Download Prognostic value of serum CA19-9 in patients with advanced

[Chinese Journal of Cancer 28:3, 240-244; March 2009]; ©2009 Landes Bioscience
Clinical Research Paper
Prognostic value of serum CA19-9 in patients with advanced
pancreatic cancer receiving gemcitabine based chemotherapy
Xin An, Yu-Hong Li,* Xu-Bin Lin, Feng-Hua Wang, Fen Feng, Rui-Hua Xu, Wen-Qi Jiang and You-Jian He
State Key Laboratory of Oncology in South China; Guangzhou, Guangdong P.R. China; and Department of Medical Oncology; Cancer Center; Sun Yat-sen University;
Guangzhou, Guangdong P.R. China
Key words: pancreatic cancer, chemotherapy, gemcitabine, serum CA19-9, treatment efficacy, prognostic factor
Background and Objective: Gemcitabine is used as an
ef­fective drug for patients with advanced pancreatic cancer.
Serum CA19-9 has been proven as the most sensitive and specific
serum marker for pancreatic cancer. This study was to investigate
the value of serum CA19-9 in evaluating treatment efficacy and
predicting prognosis of patients with pancreatic cancer treated
by gemcitabine-based chemotherapy. Methods: Seventy-one
patients with histologically confirmed, locally advanced or
metastatic pancreatic adenocarcinoma, whose Karnofsky’s performance status (KPS) score was ≥ 70 were treated with gemcitabine
alone or with gemcitabine-based chemotherapy. CA19-9 was
measured before and after chemotherapy. Results: Ten out of
71 patients had normal baseline CA19-9 levels, and 61 patients
had increased baseline CA19-9 levels. The overall survival of
patients were similar between the two groups, which were 9.0
months and 7.9 months, respectively (p = 0.797). The median
baseline CA19-9 level for patients who had increased CA19-9
was (682 ± 558) U/mL before treatment. Patients whose
pretreatment CA19-9 levels were < 682 U/mL achieved better
survival than those whose pretreatment CA19-9 levels were ≥
682 U/mL (9.6 months vs. 5.1 months, p = 0.001). In addition, patients with a pretreatment CA19-9 level of < 682 U/
mL had a better tumor response (43.5% vs. 15.8%, p = 0.051)
and clinical benefit response (48.1% vs. 29.2%, p = 0.125) than
those whose pretreatment CA19-9 level was < 682 U/mL, but
the differences were not significant. Patients with a fall of ≥25%
in the baseline CA19-9 level after chemotherapy achieved a
longer median overall survival (10.2 months vs. 5.0 months, p
*Correspondence to: Yu-Hong Li; State Key Laboratory of Oncology in South
China; and Department of Medical Oncology; Cancer Center; Sun Yat-sen
University; Guangzhou, Guangdong 510060 P.R. China; Tel.: 86.20.87343486;
Email: [email protected]
Submitted: 06/24/08; Revised: 10/24/08; Accepted: 11/05/08
This paper was translated into English from its original publication in Chinese.
Translated by: Xinglin Meditrans Center (http://www.58medtrans.com)
and Hua He on 02/20/09.
The original Chinese version of this paper is published in: Ai Zheng (Chinese
Journal of Cancer), 28(3); http://www.cjcsysu.cn/cn/article.asp?id=15113
Previously published online as a Chinese Journal of Cancer E-publication:
http://www.landesbioscience.com/journals/cjc/article/8650
240
< 0.001), better tumor response (47.8% vs. 10.5%, p = 0.002)
and better clinical benefit response (69.2% vs. 8.0%, p = 0.000)
than those without a decrease of baseline CA19-9 or with a
fall of < 25%. Multivariate analysis revealed that the baseline
CA19-9 level before chemotherapy, decreased percentage of the
CA19-9 level after chemotherapy, and the differentiation degree
of tumor cells were independent risk factors for patients whose
baseline CA19-9 levels were increased. Conclusion: The level of
pretreatment base-line CA19-9 and the decreased percentage of
baseline CA19-9 after chemotherapy are of predictive values for
survival of patients with advanced pancreatic cancer undergoing
gemcitabine-based chemotherapy and with an increased level of
baseline CA19-9.
Pancreatic cancer is one of the most common malignances in
China, with an increasing incidence rate in recent years. It often
has latent onset and is seldom accompanied with specific symptoms and body signs. Once diagnosed, less than 20% of these
patients have surgical opportunity. Even if radical surgical resection
is performed, the recurrence and metastatic rates remain as high
as 85%. For locally advanced, recurrent, and metastatic pancreatic cancer, gemcitabine monotherapy is the standard front-line
chemotherapy. However the median survival of patients receiving
this regimen is only 5 to 10 months.1,2 Serum carbohydrate
antigen 19-9 (CA19-9) is a tumor-associated antigen which has
been shown to be a highly specific and sensitive serum marker
for pancreatic cancer, thus it is recommended for the diagnosis,
staging, and postoperative surveillance of patients with pancreatic
cancer.3,4 A few studies showed that the level of serum CA19-9
could predict the survival for patient with advanced pancreatic
cancer undergoing systemic chemotherapy, although the results are
disparate. This study was to determine the value of serum CA19-9
in evaluating therapeutic efficacy and predicting prognosis in
pancreatic cancer treated with gemcitabine-based regimens.
Patients and Methods
Study subjects. A total of 71 patients with unresectable or metastatic pancreatic adenocarcinoma treated at Sun-Yet Sun University
Cancer Center from January 2000 to January 2008 were enrolled.
All patients met the following criteria: 1) Pathologically confirmed
pancreatic cancer; 2) No history of previous treatment and
Chinese Journal of Cancer
2009; Vol. 28 Issue 3
Prognostic value of serum CA19-9 in patients with advanced pancreatic cancer receiving gemcitabine based chemotherapy
receiving first-line gemicitabine based chemotherapy; 3) Karnofsky
performers status (KPS) score ≥ 70; 4) Dynamic measurement of
serum CA19-9 before and after chemotherapy; 5) No jaundice.
The median age of patients was 55 years (40–79 years). Fortyseven patients were males and 24 were females.
The primary tumor was located in pancreatic head in 21 cases,
in pancreatic body and tail in 49 cases, and in the whole pancreas
in one patient. Thirteen patients had locally advanced disease,
and 58 patients had evidence of metastatic disease. The most
common metastatic site was liver (55 cases), followed by posterior peritoneal lymph nodes and lung. Twelve cases were treated
with gemcitabine monotherapy, 35 cases with gemcitabine plus
oxaliplatin, 16 cases with gemcitabine plus 5-FU/CF in bolus,
and the rest eight cases with gemcitabine plus capecitabine or
docetaxel. The accumulated dosage for gemcitabine was 5.3 g/
m2 (2–16g/m2).
Measurement of serum CA19-9. Serum CA19-9 was measured
one week before starting chemotherapy (baseline level) and every
3–4 weeks after each cycle of chemotherapy. The concentration of
CA19-9 was measured by commercially available enzyme immunoassay (Electrochemilumnescenceimmunoassay, Elecsys 1010,
Roche Diagnostics GmbH, Mannheim, Germany). The normal
value of CA19-9 was < 35 U/mL, and the detectable range was
from 0.6 to 500,000 U/mL.
Evaluation standard for therapeutic effects. Overall survival
(OS). OS was defined as the time from beginning of the treatment
until the date of death.
Objective remission rate (ORR). The Response Evaluation
Criteria in Solid Tumor (RECIST) was used to determine the
therapeutic effect. Complete remission (CR) was defined as
complete disappearance of all target lesions without occurrence of
new lesions, a decrease in the level of tumor markers to the normal
level which was maintained for four weeks. Partial remission (PR)
was defined as a decrease in the sum of the longest lengths of
all baseline target lesions ≥ 30% which was maintained for four
weeks. Stable disease (SD) was defined as a decrease in the sum
of the longest lengths of all baseline target lesions less than PR, or
enlargement of the lesions without progression. Progressive disease
(PD) was defined as an increase in the sum of the longest lengths
of the detected smallest target lesion ≥ 20%, or occurrence of one
or more new lesion sites.
Clinical benefit response (CBR). CBR is a comprehensive
assessment for cancer related pain, body strength, and weight. The
standard is as the followings: 1) Improvement of at least one of the
three indices, a. a decrease in the dosage of the painkiller by more
than 50% compared with that given upon admission for more than
four weeks without worsening any of them, b. reduction of pain
severity for more than 50%, c. improvement of KPS score by more
than 20 points; 2) Stable assessment scores for painkiller, pain, and
body status, increase of more than 7% body weight for more than
four weeks (exclusive of fluid retention).
Statistical analysis. Data were analyzed by SPSS 12.0 software. Quantitative data between two groups were compared
using the R*C chi-square test. Comparison among multi-groups
was performed using One-Way ANOVA analysis. Survival was
www.landesbioscience.com
Table 1
Clinical characteristics of 71 patients with
pancreatic cancer
analyzed by the Kaplan-Meier test. The Cox model was used for
multi-factorial analysis to screen out independent risk factors. p <
0.05 indicated statistical significance.
Results
Overall therapeutic effects. Until April 2008, 64 patients died
and seven patients were still alive. The median OS was 7.1 months.
The one- and two-year survival rates were 13.8% and 1.7%,
respectively. Among 71 patients, 47 were evaluable, of which 14
patients achieved PR, 22 patients remained SD, and 11 patients
had PD. The overall response rate (ORR) was 29.8% (14/47).
Fifty-nine patients were able to be evaluated for clinical benefit and
the total CBR rate was 39.0% (23/59).
The relationship between serum CA19-9 and OS. Influence
of the base-line CA19-9 level on OS. The base-line serum CA19-9
level was measured in all 71 patients one week before starting
chemotherapy. Ten cases (14.1%) had normal baseline CA19-9
levels, while 61 cases (85.9%) had increased base-line CA19-9
levels. The median survival was 9.0 months for 10 patients with
normal baseline CA19-9 levels and 7.9 months for 61 patients
with increased baseline CA19-9 levels (p = 0.797) (Fig. 1A).
Sixty-one patients with elevated CA19-9 levels were further
divided into Group A (33 cases) whose baseline CA 19-9 levels
were below the medium value (682 U/mL) and Group B (28 cases)
whose baseline CA 19-9 concentrations were above the medium
value. The OS was longer for patients in Group A (9.6 months)
than in Group B (5.1 months) (p = 0.001) (Fig. 1B).
Chinese Journal of Cancer
241
Prognostic value of serum CA19-9 in patients with advanced pancreatic cancer receiving gemcitabine based chemotherapy
Figure 1. Correlation of scrum CA19-9 levels to the overall survival of patients with advanced pancreatic cancer
Table 2
Correlation of serum CA19-9 levels to tumor response, clinical benefit response (CBR)
PR: partial remission; SD: stable disease; PD: progression of disease.
Table 3
Cox regression analysis for prognostic
markers of pancreatic cancer
Influence of changes in postchemotherapy serum CA19-9
levels on OS. Postchemotherapy serum CA19-9 levels of all the
71 patients were measured every three to four weeks after each
cycle of chemotherapy. Serum CA19-9 levels of 10 patients with
normal baseline CA19-9 concentrations were maintained within
normal ranges after chemotherapy, while those of 61 patients
with increased baseline CA19-9 levels showed different extents of
changes after chemotherapy. The postchemotherapy decline rate
of CA19-9 was calculated as follows: (serum baseline CA19-9
value–the lowest postchemotherapy serum CA19-9 value)/serum
baseline CA19-9 value × 100%. As shown in Fig. 1C, when 25%
decline in the postchemotherapy serum CA19-9 was set as the
cut-off point, the median survival time of patients with a decline
in postchemotherapy CA 19-9 ≥ 25% (Group C, 32 cases) was
242
significantly longer than that of the patients with a decline in
postchemotherapy CA 19-9 < 25% or without decline (group D,
29 cases) (10.2 months vs. 5.0 months) (p < 0.001).
Correlation of CA19-9 to objected response rate (ORR) and
CBR. Among 61 patients with elevated serum CA19-9 levels, 42
were evaluable (23 from Group A and 19 from Group B; 23 from
Group C and 19 from Group D) for ORR and 51 (27 from Group
A and 24 from Group B; 26 from Group C and 25 from Group
D) were assessable for CBR. ORR were 43.5% (10/23) vs. 15.8%
(3/19) between Group A and B (p = 0.051), and 47.8% (11/23)
vs. 10.5% (2/19) between Group C and D (p = 0.002). CBR were
48.1% (13/27) vs. 29.2% (7/24) between Group A and B (p =
0.125), and 69.2% (18/26) vs. 8.0% (2/25) between Group C and
D (p < 0.001) (Table 2).
Multi-factorial analysis. Ten factors, including age, sex, KPS
score, primary tumor site, diseases extent (locally advanced or
metastatic), tumor differentiation, history of palliative therapy,
chemotherapy regimen, baseline serum CA19-9 level, decreased
percentage of the CA19-9 level after chemotherapy were analyzed
by the univariate and COX multivariate analysis. Results revealed
that baseline CA19-9 above the median value (682 U/mL),
decline in CA19-9 less than 25% and poor differentiation were
indepen­dent risk factors for advanced pancreatic cancers patients
treated with gemcitabine based chemotherapy (Table 3).
Discussion
Human pancreatic cancer is one of the most malignant diseases
in the world with highly poor prognosis. Gemcitabine is supe-
Chinese Journal of Cancer
2009; Vol. 28 Issue 3
Prognostic value of serum CA19-9 in patients with advanced pancreatic cancer receiving gemcitabine based chemotherapy
rior to 5-FU in improving CBR and prolonging the survival.
The National Comprehensive Cancer Network (NCCN) panel
recommends gemicitabine monotherapy as the standard frontline therapy for patients with advanced pancreatic cancer.1,2
Gemcitabine in combination with potentially synergistic agents,
such as cisplatin, oxaliplatin, capecitabine, irinotecan, or 5-FU,
exerts higher response rates, more favorable time to progression
and higher CBR than gemicitabine momotherapy, especially for
patients with good performance status whose ORR (CR + PR) is
around 10% to 30%, tumor control rate is around 25% to 74%,
and CBR is around 30% to 40%.5-8 In our study, most patients
(83.1%) received gemcitabine based doublets. In accordance
to previous studies, the objective effectiveness in our study was
29.8%; medium survival time was 7.1 months; one- and two-year
survival rates were 13.8% and 1.7%, respectively.
CA19-9 is a sialylated Lewis blood group antigen targeted by
the monoclonal antibody 1116 NS19-9 with the molecular weight
of 5 × 103 ku. The concentration of CA 19-9 is increased in more
than 80% of patients with advanced pancreatic cancer. However,
patients who are genotypically negative for the Lewis blood group
antigen, will not express CA19-9 even in the presence of active
pancreatic cancer. About 7–10% of the general population is negative for this antigen and, thus, will not express CA 19-9.10 In this
study, 10 of 71 patients (14.1%) had normal ranges of CA19-9
levels, whose overall survival was not significantly different from
those with elevated CA19-9 levels, consistent with that reported
by Halm et al.4 We believe that serum CA19-9 could not be
used as a prognostic factor for those patient with normal baseline
CA19-9 levels.
Maisey et al.11 conducted a retrospective study and found that
the base-line CA19-9 level prior to chemotherapy was an independent predictive factor for survival of patients with advanced
pancreatic cancer. Patients whose baseline CA19-9 was lowered
than the average value achieved a higher one-year survival rate than
those whose baseline CA19-9 was above the average. Ko et al.12
proved a positive correlation between the percentage of decline in
postchemotherapy CA19-9 and the overall survival of patient with
advanced pancreatic cancer. Halm et al.13 reported the survival of
43 consecutive patients with advanced pancreatic cancer treated
with single-agent gemcitabine and revealed that a fall of at least
20% in the CA19-9 level following the start of chemotherapy
was the only independent prognostic marker for OS. Park et al.14
conducted a phase II study and found that patients with ≥ 25%
decline in serum CA19-9 achieved better outcomes in terms of
time to progression and OS than those whose decline in CA19-9
was < 25%. However, a recent prospective study published in
Lancet Oncol concluded that only pretreatment base-line serum
CA 19-9 concentration was an independent prognostic factor for
survival, whereas a decrease in concentration during chemotherapy
was not significantly associated with lengthened survival compared
with those who did not have a corresponding decrease.15 We
found that the median OS in patients with a baseline CA19-9
below the median value was significantly superior to those with
a level above the median value. Moreover, the median OS was
significantly longer in patients with a decrease in CA 19-9 ≥ 25%
www.landesbioscience.com
than those with a decrease < 25%. Therefore, we suppose that both
pretreatment baseline CA 19-9 levels and post-treatment decline in
CA19-9 levels could predict the survival of patients with advanced
pancreatic cancer undergoing gemcitabine based chemotherapy.
However, due to the relative small number of sample size, further
prospective, large sample size studies are needed to verify the
results. We also discovered that, among patients with a decline
in post-treatment CA19-9 level ≥ 25%, most of which (88.9%)
achieved the lowest CA19-9 value within six weeks after chemotherapy (median 5.3 weeks). We suggest that measurement of CA
19-9 after six weeks of chemotherapy may assist to decide the treatment efficacy together with other clinical parameters.
RECIST is based on measuring the size changes in tumor
lesions before and after chemotherapy using imaging techniques.
Because pancreas is located behind the peritoneum, imaging
examinations have limitations in differentiating normal pancreas,
local inflammation, and fibrosis from malignant tissues, and
therefore may be inaccurate in the assessment of tumor response
during chemotherapy, even using 3D imaging techniques, such as
CT and MRI. Carmichael J et al.16 proposed using CBR as indices
to assess the treatment efficacy of pancreatic cancer. However,
these parameters are too subjective, and are difficult to be quantified, which limit their clinical application. Our study indicated
that dynamic measurements of CA 19-9 levels in gemcitabinebased chemotherapy are useful to assess the therapeutic effect.
The ORR and CBR were significantly higher in patients whose
decline in CA 19-9 was ≥ 25% than in those whose decline was
< 25%. Although the ORR and CBR were higher in patients
whose baseline CA 19-9 levels were below the median (682 U/
mL) than in those whose baseline CA 19-9 levels were above the
median, the differences were not significant, probably due to the
relatively small sample size of this study. Similar results have been
reported by other groups. Micke et al. found that patients with a
low level of baseline pretreatment CA19-9 (420U/mL) showed a
higher response rate than those whose baseline CA19-9 was above
420 U/mL.17
An important caveat in the interpretation of CA19-9 levels
in pancreatic cancer is the presence of biliary obstruction, which
results in the elevation of the marker level. In the current series,
patients with jaundice or hyperbilirubinemia were excluded.
Therefore this potential pitfall was avoided in the interpretation of
baseline CA19-9 levels.
In conclusion, dynamic measurement of CA19-9 concentrations is a useful prognostic tool in predicting the survival of
advanced pancreatic cancer patients with elevated serum CA19-9
levels receiving gemicitabine based chemotherapy
References
[1] Aapro MS, Martin C, Hatty S. Gemcitabine--a safety review [J]. Anticancer Drugs,
1998,9 (3):191-201.
[2] Burris HA 3rd, Moore MJ, Andersen J, et al. Improvements in survival and clinical
benefit with gemcitabine as first-line therapy for patients with advanced pancreas cancer:
a randomized trial [J]. J Clin Oncol, 1997,15 (6):2403-2413.
[3] Safi F, Roscher R, Beger HG. Tumor markers in pancreatic cancer: Sensitivity and specificity of CA 19-9 [J]. Hepatogastroenterology, 1989,36 (6):419-423.
[4] Halm U, Rohde N, Klapdor R, et al. Improved sensitivity of fuzzy logic based tumor
marker profiles for diagnosis of pancreatic carcinoma versus benign pancreatic disease [J].
Anticancer Res, 2000,20 (6D):4957-4960.
Chinese Journal of Cancer
243
Prognostic value of serum CA19-9 in patients with advanced pancreatic cancer receiving gemcitabine based chemotherapy
[5] Ueno H, Okusaka T, Ikeda M, et al. Phase II study of combination chemotherapy with
gemcitabine and cisplatin for patients with metastatic pancreatic cancer [J]. Jpn J Clin
Oncol, 2007,37:515-520.
[6] Baize N, Abu Shalaa A, Berthier F, et al. Gemcitabine combined with oxaliplatin is safe
and effective in patients with previously untreated advanced pancreatic adenocarcinoma
[J]. Gastroenterol Clin Biol, 2005,29 (10):1006-1009.
[7] Herrmann R, Bodoky G, Ruhstaller T, et al. Gemcitabine plus capecitabine compared
with gemcitabine alone in advanced pancreatic cancer: a randomized, multicenter, phase
III trial of the Swiss Group for Clinical Cancer Research and the Central European
Cooperative Oncology Group [J]. J Clin Oncol, 2007,25 (16):2212-2217.
[8] Endlicher E, Troppmann M, Kullmann A, et al. Irinotecan plus gemcitabine and
5-fluorouracil in advanced pancreatic cancer: a phase II study [J]. Oncology, 2007,72
(5-6):279-284.
[9] oeck S, Stieber P, Holdenrieder S, et al. Prognostic and therapeutic significance of carbohydrate antigen 19-9 as tumor marker in patients with pancreatic cancer [J]. Oncology,
2006,70 (4):255-264.
[10] chi Y, Okabe H, Inui T, et al. [Tumor marker--present and future [J]. Rinsho Byori,
1997,45 (9):875-883.
[11] aisey NR, Norman AR, Hill A, et al. CA19-9 as a prognostic factor in inoperable pancreatic cancer: the implication for clinical trials [J]. Br J Cancer, 2005,93 (7):740-743.
[12] o AH, Hwang J, Venook AP, et al. Serum CA19-9 response as a surrogate for clinical
outcome in patients receiving fixed-dose rate gemcitabine for advanced pancreatic cancer
[J]. Br J Cancer, 2005,93 (2):195-199.
[13] alm U, Schumann T, Schiefke I, et al. Decrease of CA 19-9 during chemotherapy with
gemcitabine predicts survival time in patients with advanced pancreatic cancer [J]. Br J
Cancer, 2000,82(5):1013-1016.
[14] Park BB, Park JO, Lee HR, et al. A phase II trial of gemcitabine plus capecitabine for
patients with advanced pancreatic adenocarcinoma [J]. Cancer Chemother Pharmacol,
2007,60 (4):489-494.
[15] Hess V, Glimelius B, Grawe P, et al. CA 19-9 tumour-marker response to chemotherapy
in patients with advanced pancreatic cancer enrolled in a randomised controlled trial [J].
Lancet Oncol, 2008,9 (2):132-138.
[16] Carmichael J. Clinical response benefit in patients with advanced pancreatic cancer [J].
Role of gemcitabine. Digestion, 1997,58(6):503-507.
[17] Micke O, Bruns F, Kurowski R, et al. Predictive value of carbohydrate antigen 19-9 in
pancreatic cancer treated with radiochemotherapy [J]. Int J Radiat Oncol Biol Phys,
2003,57(1):90-97.
244
Chinese Journal of Cancer
2009; Vol. 28 Issue 3