ABSTRACT
Objective
This study aimed to explore the prognostic relevance of the age‐adjusted Charlson Comorbidity Index (ACCI) in laryngeal squamous cell carcinoma (LSCC) survival.
Methods
This retrospective analysis was conducted on the medical records of 367 LSCC patients admitted to the First Hospital of Shanxi Medical University, China, from January 2012 to December 2015. The cohort (n = 367) comprised 351 males and 16 females aged 40–88 years. ACCI scores were categorized into three groups: high (7–12), moderate (5–6), and low (2–4). Prognostic factors were analyzed using the multivariable Cox proportional hazards regression model, and dynamic prognostic nomograms and calibration curves were generated based on the results. Model performance was evaluated using the C‐index.
Results
In this cohort of 367 LSCC patients, the 5‐year overall survival (OS) rate was 58.6%. Meanwhile, the 5‐year progression‐free survival (PFS) rate was 64.3%, and the cancer‐specific survival (CSS) rate was 61.9%. Cox regression analysis identified ACCI, T stage, distant metastasis, and degree of differentiation as independent risk factors for both OS and CSS (p < 0.05). Additionally, a dynamic prognostic nomogram was established and validated, with a C‐index of 0.75 (95% CI: 0.711–0.788) and 0.753 (95% CI: 0.714–0.792) for OS and CSS, respectively. High ACCI emerged as a pivotal survival prognosticator for LSCC patients. Thus, a more convenient and precise dynamic predictive model for formulating treatment strategies can be formulated by incorporating ACCI, T stage, metastasis, and differentiation.
Conclusions
Overall, this study advances our understanding of ACCI's prognostic significance and its role in guiding LSCC management to improve clinical outcomes.
Keywords: age‐adjusted Charlson Comorbidity Index, laryngeal squamous cell carcinoma, long‐term prognosis, nomograms
1. Introduction
Squamous cell carcinoma (SCC) of the head and neck is the sixth most prevalent cancer worldwide, with 25% of cases attributed to laryngeal squamous cell carcinoma (LSCC) [1]. Epidemiological studies report that approximately 180,000 new cases and 100,000 deaths were due to laryngeal cancer in 2020 [2]. In recent years, the incidence of laryngeal cancer has decreased. However, the 5‐year survival rate of LSSC patients remains suboptimal despite recently declining to 40% in advanced‐stage LLSC patients [3]. Traditionally, the treatment approach and prognosis for LSCC patients are primarily determined based on the traditional TNM staging system. Tumor size, lymph node metastasis, and the presence of distant metastasis have been identified as significant factors affecting the prognosis of LSCC [4]. However, with the growing aging population, the number of elderly cancer patients has significantly increased, which may affect the survival rate and quality of life of LSCC patients [5]. Moreover, elderly cancer patients are more likely to have comorbidities such as chronic obstructive pulmonary disease, cardiovascular disease, and diabetes, which impact the prognosis of cancer and influence the choice of treatment strategies [6]. Therefore, a thorough investigation into the various factors affecting the prognosis of LSCC can not only supplement and enhance the traditional TNM staging system but also provide crucial evidence for personalized selection of treatment planning for patients with LSCC.
In 1994, Charlson et al. established the Charlson Comorbidity Index (CCI), a scoring system initially pioneered to predict the mortality of patients seeking treatment at internal emergency departments by incorporating 19 medical comorbidities. This index has also been shown to possess prognostic value for cancer patients, such as those with glioblastoma, sarcoma, lung cancer, renal cell carcinoma, and ovarian cancer [7]. Given that age has been identified as an independent prognostic factor for survival in numerous oncologic diseases, it was incorporated into the CCI. The age‐adjusted Charlson Comorbidity Index (ACCI) is an analysis and quantification of age and comorbidities based on the number and severity of comorbidities to predict mortality risk [8]. Notably, its predictive value has been validated in various cancers, including ovarian, gastric, esophageal, pancreatic, and head and neck malignancies [6, 9, 10, 11, 12, 13]. Nonetheless, research on the predictive value of ACCI in SSC patients is scarce, particularly regarding survival rate. Additionally, there is no standardized cut‐off point for ACCI in this context.
Thus, this study aimed to explore the predictive utility of ACCI for long‐term prognosis in patients with LSCC following treatment to provide a scientific basis and practical reference for clinical decision‐making.
2. Materials and Methods
2.1. Study Design and Participants
This retrospective cohort study was conducted on 367 LSCC patients admitted to the First Hospital of Shanxi Medical University, China, from January 2012 to December 2017. The inclusion criteria were as follows: (1) primary diagnosis of LSCC; (2) complete medical records, including results of hematological tests such as comprehensive blood biochemistry, blood cell analysis, and coagulation function within 1 week before treatment. Additionally, the results of other examinations such as electrocardiogram (ECG), pulmonary ventilation function test, echocardiography of the heart, cervical ultrasound, and CT scan of the head, neck, and chest were available; (3) regular follow‐up visits after treatment, with a frequency of every 3–6 months in the first 2 years and then every 6–12 months thereafter. The follow‐up protocol included routine blood tests, cervical ultrasound, and CT scan of the neck and chest; and (4) patients or their families were fully informed and signed the informed consent form. The exclusion criteria were as follows: (1) incomplete follow‐up data or absence of a clear death date and cause of death and (2) concurrent or prior malignancies, excluding hematological malignancies. This study was approved by the Ethics Committee of Shanxi Medical University First Hospital, China (Approval number: K‐K011).
2.2. Data Collection
Demographic characteristics of patients, such as age, gender, smoking status, marital status, and so forth, were collected through a survey questionnaire. Clinical data such as disease site, pathological type, pathological differentiation degree, TNM staging, distant metastasis, treatment plan, comorbidities, and so forth, were extracted from electronic medical records. The tumor stage was assessed by pathologists based on the criteria outlined in the American Joint Committee on Cancer (AJCC) TNM Staging Manual. Following discharge, patients were followed up every 3 months via phone or outpatient visits during the first 2 years. From the third year onwards, follow‐up was performed every 6 months via phone or outpatient visits. The follow‐up period concluded on December 31, 2022, or upon patient death.
2.3. Surgical Procedures and Other Treatments
All treatment plans were based on the National Comprehensive Cancer Network (NCCN) guidelines. Surgery or radiotherapy alone is currently the first‐line option for patients with early‐stage laryngeal cancer. Surgery with concurrent chemoradiotherapy was the primary approach for patients with locally advanced laryngeal cancer. The selection of surgical modalities relied on factors such as the degree and location of the primary tumor, lymph node involvement, and patient overall health condition. Adjuvant radiotherapy was considered based on risk factors such as positive surgical margins, T4 stage tumors, two or more lymph node metastases, and extracapsular extension, and generally performed within 4–6 weeks post‐surgery. The dose for the primary tumor bed, neck with positive lymph nodes, and prophylactic irradiation area of the neck ranged from 60 to 66 Gy (2 Gy per fraction), 60 to 66 Gy (2 Gy per fraction), and 44 to 64 Gy (1.6 to 2 Gy per fraction), respectively. Various dosage regimens of radiotherapy were formulated based on disease stage and lymph node involvement. For early laryngeal cancer (T1N0M0) patients, the dose of radical radiotherapy was between 63 and 66 Gy (2.25–2.36 Gy per session). Patients with intermediate to advanced stages were administered a dose of 70 Gy over 35 sessions. Lastly, the dose for the neck area with positive lymph nodes and prophylactic irradiation of the neck lymph node was 60–66 Gy (2 Gy per session) and 44–64 Gy (1.6–2 Gy per session), respectively.
2.4. Definition of Comorbidities
Patient comorbidities were determined based on the discharge diagnosis, long‐term medical history, medical records, and reimbursement for specific medications within the 12 months before patient enrollment. The ACCI score was calculated as follows: myocardial infarction, mild liver disease, dementia, congestive heart failure, peripheral vascular disease, cerebrovascular disease, rheumatic disease, chronic pulmonary disease, peptic ulcer disease, and diabetes mellitus (DM) without end‐organ damage were each assigned 1 point. Conversely, comorbidities such as hemiplegia, renal disease, DM with end‐organ damage, malignancy, leukemia, and lymphoma were each assigned 2 points. Moderate to severe liver disease was assigned 3 points, while metastatic solid tumors or AIDS were assigned 6 points. Age was also considered in the ACCI scoring system as follows: < 50 years, Score 0; 50–59 years, Score 1; 60–69 years, Score 2; 70–79 years, Score 3 points; ≥ 80 years, Score 4 points [14]. Higher ACCI scores reflected more severe comorbidities [15].
2.5. Survival Outcomes
The follow‐up period was calculated from the date of surgery until the end of follow‐up or the occurrence of death, whichever occurred first. Overall survival (OS) was defined as the time from the initiation of treatment to the last follow‐up or death. Cancer‐specific survival (CSS) was defined as the time from the initiation of treatment to tumor progression, death, or the last follow‐up. Progression‐free survival (PFS) was defined as the time from treatment initiation to tumor progression, death, or the last follow‐up.
2.6. Statistical Analysis
The optimal cut‐off value of ACCI was determined using the X‐tile software (version 3.6.1, Yale University, USA). Patients were divided into three groups: low ACCI group (scores 2–4), medium ACCI group (scores 5–6), and high ACCI group (scores 7–12). Categorical variables were expressed as frequencies and percentages and compared using the chi‐square test. Non‐normally distributed data were presented as the median (inter‐quartile range, IQR) and compared using the Mann–Whitney U test. Survival analysis was performed using the Kaplan–Meier method. OS and CSS across the three groups were compared using the log‐rank test. Univariate and multivariate analyses were performed to identify independent prognostic factors using the Cox proportional hazards regression model. According to the Cox proportional hazards regression model, a nomogram was established. The performance of the model was evaluated using the C‐index. Statistical analyses were conducted using SPSS 20.0 software, and a p value of < 0.05 was considered statistically significant. Risk scoring, as well as nomograms and calibration curves, were generated using R software (version 3.4.3).
3. Results
3.1. Comorbidity Analysis
This study included the retrospective data of 367 LSCC patients. Among them, 210 patients (57.2%) had no comorbidities, 119 (32.4%) had one comorbidity, 33 (9.0%) had two comorbidities, and 5 (1.4%) had three comorbidities. The distribution of comorbidities is listed in Table 1.
Table 1.
Distribution of comorbidities based on age‐adjusted Charlson Comorbidity Index (n = 367).
| Comorbidity | Source | n (%) |
|---|---|---|
| Myocardial infarction | 1 | 7 (1.91) |
| Congestive heart failure | 1 | 1 (0.27) |
| Peripheral vascular disease | 1 | 1 (0.27) |
| Cerebrovascular disease | 1 | 71 (19.35) |
| Dementia | 1 | — |
| Chronic pulmonary disease | 1 | 42 (11.44) |
| Rheumatic disease | 1 | — |
| Peptic ulcer disease | 1 | 5 (1.36) |
| Mild liver disease | 1 | 57 (15.53) |
| Diabetes mellitus without end‐organ damage | 1 | 13 (3.54) |
| Hemiplegia | 2 | — |
| Renal disease | 2 | — |
| Diabetes mellitus with end‐organ damage | 2 | 1 (0.27) |
| Any malignancya | 2 | — |
| Lymphoma | 2 | — |
| Leukemia | 2 | — |
| moderate to severe Liver disease | 3 | 2 (0.54) |
| Metastatic solid tumora | 6 | — |
| Acquired immunodeficiency syndrome (AIDS) | 6 | — |
Abbreviation: —, no data.
Excluding laryngeal squamous cell carcinoma.
3.2. Demographic and Clinical Baseline Characteristics
Among the 367 LSCC patients, 351 were males and 16 were females, and their ages ranged from 40 to 88 years old, with a median age of 62 years (IQR = 57–68 years). The ACCI scores ranged from 2 to 12, with a median score of 5. Meanwhile, lymph node metastasis rates were 12.4% (26/210 cases) for glottis laryngeal cancer, 40.2% (53/132 patients) for supraglottis cancer, and 24.0% (6/25 cases) for subglottis and transglottis cancers. Follow‐up time ranged from 1 to 115 months, with a median of 60 months, concluding either with the patient's death or the last follow‐up date on December 31, 2022. Over the duration of the study, 152 deaths were recorded, among which 140 deaths were ascribed to laryngeal cancer.
In comparison, 12 deaths were attributed to nontumor‐related causes (3 deaths from coronary heart disease, 3 from pulmonary infection, 2 from cerebral infarction, 2 from cerebral hemorrhage, 1 from liver cancer, and 1 from a car accident). The 5‐year OS and PFS rates were 58.6% and 64.3%, respectively, while the CCS rate was 61.9%. At the same time, the median OS rate was 60.0 months (IQR = 27–61 months), while median CSS and PFS were 60.0 months (IQR = 26‐61 months, IQR = 25–61 months). Comparative analysis of the three ACCI groups revealed significant differences (p < 0.05) in age, site of disease, T stage, N stage, degree of differentiation, distant metastasis, treatment approach, comorbidities, smoking status, 5‐year survival outcomes, and 5‐year recurrence rates. However, gender and marital status were comparable across the three groups. The demographic and clinical characteristics of all patients are summarized in Table 2.
Table 2.
Comparison of clinical variables among different ACCI groups in patients with LSCC (n = 367).
| ACCI score | ||||||
|---|---|---|---|---|---|---|
| Variables | Number of patients (%) | 2–4 (n = 90) | 5–6 (n = 120) | 7–12 (n = 157) | χ 2 (Z) | p value |
| Gender | 3.354 | 0.188 | ||||
| Male | 351 (95.6) | 83 (92.2) | 116 (96.7) | 152 (96.8) | ||
| Female | 16 (4.4) | 7 (7.8) | 4 (3.3) | 5 (3.2) | ||
| Age, yearsa | 62.0 (57.0–68.0) | 65.0 (60.0–72.0) | 66.5.0 (61.0–72.0) | 59.0 (53.0–63.0) | 83.953 | < 0.001 |
| Tumor localization | 29.623 | < 0.001 | ||||
| Supraglottis | 132 (36) | 53 (58.9) | 38 (31.7) | 41 (26.1) | ||
| Glottis | 210 (57.2) | 32 (35.6) | 76 (63.3) | 102 (65.0) | ||
| Subglottis and transglottis | 25 (6.8) | 5 (5.6%) | 6 (5.0) | 14 (8.9) | ||
| T Stage | 19.546 | 0.003 | ||||
| T1 | 65 (17.7) | 8 (8.9) | 23 (19.2) | 34 (21.7) | ||
| T2 | 140 (38.1) | 28 (31.1%) | 43 (35.8) | 69 (43.9) | ||
| T3 | 119 (32.4) | 36 (40.0) | 43 (35.8) | 40 (25.5) | ||
| T4 | 43 (11.7) | 18 (20.0) | 11 (9.2) | 14 (8.9) | ||
| N Stage | 253.216 | < 0.001 | ||||
| N0 | 284 (77.4) | 15 (16.7) | 112 (93.3) | 157 (100%) | ||
| N1 | 32 (8.7) | 28 (31.1) | 4 (3.3) | 0 | ||
| N2 | 51 (13.9) | 47 (52.2) | 4 (3.3) | 0 | ||
| Degree of differentiation | 19.449 | 0.001 | ||||
| High | 66 (18) | 9 (10.0) | 26 (21.7) | 31 (19.7) | ||
| Medium | 220 (59.9) | 49 (54.4) | 66 (55.0) | 105 (66.9) | ||
| Poor | 81 (22.1) | 32 (35.6) | 28 (23.3) | 21 (13.4) | ||
| Distant metastasis | 51.489 | < 0.001 | ||||
| Yes | 16 (4.4) | 16 (17.8) | 0 | 0 | ||
| No | 351 (95.6) | 74 (82.2) | 120 (100) | 157 (100) | ||
| Treatment | 25.738 | < 0.001 | ||||
| Operation | 106 (28.9) | 15 (16.7) | 39 (32.5) | 52 (33.1) | ||
| Combined treatment | 221 (60.2) | 62 (68.9) | 59 (49.2) | 100 (63.7) | ||
| Radiotherapy | 40 (10.9) | 13 (14.4) | 22 (18.3) | 5 (3.2) | ||
| Comorbidity | 250.881 | < 0.001 | ||||
| No | 210 (57.2) | 50 (55.6) | 26 (21.5) | 134 (85.4) | ||
| One | 119 (32.4) | 24 (26.7) | 73 (60.8) | 22 (14.0) | ||
| Two | 33 (9.0) | 12 (13.3) | 21 (17.5) | 0 | ||
| Three | 5 (1.4) | 4 (4.4) | 0 | 1 (0.6) | ||
| Smoking | 80.354 | < 0.001 | ||||
| Yes | 270 (73.6) | 73 (81.1) | 86 (71.7) | 46 (29.3) | ||
| No | 97 (26.4) | 17 (18.9) | 34 (28.3) | 111 (70.7) | ||
| Marital status | 7.815 | 0.252 | ||||
| Married | 350 (95.4) | 84 (93.3) | 114 (95.0) | 152 (96.8) | ||
| Unmarried | 3 (0.8) | 0 | 2 (1.7) | 1 (0.6) | ||
| Divorce | 8 (2.2) | 2 (2.2) | 3 (2.5) | 3 (1.9) | ||
| Bereft of one's spouse | 6 (1.6) | 4 (4.4) | 1 (0.8) | 1 (0.6) | ||
| 5‐year survival status | 82.995 | < 0.001 | ||||
| Cancer death | 140 (38.1) | 68 (75.6) | 42 (35.0) | 30 (19.1) | ||
| Non‐cancer deaths | 12 (3.3) | 3 (3.3) | 6 (5.0) | 3 (1.9) | ||
| Survival | 215 (58.6) | 19 (21.1) | 72 (60.0) | 124 (79) | ||
| 5‐year recurrence rate | 49.833 | < 0.001 | ||||
| Yes | 131 (35.7) | 58 (64.4) | 42 (35.0) | 31 (19.7) | ||
| No | 236 (64.3) | 32 (35.6) | 78 (65.0) | 126 (80.3) | ||
Abbreviations: ACCI, age‐adjusted Charlson Comorbidity Index; LSCC, laryngeal squamous cell carcinoma.
The data are represented as median (IQR).
3.3. Survival Analysis
The 5‐year OS rates for the high, moderate, and low ACCI groups were 21.1%, 60%, and 79%, respectively. Interestingly, the low ACCI group had a higher OS rate compared to the moderate (χ 2 = 11.863, p = 0.001) and high (χ 2 = 101.402, p < 0.001) ACCI groups. In contrast, the moderate ACCI group had a higher OS rate compared to the high ACCI group (χ 2 = 43.620, p < 0.001). The 5‐year CCS rates for the high, moderate, and low ACCI groups were 24.4%, 65%, and 80.9%, respectively. Furthermore, group comparison revealed that the low ACCI group had a higher CSS rate than the moderate (χ 2 = 8.266, p = 0.004) and high (χ 2 = 94.851, p < 0.001) ACCI groups. Likewise, the moderate ACCI group also had a superior CSS rate compared to the high ACCI group (χ 2 = 45.474, p < 0.001). The Kaplan–Meier curves for 5‐year OS and CSS are illustrated in Figure 1.
Figure 1.
Kaplan–Meier curves for overall survival and cancer‐specific survival in different ACCI groups. ACCI, age‐adjusted Charlson Comorbidity Index.
3.4. Univariate and Multivariate Analysis of OS and CSS
On the one hand, univariate Cox proportional hazards regression (UCPHR) model analysis identified age, T stage, N stage, degree of differentiation, distant metastasis, comorbidity, smoking status, and ACCI as significant factors influencing the OS of LSCC patients. On the other hand, multivariate Cox proportional hazards regression (MCPHR) model analysis revealed no statistically significant difference in OS between the low and moderate ACCI groups. However, patients in the high ACCI group had a significantly higher mortality risk (HR = 3.848, 95% CI: 1.778–8.327) compared to those in the low ACCI group. As anticipated, patients with distant metastasis had a threefold higher risk of death (HR = 2.714, 95% CI: 1.419–5.190) compared to those without distant metastasis. Besides, no statistically significant difference was noted in OS between patients with T2 and T1 stages. However, patients with T3 (HR = 1.816, 95% CI: 1.006–3.279) and T4 (HR = 2.019, 95% CI: 1.024–3.982) stages were at a significantly higher risk of mortality compared to T1 stage patients. Similarly, a poorer degree of differentiation was associated with a higher risk of death (HR = 1.390, 95% CI: 1.050–1.840) (Table 3).
Table 3.
Cox proportional hazards regression model analysis of 5‐year OS for LSCC (n = 367).
| Univariate analysis | Multivariate analysisa | |||
|---|---|---|---|---|
| Variables | HR (95% CI) | p value | HR (95% CI) | p value |
| Gender | 1.187 (0.582–2.420) | 0.637 | ||
| Age | 1.047 (1.027–1.068) | < 0.001 | 1.011 (0.988–1.034) | 0.351 |
| Smoking | 1.638 (1.092–2.457) | 0.017 | 1.198 (0.784–1.830) | 0.403 |
| Marital | 1.198 (0.907–1.581) | 0.203 | ||
| Tumor localization | 0.825 (0.615–1.093) | 0.176 | ||
| T stage | < 0.001 | 0.049 | ||
| T1 | Reference | Reference | ||
| T2 | 1.533 (0.856–2.746) | 0.151 | 1.221 (0.669–2.227) | 0.515 |
| T3 | 2.867 (1.633–5.033) | < 0.001 | 1.816 (1.006–3.279) | 0.048 |
| T4 | 3.832 (2.036–7.211) | < 0.001 | 2.019 (1.024–3.982) | 0.042 |
| N stage | < 0.001 | 0.927 | ||
| N0 | Reference | Reference | ||
| N1 | 3.052 (1.912–4.871) | < 0.001 | 0.987 (0.477–2.045) | 0.973 |
| N2 | 4.440 (3.045–6.474) | < 0.001 | 1.094 (0.539–2.221) | 0.803 |
| Degree of differentiation | 1.727 (1.334–2.235) | < 0.001 | 1.390 (1.050–1.840) | 0.021 |
| Distant metastasis | 9.044 (5.251–15.577) | < 0.001 | 2.714 (1.419–5.190) | 0.003 |
| Comorbidity | < 0.001 | 0.347 | ||
| No | Reference | Reference | ||
| One | 1.209 (0.843–1.734) | 0.301 | 1.005 (0.656–1.539) | 0.981 |
| Two | 2.813 (1.773–4.463) | < 0.001 | 1.573 (0.899–2.752) | 0.112 |
| Three | 3.019 (1.102–8.268) | 0.032 | 1.574 (0.491–5.406) | 0.446 |
| ACCI | < 0.001 | 0.003 | ||
| 2–4 | Reference | Reference | ||
| 5–6 | 2.098 (1.346–3.268) | 0.001 | 1.626 (0.926–2.854) | 0.090 |
| 7–12 | 6.869 (4.525–10.426) | < 0.001 | 3.848 (1.778–8.327) | 0.001 |
| Treatment | 1.014 (0.775–1.326) | 0.919 | ||
Abbreviations: ACCI, age‐adjusted Charlson Comorbidity Index; LSCC, laryngeal squamous cell carcinoma; OS, overall survival.
Adjusted for age, T stage, N stage, degree of differentiation, distant metastasis, comorbidities, smoking, and ACCI.
On the one hand, UCPHR model analysis demonstrated that age, T stage, N stage, degree of differentiation, distant metastasis, comorbidity, smoking status, and ACCI were significant factors influencing the CSS of LSCC patients. On the other hand, MCPHR model analysis revealed no statistically significant difference in 5‐year CSS between the low ACCI group and the moderate ACCI group. However, the high ACCI group had a significantly higher mortality risk (HR = 4.276, 95% CI: 1.904–9.603) compared to the low ACCI group. Patients with distant metastasis had a threefold higher risk of death (HR = 2.946, 95% CI: 1.519–5.712) compared to those without distant metastasis. The 5‐year CSS was comparable between patients with T2 and T1 stages. However, patients in the T3 (HR = 2.066, 95% CI: 1.079–3.953) and T4 (HR = 2.339, 95% CI: 1.123–4.869) stages had approximately double the risk of death compared to those in the T1 stage. Lastly, poor differentiation was associated with a higher risk of death (HR = 1.357, 95% CI: 1.012–1.818) (Table 4).
Table 4.
Cox proportional hazards regression model analysis of 5‐year CSS for LSCC (n = 367).
| Univariate analysis | Multivariate analysisa | |||
|---|---|---|---|---|
| Variables | HR (95% CI) | p value | HR (95% CI) | p value |
| Gender | 1.133 (0.530–2.424) | 0.747 | ||
| Age | 1.042 (1.021–1.063) | < 0.001 | 1.005 (0.981–1.028) | 0.702 |
| Smoking | 1.693 (1.105–2.594) | 0.016 | 1.205 (0.771–1.882) | 0.412 |
| Marital | 1.137 (0.837–1.543) | 0.413 | ||
| Tumor localization | 0.812 (0.603–1.095) | 0.172 | ||
| T stage | < 0.001 | 0.035 | ||
| T1 | Reference | Reference | ||
| T2 | 1.749 (0.921–3.322) | 0.088 | 1.389 (0.718–2.690) | 0.329 |
| T3 | 3.290 (1.770–6.118) | < 0.001 | 2.066 (1.079–3.953) | 0.028 |
| T4 | 4.551 (2.294–9.028) | < 0.001 | 2.339 (1.123–4.869) | 0.023 |
| N stage | < 0.001 | 0.791 | ||
| N0 | Reference | Reference | ||
| N1 | 2.845 (1.724–4.694) | < 0.001 | 0.830 (0.380–1.813) | 0.640 |
| N2 | 4.663 (3.170–6.859) | < 0.001 | 1.006 (0.479–2.110) | 0.988 |
| Degree of differentiation | 1.706 (1.304–2.232) | < 0.001 | 1.357 (1.012–1.818) | 0.041 |
| Distant metastasis | 9.618 (5.575–16.593) | < 0.001 | 2.946 (1.519–5.712) | 0.001 |
| Comorbidity | < 0.001 | 0.360 | ||
| No | Reference | Reference | ||
| One | 1.128 (0.771–1.650) | 0.534 | 0.963 (0.616–1.507) | 0.870 |
| Two | 2.800 (1.746–4.491) | < 0.001 | 1.548 (0.865–2.770) | 0.141 |
| Three | 3.144 (1.146–8.619) | 0.026 | 1.580 (0.485–5.146) | 0.448 |
| ACCI | < 0.001 | 0.002 | ||
| 2–4 | Reference | Reference | ||
| 5–6 | 1.927 (1.211–3.066) | 0.006 | 1.612 (0.896–2.899) | 0.111 |
| 7–12 | 6.774 (4.405–10.416) | < 0.001 | 4.276 (1.904–9.603) | < 0.001 |
| Treatment | 0.954 (0.721–1.262) | 0.743 | ||
Abbreviations: ACCI, age‐adjusted Charlson Comorbidity Index; CSS, cancer‐specific survival; LSCC, laryngeal squamous cell carcinoma.
Adjusted for age, T stage, N stage, degree of differentiation, distant metastasis, comorbidities, smoking, and ACCI.
The MCPHR model analysis identified prognostic factors, following which nomograms and calibration curves for OS and CSS were constructed (Figures 2, 3, 4, 5). The C‐index for OS and CSS was 0.75 (95% CI: 0.711–0.788) and 0.753 (95% CI: 0.714–0.792), respectively. Superior predictive ability compared to TNM staging (OS C‐index = 0.754, p = 0.010, 95% CI: 0.022–0.162; CSS C‐index = 0.546, p < 0.001, 95% CI: 0.109–0.271), highlighting the outstanding discriminatory ability of the model. Furthermore, internal validation was performed using 1000 bootstrap resamples. At the same time, 91 cases of LSCC patients were collected as a validation set for external validation, yielding a C‐index of 0.811 (95% CI: 0.720–0.902) and 0.778 (95% CI: 0.679–0.878) for OS and CSS, respectively. In the validation set, the C‐index of ACCI for predicting OS and CSS was also higher than the TNM staging system, OS (0.811 vs. 0.677, p = 0.034, 95% CI: 0.010–0.257), CSS (0.778 vs. 0.656, p = 0.047, 95% CI: 0.001–0.243). All calibration curves for 3‐ and 5‐year OS and CSS were close to the diagonal line. This indicates that the ACCI predictive system is more capable of predicting long‐term survival rates for LSCC than the TNM staging system.
Figure 2.
Nomogram of overall survival (OS).
Figure 3.
Calibration curve of 3‐ and 5‐year overall survival (OS).
Figure 4.
Nomogram of cancer‐specific survival (CSS).
Figure 5.
Calibration curve of 3‐ and 5‐year cancer‐specific survival (CSS).
4. Discussion
This retrospective cohort study investigated the predictive ability of ACCI in LSCC patients. The results revealed that higher ACCI scores, advanced T stage, advanced M stage, and poorer degree of differentiation were significant risk factors affecting the long‐term prognosis of LSCC patients.
With the growing aging population, the proportion of elderly cancer patients has significantly increased. Notably, age has been established as an independent risk factor affecting the survival and prognosis of malignant tumors [16, 17, 18]. Meanwhile, elderly patients are more likely to have multiple comorbidities due to poor physical function, increasing mortality risk, and decreasing quality of life [19, 20]. The ACCI scoring system combines age and comorbidities to comprehensively assess a patient's health and predict disease survival outcomes [14]. Maezawa et al. [10] enrolled 2254 gastric cancer patients who underwent curative gastrectomy and noted that ACCI was an independent risk factor affecting long‐term prognosis and positively correlated with postoperative complication rates. Koppie et al. [14] analyzed the prognostic impact of ACCI on OS in 1121 bladder cancer patients after surgery and reported that higher ACCI scores were associated with worse OS. Compared to patients with lower ACCI scores (< 2), those with moderate scores (3–5) had a hazard ratio of 1.46 (95% CI: 1.20–1.78), while patients with high ACCI scores (> 5) had a hazard ratio of 2.66 (95% CI: 2.00–3.55), consistent with our results. This finding could be ascribed to high ACCI scores being linked to severe postoperative complications [21, 22], such as a systemic inflammatory state [23, 24], which can impact the host's resistance to tumors [10]. Numerous studies reported that postoperative complications, systemic inflammatory conditions, and postoperative immune suppression are associated with poor cancer prognosis [25].
This study determined the optimal cut‐off point for ACCI using the X‐tile program. Patients were divided into the high, moderate, and low ACCI groups. Data analysis exposed that higher ACCI scores were associated with advanced TNM stages, poorer differentiation, higher smoking rates, and a higher rate of comorbidities (p < 0.05), in line with the findings of previous studies. Song et al. [12] identified no significant differences in pathological differentiation and TNM stages between the high (≥ 5) and low (< 5) ACCI groups in postoperative patients with esophageal cancer. This discrepancy can be attributed to differences in disease types and cut‐off values. Mulcahy et al. [26] and Fong et al. [27] reported that the optimal cut‐off value for ACCI in LSCC patients was 3, discrepant from our results. This discrepancy may be attributed to different software used in these studies. For instance, Mulcahy et al. used the Bayesian Information Criterion (BIC) method to determine the cut‐off value of ACCI, while Fong et al. did not specify the technique applied. Herein, X‐tile software, a visualization tool developed by Yale University, USA, was used. Noteworthily, this software has been widely applied in predictive analysis and retrospective studies of various cancers [28]. Determining the optimal cut‐off value is necessary to accurately predict long‐term prognosis in LSCC patients.
In addition to the impact of ACCI on the 5‐year OS and CSS of LSCC patients, higher T stage, M stage, and poorer differentiation degree were associated with poor long‐term prognosis in LSCC patients. Gallo et al. [29] analyzed the data of 327 LSCC patients who underwent total laryngectomy and identified the degree of differentiation and local recurrence as prognostic factors influencing OS. Riele et al. [30] examined the data of 788 LSCC patients and pointed out that long‐term mortality rates were almost double for T2 stage patients, about four times for T3 patients (HR = 3.89, 95% CI: 2.41–6.28), and approximately five times higher for T4 patients (HR = 4.95, 95% CI: 2.93–8.36) compared to T1 stage patients (HR = 1.93, 95% CI: 1.15–2.97). Chen et al. [31] reported that the long‐term mortality rate for LSCC patients with distant metastasis was 2.6 times higher than those without distant metastasis (HR = 2.61, 95% CI: 2.08–3.28), in agreement with our results. Therefore, T stage, M stage, and degree of differentiation play an essential role in guiding treatment strategies for LSCC patients.
The TNM staging system, one of the most commonly used staging systems for cancer patients, has been recognized as a traditional prognostic factor in various cancers [32]. However, the system mainly focuses on tumor and does not include personal factors such as age and complications; thus, it cannot make individualized predictions for patients. Chen et al. [31] examined the data of 4237 LSCC patients and identified TNM stage and degree of differentiation as OS prognostic factors. In the present study, based on the results of the MCPHR model analysis, a dynamic prognostic nomogram was developed by integrating age, marital status, T stage, N stage, M stage, tumor size, and degree of differentiation. The C‐index of the nomogram was 0.69 (95% CI: 0.67–0.70), indicating its superior predictive performance over the TNM staging system (C‐index 0.62, 95% CI: 0.60–0.63). Cui et al. [33] conducted an MCPHR model analysis for 5‐year OS in 369 LSCC patients and identified six independent prognostic parameters, including the lymph node ratio (LNR), pack‐years, age, N‐stage, anemia, and albumin, which were incorporated into the nomogram. The C‐index of their nomogram was 0.73 (95% CI: 0.68–0.78), indicating its superior predictive performance over the TNM staging system (C‐index 0.59, 95% CI: 0.54–0.65), consistent with our results. In this study, Cox multivariate analysis was carried out to establish OS and CSS and 3‐ and 5‐year calibration curves by combining ACCI, T stage, M stage, and pathological differentiation. Each calibration curve exhibited strong concordance with actual OS and CSS outcomes. The C‐index for OS and CSS was 0.75 (95% CI: 0.711–0.788) and 0.753 (95% CI: 0.714–0.792). All are higher than the TNM staging system, the C‐index for OS and CSS was 0.654 and 0.546. The same results are also present in the validation set. Therefore, the dynamic predictive model composed of ACCI, T stage, M stage, and degree of differentiation outperformed other models in terms of predictive accuracy [31, 33].
Nonetheless, our study had some limitations that cannot be overlooked. To begin, the lack of comprehensive documentation of posttreatment complications in patients precluded further statistical analysis. Second, data were sourced from a single hospital, which may inherently introduce selection bias. Third, the sample size was relatively small, highlighting the need for larger datasets, extended follow‐up durations, and more comprehensive patient data in further studies.
5. Conclusions
In summary, elderly cancer patients are susceptible to comorbidities that can affect tumor diagnosis, treatment, and prognosis. Therefore, a multidisciplinary treatment approach should be considered based on the patient's age and comorbidities. In this study, an efficient, clinically applicable dynamic prognostic assessment system was established to guide the selection of targeted diagnostic and therapeutic strategies for LSCC patients to prolong survival.
Author Contributions
Ze‐Hui Gao was responsible for the experimental design, experimental operation, statistical analysis, and manuscript writing. Rui‐Han was responsible for data analysis, result interpretation, and article writing. Su‐Zhen She and Li‐Na Jia were responsible for data organization, formatting, and article writing. Hui Huang‐Fu and Huan‐Hai Liu were responsible for article proofreading. Professor Liu provided valuable feedback on the revisions of this article and is acknowledged as a co‐corresponding author.
Ethics Statement
The study was conducted according to the guidelines of the Declaration of Helsinki and approved by the Institutional Review Board Ethics Committee of Shanxi Medical University First Hospital, China (Approval number: K‐K011).
Conflicts of Interest
The authors declare no conflicts of interest.
Acknowledgments
We acknowledge the assistance received from the First Hospital of Shanxi Medical University. Also, we want to thank all of the organizations and groups that helped make the research possible, as well as the patients who took part in the study.
Contributor Information
Hui Huang‐Fu, Email: 13934518228@163.com.
Huan‐Hai Liu, Email: liuhuanhaiok@smmu.edu.cn.
Data Availability Statement
The data are available from the corresponding author upon reasonable request.
References
- 1. Johnson D. E., Burtness B., Leemans C. R., Lui V. W. Y., Bauman J. E., and Grandis J. R., “Head and Carcinoma,” Nature Reviews Disease Primers 6, no. 1 (November 2020): 92, 10.1038/s41572-020-00224-3. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Sung H., Ferlay J., Siegel R. L., et al., “Global Cancer Statistics 2020: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: A Cancer Journal for Clinicians 71, no. 3 (May 2021): 209–249, 10.3322/caac.21660. [DOI] [PubMed] [Google Scholar]
- 3. Obid R., Redlich M., and Tomeh C., “The Treatment of Laryngeal Cancer,” Oral and Maxillofacial Surgery Clinics of North America 31, no. 1 (2019): 1–11, 10.1016/j.coms.2018.09.001. [DOI] [PubMed] [Google Scholar]
- 4. Kim D. H., Kim S. W., Han J. S., Kim G. J., Basurrah M. A., and Hwang S. H., “The Prognostic Utilities of Various Risk Factors for Laryngeal Squamous Cell Carcinoma: A Systematic Review and Meta‐Analysis,” Medicina 59, no. 3 (March 2023): 497, 10.3390/medicina59030497. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Hurria A., Balducci L., Naeim A., et al., “Mentoring Junior Faculty in Geriatric Oncology: Report From the Cancer and Aging Research Group,” Journal of Clinical Oncology 26, no. 19 (July 2008): 3125–3127, 10.1200/JCO.2008.16.9771. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Kahl A., du Bois A., Harter P., et al., “Prognostic Value of the Age‐Adjusted Charlson Comorbidity Index (ACCI) on Short‐ and Long‐Term Outcome in Patients With Advanced Primary Epithelial Ovarian Cancer,” Annals of Surgical Oncology 24, no. 12 (November 2017): 3692–3699, 10.1245/s10434-017-6079-9. [DOI] [PubMed] [Google Scholar]
- 7. Clements N. A., Gaskins J. T., and Martin 2nd R. C. G., “Predictive Ability of Comorbidity Indices for Surgical Morbidity and Mortality: A Systematic Review and Meta‐Analysis,” Journal of Gastrointestinal Surgery 27, no. 9 (September 2023): 1971–1987, 10.1007/s11605-023-05743-4. [DOI] [PubMed] [Google Scholar]
- 8. Charlson M., Szatrowski T. P., Peterson J., and Gold J., “Validation of a Combined Comorbidity Index,” Journal of Clinical Epidemiology 47, no. 11 (November 1994): 1245–1251, 10.1016/0895-4356(94)90129-5. [DOI] [PubMed] [Google Scholar]
- 9. Shinkawa H., Tanaka S., Takemura S., et al., “Predictive Value of the Age‐Adjusted Charlson Comorbidity Index for Outcomes After Hepatic Resection of Hepatocellular Carcinoma,” World Journal of Surgery 44, no. 11 (November 2020): 3901–3914, 10.1007/s00268-020-05686-w. [DOI] [PubMed] [Google Scholar]
- 10. Maezawa Y., Aoyama T., Kano K., et al., “Impact of the Age‐Adjusted Charlson Comorbidity Index on the Short‐ and Long‐Term Outcomes of Patients Undergoing Curative Gastrectomy for Gastric Cancer,” Journal of Cancer 10, no. 22 (August 2019): 5527–5535, 10.7150/jca.35465. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Takemura K., Takenaka Y., Ashida N., et al., “Age‐Adjusted Charlson Comorbidity Index Predicts Prognosis of Laryngopharyngeal Cancer Treated With Radiation Therapy,” Acta Oto‐laryngologica 137, no. 12 (December 2017): 1307–1312, 10.1080/00016489.2017.1362112. [DOI] [PubMed] [Google Scholar]
- 12. Song J., Lin Y., Zhang J., et al., “Effect of the Age‐adjusted Charlson Comorbidity Index on the Survival of Esophageal Squamous Cell Carcinoma Patients After Radical Esophagectomy,” Journal of Clinical Medicine 11, no. 22 (November 2022): 6737, 10.3390/jcm11226737. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Cheng H., Xu J. H., He J. Q., Yang X. Y., Shen X. N., and Xu X. L., “Multivariate Analysis of Prognostic Factors in Patients With Lip Squamous Cell Carcinoma After Surgery,” World journal of Surgical Oncology 22, no. 1 (January 2024): 35, 10.1186/s12957-024-03313-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Koppie T. M., Serio A. M., Vickers A. J., et al., “Age‐Adjusted Charlson Comorbidity Score is Associated With Treatment Decisions and Clinical Outcomes for Patients Undergoing Radical Cystectomy for Bladder,” Cancer 112, no. 11 (June 2008): 2384–2392, 10.1002/cncr.23462. [DOI] [PubMed] [Google Scholar]
- 15. Bannay A., Chaignot C., Blotière P. O., et al., “The Best Use of the Charlson Comorbidity Index With Electronic Health Care Database to Predict Mortality,” Medical Care 54, no. 2 (February 2016): 188–194, 10.1097/MLR.0000000000000471. [DOI] [PubMed] [Google Scholar]
- 16. Albert A., Lee A., Allbright R., and Vijayakumar S., “Impact of Age on Receipt of Curative Treatment for Cervical Cancer: An Analysis of Patterns of Care and Survival in a Large, National Cohort,” Journal of Geriatric Oncology 10, no. 3 (May 2019): 465–474, 10.1016/j.jgo.2018.10.005. [DOI] [PubMed] [Google Scholar]
- 17. Liu K., Wang P., Zhu X., Bei Y., Zheng Z., and Yan S., “Disparities of Age‐Based Cancer‐Specific Survival Improvement With Various Clinicopathologic Characteristics for Kidney Cancer,” Cancer Management and Research 10 (July 2018): 2259–2268, 10.2147/CMAR.S169192. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Zhou A., Zhang D., Kang X., and Brooks J. D., “Identification of Age‐ and Immune‐Related Gene Signatures for Clinical Outcome Prediction in Lung Adenocarcinoma,” Cancer Medicine 12, no. 16 (August 2023): 17475–17490, 10.1002/cam4.6330. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Extermann M., Overcash J., Lyman G. H., Parr J., and Balducci L., “Comorbidity and Functional Status Are Independent in Older Cancer Patients,” Journal of Clinical Oncology 16, no. 4 (April 1998): 1582–1587, 10.1200/JCO.1998.16.4.1582. [DOI] [PubMed] [Google Scholar]
- 20. Seriu N., Tsukamoto S., Ishida Y., et al., “Influences of Comorbidities on Perioperative Rehabilitation in Patients With Gastrointestinal Cancers: A Retrospective Study,” World Journal of Surgical Oncology 21, no. 1 (October 2023): 336, 10.1186/s12957-023-03207-2. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Tian Y., Xu B., Yu G., Li Y., and Liu H., “Age‐Adjusted Charlson Comorbidity Index Score as Predictor of Prolonged Postoperative Ileus in Patients With Colorectal Cancer Who Underwent Surgical Resection,” Oncotarget 8, no. 13 (March 2017): 20794–20801, 10.18632/oncotarget.15285. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Wu J., Tian S., Xu J., et al., “Association of High‐Risk Comorbidity With Overall Survival Among Patients With Gastric Cancer and Its Sex‐Specific Differences in China: A Retrospective Observational Cohort Study,” BMC Cancer 23, no. 1 (September 2023): 916, 10.1186/s12885-023-11374-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Lin J. X., Huang Y. Q., Xie J. W., et al., “Association of the Age‐Adjusted Charlson Comorbidity Index and Systemic Inflammation With Survival in Gastric Cancer Patients After Radical Gastrectomy,” European Journal of Surgical Oncology 45, no. 12 (December 2019): 2465–2472, 10.1016/j.ejso.2019.07.010. [DOI] [PubMed] [Google Scholar]
- 24. Xu X. L., Xu J. H., He J. Q., Li Y. H., and Cheng H., “Novel Prognostic Nomograms for Postoperative Patients With Oral Cavity Squamous Cell Carcinoma in the Central Region of China,” BMC Cancer 24, no. 1 (June 2024): 730, 10.1186/s12885-024-12465-6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Cheng H., Xu J. H., He J. Q., Wu C. C., Li J. F., and Xu X. L., “Nomogram Based on Immune‐Inflammatory Indicators and Age‐Adjusted Charlson Comorbidity Index Score to Predict Prognosis of Postoperative Parotid Gland Carcinoma Patients,” BMC Oral Health 24, no. 1 (June 2024): 718, 10.1186/s12903-024-04490-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Mulcahy C. F., Mohamed A. S. R., Kanwar A., et al., “Multidisciplinary Larynx Cancer Working Group Age‐Adjusted Comorbidity and Survival in Locally Advanced Laryngeal Cancer,” Head & Neck 40, no. 9 (September 2018): 2060–2069, 10.1002/hed.25200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. Fong P. Y., Tan S. H., Lim D. W. T., et al., “Association of Clinical Factors With Survival Outcomes in Laryngeal Squamous Cell Carcinoma LSCC,” PLoS One 14, no. 11 (November 2019): e0224665, 10.1371/journal.pone.0224665. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Zhuang W., Chen J., Li Y., and Liu W., “Valuation of Lymph Node Dissection in Localized High‐Risk Renal Cell Cancer Using X‐Tile Software,” International Urology and Nephrology 52, no. 2 (February 2020): 253–262, 10.1007/s11255-019-02307-x. [DOI] [PubMed] [Google Scholar]
- 29. Gallo O., Sarno A., Baroncelli R., Bruschini L., and Boddi V., “Multivariate Analysis of Prognostic Factors in T3 N0 Laryngeal Carcinoma Treated With Total Laryngectomy,” Otolaryngology–Head and Neck Surgery 128, no. 5 (May 2003): 654–663, 10.1016/S0194-59980300228-6. [DOI] [PubMed] [Google Scholar]
- 30. te Riele R. J. L. M., Dronkers E. A. C., Wieringa M. H., et al., “Influence of Anemia and BMI on Prognosis of Laryngeal Squamous Cell Carcinoma: Development of an Updated Prognostic Model,” Oral Oncology 78 (March 2018): 25–30, 10.1016/j.oraloncology.2018.01.001. [DOI] [PubMed] [Google Scholar]
- 31. Chen L., Lin G., Qian J., et al., “A Dynamic Prognostic Nomogram to Predict the Benefit From Surgical Treatment Modality for Patients With Laryngeal Squamous Cell Carcinoma,” Head & Neck 43, no. 7 (July 2021): 2148–2158, 10.1002/hed.26683. [DOI] [PubMed] [Google Scholar]
- 32. Chen S. B., Huang S. J., Liu D. T., Weng H. R., and Chen Y. P., “Clinicopathological Features and Prognosis of Esophageal Squamous Cell Carcinoma In Young Patients,” Diseases of the Esophagus 32, no. 5 (May 2019): doy070, 10.1093/dote/doy070. [DOI] [PubMed] [Google Scholar]
- 33. Cui J., Wang L., Zhong W., et al., “Development and Validation of Nomogram to Predict Risk of Survival in Patients With Laryngeal Squamous Cell Carcinoma,” Bioscience Reports 40, no. 8 (August 2020): BSR20200228, 10.1042/BSR20200228. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The data are available from the corresponding author upon reasonable request.