Prognostic significance of advanced lung cancer inflammation index in resectable pancreatic cancer: A retrospective study

Abstract

BACKGROUND

Pancreatic cancer (PC), ranking among the most aggressive solid malignancies, currently lacks validated prognostic biomarkers to guide survival stratification. With a 5-year survival rate under 10%, this malignancy urgently requires precision tools for outcome prediction to optimize therapeutic decision-making.

AIM

To analyze whether the advanced lung cancer inflammation index (ALI) is a prognostic indicator for PC.

METHODS

Patients who were diagnosed with PC and underwent radical resection were included from January 2007 to January 2023 in a clinical center from National Cancer Center of China. The patients were divided into low and high ALI groups according to an ALI cut-off of 34.0 calculated with software X-tile. Overall survival (OS) and surgical outcomes were calculated between the two groups. Follow-up was conducted through telephone interview. Kaplan-Meier analysis was performed to estimate OS, while the log-rank test was utilized to compare OS among different tumor stages. Cox regression was used to identify independent risk factors for OS.

RESULTS

This study included 611 patients who underwent radical PC surgery. Using an ALI cutoff of 34.0, the patients were categorized into a high ALI group (n = 378) and a low ALI group (n = 233). The low ALI group had significantly lower body mass index, serum albumin, lymphocyte count, and ALI (P < 0.01), but higher neutrophil count, a higher proportion of head and neck PC, and longer operation time (P < 0.01). As for prognosis, the low ALI group had worse OS in stage I patients (P < 0.01), and low ALI (P = 0.018, hazard ratio = 0.784, 95% confidence interval: 0.641-0.960) independently conferred an increased risk for mortality.

CONCLUSION

Lower ALI is associated with worse OS for PC patients who underwent radical surgery. Patients demonstrating low ALI preoperatively require special attention from surgeons.

Key Words: Pancreatic cancer; Advanced lung cancer inflammation index; Outcomes; Overall survival; Prognostic

Core Tip: A total of 611 patients who underwent radical pancreatic cancer (PC) surgery were included in this study. The low advanced lung cancer inflammation index (ALI) group had significantly lower body mass index, serum albumin, lymphocyte count, and ALI, but higher neutrophil count, a higher proportion of head and neck PC, and longer operation time. Although there was difference in baseline information, Cox analysis was conducted. As for prognosis, the low ALI group had worse overall survival (OS) in stage I patients, and ALI was identified as an independent risk factor for OS.

INTRODUCTION

Pancreatic cancer (PC) is an aggressive malignancy with subtle early symptoms, and its global incidence and mortality rates remain high. PC continues to portend grave prognosis, with population-based studies demonstrating 5-year survival rates hovering around 10%, which plummets to a dismal 3% in metastatic disease. Early detection remains elusive, as only 15%-20% of incident cases present with anatomically resectable tumors at initial diagnosis[1-3]. Despite some advances in surgical treatment, chemotherapy, radiotherapy, and targeted therapies in recent years, the prognosis for PC remains poor[4-6]. Most patients are diagnosed at an advanced stage, missing the optimal window for treatment[7]. Radical surgical resection is currently the only treatment that offers potential curative benefits; however, significant prognostic variability remains even among patients who undergo radical surgery[8-10].

Inflammation is closely linked to the progression and prognosis of cancer[11-13]. Recently, the advanced lung cancer inflammation index (ALI) has gained significant attention in various malignant tumors as a comprehensive measure of nutritional status and systemic inflammation[14,15]. ALI is calculated by combining the body mass index (BMI), serum albumin (Alb), and the neutrophil to lymphocyte ratio (NLR). Studies have shown that ALI had a prognostic role in gastric cancer, colorectal cancer, and liver cancer[16-20].

However, research about ALI in PC is still limited. Only one study reported the effect of ALI on cancer-specific survival (CSS) in PC, and the number of included patients was 429[21]. It is necessary to investigate the precise impact of ALI on PC, especially the overall survival (OS). Thus, the purpose of this current study was to analyze whether ALI is a prognostic indicator for PC.

MATERIALS AND METHODS

Patients

Patients who were diagnosed with PC and underwent radical resection (n = 679) were included from January 2007 to January 2023 in a clinical center from National Cancer Center of China. The study was approved by the local Ethics Committee and was conducted in accordance with the World Medical Association Declaration of Helsinki. The exclusion criteria were as follows: (1) Non-R0 surgery (n = 11); (2) Incomplete clinical data (n = 13); (3) Stage IV patients (n = 23); and (4) Incomplete records of hematological tests (n = 21). Finally, a total of 611 patients were included in this study (Figure 1).

Figure 1  Study population enrollment flowchart with inclusion and exclusion criteria.

Data collection

The collected baseline characteristics included age, sex, BMI, smoking and drinking status, hypertension, type 2 diabetes mellitus, Alb, neutrophil and lymphocyte counts, and ALI, as well as tumor location, stage, and size. Outcomes assessed were operation time, intraoperative blood loss, postoperative hospital stay, and OS. Data were obtained from electronic medical records, outpatient visits, and telephone interviews.

Definitions

The definition of PC in this study followed the diagnostic criteria outlined in clinical guidelines. OS was defined as the period from surgery to death or loss to follow-up. ALI was calculated using the following formula: ALI = BMI × Alb/NLR, where BMI = weight (kg)/height² (m²), Alb = serum Alb (g/dL), and NLR = absolute neutrophil count/absolute lymphocyte count. The optimal cutoff values for the ALI were identified using X-tile software, based on the maximization of log-rank statistics for survival differences. This algorithm performs a systematic evaluation of all potential ALI thresholds via iterative partitioning and selects the value that provides the greatest statistical discrimination in OS between low-risk and high-risk groups. The optimal cutoff value was determined to be 34.0, based on which the patients were classified into a low ALI group (ALI ≤ 34.0) and a high ALI group (ALI > 34.0).

Treatment and follow-up

All patients underwent radical surgery following standard principles, with R0 resection confirmed pathologically. Follow-ups were conducted regularly thereafter. Follow-up protocols documented OS, disease recurrence, and vital status through telephone interviews, with data censoring occurring at 5-year postoperative milestone or mortality event, whichever came first.

Statistical analysis

Continuous variables, presented as the mean ± SD, were compared using an independent-sample t-test. Categorical variables, shown as counts and percentages, were analyzed by the χ² test or Fisher’s exact test. Kaplan-Meier analysis was performed to estimate OS, while the log-rank test was employed to compare OS among different tumor stages. Cox model assumptions were rigorously validated. Proportional hazards assumptions were tested via Schoenfeld residuals, with covariate-specific testing confirming no violations. Continuous variable linearity was assessed using restricted cubic splines, demonstrating no significant deviation from linearity. Martingale residual plots were used to confirm model adequacy. All covariates satisfied modeling prerequisites. Cox regression was used to identify independent risk factors for OS. Data were analyzed using Statistical Package for the Social Sciences 22.0, with P < 0.05 considered statistically significant.

RESULTS

Patients

A total of 611 patients who underwent radical PC surgery were included based on the eligibility criteria. The average age was 60.7 ± 9.3 years, with 344 males (56.3%) and 267 females (43.7%). The average BMI was 23.5 ± 3.3 kg/m² and ALI was 44.0 ± 23.5. We also included smoking and drinking history, and the surgical information was included as well. Additional baseline characteristics are presented in Table 1.

Table 1 Clinical characteristics.

Characteristic	n = 611
Age (years)	60.7 ± 9.3
Sex
Male	344 (56.3)
Female	267 (43.7)
Body mass index (kg/m2)	23.5 ± 3.3
Smoking	148 (24.2)
Drinking	109 (17.8)
Hypertension	173 (28.3)
Type 2 diabetes mellitus	177 (29.0)
Albumin (g/L)	41.8 ± 5.6
Neutrophils (× 109/L)	4.2 ± 2.0
Lymphocytes (× 109/L)	1.6 ± 0.6
Advanced lung cancer inflammation index	44.0 ± 23.5
Tumor location
Pancreatic head and neck	315 (51.6)
Pancreatic body and tail	296 (48.4)
TNM stage
I	306 (50.1)
II	158 (25.9)
III	147 (24.1)
Tumor size (cm)	3.8 ± 1.6
Operation time (min)	251.0 ± 112.0
Blood loss (mL)	411.7 ± 476.4
Hospital stays (days)	16.8 ± 10.3

Variables are expressed as the mean ± SD, n (%).

Comparison of baseline information

Based on the ALI cutoff, 233 patients were classified into the low ALI group and 378 patients into the high ALI group. The low ALI group had significantly lower BMI, serum Alb, lymphocyte count, and ALI (P < 0.01), but higher neutrophil count and a higher proportion of head and neck PC (P < 0.01) (Table 2).

Table 2 Baseline information between high advanced lung cancer inflammation index group and low advanced lung cancer inflammation index group.

Characteristic	Low ALI (n = 233)	High ALI (n = 378)	P value
Age (years)	61 + 12	62 + 11	0.508
Sex			0.418
Male	136 (58.4)	208 (55.0)
Female	97 (41.6)	170 (45.0)
Body mass index (kg/m2)	22.7 + 3.9	23.7 + 4.3	< 0.01
Smoking	54 (23.2)	94 (24.9)	0.635
Drinking	35 (15.0)	74 (19.6)	0.153
Hypertension	76 (32.6)	97 (25.7)	0.064
Type 2 diabetes mellitus	67 (28.8)	110 (29.1)	0.927
Albumin (g/L)	41.0 + 7	42.6 + 6.2	< 0.01
Neutrophils (× 109/L)	4.7 + 2.3	3.4 + 1.4	< 0.01
Lymphocytes (× 109/L)	1.2 + 0.5	1.8 + 0.7	< 0.01
ALI	24.7 + 10.9	50.1 + 24.0	< 0.01
Tumor location			< 0.01
Pancreatic head and neck	143 (61.4)	172 (45.5)
Pancreatic body and tail	90 (38.6)	206 (54.5)
TNM stage			0.691
I	116 (49.8)	190 (50.3)
II	57 (24.5)	101 (26.7)
III	60 (25.8)	87 (23.0)
Tumor size	3.5 + 1.9	3.5 + 2	0.830

Continuous variables are expressed as median and interquartile range, n (%). ALI: Advanced lung cancer inflammation index.

Surgical outcomes

The low ALI group had longer operation time than the high ALI group (P < 0.01). There was no significant difference between the two groups in terms of intraoperative blood loss (P = 0.248) or postoperative hospital stay (P = 0.207) (Table 3).

Table 3 Outcomes between high advanced lung cancer inflammation index group and low advanced lung cancer inflammation index group.

Characteristic	Low ALI (n = 233)	High ALI (n = 378)	P value
Operation time (min)	260.0 + 150.0	219.5 + 147.8	< 0.01
Intraoperative blood loss (mL)	300.0 + 500.0	200.0 + 500.0	0.248
Postoperative hospital stays (days)	14.0 + 9.0	14.0 + 10.0	0.207

Continuous variables are expressed as median and interquartile range. ALI: Advanced lung cancer inflammation index.

Kaplan-Meier curves in different tumor stages

OS was recorded with a median follow-up of 16.5 (1-200) months. OS was compared between groups among different tumor stages. Survival analysis revealed significantly worse OS in the low ALI group across all stages combined (P < 0.01) and specifically in stage I patients (P < 0.01). However, ALI showed no significant prognostic association in stage II (P = 0.863) or III (P = 0.723) disease (Figure 2).

Figure 2 Prognostic stratification by advanced lung cancer inflammation index for overall survival across all tumor stages combined and in stage I-III subgroups. A: Comparison of overall survival (OS) between high advanced lung cancer inflammation index (ALI) and low ALI in all stages combined; B: Comparison of OS between high ALI and low ALI in stage I PC patients; C: Comparison of OS between high ALI and low ALI in stage II PC patients; D: Comparison of OS between high ALI and low ALI in stage III PC patients. ALI: Advanced lung cancer inflammation index.

Cox analysis for OS

Age, tumor stage, tumor size, and ALI were significant factors in univariate analysis, thus Cox analysis was conducted using these factors. In Cox analysis, age [P < 0.01, hazard ratio (HR) = 1.015, 95% confidence interval (CI): 1.005-1.026], tumor stage (P < 0.01, HR = 1.240, 95%CI: 1.092-1.407), tumor size (P = 0.022, HR = 1.069, 95%CI: 1.010-1.131), and ALI (P = 0.018, HR = 0.784, 95%CI: 0.641-0.960) were identified to be independent prognostic factors (Table 4).

Table 4 Univariate and multivariate analysis of overall survival.

Risk factor	Univariate analysis		Multivariate analysis
	HR (95%CI)	P value	HR (95%CI)	P value
Age (years)	1.012 (1.001-1.023)	0.029	1.015 (1.005-1.026)	< 0.01
Sex (male/female)	0.844 (0.691-1.031)	0.097
Smoking (yes/no)	1.157 (0.919-1.456)	0.214
Drinking (yes/no)	1.026 (0.852-1.234)	0.789
Hypertension (yes/no)	1.128 (0.903-1.409)	0.289
Type 2 diabetes mellitus (yes/no)	1.186 (0.954-1.473)	0.123
Tumor location (body and tail/head and neck)	1.028 (0.843-1.252)	0.787
Tumor stage (III/II/I)	1.265 (1.122-1.426)	< 0.01	1.240 (1.092-1.407)	< 0.01
Tumor size (cm)	1.093 (1.036-1.154)	< 0.01	1.069 (1.010-1.131)	0.022
Advanced lung cancer inflammation index (high/Low)	0.765 (0.625-0.935)	< 0.01	0.784 (0.641-0.960)	0.018

HR: Hazard ratio.

DISCUSSION

A total of 611 patients who underwent radical PC surgery were included in this study. The low ALI group had significantly lower BMI, serum Alb, lymphocyte count, and ALI, but higher neutrophil count, a higher proportion of head and neck PC, and longer operation time. Although there was difference in baseline information, Cox analysis was conducted. As for prognosis, the low ALI group had worse OS in stage I patients, and ALI was identified as an independent risk factor for OS.

ALI has demonstrated predictive value in various diseases in previous studies and has shown strong prognostic significance in tumors[22-25]. Some studies have reported its association with CSS in patients with PC. However, there is a lack of more systematic research examining the precise relationship between ALI and PC[21].

The relationship between ALI and prognosis might be mediated through several biological mechanisms. In cancer patients, a prolonged systemic inflammatory response could impair the normal function of the immune system and disrupt tumor immune surveillance[26,27]. An increase in neutrophils with a decrease in lymphocytes typically indicated immune suppression, which might contribute to the worse prognosis observed in patients with a lower ALI. Furthermore, low serum Alb levels were indicative of poor nutritional status, which might influence PC progression through mechanisms such as immunosuppression and metabolic disturbances[28,29]. Chronic inflammation, immune evasion, and tumor-associated metabolic abnormalities were linked within PC microenvironment, and low ALI could play a potential role in these processes[30-32]. Although this study found a significant association between ALI and PC prognosis, the underlying exact mechanisms require further investigation. Future studies should explore the specific pathways and biomarkers via which ALI impacted PC progression, integrating laboratory findings with clinical data.

This study provided new evidence supporting the use of ALI as a prognostic tool for PC. Given its easy accessibility and low cost, ALI could serve as a simple and practical indicator to assist clinicians in predicting patient survival and the risk of disease progression, particularly in the treatment decision-making process. For patients with a low ALI, enhanced preoperative evaluation and postoperative follow-up should be implemented to enable timely interventions and adjustments in treatment strategies, ultimately improving the quality of life and prognosis. However, the clinical application of ALI still faces several challenges, and while ALI demonstrated strong prognostic value in this study, its applicability across different PC stages and treatment modalities remains uncertain. Future large-scale, multicenter prospective studies are necessary to validate our findings. Additionally, combining ALI with other biomarkers could further enhance the accuracy of prognostic predictions.

Overall, this study is the first to investigate the impact of ALI on OS in PC, with a relatively large number of patients. ALI is clinically significant as a comprehensive marker that effectively assesses both immune and nutritional status in patients. Our findings would provide more precise guidance for individualized treatment in PC. ALI might be an important supplementary indicator for other prognostic factors, and more comparisons are needed to verify whether ALI can replace other prognostic indicators.

However, there are some limitations to this study. First, the current study was conducted with a retrospective design at a single center, which may limit the broader applicability of the findings. Future multicenter, large-sample, prospective studies are necessary to validate the value of ALI in PC. Additionally, this study focused on the relationship between ALI and OS, and future research could explore the association between ALI and other clinical outcomes (e.g., postoperative recurrence and complications) to provide a more comprehensive assessment. Moreover, this study did not collect tumor histopathological characteristics and biomarkers such as cancer antigens-199, which also affect the prognosis of patients with PC. Since they were not included in the Cox regression analysis, the observed association may be confounded by these factors. Furthermore, OS was calculated from surgery to death or loss to follow-up, which may introduce bias. In addition, the ALI cut-off value (34.0) was derived using X-tile, which may not generalize to external cohorts. Thus, external validation is warranted.

CONCLUSION

For PC patients undergoing radical surgery, lower ALI levels independently predict worse OS, underscoring the imperative for surgical teams to closely monitor such cases preoperatively.