Med Oncol (2014) 31:287 DOI 10.1007/s12032-014-0287-y
SHORT COMMUNICATION
Glasgow Prognostic Score predicts poor prognosis among advanced biliary tract cancer patients with good performance status Toshikazu Moriwaki • Kazunori Ishige • Masahiro Araki • Shigemasa Yoshida Masaaki Nishi • Mikio Sato • Takeshi Yamada • Yoshiyuki Yamamoto • Mitsuharu Ozeki • Hiroyasu Ishida • Takashi Yamaguchi • Kenji Matsuda • Tetsuya Murashita • Masato Abei • Ichinosuke Hyodo
•
Received: 22 September 2014 / Accepted: 7 October 2014 / Published online: 16 October 2014 Ó Springer Science+Business Media New York 2014
Abstract Advanced cancer patients with good performance status (PS) sometimes show poor prognosis despite receiving some chemotherapies. We evaluated prognosis of chemo-naı¨ve advanced biliary tract cancer (ABTC) patients with good PS by Glasgow Prognostic Score (GPS). Sixty-two patients with Eastern Cooperative Oncology Group PS 0 or 1 were retrospectively analyzed, using multivariate Cox regression. GPS was defined with serum levels of two parameters, albumin[3.5 g/dl and C-reactive protein \1.0 mg/dl (both as 0, either as 1, and neither as 2). PS 0 (n = 32) and 1 (n = 30) patients had similar survival (P = 0.98). The median overall survival (OS) was 17.0 months for GPS 0 (n = 19), 14.2 months for GPS 1 (n = 17), and 6.4 months for GPS 2 (n = 26). GPS 2 had significantly shorter OS than GPS 0 (P = 0.002) or 1 (P = 0.033). Multivariate analysis identified two independent prognostic factors: GPS (hazard ratio 0.60, 95 % confidence interval
T. Moriwaki (&) K. Ishige T. Yamada Y. Yamamoto M. Abei I. Hyodo Division of Gastroenterology, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-0006, Japan e-mail: [email protected] M. Araki M. Ozeki Division of Gastroenterology and G.I. Oncology, Ibaraki Prefectural Central Hospital and Cancer Center, Kasama, Japan S. Yoshida H. Ishida T. Yamaguchi Department of Gastroenterology, Mito Medical Center, National Hospital Organization, Higashi-Ibaraki gun, Japan
0.40–0.90, P = 0.012) and liver metastasis (hazard ratio 0.43, 95 % CI 0.20–0.90, P = 0.026). GPS was useful for chemonaı¨ve ABTC patients with good PS. Keywords Prognostic factor Stratification factor Biliary tract cancer Chemotherapy Glasgow Prognostic Score ECOG PS
Introduction Biliary tract cancers (BTC) often have poor prognosis because they are diagnosed at advance stages, with distant metastasis [1]. Many patients with advanced BTC (ABTC) undergo palliative chemotherapy with gemcitabine (GEM)-based regimens. GEM plus cisplatin (CDDP) was superior to GEM alone in a randomized phase III study (ABC-02) and has become the standard first-line regimen [2]. Patients with advanced cancers and good performance status (PS) are generally eligible for chemotherapy. In fact, approximately 90 % of patients in the ABC-02 trial had Eastern Cooperative Oncology Group (ECOG) PS 0 or 1. However, some patients have poor prognosis despite having good PS. Glasgow Prognostic Score (GPS), with indices related to systemic inflammation, has a prognostic significance in various malignant solid tumors [3–9]. Here, we aimed to evaluate the prognostic value of GPS in ABTC patients with good ECOG PS undergoing chemotherapy. Materials and methods
M. Nishi K. Matsuda Division of Gastroenterology, Tsukuba Gakuen Hospital, Tsukuba, Japan
Patients and data collection
M. Sato T. Murashita Department of Gastroenterology and Hepatology, Ryugasaki Saiseikai Hospital, Ryugasaki, Japan
This study was approved by the ethics committee of each participating hospital. The following eligibility criteria were used: age C20 years; ECOG PS 0 or 1; treatment with
123
287 Page 2 of 6
GEM, S-1 (a combination of tegafur, 5-chloro-2,4-dehydroxypyridine, and potassium oxonate), GEM plus CDDP, or GEM plus S-1 as first-line chemotherapy between January 2008 and December 2010; diagnosis of unresectable or recurrent intrahepatic cholangiocarcinoma, extrahepatic cholangiocarcinoma, or gallbladder carcinoma; measurable or assessable lesions; absence of active infection; and adequate organ function. The following pretreatment clinical data were collected from clinical charts: age, sex, ECOG PS, primary tumor type, primary tumor resection, and number of metastatic organs. The following baseline laboratory values were collected: serum albumin levels, C-reactive protein (CRP), lactate dehydrogenase (LDH), alkaline phosphatase (ALP), total bilirubin, carcinoembryonic antigen (CEA), and cancer antigen (CA) 19-9. Albumin and CRP values were used for GPS [4]. Patients with a serum albumin level [3.5 g/dl and CRP \1.0 mg/dl were assessed as GPS 0; patients with either decreased albumin levels or increased CRP levels were assessed as GPS 1; and patients with both decreased albumin levels and increased CRP levels were assessed as GPS 2. Statistical analysis Overall survival (OS) was defined as the time from the first date of first-line chemotherapy to death from any cause or to the last follow-up (censored). The Kaplan–Meier method was used to estimate the OS curve with a 95 % confidence interval (CI). OS difference was analyzed by a log-lank test. Analysis of prognostic factors for OS was based on pretreatment clinical and biological variables. Factors included in the univariate analysis were ECOG PS; GPS; age; sex (male or female); total bilirubin; ALP; LDH; CEA; CA 19-9; type of primary tumor; primary resection; number of metastatic organs (B1, C2); the presence of metastases in the following organs: liver, lymph node, lung, and peritoneum; other metastases; and first-line chemotherapy regimen (GEM, S-1, GEM plus CDDP, GEM plus S-1). Factors with P \ 0.2 were included in the multivariate regression analysis. Cox proportional hazards regression model was performed to obtain adjusted hazard ratios (HRs) and to determine independent prognostic factors for OS. P \ 0.05 was considered significant for all analyses. The SPSS software package, version 22.0 (SPSS Inc., Tokyo, Japan), was used for statistical analysis.
Results Patients Clinical data were collected from 67 patients from 5 institutions. Data of five patients were excluded because of
123
Med Oncol (2014) 31:287
lack of ECOG PS and laboratory values. The numbers of patients according to GPS were 19 (31 %) for GPS 0, 17 (27 %) for GPS 1, and 26 (42 %) for GPS 2. The median age was 68 years. Extrahepatic cholangiocarcinoma was the most common ABTC, representing 50 % of the ABTCs. There were no significant differences between the GPS groups except for first-line regimen (P = 0.02) (Table 1). Survival analyses The median OS according to ECOG PS was 11 months (95 % CI 8.2–13.6) and 7.6 months (95 % CI 3.4–11.8) for patients with ECOG PS 0 and ECOG PS 1, respectively (Fig. 1a). There was no significant difference in OS between the patients with ECOG PS 0 and 1 (P = 0.98). The median OS according to GPS was 17 months (95 % CI 4.3–29.6), 14 months (95 % CI 3.2–25.2), and 6.4 months (95 % CI 4.3–8.5) for patients with GPS 0, GPS 1, and GPS 2 (Fig. 1b), respectively. OS was significantly shorter for patients with GPS 2 than that for patients with GPS 0 or 1 (GPS 0 vs. 2, P = 0.002; GPS 1 vs. 2, P = 0.033); there was no significant difference in OS between patients with GPS 0 and 1 (P = 0.32). The median OS for first-line chemotherapy regimen was 8.0 months (95 % CI 4.8–11.2) for GEM; and 10 months (95 % CI 4.8–15.2) for S-1, 11 months (95 % CI 10.6–11.2) for GEM plus CDDP, and 17 months (95 % CI 1.7–31.7) for GEM plus S-1 (P = 0.60). Analysis of prognostic factors The results of univariate and multivariate analyses for OS are shown in Table 2. Univariate analysis for OS indicated that the factors with P \ 0.2 were GPS, ALP, LDH, number of metastatic organs, liver metastasis, peritoneal metastasis, and other metastases. By multivariate analysis, independent favorable prognostic factors were better GPS (HR 0.60, 95 % CI 0.40–0.90, P = 0.012) and the absence of liver metastasis (HR 0.43, 95 % CI 0.20–0.90, P = 0.026).
Discussion GPS is a prognostic factor in various malignant tumors and superior to ECOG PS for predicting treatment response and survival in some cancers [3, 5]. Reportedly, GPS is a prognostic factor in postoperative extrahepatic cholangiocarcinoma patients or in palliative chemotherapy/care ABTC patients [7–9]. However, these studies included patients with poor PS or did not clarify the PS status. In the present study, GPS was a significant prognostic indicator in
Med Oncol (2014) 31:287
Page 3 of 6 287
Table 1 Patient characteristics in patients with good ECOG PS All n = 62
GPS 0 (%)
n = 19
GPS 1 (%)
n = 17
GPS 2 (%)
n = 26
P value (%)
Age Median (range) 68 (44–85)
68 (49–83)
70 (54–85)
67.5 (44–80)
17 (27)
7 (37)
5 (29)
5 (19)
C75 years old
0.42
Sex
0.10
Male 33 (53)
8 (42)
7 (41)
18 (69)
29 (47)
11 (58)
10 (59)
8 (31)
Female ECOG PS
0.20
0 32 (52)
13 (68)
7 (41)
12 (46)
30 (48)
6 (32)
10 (59)
14 (54)
462 (231–3,013)
389 (231–2,157)
441 (250–1,448)
542 (233–3,013)
203 (123–798)
199 (146–765)
206 (146–302)
200 (123–798)
4.9 (1.1–1,852)
5.1 (1.1–578)
3.3 (1.7–46)
6.0 (1.1–1,852)
351 (1.2–380,815)
246 (3.2–53,600)
256 (2.0–27,984)
390 (1.2–380,815)
1 ALP Median (range) LDH Median (range) CEA Median (range) CA19-9 Median (range)
Primary tumor location
0.32
Extrahepatic 31 (50)
6 (32)
11 (65)
14 (54)
13 (21)
5 (26)
2 (12)
6 (23)
18 (29)
8 (42)
4 (23)
6 (23)
Intrahepatic Gallbladder Primary tumor resection
0.91
Yes 13 (21)
4 (21)
3 (18)
6 (23)
49 (79)
15 (79)
14 (82)
20 (77)
24 (39)
7 (37)
5 (29)
12 (46)
31 (50)
11 (58)
5 (29)
15 (58)
4 (6)
1 (5)
0
3 (12)
17 (27)
4 (21)
6 (35)
7 (27)
6 (10)
2 (11)
1 (6)
3 (12)
No Metastatic organ Liver
0.53
Lymph node
0.14
Lung
0.31 0.63
Peritoneum Other Number of metastatic organs
0.82 0.71
123
287 Page 4 of 6
Med Oncol (2014) 31:287
Table 1 continued All n = 62
GPS 0 (%)
n = 19
GPS 1 (%)
n = 17
GPS 2 (%)
n = 26
P value (%)
0 12 (19)
4 (21)
5 (29)
3 (12)
27 (44)
8 (42)
8 (47)
11 (42)
15 (24)
5 (26)
3 (18)
7 (27)
8 (13)
2 (11)
1 (6)
5 (19)
1 2 C3 First-line chemotherapy
0.02
GEM alone 36 (58)
8 (42)
8 (47)
20 (77)
14 (22)
5 (26)
7 (41)
2 (8)
9 (15)
6 (32)
1 (6)
2 (8)
3 (5)
0
1 (6)
2 (8)
S-1 alone GEM ? CDDP GEM ? S-1 Subsequent treatment
0.10
Chemotherapy 37 (60)
12 (63)
12 (71)
13 (50)
22 (35)
5 (26)
4 (23)
13 (50)
3 (5)
2 (11)
1 (6)
0
BSC Treating first-line chemotherapy ECOG PS Eastern Cooperative Oncology Group Performance Status, GPS Glasgow Prognostic Score, GEM gemcitabine, CDDP cisplatin, BSC best supportive care
Fig. 1 Kaplan–Meier curves of overall survival (OS). a According to ECOG PS, the median OS was 10.9 months for the patients with PS 0 and 7.6 months for the patients with PS 1 (P = 0.98). b According to GPS in patients with PS 0 or 1, the median OS was 17.0 months for GPS 0, 14.2 months for GPS 1, and 6.4 months for GPS 2 (GPS 0 vs. GPS 2, P = 0.002; GPS 1 versus GPS 2, P = 0.033; GPS 0 versus GPS 1, P = 0.32)
chemo-naı¨ve ABTC patients with good PS: 42 % of them were classified as GPS 2 and showed significantly shorter OS than those having GPS 0 or 1. This indicates that GPS should be evaluated in ABTC patients, with seemingly good condition and PS, who are eligible for chemotherapy. Our results also suggest that GPS was more important as a stratification factor than PS in clinical trials. Generally,
123
ECOG PS up to 2 has been adopted as eligibility criteria for randomized trials, and patients are stratified based on ECOG PS 0/1 or 2. However, patients with ECOG PS 2 were only up to 10 % of all enrolled patients and were often excluded by the eligibility criteria in recent randomized studies for solid tumors, including ABTC [2, 10– 15]. Therefore, stratification with GPS would help evaluate
Med Oncol (2014) 31:287
Page 5 of 6 287
Table 2 Univariate and multivariate analyses for overall survival Variable
Univariate analysis
Multivariate analysis
Median OS, month
P value 0.98
Hazard ratio (95 % CI)
P value
0.60 (0.40–0.90)
0.012
0.54 (0.25–1.14)
0.11
1.48 (0.75–2.92)
0.26
ECOG PS
0 vs. 1
10.9 vs. 7.6
GPS
0 vs. 1 vs. 2
17.0 vs. 14.2 vs. 6.4
0.003
Age, median Sex
Low vs. high Male vs. female
7.7 vs. 10.8 8.9 vs. 11.0
0.23 0.94
ALP, median
Low vs. high
13.8 vs. 7.3
0.11
Total Bilirubin, median
Low vs. high
10.8 vs. 7.6
0.45
LDH, median
Low vs. high
8.0 vs. 10.8
0.17
CEA, median
Low vs. high
11.0 vs. 10.0
0.51
CA19-9, median
Low vs. high
8.0 vs. 11.0
0.34
Primary tumor location
Extrahepatic vs. intrahepatic vs. gallbladder
11.0 vs. 8.0 vs. 10.0
0.95
Primary tumor resection
Yes vs. no
8.0 vs. 10.0
0.84
Number of metastatic organs
B1 vs. C2
10.8 vs. 8.9
0.14
1.40 (0.59–3.28)
0.45
Liver metastasis
No vs. yes
11.0 vs. 8.9
0.038
0.43 (0.20–0.90)
0.026
Lymph node metastasis Peritoneal metastasis
No vs. yes No vs. yes
10.0 vs. 10.0 10.9 vs. 6.8
0.25 0.025
0.49 (0.22–1.07)
0.07
Lung metastasis
No vs. yes
10.0 vs. 4.4
0.52
Other metastasis
No vs. yes
10.8 vs. 7.2
0.092
0.57 (0.20–1.61)
0.29
First-line chemotherapy regimen
GEM alone vs. S-1 alone vs. GEM ? CDDP vs. GEM ? S-1
8.0 vs. 10.0 vs. 10.9 vs. 16.7
0.60
ECOG PS Eastern Cooperative Oncology Group Performance Status, GPS Glasgow Prognostic Score, ALP alkaline phosphatase, LDH lactase dehydrogenase, CEA carcinoembryonic antigen, CA19-9 cancer antigen 19-9, GEM gemcitabine, CDDP cisplatin, OS overall survival, CI confidence interval
more exactly the results of randomized studies for enriched patients with good PS. This study has several limitations: its retrospective nature, small sample size, heterogeneous ABTC origin, and different first-line chemotherapies. In our study, most patients received GEM-containing regimens as first-line chemotherapy. The results for OS were comparable with those reported in prospective studies [2, 16–18]. Further studies are needed to confirm our results. In conclusion, GPS was a useful tool to detect populations with poor prognosis among chemo-naı¨ve ABTC patients with good PS. GPS may be more useful than ECOG PS as a stratification factor in future randomized studies. Acknowledgments This work was supported in part by funding from Tsukuba Cancer Clinical Trial Group (TCTG). Conflict of interest
The authors declare no conflict of interest.
2.
3.
4.
5.
6.
7.
References 8. 1. Miyakawa S, Ishihara S, Horiguchi A, et al. Biliary tract cancer treatment: 5,584 results from the Biliary Tract Cancer Statistics
Registry from 1998 to 2004 in Japan. J Hepatobiliary Pancreat Surg. 2009;16(1):1–7. Valle J, Wasan H, Palmer DH, et al. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J Med. 2010;362(14):1273–81. Crumley AB, Stuart RC, McKernan M, et al. Comparison of an inflammation-based prognostic score (GPS) with performance status (ECOG-ps) in patients receiving palliative chemotherapy for gastroesophageal cancer. J Gastroenterol Hepatol. 2008;23(8 Pt 2):e325–9. Forrest LM, McMillan DC, McArdle CS, et al. 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):1028–30. Forrest LM, McMillan DC, McArdle CS, et al. A prospective longitudinal study of performance status, an inflammation-based score (GPS) and survival in patients with inoperable non-smallcell lung cancer. Br J Cancer. 2005;92(10):1834–6. Hwang JE, Kim HN, Kim DE, et al. Prognostic significance of a systemic inflammatory response in patients receiving first-line palliative chemotherapy for recurred or metastatic gastric cancer. BMC Cancer. 2011;11:489. Iwaku A, Kinoshita A, Onoda H, et al. The Glasgow Prognostic Score accurately predicts survival in patients with biliary tract cancer not indicated for surgical resection. Med Oncol. 2014;31(1):787. 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.
123
287 Page 6 of 6 9. Oshiro Y, Sasaki R, Fukunaga K, et al. Inflammation-based prognostic score is a useful predictor of postoperative outcome in patients with extrahepatic cholangiocarcinoma. J Hepatobiliary Pancreat Sci. 2013;20(3):389–95. 10. Falcone A, Ricci S, Brunetti I, et al. Phase III trial of infusional fluorouracil, leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) compared with infusional fluorouracil, leucovorin, and irinotecan (FOLFIRI) as first-line treatment for metastatic colorectal cancer: the Gruppo Oncologico Nord Ovest. J Clin Oncol. 2007;25(13):1670–6. 11. Buzdar AU, Suman VJ, Meric-Bernstam F, et al. Fluorouracil, epirubicin, and cyclophosphamide (FEC-75) followed by paclitaxel plus trastuzumab versus paclitaxel plus trastuzumab followed by FEC-75 plus trastuzumab as neoadjuvant treatment for patients with HER2-positive breast cancer (Z1041): a randomised, controlled, phase 3 trial. Lancet Oncol. 2013;14(13):1317–25. 12. Conroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817–25. 13. Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial. Lancet Oncol. 2014;. doi:10.1016/S1470-2045(14)70330-4.
123
Med Oncol (2014) 31:287 14. Ueno H, Ioka T, Ikeda M, et al. Randomized phase III study of gemcitabine plus S-1, S-1 alone, or gemcitabine alone in patients with locally advanced and metastatic pancreatic cancer in Japan and Taiwan: GEST study. J Clin Oncol. 2013;31(13):1640–8. 15. Wu YL, Zhou C, Hu CP, et al. Afatinib versus cisplatin plus gemcitabine for first-line treatment of Asian patients with advanced non-small-cell lung cancer harbouring EGFR mutations (LUX-Lung 6): an open-label, randomised phase 3 trial. Lancet Oncol. 2014;15(2):213–22. 16. Morizane C, Okusaka T, Mizusawa J, et al. Randomized phase II study of gemcitabine plus S-1 versus S-1 in advanced biliary tract cancer: a Japan Clinical Oncology Group trial (JCOG 0805). Cancer Sci. 2013;104(9):1211–6. 17. Okusaka T, Nakachi K, Fukutomi A, et al. Gemcitabine alone or in combination with cisplatin in patients with biliary tract cancer: a comparative multicentre study in Japan. Br J Cancer. 2010; 103(4):469–74. 18. Park JS, Oh SY, Kim SH, et al. Single-agent gemcitabine in the treatment of advanced biliary tract cancers: a phase II study. Jpn J Clin Oncol. 2005;35(2):68–73.
SHORT COMMUNICATION
Glasgow Prognostic Score predicts poor prognosis among advanced biliary tract cancer patients with good performance status Toshikazu Moriwaki • Kazunori Ishige • Masahiro Araki • Shigemasa Yoshida Masaaki Nishi • Mikio Sato • Takeshi Yamada • Yoshiyuki Yamamoto • Mitsuharu Ozeki • Hiroyasu Ishida • Takashi Yamaguchi • Kenji Matsuda • Tetsuya Murashita • Masato Abei • Ichinosuke Hyodo
•
Received: 22 September 2014 / Accepted: 7 October 2014 / Published online: 16 October 2014 Ó Springer Science+Business Media New York 2014
Abstract Advanced cancer patients with good performance status (PS) sometimes show poor prognosis despite receiving some chemotherapies. We evaluated prognosis of chemo-naı¨ve advanced biliary tract cancer (ABTC) patients with good PS by Glasgow Prognostic Score (GPS). Sixty-two patients with Eastern Cooperative Oncology Group PS 0 or 1 were retrospectively analyzed, using multivariate Cox regression. GPS was defined with serum levels of two parameters, albumin[3.5 g/dl and C-reactive protein \1.0 mg/dl (both as 0, either as 1, and neither as 2). PS 0 (n = 32) and 1 (n = 30) patients had similar survival (P = 0.98). The median overall survival (OS) was 17.0 months for GPS 0 (n = 19), 14.2 months for GPS 1 (n = 17), and 6.4 months for GPS 2 (n = 26). GPS 2 had significantly shorter OS than GPS 0 (P = 0.002) or 1 (P = 0.033). Multivariate analysis identified two independent prognostic factors: GPS (hazard ratio 0.60, 95 % confidence interval
T. Moriwaki (&) K. Ishige T. Yamada Y. Yamamoto M. Abei I. Hyodo Division of Gastroenterology, University of Tsukuba, 1-1-1 Tennodai, Tsukuba, Ibaraki 305-0006, Japan e-mail: [email protected] M. Araki M. Ozeki Division of Gastroenterology and G.I. Oncology, Ibaraki Prefectural Central Hospital and Cancer Center, Kasama, Japan S. Yoshida H. Ishida T. Yamaguchi Department of Gastroenterology, Mito Medical Center, National Hospital Organization, Higashi-Ibaraki gun, Japan
0.40–0.90, P = 0.012) and liver metastasis (hazard ratio 0.43, 95 % CI 0.20–0.90, P = 0.026). GPS was useful for chemonaı¨ve ABTC patients with good PS. Keywords Prognostic factor Stratification factor Biliary tract cancer Chemotherapy Glasgow Prognostic Score ECOG PS
Introduction Biliary tract cancers (BTC) often have poor prognosis because they are diagnosed at advance stages, with distant metastasis [1]. Many patients with advanced BTC (ABTC) undergo palliative chemotherapy with gemcitabine (GEM)-based regimens. GEM plus cisplatin (CDDP) was superior to GEM alone in a randomized phase III study (ABC-02) and has become the standard first-line regimen [2]. Patients with advanced cancers and good performance status (PS) are generally eligible for chemotherapy. In fact, approximately 90 % of patients in the ABC-02 trial had Eastern Cooperative Oncology Group (ECOG) PS 0 or 1. However, some patients have poor prognosis despite having good PS. Glasgow Prognostic Score (GPS), with indices related to systemic inflammation, has a prognostic significance in various malignant solid tumors [3–9]. Here, we aimed to evaluate the prognostic value of GPS in ABTC patients with good ECOG PS undergoing chemotherapy. Materials and methods
M. Nishi K. Matsuda Division of Gastroenterology, Tsukuba Gakuen Hospital, Tsukuba, Japan
Patients and data collection
M. Sato T. Murashita Department of Gastroenterology and Hepatology, Ryugasaki Saiseikai Hospital, Ryugasaki, Japan
This study was approved by the ethics committee of each participating hospital. The following eligibility criteria were used: age C20 years; ECOG PS 0 or 1; treatment with
123
287 Page 2 of 6
GEM, S-1 (a combination of tegafur, 5-chloro-2,4-dehydroxypyridine, and potassium oxonate), GEM plus CDDP, or GEM plus S-1 as first-line chemotherapy between January 2008 and December 2010; diagnosis of unresectable or recurrent intrahepatic cholangiocarcinoma, extrahepatic cholangiocarcinoma, or gallbladder carcinoma; measurable or assessable lesions; absence of active infection; and adequate organ function. The following pretreatment clinical data were collected from clinical charts: age, sex, ECOG PS, primary tumor type, primary tumor resection, and number of metastatic organs. The following baseline laboratory values were collected: serum albumin levels, C-reactive protein (CRP), lactate dehydrogenase (LDH), alkaline phosphatase (ALP), total bilirubin, carcinoembryonic antigen (CEA), and cancer antigen (CA) 19-9. Albumin and CRP values were used for GPS [4]. Patients with a serum albumin level [3.5 g/dl and CRP \1.0 mg/dl were assessed as GPS 0; patients with either decreased albumin levels or increased CRP levels were assessed as GPS 1; and patients with both decreased albumin levels and increased CRP levels were assessed as GPS 2. Statistical analysis Overall survival (OS) was defined as the time from the first date of first-line chemotherapy to death from any cause or to the last follow-up (censored). The Kaplan–Meier method was used to estimate the OS curve with a 95 % confidence interval (CI). OS difference was analyzed by a log-lank test. Analysis of prognostic factors for OS was based on pretreatment clinical and biological variables. Factors included in the univariate analysis were ECOG PS; GPS; age; sex (male or female); total bilirubin; ALP; LDH; CEA; CA 19-9; type of primary tumor; primary resection; number of metastatic organs (B1, C2); the presence of metastases in the following organs: liver, lymph node, lung, and peritoneum; other metastases; and first-line chemotherapy regimen (GEM, S-1, GEM plus CDDP, GEM plus S-1). Factors with P \ 0.2 were included in the multivariate regression analysis. Cox proportional hazards regression model was performed to obtain adjusted hazard ratios (HRs) and to determine independent prognostic factors for OS. P \ 0.05 was considered significant for all analyses. The SPSS software package, version 22.0 (SPSS Inc., Tokyo, Japan), was used for statistical analysis.
Results Patients Clinical data were collected from 67 patients from 5 institutions. Data of five patients were excluded because of
123
Med Oncol (2014) 31:287
lack of ECOG PS and laboratory values. The numbers of patients according to GPS were 19 (31 %) for GPS 0, 17 (27 %) for GPS 1, and 26 (42 %) for GPS 2. The median age was 68 years. Extrahepatic cholangiocarcinoma was the most common ABTC, representing 50 % of the ABTCs. There were no significant differences between the GPS groups except for first-line regimen (P = 0.02) (Table 1). Survival analyses The median OS according to ECOG PS was 11 months (95 % CI 8.2–13.6) and 7.6 months (95 % CI 3.4–11.8) for patients with ECOG PS 0 and ECOG PS 1, respectively (Fig. 1a). There was no significant difference in OS between the patients with ECOG PS 0 and 1 (P = 0.98). The median OS according to GPS was 17 months (95 % CI 4.3–29.6), 14 months (95 % CI 3.2–25.2), and 6.4 months (95 % CI 4.3–8.5) for patients with GPS 0, GPS 1, and GPS 2 (Fig. 1b), respectively. OS was significantly shorter for patients with GPS 2 than that for patients with GPS 0 or 1 (GPS 0 vs. 2, P = 0.002; GPS 1 vs. 2, P = 0.033); there was no significant difference in OS between patients with GPS 0 and 1 (P = 0.32). The median OS for first-line chemotherapy regimen was 8.0 months (95 % CI 4.8–11.2) for GEM; and 10 months (95 % CI 4.8–15.2) for S-1, 11 months (95 % CI 10.6–11.2) for GEM plus CDDP, and 17 months (95 % CI 1.7–31.7) for GEM plus S-1 (P = 0.60). Analysis of prognostic factors The results of univariate and multivariate analyses for OS are shown in Table 2. Univariate analysis for OS indicated that the factors with P \ 0.2 were GPS, ALP, LDH, number of metastatic organs, liver metastasis, peritoneal metastasis, and other metastases. By multivariate analysis, independent favorable prognostic factors were better GPS (HR 0.60, 95 % CI 0.40–0.90, P = 0.012) and the absence of liver metastasis (HR 0.43, 95 % CI 0.20–0.90, P = 0.026).
Discussion GPS is a prognostic factor in various malignant tumors and superior to ECOG PS for predicting treatment response and survival in some cancers [3, 5]. Reportedly, GPS is a prognostic factor in postoperative extrahepatic cholangiocarcinoma patients or in palliative chemotherapy/care ABTC patients [7–9]. However, these studies included patients with poor PS or did not clarify the PS status. In the present study, GPS was a significant prognostic indicator in
Med Oncol (2014) 31:287
Page 3 of 6 287
Table 1 Patient characteristics in patients with good ECOG PS All n = 62
GPS 0 (%)
n = 19
GPS 1 (%)
n = 17
GPS 2 (%)
n = 26
P value (%)
Age Median (range) 68 (44–85)
68 (49–83)
70 (54–85)
67.5 (44–80)
17 (27)
7 (37)
5 (29)
5 (19)
C75 years old
0.42
Sex
0.10
Male 33 (53)
8 (42)
7 (41)
18 (69)
29 (47)
11 (58)
10 (59)
8 (31)
Female ECOG PS
0.20
0 32 (52)
13 (68)
7 (41)
12 (46)
30 (48)
6 (32)
10 (59)
14 (54)
462 (231–3,013)
389 (231–2,157)
441 (250–1,448)
542 (233–3,013)
203 (123–798)
199 (146–765)
206 (146–302)
200 (123–798)
4.9 (1.1–1,852)
5.1 (1.1–578)
3.3 (1.7–46)
6.0 (1.1–1,852)
351 (1.2–380,815)
246 (3.2–53,600)
256 (2.0–27,984)
390 (1.2–380,815)
1 ALP Median (range) LDH Median (range) CEA Median (range) CA19-9 Median (range)
Primary tumor location
0.32
Extrahepatic 31 (50)
6 (32)
11 (65)
14 (54)
13 (21)
5 (26)
2 (12)
6 (23)
18 (29)
8 (42)
4 (23)
6 (23)
Intrahepatic Gallbladder Primary tumor resection
0.91
Yes 13 (21)
4 (21)
3 (18)
6 (23)
49 (79)
15 (79)
14 (82)
20 (77)
24 (39)
7 (37)
5 (29)
12 (46)
31 (50)
11 (58)
5 (29)
15 (58)
4 (6)
1 (5)
0
3 (12)
17 (27)
4 (21)
6 (35)
7 (27)
6 (10)
2 (11)
1 (6)
3 (12)
No Metastatic organ Liver
0.53
Lymph node
0.14
Lung
0.31 0.63
Peritoneum Other Number of metastatic organs
0.82 0.71
123
287 Page 4 of 6
Med Oncol (2014) 31:287
Table 1 continued All n = 62
GPS 0 (%)
n = 19
GPS 1 (%)
n = 17
GPS 2 (%)
n = 26
P value (%)
0 12 (19)
4 (21)
5 (29)
3 (12)
27 (44)
8 (42)
8 (47)
11 (42)
15 (24)
5 (26)
3 (18)
7 (27)
8 (13)
2 (11)
1 (6)
5 (19)
1 2 C3 First-line chemotherapy
0.02
GEM alone 36 (58)
8 (42)
8 (47)
20 (77)
14 (22)
5 (26)
7 (41)
2 (8)
9 (15)
6 (32)
1 (6)
2 (8)
3 (5)
0
1 (6)
2 (8)
S-1 alone GEM ? CDDP GEM ? S-1 Subsequent treatment
0.10
Chemotherapy 37 (60)
12 (63)
12 (71)
13 (50)
22 (35)
5 (26)
4 (23)
13 (50)
3 (5)
2 (11)
1 (6)
0
BSC Treating first-line chemotherapy ECOG PS Eastern Cooperative Oncology Group Performance Status, GPS Glasgow Prognostic Score, GEM gemcitabine, CDDP cisplatin, BSC best supportive care
Fig. 1 Kaplan–Meier curves of overall survival (OS). a According to ECOG PS, the median OS was 10.9 months for the patients with PS 0 and 7.6 months for the patients with PS 1 (P = 0.98). b According to GPS in patients with PS 0 or 1, the median OS was 17.0 months for GPS 0, 14.2 months for GPS 1, and 6.4 months for GPS 2 (GPS 0 vs. GPS 2, P = 0.002; GPS 1 versus GPS 2, P = 0.033; GPS 0 versus GPS 1, P = 0.32)
chemo-naı¨ve ABTC patients with good PS: 42 % of them were classified as GPS 2 and showed significantly shorter OS than those having GPS 0 or 1. This indicates that GPS should be evaluated in ABTC patients, with seemingly good condition and PS, who are eligible for chemotherapy. Our results also suggest that GPS was more important as a stratification factor than PS in clinical trials. Generally,
123
ECOG PS up to 2 has been adopted as eligibility criteria for randomized trials, and patients are stratified based on ECOG PS 0/1 or 2. However, patients with ECOG PS 2 were only up to 10 % of all enrolled patients and were often excluded by the eligibility criteria in recent randomized studies for solid tumors, including ABTC [2, 10– 15]. Therefore, stratification with GPS would help evaluate
Med Oncol (2014) 31:287
Page 5 of 6 287
Table 2 Univariate and multivariate analyses for overall survival Variable
Univariate analysis
Multivariate analysis
Median OS, month
P value 0.98
Hazard ratio (95 % CI)
P value
0.60 (0.40–0.90)
0.012
0.54 (0.25–1.14)
0.11
1.48 (0.75–2.92)
0.26
ECOG PS
0 vs. 1
10.9 vs. 7.6
GPS
0 vs. 1 vs. 2
17.0 vs. 14.2 vs. 6.4
0.003
Age, median Sex
Low vs. high Male vs. female
7.7 vs. 10.8 8.9 vs. 11.0
0.23 0.94
ALP, median
Low vs. high
13.8 vs. 7.3
0.11
Total Bilirubin, median
Low vs. high
10.8 vs. 7.6
0.45
LDH, median
Low vs. high
8.0 vs. 10.8
0.17
CEA, median
Low vs. high
11.0 vs. 10.0
0.51
CA19-9, median
Low vs. high
8.0 vs. 11.0
0.34
Primary tumor location
Extrahepatic vs. intrahepatic vs. gallbladder
11.0 vs. 8.0 vs. 10.0
0.95
Primary tumor resection
Yes vs. no
8.0 vs. 10.0
0.84
Number of metastatic organs
B1 vs. C2
10.8 vs. 8.9
0.14
1.40 (0.59–3.28)
0.45
Liver metastasis
No vs. yes
11.0 vs. 8.9
0.038
0.43 (0.20–0.90)
0.026
Lymph node metastasis Peritoneal metastasis
No vs. yes No vs. yes
10.0 vs. 10.0 10.9 vs. 6.8
0.25 0.025
0.49 (0.22–1.07)
0.07
Lung metastasis
No vs. yes
10.0 vs. 4.4
0.52
Other metastasis
No vs. yes
10.8 vs. 7.2
0.092
0.57 (0.20–1.61)
0.29
First-line chemotherapy regimen
GEM alone vs. S-1 alone vs. GEM ? CDDP vs. GEM ? S-1
8.0 vs. 10.0 vs. 10.9 vs. 16.7
0.60
ECOG PS Eastern Cooperative Oncology Group Performance Status, GPS Glasgow Prognostic Score, ALP alkaline phosphatase, LDH lactase dehydrogenase, CEA carcinoembryonic antigen, CA19-9 cancer antigen 19-9, GEM gemcitabine, CDDP cisplatin, OS overall survival, CI confidence interval
more exactly the results of randomized studies for enriched patients with good PS. This study has several limitations: its retrospective nature, small sample size, heterogeneous ABTC origin, and different first-line chemotherapies. In our study, most patients received GEM-containing regimens as first-line chemotherapy. The results for OS were comparable with those reported in prospective studies [2, 16–18]. Further studies are needed to confirm our results. In conclusion, GPS was a useful tool to detect populations with poor prognosis among chemo-naı¨ve ABTC patients with good PS. GPS may be more useful than ECOG PS as a stratification factor in future randomized studies. Acknowledgments This work was supported in part by funding from Tsukuba Cancer Clinical Trial Group (TCTG). Conflict of interest
The authors declare no conflict of interest.
2.
3.
4.
5.
6.
7.
References 8. 1. Miyakawa S, Ishihara S, Horiguchi A, et al. Biliary tract cancer treatment: 5,584 results from the Biliary Tract Cancer Statistics
Registry from 1998 to 2004 in Japan. J Hepatobiliary Pancreat Surg. 2009;16(1):1–7. Valle J, Wasan H, Palmer DH, et al. Cisplatin plus gemcitabine versus gemcitabine for biliary tract cancer. N Engl J Med. 2010;362(14):1273–81. Crumley AB, Stuart RC, McKernan M, et al. Comparison of an inflammation-based prognostic score (GPS) with performance status (ECOG-ps) in patients receiving palliative chemotherapy for gastroesophageal cancer. J Gastroenterol Hepatol. 2008;23(8 Pt 2):e325–9. Forrest LM, McMillan DC, McArdle CS, et al. 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):1028–30. Forrest LM, McMillan DC, McArdle CS, et al. A prospective longitudinal study of performance status, an inflammation-based score (GPS) and survival in patients with inoperable non-smallcell lung cancer. Br J Cancer. 2005;92(10):1834–6. Hwang JE, Kim HN, Kim DE, et al. Prognostic significance of a systemic inflammatory response in patients receiving first-line palliative chemotherapy for recurred or metastatic gastric cancer. BMC Cancer. 2011;11:489. Iwaku A, Kinoshita A, Onoda H, et al. The Glasgow Prognostic Score accurately predicts survival in patients with biliary tract cancer not indicated for surgical resection. Med Oncol. 2014;31(1):787. 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.
123
287 Page 6 of 6 9. Oshiro Y, Sasaki R, Fukunaga K, et al. Inflammation-based prognostic score is a useful predictor of postoperative outcome in patients with extrahepatic cholangiocarcinoma. J Hepatobiliary Pancreat Sci. 2013;20(3):389–95. 10. Falcone A, Ricci S, Brunetti I, et al. Phase III trial of infusional fluorouracil, leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) compared with infusional fluorouracil, leucovorin, and irinotecan (FOLFIRI) as first-line treatment for metastatic colorectal cancer: the Gruppo Oncologico Nord Ovest. J Clin Oncol. 2007;25(13):1670–6. 11. Buzdar AU, Suman VJ, Meric-Bernstam F, et al. Fluorouracil, epirubicin, and cyclophosphamide (FEC-75) followed by paclitaxel plus trastuzumab versus paclitaxel plus trastuzumab followed by FEC-75 plus trastuzumab as neoadjuvant treatment for patients with HER2-positive breast cancer (Z1041): a randomised, controlled, phase 3 trial. Lancet Oncol. 2013;14(13):1317–25. 12. Conroy T, Desseigne F, Ychou M, et al. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. N Engl J Med. 2011;364(19):1817–25. 13. Heinemann V, von Weikersthal LF, Decker T, et al. FOLFIRI plus cetuximab versus FOLFIRI plus bevacizumab as first-line treatment for patients with metastatic colorectal cancer (FIRE-3): a randomised, open-label, phase 3 trial. Lancet Oncol. 2014;. doi:10.1016/S1470-2045(14)70330-4.
123
Med Oncol (2014) 31:287 14. Ueno H, Ioka T, Ikeda M, et al. Randomized phase III study of gemcitabine plus S-1, S-1 alone, or gemcitabine alone in patients with locally advanced and metastatic pancreatic cancer in Japan and Taiwan: GEST study. J Clin Oncol. 2013;31(13):1640–8. 15. Wu YL, Zhou C, Hu CP, et al. Afatinib versus cisplatin plus gemcitabine for first-line treatment of Asian patients with advanced non-small-cell lung cancer harbouring EGFR mutations (LUX-Lung 6): an open-label, randomised phase 3 trial. Lancet Oncol. 2014;15(2):213–22. 16. Morizane C, Okusaka T, Mizusawa J, et al. Randomized phase II study of gemcitabine plus S-1 versus S-1 in advanced biliary tract cancer: a Japan Clinical Oncology Group trial (JCOG 0805). Cancer Sci. 2013;104(9):1211–6. 17. Okusaka T, Nakachi K, Fukutomi A, et al. Gemcitabine alone or in combination with cisplatin in patients with biliary tract cancer: a comparative multicentre study in Japan. Br J Cancer. 2010; 103(4):469–74. 18. Park JS, Oh SY, Kim SH, et al. Single-agent gemcitabine in the treatment of advanced biliary tract cancers: a phase II study. Jpn J Clin Oncol. 2005;35(2):68–73.
