J Gastrointest Surg DOI 10.1007/s11605-016-3114-2
ORIGINAL ARTICLE
Clinical Significance of the Glasgow Prognostic Score for Survival after Colorectal Cancer Surgery Tunc Eren 1 & Busra Burcu 1 & Ercument Tombalak 1 & Tugrul Ozdemir 1 & Metin Leblebici 1 & Ibrahim Ali Ozemir 1 & Sedat Ziyade 2 & Orhan Alimoglu 1
Received: 21 December 2015 / Accepted: 16 February 2016 # 2016 The Society for Surgery of the Alimentary Tract
Abstract Glasgow prognostic score (GPS) has been found to be a useful tool in various cancer types. Our aim was to evaluate the significance of GPS in patients operated on for colorectal cancer (CRC). Patients with CRC who underwent radical resections between April 2010 and January 2015 were retrospectively evaluated. GPS was estimated based on the preoperative measurement of C-reactive protein and serum albumin levels. Data including demographics, laboratory and pathological parameters, surgical outcomes, and late-term follow-up results were analyzed. The study group of 115 patients consisted of 51 (44 %) women and 64 (56 %) men with a median age of 66 (range 32–91) years. The mean follow-up period was 20 (range 7–41) months. Tumor size and wound infection rates were significantly increased in patients with higher GPS (p = 0.019 and p = 0.003, respectively). According to multivariate analyses, CEA and GPS were found to be independent risk factors significantly effecting mortality (p = 0.001 and p = 0.009, respectively). At the end of the late-term follow-up period, it was detected that cancer-specific survival significantly decreased as the GPS increased (p = 0.016). The GPS is a significant prognostic factor in CRC and should be included in the routine preoperative assessment of all surgically treated CRC patients. Keywords Colorectal neoplasms . General surgery . Survival
Introduction Colorectal cancer (CRC) is the third most common cancer worldwide and accounts for 10 % of all new cancer diagnoses.1 Consequent to the advances in surgical techniques and management, survival rate has increased substantially over the last 25 years.2 However, postoperative morbidity and mortality from CRC vary widely across the world.3
* Tunc Eren [email protected]
1
Department of General Surgery, School of Medicine, Istanbul Medeniyet University Goztepe Training & Research Hospital, Dr. Erkin Street, Kadikoy, 34730 Istanbul, Turkey
2
Department of Thoracic Surgery (Medical Statistics), School of Medicine, Bezmialem Vakif University, Istanbul, Turkey
Currently, the cornerstone of cancer prognosis is accurate staging of the tumor. However, there is increasing recognition that patient-related factors, in particular nutritional and functional decline, are associated with poorer outcome independent of tumor stage. The most commonly used measures of this decline, weight loss and performance status, are recognized to be subjective, and therefore, their reliability in predicting outcome is compromised. Moreover, they do not provide objective therapeutic targets.4 There is consistent evidence that the presence of a systemic inflammatory response is a major factor for poor prognosis.4 Malignant disease and inflammation have close relationship with each other as cancer can induce local or systemic inflammation, mediated by the activation of transcription factors and the production of major inflammatory cytokines which can influence cell proliferation, cell survival, angiogenesis, tumor cell migration, invasion, metastasis, and inhibition of adaptive immunity.5 Colorectal cancer (CRC) also has a close relationship with inflammation given that inflammatory bowel diseases are known as premalignant conditions for CRC. 6 Cyclooxygenase-2 inhibitors and nonsteroidal antiinflammatory drugs were found to decrease the incidence of
J Gastrointest Surg
colorectal adenoma and CRC.7 Glasgow prognostic score (GPS), an inflammation-based prognostic score, which is assessed by simply using the serum C-reactive protein (CRP) and albumin levels, has been found to be a useful prognostic tool in various cancer types.8 It was noted that as circulating CRP concentrations increased, albumin concentrations fell, and this relationship was similar across different tumor types.4 Also, given that albumin concentrations reflected both systemic inflammation and the amount of lean tissue, it was of interest to examine the prognostic value of the combination of an elevated CRP concentration and hypoalbuminaemia.9 In the first report by Forrest et al. in Glasgow, this objective combination compared favorably with the clinical standard combination of stage and performance status.10 The GPS which is simple to measure, routinely available, and well standardized worldwide has subsequently been the subject of prognostic studies in wide variety of operable and inoperable cancers.4 Our aim was to evaluate the prognostic significance of preoperatively estimated GPS in patients operated on with the diagnosis of CRC.
Methods Patients with biopsy-proven colorectal adenocarcinoma who underwent radical resections in the Department of General Surgery between April 2010 and January 2015 were retrospectively investigated. Approval for this study was gained from the Ethics Committee of the institution, and signed informed consent forms were obtained from all patients who were included in the study. Patients with a past medical history of any ongoing inflammatory conditions (inflammatory bowel disease, arthritis, etc.) were not included to the study group. In the early postoperative period (postoperative first month), deaths due to either surgical complications or due to non-cancer-related reasons were excluded. On the other hand, in the late-term follow-up period, patients who did not regularly attend to scheduled office visits and patients who could not be reached by various communication methods were excluded as well. Finally, 115 patients, whose preoperatively measured serum CRP and albumin levels were available, and whose early postoperative care, and late-term follow-ups could be carried out completely and reliably were enrolled to the study. Data including the following clinicopathological factors were selected and evaluated: age, gender, preoperative laboratory test values (hemogram and biochemical test parameters), neutrophil-to-lymphocyte ratio (NLR; NLR was defined as the absolute neutrophil count divided by the absolute lymphocyte count), tumor markers (CEA, CA19-9, and AFP), location of the tumor, tumor size, lymphovascular/perineural invasion, tumor differentiation, distant metastasis, TNM stage
(AJCC; American Joint Committee on Cancer 7th ed.), and early postoperative surgical outcomes (postoperative first month). After discharge from the hospital, the patients were followed up on scheduled routine office visits, and the data of the late-term follow-up period including survival outcomes was also collected and evaluated. The GPS was preoperatively estimated as described previously by Forrest et al.10 Patients with both an elevated level of CRP (>1.0 mg/dL) and hypoalbuminemia ( 0.05). The comparisons of the demographic and laboratory parameters between the three GPS groups are summarized in Table 2. Tumor size was significantly increased with increasing GPS (p = 0.019). Other clinicopathological parameters including tumor differentiation, lymphovascular invasion, perineural invasion, presence of distant metastasis, and TNM stages did
The comparison of the demographic and laboratory parameters between the three GPS groups
Parameters
Age (years) Gender Hematocrit (%) Hemoglobin (g/dL) Hemoglobin < 12 g/dL White blood cell (103/mm3) Platelet (103/mm3) Platelet ≥ 400000/mm3 Neutrophil (103/mm3) Lymphocyte (103/mm3) Neutrophil-to-lymphocyte ratio (NLR) NLR ≥ 3 Total protein (g/dL) Albumin (g/dL) C-reactive protein (mg/dL)
GPS 0 (n = 30)
GPS 1 (n = 55)
GPS 2 (n = 30)
62.2 ± 13.2 16 (53.3 %) 14 (46.7 %) 36.5 ± 5.2 12.0 ± 1.9
66.2 ± 12.2 31 (56.4 %) 24 (43.6 %) 35.6 ± 5.7 11.5 ± 2.1
69.8 ± 12.2 17 (56.7 %) 13 (43.3 %) 34.2 ± 4.7 11.1 ± 1.7
0.067 0.970b
n (%) Mean ± SD Mean ± SD n (%) Mean ± SD Median (Min-Max) Mean ± SD Median (Min-Max)
14 (46.7 %) 7.9 ± 2.7 277.4 ± 91.7 3 (10.0 %) 6.3 ± 2.9 4.9 (1.6-22.7) 1.8 ± 1.0 1.4 (0.4-5.4)
24 (43.6 %) 7.7 ± 2.1 293.0 ± 86.8 7 (12.7 %) 6.3 ± 2.9 5.4 (2.4-15.2) 2.2 ± 2.7 1.6 (0.7-2.0)
13 (43.3 %) 8.6 ± 2.7 278.6 ± 82.3 4 (13.3 %) 6.2 ± 3.4 5.4 (1.2-17.5) 1.6 ± 0.8 1.6 (0.6-3.9)
0.287 0.312 0.582c 1.000d 0.832c
Mean ± SD Median (Min-Max) n (%) Mean ± SD Mean ± SD Mean ± SD Median (Min-Max)
4.9 ± 5.2 2.7 (1.5-21.3) 12 (40.0 %) 6.7 ± 0.7 4.0 ± 0.4 0.6 ± 0.3 0.7 (0.0-1.0) n = 21
4.5 ± 3.7 3.1 (1.3-19.7) 31 (56.4 %) 6.5 ± 0.9 3.6 ± 0.8 1.9 ± 1.8 1.4 (0.0-9.9) n = 37
4.8 ± 5.5 3.6 (3.0-29.1) 18 (60.0 %) 5.5 ± 0.8 2.8 ± 0.4 4.4 ± 4.1 2.5 (1.4-19.0) n = 14
0.875c
Mean ± SD Median (Min-Max) n (%)
9.2 ± 20.7 1.9 (0.3-84.8) 6 (28.6 %) n = 12 31.4 ± 44.0 11.1 (0.8-147.2) 8 (66.7 %) n = 13 3.2 ± 1.6 3.1 (0.4-5.7) 13 (100.0 %)
13.9 ± 40.0 2.1 (0.5-229.3) 8 (21.6 %) n = 25 40.4 ± 66.8 15.2 (0.8-258.9) 20 (80.0 %) n = 19 2.3 ± 1.5 1.7 (1.2-6.7) 19 (100.0 %)
33.7 ± 94.3 2.2 (0.7-356.9) 4 (28.6 %) n = 12 20.4 ± 29.4 10.8 (2.0-109.4) 11 (91.7 %) n=9 4.2 ± 3.0 3.3 (1.0-10.5) 8 (88.9 %)
Mean ± SD Male Female Mean ± SD Mean ± SD
n (%) n (%)
Carcinoembryonic antigen (CEA) CEA (ng/mL) CEA ≥ 5 ng/mL Cancer antigen 19–9 (CA 19–9) CA 19–9 (U/mL) CA 19–9 ≥ 5 U/mL Alpha-fetoprotein (AFP) AFP (ng/mL) AFP < 9 ng/mL
Mean ± SD Median (Min-Max) n (%) Mean ± SD Median (Min-Max) n (%)
n number of patients, Mean ± SD mean ± standard deviation, Median (Min-Max) median (minimum-maximum) a
One-way ANOVA
b
Pearson’s chi-squared test
c
Kruskal-Wallis test
d
Fisher-Freeman-Halton test
*p < 0.05; **p < 0.01
pa
Glasgow prognostic score (GPS)
0.237 0.213
0.522c
0.237b 0.05). The comparisons of the surgical and clinicopathological parameters between the three GPS groups are shown in Table 3. According to univariate analyses, CEA levels exceeding 5 ng/mL, and GPS 2 were detected to be significantly related with mortality (p = 0.001 and p = 0.008, respectively). However, other variables including age, gender, hemoglobin levels, platelet levels, NLR, tumor location, tumor size, and differentiation were not found to assert significance (p > 0.05). Multivariate analyses were carried out for significant or nearly significant variables detected in the univariate analyses. According to multivariate analyses, CEA and GPS were found to be significantly effecting mortality, as well, revealing that these parameters were independent risk factors (p = 0.001 and p = 0.009, respectively). The univariate and multivariate analyses for cancer-related mortality are summarized in Table 4, whereas the independent risk factors detected in the present study are plotted in Fig. 1. At the end of the late-term follow-up period, the relation between survival and the GPS was evaluated with use of the Log rank test, and it was detected that cancer-related mortality rates increased, while cancer-specific survival significantly decreased as the GPS increased (p = 0.016). The distribution Table 3
of the survival rates and periods among the three GPS groups is summarized in Table 5. The Kaplan-Meier plot for cancerspecific survival of the three GPS groups is shown in Fig. 2.
Discussion Although it is recognized that the development of cancer has a genetic basis, there is increasing evidence that the host inflammatory response plays an important role in the development and progression of cancer.11 Inflammation can induce the initial genetic mutation and epigenetic changes for cancer initiation, can modify the tissue microenvironment that permits cancer cells to progress and metastasize, and can also suppress the immune response to tumor cells.12 Links between cancer and inflammation were first established in the nineteenth century and were based on observations that tumors often arose at sites of chronic inflammation and that inflammatory cells are present in biopsy samples collected from tumor tissues. Today, it is well known that cancer promotes release of proinflammatory cytokines from tumor cells.13 Experimental studies showed that oncogenes could activate several inflammatory factors including colony stimulating factors (CSFs), interleukin 1β, chemokines/ chemokine receptors (CCL2, CCL20, IL-8, CXCL12, CXCR4), proteases such as matrix metalloproteinases (MMP7, MMP9, MMP10), and urokinase-type plasminogen activator and its receptor.5 These cytokines interact with the
The comparison of the surgical and clinicopathological parameters between the three GPS groups
Parameters
Tumor size (cm.) Tumor size ≥ 5 cm. Tumor differentiation
Lymphovascular invasion Perineural invasion Distant metastasis TNM Stage (AJCC) Anastomotic leakage Wound infection
Glasgow prognostic score (GPS)
Mean ± SD Median (Min-Max) n (%) Well differentiated Moderately differentiated Poorly differentiated n (%) n (%) n (%) Stage I + II Stage III + IV n (%) n (%)
n (%) n (%) n (%)
n (%) n (%)
GPS 0 (n = 30)
GPS 1 (n = 55)
GPS 2 (n = 30)
4.3 ± 1.5 4.0 (2.0-8.0) 11 (36.7 %) 2 (6.7 %) 27 (90.0 %) 1 (3.3 %) 16 (53.3 %) 9 (30.0 %) 1 (3.3 %) 15 (50.0 %) 15 (50.0 %) 2 (6.7 %) 1 (3.3 %)
5.3 ± 2.1 5.0 (2.0-12.0) 35 (63.6 %) 8 (14.5 %) 39 (70.9 %) 8 (14.5 %) 36 (65.6 %) 24 (43.6 %) 3 (5.5 %) 28 (50.9 %) 27 (49.1 %) 2 (3.6 %) 1 (1.8 %)
5.4 ± 2.1 5.3 (1.0-12.0) 21 (70.0 %) 3 (10.0 %) 25 (83.3 %) 2 (6.7 %) 16 (53.3 %) 10 (33.3 %) 2 (6.7 %) 15 (50.0 %) 15 (50.0 %) 1 (3.3 %) 7 (23.3 %)
n number of patients, Mean ± SD mean ± standard deviation, Median (Min-Max) median (minimum-maximum) a
Kruskal-Wallis test
b
Pearson’s chi-squared test
c
Fisher-Freeman-Halton test
*p < 0.05; **p < 0.01
p
0.019a * 0.017b * 1.000c
0.418b 0.401b 1.000c 0.995b 0.846c 0.003c **
J Gastrointest Surg Table 4
The univariate and multivariate analyses for cancer-related mortality
Variable
Mortality No (n = 96) n (%)
Univariate Yes (n = 19) n (%)
HR
Multivariate 95 % CI
p
HR
95 % CI
p
2.132
0.66–6.91
0.207
6.932
2.27–21.16
0.001**
Age ≥ 70 years
39 (40.6 %)
12 (63.2 %)
2.257
0.89–5.74
0.088
Female gender
43 (44.8 %)
8 (42.1 %)
0.913
0.37–2.27
0.844
Hb < 12 g/dL Plt ≥ 400000/mm3 NLR ≥ 3 CEA ≥ 5 ng/mL
56 (58.3 %) 12 (12.5 %)
13 (68.4 %) 2 (10.5 %)
1.685 1.103
0.64–4.43 0.25–4.79
0.291 0.896
49 (51.0 %) 9 (15.5 %)
12 (63.2 %) 9 (64.3 %)
1.648 7.389
0.65–4.19 2.44–22.33
0.294 0.001**
Tumor location, rectum Tumor size ≥ 5 cm
34 (35.4 %) 56 (58.3 %)
5 (26.3 %) 11 (57.9 %)
0.626 1.274
0.22–1.74 0.51–3.18
0.370 0.604
Poorly differentiated tumor
7 (7.3 %)
4 (21.1 %)
2.550
0.84–7.71
0.097
2.002
0.61–6.59
0.253
GPS, 2
21 (21.9 %)
9 (47.4 %)
3.389
1.37–8.41
0.008**
4.102
1.42–11.86
0.009**
HR hazard ratio, CI confidence interval, n number of patients, Hb hemoglobin, NLR neutrophil-to-lymphocyte ratio, CEA carcinoembryonic antigen, GPS Glasgow prognostic score *p < 0.05; **p < 0.01
immunovascular system and facilitate cancer growth, tumor cell migration, proliferation, survival, angiogenesis, invasion, , dissemination, and metastasis.5 13 Cancer-producing proinflammatory cytokines induce the acute phase protein, which is a key marker of systemic inflammation.14 Since the initial studies of using CRP, as a sensitive measure of the systemic inflammatory response, stated independent prognostic value in operable CRC, there have been numerous studies reporting the prognostic value of CRP and other markers of the systemic inflammatory response in a , wide variety of operable and inoperable cancers. 8 15 Systemic inflammation has been found to be positively associated with weight loss, hypermetabolism, anorexia, and poor prognosis in cancer patients.16 In CRC, inflammation has an important role in the initiation and progression.6 In colitis-associated colon cancer, chronic inflammation causes oxidative damage to the DNA, Fig. 1 The independent risk factors effecting cancer-related mortality. [CEA: Carcinoembryonic antigen; GPS: Glasgow Prognostic Score]
leading to p53 mutations in tumor cells, and the inflamed epithelium and the inflammatory microenvironment at the tumor border can influence several key stages of invasion and metastasis.17 McMillan et al. reported that the prognostic score based on the serum CRP and albumin levels had an independent prognostic value after resection of CRC.18 Both CRP and albumin are acute phase proteins. Acute phase proteins are produced in the liver in response to inflammatory cytokines, mainly IL-6 and IL-1β. CRP level can be elevated to as much as 1000-fold after an inflammatory stimulus.19 But, the albumin level is decreased in response to inflammation. The albumin level is known to be decreased in cancer patients due to malnutrition and systemic inflammation.9 Recent studies have shown that elevated serum CRP levels may be associated with tumor size, distant metastasis, vascular invasion, lymph node metastasis, and tumor recurrence, resulting in poor prognosis.13 It has also been reported that
J Gastrointest Surg Table 5 The distribution of the survival rates and periods among the three GPS goups
GPS
n
Mortality
Survival rate
Yes
No
Cumulative survival Mean
SD
Mean survival period (months)
GPS 0
30
3
27
90.0 %
86.6
5.7
37.97 ± 1.66
GPS 1
55
7
48
87.3 %
83.6
6.0
29.85 ± 1.10
GPS 2
30
9
21
70.0 %
30.0
15.4
21.88 ± 0.89
GPS Glasgow prognostic score, n number of patients, SD standard deviation
serum albumin participate in systemic inflammatory response and that decline of its serum level is a poor prognostic factor for long-term survival in patients with various cancers.20 Based on these reports, GPS, incorporating CRP and serum albumin levels, may reflect both presence of the systemic inflammatory response (CRP), and the progressive nutritional decline (albumin) in cancer patients, resulting in poor survival outcome.13 Forrest et al. designed the GPS, an inflammationbased prognostic score, which was assessed by simply using the serum CRP and serum albumin levels.10 The GPS has been proved to be an independent prognostic factor in many studies which have been performed in unselected cohorts, operable, and inoperable cancer patients.4 In our study, we retrospectively evaluated the GPS of surgically treated CRC patients. Mauricio et al. reported that malnutrition is highly prevalent in patients with CRC, and that the nutritional status was associated with GPS.21 CRP level was reported to be higher in patients with colon cancer and larger tumors. 22 Anemia and thrombocytosis are considered to be the common manifestations Fig. 2 The Kaplan-Meier plot for cancer-specific survival among the three GPS groups
induced by inflammatory cytokines, whereas NLR is another well-known indicator of systemic inflammatory response and a , prognostic factor in CRC patients.19 23 There are also studies reporting that the GPS was positively correlated with high levels of CEA.24 In the present study, mean albumin levels were significantly decreased, as the mean CRP levels were significantly increased with increasing GPS, and these results are due to these parameters being the standard variables of the scoring system. The decrease in the mean total protein levels along with the albumin levels is also considered to be an indicator of worsened malnutrition related with higher GPS in CRC patients. However, other parameters including age, gender, hemogram values, NLR, and tumor markers did not show significant difference when compared among the GPS groups (Table 2). Choi et al. reported that in comparison with rectal cancer, colon cancer and larger tumors had a close relationship with a high GPS.14 McMillan et al. showed that the GPS was not related with Dukes’ stage of the tumor.18 But, other authors reported that the GPS was positively correlated with advanced
J Gastrointest Surg
TNM stage.24 In our study, the GPS was significantly higher in patients with larger tumors (Table 3). However, no significant difference could be detected when tumor location was analyzed (Table 4). Other clinicopathological parameters including tumor differentiation, lymphovascular invasion, perineural invasion, presence of distant metastasis, and TNM stages did not reveal significant difference among the GPS groups (Table 3). There are numerous studies revealing that increased GPS was associated with increased weight loss, poor performance status, increased comorbidity, increased proinflammatory and , , , angiogenic cytokines, and complications on treatment.4 14 21 25 When early postoperative surgical outcomes of our study group were analyzed, it was found that wound infection rates were significantly higher in the GPS 2 group. However, no significant alteration was detected among the three groups in concern of anastomotic leakage rates (Table 3). Many studies confirmed that the GPS was a good indicator of prognosis in several kinds of cancers including lung cancer,10 hepatocellular carcinoma,13 gastric cancer,26 and , , CRC.14 18 21 Additionally, it was reported that preoperative CEA has independent prognostic values in CRC, and that preoperatively elevated CEA is correlated with worsened , outcomes.27 28 In the present study, according to multivariate analyses, preoperative CEA levels exceeding 5 ng/mL and GPS 2 were found to be significantly effecting mortality, revealing that these parameters were independent risk factors (Table 4, Fig. 1). At the end of the late-term follow-up period of our patients, it was detected that cancer-specific survival significantly decreased as the GPS increased (Table 5, Fig. 2). The systemic inflammatory response is a major factor in the relationship between nutritional decline and poor outcome in patients with cancer.4 GPS may be included, in addition or in preference to the current definitions of cachexia, with tumor staging as part of the routine assessment of cancer patients. MacDonald recently concluded that particularly in the more aggressive tumor types (e.g., pancreas and lung), the future of patients with elevated GPS is so grim that they should be given precachexia status and offered multimodal therapy which may delay the onset of cachexia and/or death.29 As a consequence, this will highlight the need not only to treat the tumor but also the systemic inflammatory response, a potentially less intractable target compared with established weight loss and/or poor performance status. Further work is required to define the value of GPS as a stratification factor, as a selection criteria in randomized clinical trials and as a therapeutic target in patients with cancer. Also, how the prognostic value of the GPS compares with that of other measures of the systemic inflammatory response, such as the NLR.4 We believe that the present results provide good evidence of systemic inflammation acting as a factor promoting the fatal progression in most CRC patients, and the moderation of the
systemic inflammatory response will become, in the future, as important a therapeutic target as the tumor itself. Our results suggest that anti-inflammatory agents with other therapeutic modalities may prolong the survival of CRC patients.30
Conclusion In conclusion, our study showed that the GPS was an independent variable from tumor stage and a significant prognostic factor in CRC patients. The GPS should be included in the routine preoperative assessment of all surgically treated CRC patients. Acknowledgments The authors would like to thank Mrs. Emire BORα for the statistical work-up of this study. α Biostatistics Specialist, EMPIAR Statistics, Istanbul, Turkey Compliance with Ethical Standards Approval for this study was gained from the Ethics Committee of Istanbul Medeniyet University, School of Medicine, Goztepe Training and Research Hospital, and signed informed consent forms were obtained from all patients. Each author hereby acknowledges that the final state of this manuscript is prepared and sent with his/her approval having been taken. Additionally, all authors declare that no source of funding exists, and that they have no direct or indirect commercial financial incentive associated with publishing this article. The authors also confirm that this manuscript has not been published and is not under simultaneous consideration by another journal or electronic publication. All authors have read and complied with the requirements set forth in the Instructions to Authors. Conflict of Interest The corresponding author and all co-authors declare that they have no conflict of interest.
References 1.
Global Burden of Disease Cancer Collaboration, Fitzmaurice C, Dicker D, Pain A, Hamavid H, Moradi-Lakeh M, MacIntyre MF, Allen C, Hansen G, Woodbrook R, Wolfe C, Hamadeh RR, Moore A, Werdecker A, Gessner BD, Te Ao B, McMahon B, Karimkhani C, Yu C, Cooke GS, Schwebel DC, Carpenter DO, Pereira DM, Nash D, Kazi DS, De Leo D, Plass D, Ukwaja KN, Thurston GD, Yun Jin K, Simard EP, Mills E, Park EK, Catalá-López F, de Veber G, Gotay C, Khan G, Hosgood HD 3rd, Santos IS, Leasher JL, Singh J, Leigh J, Jonas J, Sanabria J, Beardsley J, Jacobsen KH, Takahashi K, Franklin RC, Ronfani L, Montico M, Naldi L, Tonelli M, Geleijnse J, Petzold M, Shrime MG, Younis M, Yonemoto N, Breitborde N, Yip P, Pourmalek F, Lotufo PA, Esteghamati A, Hankey GJ, Ali R, Lunevicius R, Malekzadeh R, Dellavalle R, Weintraub R, Lucas R, Hay R, Rojas-Rueda D, Westerman R, Sepanlou SG, Nolte S, Patten S, Weichenthal S, Abera SF, Fereshtehnejad SM, Shiue I, Driscoll T, Vasankari T, Alsharif U, Rahimi-Movaghar V, Vlassov VV, Marcenes WS, Mekonnen W, Melaku YA, Yano Y, Artaman A, Campos I, MacLachlan J, Mueller U, Kim D, Trillini M, Eshrati B, Williams HC, Shibuya K, Dandona R, Murthy K, Cowie B, Amare AT, Antonio CA, Castañeda-Orjuela C, van Gool CH, Violante F, Oh IH, Deribe K, Soreide K, Knibbs L, Kereselidze M, Green M, Cardenas R, Roy N, Tillman T, Li Y,
J Gastrointest Surg
2.
3.
4.
5. 6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
Krueger H, Monasta L, Dey S, Sheikhbahaei S, Hafezi-Nejad N, Kumar GA, Sreeramareddy CT, Dandona L, Wang H, Vollset SE, Mokdad A, Salomon JA, Lozano R, Vos T, Forouzanfar M, Lopez A, Murray C, Naghavi M. The Global Burden of Cancer 2013. JAMA Oncol. 2015;1(4):505–27. Coleman MP, Rachet B, Woods LM, Mitry E, Riga M, Cooper N, Quinn MJ, Brenner H, Estève J. Trends and socioeconomic inequalities in cancer survival in England and Wales up to 2001. Br J Cancer. 2004;90(7):1367–73. McArdle CS, Hole DJ. Outcome following surgery for colorectal cancer: analysis by hospital after adjustment for case-mix and deprivation. Br J Cancer. 2002;86(3):331–5. McMillan DC. The systemic inflammation-based Glasgow Prognostic Score: a decade of experience in patients with cancer. Cancer Treat Rev. 2013;39(5):534–40. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflammation. Nature. 2008;454(7203):436–44. Triantafillidis JK, Nasioulas G, Kosmidis PA. Colorectal cancer and inflammatory bowel disease: epidemiology, risk factors, mechanisms of carcinogenesis and prevention strategies. Anticancer Res. 2009;29(7):2727–37. Rostom A, Dubé C, Lewin G, Tsertsvadze A, Barrowman N, Code C, Sampson M, Moher D; U.S. Preventive Services Task Force. Nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 inhibitors for primary prevention of colorectal cancer: a systematic review prepared for the U.S. Preventive Services Task Force. Ann Intern Med. 2007;146(5):376–89. Roxburgh CS, McMillan DC. Role of systemic inflammatory response in predicting survival in patients with primary operable cancer. Future Oncol. 2010;6(1):149–63. McMillan DC, Watson WS, O’Gorman P, Preston T, Scott HR, McArdle CS. Albumin concentrations are primarily determined by the body cell mass and the systemic inflammatory response in cancer patients with weight loss. Nutr Cancer. 2001;39(2):210–3. 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-smallcell lung cancer. Br J Cancer. 2003;89(6):1028–30. Colotta F, Allavena P, Sica A, Garlanda C, Mantovani A. Cancerrelated inflammation, the seventh hallmark of cancer: links to genetic instability. Carcinogenesis. 2009;30(7):1073–81. Trinchieri G. Cancer and inflammation: an old intuition with rapidly evolving new concepts. Annu Rev Immunol. 2012;30:677–706. Kinoshita A, Onoda H, Imai N, Iwaku A, Oishi M, Tanaka K, Fushiya N, Koike K, Nishino H, Matsushima M, Saeki C, Tajiri H. The Glasgow Prognostic Score, an inflammation based prognostic score, predicts survival in patients with hepatocellular carcinoma. BMC Cancer. 2013;13:52. Choi KW, Hong SW, Chang YG, Lee WY, Lee B, Paik IW, Lee H. Inflammation-based score (Glasgow prognostic score) as an independent prognostic factor in colorectal cancer patients. Ann Surg Treat Res. 2014;86(6):309–13. McMillan DC, Canna K, McArdle CS. Systemic inflammatory response predicts survival following curative resection of colorectal cancer. Br J Surg. 2003;90(2):215–9.
16.
17. 18.
19. 20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30.
Deans C, Wigmore SJ. Systemic inflammation, cachexia and prognosis in patients with cancer. Curr Opin Clin Nutr Metab Care. 2005;8(3):265–9. Terzić J, Grivennikov S, Karin E, Karin M. Inflammation and colon cancer. Gastroenterology. 2010;138(6):2101–2114.e5. McMillan DC, Crozier JE, Canna K, Angerson WJ, McArdle CS. Evaluation of an inflammation-based prognostic score (GPS) in patients undergoing resection for colon and rectal cancer. Int J Colorectal Dis. 2007;22(8):881–6. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med. 1999;340(6):448–54. Kobayashi T, Teruya M, Kishiki T, Endo D, Takenaka Y, Tanaka H, Miki K, Kobayashi K, Morita K. Inflammation-based prognostic score, prior to neoadjuvant chemoradiotherapy, predicts postoperative outcome in patients with esophageal squamous cell carcinoma. Surgery. 2008;144(5):729–35. Maurício SF, da Silva JB, Bering T, Correia MI. Relationship between nutritional status and the Glasgow Prognostic Score in patients with colorectal cancer. Nutrition. 2013;29(4):625–9. Shiu YC, Lin JK, Huang CJ, Jiang JK, Wang LW, Huang HC, Yang SH. Is C-reactive protein a prognostic factor of colorectal cancer? Dis Colon Rectum. 2008;51(4):443–9. Walsh SR, Cook EJ, Goulder F, Justin TA, Keeling NJ. Neutrophillymphocyte ratio as a prognostic factor in colorectal cancer. J Surg Oncol. 2005;91(3):181–4. Ishizuka M, Nagata H, Takagi K, Horie T, Kubota K. Inflammationbased prognostic score is a novel predictor of postoperative outcome in patients with colorectal cancer. Ann Surg. 2007;246(6): 1047–51. Proctor MJ, Morrison DS, Talwar D, Balmer SM, Fletcher CD, O’Reilly DS, Foulis AK, Horgan PG, McMillan DC. A comparison of inflammation-based prognostic scores in patients with cancer. A Glasgow Inflammation Outcome Study. Eur J Cancer. 2011;47(17):2633–41. Mohri Y, Tanaka K, Ohi M, Toiyama Y, Yasuda H, Inoue Y, Uchida K, Kusunoki M. Inflammation-based prognostic score as a predictor of postoperative gastric cancer recurrence. Anticancer Res. 2012;32(10):4581–4. Wang RF, Song BR, Peng JJ, Cai GX, Liu FQ, Wang MH, Cai SJ, Ye X. The Prognostic Value of Preoperative Serum CEA and CA19-9 Values in Stage I-III Colorectal Cancer. Hepatogastroenterology. 2014;61(132):994–9. Chen L, Jiang B, Di J, Zhang C, Wang Z, Zhang N, Xing J, Cui M, Yang H, Yao Z, Su X. Predictive value of preoperative detection of CEA and CA199 for prognosis in patients with stage II-III colorectal cancer. Zhonghua Wei Chang Wai Ke Za Zhi. 2015;18(9):914–9. MacDonald N. Terminology in cancer cachexia: importance and status. Curr Opin Clin Nutr Metab Care. 2012;15(3):220–5. Bastiaannet E, Sampieri K, Dekkers OM, de Craen AJ, van HerkSukel MP, Lemmens V, van den Broek CB, Coebergh JW, Herings RM, van de Velde CJ, Fodde R, Liefers GJ. Use of aspirin postdiagnosis improves survival for colon cancer patients. Br J Cancer. 2012;106(9):1564–70.
ORIGINAL ARTICLE
Clinical Significance of the Glasgow Prognostic Score for Survival after Colorectal Cancer Surgery Tunc Eren 1 & Busra Burcu 1 & Ercument Tombalak 1 & Tugrul Ozdemir 1 & Metin Leblebici 1 & Ibrahim Ali Ozemir 1 & Sedat Ziyade 2 & Orhan Alimoglu 1
Received: 21 December 2015 / Accepted: 16 February 2016 # 2016 The Society for Surgery of the Alimentary Tract
Abstract Glasgow prognostic score (GPS) has been found to be a useful tool in various cancer types. Our aim was to evaluate the significance of GPS in patients operated on for colorectal cancer (CRC). Patients with CRC who underwent radical resections between April 2010 and January 2015 were retrospectively evaluated. GPS was estimated based on the preoperative measurement of C-reactive protein and serum albumin levels. Data including demographics, laboratory and pathological parameters, surgical outcomes, and late-term follow-up results were analyzed. The study group of 115 patients consisted of 51 (44 %) women and 64 (56 %) men with a median age of 66 (range 32–91) years. The mean follow-up period was 20 (range 7–41) months. Tumor size and wound infection rates were significantly increased in patients with higher GPS (p = 0.019 and p = 0.003, respectively). According to multivariate analyses, CEA and GPS were found to be independent risk factors significantly effecting mortality (p = 0.001 and p = 0.009, respectively). At the end of the late-term follow-up period, it was detected that cancer-specific survival significantly decreased as the GPS increased (p = 0.016). The GPS is a significant prognostic factor in CRC and should be included in the routine preoperative assessment of all surgically treated CRC patients. Keywords Colorectal neoplasms . General surgery . Survival
Introduction Colorectal cancer (CRC) is the third most common cancer worldwide and accounts for 10 % of all new cancer diagnoses.1 Consequent to the advances in surgical techniques and management, survival rate has increased substantially over the last 25 years.2 However, postoperative morbidity and mortality from CRC vary widely across the world.3
* Tunc Eren [email protected]
1
Department of General Surgery, School of Medicine, Istanbul Medeniyet University Goztepe Training & Research Hospital, Dr. Erkin Street, Kadikoy, 34730 Istanbul, Turkey
2
Department of Thoracic Surgery (Medical Statistics), School of Medicine, Bezmialem Vakif University, Istanbul, Turkey
Currently, the cornerstone of cancer prognosis is accurate staging of the tumor. However, there is increasing recognition that patient-related factors, in particular nutritional and functional decline, are associated with poorer outcome independent of tumor stage. The most commonly used measures of this decline, weight loss and performance status, are recognized to be subjective, and therefore, their reliability in predicting outcome is compromised. Moreover, they do not provide objective therapeutic targets.4 There is consistent evidence that the presence of a systemic inflammatory response is a major factor for poor prognosis.4 Malignant disease and inflammation have close relationship with each other as cancer can induce local or systemic inflammation, mediated by the activation of transcription factors and the production of major inflammatory cytokines which can influence cell proliferation, cell survival, angiogenesis, tumor cell migration, invasion, metastasis, and inhibition of adaptive immunity.5 Colorectal cancer (CRC) also has a close relationship with inflammation given that inflammatory bowel diseases are known as premalignant conditions for CRC. 6 Cyclooxygenase-2 inhibitors and nonsteroidal antiinflammatory drugs were found to decrease the incidence of
J Gastrointest Surg
colorectal adenoma and CRC.7 Glasgow prognostic score (GPS), an inflammation-based prognostic score, which is assessed by simply using the serum C-reactive protein (CRP) and albumin levels, has been found to be a useful prognostic tool in various cancer types.8 It was noted that as circulating CRP concentrations increased, albumin concentrations fell, and this relationship was similar across different tumor types.4 Also, given that albumin concentrations reflected both systemic inflammation and the amount of lean tissue, it was of interest to examine the prognostic value of the combination of an elevated CRP concentration and hypoalbuminaemia.9 In the first report by Forrest et al. in Glasgow, this objective combination compared favorably with the clinical standard combination of stage and performance status.10 The GPS which is simple to measure, routinely available, and well standardized worldwide has subsequently been the subject of prognostic studies in wide variety of operable and inoperable cancers.4 Our aim was to evaluate the prognostic significance of preoperatively estimated GPS in patients operated on with the diagnosis of CRC.
Methods Patients with biopsy-proven colorectal adenocarcinoma who underwent radical resections in the Department of General Surgery between April 2010 and January 2015 were retrospectively investigated. Approval for this study was gained from the Ethics Committee of the institution, and signed informed consent forms were obtained from all patients who were included in the study. Patients with a past medical history of any ongoing inflammatory conditions (inflammatory bowel disease, arthritis, etc.) were not included to the study group. In the early postoperative period (postoperative first month), deaths due to either surgical complications or due to non-cancer-related reasons were excluded. On the other hand, in the late-term follow-up period, patients who did not regularly attend to scheduled office visits and patients who could not be reached by various communication methods were excluded as well. Finally, 115 patients, whose preoperatively measured serum CRP and albumin levels were available, and whose early postoperative care, and late-term follow-ups could be carried out completely and reliably were enrolled to the study. Data including the following clinicopathological factors were selected and evaluated: age, gender, preoperative laboratory test values (hemogram and biochemical test parameters), neutrophil-to-lymphocyte ratio (NLR; NLR was defined as the absolute neutrophil count divided by the absolute lymphocyte count), tumor markers (CEA, CA19-9, and AFP), location of the tumor, tumor size, lymphovascular/perineural invasion, tumor differentiation, distant metastasis, TNM stage
(AJCC; American Joint Committee on Cancer 7th ed.), and early postoperative surgical outcomes (postoperative first month). After discharge from the hospital, the patients were followed up on scheduled routine office visits, and the data of the late-term follow-up period including survival outcomes was also collected and evaluated. The GPS was preoperatively estimated as described previously by Forrest et al.10 Patients with both an elevated level of CRP (>1.0 mg/dL) and hypoalbuminemia ( 0.05). The comparisons of the demographic and laboratory parameters between the three GPS groups are summarized in Table 2. Tumor size was significantly increased with increasing GPS (p = 0.019). Other clinicopathological parameters including tumor differentiation, lymphovascular invasion, perineural invasion, presence of distant metastasis, and TNM stages did
The comparison of the demographic and laboratory parameters between the three GPS groups
Parameters
Age (years) Gender Hematocrit (%) Hemoglobin (g/dL) Hemoglobin < 12 g/dL White blood cell (103/mm3) Platelet (103/mm3) Platelet ≥ 400000/mm3 Neutrophil (103/mm3) Lymphocyte (103/mm3) Neutrophil-to-lymphocyte ratio (NLR) NLR ≥ 3 Total protein (g/dL) Albumin (g/dL) C-reactive protein (mg/dL)
GPS 0 (n = 30)
GPS 1 (n = 55)
GPS 2 (n = 30)
62.2 ± 13.2 16 (53.3 %) 14 (46.7 %) 36.5 ± 5.2 12.0 ± 1.9
66.2 ± 12.2 31 (56.4 %) 24 (43.6 %) 35.6 ± 5.7 11.5 ± 2.1
69.8 ± 12.2 17 (56.7 %) 13 (43.3 %) 34.2 ± 4.7 11.1 ± 1.7
0.067 0.970b
n (%) Mean ± SD Mean ± SD n (%) Mean ± SD Median (Min-Max) Mean ± SD Median (Min-Max)
14 (46.7 %) 7.9 ± 2.7 277.4 ± 91.7 3 (10.0 %) 6.3 ± 2.9 4.9 (1.6-22.7) 1.8 ± 1.0 1.4 (0.4-5.4)
24 (43.6 %) 7.7 ± 2.1 293.0 ± 86.8 7 (12.7 %) 6.3 ± 2.9 5.4 (2.4-15.2) 2.2 ± 2.7 1.6 (0.7-2.0)
13 (43.3 %) 8.6 ± 2.7 278.6 ± 82.3 4 (13.3 %) 6.2 ± 3.4 5.4 (1.2-17.5) 1.6 ± 0.8 1.6 (0.6-3.9)
0.287 0.312 0.582c 1.000d 0.832c
Mean ± SD Median (Min-Max) n (%) Mean ± SD Mean ± SD Mean ± SD Median (Min-Max)
4.9 ± 5.2 2.7 (1.5-21.3) 12 (40.0 %) 6.7 ± 0.7 4.0 ± 0.4 0.6 ± 0.3 0.7 (0.0-1.0) n = 21
4.5 ± 3.7 3.1 (1.3-19.7) 31 (56.4 %) 6.5 ± 0.9 3.6 ± 0.8 1.9 ± 1.8 1.4 (0.0-9.9) n = 37
4.8 ± 5.5 3.6 (3.0-29.1) 18 (60.0 %) 5.5 ± 0.8 2.8 ± 0.4 4.4 ± 4.1 2.5 (1.4-19.0) n = 14
0.875c
Mean ± SD Median (Min-Max) n (%)
9.2 ± 20.7 1.9 (0.3-84.8) 6 (28.6 %) n = 12 31.4 ± 44.0 11.1 (0.8-147.2) 8 (66.7 %) n = 13 3.2 ± 1.6 3.1 (0.4-5.7) 13 (100.0 %)
13.9 ± 40.0 2.1 (0.5-229.3) 8 (21.6 %) n = 25 40.4 ± 66.8 15.2 (0.8-258.9) 20 (80.0 %) n = 19 2.3 ± 1.5 1.7 (1.2-6.7) 19 (100.0 %)
33.7 ± 94.3 2.2 (0.7-356.9) 4 (28.6 %) n = 12 20.4 ± 29.4 10.8 (2.0-109.4) 11 (91.7 %) n=9 4.2 ± 3.0 3.3 (1.0-10.5) 8 (88.9 %)
Mean ± SD Male Female Mean ± SD Mean ± SD
n (%) n (%)
Carcinoembryonic antigen (CEA) CEA (ng/mL) CEA ≥ 5 ng/mL Cancer antigen 19–9 (CA 19–9) CA 19–9 (U/mL) CA 19–9 ≥ 5 U/mL Alpha-fetoprotein (AFP) AFP (ng/mL) AFP < 9 ng/mL
Mean ± SD Median (Min-Max) n (%) Mean ± SD Median (Min-Max) n (%)
n number of patients, Mean ± SD mean ± standard deviation, Median (Min-Max) median (minimum-maximum) a
One-way ANOVA
b
Pearson’s chi-squared test
c
Kruskal-Wallis test
d
Fisher-Freeman-Halton test
*p < 0.05; **p < 0.01
pa
Glasgow prognostic score (GPS)
0.237 0.213
0.522c
0.237b 0.05). The comparisons of the surgical and clinicopathological parameters between the three GPS groups are shown in Table 3. According to univariate analyses, CEA levels exceeding 5 ng/mL, and GPS 2 were detected to be significantly related with mortality (p = 0.001 and p = 0.008, respectively). However, other variables including age, gender, hemoglobin levels, platelet levels, NLR, tumor location, tumor size, and differentiation were not found to assert significance (p > 0.05). Multivariate analyses were carried out for significant or nearly significant variables detected in the univariate analyses. According to multivariate analyses, CEA and GPS were found to be significantly effecting mortality, as well, revealing that these parameters were independent risk factors (p = 0.001 and p = 0.009, respectively). The univariate and multivariate analyses for cancer-related mortality are summarized in Table 4, whereas the independent risk factors detected in the present study are plotted in Fig. 1. At the end of the late-term follow-up period, the relation between survival and the GPS was evaluated with use of the Log rank test, and it was detected that cancer-related mortality rates increased, while cancer-specific survival significantly decreased as the GPS increased (p = 0.016). The distribution Table 3
of the survival rates and periods among the three GPS groups is summarized in Table 5. The Kaplan-Meier plot for cancerspecific survival of the three GPS groups is shown in Fig. 2.
Discussion Although it is recognized that the development of cancer has a genetic basis, there is increasing evidence that the host inflammatory response plays an important role in the development and progression of cancer.11 Inflammation can induce the initial genetic mutation and epigenetic changes for cancer initiation, can modify the tissue microenvironment that permits cancer cells to progress and metastasize, and can also suppress the immune response to tumor cells.12 Links between cancer and inflammation were first established in the nineteenth century and were based on observations that tumors often arose at sites of chronic inflammation and that inflammatory cells are present in biopsy samples collected from tumor tissues. Today, it is well known that cancer promotes release of proinflammatory cytokines from tumor cells.13 Experimental studies showed that oncogenes could activate several inflammatory factors including colony stimulating factors (CSFs), interleukin 1β, chemokines/ chemokine receptors (CCL2, CCL20, IL-8, CXCL12, CXCR4), proteases such as matrix metalloproteinases (MMP7, MMP9, MMP10), and urokinase-type plasminogen activator and its receptor.5 These cytokines interact with the
The comparison of the surgical and clinicopathological parameters between the three GPS groups
Parameters
Tumor size (cm.) Tumor size ≥ 5 cm. Tumor differentiation
Lymphovascular invasion Perineural invasion Distant metastasis TNM Stage (AJCC) Anastomotic leakage Wound infection
Glasgow prognostic score (GPS)
Mean ± SD Median (Min-Max) n (%) Well differentiated Moderately differentiated Poorly differentiated n (%) n (%) n (%) Stage I + II Stage III + IV n (%) n (%)
n (%) n (%) n (%)
n (%) n (%)
GPS 0 (n = 30)
GPS 1 (n = 55)
GPS 2 (n = 30)
4.3 ± 1.5 4.0 (2.0-8.0) 11 (36.7 %) 2 (6.7 %) 27 (90.0 %) 1 (3.3 %) 16 (53.3 %) 9 (30.0 %) 1 (3.3 %) 15 (50.0 %) 15 (50.0 %) 2 (6.7 %) 1 (3.3 %)
5.3 ± 2.1 5.0 (2.0-12.0) 35 (63.6 %) 8 (14.5 %) 39 (70.9 %) 8 (14.5 %) 36 (65.6 %) 24 (43.6 %) 3 (5.5 %) 28 (50.9 %) 27 (49.1 %) 2 (3.6 %) 1 (1.8 %)
5.4 ± 2.1 5.3 (1.0-12.0) 21 (70.0 %) 3 (10.0 %) 25 (83.3 %) 2 (6.7 %) 16 (53.3 %) 10 (33.3 %) 2 (6.7 %) 15 (50.0 %) 15 (50.0 %) 1 (3.3 %) 7 (23.3 %)
n number of patients, Mean ± SD mean ± standard deviation, Median (Min-Max) median (minimum-maximum) a
Kruskal-Wallis test
b
Pearson’s chi-squared test
c
Fisher-Freeman-Halton test
*p < 0.05; **p < 0.01
p
0.019a * 0.017b * 1.000c
0.418b 0.401b 1.000c 0.995b 0.846c 0.003c **
J Gastrointest Surg Table 4
The univariate and multivariate analyses for cancer-related mortality
Variable
Mortality No (n = 96) n (%)
Univariate Yes (n = 19) n (%)
HR
Multivariate 95 % CI
p
HR
95 % CI
p
2.132
0.66–6.91
0.207
6.932
2.27–21.16
0.001**
Age ≥ 70 years
39 (40.6 %)
12 (63.2 %)
2.257
0.89–5.74
0.088
Female gender
43 (44.8 %)
8 (42.1 %)
0.913
0.37–2.27
0.844
Hb < 12 g/dL Plt ≥ 400000/mm3 NLR ≥ 3 CEA ≥ 5 ng/mL
56 (58.3 %) 12 (12.5 %)
13 (68.4 %) 2 (10.5 %)
1.685 1.103
0.64–4.43 0.25–4.79
0.291 0.896
49 (51.0 %) 9 (15.5 %)
12 (63.2 %) 9 (64.3 %)
1.648 7.389
0.65–4.19 2.44–22.33
0.294 0.001**
Tumor location, rectum Tumor size ≥ 5 cm
34 (35.4 %) 56 (58.3 %)
5 (26.3 %) 11 (57.9 %)
0.626 1.274
0.22–1.74 0.51–3.18
0.370 0.604
Poorly differentiated tumor
7 (7.3 %)
4 (21.1 %)
2.550
0.84–7.71
0.097
2.002
0.61–6.59
0.253
GPS, 2
21 (21.9 %)
9 (47.4 %)
3.389
1.37–8.41
0.008**
4.102
1.42–11.86
0.009**
HR hazard ratio, CI confidence interval, n number of patients, Hb hemoglobin, NLR neutrophil-to-lymphocyte ratio, CEA carcinoembryonic antigen, GPS Glasgow prognostic score *p < 0.05; **p < 0.01
immunovascular system and facilitate cancer growth, tumor cell migration, proliferation, survival, angiogenesis, invasion, , dissemination, and metastasis.5 13 Cancer-producing proinflammatory cytokines induce the acute phase protein, which is a key marker of systemic inflammation.14 Since the initial studies of using CRP, as a sensitive measure of the systemic inflammatory response, stated independent prognostic value in operable CRC, there have been numerous studies reporting the prognostic value of CRP and other markers of the systemic inflammatory response in a , wide variety of operable and inoperable cancers. 8 15 Systemic inflammation has been found to be positively associated with weight loss, hypermetabolism, anorexia, and poor prognosis in cancer patients.16 In CRC, inflammation has an important role in the initiation and progression.6 In colitis-associated colon cancer, chronic inflammation causes oxidative damage to the DNA, Fig. 1 The independent risk factors effecting cancer-related mortality. [CEA: Carcinoembryonic antigen; GPS: Glasgow Prognostic Score]
leading to p53 mutations in tumor cells, and the inflamed epithelium and the inflammatory microenvironment at the tumor border can influence several key stages of invasion and metastasis.17 McMillan et al. reported that the prognostic score based on the serum CRP and albumin levels had an independent prognostic value after resection of CRC.18 Both CRP and albumin are acute phase proteins. Acute phase proteins are produced in the liver in response to inflammatory cytokines, mainly IL-6 and IL-1β. CRP level can be elevated to as much as 1000-fold after an inflammatory stimulus.19 But, the albumin level is decreased in response to inflammation. The albumin level is known to be decreased in cancer patients due to malnutrition and systemic inflammation.9 Recent studies have shown that elevated serum CRP levels may be associated with tumor size, distant metastasis, vascular invasion, lymph node metastasis, and tumor recurrence, resulting in poor prognosis.13 It has also been reported that
J Gastrointest Surg Table 5 The distribution of the survival rates and periods among the three GPS goups
GPS
n
Mortality
Survival rate
Yes
No
Cumulative survival Mean
SD
Mean survival period (months)
GPS 0
30
3
27
90.0 %
86.6
5.7
37.97 ± 1.66
GPS 1
55
7
48
87.3 %
83.6
6.0
29.85 ± 1.10
GPS 2
30
9
21
70.0 %
30.0
15.4
21.88 ± 0.89
GPS Glasgow prognostic score, n number of patients, SD standard deviation
serum albumin participate in systemic inflammatory response and that decline of its serum level is a poor prognostic factor for long-term survival in patients with various cancers.20 Based on these reports, GPS, incorporating CRP and serum albumin levels, may reflect both presence of the systemic inflammatory response (CRP), and the progressive nutritional decline (albumin) in cancer patients, resulting in poor survival outcome.13 Forrest et al. designed the GPS, an inflammationbased prognostic score, which was assessed by simply using the serum CRP and serum albumin levels.10 The GPS has been proved to be an independent prognostic factor in many studies which have been performed in unselected cohorts, operable, and inoperable cancer patients.4 In our study, we retrospectively evaluated the GPS of surgically treated CRC patients. Mauricio et al. reported that malnutrition is highly prevalent in patients with CRC, and that the nutritional status was associated with GPS.21 CRP level was reported to be higher in patients with colon cancer and larger tumors. 22 Anemia and thrombocytosis are considered to be the common manifestations Fig. 2 The Kaplan-Meier plot for cancer-specific survival among the three GPS groups
induced by inflammatory cytokines, whereas NLR is another well-known indicator of systemic inflammatory response and a , prognostic factor in CRC patients.19 23 There are also studies reporting that the GPS was positively correlated with high levels of CEA.24 In the present study, mean albumin levels were significantly decreased, as the mean CRP levels were significantly increased with increasing GPS, and these results are due to these parameters being the standard variables of the scoring system. The decrease in the mean total protein levels along with the albumin levels is also considered to be an indicator of worsened malnutrition related with higher GPS in CRC patients. However, other parameters including age, gender, hemogram values, NLR, and tumor markers did not show significant difference when compared among the GPS groups (Table 2). Choi et al. reported that in comparison with rectal cancer, colon cancer and larger tumors had a close relationship with a high GPS.14 McMillan et al. showed that the GPS was not related with Dukes’ stage of the tumor.18 But, other authors reported that the GPS was positively correlated with advanced
J Gastrointest Surg
TNM stage.24 In our study, the GPS was significantly higher in patients with larger tumors (Table 3). However, no significant difference could be detected when tumor location was analyzed (Table 4). Other clinicopathological parameters including tumor differentiation, lymphovascular invasion, perineural invasion, presence of distant metastasis, and TNM stages did not reveal significant difference among the GPS groups (Table 3). There are numerous studies revealing that increased GPS was associated with increased weight loss, poor performance status, increased comorbidity, increased proinflammatory and , , , angiogenic cytokines, and complications on treatment.4 14 21 25 When early postoperative surgical outcomes of our study group were analyzed, it was found that wound infection rates were significantly higher in the GPS 2 group. However, no significant alteration was detected among the three groups in concern of anastomotic leakage rates (Table 3). Many studies confirmed that the GPS was a good indicator of prognosis in several kinds of cancers including lung cancer,10 hepatocellular carcinoma,13 gastric cancer,26 and , , CRC.14 18 21 Additionally, it was reported that preoperative CEA has independent prognostic values in CRC, and that preoperatively elevated CEA is correlated with worsened , outcomes.27 28 In the present study, according to multivariate analyses, preoperative CEA levels exceeding 5 ng/mL and GPS 2 were found to be significantly effecting mortality, revealing that these parameters were independent risk factors (Table 4, Fig. 1). At the end of the late-term follow-up period of our patients, it was detected that cancer-specific survival significantly decreased as the GPS increased (Table 5, Fig. 2). The systemic inflammatory response is a major factor in the relationship between nutritional decline and poor outcome in patients with cancer.4 GPS may be included, in addition or in preference to the current definitions of cachexia, with tumor staging as part of the routine assessment of cancer patients. MacDonald recently concluded that particularly in the more aggressive tumor types (e.g., pancreas and lung), the future of patients with elevated GPS is so grim that they should be given precachexia status and offered multimodal therapy which may delay the onset of cachexia and/or death.29 As a consequence, this will highlight the need not only to treat the tumor but also the systemic inflammatory response, a potentially less intractable target compared with established weight loss and/or poor performance status. Further work is required to define the value of GPS as a stratification factor, as a selection criteria in randomized clinical trials and as a therapeutic target in patients with cancer. Also, how the prognostic value of the GPS compares with that of other measures of the systemic inflammatory response, such as the NLR.4 We believe that the present results provide good evidence of systemic inflammation acting as a factor promoting the fatal progression in most CRC patients, and the moderation of the
systemic inflammatory response will become, in the future, as important a therapeutic target as the tumor itself. Our results suggest that anti-inflammatory agents with other therapeutic modalities may prolong the survival of CRC patients.30
Conclusion In conclusion, our study showed that the GPS was an independent variable from tumor stage and a significant prognostic factor in CRC patients. The GPS should be included in the routine preoperative assessment of all surgically treated CRC patients. Acknowledgments The authors would like to thank Mrs. Emire BORα for the statistical work-up of this study. α Biostatistics Specialist, EMPIAR Statistics, Istanbul, Turkey Compliance with Ethical Standards Approval for this study was gained from the Ethics Committee of Istanbul Medeniyet University, School of Medicine, Goztepe Training and Research Hospital, and signed informed consent forms were obtained from all patients. Each author hereby acknowledges that the final state of this manuscript is prepared and sent with his/her approval having been taken. Additionally, all authors declare that no source of funding exists, and that they have no direct or indirect commercial financial incentive associated with publishing this article. The authors also confirm that this manuscript has not been published and is not under simultaneous consideration by another journal or electronic publication. All authors have read and complied with the requirements set forth in the Instructions to Authors. Conflict of Interest The corresponding author and all co-authors declare that they have no conflict of interest.
References 1.
Global Burden of Disease Cancer Collaboration, Fitzmaurice C, Dicker D, Pain A, Hamavid H, Moradi-Lakeh M, MacIntyre MF, Allen C, Hansen G, Woodbrook R, Wolfe C, Hamadeh RR, Moore A, Werdecker A, Gessner BD, Te Ao B, McMahon B, Karimkhani C, Yu C, Cooke GS, Schwebel DC, Carpenter DO, Pereira DM, Nash D, Kazi DS, De Leo D, Plass D, Ukwaja KN, Thurston GD, Yun Jin K, Simard EP, Mills E, Park EK, Catalá-López F, de Veber G, Gotay C, Khan G, Hosgood HD 3rd, Santos IS, Leasher JL, Singh J, Leigh J, Jonas J, Sanabria J, Beardsley J, Jacobsen KH, Takahashi K, Franklin RC, Ronfani L, Montico M, Naldi L, Tonelli M, Geleijnse J, Petzold M, Shrime MG, Younis M, Yonemoto N, Breitborde N, Yip P, Pourmalek F, Lotufo PA, Esteghamati A, Hankey GJ, Ali R, Lunevicius R, Malekzadeh R, Dellavalle R, Weintraub R, Lucas R, Hay R, Rojas-Rueda D, Westerman R, Sepanlou SG, Nolte S, Patten S, Weichenthal S, Abera SF, Fereshtehnejad SM, Shiue I, Driscoll T, Vasankari T, Alsharif U, Rahimi-Movaghar V, Vlassov VV, Marcenes WS, Mekonnen W, Melaku YA, Yano Y, Artaman A, Campos I, MacLachlan J, Mueller U, Kim D, Trillini M, Eshrati B, Williams HC, Shibuya K, Dandona R, Murthy K, Cowie B, Amare AT, Antonio CA, Castañeda-Orjuela C, van Gool CH, Violante F, Oh IH, Deribe K, Soreide K, Knibbs L, Kereselidze M, Green M, Cardenas R, Roy N, Tillman T, Li Y,
J Gastrointest Surg
2.
3.
4.
5. 6.
7.
8.
9.
10.
11.
12. 13.
14.
15.
Krueger H, Monasta L, Dey S, Sheikhbahaei S, Hafezi-Nejad N, Kumar GA, Sreeramareddy CT, Dandona L, Wang H, Vollset SE, Mokdad A, Salomon JA, Lozano R, Vos T, Forouzanfar M, Lopez A, Murray C, Naghavi M. The Global Burden of Cancer 2013. JAMA Oncol. 2015;1(4):505–27. Coleman MP, Rachet B, Woods LM, Mitry E, Riga M, Cooper N, Quinn MJ, Brenner H, Estève J. Trends and socioeconomic inequalities in cancer survival in England and Wales up to 2001. Br J Cancer. 2004;90(7):1367–73. McArdle CS, Hole DJ. Outcome following surgery for colorectal cancer: analysis by hospital after adjustment for case-mix and deprivation. Br J Cancer. 2002;86(3):331–5. McMillan DC. The systemic inflammation-based Glasgow Prognostic Score: a decade of experience in patients with cancer. Cancer Treat Rev. 2013;39(5):534–40. Mantovani A, Allavena P, Sica A, Balkwill F. Cancer-related inflammation. Nature. 2008;454(7203):436–44. Triantafillidis JK, Nasioulas G, Kosmidis PA. Colorectal cancer and inflammatory bowel disease: epidemiology, risk factors, mechanisms of carcinogenesis and prevention strategies. Anticancer Res. 2009;29(7):2727–37. Rostom A, Dubé C, Lewin G, Tsertsvadze A, Barrowman N, Code C, Sampson M, Moher D; U.S. Preventive Services Task Force. Nonsteroidal anti-inflammatory drugs and cyclooxygenase-2 inhibitors for primary prevention of colorectal cancer: a systematic review prepared for the U.S. Preventive Services Task Force. Ann Intern Med. 2007;146(5):376–89. Roxburgh CS, McMillan DC. Role of systemic inflammatory response in predicting survival in patients with primary operable cancer. Future Oncol. 2010;6(1):149–63. McMillan DC, Watson WS, O’Gorman P, Preston T, Scott HR, McArdle CS. Albumin concentrations are primarily determined by the body cell mass and the systemic inflammatory response in cancer patients with weight loss. Nutr Cancer. 2001;39(2):210–3. 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-smallcell lung cancer. Br J Cancer. 2003;89(6):1028–30. Colotta F, Allavena P, Sica A, Garlanda C, Mantovani A. Cancerrelated inflammation, the seventh hallmark of cancer: links to genetic instability. Carcinogenesis. 2009;30(7):1073–81. Trinchieri G. Cancer and inflammation: an old intuition with rapidly evolving new concepts. Annu Rev Immunol. 2012;30:677–706. Kinoshita A, Onoda H, Imai N, Iwaku A, Oishi M, Tanaka K, Fushiya N, Koike K, Nishino H, Matsushima M, Saeki C, Tajiri H. The Glasgow Prognostic Score, an inflammation based prognostic score, predicts survival in patients with hepatocellular carcinoma. BMC Cancer. 2013;13:52. Choi KW, Hong SW, Chang YG, Lee WY, Lee B, Paik IW, Lee H. Inflammation-based score (Glasgow prognostic score) as an independent prognostic factor in colorectal cancer patients. Ann Surg Treat Res. 2014;86(6):309–13. McMillan DC, Canna K, McArdle CS. Systemic inflammatory response predicts survival following curative resection of colorectal cancer. Br J Surg. 2003;90(2):215–9.
16.
17. 18.
19. 20.
21.
22.
23.
24.
25.
26.
27.
28.
29. 30.
Deans C, Wigmore SJ. Systemic inflammation, cachexia and prognosis in patients with cancer. Curr Opin Clin Nutr Metab Care. 2005;8(3):265–9. Terzić J, Grivennikov S, Karin E, Karin M. Inflammation and colon cancer. Gastroenterology. 2010;138(6):2101–2114.e5. McMillan DC, Crozier JE, Canna K, Angerson WJ, McArdle CS. Evaluation of an inflammation-based prognostic score (GPS) in patients undergoing resection for colon and rectal cancer. Int J Colorectal Dis. 2007;22(8):881–6. Gabay C, Kushner I. Acute-phase proteins and other systemic responses to inflammation. N Engl J Med. 1999;340(6):448–54. Kobayashi T, Teruya M, Kishiki T, Endo D, Takenaka Y, Tanaka H, Miki K, Kobayashi K, Morita K. Inflammation-based prognostic score, prior to neoadjuvant chemoradiotherapy, predicts postoperative outcome in patients with esophageal squamous cell carcinoma. Surgery. 2008;144(5):729–35. Maurício SF, da Silva JB, Bering T, Correia MI. Relationship between nutritional status and the Glasgow Prognostic Score in patients with colorectal cancer. Nutrition. 2013;29(4):625–9. Shiu YC, Lin JK, Huang CJ, Jiang JK, Wang LW, Huang HC, Yang SH. Is C-reactive protein a prognostic factor of colorectal cancer? Dis Colon Rectum. 2008;51(4):443–9. Walsh SR, Cook EJ, Goulder F, Justin TA, Keeling NJ. Neutrophillymphocyte ratio as a prognostic factor in colorectal cancer. J Surg Oncol. 2005;91(3):181–4. Ishizuka M, Nagata H, Takagi K, Horie T, Kubota K. Inflammationbased prognostic score is a novel predictor of postoperative outcome in patients with colorectal cancer. Ann Surg. 2007;246(6): 1047–51. Proctor MJ, Morrison DS, Talwar D, Balmer SM, Fletcher CD, O’Reilly DS, Foulis AK, Horgan PG, McMillan DC. A comparison of inflammation-based prognostic scores in patients with cancer. A Glasgow Inflammation Outcome Study. Eur J Cancer. 2011;47(17):2633–41. Mohri Y, Tanaka K, Ohi M, Toiyama Y, Yasuda H, Inoue Y, Uchida K, Kusunoki M. Inflammation-based prognostic score as a predictor of postoperative gastric cancer recurrence. Anticancer Res. 2012;32(10):4581–4. Wang RF, Song BR, Peng JJ, Cai GX, Liu FQ, Wang MH, Cai SJ, Ye X. The Prognostic Value of Preoperative Serum CEA and CA19-9 Values in Stage I-III Colorectal Cancer. Hepatogastroenterology. 2014;61(132):994–9. Chen L, Jiang B, Di J, Zhang C, Wang Z, Zhang N, Xing J, Cui M, Yang H, Yao Z, Su X. Predictive value of preoperative detection of CEA and CA199 for prognosis in patients with stage II-III colorectal cancer. Zhonghua Wei Chang Wai Ke Za Zhi. 2015;18(9):914–9. MacDonald N. Terminology in cancer cachexia: importance and status. Curr Opin Clin Nutr Metab Care. 2012;15(3):220–5. Bastiaannet E, Sampieri K, Dekkers OM, de Craen AJ, van HerkSukel MP, Lemmens V, van den Broek CB, Coebergh JW, Herings RM, van de Velde CJ, Fodde R, Liefers GJ. Use of aspirin postdiagnosis improves survival for colon cancer patients. Br J Cancer. 2012;106(9):1564–70.
