ABSTRACT
Given the limited evidence on prognostic factors specifically for older patients with extensive‐stage small cell lung cancer (ES‐SCLC), a population with distinct clinical characteristics, this study aimed to validate whether previously reported prognostic indicators retain their predictive value in this vulnerable group. A retrospective analysis was conducted on data from 270 older ES‐SCLC patients who received treatment at the Fourth Hospital of Hebei Medical University between December 2016 and June 2024. By the final follow‐up date of October 15, 2024, 212 deaths had been recorded. The median progression‐free survival (mPFS) was 6.7 months (95% confidence interval [CI] 6.0–7.4), and the median overall survival (mOS) was 13.1 months (95% CI 11.8–14.4). For PFS, univariate and multivariate Cox analyses identified first‐line chemotherapy (CT) and old‐old (≥ 75 years) as independent adverse prognostic factors. For OS, old‐old, a positive smoking history, bone metastasis, and high‐lactate dehydrogenase (> 250 U/L) were identified as significant adverse prognostic factors. Notably, high‐pro‐gastrin‐releasing peptide (ProGRP) (> 69.2 pg/mL) was significantly associated with an increased risk of death during the follow‐up period beyond 10 months (HR = 1.85, 95% CI 1.05–3.26, p = 0.032); conversely, no significant association was observed within the initial 10 months of follow‐up (HR = 0.84, 95% CI 0.44–1.60, p = 0.604). In conclusion, these findings not only corroborate the prognostic value of previously identified risk factors in older patients with ES‐SCLC but also demonstrate that the prognostic impact of ProGRP is distinctly time‐dependent.
Keywords: extensive‐stage small cell lung cancer, older patients, prognostic risk factors, survival
The study corroborates the prognostic value of previously identified risk factors impacting progression‐free survival (PFS) and overall survival (OS) in older patients with extensive‐stage small cell lung cancer (ES‐SCLC). The figure was created with Microsoft PowerPoint. mPFS, Median progression‐free survival; mOS, Median overall survival; CI, Confidence interval; HR, Hazard ratio; CT, Chemotherapy; LDH, Lactate dehydrogenase; ProGRP, Pro‐gastrin‐releasing peptide.
1. Introduction
Lung cancer is the most commonly diagnosed malignancy and the leading cause of cancer‐related death worldwide. According to GLOBOCAN 2022, it accounted for 2.5 million new cases (12.4% of all cancers) and 1.8 million deaths (18.7% of all cancer mortality) [1]. In China, lung cancer likewise leads in incidence and mortality, with 1.06 million new cases and over 730,000 deaths reported in 2022 [2]. Tobacco exposure remains the principal risk factor, with 95% of small cell lung cancer (SCLC) patients having a smoking history [3]. Cigarette smoking increases the risk of developing lung cancer by 10‐ to 30‐fold [4]. SCLC constitutes about 13.8% of lung cancers [5], characterized by early metastatic spread, rapid tumor growth, and poor prognosis [6]; over 70% of patients are diagnosed at the extensive stage [7], and the 5‐year survival rate remains below 5% [8].
With population aging, the burden of lung cancer among the elderly continues to rise. The National Comprehensive Cancer Network (NCCN) defines older patients as those aged ≥ 65 years [9]. According to data from the National Bureau of Statistics of China, the proportion of the older population aged ≥ 65 years has increased from 8.9% in 2010 to 15.6% in 2024 [10]. Among this group, lung cancer incidence and mortality have shown sustained growth [11]. Similarly, the proportion of SCLC patients aged > 70 years in the United States increased from 23% in 1975 to 44% in 2010 [12]. However, therapeutic decision‐making for older adults remained challenging, largely due to their underrepresentation in clinical trials—only 25% of participants are aged ≥ 65 years [13]. The scarcity of robust evidence regarding the benefit–risk balance of therapeutic strategies in older adults may lead to nonstandard treatment approaches, compromising therapeutic benefit in this population [14].
Prognostic research has been a longstanding focus in oncology, but in fact, few studies on ES‐SCLC have conducted dedicated analyses in older populations. To address this gap, this study conducted a retrospective analysis of survival outcomes in a real‐world cohort of 270 older patients with ES‐SCLC, aiming to determine whether previously reported prognostic factors remain predictive in this specific population.
2. Materials and Methods
2.1. The Study Cohort
This retrospective real‐world study included 270 older patients with ES‐SCLC who received treatment at The Fourth Hospital of Hebei Medical University between December 2016 and June 2024. This study protocol was approved by the Ethics Committee of the Fourth Hospital of Hebei Medical University and conducted in accordance with the principles of the Declaration of Helsinki. Given the retrospective design and the de‐identification of all patient data, the requirement for written informed consent was waived by the Ethics Committee.
Eligible patients were aged ≥ 65 years at initial diagnosis, with histologically or cytologically confirmed primary ES‐SCLC (as defined by the Veterans Administration Lung Study Group [VALG] staging system). Patients were required to complete at least two cycles of first‐line chemoimmunotherapy (CIT) or chemotherapy (CT) and had evaluable treatment response according to RECIST version 1.1 [15]. Additional inclusion criteria required complete baseline sociodemographic data, peripheral blood laboratory results, documentation of diagnostic and therapeutic procedures, and traceable follow‐up records. Exclusion criteria included prior exposure to any other anti‐tumor therapy (e.g., radiotherapy or surgery) before first‐line treatment, severe comorbidities involving major organ dysfunction (heart, liver, or kidney), severe active autoimmune diseases, or the presence of other concurrent primary malignancies.
2.2. Data Collection
The following data were collected: (1) Baseline sociodemographic data: Age at diagnosis, sex, and smoking history. (2) Baseline peripheral blood laboratory results: Lactate dehydrogenase (LDH), neutrophil‐to‐lymphocyte ratio (NLR), derived neutrophil‐to‐lymphocyte ratio (dNLR), pro‐gastrin‐releasing peptide (ProGRP), and neuron‐specific enolase (NSE) levels, measured at the most recent assessment prior to treatment initiation. For LDH, ProGRP, and NSE, the laboratory upper limit of normal (ULN) was used as the cutoff value. Based on established thresholds from prior relevant studies, the cutoff values were defined as 4.0 for NLR [16] and 3.0 for dNLR [17]. (3) Diagnostic and therapeutic course: Eastern Cooperative Oncology Group (ECOG) performance status (PS) score, sites of distant metastasis, administration of thoracic radiotherapy (TRT), cranial irradiation, or prophylactic cranial irradiation (PCI), first‐line therapeutic regimen (CT or CIT), and best overall response (BOR) according to RECIST 1.1 criteria (including complete response [CR], partial response [PR], stable disease [SD], and progressive disease [PD]) [15].
2.3. Follow‐Up and Study Endpoints
Patient outcomes and survival data were obtained through the institutional follow‐up registry, inpatient and outpatient medical records, and telephone follow‐ups. The follow‐up cutoff date was October 15, 2024. The study endpoints were progression‐free survival (PFS) and overall survival (OS). PFS was defined as the time from the initiation of first‐line systemic therapy to the first documented disease progression or death from any cause. OS was defined as the time from the initiation of first‐line systemic therapy to death from any cause. For patients who had not experienced progression or death by the cutoff date, data were censored at the date of last confirmed follow‐up. All survival times were calculated in months.
2.4. Statistical Analysis
Statistical analyses were performed using SPSS version 26.0 (IBM Corp., Armonk, NY, USA), Graphpad Prism 10.1.2, and R version 4.5.2 (R Foundation for Statistical Computing, Vienna, Austria). Categorical variables were presented as frequencies and percentages (%). Median survival times were estimated using the Kaplan–Meier method, with survival curves plotted accordingly. Univariate survival analyses were conducted using the Cox proportional hazards regression. Variables with p < 0.1 in univariate analysis were entered into multivariable Cox regression models to identify independent prognostic factors. The proportional hazards (PH) assumption for covariates was evaluated using Schoenfeld residuals with the cox.zph function in the survival package. A significant p < 0.05 indicated a violation of the PH assumption. For key predictors exhibiting non‐proportional hazards (NPH), a time‐dependent Cox model was fitted using a time‐splitting approach [18]. Based on visual inspection of the Schoenfeld residual plots for NPH variables, the analysis time was divided into two intervals. The dataset was restructured into a counting process style using the survSplit function in the survival package. After model refitting, the PH assumption was re‐evaluated using the cox.zph function to confirm the adequacy of the revised model. Hazard ratios (HRs) and 95% confidence intervals (CIs) were calculated, with statistical significance defined as two‐sided p < 0.05.
To address potential selection bias and ensure comparability between the CT and CIT groups, propensity score matching (PSM) was applied using a 1:1 nearest‐neighbor matching algorithm with a caliper width of 0.2 standard deviations of the propensity score. Propensity scores were estimated through logistic regression based on the following covariates: Age, sex, ECOG PS score, smoking history, presence of brain, bone, pulmonary, hepatic, and adrenal metastases, number of metastatic organs, and baseline levels of LDH, NLR, dNLR, ProGRP, and NSE. Matching was performed using the MatchIt package, and baseline characteristics were summarized using the tableone package. After matching, survival outcomes were reanalyzed in the balanced cohort.
3. Results
3.1. Patient Baseline Characteristics and Follow‐Up Outcomes
The final analysis encompassed a cohort of 270 older patients diagnosed with ES‐SCLC. Patients were categorized into two age groups: Young‐old (65–74 years) and old‐old (≥ 75 years). Best overall response (BOR) to treatment was classified as either optimal treatment response (CR/PR) or suboptimal treatment response (SD/PD). Clinical characteristics are summarized in Table 1. Among the cohort, 35 patients (13.0%) were aged ≥ 75 years; 96 (35.6%) received first‐line CIT; 222 (82.2%) were male; 54 (20.0%) had ECOG PS score = 0, and 216 (80.0%) had 1–2. The distribution of metastatic sites included the liver (number[n] = 95, 35.2%), bone (n = 83, 30.7%), adrenal gland (n = 42, 15.6%), and lung (n = 38, 14.1%); 51 patients (18.9%) presented with metastases involving more than two organs. Therapeutic interventions comprised TRT in 83 (30.7%), head radiotherapy in 40 (14.8%), and PCI in 8 (3.0%). According to RECIST version 1.1 criteria, the BOR during first‐line treatment was: CR in 0 (0%), PR in 184 (68.2%), SD in 67 (24.8%), and PD in 19 (7.0%).
TABLE 1.
Clinical characteristics of older patients with ES‐SCLC.
| Variable | n (%) | Variable | n (%) | ||
|---|---|---|---|---|---|
| Age (years) | 65–74 | 235 (87.0) | Therapeutic regimen | CIT | 96 (35.6) |
| ≥ 75 | 35 (13.0) | CT | 174 (64.4) | ||
| Sex | Female | 48 (17.8) | TRT | Yes | 83 (30.7) |
| Male | 222 (82.2) | No | 187 (69.3) | ||
| ECOG PS score | 0 | 54 (20.0) | Head radiotherapy | Yes | 40 (14.8) |
| 1–2 | 216 (80.0) | No | 230 (85.2) | ||
| Smoking history | No | 81 (30.0) | PCI | Yes | 8 (3.0) |
| Yes | 189 (70.0) | No | 262 (97.0) | ||
| Brain metastasis | No | 210 (77.8) | LDH (U/L) | ≤ 250 | 134 (49.6) |
| Yes | 60 (22.2) | > 250 | 136 (50.4) | ||
| Bone metastasis | No | 187 (69.3) | NLR | ≤ 4.0 | 177 (65.6) |
| Yes | 83 (30.7) | > 4.0 | 93 (34.4) | ||
| Pulmonary metastasis | No | 232 (85.9) | dNLR | ≤ 3.0 | 198 (73.3) |
| Yes | 38 (14.1) | > 3.0 | 72 (26.7) | ||
| Hepatic metastasis | No | 175 (64.8) | ProGRP (mg/ml) | ≤ 69.2 | 36 (13.3) |
| Yes | 95 (35.2) | > 69.2 | 234 (86.7) | ||
| Adrenal metastasis | No | 228 (84.4) | NSE (ng/ml) | ≤ 16.3 | 10 (3.7) |
| Yes | 42 (15.6) | > 16.3 | 260 (96.3) | ||
| Number of organs with metastasis | ≤ 2 | 219 (81.1) | BOR | CR/PR | 184 (68.1) |
| > 2 | 51 (18.9) | SD/PD | 86 (31.9) | ||
Abbreviations: BOR, Best overall response; CIT, Chemoimmunotherapy; CR, Complete response; CT, Chemotherapy; dNLR, Derived neutrophil‐to‐lymphocyte ratio; ECOG PS, Eastern Cooperative Oncology Group performance status; LDH, Lactate dehydrogenase; NLR, Neutrophil‐to‐lymphocyte ratio; n, Number; NSE, Neuron‐specific enolase; PCI, Prophylactic cranial irradiation; PD, Progressive disease; PR, Partial response; ProGRP, Pro‐gastrin‐releasing peptide; SD, Stable disease; TRT, Thoracic radiotherapy.
As of the final follow‐up cutoff date on October 15, 2024, the median follow‐up duration was 14.2 months (interquartile range [IQR]: 10.0–20.5). A total of 212 death events were recorded. Among all patients, the median progression‐free survival (mPFS) was 6.7 months (95% confidence interval [CI] 6.0–7.4), and the median overall survival (mOS) was 13.1 months (95% CI 11.8–14.4), as shown in Figure 1. The 1‐, 2‐, 3‐, 4‐, and 5‐year OS rates were 57%, 23%, 16%, 12%, and 8%, respectively.
FIGURE 1.
The mPFS (A) and mOS (B) of older patients with ES‐SCLC.
3.2. Analysis of Prognostic Factors
3.2.1. Analysis of Prognostic Factors for PFS
The results are summarized in Table 2. Univariate analysis showed that first‐line CT (HR 1.88, 95% CI 1.40–2.52; p < 0.001) and old‐old (HR 2.27, 95% CI 1.58–3.27; p < 0.001) were significantly correlated with reduced PFS in patients with ES‐SCLC (Figure 2). Variables with p < 0.1 in the univariate analysis were subsequently entered into a multivariate Cox proportional hazards regression model. All variables satisfied the PH assumption, as confirmed by the Schoenfeld residual test (all p > 0.05; Table 3). In the multivariate analysis, first‐line CT (adjusted HR = 1.79, 95% CI 1.33–2.40; p < 0.001) and old‐old (adjusted HR = 2.06, 95% CI 1.43–2.97; p < 0.001) were identified as independent predictors of PFS.
TABLE 2.
Analysis of factors influencing PFS in older patients with ES‐SCLC.
| Variable | Univariate analysis | Multivariate analysis | ||||
|---|---|---|---|---|---|---|
| mPFS (months) (95% CI) | HR (95% CI) | p | HR (95% CI) | p | ||
| Therapeutic regimen | CIT | 7.8 (6.9–8.7) | — | < 0.001 | — | < 0.001 |
| CT | 5.8 (5.1–6.5) | 1.88 (1.40–2.52) | 1.79 (1.33–2.40) | |||
| Sex | Female | 6.7 (5.5–7.9) | — | 0.759 | ||
| Male | 6.7 (5.8–7.6) | 1.06 (0.75–1.48) | ||||
| Age (years) | 65–74 | 7.1 (6.5–7.7) | — | < 0.001 | — | < 0.001 |
| ≥ 75 | 4.0 (3.5–4.5) | 2.27 (1.58–3.27) | 2.06 (1.43–2.97) | |||
| ECOG PS score | 0 | 7.3 (5.4–9.2) | — | 0.262 | ||
| 1–2 | 6.3 (5.5–7.1) | 1.21 (0.87–1.70) | ||||
| Smoking history | No | 7.3 (6.5–8.1) | — | 0.212 | ||
| Yes | 6.3 (5.4–7.2) | 1.20 (0.90–1.60) | ||||
| Brain metastasis | No | 6.7 (5.9–7.5) | — | 0.201 | ||
| Yes | 6.7 (5.0–8.4) | 0.81 (0.58–1.12) | ||||
| Bone metastasis | No | 7.3 (6.5–8.1) | — | 0.067 | — | 0.160 |
| Yes | 5.7 (4.7–6.7) | 1.30 (0.98–1.72) | 1.22 (0.92–1.62) | |||
| Pulmonary metastasis | No | 6.6 (5.8–7.4) | — | 0.564 | ||
| Yes | 6.8 (5.9–7.7) | 1.11 (0.77–1.61) | ||||
| Hepatic metastasis | No | 6.6 (5.9–7.3) | — | 0.851 | ||
| Yes | 6.8 (5.6–8.0) | 1.03 (0.78–1.35) | ||||
| Adrenal metastasis | No | 6.8 (6.0–7.6) | — | 0.375 | ||
| Yes | 6.3 (4.4–8.2) | 1.18 (0.82–1.69) | ||||
| Number of organs with metastasis | ≤ 2 | 6.8 (6.0–7.6) | — | 0.236 | ||
| > 2 | 6.3 (5.5–7.1) | 1.22 (0.88–1.70) | ||||
| LDH (U/L) | ≤ 250 | 6.8 (5.9–7.7) | — | 0.262 | ||
| > 250 | 6.2 (5.0–7.4) | 1.16 (0.89–1.51) | ||||
| NLR | ≤ 4.0 | 6.5 (5.5–7.5) | — | 0.719 | ||
| > 4.0 | 6.7 (5.9–7.5) | 0.95 (0.72–1.25) | ||||
| dNLR | ≤ 3.0 | 6.6 (5.7–7.5) | — | 0.908 | ||
| > 3.0 | 6.7 (5.7–7.7) | 0.98 (0.73–1.32) | ||||
| ProGRP (pg/ml) | ≤ 69.2 | 6.6 (4.8–8.4) | — | 0.384 | ||
| > 69.2 | 6.7 (6.0–7.4) | 1.19 (0.80–1.76) | ||||
| NSE (ng/ml) | ≤ 16.3 | 6.1 (4.7–7.5) | — | 0.918 | ||
| > 16.3 | 6.7 (6.0–7.4) | 0.97 (0.50–1.88) | ||||
Abbreviations: CI, Confidence interval; CIT, Chemoimmunotherapy; CT, Chemotherapy; dNLR, Derived neutrophil‐to‐lymphocyte ratio; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, Hazard ratio; LDH, Lactate dehydrogenase; mPFS, Median progression‐free survival; NLR, Neutrophil‐to‐lymphocyte ratio; NSE, Neuron‐specific enolase; ProGRP, Pro‐gastrin‐releasing peptide.
FIGURE 2.
Prognostic factors for PFS in older patients with ES‐SCLC. Kaplan–Meier estimates of PFS in the (A) Therapeutic regimen, (B) Age.
TABLE 3.
Schoenfeld residual tests for the proportional hazards assumption in the multivariate cox model of PFS.
| Variable | χ2 | df | p |
|---|---|---|---|
| Therapeutic regimen | 0.960 | 1 | 0.327 |
| Age (years) | 0.599 | 1 | 0.439 |
| Bone metastasis | 0.147 | 1 | 0.702 |
| GLOBAL | 1.783 | 3 | 0.619 |
3.2.2. Analysis of Prognostic Factors for OS
Results of the univariate analysis are presented in Table 4 and Figure 3. Old‐old (HR = 1.79, 95% CI 1.23–2.60; p = 0.002), a positive smoking history (HR = 1.44, 95% CI 1.06–1.95; p = 0.020), bone metastasis (HR = 1.65, 95% CI 1.24–2.20; p = 0.001), and high‐LDH (> 250 U/L; HR = 1.69, 95% CI 1.27–2.23; p < 0.001) were significantly associated with inferior OS.
TABLE 4.
Univariate analysis of factors influencing OS in older patients with ES‐SCLC.
| Variable | mOS (months) (95% CI) | HR (95% CI) | p | |
|---|---|---|---|---|
| Therapeutic regimen | CIT | 16.0 (13.2–18.8) | — | 0.088 |
| CT | 12.3 (10.8–13.8) | 1.31 (0.96–1.78) | ||
| Sex | Female | 14.0 (9.6–18.4) | — | 0.433 |
| Male | 12.9 (11.5–14.3) | 1.15 (0.81–1.64) | ||
| Age (years) | 65–74 | 13.6 (11.8–15.4) | — | 0.002 |
| ≥ 75 | 10.5 (8.8–12.2) | 1.79 (1.23–2.60) | ||
| ECOG PS score | 0 | 13.6 (11.4–15.8) | — | 0.277 |
| 1–2 | 12.9 (11.5–14.3) | 1.22 (0.85–1.74) | ||
| Smoking history | No | 14.4 (10.6–18.2) | — | 0.020 |
| Yes | 12.3 (11.0–13.6) | 1.44 (1.06–1.95) | ||
| Brain metastasis | No | 12.8 (11.5–14.1) | — | 0.140 |
| Yes | 14.2 (10.3–18.1) | 0.77 (0.55–1.09) | ||
| Bone metastasis | No | 14.9 (12.9–16.9) | — | 0.001 |
| Yes | 10.8 (9.9–11.7) | 1.65 (1.24–2.20) | ||
| Pulmonary metastasis | No | 12.5 (11.2–13.8) | — | 0.613 |
| Yes | 14.1 (12.8–15.4) | 1.10 (0.76–1.59) | ||
| Hepatic metastasis | No | 13.5 (11.1–15.9) | — | 0.102 |
| Yes | 12.5 (10.5–14.5) | 1.27 (0.96–1.68) | ||
| Adrenal metastasis | No | 13.5 (12.0–15.0) | — | 0.184 |
| Yes | 12.2 (9.4–15.0) | 1.28 (0.89–1.86) | ||
| Number of organs with metastasis | ≤ 2 | 13.1 (10.9–15.3) | — | 0.061 |
| > 2 | 13.1 (10.3–15.9) | 1.38 (0.99–1.93) | ||
| LDH (U/L) | ≤ 250 | 14.1 (11.5–16.7) | — | < 0.001 |
| > 250 | 12.2 (10.4–14.0) | 1.69 (1.27–2.23) | ||
| NLR | ≤ 4.0 | 13.1 (11.4–14.8) | — | 0.983 |
| > 4.0 | 13.1 (11.7–14.5) | 1.00 (0.75–1.34) | ||
| dNLR | ≤ 3.0 | 13.5 (12.0–15.0) | — | 0.678 |
| > 3.0 | 12.5 (10.7–14.3) | 1.07 (0.79–1.45) | ||
| ProGRP (pg/ml) | ≤ 69.2 | 13.6 (8.4–18.8) | — | 0.092 |
| > 69.2 | 13.1 (11.9–14.3) | 1.44 (0.94–2.20) | ||
| NSE (ng/ml) | ≤ 16.3 | 13.6 (11.7–15.5) | — | 0.743 |
| > 16.3 | 13.1 (11.8–14.4) | 1.12 (0.57–2.18) | ||
Abbreviations: CI, Confidence interval; CIT, Chemoimmunotherapy; CT, Chemotherapy; dNLR, Derived neutrophil‐to‐lymphocyte ratio; ECOG PS, Eastern Cooperative Oncology Group performance status; HR, Hazard ratio; LDH, Lactate dehydrogenase; mOS, Median overall survival; NLR, Neutrophil‐to‐lymphocyte ratio; NSE, Neuron‐specific enolase; ProGRP, Pro‐gastrin‐releasing peptide.
FIGURE 3.
Prognostic factors for OS in older patients with ES‐SCLC. Kaplan–Meier estimates of OS in the (A) Therapeutic regimen, (B) Age, (C) Smoking history, (D) Bone metastasis, (E) LDH, and (F) ProGRP. LDH, Lactate dehydrogenase; ProGRP, Pro‐gastrin‐releasing peptide.
Variables with p < 0.1 in the univariate analysis were further evaluated for the PH assumption. The results indicated that ProGRP violated the PH assumption (p = 0.043; Table 5). The Schoenfeld residual plot revealed a distinct upward trend in the time‐varying regression coefficient β(t) for ProGRP, with values transitioning from negative in the early follow‐up period to positive beyond approximately 10 months (Figure 4). Accordingly, 10 months was selected as the cutoff point, and the analysis time was divided into two intervals.
TABLE 5.
Schoenfeld residual tests for the proportional hazards assumption in the multivariate cox model of OS before and after time‐splitting.
| Variable | Before time‐splitting | After time‐splitting | ||||
|---|---|---|---|---|---|---|
| χ2 | df | p | χ2 | df | p | |
| Therapeutic regimen | 0.340 | 1 | 0.560 | 0.324 | 1 | 0.569 |
| Age (years) | 0.008 | 1 | 0.927 | 0.003 | 1 | 0.953 |
| Smoking history | 0.249 | 1 | 0.618 | 0.233 | 1 | 0.630 |
| Bone metastasis | 0.797 | 1 | 0.372 | 1.032 | 1 | 0.310 |
| Number of organs with metastasis | 0.005 | 1 | 0.941 | 0.004 | 1 | 0.951 |
| LDH (U/L) | 0.366 | 1 | 0.545 | 0.451 | 1 | 0.502 |
| ProGRP (pg/ml) | 4.089 | 1 | 0.043 | 0.994 | 2 | 0.608 |
| GLOBAL | 6.629 | 7 | 0.469 | 3.535 | 8 | 0.896 |
Abbreviations: LDH, Lactate dehydrogenase; ProGRP, Pro‐gastrin‐releasing peptide.
FIGURE 4.
Schoenfeld residual plots of the non‐proportional hazard variable for OS. The x‐axis represents time, and the y‐axis shows the estimated time‐varying regression coefficient β(t). The solid black line denotes the estimated time‐varying hazard, the dashed black lines indicate the 95% confidence interval, and the green dashed line represents the average hazard over time. The horizontal dotted line marks the null effect reference (β = 0). Each circle represents a residual at a given time point, illustrating deviations from the proportional hazards assumption. ProGRP, Pro‐gastrin‐releasing peptide.
In the subsequent multivariable analysis, several variables were identified as significant independent predictors of OS (Table 6). Old‐old (adjusted HR = 1.86, 95% CI 1.27–2.73; p = 0.001), a positive smoking history (adjusted HR = 1.42, 95% CI 1.04–1.93; p = 0.028), bone metastasis (adjusted HR = 1.54, 95% CI 1.14–2.07; p = 0.005), and high‐LDH (adjusted HR = 1.63, 95% CI 1.23–2.17; p < 0.001) were independently associated with shorter OS. The time‐split analysis further revealed that the prognostic effect of ProGRP varied across time intervals. Beyond 10 months of follow‐up, high‐ProGRP (> 69.2 pg/mL) was independently associated with a significantly increased risk of death (adjusted HR = 1.85, 95% CI: 1.05–3.26, p = 0.032). Conversely, no significant association was observed during the initial 10 months (adjusted HR = 0.84, 95% CI: 0.44–1.60, p = 0.604).
TABLE 6.
Multivariate analysis of factors influencing OS after time‐splitting in older patients with ES‐SCLC.
| Variable | n (%) | HR (95% CI) | p | |
|---|---|---|---|---|
| Therapeutic regimen | CIT | 96 (35.6) | — | 0.087 |
| CT | 174 (64.4) | 1.31 (0.96–1.79) | ||
| Age (years) | 65–74 | 235 (87.0) | — | 0.001 |
| ≥ 75 | 35 (13.0) | 1.86 (1.27–2.73) | ||
| Smoking history | No | 81 (30.0) | — | 0.028 |
| Yes | 189 (70.0) | 1.42 (1.04–1.93) | ||
| Bone metastasis | No | 187 (69.3) | — | 0.005 |
| Yes | 83 (30.7) | 1.54 (1.14–2.07) | ||
| Number of organs with metastasis | ≤ 2 | 219 (81.1) | — | 0.275 |
| > 2 | 51 (18.9) | 1.22 (0.86–1.73) | ||
| LDH (U/L) | ≤ 250 | 134 (49.6) | — | < 0.001 |
| > 250 | 136 (50.4) | 1.63 (1.23–2.17) | ||
| ProGRP (tgroup = 1) | ≤ 69.2 | 36 (13.3) | — | 0.604 |
| > 69.2 | 234 (86.7) | 0.84 (0.44–1.60) | ||
| ProGRP (tgroup = 2) | ≤ 69.2 | 21 (12.6) | — | 0.032 |
| > 69.2 | 146 (87.4) | 1.85 (1.05–3.26) | ||
Note: The analysis time was divided into two intervals based on a 10‐month cutoff (tgroup = 1, ≤ 10 months; tgroup = 2, > 10 months) to address the non‐proportional hazards observed in the Cox regression model. LDH, Lactate dehydrogenase; ProGRP, Pro‐gastrin‐releasing peptide.
3.2.3. PSM Results
After PSM, 92 patients who received CIT were successfully matched with 92 patients who received CT (total n = 184). The baseline characteristics between the two groups were well balanced, with no significant differences across all variables (all p > 0.05, Table S1). In the matched cohort, the mPFS in the CT group remained significantly shorter than that in the CIT group (6.8 vs. 7.8 months; HR = 1.60, 95% CI 1.14–2.23; p = 0.006). The difference in mOS between the two groups was still not statistically significant (14.3 vs. 16.0 months; HR = 1.09, 95% CI 0.77–1.56; p = 0.631). Multivariate analysis of the matched cohort further confirmed that first‐line CT (compared with CIT) remained an independent adverse prognostic factor for PFS (adjusted HR = 1.60, 95% CI 1.15–2.24; p = 0.006), which was consistent with the findings from the overall cohort analysis (Table S2 to Table S5, Figure S1).
4. Discussion
Although numerous studies have explored prognostic indicators for SCLC, research focusing specifically on the older population remains limited. However, this population often exhibits distinct clinical characteristics, such as a higher prevalence of comorbidities, altered physiological functions, and varying responses to treatment compared to younger individuals. In this retrospective analysis of real‐world data, we sought to determine whether conventional cancer prognostic factors preserve their prognostic validity among older patients with ES‐SCLC.
In this study, patients receiving first‐line CIT achieved significantly longer PFS than those treated with CT alone, although no OS benefit was observed. A growing body of randomized controlled trials (RCTs), such as IMpower133 [19], CASPIAN [20], ASTRUM‐005 [21], and CAPSTONE‐1 [22], has demonstrated that first‐line CIT yields significant improvements in PFS and OS relative to CT alone for patients with ES‐SCLC. The results of this study suggest that immunotherapy confers better disease control through antitumor immune activation. However, the survival benefit may be offset in older patients by comorbidities (such as coronary atherosclerotic heart disease and chronic obstructive pulmonary disease), frailty, and treatment intolerance. Treatment‐related toxicity can substantially compromise treatment intensity and quality of life, ultimately undermining OS. Moreover, patients initially treated with CT alone still had the opportunity to receive immunotherapy in subsequent lines and could also undergo local treatments such as radiotherapy, which may have improved their survival outcomes. For the reasons outlined above, the first‐line therapeutic regimen may not confer a significant OS benefit in this specific population with ES‐SCLC.
Age is a critical consideration in the treatment of older patients with ES‐SCLC. Age subgroup analysis in this study revealed that old‐old (≥ 75 years) constitutes an independent risk factor for both PFS and OS, implying a diminished therapeutic response among old‐old patients. This phenomenon may be attributed to the progression of immunosenescence associated with aging, characterized by the deterioration of immune organs and consequent impairment of immune cell function [23]. Immunosenescence is also closely linked to inflammation. Cells characterized by the senescence‐associated secretory phenotype (SASP) sustain the secretion of multiple factors, such as chemokines, angiogenic factors, and pro‐inflammatory cytokines. This persistent secretion contributes to chronic low‐grade inflammation, which in turn facilitates the recruitment of immunosuppressive cells, ultimately resulting in the attenuation of immune function [24]. Therefore, older patients may exhibit a diminished response to therapy, particularly to immunotherapy. Increasing age is also associated with a decline in organ function and diminished tolerance to therapeutic interventions, thereby elevating the risk of severe toxicities such as myelotoxicity, fatigue, and gastrointestinal effects, as well as a higher likelihood of treatment discontinuation. Clinically, it is essential for physicians to consider not only patients' chronological age but, more importantly, their physiological age, especially in the older population. This approach is in alignment with current NCCN recommendations [9].
A history of smoking, bone metastasis and high‐LDH (> 250 U/L) were also identified as independent adverse prognostic factors for OS in older patients with ES‐SCLC, consistent with previous findings. Smoking history has been recognized as an independent adverse prognostic indicator that impacts both PFS and OS in patients with SCLC [25]. Therefore, smoking cessation remains a critical component of supportive care and should be strongly promoted throughout the continuum of SCLC management. A study based on the SEER database, which included 4201 patients with SCLC, found that patients with bone metastasis had a significantly shorter mOS compared to those without bone metastasis (6 vs. 10 months; p = 0.001) [26]. Clinically, bone metastasis can lead to severe complications, such as debilitating bone pain, pathological fractures, and spinal cord compression. These complications collectively exert a profoundly negative impact on patients' quality of life and contribute to reduced survival. Cancer cells preferentially metabolize glucose to pyruvate via glycolysis, subsequently converting pyruvate to lactate through LDH‐A catalysis—even under normoxic conditions. This constitutes the Warburg effect [27]. Elevated LDH is recognized as a marker of high tumor burden and is associated with an unfavorable prognosis in cancer patients [28]. A large‐scale meta‐analysis encompassing 31,857 patients from multiple clinical studies established that elevated LDH levels significantly correlate with poorer OS across diverse solid tumor types. Notably, LDH demonstrated enhanced prognostic impact in metastatic disease (reflecting higher tumor burden) compared to non‐metastatic disease [29]. Analysis of baseline LDH levels and treatment efficacy in 585 ES‐SCLC patients from the phase 3 ASTRUM‐005 trial revealed significantly longer OS in patients with LDH≤ULN vs. those with LDH>ULN, and LDH is an independent prognostic biomarker for ES‐SCLC [30].
ProGRP exhibited a clear time‐dependent prognostic trend. While it showed no significant association with mortality within the first 10 months, it emerged as a significant prognostic factor for long‐term survival thereafter (HR = 1.85, p = 0.032). As a well‐established biomarker in SCLC, ProGRP has been widely validated for diagnosis, disease monitoring, and evaluation of treatment response [31, 32]. This delayed prognostic effect suggests that ProGRP is more likely to reflect the biological aggressiveness of the tumor rather than short‐term treatment response. Therefore, dynamic monitoring of ProGRP during follow‐up may provide valuable information for predicting long‐term prognosis.
This study has several limitations inherent to its retrospective, real‐world design. First, the potential for selection and confounding biases cannot be fully eliminated, as treatment choices were not randomized and may reflect underlying patient fitness or physician preference. Despite multivariate adjustments, unmeasured confounders may persist. Second, the completeness and accuracy of the data were dependent on medical records, which may introduce information bias from sources such as incomplete or non‐standardized documentation; additionally, the assessment of treatment response could be affected by inconsistent intervals in follow‐up imaging. Third, the lack of detailed data on treatment‐related adverse events prevented further evaluation of therapy tolerability and safety. Lastly, the single‐center nature of the data may also limit the generalizability of the findings. Therefore, the identified factors should be interpreted as independent prognostic factors rather than definitive causal factors. Further prospective validation is required.
Future research should focus on advancing prospective studies specifically designed for older patients with ES‐SCLC, integrating multidimensional data such as clinical characteristics, comprehensive geriatric assessment (CGA) parameters, radiomics, liquid biopsies (e.g., ctDNA), and tumor microenvironment molecular features. This integration will facilitate the development of more refined prognostic and predictive models, enabling truly individualized treatment decision‐making. Furthermore, leveraging large‐scale, multicenter real‐world databases with standardized data collection for external validation and in‐depth exploration of long‐term outcomes will be crucial.
5. Conclusion
These findings not only corroborate the prognostic value of previously identified risk factors in older patients with ES‐SCLC but also demonstrate that the prognostic impact of ProGRP is distinctly time‐dependent.
Author Contributions
Bo Wang: investigation. Jiayin Liu: software, data curation, investigation, formal analysis, writing – original draft, writing – review and editing, visualization. Xiaolin Li: methodology, investigation. Li Feng: resources, formal analysis. Ning Liu: investigation. Yan Liu: investigation. Xun Liu: data curation, investigation, writing – original draft, visualization. Jing Han: methodology, validation. Hui Jin: investigation. Zhisong Fan: validation, resources. Long Wang: writing – review and editing. Xue Zhang: validation. Dan Li: validation, writing – review and editing. Jing Zuo: writing – review and editing, resources, supervision. Yudong Wang: conceptualization, resources, project administration, funding acquisition, supervision, writing – review and editing.
Funding
This work was supported by Hebei Natural Science Foundation (H2025206698) and Wu Jieping Medical Foundation (320.6750.2021‐01‐10).
Disclosure
The authors declare that the content and conclusions of this article are solely the views of the authors and should not be interpreted as representing the views of The Fourth Hospital of Hebei Medical University or the funding agencies.
Ethics Statement
This study protocol was approved by the Ethics Committee of the Fourth Hospital of Hebei Medical University and was conducted in accordance with the Declaration of Helsinki and Clinical Practice guidelines.
Conflicts of Interest
The authors declare no conflicts of interest.
Supporting information
Table S1: Comparison of baseline characteristics before and after matching between the CT group and the CIT group.
Table S2: Analysis of factors influencing PFS in older ES‐SCLC patients after matching.
Table S3: Schoenfeld residual tests for the proportional hazards assumption in the multivariate Cox model of PFS after matching.
Table S4: Univariate analysis of factors influencing OS in older ES‐SCLC patients after matching.
Table S5: Schoenfeld residual tests for the proportional hazards assumption in the multivariate Cox model of OS after matching.
Figure S1: Survival analysis of the CT and CIT groups after matching. (A) Kaplan–Meier curve for PFS. (B) Kaplan–Meier curve for OS.
Acknowledgements
The authors have nothing to report.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.
References
- 1. Bray F., Laversanne M., Sung H., et al., “Global Cancer Statistics 2022: GLOBOCAN Estimates of Incidence and Mortality Worldwide for 36 Cancers in 185 Countries,” CA: a Cancer Journal for Clinicians 74 (2024): 229–263. [DOI] [PubMed] [Google Scholar]
- 2. Han B., Zheng R., Zeng H., et al., “Cancer Incidence and Mortality in China, 2022,” Journal of the National Cancer Center 4 (2024): 47–53. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Kim S. Y., Park H. S., and Chiang A. C., “Small Cell Lung Cancer: A Review,” JAMA 333 (2025): 1906–1917. [DOI] [PubMed] [Google Scholar]
- 4. Leiter A., Veluswamy R. R., and Wisnivesky J. P., “The Global Burden of Lung Cancer: Current Status and Future Trends,” Nature Reviews. Clinical Oncology 20 (2023): 624–639. [DOI] [PubMed] [Google Scholar]
- 5. Cittolin‐Santos G. F., Knapp B., Ganesh B., et al., “The Changing Landscape of Small Cell Lung Cancer,” Cancer 130 (2024): 2453–2461. [DOI] [PubMed] [Google Scholar]
- 6. Zhang J., Zeng X., Guo Q., et al., “Small Cell Lung Cancer: Emerging Subtypes, Signaling Pathways, and Therapeutic Vulnerabilities,” Experimental Hematology & Oncology 13 (2024): 78. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Pavan A., Attili I., Pasello G., Guarneri V., Conte P. F., and Bonanno L., “Immunotherapy in Small‐Cell Lung Cancer: From Molecular Promises to Clinical Challenges,” Journal for Immunotherapy of Cancer 7 (2019): 205. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Jeon D. S., Kim H. C., Kim S. H., et al., “Five‐Year Overall Survival and Prognostic Factors in Patients With Lung Cancer: Results From the Korean Association of Lung Cancer Registry (KALC‐R) 2015,” Cancer Research and Treatment 55 (2023): 103–111. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. National Comprehensive Cancer Network , “NCCN Clinical Practice Guidelines in Oncology: Older Adult Oncology,” NCCN Clinical Practice Guidelines in Oncology. National Comprehensive Cancer Network. Available from, https://www.nccn.org/guidelines.
- 10. National Bureau of Statistics , “Annual data on China's population age structure and dependency ratio for the year 2024,” National Data—National Bureau of Statistics of China. Available from, https://data.stats.gov.cn/.
- 11. Meng F., Yu L., Ma X., and Sun H., “Incidence and Mortality of Cancer in Elderly Population in China,” Journal of Evidence⁃Based Medicine 23 (2023): 169–173. [Google Scholar]
- 12. Abdel‐Rahman O., “Changing Epidemiology of Elderly Small Cell Lung Cancer Patients Over the Last 40 Years; a SEER Database Analysis,” Clinical Respiratory Journal 12 (2018): 1093–1099. [DOI] [PubMed] [Google Scholar]
- 13. Lewis J. H., Kilgore M. L., Goldman D. P., et al., “Participation of Patients 65 Years of Age or Older in Cancer Clinical Trials,” Journal of Clinical Oncology 21 (2003): 1383–1389. [DOI] [PubMed] [Google Scholar]
- 14. Rocque G. B. and Williams G. R., “Bridging the Data‐Free Zone: Decision Making for Older Adults With Cancer,” Journal of Clinical Oncology 37 (2019): 3469–3471. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. Eisenhauer E. A., Therasse P., Bogaerts J., et al., “New Response Evaluation Criteria in Solid Tumours: Revised RECIST Guideline (Version 1.1),” European Journal of Cancer 45 (2009): 228–247. [DOI] [PubMed] [Google Scholar]
- 16. Kumar N. S., Hoesley J., Wooldridge J. T., and Arnaoutakis K., “Prognostic Significance of Neutrophil‐To‐Lymphocyte Ratio in Small Cell Lung Cancer in the Era of Immunotherapy,” Journal of Clinical Oncology 43 (2025): e20129‐e20129. [Google Scholar]
- 17. Yang T., Hao L., Yang X., et al., “Prognostic Value of Derived Neutrophil‐To‐Lymphocyte Ratio (dNLR) in Patients With Non‐Small Cell Lung Cancer Receiving Immune Checkpoint Inhibitors: A Meta‐Analysis,” BMJ Open 11 (2021): e049123. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Zhang Z., Reinikainen J., Adeleke K. A., Pieterse M. E., and Groothuis‐Oudshoorn C. G. M., “Time‐Varying Covariates and Coefficients in Cox Regression Models,” Annals of Translational Medicine 6 (2018): 121. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Horn L., Mansfield A. S., Szczęsna A., et al., “First‐Line Atezolizumab Plus Chemotherapy in Extensive‐Stage Small‐Cell Lung Cancer,” New England Journal of Medicine 379 (2018): 2220–2229. [DOI] [PubMed] [Google Scholar]
- 20. Paz‐Ares L., Dvorkin M., Chen Y., et al., “Durvalumab Plus Platinum‐Etoposide Versus Platinum‐Etoposide in First‐Line Treatment of Extensive‐Stage Small‐Cell Lung Cancer (CASPIAN): A Randomised, Controlled, Open‐Label, Phase 3 Trial,” Lancet 394 (2019): 1929–1939. [DOI] [PubMed] [Google Scholar]
- 21. Cheng Y., Han L., Wu L., et al., “Effect of First‐Line Serplulimab vs Placebo Added to Chemotherapy on Survival in Patients With Extensive‐Stage Small Cell Lung Cancer: The ASTRUM‐005 Randomized Clinical Trial,” JAMA 328 (2022): 1223–1232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Wang J., Zhou C., Yao W., et al., “Adebrelimab or Placebo Plus Carboplatin and Etoposide as First‐Line Treatment for Extensive‐Stage Small‐Cell Lung Cancer (CAPSTONE‐1): A Multicentre, Randomised, Double‐Blind, Placebo‐Controlled, Phase 3 Trial,” Lancet Oncology 23 (2022): 739–747. [DOI] [PubMed] [Google Scholar]
- 23. Wang Y., Dong C., Han Y., Gu Z., and Sun C., “Immunosenescence, Aging and Successful Aging,” Frontiers in Immunology 13 (2022): 942796. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Liu Z., Liang Q., Ren Y., et al., “Immunosenescence: Molecular Mechanisms and Diseases,” Signal Transduction and Targeted Therapy 8 (2023): 200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. Liu X., Jiang T., Li W., et al., “Characterization of Never‐Smoking and Its Association With Clinical Outcomes in Chinese Patients With Small‐Cell Lung Cancer,” Lung Cancer 115 (2018): 109–115. [DOI] [PubMed] [Google Scholar]
- 26. Chen Q., Liang H., Zhou L., et al., “Deep Learning of Bone Metastasis in Small Cell Lung Cancer: A Large Sample‐Based Study,” Frontiers in Oncology 13 (2023): 1097897. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27. DeBerardinis R. J. and Chandel N. S., “Fundamentals of Cancer Metabolism,” Science Advances 2 (2016): e1600200. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Claps G., Faouzi S., Quidville V., et al., “The Multiple Roles of LDH in Cancer,” Nature Reviews. Clinical Oncology 19 (2022): 749–762. [DOI] [PubMed] [Google Scholar]
- 29. Zhang J., Yao Y. H., Li B. G., Yang Q., Zhang P. Y., and Wang H. T., “Prognostic Value of Pretreatment Serum Lactate Dehydrogenase Level in Patients With Solid Tumors: A Systematic Review and Meta‐Analysis,” Scientific Reports 5 (2015): 9800. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Cheng Y., Han L., Wu L., et al., “Abstract 6390: Exploratory Biomarker Analysis of Phase 3 ASTRUM‐005 Study: Serplulimab Versus Placebo Plus Chemotherapy for Extensive‐Stage Small Cell Lung Cancer,” Cancer Research 84 (2024): 6390. [Google Scholar]
- 31. Takada M., Kusunoki Y., Masuda N., et al., “Pro‐Gastrin‐Releasing Peptide (31‐98) as a Tumour Marker of Small‐Cell Lung Cancer: Comparative Evaluation With Neuron‐Specific Enolase,” British Journal of Cancer 73 (1996): 1227–1232. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Grønberg M., Aanerud M., Halvorsen T. O., Killingberg K. T., and Grønberg B. H., “Associations Between Progastrin‐Releasing Peptide (ProGRP) and Neuron‐Specific Enolase (NSE) and Survival in Patients With Limited‐Stage Small Cell Lung Cancer (LS SCLC) Receiving Chemoradiotherapy (CRT),” Lung Cancer (Amsterdam, Netherlands) 207 (2025): 108678. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Table S1: Comparison of baseline characteristics before and after matching between the CT group and the CIT group.
Table S2: Analysis of factors influencing PFS in older ES‐SCLC patients after matching.
Table S3: Schoenfeld residual tests for the proportional hazards assumption in the multivariate Cox model of PFS after matching.
Table S4: Univariate analysis of factors influencing OS in older ES‐SCLC patients after matching.
Table S5: Schoenfeld residual tests for the proportional hazards assumption in the multivariate Cox model of OS after matching.
Figure S1: Survival analysis of the CT and CIT groups after matching. (A) Kaplan–Meier curve for PFS. (B) Kaplan–Meier curve for OS.
Data Availability Statement
The data that support the findings of this study are available from the corresponding author upon reasonable request.