HALP–GRIms Synergy Offers Superior Prognostic and Predictive Precision in Advanced Pancreatic Ductal Adenocarcinoma

HALP–GRIms Synergy Offers Superior Prognostic and Predictive Precision in Advanced Pancreatic Ductal Adenocarcinoma | Research Square window.SnipcartSettings = { analytics: { enabled: false } }; (function() { var accessVector = localStorage.getItem('access_vector') || ''; window.dataLayer = window.dataLayer || []; if (accessVector) { window.dataLayer.push({ user: { profile: { profileInfo: { snid: accessVector } } } }); } })(); (function(w,d,s,l,i){w[l]=w[l]||[];w[l].push({'gtm.start':new Date().getTime(),event:'gtm.js'});var f=d.getElementsByTagName(s)[0],j=d.createElement(s),dl=l!='dataLayer'?'&l='+l:'';j.async=true;j.src='https://www.googletagmanager.com/gtm.js?id='+i+dl;f.parentNode.insertBefore(j,f);})(window,document,'script','dataLayer','GTM-K279D39R'); Browse Preprints In Review Journals COVID-19 Preprints AJE Video Bytes Research Tools Research Promotion AJE Professional Editing AJE Rubriq About Preprint Platform In Review Editorial Policies Our Team Advisory Board Help Center Sign In Submit a Preprint Cite Share Download PDF Research Article HALP–GRIms Synergy Offers Superior Prognostic and Predictive Precision in Advanced Pancreatic Ductal Adenocarcinoma Berkay Yeşilyurt, Fahriye Tuğba Köş, İsmet Seven This is a preprint; it has not been peer reviewed by a journal. https://doi.org/ 10.21203/rs.3.rs-7277845/v1 This work is licensed under a CC BY 4.0 License Status: Posted Version 1 posted You are reading this latest preprint version Abstract Background : In the challenging landscape of metastatic pancreatic ductal adenocarcinoma (mPDAC), clinicians urgently need accessible and reliable tools to predict outcomes and guide treatment. Immunonutritional indices like the Gustave Roussy Immune Score (GRIms) and the hemoglobin, albumin, lymphocyte, and platelet (HALP) index offer promise, but their full prognostic/predictive potential remained unclear. Methods : We retrospectively analysed 201 patients with mPDAC who received first-line chemotherapy between 2020 and 2024. GRIms and HALP scores were calculated at baseline and at interval using routine blood tests. We assessed their relationship with overall survival (OS) and progression-free survival (PFS) through univariate and multivariate Cox regression models. Results : Patients with high GRIms or low HALP at baseline had significantly shorter OS and PFS (all p < 0.05), with GRIms emerging as a stronger independent predictor (HR for OS: 2.3, p < 0.001). When combined, these two scores offered even greater prognostic precision, clearly identifying the highest-risk patient for mOS (HR: 2.5, p < 0.001). Dynamic changes also mattered; patients whose GRIms increased or HALP declined over time had notably worse survival (mOS: 7 vs 12 months; HR: 3.9, p < 0.001) with combined use suggest stronger prediction (HR: 5.9, p<0.001). A high combined score was also associated with early chemotherapy resistance and poor radiologic response. Conclusion : GRIms and HALP are inexpensive, non-invasive, and readily available markers that offer stronger prognosis in mPDAC. Their combined and longitudinal use may improve clinical decision-making, help flag patients at high risk, and ultimately support more personalized and proactive treatment approaches. Figures Figure 1 Figure 2 Introduction Metastatic pancreatic ductal adenocarcinoma (mPDAC) stands amongst the most lethal malignant tumours worldwide and projected to be second most common cause of cancer-related death in the United States by 2030 1,2 . Roughly 50–55% present with metastatic disease at diagnosis 1 . Therefore, outcomes are poor; the 5-year overall survival rate (OS) is hardly 3% 1,3 . Current treatment for mPDAC is mainly systemic chemotherapy, which improves survival. Multi-chemotherapeutic regimen protocols such as irinotecan, oxaliplatin plus infusional fluorouracil (FOLFIRINOX) and gemcitabine plus nab-paclitaxel can achieve median OS (mOS) rates around 8–11 months 4,5 , but almost all patients encounter disease progression within a year of diagnosis 6 . Unlike many solid tumours, immunotherapy (IO) in mPDAC has not yet been promising, as IO’s have shown indeed only a slight potency in mPDAC patients in some trials 2 . mPDAC is considered as an immunologically ‘cold tumours’, which is defined as having a low mutational burden, with an impenetrable desmoplasia and stiffened matrix architecture, makes it hard for effector immune cells to interact with tumour cells to have required contact for anti-tumours immune response 7 . Given these therapeutic challenges, there is a critical demand for decent prognostic biomarkers to have improved stratification factors and guide treatment choices, and indicators based on systemic inflammation and nutritional status have demonstrated promise in mPDAC, in contrast to tumor markers 6 . One of the aforementioned markers is the hemoglobin, albumin, lymphocyte, and platelet (HALP) score, a combined biomarker contemplative both the nutritional state and systemic inflammatory response 8 . A lower HALP score components often associated with high tumor load, aggressive biology and tumor cachexia. Growing evidence indicates that the HALP score has prognostic implication especially in gastrointestinal malignancies. In mPDAC, a low pretreatment HALP has been correlated with more advanced disease and worse outcomes. For instance, in resected PDAC patients, those with low HALP had a significantly shorter mOS compared to patients with higher HALP (11.5 vs 23.6 months) 8 . Besides, in a cohort of mPDAC patients, a low HALP score was related with markedly shorter median progression-free survival (mPFS) and mOS compared to a high HALP score (mOS 10.6 vs 18.0 months, p < 0.001) 9 . These findings are consistent with a recent meta-analysis of over 13,000 patients, which illustrated that a low HALP is linked to significantly poorer OS for patients with different types of malignancies (pooled hazard ratio (HR):1.6) 10 . Altogether, this evidence suggests that HALP is an acceptable and clinically useful prognostic marker in mPDAC. Another promising prognostic tool is the Gustave Roussy Immune Score (GRIms), which integrates laboratory values of the neutrophil-to-lymphocyte ratio (NLR), serum albumin, and lactate dehydrogenase (LDH) 11 . The GRIms was originally planned to improve patient selection for early-phase trials relating to IO, but it has since proven to be a valuable prognostic index in various cancers and treatment settings 11,12 . A current meta-analysis of nearly 5,000 malignant patients found that a high GRIms was associated with significantly shorter PFS (HR:1.42) and more than doubled the risk of death (HR: 2.07) compared to a low GRIms 12 . In patients with mPDAC, the GRIms has demonstrated prognostic utility (High vs low GRIms mOS: 7 vs 10 months, p < 0.001) 13 . Remarkably, in that study the GRIms shown the highest predictive performance for OS among several inflammation-based scores evaluated. These data emphasize the value of the GRIms as a simple yet effective prognostic marker. As GRIms and HALP pick up distinct prospects of the host-tumour interplay (immunologic/nutritional status vs. tumour burden and systemic inflammation), their combined use may provide a more robust prognostic evaluation than either score alone. Both metrics are low-priced and effortlessly obtained from routine laboratory tests, enhancing their appeal for clinical use. Patients displaying both a high GRIms and a low HALP score could be identified as extremely high-risk, prompting application of intensified therapy or prioritizing clinical trial enrolment, whereas the ones with favourable scores might be managed with standard approaches and exempted unnecessary toxicity. Eventually, integrating GRIms and HALP into prognostic models could enable more customized treatment preparing in mPDAC. Considering this rationale, the existing study evaluates the combined prognostic value of HALP and the GRIms in patients with mPDAC, with particular focus on their ability to show prognostic power and predict OS and PFS. Methods Patient selection In this retrospective cohort study, anonymized patients over the age of 18 with mPDAC who diagnosed between June 2020 to June 2024 were included. Follow-up ended in April 2025. Patients whose treatment started within these dates were eligible, excluding those in clinical trials, patients who receive therapies other than systemic chemotherapy, who rejected/discontinued treatment or follow-up for any reason prior to completing first line treatment or with incomplete laboratory results to calculate HALP and GRIms. Patient Eastern Cooperative Oncology Group (ECOG) PS, demographics, metastatic disease sites, baseline NLR, baseline HALP and GRIms levels, response to first line treatment and survival outcomes were recorded for each participant. The HALP score was calculated using the formula: hemoglobin (g/L) ×albumin (Alb) (g/L) × lymphocyte (/L) divided by platelets (/L). The GRIms was calculated by the following three variables: LDH (within normal range: 0 vs. >upper limit of normal (ULN), 240 U/L in our centre: +1), Alb (≥ 35 g/L: 0 vs. 6: +1). The HALP score was dichotomized using the optimal cutoff value determined by ROC analysis, while the GRIms was categorized based on the median value of the study population. Also, a follow-up value of HALP and GRIms was calculated by the exact method at patients’ interval response assessment (at third month or at progression) after treatment initiation. A combined prognostic score was then generated by integrating HALP and GRIms: patients who were classified as high-risk by both scores were defined as high-risk in the combined model, whereas all other patients were considered low risk. Patients’ first-line treatment regimens and best responses to first-line therapy were recorded. Statistical analyses Patient descriptive characteristics for continuous variables were expressed as the median and interquartile range (IQR; 25th-75th percentile), while categorical variables were summarized using percentage and frequency. For comparing independent groups, independent samples t-tests and Mann–Whitney U tests were used for continuous variables, while fisher’s exact test/chi-square tests were engaged for categorical variables. The optimal HALP cutoff or survival prediction was evaluated with receptor operating curve (ROC) analysis, drawing the common summary measure of the ROC curve by J point to identify the threshold that achieved the optimal balance between sensitivity and specificity. The reverse Kaplan–Meier method was utilized to calculate the follow-up time. OS was calculated from diagnosis of adenocarcinoma to the date of death or last follow-up. PFS was defined as the time elapsed from treatment initiation to disease progression or death. Kaplan–Meier survival analysis was employed to estimate survival probabilities, and the log-rank test was used to compare survival differences between patient groups. Cox proportional hazards regression was utilized for multivariate survival analysis, with variables with a p < .15 in the univariable analysis were included in the multivariable Cox models to evaluate their association with OS and yielding HR’s with associated 95% confidence intervals. Since the HALP score, GRIms, and the combined score share common variables, each was entered into the regression models separately to avoid multicollinearity. Statistical analyses were performed using SPSS version 25.0 (IBM Inc., Armonk, NY, USA), with statistical significance defined as a p -value of less than 0.05. Ethics approval This study was conducted with the approval of local ethics committee (TABED 1-25-1142) and conducted according to the Declaration of Helsinki (revised version 2013). Results There were no distinguishing features in baseline characteristics (table 1). The mPFS of the cohort was 6.5 months (range: 0.1–55.4), and the mOS was 10.56 months (range: 2.0–39.9). By the end of follow-up, 172 patients (85.6%) had died, and 29 patients (14.4%) were still alive. By comparing high-risk and low-risk groups defined by baseline HALP and GRIms, no significant differences were observed for age or ECOG PS. However, patients in the high GRIms group had significantly elevated baseline CEA ( p = 0.001) and Ca19-9 levels ( p < 0.001) compared to the low GRIms group. In contrast, HALP-based stratification did not generate significant differences for CEA, Ca19-9, or age ( p : 0.634, p : 0.647, p : 0.395, respectively). Patients with ECOG 0 were predominantly treated with triplet chemotherapy, while those with ECOG ≥1 were mostly managed with doublet or monotherapy ( p < 0.001), showing statistically significant differences. No statistically significant association was observed between baseline HALP or GRIms and the type of chemotherapy regimen ( p : 0.134 and 0.688, respectively) or primary tumour size ( p = 0.866, p = 0.179, respectively). Patients who received FOLFIRINOX demonstrated a longer mPFS and mOS (7.06 and 11.1 months) compared to those who received other regimens (5.36 and 9.44 months, respectively). However, in Kaplan–Meier analyses, only the difference in PFS reached statistical significance. In the univariate Cox regression model for HALP as a continuous variable, the HALP change score was significantly associated with PFS and OS (HR: 0.984, 95% CI: 0.976–0.992, p < 0.001 and HR: 0.979, 95% CI: 0.971–0.986, p < 0.001- regression coefficient (B): −0.016 and B: −0.022), with a greater decline in HALP score over time is associated with poorer survival outcomes (Figure 1, 2)(Table 2). In multivariate Cox regression analysis for the HALP numerical change score alongside clinical parameters, a significant association was found between HALP decline and poorer outcomes in both PFS and OS. For PFS, each unit decrease in HALP score was associated with a 2.6% increase in the risk of progression or death (HR: 0.974, 95% CI: 0.966–0.982, p < 0.001). Similarly, in the OS model, the risk increased by 2.0% per unit decrease (HR: 0.980, 95% CI: 0.971–0.989, p < 0.001) (Table 3,4). These associations remained significant after adjusting for ECOG PS, age, treatment regimen, number of metastases, and CA19-9 levels. In subgroup analysis, among patients treated with FOLFIRINOX, those classified as high-risk by the combined score had significantly lower mOS compared to low-risk patients (10.1 vs 11.2 months, log-rank p = 0.008; HR = 2.32, 95% CI: 1.23–4.39, p = 0.009). Among patients treated with monotherapy/doublet therapy, those classified as high-risk by the combined score had even more significantly shorter mOS compared to low-risk patients (5.7 vs 10.5 months, log-rank p <0.005; HR = 3.25, 95% CI: 1.91–5.50, p <0.005). Logistic regression analyses was conducted to evaluate whether baseline HALP, GRIms and combined scores could predict radiological response to first-line chemotherapy (classified as CR/PR/SD vs PD). Between HALP, GRIms and combined score high/low risk groups, no statistically significant difference in the best responses were recorded. The GRIms and combined score demonstrated a statistically significant association with PD ( p = 0.045; OR = 1.84; 95% CI: 1.02–3.33 and p = 0.048, OR=2.07, 95% CI: 0.99–4.34, respectively). The GRIms and combined score might also serve as a predictive marker of chemotherapy resistance in mPDAC. Discussion The study identified that systemic immunonutritional scores, notably the GRIms and combined score, surpassed the HALP score in foreshadowing outcomes for advanced PDAC patients. Consistent with preceding studies that a lower HALP score indicates worse survival in mPDAC, our data affirm HALP as a compelling univariate predictor of OS and PFS 10,14 . Nonetheless on multivariate analyses, HALP lost its power compared to GRIms and the combined score, recommending that the added factors in GRIms and their integration with HALP catching additional prognostic ability. This is in line with recent large research in mPDAC which high risk GRIms was related with a 2.3-fold higher risk of death 13 . Our data bolster that integrating nutritional and inflammatory markers provides a more powerful score, recalling the Memorial Sloan Kettering Score (albumin + NLR) which also predicted survival 15 . The separate prognostic value of GRIms and the combined score, indeed after compensating for other variables, underscores their clinical relevance. These scores likely provide as proxies of a host frailty and aggressive tumour microenvironment– conditions not fully apprehended by PS level alone 15,16 . The supremacy of the combined score and GRIms could be associated to their inclusion of the systemic inflammatory response. NLR is a settled marker of inflammation that is cancer-related interacts with cytokine-directed tumour progression and immunosuppressive neutrophil activity, so that high NLR mean imply lower tumour-infiltrating lymphocytes (TIL) which indicates an immune ‘cold’ microenvironment, and consecutively predicts therapy resistance and lower OS 13,17,18 . In our study, high GRIms scores (high NLR affects GRIms) might coincided to lower CD8 + TIL levels in tumour histology 13 . Also, meta-analysis data shows that high GRIms level deliberate shorter mPFS and a 2-times higher mortality risk 12 . Inclusion of LDH -a marker of disease burden and endproduct of anaerobic glycolysis– increases prognostic efficiency 19 . By incorporating GRIms with HALP, combined score catches nutritional and inflammatory condition. Anaemia and malnutrition (HALP’s Hb and Alb units) presumably aggravate immune dysfunction and intolerance to treatment, as thrombocytosis and lymphopenia suggest cytokine increase that sustains malady and tumour progression 17,20,21 . Hence, patients with worse combined scores display a specifically high-risk phenotype with both compromised host reserves and aggressive tumour biology. This subgroups’ shorter mPFS and mOS, also the risk of early death was being 2.5–3.9 times higher on multivariate analysis were seen in this study is consistent with prior analyses showing that lower nutritional scores (Alb, prognostic nutritional index) and elevated inflammation-based scores (CRP/Alb, NLR) separately anticipate worse survival in mPDAC 22 . Unsurprisingly, patients with better combined scores with mOS goes beyond 18 months were also compatible with literature. Another purpose was showing HALP–GRIms combination might also provide predictive vision for response to chemotherapy in mPDAC. We found that patients with high combined scores or high GRIms were naturally more prone to display progressive disease on first-line chemotherapy. Specifically, patients with worse immunonutritional status had lesser probability to benefit from approved regimens. There is coherent evidence for IO in urothelial carcinoma that high GRIms predicts shorter mPFS and lack of response 19 . It also reverberate with a prior study in PDAC that initial NLR < 3.1 was a separate predictor of response to first-line gemcitabine-based chemotherapy 23 . Besides, Okabe et al. showed that a dynamic NLR shift was meaningfully related with disease control rate following chemotherapy 24 . Our results additionally revealed a initial combined score might find patients at risk of resistance. Patients with high GRIms presumably have systemic inflammation and tumour-driven granulocytosis that advocates desmoplasia induced immunosuppression, therapy resistance 18 . Granulocyte-derived factors (e.g. IL-1β, IL-6/STAT3 activation) have been demonstrated to weaken chemotherapeutic potency in PDAC 17,18 . So, the primary progressive patients in our study may be induced as an intrinsically chemorefractory tumour microenvironment. However, HALP alone was an inadequate predictor, indicating that inflammatory status is the basic driver of chemosensitivity. These data suggest that immunonutritional scoring could be used in response estimate. Whose HALP score–GRIms deteriorate on therapy might benefit closer surveillance or early treatment modification, yet since progression becomes apparent. Patients with both a GRIms rise and HALP decline had the poorest results. Dynamic layering were highly significant, and the combined change presented almost a four-fold hazard of death in univariate analyses. This highlights that trajectories of these markers during treatment summarize the response and the tolerance to therapy. Decreasing HALP probably indicates cachexia or nutritional deficits, while increasing GRIms may reflect unresolving tumour inflammation under therapy. In a new prospective study of advanced cancers, whose NLR decreased after palliative care had better mOS compared to those with increasing NLR 17 . In PDAC especially, Datta et al. detected that NLR fading throughout preoperative chemotherapy correlated with longer survival and trend to major pathologic response 18 . Also, a study in NSCLC IO revealed that whose HALP scores increased during treatment had considerably longer mOS and mPFS 25 . Checking HALP and GRIms during interval response assessment of first-line therapy might assist to recognize who could respond to early switch to second-line options, these patients may have rising host stress or microscopic progression and threshold of suspicion for progression should be lower. This treatment personalization approach is also shown in a new real-world analyses remains to find CA19-9 and PS as important prognostic factors, but our evidence propose immunonutritional dynamics incorporate litigable elements ahead static baseline levels 16 . Moreover, prospective studies should assess adding HALP/GRIms or other score changes to be a criterion for early modification or for stratifying maintenance vs. switch strategies. Inflammatory indices like GRIms have been utilized to choose patients for IO trials, and it is known that PDAC patients with a significant inflammatory load experience less benefit from both IO’s and standard chemotherapy 13,19 . Stratifying through HALP–GRIms could classify patients who may need auxiliary cachexia-targeted or anti-inflammatory approaches beside chemotherapy. Neutrophil function inhibition, IL-6 blockade, or nutritional support might particularly support the high-risk group. Preclinical studies have concluded that disrupting the inflammation–immunosuppression cycle (like with CXCR1/2 or CCR2 inhibitors) may recover chemosensitivity in PDAC models 18 . And the HALP–GRIms could perform as a stratification factor in future trials, assuring fair evaluation of novel treatments among risk subgroups 13 . Conclusion In summary, our study emphasize that combined use of immunonutritional scores provide superior predictive and prognostic data in mPDAC. The GRIms and combined score more powerfully predicted mPFS and mOS than HALP alone and also predicted radiologic non-response to first-line chemotherapy. Combined scores’ power being higher than GRIms suggest HALP adds additional capability to GRIms, therefore should be used together. Dynamic changes appeared as effective early indicators of treatment potency, surpassing initial levels and assist identifying patients at risk for rapid progression to tailor therapy appropriately. The scores can be extracted, calculated using hospital-based laboratory test results and automatically presented on electronic health records. Routine combined use of these scores in clinical practice could ameliorate risk stratification above traditional factors, promoting personalized treatment in mPDAC. Our results necessitates prospective validation but provide important evidence to use immunonutritional scores both as prognostic markers and as early response metrics to optimize outcomes in PDAC by demonstrating interaction between tumour biology and host condition, holding serious hope to improve management and trial design in the future 13,14,16 . Declarations Conflict of Interest Statement The authors declare that they have no conflict of interest. Acknowledgements The authors would like to thank the medical and administrative staff of the Ankara City Hospital for their support in facilitating access to patient data and clinical records. Data Availability Statement The data that support the findings of this study are available from the corresponding author upon reasonable request. References Mizrahi JD, Surana R, Valle JW, Shroff RT. Pancreatic cancer. Lancet 2020;395(10242):2008-2020. (In eng). DOI: 10.1016/s0140-6736(20)30974-0. Park W, Chawla A, O'Reilly EM. Pancreatic Cancer: A Review. Jama 2021;326(9):851-862. (In eng). DOI: 10.1001/jama.2021.13027. Lippi G, Mattiuzzi C. The global burden of pancreatic cancer. Arch Med Sci 2020;16(4):820-824. (In eng). DOI: 10.5114/aoms.2020.94845. Conroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus Gemcitabine for Metastatic Pancreatic Cancer. New England Journal of Medicine 2011;364(19):1817-1825. DOI: doi:10.1056/NEJMoa1011923. Hoff DDV, Ervin T, Arena FP, et al. Increased Survival in Pancreatic Cancer with nab-Paclitaxel plus Gemcitabine. New England Journal of Medicine 2013;369(18):1691-1703. DOI: doi:10.1056/NEJMoa1304369. Rochefort P, Lardy-Cleaud A, Sarabi M, Desseigne F, Cattey-Javouhey A, de la Fouchardière C. Long-Term Survivors in Metastatic Pancreatic Ductal Adenocarcinoma: A Retrospective and Matched Pair Analysis. Oncologist 2019;24(12):1543-1548. (In eng). DOI: 10.1634/theoncologist.2018-0786. Dell'Aquila E, Fulgenzi CAM, Minelli A, et al. Prognostic and predictive factors in pancreatic cancer. Oncotarget 2020;11(10):924-941. (In eng). DOI: 10.18632/oncotarget.27518. Xu SS, Li S, Xu HX, et al. Haemoglobin, albumin, lymphocyte and platelet predicts postoperative survival in pancreatic cancer. World J Gastroenterol 2020;26(8):828-838. (In eng). DOI: 10.3748/wjg.v26.i8.828. Balçık OY, Aytaç A, Ekinci F, et al. Clinical Importance of the Hemoglobin, Albumin, Lymphocyte, Platelet Score in Metastatic Pancreatic Cancer. Acta Oncologica Turcica 2023;56(3):250-258. DOI: 10.5505/aot.2023.26779. Xu H, Zheng X, Ai J, Yang L. Hemoglobin, albumin, lymphocyte, and platelet (HALP) score and cancer prognosis: A systematic review and meta-analysis of 13,110 patients. Int Immunopharmacol 2023;114:109496. (In eng). DOI: 10.1016/j.intimp.2022.109496. Bigot F, Castanon E, Baldini C, et al. Prospective validation of a prognostic score for patients in immunotherapy phase I trials: The Gustave Roussy Immune Score (GRIm-Score). Eur J Cancer 2017;84:212-218. (In eng). DOI: 10.1016/j.ejca.2017.07.027. Sun KX, Xu RQ, Rong H, Pang HY, Xiang TX. Prognostic significance of the Gustave Roussy immune (GRIm) score in cancer patients: a meta-analysis. Ann Med 2023;55(2):2236640. (In eng). DOI: 10.1080/07853890.2023.2236640. Ma LX, Wang Y, Espin-Garcia O, et al. Systemic inflammatory prognostic scores in advanced pancreatic adenocarcinoma. Br J Cancer 2023;128(10):1916-1921. (In eng). DOI: 10.1038/s41416-023-02214-0. Demir N, Gökmen İ, Sağdıç Karateke Y, et al. HALP score as a prognostic marker for overall survival in advanced pancreatic cancer. Front Oncol 2025;15:1542463. (In eng). DOI: 10.3389/fonc.2025.1542463. Lebenthal JM, Zheng J, Glare PA, O'Reilly EM, Yang AC, Epstein AS. Prognostic value of the Memorial Sloan Kettering Prognostic Score in metastatic pancreatic adenocarcinoma. Cancer 2021;127(10):1568-1575. (In eng). DOI: 10.1002/cncr.33420. Taieb J, Seufferlein T, Reni M, et al. Treatment sequences and prognostic/predictive factors in metastatic pancreatic ductal adenocarcinoma: univariate and multivariate analyses of a real-world study in Europe. BMC Cancer 2023;23(1):877. DOI: 10.1186/s12885-023-11377-1. Zhao W, Wu Z, Li Y, et al. Pretreatment neutrophil-to-lymphocyte ratio and its dynamic changes are associated with the overall survival in advanced cancer patients undergoing palliative care. Scientific Reports 2016;6(1):31394. DOI: 10.1038/srep31394. de Castro Silva I, Bianchi A, Deshpande NU, et al. Neutrophil-mediated fibroblast-tumor cell il-6/stat-3 signaling underlies the association between neutrophil-to-lymphocyte ratio dynamics and chemotherapy response in localized pancreatic cancer: A hybrid clinical-preclinical study. Elife 2022;11 (In eng). DOI: 10.7554/eLife.78921. Tanabe K, Kobayashi S, Maezawa Y, et al. Gustave Roussy Immune score as a prognostic biomarker in patients with platinum-refractory metastatic urothelial carcinoma treated with pembrolizumab: YUSHIMA study. Int J Clin Oncol 2024;29(9):1302-1310. (In eng). DOI: 10.1007/s10147-024-02563-7. Han R, Tian Z, Jiang Y, et al. Prognostic significance of the systemic immune inflammation index in patients with metastatic and unresectable pancreatic cancer. Frontiers in Surgery 2022;Volume 9 - 2022 (Original Research) (In English). DOI: 10.3389/fsurg.2022.915599. GÜLbaĞCi BB, ÖZen M, ÖZen EngİN E, AkdaĞ KahvecİOĞLu F, ÇİFtÇİ E, HacibekİRoĞLu İ. Does Delta Hemoglobin-Albumin-Lymphocyte Platelet (HALP) Score Predict the Risk of Early Progression in Patients Treated with CDK4/6 Inhibitors? Namık Kemal Tıp Dergisi 2024;12(4):260-265. DOI: 10.4274/nkmj.galenos.2024.12599. Hackner D, Merkel S, Weiß A, et al. Neutrophil-to-Lymphocyte Ratio and Prognostic Nutritional Index Are Predictors for Overall Survival after Primary Pancreatic Resection of Pancreatic Ductal Adenocarcinoma: A Single Centre Evaluation. Cancers 2024;16(16):2911. (https://www.mdpi.com/2072-6694/16/16/2911). Kitsugi K, Kawata K, Noritake H, et al. Prognostic value of neutrophil to lymphocyte ratio in patients with advanced pancreatic ductal adenocarcinoma treated with systemic chemotherapy. Ann Med 2024;56(1):2398725. (In eng). DOI: 10.1080/07853890.2024.2398725. Okabe H, Masuda T, Tomita M, et al. Combined Neutrophil-to-Lymphocyte Ratio Score Is Associated With Chemotherapeutic Response and Predicts Prognosis in Patients With Advanced Pancreatic Cancer. Anticancer Res 2024;44(4):1575-1582. (In eng). DOI: 10.21873/anticanres.16955. Koçanoğlu A, Karakaya S, Zeynelgil E, Düzköprü Y, Doğan Ö. Dynamic Alteration of HALP Score as a Predictor in Patients with Receiving Immunotherapy for Advanced Non-Small Cell Lung Cancer. Medicina 2025;61(6):989. (https://www.mdpi.com/1648-9144/61/6/989). Tables Table 1: Baseline characteristics Characteristic Value Number of patients (n) 201 Age at diagnosis (years) Mean ± SD (range) 62.9 ± 10.7 (33–89) Sex Male Female 120 (59.7%) 81 (40.3%) ECOG PS 0–1 2–3 122 (60.7%) 79 (39.3%) Follow-up duration (months) Median (95% CI) 23.2 (3.83–42.57) Chemotherapy regimen Triplet (FOLFIRINOX) Monotherapy/Doublet 89 (44.3%) 112 (55.7%) Albumin level (g/L) Median (IQR) 40.40 (34.0–42.1) LDH (U/L) Median (IQR) 223.50 (136.8–224.6) Hemoglobin (g/dL) Median (IQR) 12.70 (11.0–13.3) NLR Median (IQR) 3.13 (2.21–5.22) CEA (ng/mL) Median (IQR) 3.65 (0.1–12.88) CA 19-9 (U/mL) Median (IQR) 627.10 (85.5–4832.7) Initial HALP score Median (IQR) 30.66 (28.99) Initial GRIms Median (IQR) 1.00 (1.00) Table 2: Univariate Cox regression models for initial HALP, GRIms and combined scores: Score OS HR (95% CI) p -value OS PFS HR (95% CI) p -value PFS HALP 1.373 (1.015-1.857) 0.04 1.339 (1.012-1.770) 0.041 GRIm 2.410 (1.730-3.356) <0.001 1.960 (1.448-2.653) <0.001 Combined Score 2.641 (1.766-3.948) <0.001 1.868 (1.289-2.707) 0.001 Table 3: Multivariate Cox regression models for initial values and interval (follow-up) HALP, GRIms, combined score; adjusted for age, ECOG PS, chemotherapy regimen, and tumour markers: Variable Adjusted HR (95% CI) for OS p -value OS Adjusted HR (95% CI) for PFS p -value PFS Combined Score (High vs Low) 2.524 (1.677-3.801) <0.001 1.804 (1.240-2.627) 0.002 GRIms (High vs Low) 2.330 (1.670-3.260) <0.001 1.985 (1.463–2.702) <0.001 HALP (High vs Low) 1.290 (0.950–1.760) 0.108 1.290 (0.970–1.710) 0.078 ECOG 1.759 (1.216-2.544) 0.003 1.321 (0.954-1.830) 0.094 CA 19-9 (High vs Low) 1.490 (1.084-2.046) 0.014 1.482 (1.109-1.981) 0.008 Age 1.169 (0.842-1.623) 0.352 1.181 (0.874-1.595) 0.278 CEA (High vs Low) 0.997 (0.730-1.362) 0.986 1.091 (0.817-1.457) 0.555 Chemotherapy Regimen 1.000 (0.686-1.457) 0.999 1.242 (0.892-1.728) 0.199 Interval HALP (High vs Low) 2.402 (1.750-3.290) <0.001 Interval GRIm (High vs Low) 5.481 (3.690-8.142) <0.001 Interval Combined Score (High vs Low) 5.904 (4.051-8.594) <0.001 Table 4: Univariate Association of Post-Treatment Score Changes with mOS and mPFS Variable mOS (months) mPFS (months) Univariate Cox HR (95% CI) Decrease in HALP 8.87 vs 12.21 4.97 vs 7.93 2.34 (1.741–3.138) Increase in GRIms 7.54 vs 12.38 4.07 vs 7.67 2.48 (1.842–3.342) Higher Combined score 7.36 vs 11.93 3.60 vs 7.40 3.85 (2.720–5.442) Cite Share Download PDF Status: Posted Version 1 posted You are reading this latest preprint version Research Square lets you share your work early, gain feedback from the community, and start making changes to your manuscript prior to peer review in a journal. As a division of Research Square Company, we’re committed to making research communication faster, fairer, and more useful. We do this by developing innovative software and high quality services for the global research community. Our growing team is made up of researchers and industry professionals working together to solve the most critical problems facing scientific publishing. Also discoverable on Platform About Our Team In Review Editorial Policies Advisory Board Help Center Resources Author Services Accessibility API Access RSS feed Manage Cookie Preferences © Research Square 2026 | ISSN 2693-5015 (online) Privacy Policy Terms of Service Do Not Sell My Personal Information {"props":{"pageProps":{"initialData":{"identity":"rs-7277845","acceptedTermsAndConditions":true,"allowDirectSubmit":true,"archivedVersions":[],"articleType":"Research Article","associatedPublications":[],"authors":[{"id":500376372,"identity":"85fe314d-3664-4405-93f8-9f11c8f0d9fb","order_by":0,"name":"Berkay Yeşilyurt","email":"data:image/png;base64,iVBORw0KGgoAAAANSUhEUgAAAZAAAAAyAQMAAABI0h/eAAAABlBMVEX///8AAABVwtN+AAAACXBIWXMAAA7EAAAOxAGVKw4bAAAA50lEQVRIie3PuwrCMBSA4SMH0uWoawRRHyFQ8II+jCJ0qi6Ck4hTJ/VZnJwrwbrp2s2h4CAOgiAIKqZ172UTzD8kGfKREwCd7gcTABjuXUB01c7LWQjrhoQyECARHpJJw1gE4uHAUOzodvYnTQJDbldxpDXfmb2FAyMh8+u27anByLL82MF8C928A72VIqbNFOFUjyfHE25eEaGTab/TEJ9h//sKYTBwUpDW3EKzvOejkmR1HCw5saS/NAwPS5dxZ1g4yOBm36eVoiG9WBKVYxxqLqhVxRKvRz0BqjPAa7rbOp1O9299APQpRJeg7dXdAAAAAElFTkSuQmCC","orcid":"https://orcid.org/0000-0003-0130-7111","institution":"Ankara City Hospital: Ankara Sehir Hastanesi","correspondingAuthor":true,"prefix":"","firstName":"Berkay","middleName":"","lastName":"Yeşilyurt","suffix":""},{"id":500376373,"identity":"49f4e2de-4740-4bdf-a93d-66e06c840684","order_by":1,"name":"Fahriye Tuğba Köş","email":"","orcid":"","institution":"Ankara City Hospital: Ankara Sehir Hastanesi","correspondingAuthor":false,"prefix":"","firstName":"Fahriye","middleName":"Tuğba","lastName":"Köş","suffix":""},{"id":500376374,"identity":"ff76019f-7e0d-4eba-aa0d-5ab673f35bc9","order_by":2,"name":"İsmet Seven","email":"","orcid":"","institution":"Ankara City Hospital: Ankara Sehir Hastanesi","correspondingAuthor":false,"prefix":"","firstName":"İsmet","middleName":"","lastName":"Seven","suffix":""}],"badges":[],"createdAt":"2025-08-02 11:21:51","currentVersionCode":1,"declarations":"","doi":"10.21203/rs.3.rs-7277845/v1","doiUrl":"https://doi.org/10.21203/rs.3.rs-7277845/v1","draftVersion":[],"editorialEvents":[],"editorialNote":"","failedWorkflow":false,"files":[{"id":89654855,"identity":"c6c02116-4a96-4b78-a77b-39633ce4f6be","added_by":"auto","created_at":"2025-08-22 10:19:47","extension":"png","order_by":1,"title":"Figure 1","display":"","copyAsset":false,"role":"figure","size":81822,"visible":true,"origin":"","legend":"\u003cp\u003eKaplan-meier curves of initial HALP, GRIms and combined score for PFS\u003c/p\u003e","description":"","filename":"floatimage1.png","url":"https://assets-eu.researchsquare.com/files/rs-7277845/v1/a29829af1ad4c7ba8e35e28d.png"},{"id":89654857,"identity":"86239d3a-d0fc-46f0-bd48-0af30262731b","added_by":"auto","created_at":"2025-08-22 10:19:47","extension":"png","order_by":2,"title":"Figure 2","display":"","copyAsset":false,"role":"figure","size":78674,"visible":true,"origin":"","legend":"\u003cp\u003eKaplan-meier curves of initial HALP, GRIms and combined score for OS\u003c/p\u003e","description":"","filename":"floatimage2.png","url":"https://assets-eu.researchsquare.com/files/rs-7277845/v1/798878dfc9ecf03dd2b48f19.png"},{"id":89875343,"identity":"5e207067-6a09-4d91-8823-a367cccafa3e","added_by":"auto","created_at":"2025-08-26 03:57:38","extension":"pdf","order_by":0,"title":"","display":"","copyAsset":false,"role":"manuscript-pdf","size":741751,"visible":true,"origin":"","legend":"","description":"","filename":"manuscript.pdf","url":"https://assets-eu.researchsquare.com/files/rs-7277845/v1/3e32e9a1-220b-411f-bb53-2c8bbd151751.pdf"}],"financialInterests":"","formattedTitle":"HALP–GRIms Synergy Offers Superior Prognostic and Predictive Precision in Advanced Pancreatic Ductal Adenocarcinoma","fulltext":[{"header":"Introduction","content":"\u003cp\u003eMetastatic pancreatic ductal adenocarcinoma (mPDAC) stands amongst the most lethal malignant tumours worldwide and projected to be second most common cause of cancer-related death in the United States by 2030\u003csup\u003e1,2\u003c/sup\u003e. Roughly 50–55% present with metastatic disease at diagnosis\u003csup\u003e1\u003c/sup\u003e. Therefore, outcomes are poor; the 5-year overall survival rate (OS) is hardly 3%\u003csup\u003e1,3\u003c/sup\u003e. Current treatment for mPDAC is mainly systemic chemotherapy, which improves survival. Multi-chemotherapeutic regimen protocols such as irinotecan, oxaliplatin plus infusional fluorouracil (FOLFIRINOX) and gemcitabine plus nab-paclitaxel can achieve median OS (mOS) rates around 8–11 months\u003csup\u003e4,5\u003c/sup\u003e, but almost all patients encounter disease progression within a year of diagnosis\u003csup\u003e6\u003c/sup\u003e. Unlike many solid tumours, immunotherapy (IO) in mPDAC has not yet been promising, as IO’s have shown indeed only a slight potency in mPDAC patients in some trials\u003csup\u003e2\u003c/sup\u003e. mPDAC is considered as an immunologically ‘cold tumours’, which is defined as having a low mutational burden, with an impenetrable desmoplasia and stiffened matrix architecture, makes it hard for effector immune cells to interact with tumour cells to have required contact for anti-tumours immune response\u003csup\u003e7\u003c/sup\u003e. Given these therapeutic challenges, there is a critical demand for decent prognostic biomarkers to have improved stratification factors and guide treatment choices, and indicators based on systemic inflammation and nutritional status have demonstrated promise in mPDAC, in contrast to tumor markers\u003csup\u003e6\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eOne of the aforementioned markers is the hemoglobin, albumin, lymphocyte, and platelet (HALP) score, a combined biomarker contemplative both the nutritional state and systemic inflammatory response\u003csup\u003e8\u003c/sup\u003e. A lower HALP score components often associated with high tumor load, aggressive biology and tumor cachexia. Growing evidence indicates that the HALP score has prognostic implication especially in gastrointestinal malignancies. In mPDAC, a low pretreatment HALP has been correlated with more advanced disease and worse outcomes. For instance, in resected PDAC patients, those with low HALP had a significantly shorter mOS compared to patients with higher HALP (11.5 vs 23.6 months)\u003csup\u003e8\u003c/sup\u003e. Besides, in a cohort of mPDAC patients, a low HALP score was related with markedly shorter median progression-free survival (mPFS) and mOS compared to a high HALP score (mOS 10.6 vs 18.0 months, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001)\u003csup\u003e9\u003c/sup\u003e. These findings are consistent with a recent meta-analysis of over 13,000 patients, which illustrated that a low HALP is linked to significantly poorer OS for patients with different types of malignancies (pooled hazard ratio (HR):1.6)\u003csup\u003e10\u003c/sup\u003e. Altogether, this evidence suggests that HALP is an acceptable and clinically useful prognostic marker in mPDAC.\u003c/p\u003e\u003cp\u003eAnother promising prognostic tool is the Gustave Roussy Immune Score (GRIms), which integrates laboratory values of the neutrophil-to-lymphocyte ratio (NLR), serum albumin, and lactate dehydrogenase (LDH)\u003csup\u003e11\u003c/sup\u003e. The GRIms was originally planned to improve patient selection for early-phase trials relating to IO, but it has since proven to be a valuable prognostic index in various cancers and treatment settings\u003csup\u003e11,12\u003c/sup\u003e. A current meta-analysis of nearly 5,000 malignant patients found that a high GRIms was associated with significantly shorter PFS (HR:1.42) and more than doubled the risk of death (HR: 2.07) compared to a low GRIms\u003csup\u003e12\u003c/sup\u003e. In patients with mPDAC, the GRIms has demonstrated prognostic utility (High vs low GRIms mOS: 7 vs 10 months, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001)\u003csup\u003e13\u003c/sup\u003e. Remarkably, in that study the GRIms shown the highest predictive performance for OS among several inflammation-based scores evaluated. These data emphasize the value of the GRIms as a simple yet effective prognostic marker.\u003c/p\u003e\u003cp\u003eAs GRIms and HALP pick up distinct prospects of the host-tumour interplay (immunologic/nutritional status vs. tumour burden and systemic inflammation), their combined use may provide a more robust prognostic evaluation than either score alone. Both metrics are low-priced and effortlessly obtained from routine laboratory tests, enhancing their appeal for clinical use. Patients displaying both a high GRIms and a low HALP score could be identified as extremely high-risk, prompting application of intensified therapy or prioritizing clinical trial enrolment, whereas the ones with favourable scores might be managed with standard approaches and exempted unnecessary toxicity. Eventually, integrating GRIms and HALP into prognostic models could enable more customized treatment preparing in mPDAC. Considering this rationale, the existing study evaluates the combined prognostic value of HALP and the GRIms in patients with mPDAC, with particular focus on their ability to show prognostic power and predict OS and PFS.\u003c/p\u003e\u003cp\u003e\u003c/p\u003e"},{"header":"Methods","content":"\u003cp\u003e\u003cb\u003ePatient selection\u003c/b\u003e\u003c/p\u003e\u003cp\u003eIn this retrospective cohort study, anonymized patients over the age of 18 with mPDAC who diagnosed between June 2020 to June 2024 were included. Follow-up ended in April 2025. Patients whose treatment started within these dates were eligible, excluding those in clinical trials, patients who receive therapies other than systemic chemotherapy, who rejected/discontinued treatment or follow-up for any reason prior to completing first line treatment or with incomplete laboratory results to calculate HALP and GRIms. Patient Eastern Cooperative Oncology Group (ECOG) PS, demographics, metastatic disease sites, baseline NLR, baseline HALP and GRIms levels, response to first line treatment and survival outcomes were recorded for each participant. The HALP score was calculated using the formula: hemoglobin (g/L) ×albumin (Alb) (g/L) × lymphocyte (/L) divided by platelets (/L). The GRIms was calculated by the following three variables: LDH (within normal range: 0 vs. \u0026gt;upper limit of normal (ULN), 240 U/L in our centre: +1), Alb (≥ 35 g/L: 0 vs. \u0026lt; 35g/L: +1), and NLR (≤ 6: 0 vs. \u0026gt;6: +1). The HALP score was dichotomized using the optimal cutoff value determined by ROC analysis, while the GRIms was categorized based on the median value of the study population. Also, a follow-up value of HALP and GRIms was calculated by the exact method at patients’ interval response assessment (at third month or at progression) after treatment initiation. A combined prognostic score was then generated by integrating HALP and GRIms: patients who were classified as high-risk by both scores were defined as high-risk in the combined model, whereas all other patients were considered low risk. Patients’ first-line treatment regimens and best responses to first-line therapy were recorded.\u003c/p\u003e\u003cp\u003e\u003cb\u003eStatistical analyses\u003c/b\u003e\u003c/p\u003e\u003cp\u003ePatient descriptive characteristics for continuous variables were expressed as the median and interquartile range (IQR; 25th-75th percentile), while categorical variables were summarized using percentage and frequency. For comparing independent groups, independent samples t-tests and Mann–Whitney U tests were used for continuous variables, while fisher’s exact test/chi-square tests were engaged for categorical variables. The optimal HALP cutoff or survival prediction was evaluated with receptor operating curve (ROC) analysis, drawing the common summary measure of the ROC curve by J point to identify the threshold that achieved the optimal balance between sensitivity and specificity. The reverse Kaplan–Meier method was utilized to calculate the follow-up time. OS was calculated from diagnosis of adenocarcinoma to the date of death or last follow-up. PFS was defined as the time elapsed from treatment initiation to disease progression or death. Kaplan–Meier survival analysis was employed to estimate survival probabilities, and the log-rank test was used to compare survival differences between patient groups. Cox proportional hazards regression was utilized for multivariate survival analysis, with variables with a \u003cem\u003ep\u003c/em\u003e \u0026lt; .15 in the univariable analysis were included in the multivariable Cox models to evaluate their association with OS and yielding HR’s with associated 95% confidence intervals. Since the HALP score, GRIms, and the combined score share common variables, each was entered into the regression models separately to avoid multicollinearity. Statistical analyses were performed using SPSS version 25.0 (IBM Inc., Armonk, NY, USA), with statistical significance defined as a \u003cem\u003ep\u003c/em\u003e-value of less than 0.05.\u003c/p\u003e\u003cp\u003e\u003cstrong\u003eEthics approval\u003c/strong\u003e\u003c/p\u003e\u003cp\u003e This study was conducted with the approval of local ethics committee (TABED 1-25-1142) and conducted according to the Declaration of Helsinki (revised version 2013).\u003c/p\u003e"},{"header":"Results","content":"\u003cp\u003eThere were no distinguishing features in baseline characteristics (table 1). The mPFS of the cohort was 6.5 months (range: 0.1\u0026ndash;55.4), and the mOS was 10.56 months (range: 2.0\u0026ndash;39.9). By the end of follow-up, 172 patients (85.6%) had died, and 29 patients (14.4%) were still alive.\u003c/p\u003e\n\u003cp\u003eBy comparing high-risk and low-risk groups defined by baseline HALP and GRIms, no significant differences were observed for age or ECOG PS. However, patients in the high GRIms group had significantly elevated baseline CEA (\u003cem\u003ep\u003c/em\u003e = 0.001) and Ca19-9 levels (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001) compared to the low GRIms group. In contrast, HALP-based stratification did not generate significant differences for CEA, Ca19-9, or age (\u003cem\u003ep\u003c/em\u003e: 0.634, \u003cem\u003ep\u003c/em\u003e: 0.647,\u003cem\u003e\u0026nbsp;p\u003c/em\u003e: 0.395, respectively).\u0026nbsp;\u003c/p\u003e\n\u003cp\u003ePatients with ECOG 0 were predominantly treated with triplet chemotherapy, while those with ECOG \u0026ge;1 were mostly managed with doublet or monotherapy (\u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001), showing statistically significant differences.\u003c/p\u003e\n\u003cp\u003eNo statistically significant association was observed between baseline HALP or GRIms and the type of chemotherapy regimen (\u003cem\u003ep\u003c/em\u003e: 0.134 and 0.688, respectively) or primary tumour size (\u003cem\u003ep\u003c/em\u003e = 0.866, \u003cem\u003ep\u003c/em\u003e = 0.179, respectively).\u003c/p\u003e\n\u003cp\u003ePatients who received FOLFIRINOX demonstrated a longer mPFS and mOS (7.06 and 11.1 months) compared to those who received other regimens (5.36 and 9.44 months, respectively). However, in Kaplan\u0026ndash;Meier analyses, only the difference in PFS reached statistical significance.\u003c/p\u003e\n\u003cp\u003eIn the univariate Cox regression model for HALP as a continuous variable, the HALP change score was significantly associated with PFS and OS (HR: 0.984, 95% CI: 0.976\u0026ndash;0.992, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001 and HR: 0.979, 95% CI: 0.971\u0026ndash;0.986, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001- regression coefficient (B): \u0026minus;0.016 and B: \u0026minus;0.022), with a greater decline in HALP score over time is associated with poorer survival outcomes (Figure 1, 2)(Table 2).\u003c/p\u003e\n\u003cp\u003eIn multivariate Cox regression analysis for the HALP numerical change score alongside clinical parameters, a significant association was found between HALP decline and poorer outcomes in both PFS and OS. For PFS, each unit decrease in HALP score was associated with a 2.6% increase in the risk of progression or death (HR: 0.974, 95% CI: 0.966\u0026ndash;0.982, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001). Similarly, in the OS model, the risk increased by 2.0% per unit decrease (HR: 0.980, 95% CI: 0.971\u0026ndash;0.989, \u003cem\u003ep\u003c/em\u003e \u0026lt; 0.001) (Table 3,4). These associations remained significant after adjusting for ECOG PS, age, treatment regimen, number of metastases, and CA19-9 levels.\u003c/p\u003e\n\u003cp\u003eIn subgroup analysis, among patients treated with FOLFIRINOX, those classified as high-risk by the combined score had significantly lower mOS compared to low-risk patients (10.1 vs 11.2 months, log-rank p = 0.008; HR = 2.32, 95% CI: 1.23\u0026ndash;4.39, \u003cem\u003ep\u003c/em\u003e = 0.009). Among patients treated with monotherapy/doublet therapy, those classified as high-risk by the combined score had even more significantly shorter mOS compared to low-risk patients (5.7 vs 10.5 months, log-rank p \u0026lt;0.005; HR = 3.25, 95% CI: 1.91\u0026ndash;5.50, \u003cem\u003ep\u003c/em\u003e \u0026lt;0.005).\u003c/p\u003e\n\u003cp\u003eLogistic regression analyses was conducted to evaluate whether baseline HALP, GRIms and combined scores could predict radiological response to first-line chemotherapy (classified as CR/PR/SD vs PD). Between HALP, GRIms and combined score high/low risk groups, no statistically significant difference in the best responses were recorded. The GRIms and combined score demonstrated a statistically significant association with PD (\u003cem\u003ep\u003c/em\u003e = 0.045; OR = 1.84; 95% CI: 1.02\u0026ndash;3.33 and \u003cem\u003ep\u003c/em\u003e = 0.048, OR=2.07, 95% CI: 0.99\u0026ndash;4.34, respectively). The GRIms and combined score might also serve as a predictive marker of chemotherapy resistance in mPDAC.\u003c/p\u003e"},{"header":"Discussion","content":"\u003cp\u003eThe study identified that systemic immunonutritional scores, notably the GRIms and combined score, surpassed the HALP score in foreshadowing outcomes for advanced PDAC patients. Consistent with preceding studies that a lower HALP score indicates worse survival in mPDAC, our data affirm HALP as a compelling univariate predictor of OS and PFS\u003csup\u003e10,14\u003c/sup\u003e. Nonetheless on multivariate analyses, HALP lost its power compared to GRIms and the combined score, recommending that the added factors in GRIms and their integration with HALP catching additional prognostic ability. This is in line with recent large research in mPDAC which high risk GRIms was related with a 2.3-fold higher risk of death\u003csup\u003e13\u003c/sup\u003e. Our data bolster that integrating nutritional and inflammatory markers provides a more powerful score, recalling the Memorial Sloan Kettering Score (albumin\u0026thinsp;+\u0026thinsp;NLR) which also predicted survival\u003csup\u003e15\u003c/sup\u003e. The separate prognostic value of GRIms and the combined score, indeed after compensating for other variables, underscores their clinical relevance. These scores likely provide as proxies of a host frailty and aggressive tumour microenvironment\u0026ndash; conditions not fully apprehended by PS level alone\u003csup\u003e15,16\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eThe supremacy of the combined score and GRIms could be associated to their inclusion of the systemic inflammatory response.\u003c/p\u003e\u003cp\u003eNLR is a settled marker of inflammation that is cancer-related interacts with cytokine-directed tumour progression and immunosuppressive neutrophil activity, so that high NLR mean imply lower tumour-infiltrating lymphocytes (TIL) which indicates an immune \u0026lsquo;cold\u0026rsquo; microenvironment, and consecutively predicts therapy resistance and lower OS\u003csup\u003e13,17,18\u003c/sup\u003e. In our study, high GRIms scores (high NLR affects GRIms) might coincided to lower CD8\u003csup\u003e+\u003c/sup\u003e TIL levels in tumour histology\u003csup\u003e13\u003c/sup\u003e. Also, meta-analysis data shows that high GRIms level deliberate shorter mPFS and a 2-times higher mortality risk\u003csup\u003e12\u003c/sup\u003e. Inclusion of LDH -a marker of disease burden and endproduct of anaerobic glycolysis\u0026ndash; increases prognostic efficiency\u003csup\u003e19\u003c/sup\u003e. By incorporating GRIms with HALP, combined score catches nutritional and inflammatory condition. Anaemia and malnutrition (HALP\u0026rsquo;s Hb and Alb units) presumably aggravate immune dysfunction and intolerance to treatment, as thrombocytosis and lymphopenia suggest cytokine increase that sustains malady and tumour progression\u003csup\u003e17,20,21\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eHence, patients with worse combined scores display a specifically high-risk phenotype with both compromised host reserves and aggressive tumour biology. This subgroups\u0026rsquo; shorter mPFS and mOS, also the risk of early death was being 2.5\u0026ndash;3.9 times higher on multivariate analysis were seen in this study is consistent with prior analyses showing that lower nutritional scores (Alb, prognostic nutritional index) and elevated inflammation-based scores (CRP/Alb, NLR) separately anticipate worse survival in mPDAC\u003csup\u003e22\u003c/sup\u003e. Unsurprisingly, patients with better combined scores with mOS goes beyond 18 months were also compatible with literature.\u003c/p\u003e\u003cp\u003eAnother purpose was showing HALP\u0026ndash;GRIms combination might also provide predictive vision for response to chemotherapy in mPDAC. We found that patients with high combined scores or high GRIms were naturally more prone to display progressive disease on first-line chemotherapy. Specifically, patients with worse immunonutritional status had lesser probability to benefit from approved regimens. There is coherent evidence for IO in urothelial carcinoma that high GRIms predicts shorter mPFS and lack of response\u003csup\u003e19\u003c/sup\u003e. It also reverberate with a prior study in PDAC that initial NLR\u0026thinsp;\u0026lt;\u0026thinsp;3.1 was a separate predictor of response to first-line gemcitabine-based chemotherapy\u003csup\u003e23\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eBesides, Okabe et al. showed that a dynamic NLR shift was meaningfully related with disease control rate following chemotherapy\u003csup\u003e24\u003c/sup\u003e. Our results additionally revealed a initial combined score might find patients at risk of resistance. Patients with high GRIms presumably have systemic inflammation and tumour-driven granulocytosis that advocates desmoplasia induced immunosuppression, therapy resistance\u003csup\u003e18\u003c/sup\u003e. Granulocyte-derived factors (e.g. IL-1β, IL-6/STAT3 activation) have been demonstrated to weaken chemotherapeutic potency in PDAC\u003csup\u003e17,18\u003c/sup\u003e. So, the primary progressive patients in our study may be induced as an intrinsically chemorefractory tumour microenvironment. However, HALP alone was an inadequate predictor, indicating that inflammatory status is the basic driver of chemosensitivity. These data suggest that immunonutritional scoring could be used in response estimate. Whose HALP score\u0026ndash;GRIms deteriorate on therapy might benefit closer surveillance or early treatment modification, yet since progression becomes apparent.\u003c/p\u003e\u003cp\u003ePatients with both a GRIms rise and HALP decline had the poorest results. Dynamic layering were highly significant, and the combined change presented almost a four-fold hazard of death in univariate analyses.\u003c/p\u003e\u003cp\u003eThis highlights that trajectories of these markers during treatment summarize the response and the tolerance to therapy. Decreasing HALP probably indicates cachexia or nutritional deficits, while increasing GRIms may reflect unresolving tumour inflammation under therapy. In a new prospective study of advanced cancers, whose NLR decreased after palliative care had better mOS compared to those with increasing NLR\u003csup\u003e17\u003c/sup\u003e. In PDAC especially, Datta \u003cem\u003eet al.\u003c/em\u003e detected that NLR fading throughout preoperative chemotherapy correlated with longer survival and trend to major pathologic response\u003csup\u003e18\u003c/sup\u003e. Also, a study in NSCLC IO revealed that whose HALP scores increased during treatment had considerably longer mOS and mPFS\u003csup\u003e25\u003c/sup\u003e.\u003c/p\u003e\u003cp\u003eChecking HALP and GRIms during interval response assessment of first-line therapy might assist to recognize who could respond to early switch to second-line options, these patients may have rising host stress or microscopic progression and threshold of suspicion for progression should be lower. This treatment personalization approach is also shown in a new real-world analyses remains to find CA19-9 and PS as important prognostic factors, but our evidence propose immunonutritional dynamics incorporate litigable elements ahead static baseline levels\u003csup\u003e16\u003c/sup\u003e. Moreover, prospective studies should assess adding HALP/GRIms or other score changes to be a criterion for early modification or for stratifying maintenance vs. switch strategies.\u003c/p\u003e\u003cp\u003eInflammatory indices like GRIms have been utilized to choose patients for IO trials, and it is known that PDAC patients with a significant inflammatory load experience less benefit from both IO\u0026rsquo;s and standard chemotherapy\u003csup\u003e13,19\u003c/sup\u003e. Stratifying through HALP\u0026ndash;GRIms could classify patients who may need auxiliary cachexia-targeted or anti-inflammatory approaches beside chemotherapy. Neutrophil function inhibition, IL-6 blockade, or nutritional support might particularly support the high-risk group. Preclinical studies have concluded that disrupting the inflammation\u0026ndash;immunosuppression cycle (like with CXCR1/2 or CCR2 inhibitors) may recover chemosensitivity in PDAC models\u003csup\u003e18\u003c/sup\u003e. And the HALP\u0026ndash;GRIms could perform as a stratification factor in future trials, assuring fair evaluation of novel treatments among risk subgroups\u003csup\u003e13\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Conclusion","content":"\u003cp\u003eIn summary, our study emphasize that combined use of immunonutritional scores provide superior predictive and prognostic data in mPDAC. The GRIms and combined score more powerfully predicted mPFS and mOS than HALP alone and also predicted radiologic non-response to first-line chemotherapy. Combined scores\u0026rsquo; power being higher than GRIms suggest HALP adds additional capability to GRIms, therefore should be used together. Dynamic changes appeared as effective early indicators of treatment potency, surpassing initial levels and assist identifying patients at risk for rapid progression to tailor therapy appropriately. The scores can be extracted, calculated using hospital-based laboratory test results and automatically presented on electronic health records. Routine combined use of these scores in clinical practice could ameliorate risk stratification above traditional factors, promoting personalized treatment in mPDAC. Our results necessitates prospective validation but provide important evidence to use immunonutritional scores both as prognostic markers and as early response metrics to optimize outcomes in PDAC by demonstrating interaction between tumour biology and host condition, holding serious hope to improve management and trial design in the future\u003csup\u003e13,14,16\u003c/sup\u003e.\u003c/p\u003e"},{"header":"Declarations","content":"\u003cp\u003e\u003ch2\u003eConflict of Interest Statement\u003c/h2\u003e\u003cp\u003eThe authors declare that they have no conflict of interest.\u003c/p\u003e\u003c/p\u003e\u003ch2\u003eAcknowledgements\u003c/h2\u003e\u003cp\u003eThe authors would like to thank the medical and administrative staff of the Ankara City Hospital for their support in facilitating access to patient data and clinical records.\u003c/p\u003e\u003ch2\u003eData Availability Statement\u003c/h2\u003e\u003cp\u003eThe data that support the findings of this study are available from the corresponding author upon reasonable request.\u003c/p\u003e"},{"header":"References","content":"\u003col\u003e\n\u003cli\u003eMizrahi JD, Surana R, Valle JW, Shroff RT. Pancreatic cancer. Lancet 2020;395(10242):2008-2020. (In eng). DOI: 10.1016/s0140-6736(20)30974-0.\u003c/li\u003e\n\u003cli\u003ePark W, Chawla A, O\u0026apos;Reilly EM. Pancreatic Cancer: A Review. Jama 2021;326(9):851-862. (In eng). DOI: 10.1001/jama.2021.13027.\u003c/li\u003e\n\u003cli\u003eLippi G, Mattiuzzi C. The global burden of pancreatic cancer. Arch Med Sci 2020;16(4):820-824. (In eng). DOI: 10.5114/aoms.2020.94845.\u003c/li\u003e\n\u003cli\u003eConroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus Gemcitabine for Metastatic Pancreatic Cancer. New England Journal of Medicine 2011;364(19):1817-1825. DOI: doi:10.1056/NEJMoa1011923.\u003c/li\u003e\n\u003cli\u003eHoff DDV, Ervin T, Arena FP, et al. Increased Survival in Pancreatic Cancer with nab-Paclitaxel plus Gemcitabine. New England Journal of Medicine 2013;369(18):1691-1703. DOI: doi:10.1056/NEJMoa1304369.\u003c/li\u003e\n\u003cli\u003eRochefort P, Lardy-Cleaud A, Sarabi M, Desseigne F, Cattey-Javouhey A, de la Fouchardi\u0026egrave;re C. Long-Term Survivors in Metastatic Pancreatic Ductal Adenocarcinoma: A Retrospective and Matched Pair Analysis. Oncologist 2019;24(12):1543-1548. (In eng). DOI: 10.1634/theoncologist.2018-0786.\u003c/li\u003e\n\u003cli\u003eDell\u0026apos;Aquila E, Fulgenzi CAM, Minelli A, et al. Prognostic and predictive factors in pancreatic cancer. Oncotarget 2020;11(10):924-941. (In eng). DOI: 10.18632/oncotarget.27518.\u003c/li\u003e\n\u003cli\u003eXu SS, Li S, Xu HX, et al. Haemoglobin, albumin, lymphocyte and platelet predicts postoperative survival in pancreatic cancer. World J Gastroenterol 2020;26(8):828-838. (In eng). DOI: 10.3748/wjg.v26.i8.828.\u003c/li\u003e\n\u003cli\u003eBal\u0026ccedil;ık OY, Ayta\u0026ccedil; A, Ekinci F, et al. Clinical Importance of the Hemoglobin, Albumin, Lymphocyte, Platelet Score in Metastatic Pancreatic Cancer. Acta Oncologica Turcica 2023;56(3):250-258. DOI: 10.5505/aot.2023.26779.\u003c/li\u003e\n\u003cli\u003eXu H, Zheng X, Ai J, Yang L. Hemoglobin, albumin, lymphocyte, and platelet (HALP) score and cancer prognosis: A systematic review and meta-analysis of 13,110 patients. Int Immunopharmacol 2023;114:109496. (In eng). DOI: 10.1016/j.intimp.2022.109496.\u003c/li\u003e\n\u003cli\u003eBigot F, Castanon E, Baldini C, et al. Prospective validation of a prognostic score for patients in immunotherapy phase I trials: The Gustave Roussy Immune Score (GRIm-Score). Eur J Cancer 2017;84:212-218. (In eng). DOI: 10.1016/j.ejca.2017.07.027.\u003c/li\u003e\n\u003cli\u003eSun KX, Xu RQ, Rong H, Pang HY, Xiang TX. Prognostic significance of the Gustave Roussy immune (GRIm) score in cancer patients: a meta-analysis. Ann Med 2023;55(2):2236640. (In eng). DOI: 10.1080/07853890.2023.2236640.\u003c/li\u003e\n\u003cli\u003eMa LX, Wang Y, Espin-Garcia O, et al. Systemic inflammatory prognostic scores in advanced pancreatic adenocarcinoma. Br J Cancer 2023;128(10):1916-1921. (In eng). DOI: 10.1038/s41416-023-02214-0.\u003c/li\u003e\n\u003cli\u003eDemir N, G\u0026ouml;kmen İ, Sağdı\u0026ccedil; Karateke Y, et al. HALP score as a prognostic marker for overall survival in advanced pancreatic cancer. Front Oncol 2025;15:1542463. (In eng). DOI: 10.3389/fonc.2025.1542463.\u003c/li\u003e\n\u003cli\u003eLebenthal JM, Zheng J, Glare PA, O\u0026apos;Reilly EM, Yang AC, Epstein AS. Prognostic value of the Memorial Sloan Kettering Prognostic Score in metastatic pancreatic adenocarcinoma. Cancer 2021;127(10):1568-1575. (In eng). DOI: 10.1002/cncr.33420.\u003c/li\u003e\n\u003cli\u003eTaieb J, Seufferlein T, Reni M, et al. Treatment sequences and prognostic/predictive factors in metastatic pancreatic ductal adenocarcinoma: univariate and multivariate analyses of a real-world study in Europe. BMC Cancer 2023;23(1):877. DOI: 10.1186/s12885-023-11377-1.\u003c/li\u003e\n\u003cli\u003eZhao W, Wu Z, Li Y, et al. Pretreatment neutrophil-to-lymphocyte ratio and its dynamic changes are associated with the overall survival in advanced cancer patients undergoing palliative care. Scientific Reports 2016;6(1):31394. DOI: 10.1038/srep31394.\u003c/li\u003e\n\u003cli\u003ede Castro Silva I, Bianchi A, Deshpande NU, et al. Neutrophil-mediated fibroblast-tumor cell il-6/stat-3 signaling underlies the association between neutrophil-to-lymphocyte ratio dynamics and chemotherapy response in localized pancreatic cancer: A hybrid clinical-preclinical study. Elife 2022;11 (In eng). DOI: 10.7554/eLife.78921.\u003c/li\u003e\n\u003cli\u003eTanabe K, Kobayashi S, Maezawa Y, et al. Gustave Roussy Immune score as a prognostic biomarker in patients with platinum-refractory metastatic urothelial carcinoma treated with pembrolizumab: YUSHIMA study. Int J Clin Oncol 2024;29(9):1302-1310. (In eng). DOI: 10.1007/s10147-024-02563-7.\u003c/li\u003e\n\u003cli\u003eHan R, Tian Z, Jiang Y, et al. Prognostic significance of the systemic immune inflammation index in patients with metastatic and unresectable pancreatic cancer. Frontiers in Surgery 2022;Volume 9 - 2022 (Original Research) (In English). DOI: 10.3389/fsurg.2022.915599.\u003c/li\u003e\n\u003cli\u003eG\u0026Uuml;LbaĞCi BB, \u0026Ouml;Zen M, \u0026Ouml;Zen EngİN E, AkdaĞ KahvecİOĞLu F, \u0026Ccedil;İFt\u0026Ccedil;İ E, HacibekİRoĞLu İ. Does Delta Hemoglobin-Albumin-Lymphocyte Platelet (HALP) Score Predict the Risk of Early Progression in Patients Treated with CDK4/6 Inhibitors? Namık Kemal Tıp Dergisi 2024;12(4):260-265. DOI: 10.4274/nkmj.galenos.2024.12599.\u003c/li\u003e\n\u003cli\u003eHackner D, Merkel S, Wei\u0026szlig; A, et al. Neutrophil-to-Lymphocyte Ratio and Prognostic Nutritional Index Are Predictors for Overall Survival after Primary Pancreatic Resection of Pancreatic Ductal Adenocarcinoma: A Single Centre Evaluation. Cancers 2024;16(16):2911. (https://www.mdpi.com/2072-6694/16/16/2911).\u003c/li\u003e\n\u003cli\u003eKitsugi K, Kawata K, Noritake H, et al. Prognostic value of neutrophil to lymphocyte ratio in patients with advanced pancreatic ductal adenocarcinoma treated with systemic chemotherapy. Ann Med 2024;56(1):2398725. (In eng). DOI: 10.1080/07853890.2024.2398725.\u003c/li\u003e\n\u003cli\u003eOkabe H, Masuda T, Tomita M, et al. Combined Neutrophil-to-Lymphocyte Ratio Score Is Associated With Chemotherapeutic Response and Predicts Prognosis in Patients With Advanced Pancreatic Cancer. Anticancer Res 2024;44(4):1575-1582. (In eng). DOI: 10.21873/anticanres.16955.\u003c/li\u003e\n\u003cli\u003eKo\u0026ccedil;anoğlu A, Karakaya S, Zeynelgil E, D\u0026uuml;zk\u0026ouml;pr\u0026uuml; Y, Doğan \u0026Ouml;. Dynamic Alteration of HALP Score as a Predictor in Patients with Receiving Immunotherapy for Advanced Non-Small Cell Lung Cancer. Medicina 2025;61(6):989. (https://www.mdpi.com/1648-9144/61/6/989).\u003c/li\u003e\n\u003c/ol\u003e"},{"header":"Tables","content":"\u003cp\u003eTable 1: Baseline characteristics\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCharacteristic\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eValue\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNumber of patients (n)\u003c/strong\u003e\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e201\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAge at diagnosis (years)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Mean \u0026plusmn; SD (range)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e62.9 \u0026plusmn; 10.7 (33\u0026ndash;89)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eSex\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Male\u003cbr\u003e\u0026nbsp;Female\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e120 (59.7%)\u003cbr\u003e\u0026nbsp;81 (40.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eECOG PS\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; 0\u0026ndash;1\u003cbr\u003e\u0026nbsp;2\u0026ndash;3\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e122 (60.7%)\u003cbr\u003e\u0026nbsp;79 (39.3%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eFollow-up duration (months)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e23.2 (3.83\u0026ndash;42.57)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eChemotherapy regimen\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Triplet (FOLFIRINOX)\u003cbr\u003e\u0026nbsp;Monotherapy/Doublet\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003cp\u003e89 (44.3%)\u003cbr\u003e\u0026nbsp;112 (55.7%)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eAlbumin level (g/L)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e40.40 (34.0\u0026ndash;42.1)\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eLDH (U/L)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e223.50 (136.8\u0026ndash;224.6)\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eHemoglobin (g/dL)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e12.70 (11.0\u0026ndash;13.3)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eNLR\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e3.13 (2.21\u0026ndash;5.22)\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCEA (ng/mL)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e3.65 (0.1\u0026ndash;12.88)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eCA 19-9 (U/mL)\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e627.10 (85.5\u0026ndash;4832.7)\u003cbr\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInitial HALP score\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e30.66 (28.99)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"top\" style=\"width: 245px;\"\u003e\n \u003cp\u003e\u003cstrong\u003eInitial GRIms\u003c/strong\u003e\u003c/p\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp; \u0026nbsp; Median (IQR)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 236px;\"\u003e\n \u003cp\u003e1.00 (1.00)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eTable 2: Univariate Cox regression models for initial HALP, GRIms and combined scores:\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"576\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eScore\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 132px;\"\u003e\n \u003cp\u003eOS HR (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e-value OS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"top\" style=\"width: 132px;\"\u003e\n \u003cp\u003ePFS HR (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e-value PFS\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHALP\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 132px;\"\u003e\n \u003cp\u003e1.373 (1.015-1.857)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;0.04\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 132px;\"\u003e\n \u003cp\u003e1.339 (1.012-1.770)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;0.041\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eGRIm\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 132px;\"\u003e\n \u003cp\u003e2.410 (1.730-3.356)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 132px;\"\u003e\n \u003cp\u003e1.960 (1.448-2.653)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eCombined Score\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 132px;\"\u003e\n \u003cp\u003e2.641 (1.766-3.948)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 132px;\"\u003e\n \u003cp\u003e1.868 (1.289-2.707)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 66px;\"\u003e\n \u003cp\u003e\u0026nbsp; \u0026nbsp;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eTable 3: Multivariate Cox regression models for initial values and interval (follow-up) HALP, GRIms, combined score; adjusted for age, ECOG PS, chemotherapy regimen, and tumour markers:\u0026nbsp;\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"578\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003eAdjusted HR (95% CI) for OS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e-value OS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003eAdjusted HR (95% CI) for PFS\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u003cem\u003ep\u003c/em\u003e-value PFS\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eCombined Score (High vs Low)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e2.524 (1.677-3.801)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.804 (1.240-2.627)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.002\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eGRIms (High vs Low)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e2.330 (1.670-3.260)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.985 (1.463\u0026ndash;2.702)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHALP (High vs Low)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.290 (0.950\u0026ndash;1.760)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026nbsp; 0.108\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.290 (0.970\u0026ndash;1.710)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.078\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eECOG\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.759 (1.216-2.544)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.003\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.321 (0.954-1.830)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.094\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eCA 19-9 (High vs Low)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.490 (1.084-2.046)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.014\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.482 (1.109-1.981)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.008\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eAge\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.169 (0.842-1.623)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.352\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.181 (0.874-1.595)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.278\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eCEA (High vs Low)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e0.997 (0.730-1.362)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.986\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.091 (0.817-1.457)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.555\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eChemotherapy Regimen\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.000 (0.686-1.457)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.999\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e1.242 (0.892-1.728)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e0.199\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 180px;\"\u003e\n \u003cp\u003eInterval HALP\u0026nbsp;(High vs Low)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 135px;\"\u003e\n \u003cp\u003e2.402 (1.750-3.290)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 180px;\"\u003e\n \u003cp\u003eInterval GRIm\u0026nbsp;(High vs Low)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 135px;\"\u003e\n \u003cp\u003e5.481 (3.690-8.142)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd style=\"width: 180px;\"\u003e\n \u003cp\u003eInterval Combined Score\u0026nbsp;(High vs Low)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 135px;\"\u003e\n \u003cp\u003e5.904 (4.051-8.594)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026lt;0.001\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 64px;\"\u003e\n \u003cp\u003e\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u003cbr\u003e\u003c/p\u003e\n\u003cp\u003eTable 4: Univariate Association of Post-Treatment Score Changes with mOS and mPFS\u003c/p\u003e\n\u003ctable border=\"1\" cellspacing=\"0\" cellpadding=\"0\" width=\"586\"\u003e\n \u003ctbody\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eVariable\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003emOS (months)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003emPFS (months)\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003eUnivariate Cox HR (95% CI)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eDecrease in HALP\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e8.87 vs 12.21\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e4.97 vs 7.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e2.34 (1.741\u0026ndash;3.138)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eIncrease in GRIms\u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp; \u0026nbsp;\u0026nbsp;\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e7.54 vs 12.38\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e4.07 vs 7.67\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e2.48 (1.842\u0026ndash;3.342)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003ctr\u003e\n \u003ctd valign=\"bottom\" style=\"width: 180px;\"\u003e\n \u003cp\u003eHigher Combined score\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e7.36 vs 11.93\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e3.60 vs 7.40\u003c/p\u003e\n \u003c/td\u003e\n \u003ctd valign=\"bottom\" style=\"width: 135px;\"\u003e\n \u003cp\u003e3.85 (2.720\u0026ndash;5.442)\u003c/p\u003e\n \u003c/td\u003e\n \u003c/tr\u003e\n \u003c/tbody\u003e\n\u003c/table\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u0026nbsp;\u003c/p\u003e"}],"fulltextSource":"","fullText":"","funders":[],"hasAdminPriorityOnWorkflow":false,"hasManuscriptDocX":true,"hasOptedInToPreprint":true,"hasPassedJournalQc":"","hasAnyPriority":false,"hideJournal":true,"highlight":"","institution":"","isAcceptedByJournal":false,"isAuthorSuppliedPdf":false,"isDeskRejected":"","isHiddenFromSearch":false,"isInQc":false,"isInWorkflow":false,"isPdf":false,"isPdfUpToDate":true,"isWithdrawnOrRetracted":false,"journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true},"keywords":"","lastPublishedDoi":"10.21203/rs.3.rs-7277845/v1","lastPublishedDoiUrl":"https://doi.org/10.21203/rs.3.rs-7277845/v1","license":{"name":"CC BY 4.0","url":"https://creativecommons.org/licenses/by/4.0/"},"manuscriptAbstract":"\u003cp\u003e\u003cstrong\u003eBackground\u003c/strong\u003e: In the challenging landscape of metastatic pancreatic ductal adenocarcinoma (mPDAC), clinicians urgently need accessible and reliable tools to predict outcomes and guide treatment. Immunonutritional indices like the Gustave Roussy Immune Score (GRIms) and the hemoglobin, albumin, lymphocyte, and platelet (HALP) index offer promise, but their full prognostic/predictive potential remained unclear.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eMethods\u003c/strong\u003e: We retrospectively analysed 201 patients with mPDAC who received first-line chemotherapy between 2020 and 2024. GRIms and HALP scores were calculated at baseline and at interval using routine blood tests. We assessed their relationship with overall survival (OS) and progression-free survival (PFS) through univariate and multivariate Cox regression models.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eResults\u003c/strong\u003e: Patients with high GRIms or low HALP at baseline had significantly shorter OS and PFS (all p \u0026lt; 0.05), with GRIms emerging as a stronger independent predictor (HR for OS: 2.3, p \u0026lt; 0.001). When combined, these two scores offered even greater prognostic precision, clearly identifying the highest-risk patient for mOS (HR: 2.5, p \u0026lt; 0.001). Dynamic changes also mattered; patients whose GRIms increased or HALP declined over time had notably worse survival (mOS: 7 vs 12 months; HR: 3.9, p \u0026lt; 0.001) with combined use suggest stronger prediction (HR: 5.9, p\u0026lt;0.001). A high combined score was also associated with early chemotherapy resistance and poor radiologic response.\u0026nbsp;\u003c/p\u003e\n\u003cp\u003e\u003cstrong\u003eConclusion\u003c/strong\u003e: GRIms and HALP are inexpensive, non-invasive, and readily available markers that offer stronger prognosis in mPDAC. Their combined and longitudinal use may improve clinical decision-making, help flag patients at high risk, and ultimately support more personalized and proactive treatment approaches.\u003c/p\u003e","manuscriptTitle":"HALP–GRIms Synergy Offers Superior Prognostic and Predictive Precision in Advanced Pancreatic Ductal Adenocarcinoma","msid":"","msnumber":"","nonDraftVersions":[{"code":1,"date":"2025-08-22 10:19:42","doi":"10.21203/rs.3.rs-7277845/v1","editorialEvents":[{"type":"communityComments","content":0}],"status":"published","journal":{"display":true,"email":"[email protected]","identity":"researchsquare","isNatureJournal":false,"hasQc":true,"allowDirectSubmit":true,"externalIdentity":"","sideBox":"","snPcode":"","submissionUrl":"/submission","title":"Research Square","twitterHandle":"researchsquare","acdcEnabled":true,"dfaEnabled":false,"editorialSystem":"","reportingPortfolio":"","inReviewEnabled":false,"inReviewRevisionsEnabled":true}}],"origin":"","ownerIdentity":"18592e7a-0630-4520-8b4d-877219b4393d","owner":[],"postedDate":"August 22nd, 2025","published":true,"recentEditorialEvents":[],"rejectedJournal":[],"revision":"","amendment":"","status":"posted","subjectAreas":[],"tags":[],"updatedAt":"2025-08-26T03:49:31+00:00","versionOfRecord":[],"versionCreatedAt":"2025-08-22 10:19:42","video":"","vorDoi":"","vorDoiUrl":"","workflowStages":[]},"version":"v1","identity":"rs-7277845","journalConfig":"researchsquare"},"__N_SSP":true},"page":"/article/[identity]/[[...version]]","query":{"redirect":"/article/rs-7277845","identity":"rs-7277845","version":["v1"]},"buildId":"8U1c8b4HqxoKbykW_rLl7","isFallback":false,"isExperimentalCompile":false,"dynamicIds":[84888],"gssp":true,"scriptLoader":[]}