Prognostic Significance of Inflammation-based Scores in Pancreatic Cancer Patients Treated With Palliative Chemotherapy: A Single Institution Experience

Abstract

Background/Aim

Inflammation-based prognostic scores have shown prognostic significance and have been associated with clinical outcomes in various types of cancer. Inflammation is known to promote tumor progression leading to reduced survival. In pancreatic cancer, systemic inflammation is common and contributes to its dismal prognosis. Although the prognosis of pancreatic cancer is improving with the introduction of new drugs, the prognostic indicators are still poorly understood. The present study aimed to evaluate inflammation-based prognostic scores in patients with metastatic pancreatic cancer receiving first-line chemotherapy.

Patients and Methods

A total of 43 patients with metastatic pancreatic cancer undergoing first-line chemotherapy (gemcitabine+nab-paclitaxel and mFOLFIRINOX) in our institution were analyzed. Baseline clinicopathological and pre-treatment laboratory data were collected. Survival was estimated using the Kaplan-Meier method and survival differences were evaluated using the log-rank test.

Results

In the whole cohort, we identified lymphocyte-to-monocyte ratio ≥3, systemic inflammatory response index <2.3, carcinoembryonic antigen <2.5, neutrophil-to-lymphocyte ratio <5, Memorial Sloane Kettering score <2, and prognostic index <2 as prognostic markers associated with improved overall survival in patients receiving first-line chemotherapy.

Conclusion

The current analysis showed an association between inflammatory-based prognostic markers and overall survival in patients with metastatic pancreatic cancer treated in a real-world setting at a single institution.

Pancreatic carcinoma (PC) is an aggressive cancer with gradually increasing incidence and mortality. The highest incidence is observed in Europe, Northern America, Australia, and New Zealand. In general, PC is relatively uncommon compared to other cancers but has a high mortality rate. PC is currently the 7-th most common cause of cancer death worldwide (1). PC is the 4-th most lethal cancer in Europe and the 3-rd in Northern America; with a 5-year survival of 10%, the prognosis of this malignancy remains dismal (2). The incidence and mortality of PC are expected to increase by 25% by 2025, which will place PC as the 2nd most common cause of death among malignancies in 2040 (3).

Systemic chemotherapy has long been the standard treatment for pancreatic cancer at all stages, notwithstanding its limited efficacy compared to treatments for other malignancies. Despite recent improvements in survival rates with the introduction of modern chemotherapy regimens, the prognosis of PC remains dismal. Gemcitabine monotherapy is also an option nowadays in older patients, or in patients with comorbidities (4). Currently, more active regimens with a higher response rate are preferred, namely gemcitabin + nab-paclitaxel (GnP) (5) or folinic acid (leucovorin)-fluorouracil-irinotecan-oxaliplatin (mFOLFIRINOX) (6). Due to the higher incidence of grade 3 toxicities, they are reserved mainly for younger patients in a good performance status. As an alternative, 5-fluorouracil, leucovorin, liposomal irinotecan and oxaliplatin (NALIRIFOX) can be considered as a 1st line regimen in metastatic PC (7). Nevertheless, the median overall survival (mOS) observed with NALIRIFOX was the same as that with the mFOLFIRINOX regimen (noting the limitations of cross-trial comparisons), despite significantly higher economic toxicity associated with NALIRIFOX.

Nevertheless, in daily clinical practice, we observe high variability in the outcomes of PC patients treated with these standard-of-care regimens. Therefore, it is important to identify prognostic and/or predictive markers that could help us better predict patient survival and tailor future therapies. However, we still lack prognostic and predictive markers that can identify patients who would benefit from intensified treatment, those who will rapidly progress despite an aggressive approach, and those who need early symptomatic treatment. At present, both upfront radiological staging and resectability status are used to assist with prognostication; however, biological markers in this area are lacking. Cancer antigen 19-9 (Ca 19-9) is currently used as a marker to assess response to therapy; however, its use as a prognostic biomarker has not been validated.

As observed in many cancer types, chronic inflammatory states also promote the development of pancreatic neoplasms (8). Moreover, one of the hallmarks of cancer is its ability to induce an inflammatory response in the body. This inflammation is driven by various factors, including the release of cytokines and other immune system modulators by cancer cells (9,10). Elevated levels of these markers are often associated with a more aggressive disease course, poorer response to treatment, and decreased OS in cancer patients.

The routine indicators of the systemic inflammatory response are circulating white blood cells and acute-phase proteins. It has been reported that white blood cell counts, which include neutrophil, lymphocyte, and monocyte counts, and levels of acute-phase proteins, such as C-reactive protein, have prognostic value in many types of cancer, including PC (11). Because of this association, numerous inflammation-based scores have emerged as prognostic tools in PC and various other malignancies over the past few decades (12,13).

In general, inflammation-based scores are inexpensive and easy-to-measure parameters used to develop prognostic scores, indexes and nomograms, which can be used to better stratify patients and identify those with either a good or poor prognosis. The majority of these prognostic scores evaluate various common laboratory parameters, such as blood counts and specific biochemical parameters. They often assess the impact of the systemic inflammatory response, which is common in cancer, on different treatment outcomes (e.g., mOS).

The aim of the current study was to evaluate the prognostic impact of a wide set of already established circulating biomarkers of inflammatory response in a cohort of metastatic PC patients treated with palliative chemotherapy (GnP and mFOLFIRINOX) in a real-world setting at a single institution. We believe that our findings contribute to the growing body of evidence supporting these scores as vital prognostic indicators in the treatment of PC patients in daily clinical practice.

Patients and Methods

Study design. From June 2019 to April 2023, 110 patients with diagnosed PC at the University Hospital Kralovske Vinohrady were prospectively enrolled into this study. After excluding 54 non-metastatic patients, 56 patients with PC who received first-line chemotherapy were further analyzed. The presence of metastatic disease was evaluated in all enrolled patients by means of contrast-enhanced computed tomography scan within one month before initiation of systemic treatment. All investigated laboratory parameters were assessed before the initiation of systemic chemotherapy (within a maximum of a 14-day window) and recorded in the institutional electronic medical report. The choice of systemic treatment was selected based on current guidelines, reflecting patients’ performance status, comorbidities, and preference. Treatment continued until unacceptable toxicity, progression of disease, or patient-initiated termination. Follow-up visits were routinely assessed and documented in the electronic medical record.

All procedures performed in this study were in accordance with the ethical standards of the institutional research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards. The study was approved by the Ethics Committee of the University Hospital Kralovske Vinohrady, Prague (EK-VP/30/0/2021).

Treatment methods and assessments. GnP was administered as follows: gemcitabine 1,000 mg/m² and nab-paclitaxel 125 mg/m² on days 1, 8, and 15 every four weeks. mFOLFIRINOX was administered as follows: oxaliplatin 85 mg/m², leucovorin 200 mg/m², irinotecan 150 mg/m², and continuous intravenous 2,400 mg/m² 5-fluorouracil for 46 h every two weeks. Capecitabine was administered as follows: 1,000 mg/m² bid (day 1 till day 14) every three weeks. Doxorubicin was administered as follows: 75 mg/m² every three weeks.

Tumor staging was performed according to the American Joint Committee on Cancer (AJCC)/International Union against Cancer (UICC) TNM classification (8th Edition). Clinical staging was routinely based on contrast-enhanced computed tomography before initiation of palliative chemotherapy and subsequently every ±12 weeks.

OS was calculated from the date of administration of the first line palliative chemotherapy to death or censored at the time of the final data cut-off if patients were still alive.

Definition of investigated inflammatory markers and other parameters. Table I summarizes systemic inflammation-based prognostic scores and ratios analyzed in the current study. The cut-off values for all investigated parameters were adopted based on previously reported studies. Subsequently, patients were divided into corresponding groups according to each of the inflammatory markers.

Table I. Overview of systemic inflammation-based prognostic scores, ratios, and values investigated in the current study.

Statistical analysis. Data were collected by clinicians with expertise in clinical research under the supervision of the statistician and then centrally managed. Survival was estimated using the Kaplan-Meier method, with comparisons of the probability of survival performed using the log-rank test. Log-rank test was applied to estimate the corresponding hazard ratios (HRs). HRs are expressed with 95% confidence interval (Cis). Univariate analyses were performed by using Student’s t-test. p-Values <0.05 were considered statistically significant.

Results

Patient characteristics. Between June 2019 and April 2023, 110 patients with ECOG PS 0-2, diagnosed and treated at the Department of Oncology, University Hospital Kralovske Vinohrady were enrolled in the current study. After excluding 54 non-metastatic patients, 56 patients with metastatic disease were further analyzed. Baseline clinical characteristics of the analyzed patients are summarized in Table II.

Table II. Baseline clinical characteristics of the metastatic pancreatic patients.

WHO: World Health Organization; CEA: carcinoembryonic antigen; Ca19-9: carbohydrate antigen 19-9; 5-FU: 5 fluorouracil. ¹In four patients, we were not able to identify dysplastic or malignant cells in the biopsy due to the very limited sample size, nevertheless, these patients were considered as having metastatic pancreatic cancer due to elevation of markers (CEA, Ca 19-9) and confirmation of metastatic disease through contrast-enhanced computed tomography scan. ²Grade cannot be assessed due to the very limited sample size of the biopsy specimen. ³One patient with BRCA2 mutation was treated with olaparib and one patient with leiomyosarcoma was treated with pazopanib. ⁴One patient with leiomyosarcoma was treated with gemcitabine+docetaxel and one patient with adenocarcinoma was treated with cisplatin (after 1^st line mFOLFIRINOX and 2^nd line gemcitabine+nab-paclitaxel).

The median age of patients was 65 years (range=41-82 years). In total, 41 (73.2%) patients were women. The ECOG performance status (PS) was 0, 1, and 2 in 20 (35.7%), 23 (41.4%), and 13 (23.2%) patients, respectively. The majority of patients, 46 (82.1%), were diagnosed with an invasive ductal adenocarcinoma. We also observed and enrolled other histological subtypes of PC. One (1.8%) patient was diagnosed with adenosquamous carcinoma, one (1.8%) with anaplastic carcinoma, and one (1.8%) with leiomyosarcoma. In total, 3 (5.4%) patients were diagnosed with Pan I-III. These patients were also included in the current analysis and considered as metastatic PC patients due to elevation of markers CEA and Ca 19-9, and confirmation of metastatic disease through contrast-enhanced computed tomography scan. Finally, in 4 (7.1%) patients, we were unable to identify dysplastic or malignant cells in the biopsy due to the very limited sample size. Nevertheless, these patients were also included in the current analysis using the same criteria as in patients with histologically proven PanIN I-III. Distribution of grade in our cohort was as follows: grade 1 was observed in 1 (1.8%), grade 2 in 16 (28.6%), grade 3 in 18 (32.1%) and grade 4 in 1 (1.8%) patient. In total, in 20 (35.7%) patients, grade could not be assessed due to a very limited sample size.

All 56 (100%) enrolled and analyzed patients in the current study were confirmed to have metastatic disease via contrast-enhanced computed tomography scan. The most prevalent metastatic site in our cohort was the liver (42 patients, 75%), followed by the peritoneum (10 patients, 18%), non-regional lymph nodes (9 patients, 16%), and lungs (5 patients, 9%). In total, seven (13%) patients were diagnosed with metastatic disease in other areas (e.g., spleen, skin, adrenal gland, bones).

First-line systemic treatment was administered in all 56 (100%) patients. Eleven (19.6%) patients were treated with gemcitabine monotherapy, 27 (48.2%) patients with GnP, and 16 (28.6%) patients with mFOLFIRINOX. One (1.8%) patient was treated initially with doxorubicin monotherapy (histologically confirmed leiomyosarcoma of pancreas) and one (1.8%) patient was treated with capecitabine (patients’ preference). After progression on first-line treatment, 22 (39.3%) patients received second-line treatment (mainly gemcitabine- and 5-FU-based regimens). Third-line treatment was used in a minority of patients. In total, only two (3.6%) patients received third-line treatment (gemcitabine+docetaxel in a patient with leiomyo-sarcoma; cisplatin+gemcitabine in a patient after progression on mFOLFIRINOX and GnP). Median CEA level in our cohort was 7.7 μg/l (range=0.5-909.6 μg/l); median Ca 19-9 level was 1,629.2 kU/l (range=1.2-70,000 kU/l).

By the end of the follow-up period on December 11, 2023, 51 (91.1%) out of 56 patients died. All died from PC. The mOS was 11.4 months (95%CI=8.0-14.7) for the mFOLFIRINOX group and 8.8 months (95%CI=6.3-11.4) for the GnP group, (HR=0.594, 95%CI=0.316-1.114, p=0.165) (Figure 1).

Figure 1.

Kaplan-Meier (KM) curves in patients treated with first line chemotherapy [gemcitabine+nab-paclitaxel (GnP) and mFOLFIRINOX]. mOS: Median overall survival; CI: confidence interval.

Circulating biomarkers as prognostic markers in patients receiving GnP and mFOLFIRINOX. In the current study, we evaluated the prognostic significance of inflammatory-based parameters with respect to mOS in patients treated with GnP and mFOLFIRINOX, which are currently considered as standard treatment options in newly diagnosed metastatic PC. This analysis included 43 patients. We assessed the following parameters: C-reactive protein (CRP), C-reactive protein-to-albumin ratio (CAR), neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), lymphocyte-to-monocyte ratio (LMR), systemic inflammatory response index (SIRI), prognostic nutritional index (PNI), prognostic index (PI), Glasgow prognostic score (GPS), modified Glasgow prognostic score (mGPS), Memorial Sloane Kettering center score (MPS), albumin, albumin-bilirubin score (ALBI), carcinoembryonic antigen (CEA), and carbohydrate antigen 19-9 (Ca19-9). All parameters were assessed before the administration of the first line palliative treatment.

In this cohort, we identified six parameters that were significantly associated with mOS, namely CEA, LMR, NLR, SIRI, MPS, and PI. Results of all investigated parameters in this cohort are summarized in Table III and Table IV.

Table III. Prognostic significance of investigated inflammatory-based parameters in patients treated with chemotherapy (GnP and mFOLFIRINOX) based on cut-off levels.

CRP: C-reactive protein; CAR: C-reactive protein-to-albumin ratio; NLR: neutrophil-to-lymphocyte ratio; PLR: platelet-to-lymphocyte ratio; LMR: lymphocyte-to-monocyte ratio; SIRI: systemic inflammatory response index; PNI: prognostic nutritional index; CEA: carcinoembryonic antigen; Ca19-9: carbohydrate antigen 19-9. Statistically significant p-values are shown in bold.

Table IV. Prognostic significance of investigated inflammatory-based parameters in patients treated with chemotherapy (GnP and mFOLFIRINOX) based on score levels.

mGPS: Modified Glasgow prognostic score; PI: prognostic index; MPS: Memorial Sloane Kettering center score; ALBI: albumin, albumin-bilirubin score; GPS: Glasgow prognostic score. Statistically significant p-values are shown in bold.

Specifically, patients with a CEA <2.5 μg/l had a mOS of 14.1 (95%CI=10.5-17.8) months compared to patients with CEA ≥2.5 μg/l of only 8.8 (95%CI=6.7-11.0) months (HR=1.992, 95%CI=1.062-3.7436, p=0.004). Patients with a LMR ≥3 had a mOS of 12.1 (95%CI=8.9-15.3) months compared to patients with LMR <3 of only 8.3 (95%CI=6.5-10.1) months (HR=2.357, 95%CI=1.223-4.543, p=0.001). In addition, patients with NLR <5 had a mOS of 11.3 (95%CI=8.7-13.9) months compared to the patients with NLR ≥5 of only 7.5 (95%CI=5.0-9.9) months (HR=1.813, 95%CI=0.846-3.886, p=0.012). Moreover, patients with SIRI <2.3 had a mOS of 11.7 (95%CI=8.9-14.4) months compared to the patients with SIRI ≥2.3 of only 7.1 (95%CI=5.1-9.1) months (HR=2.638, 95%CI=1.295-5.373, p<0.001). Next, patients with MPS 0/1 had a mOS of 11.3 (95%CI=8.3-14.4) months compared to the patients with MPS 2 of only 1.2 (95%CI=-1.2-3.5) months, HR=4.288 (95%CI=1.006-18.272), p<0.001. Finally, patients with PI 0/1 had a mOS of 11.1 (95%CI=7.7-14.5) months compared to the patients with PI 2 of only 6.0 (95%CI=3.9-8.1) months (HR=1.229, 95%CI=1.155-10.114, p=0.001) (Figure 2A-F, Table V, and Table VI).

Figure 2.

Kaplan-Meier curves for overall survival of patients treated with chemotherapy [gemcitabine+nab-paclitaxel (GnP) and mFOLFIRINOX] based on baseline A) lymphocyte-to-monocyte ratio (LMR), B) systemic inflammation response index (SIRI), C) carcinoembryonic antigen (CEA), D) neutrophil-to-lymphocyte ratio (NLR), E) Memorial Sloane Kettering prognostic score (MPS), and F) prognostic index (PI). HR: Hazard ratio; CI: confidence interval.

Table V. Distribution of inflammation-based prognostic parameters < or ≥cut-off in patients treated with chemotherapy (GnP and mFOLFIRINOX).

LMR: Lymphocyte-to-monocyte ratio; SIRI: systemic inflammatory response index; CEA: carcinoembryonic antigen; NLR: neutrophil-to-lymphocyte ratio; MPS: Memorial Sloane Kettering center score; PI: prognostic index.

Table VI. Statistically significant inflammation-based prognostic parameters in patients treated with chemotherapy (GnP and mFOLFIRINOX) and corresponding HR.

LMR: Lymphocyte-to-monocyte ratio; SIRI: systemic inflammatory response index; CEA: carcinoembryonic antigen; NLR: neutrophil-to-lymphocyte ratio; MPS: Memorial Sloane Kettering center score; PI: prognostic index.

Then, we hypothesized, that the combination of statistically significant prognostic parameters (CEA, LMR, NLR, SIRI, MPS, and PI) could more effectively stratify patients in different prognostic risk groups. We proposed the „pancreatic score“, where for each positive marker, one point was allocated (CEA ≥2.5, LMR <3, NLR ≥5, SIRI ≥2.3, MPS >0/1 and PI >0/1).

First, we assigned patients into groups (0, 1, 2, 3, 4, 5, and 6 points) and correlated the different groups with the median OS. Patients without any positive marker had the best prognosis leading to a mOS of 19.6 months (95%CI=15.3-23.9). On the other hand, patients with all six identified risk factors had markedly worse prognosis with a mOS of 1.3 months (95%CI=-1.4-3.9). Based on the observed mOS, we grouped patients with 0-4 risk parameters together and compared them to the patients having 5-6 parameters. Patients with 0-4 risk parameters had a significantly longer mOS of 11.9 (95%CI=8.0-15.7) months compared to those with 5-6 risk parameters, who had an OS of 7.5 (95%CI=5.4-9.5) months (HR=4.214, 95%CI=1.286-13.81, p<0.001) (Figure 3, Table VII).

Figure 3.

Kaplan-Meier curve for overall survival of patients treated with chemotherapy [gemcitabine+nab-paclitaxel (GnP) and mFOLFIRINOX] based on ‘pancreatic score’. SIS: Systemic inflammation score.

Table VII. Median overall survival of patients based on the number of risk factors in “pancreatic score”.

SIS: Systemic inflammation score.

Discussion

Metastatic PC remains a major health problem with significant unmet need. Despite improvements in systemic treatment in recent years, the vast majority of patients diagnosed with metastatic disease will inevitably die within one year from diagnosis. The majority of PC patients are nowadays treated with non-specific chemotherapy regimens. From daily clinical practice, it is well known, that prognosis of certain patients may differ substantially even when treated with the same regimen and having more or less the same extent of disease. Moreover, currently we are lacking more precise prognostic markers, which could guide us in selecting appropriate treatments for individual patients.

Bearing this in mind, we investigated the value of easily obtainable blood-based parameters on the prognosis of patients with metastatic PC at the start of first-line palliative chemotherapy in a real-world setting treated at a single institution. We observed similar OS results in our cohort compared to the results reported in large phase III registration studies of GnP and mFOLFIRINOX. In patients treated with first-line GnP, we observed a mOS of 8.8 months, which is nearly similar to the mOS observed in the MPACT trial (5), namely 8.5 months. Moreover, a smaller proportion of patients treated with first-line GnP in our study received subsequent second-line treatment when compared to the MPACT trial (33.3% vs. 38%) (5). These observations allow us to hypothesize, that despite the lower number of patients receiving second line treatment in our study, the nearly identical mOS benefit compared to the MPACT study may be due to a higher proportion of patients presenting with “good” inflammation-based prognostic scores at baseline.

Patients treated with first-line mFOLFIRINOX in our study achieved a mOS of 11.4 months, which is again very comparable to the 11.1 months reported in the PRODIGE/ACCORD 11 trial (6). When looking at the second-line treatment after progression on mFOLFIRINOX, a high proportion of patients received second-line treatment in our cohort compared to the PRODIGE/ACCORD 11 trial (6) (62.5% vs. 46.8%). Nevertheless, we acknowledge the small number of patients treated with mFOLFIRINOX in our study, which could play a role in this discrepancy.

In addition, among the entire study cohort (n=56), we observed that only 39.3% of patients received further line therapy after progression on first-line treatment. This numbers are in line with other published real-world studies of patients with metastatic PC, namely 38.3% (14) and 46% (15).

Interestingly, in our analysis, we did not find a statistically significant difference in mOS when analyzing patients receiving first-line mFOLFIRINOX or GnP. The mOS was 11.4 months (95%CI=8.0-14.7) for the mFOLFIRINOX group and 8.8 months (95%CI=6.3-11.4) for the GnP group (HR=0.594, 95%CI=0.316-1.114, p=0.165). Nevertheless, similar efficacy between these two regimens has been reported in real-world settings (16,17), supporting their use in the first-line setting based on similar mOS results, despite their different toxicity profiles.

In PC, several inflammatory-based prognostic markers have been studied to assess the severity of the disease, predict patient outcomes, and guide treatment decisions. These markers reflect the host’s antitumor inflammation capacity and are of prognostic value in PC patients (18).

We were able to confirm the prognostic significance of LMR on survival in a cohort of patients treated with GnP and mFOLFIRNOX. The definitive mechanism linking pretreatment LMR to survival outcomes remains unclear. There is no clear cut-off value for LMR, as shown by Lin et al. in a meta-analysis of 3,338 patients with PC (19). The cut-off varied and ranged from 2.05 to 4.6. They revealed, that high LMR was significantly associated with better OS in PC patients irrespective of the cut-off value. In general, lymphocytes play a critical part in cellular immunity, especially tumor-infiltrating lymphocytes (TIL). On the other hand, monocytes can potentially differentiate into tumor-associated macrophages (TAM) promoting angiogenesis, extracellular matrix remodeling, tumor invasion, and metastasis (9,10). Taken together, the potential of LMR as a prognostic biomarker in PC may be demonstrated by a high lymphocyte count and low monocyte count, which may reflect systemic inflammation that inhibits tumor progression and migration. Prognostic significance of LMR on survival was confirmed also by Takeuchi et al. in a study of 32 patients who underwent curative surgery for pancreatic head cancer. Using a cut-off value of LMR </>3.0 they confirmed, that low LMR (<3.0) was associated with a decreased overall survival and that preoperative LMR was an independent prognostic factor in patients with pancreatic head cancer (20).

SIRI, which integrates clinically available neutrophil, monocyte, and lymphocyte counts was initially introduced by Qi et al. (21) and validated on a cohort of gemcitabine-based treated patients. Patients with SIRI <cut-off had significantly longer OS compared to patients with SIRI >cut-off (21). At the time Qi et al. published their results, GnP and mFOLFIRINOX had not yet been widely adopted worldwide. Validation of prognostic significance of SIRI in a cohort of patients treated with modern regimens was confirmed by Pacheco-Barcia et al. (22). They demonstrated that elevated SIRI was an independent prognostic factor for patients with metastatic PC. Our results are in line with previously published studies.

Elevated tumor markers in general are associated with worse prognosis, irrespective of malignancy. However, although it has been demonstrated that Ca 19-9 is useful for predicting prognosis in patients with PC, little is known about the implications of CEA levels as a prognostic marker for these patients. We identified elevated CEA level as a negative prognostic parameter in a cohort of patients treated with GnP and mFOLFIRINOX. There are several studies confirming this results. Imaoka et al. evaluated the prognostic impact of CEA in a cohort of 433 patients with metastatic disease. They showed that elevated CEA (>5 μg/l) was associated with a significantly shorter mOS compared to normal CEA levels (<5 μg/l) (23). The prognostic value of CEA levels was confirmed also in a systematic review and meta-analysis by Meng et al. (24). They evaluated 3,650 participants enrolled in 19 studies and found that elevated CEA levels (a cut-off of 5 was used in the majority of the studies) were significantly associated with worse OS (HR=1.43, 95%CI=1.31-1.56). Therefore, measurement of serum CEA levels is an important addition to Ca 19-9, when considering clinical decision making and the prognosis of patients (24).

NLR is a widely accepted marker of prognosis in different malignancies. The first study to examine NLR in patients with PC was reported in 2010 and focused on patients who had undergone pancreaticoduodenectomy. This study found that an elevated NLR is associated with markedly worse survival outcomes (25). The prognostic significance of an elevated NLR was confirmed also in a cohort of advanced PC patients (26). So far, 44 studies have shown the significant prognostic value of NLR in PC, although different cut-offs were used, ranging from 1.7 to 14.1 (27). It is well known that inflammation contributes to cancer development and progression and that a high neutrophil count is a hallmark of systemic inflammation. In particular, neutrophils secreting inflammatory cytokines, such as interleukin-2 (IL-2), interleukin-6 (IL-6), interleukin-10 (IL-10), tumor necrosis factor α (TNF-α), and proangiogenic factors including vascular endothelial growth factor (VEGF), provide a favorable tumor microenvironment for cancer progression. Furthermore, increased IL-10 and TNF-α levels lead to a decrease in the lymphocyte count, with a relevant role in the immune surveillance process towards cancer cells. Therefore, a high NLR could indicate an increased neutrophil-dependent inflammatory response and a reduced lymphocyte-mediated antitumor immune response, resulting in tumor progression and poor outcomes (9,10).

More specifically, the prognostic value of NLR was confirmed in a cohort of 70 patients treated with GnP in real-world practice (28). Prognostic value of NLR was confirmed also recently by Kawahara et al. in a retrospective analysis of 461 patients with pancreatic cancer who underwent surgery. They showed that NLR >3.2 was associated with a decreased OS (21.6 month vs. 25.8 months). In univariate and multivariate analyses, the preoperative NLR was an independent prognostic factor for OS (p=0.022) (29). We also identified elevated NLR as a negative prognostic factor in patients treated with GnP and mFOLFIRINOX. Moreover, NLR was found to be highly significant in our cohort (p=0.012).

The MPS was introduced relatively recently. It combines NLR and albumin levels and is easy to use in clinical practice. Lebenthal et al. found prognostic significance of MPS in a cohort of patients with metastatic PC treated with different first-line regimens (mOS: 12.9 months, 9.0 months, and 5.4 months for MPS 0, 1, and 2, respectively) (30). These results were confirmed in another cohort of metastatic PC patients treated with different first-line regimens (31). We validated and confirmed the prognostic significance of MPS in patients treated with GnP and mFOLFIRINOX.

Finally, PI was first described in a cohort of patients with advanced non-small-cell lung cancer, where it showed significant prognostic value for survival (32). There is also a study showing its prognostic value in PC patients undergoing resection (33). Yet, to the best of our knowledge, its impact on survival in a selected cohort of patients with metastatic PC has not been investigated to date in a real-world setting. Here, we showed that PI is of prognostic significance in a cohort of metastatic patients treated with GnP and mFOLFIRINOX.

We fully acknowledge the limitations of our study, including the small sample size and single-institution design. These factors could introduce bias in patient enrollment and selection of chemotherapy regimens, potentially affecting the interpretation of our results and their generalizability. Despite these shortcomings, we were still able to prove the prognostic significance of several inflammation-based prognostic scores on the survival of patients receiving first-line chemotherapy in a single institution. Finally, based on indirect comparisons, the prognosis of our patients in this study was consistent with published data. Therefore, we believe that inflammation-based prognostic scores are a valuable tool for assessing the prognosis of patients with PC receiving first-line palliative treatment and are readily prepared for routine clinical use.

Conclusion

In this single institution, real-world study, we provide additional evidence on the value of inflammation-based prognostic scores as surrogates for inflammatory status and their clinical relevance for the prognosis of PC patients treated with palliative chemotherapy. We showed that systemic inflammation is associated with reduced OS in patients receiving chemotherapy. This knowledge could assist clinicians in selecting appropriate therapeutic strategies for individualized management, such as intensification of treatment or early introduction of symptomatic treatment in patients with high levels of systemic inflammation markers.

Nevertheless, while inflammatory markers provide valuable prognostic information, they represent just one aspect of a comprehensive assessment of a patient’s condition. Clinical decisions should be based on a combination of several factors, including imaging studies, tumor markers, and the overall health status of the patient. Additionally, ongoing research may provide new insights that further refine our understanding of the relationship between inflammation and metastatic PC.

Funding

This research received no external funding.

Conflicts of Interest

The Authors declare no conflicts of interest in relation to this study.

Authors’ Contributions

M.L. conceptualization, data curation, writing – original draft; T.S. data curation; M.O. writing – review and editing; R.S. supervision, writing – review and editing.

Acknowledgements

The Authors would like to express their gratitude to Michal Tenčík for his invaluable help in statistical data analysis and his valuable suggestions regarding this study.