Preoperative C-reactive protein-augmented CONUT score as a better prognostic indicator than CONUT alone in non-small cell lung cancer across age groups

Introduction

Background

Preoperative nutritional status has been shown to correlate with postoperative outcomes across various types of cancer, and in recent years, its importance has also been increasingly recognized in the field of thoracic surgery (1-3). We previously reported that, among patients aged 80 years and older with non-small cell lung cancer (NSCLC), the Controlling Nutritional Status (CONUT) score was the most sensitive and useful prognostic indicator for overall survival (OS), outperforming other preoperative inflammation- and nutrition-based scores such as the Prognostic Nutritional Index (PNI), Geriatric Nutritional Risk Index (GNRI), C-reactive protein-to-lymphocyte ratio (CLR), C-reactive protein-to-albumin ratio (CAR), Hemoglobin, Albumin, Lymphocyte, and Platelet (HALP) score, neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), platelet-to-albumin ratio (PAR), and neutrophil-to-albumin ratio (NAR) (4).

Rationale and knowledge gap

The CONUT score is calculated based on three routinely available laboratory parameters: serum albumin concentration, total cholesterol level, and total lymphocyte count. Each parameter is scored on a point scale, and the total score ranges from 0 to 12, with higher scores indicating poorer nutritional status. Specifically, low albumin and cholesterol levels, as well as lymphopenia, contribute to a higher CONUT score, reflecting malnutrition and impaired immune function. Detailed scoring criteria are summarized in previous literature (4), but a brief summary is as follows: albumin (0–6 points), lymphocyte count (0–3 points), and total cholesterol (0–3 points).

However, it does not include C-reactive protein (CRP), a widely recognized inflammation marker often used as a prognostic indicator in cancer. During the course of our previous research, we found that the components of the CONUT score and CRP function as independent indicators of nutritional status. Based on these findings, we proposed a novel modified index, the combined CRP and CONUT (C-CONUT) score, which incorporates CRP into the original CONUT scoring system. We subsequently demonstrated that, among patients aged 80 years and older, the C-CONUT score showed greater sensitivity as a prognostic marker compared to the original CONUT score (5). Whether these nutritional scores also serve as reliable prognostic factors in younger patient populations remains unclear.

Objective

Therefore, the aim of the present study was to evaluate the utility of both the CONUT and C-CONUT scores as preoperative prognostic indicators in NSCLC patients across various age groups, with a particular focus on comparing the predictive performance of the two scoring systems. We present this article in accordance with the STROBE reporting checklist (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1623/rc).

Methods

Study design

The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Kochi Medical School (No. ERB-109931). Informed patient consent was not necessary because of the retrospective nature of our single-center study.

Patients

From January 2012 to December 2022, at the Kochi Medical School Thoracic Surgery Department, surgical resection was performed for 923 primary lung cancer patients. Exclusion criteria were patients who underwent neoadjuvant treatment, pathological stage 0 and IIIB, or had missing data that were necessary for the calculation of prognostic indexes. Ultimately, 738 patients were enrolled in the analysis, which included the preoperative measurement of their nutritional parameters, including total cholesterol, serum albumin, neutrophil, lymphocyte, and CRP values.

Subgroup analysis was performed by dividing the patients into three cohorts (Figure 1). Study flow diagram of patient selection and age-based subgroup classification. The three age cohorts (≤79, ≤75, and ≤65 years) are overlapping and represent nested subsets. Patients aged ≥80 years were excluded from the study.

Figure 1 Flow chart of patient selection. NSCLC, non-small cell lung cancer.

Data collection and follow-up

Patient data, such as age, gender, the body mass index (BMI), smoking history, surgical procedure, histological subtype, pathological stage, standard blood and biochemical measurements, lymphatic, vascular and pleural invasion, and Charlson Comorbidity Index (CCI), were retrospectively obtained from pathology reports, electronic patient records, and case notes. Postoperatively, patients underwent follow-ups with regular physical examinations, blood tests, and chest X-rays every 3 months during the first three years and then every 6 months afterwards. Computed tomography scans of the chest and abdomen were also performed at least once a year. Clinicopathological characteristics and OS were retrospectively analyzed.

New C-CONUT score and immune-nutritional indicators

The CONUT score was calculated based on preoperative albumin concentration, total cholesterol concentration and lymphocyte counts. The following formulas were used to calculate other immune-nutritional indicators:

CAR = CRP/albuminconcentration[1]

NLR=neutrophil/lymphocytecount(×109/L)[2]

C-CONUTscore=CRPscore+conventionalCONUTscore[3]

Further details regarding Eq. [3] can be found in reference (5).

The C-CONUT score was calculated by adding a CRP score to the conventional CONUT score. The CRP score was defined as follows: patients with a serum CRP level <0.1 mg/dL were assigned 0 points, and those with 0.1–0.99 mg/dL were assigned 1 point, and ≥1 mg/dL were assigned 2 points. This cutoff value was determined based on prior clinical studies and internal validation. The resulting C-CONUT score therefore ranges from 0 to 13 points, allowing for a composite evaluation of both nutritional status and systemic inflammation.

Statistical analysis

Data were summarized as medians with interquartile ranges or as counts with corresponding percentages. OS was defined as the time from the date of surgery to the date of death or last follow-up. To evaluate the predictive ability for 5-year OS, receiver operating characteristic (ROC) curves were constructed for each immunonutritional index. In the ROC analysis, 5-year survival status was treated as a binary outcome, and the optimal cutoff values and area under the curve (AUC) for each index were determined.

Survival analyses for clinical, pathological, and prognostic factors were performed using the Kaplan-Meier method, and differences between groups were compared using the log-rank test. To identify independent prognostic factors, multivariate analysis was conducted using a Cox proportional hazards regression model with a stepwise selection method. Variables with a P value of less than 0.05 in the univariate analysis were included in the multivariate model to avoid omitting clinically important factors. Associations between each independent variable and survival were expressed as hazard ratios (HRs) with 95% confidence intervals (CIs).

To assess the potential multicollinearity between the CONUT score and the C-CONUT score, variance inflation factors (VIFs) were calculated. Since moderate multicollinearity (VIF >3) was observed, both scores were not included simultaneously in the same multivariate model. Instead, separate Cox regression models were developed for the CONUT and C-CONUT scores, respectively, using the same covariates to adjust for confounding, and their prognostic performances were individually evaluated. To further address the potential non-linearity of age as a prognostic factor in cohort 3 (≤65 years), we conducted an additional univariate analysis using a cut-off value of 62 years, determined via ROC curve analysis. However, age remained non-significant in this alternative model and was therefore excluded from the final multivariate analysis.

All statistical analyses were performed using JMP Pro (version 12; SAS Institute Inc., Cary, NC, USA). A two-sided P value of less than 0.05 was considered statistically significant.

Results

Patient characteristics

Overall, 738 patients were analyzed. The median follow-up period of patients spanned 56 (range, 7–121) months. Their clinicopathological data is provided in Table 1. The median patient age was 73 (range, 31–92) years, 369 patients (50.0%) had smoked previously, and 450 patients (60.9%) were male. Furthermore, 258 patients (35.0%) underwent sublobar resections. Most patients (71.6%) had a diagnosis of adenocarcinoma. Five hundred fifty-two (74.8%), 111 (15.0%), and 75 (10.2%) patients were ultimately diagnosed with pathological I, II, and III, respectively.

ROC curve analysis and cutoff values

ROC curves were constructed to determine the optimal cutoff values for each prognostic indicator. The identified cutoffs were as follows: CONUT score ≥2, C-CONUT score ≥3, CAR ≥2.56, and NLR ≥2.83.

Among these, the C-CONUT score demonstrated the highest AUC (0.654; 95% CI: 0.653–0.781; P<0.001), outperforming the conventional CONUT score and other markers (Figure 2).

Figure 2 The ROC curve of various markers for predicting OS. AUC, area under the curve; C-CONUT, combined C-reactive protein and Controlling Nutritional Status; CAR, C-reactive protein-to-albumin ratio; CONUT, Controlling Nutritional Status; NLR, neutrophil-to-lymphocyte ratio; OS, overall survival; ROC, receiver operating characteristic.

Kaplan-Meier survival analysis

Patients were stratified into low and high groups based on the predefined cutoffs for CONUT and C-CONUT scores.

CONUT score

The high CONUT group (score ≥2; n=305) showed significantly poorer OS compared to the low CONUT group (score 0–1; n=433). The 5-year OS was 69.8% in the high group vs. 81.0% in the low group (P<0.001; Figure 3A). This prognostic significance was maintained in the ≤79 years cohort (P=0.003; Figure 3B) and in the ≤75 years cohort (P=0.02; Figure 3C), but not in the ≤65 years cohort (P=0.85; Figure 3D).

Figure 3 OS between high and low CONUT score group. (A) Overall (n=738); (B) age less than 79 years (cohort 1: n=581); (C) age less than 75 years (cohort 2: n=449); (D) age less than 65 years (cohort 3: n=150). CONUT, Controlling Nutritional Status; OS, overall survival.

C-CONUT score

The high C-CONUT group (score ≥3) consistently exhibited worse OS across all age groups, including the ≤65 years cohorts, where the CONUT score did not retain significance (Figure 4). This suggests that the C-CONUT score provides more robust prognostic discrimination in younger populations.

Figure 4 OS between high and low and high C-CONUT score group. (A) Overall (n=738); (B) age less than 79 years (cohort 1: n=581); (C) age less than 75 years (cohort 2: n=449); (D) age less than 65 years (cohort 3: n=150). C-CONUT, combined C-reactive protein and Controlling Nutritional Status; OS, overall survival.

Multivariate prognostic analysis

In the overall cohort, multivariate Cox regression analysis identified the following as independent prognostic factors for OS: age (P<0.001), gender (P<0.001), pathological stage (P=0.009), lymphatic invasion (P=0.01), pleural invasion (P=0.003), CONUT score (P=0.01) and C-CONUT score (HR =1.71; 95% CI: 1.15–3.12; P=0.003) (Table 2).

In age-specific analyses

Cohort 2 (≤75 years): the CONUT score was not a significant factor (P=0.07), whereas the C-CONUT score remained significant (P=0.005; Table 3).

Cohort 3 (≤65 years): similarly, only the C-CONUT score showed a significant association with OS (P=0.044; Table 4).

These findings further support the superior prognostic value of the C-CONUT score over CONUT alone in younger NSCLC patients. During multivariate modeling, we noted that the inclusion of both CONUT and C-CONUT scores led to moderate multicollinearity, as assessed by VIF analysis. To avoid distortion of the model estimates, we performed separate multivariate analyses for each score. This approach confirmed the superior prognostic performance of the C-CONUT score over the CONUT score, particularly in younger cohorts.

Discussion

Malnutrition is widely recognized as a significant prognostic factor in patients with malignant tumors, and the importance of early nutritional screening and timely intervention has been highlighted in numerous studies (6,7). However, in clinical practice, there remains a need for simple, reproducible, and sensitive nutritional assessment tools (8). In our previous study, we demonstrated that the C-CONUT score—an adaptation of the traditional CONUT score incorporating CRP, a marker of systemic inflammation—was a useful prognostic indicator in elderly patients (≥80 years) with lung cancer (4,5).

In the present study, we evaluated the prognostic value of both CONUT and C-CONUT scores in a relatively younger cohort of lung cancer patients (≤75 years). Our findings revealed that, unlike in the elderly cohort, the conventional CONUT score alone was not significantly associated with prognosis in this younger population. In contrast, the C-CONUT score emerged as an independent prognostic factor, suggesting the importance of age-specific application of nutritional indices. Younger patients tend to have greater physiological reserve than older individuals, which may allow for better recovery from transient preoperative nutritional deficits. Nevertheless, the C-CONUT score—which reflects both nutritional status and systemic inflammation—may more sensitively detect latent risks even in this relatively robust patient group.

The CONUT score is based on serum albumin levels, total cholesterol, and lymphocyte count, providing an integrated assessment of immunonutritional status. However, these components can be influenced by non-nutritional factors such as liver dysfunction, infection, or systemic inflammation (9). CRP, on the other hand, is a well-established biomarker of inflammation and has been associated with tumor-related cytokine activity and poor prognosis (10). By combining these parameters, the C-CONUT score offers a more balanced risk assessment, potentially overcoming the limitations of individual components.

Of particular interest in this study is the significant association observed between the C-CONUT score and OS in patients aged ≤75 years. This age group is less affected by frailty or cognitive decline and generally exhibits higher compliance with preoperative nutritional guidance, thereby enhancing the potential benefit of early risk stratification and clinical intervention. In fact, in gastrointestinal cancers, preoperative nutritional support has been shown to reduce postoperative complications, even in patients without overt malnutrition (11). Future prospective studies are warranted to determine whether interventions targeting preoperative nutritional and inflammatory status can improve outcomes in high-risk patients identified by elevated C-CONUT scores. Notably, in advanced lung cancer, oral nutritional supplementation has been suggested to reduce CRP levels and inflammatory cytokines, potentially contributing to improved progression-free survival (PFS) (12).

In the present study, the C-CONUT score demonstrated moderate discriminative ability, with an AUC of 0.654. Although this may not represent a high level of accuracy, the ease of calculation, non-invasive nature, and lack of additional cost render it a practical and clinically meaningful tool. Furthermore, the fact that the C-CONUT score was an independent prognostic factor in younger patients—whereas the CONUT score was not—suggests that it may reflect latent risk factors not adequately captured by traditional nutritional assessments alone. Nonetheless, the moderate AUC indicates that the C-CONUT score should not be used in isolation for definitive clinical decision-making. Rather, it should be integrated into a comprehensive risk assessment model that includes clinical, pathological, and radiological parameters. Additional prospective studies are needed to further validate the additive prognostic value of this score and to evaluate the effectiveness of targeted interventions based on its stratification.

It is important to note that the C-CONUT score in this study was calculated by simply adding the CRP score and the CONUT score with equal weighting (1:1), and this weighting has not been statistically validated. This represents a key limitation, and future studies should aim to optimize the weighting of individual components to enhance the score’s prognostic accuracy.

This study also has several limitations. It was a single-center retrospective analysis with a relatively small sample size and potential variability in blood sampling timing. Moreover, the study population consisted exclusively of Japanese patients, which may limit the generalizability of the findings. External validation through multicenter, multi-ethnic prospective studies is needed, considering potential differences in genetic background and healthcare systems. The C-CONUT score represents a simple, non-invasive, and cost-effective tool that can be readily calculated from routine laboratory tests. Our findings suggest that it may be valuable for early risk stratification and perioperative management in patients with lung cancer. Further large-scale prospective studies are warranted to clarify its clinical utility and to establish its role in guiding nutritional and inflammatory interventions.

Clinically, the C-CONUT score, easily calculated from routine preoperative labs, may assist in personalized perioperative management. High-risk patients could be considered for closer monitoring, such as ICU admission or early nutritional intervention. Postoperatively, these patients may benefit from more frequent imaging or shorter follow-up intervals. Moreover, a high C-CONUT score could help identify candidates for adjuvant therapy, even among patients with similar pathological stages. These potential applications warrant further prospective validation.

Conclusions

In conclusion, this exploratory study suggests that the C-CONUT score may provide better prognostic prediction than the conventional CONUT score in relatively younger patients with NSCLC aged 75 years or younger. However, as this study was retrospective and conducted at a single institution, the findings should be interpreted with caution and require validation in future prospective multicenter studies.

Given its ease of calculation from routine laboratory data, the C-CONUT score serves as a practical and non-invasive tool for risk assessment in clinical settings. Further prospective studies are warranted to determine whether interventions based on C-CONUT stratification can improve patient prognosis.

Acknowledgments

None.

Footnote

Reporting Checklist: The authors have completed the STROBE reporting checklist. Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1623/rc

Data Sharing Statement: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1623/dss

Peer Review File: Available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1623/prf

Funding: None.

Conflicts of Interest: All authors have completed the ICMJE uniform disclosure form (available at https://jtd.amegroups.com/article/view/10.21037/jtd-2025-1623/coif). The authors have no conflicts of interest to declare.

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the Ethics Committee of Kochi Medical School (No. ERB-109931) and individual consent for this retrospective analysis was waived.

Open Access Statement: This is an Open Access article distributed in accordance with the Creative Commons Attribution-NonCommercial-NoDerivs 4.0 International License (CC BY-NC-ND 4.0), which permits the non-commercial replication and distribution of the article with the strict proviso that no changes or edits are made and the original work is properly cited (including links to both the formal publication through the relevant DOI and the license). See: https://creativecommons.org/licenses/by-nc-nd/4.0/.

References

1. Takahashi M, Sowa T, Tokumasu H, et al. Comparison of three nutritional scoring systems for outcomes after complete resection of non-small cell lung cancer. J Thorac Cardiovasc Surg 2021;162:1257-1268.e3. [Crossref] [PubMed]
2. Okada S, Shimada J, Kato D, et al. Clinical Significance of Prognostic Nutritional Index After Surgical Treatment in Lung Cancer. Ann Thorac Surg 2017;104:296-302. [Crossref] [PubMed]
3. Chen Y, Cao H, Yao J. Controlling nutritional status score is a predictor of survival in hepatocellular carcinoma: A meta-analysis and meta-regression. Pak J Med Sci 2025;41:1226-33. [Crossref] [PubMed]
4. Tamura M, Sakai T, Furukawa N, et al. Prognostic Significance of CONUT Score in Elderly NSCLC. Ann Thorac Cardiovasc Surg 2024;30:24-00009. [Crossref] [PubMed]
5. Miyazaki R, Tamura M, Sakai T, et al. Using the combined C-reactive protein and controlling nutritional status index for elderly non-small cell lung cancer. J Thorac Dis 2024;16:4400-8. [Crossref] [PubMed]
6. Shoji F. Clinical impact of preoperative immunonutritional status in patients undergoing surgical resection of lung cancer. J Thorac Dis 2019;11:S408-12. [Crossref] [PubMed]
7. Xie T, Dong Z, Wu C, et al. Association between CONUT scores and survival outcomes in patients with non-small cell lung cancer: meta-analysis from 4973 Asian cases. Front Oncol 2025;15:1522368. [Crossref] [PubMed]
8. Liu H, Yang XC, Liu DC, et al. Clinical significance of the controlling nutritional status (CONUT) score in gastric cancer patients: A meta-analysis of 9,764 participants. Front Nutr 2023;10:1156006. [Crossref] [PubMed]
9. Xie H, Nong C, Yuan G, et al. The value of preoperative controlling nutritional status score in evaluating short-term and long-term outcomes of patients with colorectal cancer following surgical resection. J Cancer 2020;11:7045-56. [Crossref] [PubMed]
10. Koch A, Fohlin H, Sörenson S. Prognostic significance of C-reactive protein and smoking in patients with advanced non-small cell lung cancer treated with first-line palliative chemotherapy. J Thorac Oncol 2009;4:326-32. [Crossref] [PubMed]
11. Kabata P, Jastrzębski T, Kąkol M, et al. Preoperative nutritional support in cancer patients with no clinical signs of malnutrition--prospective randomized controlled trial. Support Care Cancer 2015;23:365-370. [Crossref] [PubMed]
12. Sánchez-Lara K, Turcott JG, Juárez-Hernández E, et al. Effects of an oral nutritional supplement containing eicosapentaenoic acid on nutritional and clinical outcomes in patients with advanced non-small cell lung cancer: randomised trial. Clin Nutr 2014;33:1017-23. [Crossref] [PubMed]