Glasgow prognostic score serve as a predictor of clinical outcome in p

Introduction

Multiple myeloma (MM) is a heterogeneous, hematological malignancy, which is characterized by monoclonal proliferation of malignant plasma cells accumulated in the bone marrow and production of abnormal monoclonal immunoglobulin.1 Multiple myeloma, the second most prevalent blood cancer in high-income countries, has an increasing global incidence rate, which is approximately 6–7 per 100,000 people per year.2,3

Although, with the application of novel agents, including proteasome inhibitors,4 immunomodulatory drugs,5 CD38 targeting antibodies,6 as well as CAR-T cell therapy,7 bispecific antibodies,8 and immune checkpoint inhibitors,9 the life expectancy of MM has achieved encouraging improvement, but the majority of MM patients will experience the relapses or progression of the disease, for which MM is still considered to be incurable. For that reason, it is important to perform risk stratification within the MM population for their individualized precision therapy. Nowadays, several staging systems, including Durie-Salmon (DS) stage,10 International Staging System (ISS),11 the Revised ISS (R-ISS),12 and Mayo Stratification of Myeloma and Risk-Adapted Therapy (mSMART)13 have been used to guide clinical work. However, even for R-ISS, the most widely applied staging system, its capability of predicting survival or performing stratification of patients is still limited.14 Therefore, a new staging system with convenient indicators that improve the accuracy and sensitivity of prognosis is highly required for precision treatment in MM.

The Glasgow prognostic score (GPS)15 is a simple index that expresses the degree of inflammation within the body. In this score, there are two biomarkers, C-reactive protein (CRP) and serum albumin. CRP is an acute-phase protein that reflects the systemic inflammation levels. Furthermore, the serum concentration of CRP is negatively correlated with prognosis within a variety of cancers, probably because it is modulated by several proinflammatory cytokines like IL-1, IL-6, and TNF-ɑ.16 And for serum albumin, research has indicated that hypoalbuminemia is associated with poor survival chances.17 GPS ranges from 0 to 2 points, with 1 point assigned for each of the following: increasing CRP level (>10mg/L) and hypoalbuminemia (<35g/L). The modified GPS (mGPS) and the high-sensitivity modified GPS (Hs-mGPS) are two modified versions of the original GPS. mGPS highlights the significance of CRP, where cases with hypoalbuminemia but without elevated CRP levels remain scored as 0 points.15 Hs-mGPS, which has a more sensitive cut-off value for CRP (3mg/L), has been proven to enhance the prognostic value of the mGPS.18

Several previous studies have proven that GPS has the capabilities of predicting patients’ survival in various types of cancers, including non-small cell lung cancer,19 colorectal cancer,20 prostate cancer21 and so on. However, in MM, the prognostic value of GPS has not been verified yet. Considering that immunity and inflammatory response contribute to the progression of MM,22 we assume that GPS may also be applicable for prognostic prediction in MM. Therefore, the current study aimed to validate the prognostic value of GPS in a large population of newly diagnosed multiple myeloma (NDMM).

Patients and Methods
Patient Selection

In this retrospective study, the clinical and laboratory data of a total of 865 patients with NDMM between July 2001 and August 2021 were collected and analyzed. The specific inclusion criteria for our study were as follows: 1) age ≥18 years; 2) newly diagnosed multiple myeloma based on the International Myeloma Working Group (IMWG) diagnostic criteria; 3) all patients should receive antineoplastic therapy; 4) complete baseline clinical and laboratory data were available before treatment. Specific data were shown in the Results section. All these data were collected from the electronic medical record. Patients with any exclusion criteria as follows were excluded from this study: 1) diagnosed with MM and comorbid with other neoplastic disease; 2) had received antineoplastic therapy before being diagnosed in our center. 3) did not receive any antineoplastic therapy after being diagnosed; 4) diagnosed with other plasma cell dyscrasias such as Waldenström macroglobulinemia, primary AL amyloidosis, systemic light chain deposition disease; 5) lacked sufficient baseline data for analysis. All the patients were divided into three subgroups based on their GPS at diagnosis, and these groups were designated as GPS-0, GPS-1, and GPS-2 in the subsequent text.

The primary endpoint we measured was overall survival (OS), which was defined as the time from the date of diagnosis to the date of death for any reason or the time of last follow-up, and the progression-free survival (PFS) was defined as the time interval between the date of diagnosis to the date of progression based on the IMWG criteria, death from any cause, or the last follow-up.

GPS/mGPS/Hs-mGPS Calculation

Table 1 presents the specific calculation rules for the scores of GPS, mGPS, and Hs-mGPS.

Table 1 The Scoring Criteria for GPS, mGPS, and Hs-mGPS

Statistical Analysis

All clinical and laboratory data were presented as categorical variables and were analyzed by the chi-square test. To assess the relationship between baseline GPS/mGPS/Hs-mGPS and the survival outcome (OS and PFS), Kaplan-Meier curves were drawn and compared by the Log rank test. The median follow-up, median OS, median PFS, five-year OS rate, and five-year PFS rate were obtained through Kaplan-Meier analysis. Univariate (UVA) and multivariate (MVA) analyses were carried out using the Cox proportional hazard model for OS and PFS. The performance of each GPS variant was assessed by area under the curve (AUC) values at 1, 3, and 5 years, based on time-dependent receiver operating characteristic (ROC) analysis. A two-sided p-value less than 0.05 was considered statistically significant. All statistical analyses were conducted using R 4.3.1.

Results
Patient Characteristics

A total of 865 eligible patients were included in this study. Patients’ characteristics are provided in Table 2. The median age was 60 years, and 83.5% of patients were over 50 years old when diagnosed. Of the included patients, 60.2% were male. 89.2% of patients showed a good performance status (ECOG PS 0~1).

Table 2 Clinical and Laboratory Characteristics for All Patients Based on GPS Level

The median GPS at diagnosis was 1, and at baseline, the number of patients with GPS of 0, 1, and 2 were 402, 343, and 120, respectively. Patients with high GPS were more likely to have a worse performance status (ECOG PS ≥2 (GPS ranges from 0 to 2): 6.2% vs 12.8% vs 20.0%, p<0.001), higher DS stage (percentage of DS III (GPS ranges from 0 to 2): 65.2% vs 79.6% vs 85.0%, p<0.001), and higher ISS stage (percentage of ISS III (GPS ranges from 0 to 2): 29.6% vs 46.6% vs 60.0%, p<0.001). Among all the patients, the patients with high GPS at baseline had lower β2-MG level (β2-MG≤3.5g/L (GPS ranges from 0 to 2): 38.1% vs 66.5% vs 80.0%, p<0.001) and higher LDH level (LDH>265U/L (GPS ranges from 0 to 2): 7.5% vs 12.5% vs 23.3%, p<0.001). Compared with the lower GPS group, the group with high GPS tended to have impaired renal function, manifesting as elevated creatinine level (CRE> upper limit of normal (GPS ranges from 0 to 2): 20.1% vs 29.7% vs 43.3%, p<0.001). However, we found that patients with high GPS were less frequently comorbid with extramedullary disease (percentage of EMD (GPS ranges from 0 to 2): 26.6% vs 20.7% vs 15.8%, p=0.024). In the following treatment, 12.5% of patients received a transplant, and the chemotherapy regimens patients received were proteasome-based regimens in 29.4%, immunomodulatory-based regimens in 23.3%, regimens combining proteasome and immunomodulatory agents in 23.7%, and other types of chemotherapy in 23.6%.

Baseline GPS Correlated with Survival Outcomes in Kaplan-Meier Analysis

Firstly, we evaluated the association between patients’ survival outcomes and baseline GPS in three subgroups. The median follow-up of the study was 47 months. For all 865 patients, the median OS was 58 months (95% CI, 53–66), and the median PFS was 35 months (95% CI, 30–39), as shown in Figure 1. The 5-year OS of patients with baseline GPS of 0, 1, and 2 were 56.6%, 42.2%, and 37.9%, respectively. The 5-year PFS of patients in three subgroups were 39.8%, 32.7%, and 24.3%, respectively.

Figure 1 Survival curves for all patients (n=865) obtained with Kaplan-Meier analysis. (A) Survival curves for OS. (B) Survival curves for PFS.

Considering that GPS is a categorical variable, Kaplan-Meier survival curves were constructed for each subgroup. And we applied the Log rank test to compare the OS and PFS between the three GPS subgroups, for which we confirmed that the higher the GPS at baseline was, the worse the survival outcomes would be. (median OS (GPS ranges from 0 to 2): 77 months (95% CI, 63–138) vs 48 months (95% CI, 42–60) vs 44 months (95% CI, 31–62), p<0.001; median PFS (GPS ranges from 0 to 2): 42 months (95% CI, 36–55) vs 27 months (95% CI, 23–38) vs 24 months (95% CI, 17–36), p<0.001; Figure 2).

Figure 2 Survival curves obtained with Kaplan-Meier analysis between different GPS groups. (A) Survival curves for OS. (B) Survival curves for PFS.

Baseline GPS Serves as an Independent Prognostic Factor of PFS and OS in Univariate and Multivariate Analyses

Tables 3 and 4 demonstrated the correlation derived from Cox regression analyses between multiple variables, including GPS and survival outcomes (PFS and OS), for both univariable and multivariable assessments. In UVA, GPS was significantly associated with both OS and PFS. (OS: HR 1.721 (95% CI 1.394–2.125), p<0.001, Table 3; PFS: HR 1.487 (95% CI 1.233–1.794), p<0.001, Table 4) The result of following MVA revealed that baseline GPS is an independent prognostic factor for OS and PFS in patients with newly diagnosed multiple myeloma. (OS: HR 1.290 (95% CI 1.025–1.623), p=0.030, Table 3; PFS: HR 1.245 (95% CI 1.017–1.524), p=0.034, Table 4) Besides GPS, poor ECOG PS (OS: HR 1.842 (95% CI 1.399–2.425), p<0.001, Table 3; PFS: HR 1.558 (95% CI 1.197–2.027), p<0.001, Table 4) as well as not receiving transplant (OS: HR 0.578 (95% CI 0.387–0.862), p=0.007, Table 3; PFS: HR 0.693 (95% CI 0.505–0.953), p=0.024, Table 4) and LDH>265U/L (OS: HR 1.838 (95% CI 1.402–2.410), p<0.001, Table 3; PFS: HR 1.891 (95% CI 1.471–2.432), p<0.001, Table 4) were found to be independent risk factors for both OS and PFS in MVA. However, we found that the DS stage and ISS, which are widely used in clinical practice, only showed differences among the three groups in UVA, but not in MVA (Table 3 and Table 4).

Table 3 Univariate and Multivariate Analysis of Variables Associated with Overall Survival

Table 4 Univariate and Multivariate Analysis of Variables Associated with Progression-Free Survival

GPS Has Better Predictive Value Than mGPS and Hs-mGPS

To compare the predictive capacity of GPS, mGPS, and Hs-mGPS systems, we generated Kaplan-Meier curves stratified by mGPS and Hs-mGPS, respectively, with inter-group differences assessed with the Log rank test. Figure 3A and B showed the survival curves of GPS. Although overall statistically significant differences in OS and PFS were observed across the subgroups classified by both mGPS (OS: p<0.001; PFS: p=0.003; Figure 3C and D) and Hs-mGPS (OS: p<0.001; PFS: p=0.002; Figure 3E and F), the survival curves of mGPS and Hs-mGPS demonstrated closer distance within adjacent curves and more frequent crossings compared to the GPS-based stratified curves. Subsequent pairwise comparisons revealed no significant prognostic distinction between adjacent mGPS categories (0 vs 1, 1 vs 2) in either OS or PFS analyses (Figure 3C and D), indicating that mGPS has suboptimal predictive performance compared to GPS or Hs-mGPS. As shown in Figure 3F, the PFS curves for Hs-mGPS scores 0 and 1 demonstrated significant overlap (p = 0.882), indicating compromised discriminative capacity of Hs-mGPS in distinguishing low-risk from intermediate-risk subgroups. In the meanwhile, GPS also showed limited capacity to differentiate between intermediate-risk and high-risk subgroups, but the corresponding survival curves exhibited distinct separation trends with minimal curve crossover (Figure 3A and B).

Figure 3 Survival curves and inter-group differences analysis results obtained with Kaplan-Meier analysis between different GPS/mGPS/Hs-mGPS groups. (A) Survival curves for OS between different GPS groups. (B) Survival curves for PFS between different GPS groups. (C) Survival curves for OS between different mGPS groups.(D) Survival curves for PFS between different mGPS groups.(E) Survival curves for OS between different Hs-mGPS groups.(F) Survival curves for PFS between different Hs-mGPS groups.

We subsequently constructed time-dependent ROC curves for GPS, mGPS, and Hs-mGPS, with corresponding AUC values calculated (Figure 4). Except for AUC values from ROC curves for 3-year OS (GPS: 58.41% vs Hs-mGPS: 58.46%), GPS consistently achieved higher AUC values than both mGPS and Hs-mGPS at all other timepoints for both OS and PFS predictions (Figure 4). Additionally, the AUC values of mGPS and Hs-mGPS declined more sharply over time than those of GPS. To compare the prognostic accuracy of GPS, mGPS, and Hs-mGPS for short-term and long-term survival, we selected 1-year and 5-year ROC curves for subsequent comparative analysis (Figure 5). For short-term survival prediction, GPS demonstrated no statistically significant advantage over mGPS or Hs-mGPS (OS: GPS vs mGPS: p= 0.092; GPS vs Hs-mGPS: p= 0.480; PFS: GPS vs mGPS: p= 0.105; GPS vs Hs-mGPS: p= 0.177; Figure 5A and B). In contrast, for long-term survival prognosis, GPS exhibited significantly superior discriminatory power, particularly in PFS prediction (OS: GPS vs mGPS: p= 0.002; GPS vs Hs-mGPS: p= 0.094; PFS: GPS vs mGPS: p= 0.035; GPS vs Hs-mGPS: p= 0.021; Figure 5C and D).

Figure 4 Time-dependent ROC curves and time-dependent AUC values of the GPS/mGPS/Hs-mGPS for predicting OS (A–C) and PFS (D–F). (A and D) ROC curves and AUC values of GPS. (B and E) ROC curves and AUC values of mGPS. (C and F) ROC curves and AUC values of Hs-mGPS.

Figure 5 Comparison of prognostic performance for predicting OS (A and B) and PFS (C and D) between GPS and mGPS or Hs-mGPS. (A and C) in terms of short-term survival. (B and D) in terms of long-term survival.

Comprehensive evaluation of these prognostic models demonstrated that GPS possesses superior predictive validity in clinical outcome stratification.

Discussion

In this study, we evaluated the prognostic value of GPS in a large, real-world cohort of 865 NDMM patients. Our findings indicate that a higher GPS is significantly associated with inferior OS and PFS. Importantly, this association remained robust in multivariate analyses adjusting for confounding factors, indicating that GPS serves as an independent prognostic marker in NDMM.

Our findings extend previous mechanistic studies demonstrating inflammation to adverse prognosis in NDMM. Elevated inflammatory mediators, including CRP, IL-18, and IL-17, promote disease progression through multiple pathways that include the inhibition of CD8+ T-cell function and the induction of myeloid-derived suppressor cells (MDSCs).23–26 These findings provide biological plausibility for the adverse impact of a heightened inflammatory state on patient outcomes. Based on these proven mechanisms, many prognostic models based on various inflammatory indicators such as CRP, lactate dehydrogenase (LDH), and immune cell ratio have been developed.27–29 These inflammatory prognostic models have shown better risk stratification ability compared to the traditional International Staging System (ISS) and revised ISS (R-ISS).

The GPS is also a prognostic assessment model based on the systemic inflammatory response, incorporating serum CRP and serum albumin levels. It is one of the most useful prognostic models for various common solid tumors.19,20 Recent studies have also shown that GPS is an independent prognostic factor of PFS and OS in patients with hematologic malignancies, including diffuse large B-cell lymphoma,30 NK/T-cell lymphoma,31 and Hodgkin lymphoma.32 In the context of multiple myeloma, our study further extends these observations. We found that patients with elevated GPS scores also presented with more advanced disease stages, as reflected by higher Durie-Salmon and ISS classifications. This suggests that the inflammatory milieu, as captured by the GPS, is intricately linked with disease burden and progression. We demonstrated that GPS is significantly associated with survival outcomes, including OS and PFS. Patients with higher GPS scores exhibited worse survival outcomes, a correlation that remained robust even after adjusting for confounders in MVA.

Furthermore, the original GPS shows better prognostic predictive value than its two modified versions, mGPS and Hs-mGPS. Several studies reported that hypoalbuminemia harms the survival outcome of MM patients.33,34 However, under the scoring rules of mGPS and Hs-mGPS, the role of hypoproteinemia is excluded when identifying the low-risk patient groups, which may lead to some intermediate-risk patients who present with hypoalbuminemia but non-elevated CRP being incorrectly classified into the low-risk group. In our cohort, among 343 patients classified as GPS-1, over 70% (246/343) were classified into the low-risk group by mGPS. The exclusion of hypoalbuminemia would contribute to the overlap of the curves between the low-risk and intermediate-risk groups. Furthermore, under the mGPS stratification framework, only 25.1% (217/865) of patients were categorized into intermediate/high-risk groups. Previous studies have indicated that the majority of patients were assigned to the low-risk group by mGPS, thereby compromising the prognostic stratification capability of mGPS.35 And for Hs-mGPS, it adopts a stricter CRP threshold, which to some extent enhances the discrimination capability for identifying the intermediate- and high-risk groups. Hs-mGPS may also demonstrate potential for identifying patients with exceptional prognosis.18 As shown in Figure 3E and F, employing high-sensitivity CRP indeed enhanced the identification of high-risk patients but still failed to distinguish intermediate-risk patients from low-risk groups. Collectively, we propose that optimizing the thresholds for CRP and albumin may significantly enhance the prognostic performance of these readily available biomarkers. In summary, these findings highlight the potential of GPS as an independent prognostic marker for MM and suggest its value in enhancing the current risk stratification system.

Besides the prognostic impact of GPS, this score has actionable clinical importance and can be assessed quickly. GPS only requires CRP and serum albumin, two easily available clinical biomarkers, to complete the risk assessment of patients. GPS has advantages over R-ISS, which has widely used clinical staging systems. Calculate the R-ISS, requiring a cytogenetic diagnosis, which causes a certain delay, while the components of the GPS are always available within the initial blood sampling range. Furthermore, the collected biomarkers used for calculating R-ISS were less representative in reflecting general inflammatory activity, while GPS can reflect the inflammatory state of the organism. Considering that the systemic inflammation level undergoes dynamic evolution throughout tumorigenesis and tumor progression of MM,36 we wonder whether GPS has the potential to serve as a dynamic prognostic model for MM. Firstly, it is simple and feasible to conduct dynamic monitoring of GPS because of its data accessibility. Secondly, several studies have confirmed that dynamic changes of GPS or its modified variants exhibit predictive capacity for disease progression in some other cancers.37–39 However, it is regrettable that our study only collected baseline GPS values from MM patients before treatment initiation. In future work, we intend to gather GPS values at multiple time points, including post-treatment remission and disease progression. Furthermore, following the methodological blueprint of the DIPSS model,40 we are going to evaluate the acquisition of risk factors (elevated CRP and hypoalbuminemia) during follow-up. Integrating these serial GPS values with treatment response, survival status, and other clinical variables, we will comprehensively explore the potential of GPS as a dynamic prognostic model.

Our study has several limitations. First, this is a single-center retrospective study; therefore, selection bias was inevitable, and our results must be validated in large-scale, prospective studies conducted across multiple centers. Secondly, this is a retrospective study with a large period. Due to differences in treatment protocols across regions and variations in individual patient treatment regimens, these factors may contribute to differences in survival outcomes. Lastly, this research did not include the cytogenetic data. Since the 2010s, with the role of high-risk cytogenetics being elucidated,41,42 FISH has gradually become more widely used in clinical practice.43,44 However, the wide period of the initial diagnosis of the enrolled patients in our study means that many patients did not undergo FISH or other cytogenetic tests. Therefore, whether there exists an association between GPS and high-risk cytogenetic results, and whether combining GPS with cytogenetic data demonstrates better prognostic stratification capability, needs to be further elucidated in future studies.

In summary, our study provides novel evidence that the Glasgow Prognostic Score is an independent prognostic marker in NDMM, highlighting the pivotal role of systemic inflammation in disease progression. Given its simplicity and rapid availability, the GPS has the potential to enhance risk stratification and guide therapeutic strategies in clinical practice.

Conclusion

The baseline GPS at the time of diagnosis is an independent prognostic factor that negatively relates to the MM patient survival, for which GPS may serve as a supplement tool in risk stratification upon the primary medical assessment.

Data Sharing Statement

The data of this study can be obtained from the corresponding author Yang Liang upon reasonable request.

Ethics Approval and Informed Consent

This study got the approval of the Institutional Ethics Committee of Sun Yat-sen University Cancer Center. According to the regulations of the Ethics Committee, the criteria for granting a waiver of informed consent included: 1) all human specimens and data for research are previously collected and properly preserved; 2) all included patients are uncontactable; 3) the research involves no risks to personal privacy or commercial interests; 4) all included patients had previously provided informed consent for authorizing the use of donated materials and associated information for all medical research purposes. The requirement for informed consent was waived because it was a retrospective study and all included patients were no longer contactable The waiver of informed consent will not adversely affect the rights or health of the participants, and we covered patient data confidentially. This retrospective study was performed in accordance with the Declaration of Helsinki.

Funding

YL is supported by Beijing Xisike Clinical Oncology Research Foundation (Y-SYBLD2022MS-0140) and Guangdong Basic and Applied Basic Research Foundation (Grant No. 2025A1515010717). YW is supported by the National Natural Science Foundation of China (Grant No. 82470162), the Guangdong Basic and Applied Basic Research Foundation (Grant No. 2025A1515012632), and Beijing Xisike Clinical Oncology Research Foundation (Y-Young2022–0281).

Disclosure

The authors report no conflicts of interest in this work.

References

1. van de Donk N, Pawlyn C, Yong KL. Multiple myeloma. Lancet. 2021;397(10272):410–427. doi:10.1016/S0140-6736(21)00135-5

2. Carreras E, Dufour C, Mohty M, Kröger N. The EBMT Handbook: Hematopoietic Stem Cell Transplantation and Cellular Therapies. Cham (CH): SpringerCopyright 2019, The Editor(s) (if applicable) and The Author(s); 2019.

3. GBD 2017 Disease and Injury Incidence and Prevalence Collaborators. Global, regional, and national incidence, prevalence, and years lived with disability for 354 diseases and injuries for 195 countries and territories, 1990-2017: a systematic analysis for the global burden of disease study 2017. Lancet. 2018;392(10159):1789–1858. doi:10.1016/S0140-6736(18)32279-7

4. Mateos MV, Hernández MT, Giraldo P, et al. Lenalidomide plus dexamethasone versus observation in patients with high-risk smouldering multiple myeloma (QuiRedex): long-term follow-up of a randomised, controlled, Phase 3 trial. Lancet Oncol. 2016;17(8):1127–1136. doi:10.1016/S1470-2045(16)30124-3

5. Landgren O, Sonneveld P, Jakubowiak A, et al. Carfilzomib with immunomodulatory drugs for the treatment of newly diagnosed multiple myeloma. Leukemia. 2019;33(9):2127–2143. doi:10.1038/s41375-019-0517-6

6. Moreau P, Attal M, Hulin C, et al. Bortezomib, thalidomide, and dexamethasone with or without daratumumab before and after autologous stem-cell transplantation for newly diagnosed multiple myeloma (CASSIOPEIA): a randomised, open-label, phase 3 study. Lancet. 2019;394(10192):29–38. doi:10.1016/S0140-6736(19)31240-1

7. Dima D, Rashid A, Davis JA, et al. Efficacy and safety of idecabtagene vicleucel in patients with relapsed-refractory multiple myeloma not meeting the KarMMa-1 trial eligibility criteria: a real-world multicentre study. Br J Haematol. 2024;204(4):1293–1299. doi:10.1111/bjh.19302

8. Devasia AJ, Chari A, Lancman G. Bispecific antibodies in the treatment of multiple myeloma. Blood Cancer J. 2024;14(1):158. doi:10.1038/s41408-024-01139-y

9. Merz M. A comeback for checkpoint inhibition in multiple myeloma. Nat Cancer. 2024;5(10):1449–1451. doi:10.1038/s43018-024-00803-3

10. Durie BG. The role of anatomic and functional staging in myeloma: description of Durie/Salmon plus staging system. Eur J Cancer. 2006;42(11):1539–1543. doi:10.1016/j.ejca.2005.11.037

11. Greipp PR, San Miguel J, Durie BG, et al. International staging system for multiple myeloma. J Clin Oncol. 2005;23(15):3412–3420. doi:10.1200/JCO.2005.04.242

12. Palumbo A, Avet-Loiseau H, Oliva S, et al. Revised international staging system for multiple myeloma: a report from international myeloma working group. J Clin Oncol. 2015;33(26):2863–2869. doi:10.1200/JCO.2015.61.2267

13. Mikhael JR, Dingli D, Roy V, et al. Management of newly diagnosed symptomatic multiple myeloma: updated mayo stratification of myeloma and risk-adapted therapy (mSMART) consensus guidelines 2013. Mayo Clin Proc. 2013;88(4):360–376. doi:10.1016/j.mayocp.2013.01.019

14. Cho H, Yoon DH, Lee JB, et al. Comprehensive evaluation of the revised international staging system in multiple myeloma patients treated with novel agents as a primary therapy. Am J Hematol. 2017;92(12):1280–1286. doi:10.1002/ajh.24891

15. McMillan DC. The systemic inflammation-based glasgow prognostic score: a decade of experience in patients with cancer. Cancer Treat Rev. 2013;39(5):534–540. doi:10.1016/j.ctrv.2012.08.003

16. Heikkilä K, Ebrahim S, Lawlor DA. A systematic review of the association between circulating concentrations of C reactive protein and cancer. J Epidemiol Community Health. 2007;61(9):824–833. doi:10.1136/jech.2006.051292

17. Gupta D, Lis CG. Pretreatment serum albumin as a predictor of cancer survival: a systematic review of the epidemiological literature. Nutr J. 2010;9(1):69. doi:10.1186/1475-2891-9-69

18. Proctor MJ, Horgan PG, Talwar D, Fletcher CD, Morrison DS, McMillan DC. Optimization of the systemic inflammation-based glasgow prognostic score: a glasgow inflammation outcome study. Cancer. 2013;119(12):2325–2332. doi:10.1002/cncr.28018

19. Forrest LM, McMillan DC, McArdle CS, Angerson WJ, Dunlop DJ. Evaluation of cumulative prognostic scores based on the systemic inflammatory response in patients with inoperable non-small-cell lung cancer. Br J Cancer. 2003;89(6):1028–1030. doi:10.1038/sj.bjc.6601242

20. Ishizuka M, Nagata H, Takagi K, Horie T, Kubota K. Inflammation-based prognostic score is a novel predictor of postoperative outcome in patients with colorectal cancer. Ann Surg. 2007;246(6):1047–1051. doi:10.1097/SLA.0b013e3181454171

21. Ando K, Sakamoto S, Saito S, et al. Prognostic value of high-sensitivity modified glasgow prognostic score in castration-resistant prostate cancer patients who received docetaxel. Cancers. 2021;13(4):773. doi:10.3390/cancers13040773

22. Botta C, Di Martino MT, Ciliberto D, et al. A gene expression inflammatory signature specifically predicts multiple myeloma evolution and patients survival. Blood Cancer J. 2016;6(12):e511. doi:10.1038/bcj.2016.118

23. Nakamura K, Kassem S, Cleynen A, et al. Dysregulated IL-18 is a key driver of immunosuppression and a possible therapeutic target in the multiple myeloma microenvironment. Cancer Cell. 2018;33(4):634–648.e635. doi:10.1016/j.ccell.2018.02.007

24. Jiang J, Peng Z, Wang J, et al. C-reactive protein impairs immune response of CD8+ T cells via FcγRIIb-p38MAPK-ROS axis in multiple myeloma. J ImmunoTher Cancer. 2023;11(10):e007593. doi:10.1136/jitc-2023-007593

25. Prabhala RH, Pelluru D, Fulciniti M, et al. Elevated IL-17 produced by TH17 cells promotes myeloma cell growth and inhibits immune function in multiple myeloma. Blood. 2010;115(26):5385–5392. doi:10.1182/blood-2009-10-246660

26. Rossi M, Altomare E, Botta C, et al. miR-21 antagonism abrogates Th17 tumor promoting functions in multiple myeloma. Leukemia. 2021;35(3):823–834. doi:10.1038/s41375-020-0947-1

27. Ren L, Xu J, Li J, et al. A prognostic model incorporating inflammatory cells and cytokines for newly diagnosed multiple myeloma patients. Clin Exp Med. 2023;23(6):2583–2591. doi:10.1007/s10238-023-00992-8

28. Kim DS, Yu ES, Kang KW, et al. Myeloma prognostic index at diagnosis might be a prognostic marker in patients newly diagnosed with multiple myeloma. Korean J Internal Med. 2017;32(4):711–721. doi:10.3904/kjim.2016.054

29. Lee JH, Kim S-H, Lee H-S, et al. Prognostic implication of inflammatory factor-based staging system in multiple myeloma in the new agent era. Blood. 2019;134(Supplement_1):4372. doi:10.1182/blood-2019-127707

30. Kim Y, Kim SJ, Hwang D, et al. The modified glasgow prognostic scores as a predictor in diffuse large B cell lymphoma treated with R-CHOP regimen. Yonsei Med J. 2014;55(6):1568–1575. doi:10.3349/ymj.2014.55.6.1568

31. Huh SJ, Oh SY, Lee S, et al. The Glasgow Prognostic Score is a significant predictor of peripheral T-cell lymphoma (PTCL) treated with CHOP-based chemotherapy and comparable with PTCL prognostic scores. Int J Hematol. 2019;110(4):438–446. doi:10.1007/s12185-019-02693-z

32. Witte H, Biersack H, Kopelke S, et al. The Glasgow prognostic score at diagnosis is an independent predictor of survival in advanced stage classical Hodgkin lymphoma. Br J Haematol. 2019;184(5):869–873. doi:10.1111/bjh.15198

33. Richardson PG, Barlogie B, Berenson J, et al. Clinical factors predictive of outcome with bortezomib in patients with relapsed, refractory multiple myeloma. Blood. 2005;106(9):2977–2981. doi:10.1182/blood-2005-02-0691

34. Kaddoura M, Binder M, Dingli D, et al. Impact of achieving a complete response to initial therapy of multiple myeloma and predictors of subsequent outcome. Am J Hematol. 2022;97(3):267–273. doi:10.1002/ajh.26439

35. Nakamura T, Asanuma K, Hagi T, Sudo A. Modified glasgow prognostic score is better for predicting oncological outcome in patients with soft tissue sarcoma, compared to high-sensitivity modified glasgow prognostic score. J Inflamm Res. 2022;15:3891–3899. doi:10.2147/JIR.S369993

36. Zheng MM, Zhang Z, Bemis K, et al. The systemic cytokine environment is permanently altered in multiple myeloma. PLoS One. 2013;8(3):e58504. doi:10.1371/journal.pone.0058504

37. Saal J, Bald T, Eckstein M, et al. Integration of on-treatment modified Glasgow prognostic score (mGPS) to improve imaging-based prediction of outcomes in patients with non-small cell lung cancer on immune checkpoint inhibition. Lung Cancer. 2024;189:107505. doi:10.1016/j.lungcan.2024.107505

38. Patil D, Le TL, Bens KB, et al. Dynamic evaluation of the modified glasgow prognostic scale in patients with resected, localized clear cell renal cell carcinoma. Urology. 2020;141:101–107. doi:10.1016/j.urology.2020.03.024

39. Pang S, Zhou Z, Yu X, et al. The predictive value of integrated inflammation scores in the survival of patients with resected hepatocellular carcinoma: a retrospective cohort study. Int J Surg. 2017;42:170–177. doi:10.1016/j.ijsu.2017.04.018

40. Passamonti F, Cervantes F, Vannucchi AM, et al. A dynamic prognostic model to predict survival in primary myelofibrosis: a study by the IWG-MRT (international working group for myeloproliferative neoplasms research and treatment). Blood. 2010;115(9):1703–1708. doi:10.1182/blood-2009-09-245837

41. Neben K, Jauch A, Hielscher T, et al. Progression in smoldering myeloma is independently determined by the chromosomal abnormalities del(17p), t(4;14), gain 1q, hyperdiploidy, and tumor load. J Clin Oncol. 2013;31(34):4325–4332. doi:10.1200/JCO.2012.48.4923

42. Rajkumar SV, Gupta V, Fonseca R, et al. Impact of primary molecular cytogenetic abnormalities and risk of progression in smoldering multiple myeloma. Leukemia. 2013;27(8):1738–1744. doi:10.1038/leu.2013.86

43. Ross FM, Avet-Loiseau H, Ameye G, et al. Report from the European myeloma network on interphase FISH in multiple myeloma and related disorders. Haematologica. 2012;97(8):1272–1277. doi:10.3324/haematol.2011.056176

44. Rajkumar SV, Dimopoulos MA, Palumbo A, et al. International myeloma working group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 2014;15(12):e538–548. doi:10.1016/S1470-2045(14)70442-5