.2) | 11087 (72.8) | 7754 (69.7) | 4620 (70.3) | 1100 (76.7) | <0.001 |
| 1 | 2777 (13.6) | 3062 (20.1) | 2380 (21.4) | 1350 (20.5) | 235 (16.4) | ||
| 2 | 546 (2.7) | 758 (5.0) | 701 (6.3) | 417 (6.3) | 65 (4.5) | ||
| 3+ | 313 (1.5) | 320 (2.1) | 288 (2.6) | 187 (2.8) | 34 (2.4) | ||
| T Stage | 1 | 7636 (42.4) | 5789 (42.9) | 4210 (43.4) | 2218 (38.9) | 431 (34.9) | <0.001 |
| 2 | 5205 (28.9) | 3746 (27.7) | 2663 (27.4) | 1711 (30.0) | 366 (29.6) | ||
| 3 | 1464 (8.1) | 1097 (8.1) | 715 (7.4) | 456 (8.0) | 139 (11.2) | ||
| 4 | 343 (1.9) | 272 (2.0) | 199 (2.1) | 108 (1.9) | 19 (1.5) | ||
| 4A | 3106 (17.2) | 2369 (17.5) | 1809 (18.6) | 1140 (20.0) | 265 (21.4) | ||
| 4B | 275 (1.5) | 228 (1.7) | 110 (1.1) | 76 (1.3) | 16 (1.3) | ||
| N Stage | 0 | 11757 (70.2) | 9224 (72.9) | 7015 (77.5) | 4250 (78.7) | 945 (79.6) | <0.001 |
| 1 | 1819 (10.9) | 1273 (10.1) | 854 (9.4) | 530 (9.8) | 113 (9.5) | ||
| 2 | 2992 (17.9) | 2037 (16.1) | 1123 (12.4) | 0 (11.2) | 125 (10.5) | ||
| 3 | 183 (1.1) | 112 (0.9) | 62 (0.7) | 14 (0.3) | 4 (0.3) | ||
| M Stage | 0 | 19088 (98.2) | 14213 (98.1) | 10461 (98.4) | 6157 (98.2) | 1345 (98.5) | |
| 1 | 356 (1.8) | 279 (1.9) | 170 (1.6) | 114 (1.8) | 20 (1.5) | ||
| Overall Stage | 0 | 145 (0.8) | 113 (0.8) | 84 (0.8) | 37 (0.6) | 3 (0.2) | 0.32 |
| 1 | 7556 (39.5) | 5589 (39.3) | 4129 (40.3) | 2159 (35.9) | 419 (32.9) | ||
| 2 | 2976 (15.6) | 2279 (16.0) | 1759 (17.2) | 1261 (21.0) | 275 (21.6) | ||
| 3 | 2006 (10.5) | 1516 (10.7) | 1011 (9.9) | 617 (10.3) | 156 (12.2) | ||
| 4 | 6455 (33.7) | 4721 (33.2) | 3267 (31.9) | 1933 (32.2) | 421 (33.0) | ||
The percentage of patients presenting with T4a disease increased with increasing age group (e.g. 21.4% of patients over 90 years old versus 17.2% of patients aged 40–59 years) (Table 1). Interestingly, the percentage of patients with N0 nodal status increased with increasing age group (e.g. 70.2% of patients aged 40–59 years versus 77.5% of patients aged 70–79 years versus 78.7% of patients aged 80–89 years), whereas the inverse was true for N1–3.
Age and Treatment
In the univariate analysis, the percentage of patients who received surgery alone increased with increasing age, with the exception of patients over 90 years of age (Figure 1), who were more likely to receive palliative therapy. Conversely, the percentage of patients who received either definitive surgery with adjuvant radiation or surgery and chemoradiation therapy decreased with increasing age. Palliative therapy was provided to 20.5% of patients aged 80–89 years and 38.3% of patients over 90 years of age.
Figure 1.
Percentage of patients receiving each treatment type by age cohort, including both early stage and advanced stage patients. The percentage of patients receiving surgery alone increased with increasing age for patients aged 40 through 89. The percentage of patients receiving surgery and adjuvant radiation therapy or triple modality therapy decreased with increasing age. The percentage of patients per cohort receiving palliative therapy was markedly higher among elderly patients (>80 years).
Figures 2 and 3 summarize the multinomial logistic regression model calculating the odds of receiving palliative therapy or surgery only compared to multi-modality therapy (i.e. surgery and radiation therapy or surgery and adjuvant chemoradiation) when controlling for the confounding effects of comorbidity, insurance, race, and overall stage (Supplemental Table 1). Older adults with both early stage and advanced stage disease were more likely to receive palliative treatment or surgery alone compared to younger patients: These OR increased with increasing age group. For example, early-stage patients aged 70–79 years had an OR of 3.87 (95% CI 2.96–5.06) for receiving palliative therapy compared to patients aged 40–59 years, while for adults ≥90 years this same comparison yielded an OR of 17.86 (95% CI 9.97–31.99) (Figure 2). Similarly, patients with advanced stage disease (overall Stage III or IV) were less likely to receive multi-modality treatment with increasing age (Figure 3). This effect was largest with older cohorts. For example, patients aged 90 years or older had an OR of 6.13 (95% CI 4.06–9.26) of receiving surgery alone instead of surgery with radiation when compared to patients aged 40–59 years. Additionally, patients aged 80–89 years of age had an OR of 0.64 of receiving surgery with chemoradiation therapy compared to surgery with radiation when compared to patients aged 40–59 years. As expected, patients with increasing comorbidity burden had increased odds of receiving single modality therapy or palliative therapy compared to multimodality therapy.
Figure 2.
Multinomial logistic regression model estimating the odds of early-stage patients receiving palliative therapy or surgery alone vs multi-modality treatment (i.e. surgery and radiation or surgery with chemoradiation) relative to 40–59 year old patients, adjusted for the effects of age, comorbidity, insurance, race, and overall stage. Full model included as Supplemental Table 1.
Figure 3.
Multinomial logistic regression model estimating the odds of advanced-stage patients receiving palliative therapy, surgery, and surgery with chemoradiation vs surgery with radiation therapy (i.e. standard of care for advanced OCSCC), relative to 40–59 year old patients, adjusted for the effects of age, comorbidity, insurance, race, and overall stage. Full model included as Supplemental Table 2.
Survival Analysis
Tables 2 and 3 summarize the Cox Proportional Hazard Regression analysis for 5-year OS controlling for the relative effects of insurance, treatment, comorbidity, and overall stage for Stage I/II and Stage III/IV patients, respectively. Early overall stage patients of all age groups who received multiple modality therapy had higher hazard ratios (HR) for 5-year mortality, with the exception of 70–79 year-olds receiving surgery with adjuvant chemoradiation, in whom there was no significant difference from the reference level (Table 2). Notably, patients who received surgery and adjuvant radiation therapy for advanced stage disease had improved 5-year survival rates across all age groups (Table 3). The same trend was observed for patients with advanced stage disease who received surgery with adjuvant chemoradiation, with the exception of 40–59 year-olds, in whom there was no statistically significant difference (HR 0.95, 95% CI 0.86–1.04). As expected, both early-stage and advanced-stage patients with higher comorbidity scores had worse 5-year survival compared to healthier patients across all age cohorts.
Table 2.
Cox proportional hazard regression analysis for hazard of death at 5 years among patients with overall early stage (Stage I/II) disease. Patients in each age cohort who received surgery and adjuvant radiation therapy had worse survival at 5 years when compared to single modality therapy across all age groups, even when controlling for insurance, comorbidity, site, race, and overall stage. Comorbidity was independently associated with increased risk of 5-year mortality across all age cohorts. Ref = reference level.
| Stage I/II (Early Stage) | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| 40–59 yrs | 60–69 yrs | 70–79 yrs | 80+ yrs | ||||||
| HR (95% CI) | p-value | HR (95% CI) | p-value | HR (95% CI) | p-value | HR (95% CI) | p-value | ||
| Insurance | Uninsured | Ref | Ref | Ref | Ref | ||||
| Private | 0.77 (0.62, 0.94) | 0.011 | 0.69 (0.49, 0.96) | 0.026 | 2.63 (0.37, 18.86) | 0.337 | 1.01 (0.49, 2.07) | 0.976 | |
| Medicaid | 1.48 (1.16, 1.89) | 0.002 | 1.48 (1.01, 2.18) | 0.044 | 1.75 (0.23, 13.58) | 0.591 | 1.61 (0.7, 3.69) | 0.264 | |
| Medicare | 1.66 (1.3, 2.13) | 0.000 | 1.01 (0.73, 1.4) | 0.942 | 2.91 (0.41, 20.71) | 0.287 | 1 (0.49, 2.01) | 0.993 | |
| Other government | 1.69 (1.07, 2.68) | 0.025 | 0.73 (0.39, 1.36) | 0.321 | 4.17 (0.52, 33.5) | 0.179 | 1.52 (0.61, 3.79) | 0.368 | |
| Treatment | Surgery Alone | Ref | Ref | Ref | Ref | ||||
| Surgery/Radiation | 1.94 (1.65, 2.3) | 0.000 | 1.48 (1.23, 1.78) | 0 | 1.35 (1.1, 1.65) | 0.004 | 1.66 (1.32, 2.08) | <0.001 | |
| Triple Modality | 2.76 (2.16, 3.53) | 0.000 | 1.98 (1.43, 2.74) | 0 | 0.81 (0.46, 1.4) | 0.446 | 1.97 (0.94, 4.16) | 0.074 | |
| Comorbidity | 0 | Ref | Ref | Ref | Ref | ||||
| 1 | 1.34 (1.15, 1.56) | 0.000 | 1.33 (1.16, 1.52) | 0 | 1.39 (1.23, 1.59) | 0.000 | 1.24 (1.09, 1.4) | 0.001 | |
| 2+ | 1.43 (1.11, 1.84) | 0.005 | 1.99 (1.64, 2.42) | 0 | 1.84 (1.54, 2.2) | 0.000 | 1.65 (1.38, 1.97) | <0.001 | |
| Race | White | Ref | Ref | Ref | Ref | ||||
| Black | 1.01 (0.79, 1.29) | 0.923 | 1.11 (0.85, 1.46) | 0.439 | 1.51 (1.14, 1.99) | 0.004 | 0.78 (0.52, 1.18) | 0.238 | |
| Hispanic | 0.87 (0.63, 1.21) | 0.407 | 0.76 (0.53, 1.11) | 0.154 | 1.04 (0.73, 1.48) | 0.824 | 0.63 (0.39, 1) | 0.051 | |
| Other | 0.85 (0.6, 1.19) | 0.336 | 1 (0.71, 1.42) | 0.98 | 0.72 (0.49, 1.05) | 0.088 | 0.64 (0.42, 0.96) | 0.032 | |
| Overall Stage | I | Ref | Ref | Ref | Ref | ||||
| II | 1.57 (1.39, 1.78) | 0.00 | 1.5 (1.32, 1.69) | 0.000 | 1.48 (1.31, 1.66) | 0.000 | 1.42 (1.28, 1.58) | <0.001 | |
| III | -- | -- | -- | -- | |||||
| IV | -- | -- | -- | -- | |||||
Table 3.
Cox proportional hazard regression analysis for hazard of death at 5 years among patients with overall advanced stage (Stage III/IV) disease. Patients in each age cohort who received surgery and adjuvant radiation therapy had improved survival at 5 years when compared to single modality therapy across all age groups, even when controlling for insurance, comorbidity, site, race, and overall stage. Comorbidity was independently associated with increased risk of 5-year mortality across all age cohorts.
| Stage III/IV (Advanced Stage) | |||||||||
|---|---|---|---|---|---|---|---|---|---|
| 40–59 yrs | 60–69 yrs | 70–79 yrs | 80+ yrs | ||||||
| HR (95% CI) | p-value | HR (95% CI) | p-value | HR (95% CI) | p-value | HR (95% CI) | p-value | ||
| Insurance | Uninsured | Ref | Ref | Ref | Ref | ||||
| Private | 0.74 (0.65, 0.85) | <0.001 | 0.89 (0.71, 1.12) | 0.328 | 0.76 (0.45, 1.29) | 0.313 | 1.65 (0.52, 5.21) | 0.395 | |
| Medicaid | 1.19 (1.03, 1.37) | 0.021 | 1.31 (1.02, 1.69) | 0.036 | 0.77 (0.41, 1.45) | 0.423 | 1.63 (0.47, 5.67) | 0.439 | |
| Medicare | 1.1 (0.93, 1.31) | 0.253 | 1.12 (0.9, 1.39) | 0.322 | 0.89 (0.54, 1.48) | 0.661 | 1.82 (0.58, 5.66) | 0.304 | |
| Other government | 0.93 (0.69, 1.25) | 0.620 | 1.07 (0.77, 1.5) | 0.678 | 0.75 (0.37, 1.54) | 0.436 | 4.25 (1.16, 15.53) | 0.029 | |
| Treatment | Surgery Alone | Ref | Ref | Ref | Ref | ||||
| Surgery/Radiation | 0.72 (0.65, 0.81) | <0.001 | 0.66 (0.59, 0.75) | <0.001 | 0.62 (0.54, 0.7) | <0.001 | 0.64 (0.56, 0.75) | <0.001 | |
| Triple Modality | 0.95 (0.86, 1.04) | 0.281 | 0.82 (0.74, 0.92) | <0.001 | 0.78 (0.68, 0.89) | <0.001 | 0.75 (0.56, 0.99) | 0.045 | |
| Comorbidity | 0 | Ref | Ref | Ref | Ref | ||||
| 1 | 1.13 (1.02, 1.26) | 0.023 | 1.22 (1.09, 1.35) | <0.001 | 1.18 (1.05, 1.33) | 0.005 | 1.23 (1.07, 1.43) | 0.005 | |
| 2+ | 1.51 (1.27, 1.8) | <0.001 | 1.37 (1.17, 1.61) | <0.001 | 1.45 (1.23, 1.72) | <0.001 | 1.51 (1.24, 1.83) | <0.001 | |
| Race | White | Ref | Ref | Ref | Ref | ||||
| Black | 0.96 (0.85, 1.1) | 0.585 | 1.24 (1.06, 1.45) | 0.007 | 1.3 (1.06, 1.61) | 0.012 | 0.97 (0.7, 1.34) | 0.835 | |
| Hispanic | 0.82 (0.67, 1) | 0.052 | 1.03 (0.82, 1.31) | 0.783 | 1.03 (0.8, 1.33) | 0.799 | 1.14 (0.85, 1.54) | 0.369 | |
| Other | 0.85 (0.67, 1.06) | 0.150 | 1.07 (0.82, 1.38) | 0.628 | 1.02 (0.77, 1.35) | 0.877 | 1.04 (0.68, 1.6) | 0.858 | |
| Overall Stage | I | -- | -- | -- | -- | ||||
| II | -- | -- | -- | -- | |||||
| III | Ref | Ref | Ref | Ref | |||||
| IV | 1.75 (1.58, 1.95) | <0.001 | 1.68 (1.5, 1.88) | <0.001 | 1.47 (1.3, 1.66) | <0.001 | 1.24 (1.08, 1.43) | 0.003 | |
DISCUSSION
With over 54,741 included patients from the NCDB, our study represents the single largest study to evaluate treatment patterns in the United States for elderly patients with OCSCC. Our findings demonstrate that the likelihood of receiving multimodality therapy decreases as a function of age. Furthermore, our findings demonstrate that patients of all ages who receive multimodality therapy consisting of surgery with adjuvant therapy experience a decreased risk of death at 5 years, even after controlling for multiple factors, suggesting that treatment decisions based solely on age have the potential to negatively influence patient outcomes.
In our study, patients over the age of 80 were more frequently female compared to the younger patient cohorts, possibly due to the increased life expectancy of women relative to men [22]. Additionally, older patients more frequently presented with advanced T stage; conversely, younger patients were more likely to present with advanced nodal disease. While prior studies have suggested similar trends in demographics and overall cancer stage [3,23–25], our study adds to this body of literature through the inclusion of T and N stage data available in the NCDB. The NCDB does not include information regarding tobacco and alcohol exposure, however prior studies have found that elderly patients report these exposures much less frequently than their younger counterparts [25]. These findings suggest that some elderly patients with HNSCC may present with tumors that are associated with a unique set of underlying risk factors or alternative mechanisms of carcinogenesis, though the influence of sex and socioeconomic factors may also influence these patterns in ways not accounted for in the NCDB.
Our findings demonstrate that elderly patients with advanced disease are less likely to receive multimodal, standard-of-care therapy. Even after controlling for insurance, comorbidity, site, race, and overall stage, we found that patients over the age of 70 had significantly lower odds of receiving multimodal therapy (i.e. surgery with radiation or surgery with chemoradiation) compared to younger patients. Furthermore, the odds of receiving multimodality therapy decreased with increasing age, whereby the eldest of the elderly were most likely to receive surgery alone or palliative therapy. Prior single-center and multi-institutional national database studies utilizing the Surveillance, Epidemiology, and End Results (SEER) have similarly found lower utilization of multimodality therapy in elderly patients, particularly those over the age of 80 [3,18–21].
There are several possible explanations for the discrepancy in treatment of older and younger patients with OCSCC. One possible explanation is that elderly patients have a higher comorbidity burden than younger patients, thereby making them less fit candidates for extensive surgery and systemic therapy. Comorbidity has been shown to be an independent predictor of survival across a broad range of malignancies, including head and neck SCC (HNSCC), even when controlling for age [26]. Furthermore, prior work has shown that older HNSCC patients are more likely to die of a competing cause of death (i.e. a cause other than their index cancer) than are younger patients [4]. Indeed, in our study, comorbidity score independently predicted worse 5-year overall survival in patients over the age of 80. This finding supports prior work utilizing the SEER database by Bhattacharyya and colleagues who found that, while overall survival in patients over 70 years with oral tongue or oropharyngeal SCC was worse than a younger cohort, disease-specific survival was comparable [27]. Their findings suggest that survival differences in elderly patients may be attributable to comorbid conditions. In contrast, our analysis showed that older patients were much less likely to receive multimodality therapy even after adjusting for comorbidity, underscoring the point that factors other than comorbidity may also influence treatment choice in elderly patient populations.
Elderly patients are a heterogeneous population. Treatment decisions for elderly patients with OCSCC are therefore inherently complex and require consideration of many medical and non-medical factors, including comorbidity, functional status, frailty, neurocognitive status, psychosocial supports, individual values, and overall goals of care [28]. Of these other factors, frailty may perhaps be the most important consideration. Frailty overlaps with, but is distinct from, comorbidity and disability [29]. It is increasingly recognized as a major determinant in health outcomes in elderly patients [30]. The NCDB does not record a frailty measure for included patients. It is possible that a number of patients were deemed inappropriate candidates for multimodality therapy based on factors not adequately captured by this database.
Another potential reason for decreased utilization of multimodal treatment in elderly patients is that elderly patients may elect to undergo single modality or palliative therapy at a higher rate than younger patients based on their own preferences and goals. Social isolation, functional decline, and geriatric syndromes disproportionately affect elderly patients, and multimodal therapy with curative intent may not be the optimal decision for every patient [28]. In a prospective cohort study of HNSCC patient preferences, older patients prioritized “living as long as possible” less frequently relative to other potential priorities, including “being cured of my cancer” and other non-oncologic priorities, including “being able to swallow all foods and liquids” [31]. Thus, perceived burden of treatment, potential complications, decrease in post-treatment quality of life, increased burden on family members, or reduced life expectancy are all possible motivating factors for this demographic. Individual patient treatment preferences are inconsistently captured in the NCDB, and our analysis is unable to explicitly account for this potential consideration. Palliative strategies have continued to evolve and may minimize symptom burden, preserving quality of life, and even extending life without achieving oncologic clearance; their potential benefit for appropriately selected patients should not be minimized [32]. From the clinician’s standpoint, a keen awareness of elderly patient preferences and priorities as well as the treatment options available will facilitate a more nuanced and complete discussion of treatment decision-making [18,33].
Notably, our analysis suggests that elderly OCSCC patients who receive multi-modality therapy have a survival benefit compared to patients treated with surgery alone. However, undoubtedly many factors that influence elderly patients’ treatment are not captured adequately by retrospective datasets including NCDB. Other studies have also found a survival benefit for elderly patients with HNSCC of other anatomic subsites treated with standard of care therapy [19,34]. In particular, Ward and colleagues found that the use of systemic therapy improved survival in patients over 71 years of age with advanced stage HNSCC across all mucosal subsites who were treated non-surgically [34]. Importantly, they did not find an age cutoff at which this survival benefit was lost. Similarly, Camilon and colleagues used SEER to evaluate treatment patterns and outcomes in patients with oropharyngeal SCC. In addition to finding that patients received less treatment with increasing age, they found that older patients who did receive appropriate treatment benefited, with comparable HR across all age groups [19]. Several retrospective studies suggest that older adults have comparable perioperative complication rates when compared to younger patients, particularly when controlling for comorbidity burden [35–38]. However, an analysis of the SEER database by Chaudary et al characterized risk factors for 30-day readmission for patients undergoing primary surgery for oropharyngeal and laryngeal cancer. They found that age ≥ 80 years was predictive of 30-day readmission (OR 1.54, 95% CI 1.02–2.32), and that 30-day readmission was associated with increased risk of 30-day mortality for all patients (OR 5.89, 95% CI 2.21–15.70) [39]. In a recent multi-institutional study of elderly patients undergoing major ablation and reconstruction for HNSCC (age ≥ 80 years, n = 376), Fancy et al proposed risk-stratification system based on age (older or younger than 85 years), comorbidity score (Adult Comorbidity Evaluation 27), BMI, and frailty (as measured by the modified frailty index) that stratified patients into three classes [12]. Relative to Class I patients (lowest risk), Class III patients were more likely to have 30-day severe complications (OR 4.50, 95% CI 2.44–8.28), 90-day mortality (OR 6.67, 95% CI 1.40–31.85), and 90-day functional decline (OR 2.49, 95% CI 1.11–5.57). Overall, these data support the notion that patient age should only guide treatment decisions when considered in the broader context of each patient’s individual comorbidities, functional status, oncologic condition, psychosocial support, and personal goals of care.
There are several limitations to our study. The data from the NCDB, while robust in quality and quantity, is nonetheless retrospective data. All data reported in the NCDB are from CoC sites, and it is possible that some included patients who underwent adjuvant treatment at non-CoC sites would be misclassified. Rates of unknown radiation therapy status, however, were comparable across groups (mean 2.72%, SD 0.278%). Additionally, specific causes of death are not captured by the NCDB; consequently, we were unable to calculate disease-specific survival. Exposures, such as tobacco use and alcohol consumption, are also not captured. While the inclusion of comorbidity information is a relative strength of the NCDB over comparable databases, the Charlson-Deyo score can be regarded as a crude metric and specific comorbidity data for individual patients were not available. Future work must investigate these findings in a broader multi-institutional, retrospective cohort to more extensively profile the treatment patterns and outcomes among this select group of patients.
CONCLUSION
Our findings demonstrate that older adults with both early and advanced overall stage OCSCC receive less intense therapy than younger patients. Furthermore, our findings demonstrate that elderly patients who received multimodality therapy with surgery and adjuvant therapy experienced a decreased risk of death at 5 years, even when controlling for stage, tumor site, comorbidity, race, gender, and insurance status. Together, these data suggest an urgent need to critically appraise the oncologic management of elderly patients with OCSCC to identify potential areas for the improved delivery of care.
Supplementary Material
Highlights.
Describes patterns of care for over 54,000 oral squamous cell carcinoma (OSCC) patients
Elderly patients are less likely to receive multi-modality therapy for advanced OSCC
Multi-modality therapy may be beneficial in select elderly OSCC patients
FUNDING SOURCES:
This work was supported by the National Institute of Deafness and Other Communication Disorders (T32DC000022).
Abbreviations:
- OCSCC
oral squamous cell carcinoma
- NCDB
National Cancer Database
- SEER
Surveillance, Epidemiology, and End Results
Footnotes
Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
CONFLICT OF INTERESTS: None declared.
References
- [1].Siegel RL, Miller KD, Jemal A. Cancer statistics, 2019. CA Cancer J Clin 2019;69:7–34. 10.3322/caac.21551. [DOI] [PubMed] [Google Scholar]
- [2].Smith BD, Smith GL, Hurria A, Hortobagyi GN, Buchholz TA. Future of Cancer Incidence in the United States: Burdens Upon an Aging, Changing Nation. J Clin Oncol 2009;27:2758–65. 10.1200/JCO.2008.20.8983. [DOI] [PubMed] [Google Scholar]
- [3].Massa ST, Cass LM, Challapalli S, Zahirsha Z, Simpson M, Ward G, et al. Demographic predictors of head and neck cancer survival differ in the elderly: HNSCC Survival Predictors in the Elderly. The Laryngoscope 2019;129:146–53. 10.1002/lary.27289. [DOI] [PubMed] [Google Scholar]
- [4].Massa ST, Osazuwa-Peters N, Christopher KM, Arnold LD, Schootman M, Walker RJ, et al. Competing causes of death in the head and neck cancer population. Oral Oncol 2017;65:8–15. 10.1016/j.oraloncology.2016.12.006. [DOI] [PubMed] [Google Scholar]
- [5].VanderWalde NA, Fleming M, Weiss J, Chera BS. Treatment of Older Patients With Head and Neck Cancer: A Review. The Oncologist 2013;18:568–78. 10.1634/theoncologist.2012-0427. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [6].Pignon J-P, le Maître A, Maillard E, Bourhis J. Meta-analysis of chemotherapy in head and neck cancer (MACH-NC): An update on 93 randomised trials and 17,346 patients. Radiother Oncol 2009;92:4–14. 10.1016/j.radonc.2009.04.014. [DOI] [PubMed] [Google Scholar]
- [7].Bernier J, Ozsahin M, Lefèbvre J-L, Maingon P, Cognetti F, van Glabbeke M. Postoperative Irradiation with or without Concomitant Chemotherapy for Locally Advanced Head and Neck Cancer. N Engl J Med 2004:8. [DOI] [PubMed] [Google Scholar]
- [8].Cooper JS, Pajak TF, Forastiere AA, Jacobs J, Campbell BH, Saxman SB, et al. Postoperative Concurrent Radiotherapy and Chemotherapy for High-Risk Squamous-Cell Carcinoma of the Head and Neck. N Engl J Med 2004;350:1937–44. 10.1056/NEJMoa032646. [DOI] [PubMed] [Google Scholar]
- [9].Jubbal KT, Zavlin D, Suliman A. The effect of age on microsurgical free flap outcomes: An analysis of 5,951 cases. Microsurgery 2017;37:858–64. 10.1002/micr.30189. [DOI] [PubMed] [Google Scholar]
- [10].Goel AN, Raghavan G, St John MA, Long JL. Risk Factors, Causes, and Costs of Hospital Readmission After Head and Neck Cancer Surgery Reconstruction. JAMA Facial Plast Surg 2019;21:137. 10.1001/jamafacial.2018.1197. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [11].Crippen MM, Brady JS, Mozeika AM, Eloy JA, Baredes S, Park RCW. Impact of Body Mass Index on Operative Outcomes in Head and Neck Free Flap Surgery. Otolaryngol Neck Surg 2018;159:817–23. 10.1177/0194599818777240. [DOI] [PubMed] [Google Scholar]
- [12].Fancy T, Huang AT, Kass JI, Lamarre ED, Tassone P, Mantravadi AV, et al. Complications, Mortality, and Functional Decline in Patients 80 Years or Older Undergoing Major Head and Neck Ablation and Reconstruction. JAMA Otolaryngol Neck Surg 2019. 10.1001/jamaoto.2019.2768. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [13].Mueller SA, Mayer C, Bojaxhiu B, Aeberhard C, Schuetz P, Stanga Z, et al. Effect of preoperative immunonutrition on complications after salvage surgery in head and neck cancer. J Otolaryngol - Head Neck Surg 2019;48:25. 10.1186/s40463-019-0345-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [14].Abt N. Frailty Index: Intensive Care Unit Complications in Head and Neck Oncologic Regional and Free Flap Reconstruction n.d [DOI] [PubMed]
- [15].Kim B, Ver Halen J, Grant D, Kim J. Anesthesia Duration as an Independent Risk Factor for Postoperative Complications in Free Flap Surgery: A Review of 1,305 Surgical Cases. J Reconstr Microsurg 2013;30:217–26. 10.1055/s-0033-1358382. [DOI] [PubMed] [Google Scholar]
- [16].Social Security Administration. Actuarial Life Table n.d https://www.ssa.gov/OACT/STATS/table4c6.html (accessed April 8, 2021).
- [17].Puram SV, Bhattacharyya N. Quality Indicators for Head and Neck Oncologic Surgery: Academic versus Nonacademic Outcomes. Otolaryngol Neck Surg 2016;155:733–9. 10.1177/0194599816654689. [DOI] [PubMed] [Google Scholar]
- [18].Derks W, Leeuw JRJ, Hordijk GJ, Winnubst JAM. Reasons for non-standard treatment in elderly patients with advanced head and neck cancer. Eur Arch Otorhinolaryngol 2005;262:21–6. 10.1007/s00405-004-0744-x. [DOI] [PubMed] [Google Scholar]
- [19].Camilon PR, Stokes WA, Nguyen SA, Lentsch EJ. Are the elderly with oropharyngeal carcinoma undertreated?: Undertreatment of the Elderly with OPSCC. The Laryngoscope 2014;124:2057–63. 10.1002/lary.24660. [DOI] [PubMed] [Google Scholar]
- [20].Sanabria A, Carvalho AL, Vartanian JG, Magrin J, Ikeda MK, Kowalski LP. Factors That Influence Treatment Decision in Older Patients With Resectable Head and Neck Cancer: The Laryngoscope 2007;117:835–40. 10.1097/MLG.0b013e3180337827. [DOI] [PubMed] [Google Scholar]
- [21].Moye VA, Chandramouleeswaran S, Zhao N, Muss HB, Weissler MC, Hayes DN, et al. Elderly Patients With Squamous Cell Carcinoma of the Head and Neck and the Benefit of Multimodality Therapy. The Oncologist 2015;20:159–65. 10.1634/theoncologist.2013-0325. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [22].Kontis V, Bennett JE, Mathers CD, Li G, Foreman K, Ezzati M. Future life expectancy in 35 industrialised countries: projections with a Bayesian model ensemble. The Lancet 2017;389:1323–35. 10.1016/S0140-6736(16)32381-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [23].Sarini J, Fournier C, Lefebvre J-L, Bonafos G, Van JT, Coche-Dequéant B. Head and Neck Squamous Cell Carcinoma in Elderly Patients: A Long-term Retrospective Review of 273 Cases. Arch Otolaryngol Neck Surg 2001;127:1089. 10.1001/archotol.127.9.1089. [DOI] [PubMed] [Google Scholar]
- [24].Barzan L, Veronesi A, Caruso G, Serraino D, Magri D, Zagonel V, et al. Head and neck cancer and ageing: a retrospective study in 438 patients. J Laryngol Otol 1990;104:634–40. 10.1017/s0022215100113453. [DOI] [PubMed] [Google Scholar]
- [25].Gugić J, Strojan P. Squamous cell carcinoma of the head and neck in the elderly. Rep Pract Oncol Radiother 2013;18:16–25. 10.1016/j.rpor.2012.07.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [26].Piccirillo JF. Prognostic Importance of Comorbidity in a Hospital-Based Cancer Registry. JAMA 2004;291:2441. 10.1001/jama.291.20.2441. [DOI] [PubMed] [Google Scholar]
- [27].Bhattacharyya N. A Matched Survival Analysis for Squamous Cell Carcinoma of the Head and Neck in the Elderly: The Laryngoscope 2003;113:368–72. 10.1097/00005537-200302000-00030. [DOI] [PubMed] [Google Scholar]
- [28].Soto-Perez-de-Celis E, Li D, Yuan Y, Lau YM, Hurria A. Functional versus chronological age: geriatric assessments to guide decision making in older patients with cancer. Lancet Oncol 2018;19:e305–16. 10.1016/S1470-2045(18)30348-6. [DOI] [PubMed] [Google Scholar]
- [29].Noor A, Gibb C, Boase S, Hodge J-C, Krishnan S, Foreman A. Frailty in geriatric head and neck cancer: A contemporary review: Frailty in Geriatric Head and Neck Cancer. The Laryngoscope 2018;128:E416–24. 10.1002/lary.27339. [DOI] [PubMed] [Google Scholar]
- [30].Abt NB, Xie Y, Puram SV, Richmon JD, Varvares MA. Frailty index: Intensive care unit complications in head and neck oncologic regional and free flap reconstruction. Head Neck 2017;39:1578–85. 10.1002/hed.24790. [DOI] [PubMed] [Google Scholar]
- [31].Windon MJ, D’Souza G, Faraji F, Troy T, Koch WM, Gourin CG, et al. Priorities, concerns, and regret among patients with head and neck cancer. Cancer 2019;125:1281–9. 10.1002/cncr.31920. [DOI] [PMC free article] [PubMed] [Google Scholar]
- [32].Ferrell BR, Temel JS, Temin S, Alesi ER, Balboni TA, Basch EM, et al. Integration of Palliative Care Into Standard Oncology Care: American Society of Clinical Oncology Clinical Practice Guideline Update. J Clin Oncol 2017;35:96–112. 10.1200/JCO.2016.70.1474. [DOI] [PubMed] [Google Scholar]
- [33].Derks W, Leeuw RD, Winnubst J, Hordijk GJ. Elderly patients with head and neck cancer: physical, social and psychological aspects after 1 year. Acta Otolaryngol (Stockh) 2004;124:509–14. 10.1080/00016480310000656. [DOI] [PubMed] [Google Scholar]
- [34].Ward MC, Reddy CA, Adelstein DJ, Koyfman SA. Use of systemic therapy with definitive radiotherapy for elderly patients with head and neck cancer: A National Cancer Data Base analysis: Elderly Head and Neck Cancer Therapy. Cancer 2016;122:3472–83. 10.1002/cncr.30214. [DOI] [PubMed] [Google Scholar]
- [35].Milet PR, Mallet Y, El Bedoui S, Penel N, Servent V, Lefebvre J-L. Head and neck cancer surgery in the elderly – Does age influence the postoperative course? Oral Oncol 2010;46:92–5. 10.1016/j.oraloncology.2009.10.002. [DOI] [PubMed] [Google Scholar]
- [36].Patel V, Stern C, Miglani A, Weichman K, Lin J, Ow T, et al. Evaluation of the Relationship between Age and Outcome after Microvascular Reconstruction among Patients with Recurrent Head and Neck Squamous Cell Carcinoma. J Reconstr Microsurg 2017;33:336–42. 10.1055/s-0037-1598621. [DOI] [PubMed] [Google Scholar]
- [37].Kowalski LP, Alcantara PSM, Magrin J, Parise O. A case-control study on complications and survival in elderly patients undergoing major head and neck surgery. Am J Surg 1994;168:485–90. 10.1016/S0002-9610(05)80107-2. [DOI] [PubMed] [Google Scholar]
- [38].Clayman GL, Eicher SA, Sicard MW, Razmpa E, Goepfert H. Surgical outcomes in head and neck cancer patients 80 years of age and older. Head Neck 1998;20:216–23. . [DOI] [PubMed] [Google Scholar]
- [39].Chaudhary H, Stewart CM, Webster K, Herbert RJ, Frick KD, Eisele DW, et al. Readmission following primary surgery for larynx and oropharynx cancer in the elderly: Readmission After Surgery in the Elderly. The Laryngoscope 2017;127:631–41. 10.1002/lary.26311. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.