- Email: [email protected]
Postoperative Chemoradiotherapy for Extrahepatic Bile Duct Cancer
Int. J. Radiation Oncology Biol. Phys., Vol. 79, No. 3, pp. 696–704, 2011 Copyright Ó 2011 Elsevier Inc. Printed in the USA. All rights reserved 0360-3016/$–see front matter
doi:10.1016/j.ijrobp.2009.12.031
CLINICAL INVESTIGATION
Bile Ducts
POSTOPERATIVE CHEMORADIOTHERAPY FOR EXTRAHEPATIC BILE DUCT CANCER JIN-HONG PARK, M.D.,* EUN KYUNG CHOI, M.D., PH.D.,* SEUNG DO AHN, M.D., PH.D.,* SANG-WOOK LEE, M.D., PH.D.,* SI YEOL SONG, M.D., PH.D.,* SANG MIN YOON, M.D., PH.D.,* YOUNG SEOK KIM, M.D.,* YU SUN LEE, M.D.,* SUNG-GYU LEE, M.D., PH.D.,y SHIN HWANG, M.D., PH.D.,y YOUNG-JOO LEE, M.D., PH.D.,y KWANG-MIN PARK, M.D., PH.D.,y TAE WON KIM, M.D., PH.D.,z HEUNG MOON CHANG, M.D., PH.D.,z JAE-LYUN LEE, M.D., PH.D.,z AND JONG HOON KIM, M.D., PH.D.* Departments of *Radiation Oncology, yGeneral Surgery, and zInternal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea Purpose: To evaluate the effect of postoperative concurrent chemoradiotherapy using three-dimensional conformal radiotherapy and to identify the prognostic factors that influence survival in patients with extrahepatic bile duct cancer. Methods and Materials: We retrospectively analyzed the data from 101 patients with extrahepatic bile duct cancer who had undergone postoperative concurrent chemoradiotherapy using three-dimensional conformal radiotherapy. Of the 101 patients, 52 (51%) had undergone complete resection (R0 resection) and 49 (49%) had microscopic or macroscopic residual tumors (R1 or R2 resection). The median radiation dose was 50 Gy. Also, 85 patients (84%) underwent concurrent chemotherapy with 5-fluorouracil. Results: The median follow-up period was 47 months for the surviving patients. The 5-year overall survival rate was 34% for all patients. A comparison between patients with R0 and R1 resection indicated no significant difference in the 5-year overall survival (44% vs. 33%, p = .2779), progression-free survival (35% vs. 22%, p = .3107), or locoregional progression-free survival (75% vs. 63%, p = .2784) rates. An analysis of the first failure site in the 89 patients with R0 or R1 resection indicated isolated locoregional recurrence in 7 patients. Elevated postoperative carbohydrate antigen 19-9 level was an independent prognostic factor for overall survival (p = .001) and progression-free survival (p = .033). A total of 3 patients developed Grade 3 or greater late toxicity. Conclusion: Adjuvant concurrent chemoradiotherapy using three-dimensional conformal radiotherapy appears to improve locoregional control and survival in extrahepatic bile duct cancer patients with R1 resection. The postoperative carbohydrate antigen 19-9 level might be a useful prognostic marker to select patients for more intensified adjuvant therapy. Ó 2011 Elsevier Inc. Extrahepatic bile duct cancer, postoperative chemoradiotherapy, carbohydrate antigen 19-9.
radiotherapy (RT). A recent report, using data from the Surveillance, Epidemiology, and End Results database, suggested that postoperative RT provides a survival benefit in biliary tract cancer patients (6, 7). However, the benefits of adjuvant RT in EHBDC is still a matter of debate (8–16). Because of the relatively low incidence of EHBDC, most reports concerning adjuvant RT have been retrospective analyses of data that included heterogeneous treatments over long periods. Very few reports have described the use of modern RT or concurrent chemoradiotherapy (CCRT) techniques as adjuvant therapy for EHBDC (10, 12–14). Thus, the aim of the present study was to evaluate the results of adjuvant CCRT using three-dimensional conformal RT
INTRODUCTION Cholangiocarcinoma is a relatively uncommon malignancy that arises from the epithelial cells of the biliary tract and has a tendency for extension along the biliary tract (1). Surgical resection of extrahepatic bile duct cancer (EHBDC) is the only treatment that allows extended survival, but complete resection is often limited because of locally advanced stage and its close proximity to adjacent critical structures, such as major vessels of the liver (2, 3). Even after curative resection, locoregional relapse is common and a significant cause of death (4, 5). To improve the poor results of surgical resection, numerous attempts have been made to find a role for adjuvant
Conflict of interest: none. Received Sept 13, 2009, and in revised form Dec 1, 2009. Accepted for publication Dec 3, 2009.
Reprint requests to: Jong Hoon Kim, M.D., Ph.D., Department of Radiation Oncology, Asan Medical Center, University of Ulsan College of Medicine, 388-1, PoongNap-Dong, SongPa-Gu, Seoul, Korea. Tel: (+82) 2-3010-4434; Fax: (+82) 2-486-7258; E-mail: [email protected] 696
Postoperative CCRT for extrahepatic bile duct cancer d J. PARK et al.
(3D-CRT) and to identify the prognostic factors that influence survival in EHBDC patients treated at a single institution. METHODS AND MATERIALS Eligibility criteria The eligibility criteria for the present study were as follows: histologically proven EHBDC, no distant metastasis, no history of malignancy other than skin cancer, Eastern Cooperative Oncology Group performance status of #2, radiation dose of >30 Gy. Patients who had undergone only a simple biliary bypass procedure without resection of gross tumor were excluded from the present study.
697
Chemotherapy Concurrent chemotherapy was administered to 85 (84%) of the 101 patients. Of the 85 patients, 7 received two cycles of 5-fluorouracil (375 mg/m2/d) and leucovorin (20 mg/m2/d) by rapid intravenous injection for 3 days per cycle from Days 1 to 28 of RT. Of the 85 patients, 78 (92%) received oral uracil/tegafur or capecitabine. Of these 78 patients, 42 were prescribed oral uracil/tegafur (300 mg/m2/ d) and leucovorin (90 mg/d) in three divided doses, and 36 received capecitabine (1,650 mg/m2/d) in two divided doses. Oral chemotherapy was administered throughout the entire RT course, without weekend breaks. Adjuvant chemotherapy based on 5-fluorouracil was given to 29 patients who showed postoperative residual disease or abnormal postoperative serum CA 19-9 levels.
Statistical analysis Patient evaluation The preoperative evaluation included measurement of serum bilirubin and carbohydrate antigen 19-9 (CA 19-9) levels, chest radiography, computed tomography of the abdomen, magnetic resonance imaging of the abdomen, and endoscopic retrograde cholangiography or percutaneous transhepatic cholangiography. All patients underwent disease staging according to the sixth edition of the American Joint Committee on Cancer System. We classified EHBDC patients into two groups using the system proposed by Nakeeb et al. (2). Proximal EHBDC was defined as tumor that involved the proximal bile duct, including the confluent portion of the bilateral hepatic ducts or the common hepatic duct. Distal EHBDC was defined as tumor that involved the common bile duct. A diffuse tumor that involved both the proximal and distal regions was classified as proximal EHBDC. The extent of residual disease after resection was classified as R0 (no residual tumor), R1 (microscopic residual tumor), or R2 (macroscopic residual tumor). A postoperative computed tomography scan was performed before RT, and 1–3 months after surgical resection, the CA 19-9 level was remeasured. Treatment-related toxicity was graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. After RT completion, computed tomography of the abdomen was performed every 3 months for the first 2 years and every 6 months thereafter.
Postoperative RT Postoperative RT was generally recommended for patients with incomplete resection (R1 or R2 resection) or R0 resection with nodal metastases and/or invasion to a neighboring organ or major vessel. The RT was typically started 4–6 weeks after resection. All patients were treated with 3D-CRT. The clinical target volume included the primary tumor bed with a 2–3-cm margin and the regional lymphatics. The planning target volume included the clinical target volume and a nonuniform 1–2-cm margin, considering setup uncertainty and breathing motion. External beam RT was delivered with multiple fields using megavoltage photon beams (6 or 15 MV) at 1.8 or 2.0 Gy daily for 5 days each week. The dose for the initial field was 50 Gy, followed by a boost dose in patients with R1 or R2 resection. An additional boost dose of 9–10 Gy, using three-to-six coplanar or noncoplanar beams, was delivered to the tumor bed and/or gross residual disease, with a 1–2-cm margin, in patients with R1 or R2 resection. The median total dose for all patients was 50 Gy (range, 32–60), and 92 patients (91%) received a dose of $50 Gy. The median dose for patients with R1 or R2 resection was 60 Gy (range, 35–60).
All events were measured from the date of surgery to the date of occurrence or the last follow-up visit. Locoregional recurrence was defined as any recurrence in the primary tumor bed and regional lymphatic areas. Distant metastasis was defined as any recurrence in a systemic organ, the peritoneum, or a distant lymph node. The chi-square test was used to compare failure rates. The Kaplan-Meier method was used to estimate overall survival (OS), progression-free survival (PFS), and locoregional progression-free survival (LRPFS). Survival curves were compared using the log– rank test. A Cox proportional hazards model was used to identify prognostic factors on multivariate analysis.
RESULTS Patient selection and characteristics A total of 236 patients with EHBDC underwent RT between January 1998 and October 2007 at the Asan Medical Center (Seoul, Korea). Of these, 126 patients were treated definitively without surgical resection or with only a palliative bypass operation. Of the 110 patients who received postoperative RT, 9 were ineligible (6 had received a radiation dose of <30 Gy, 2 had recurrent cancer, and 1 had simultaneous metastatic disease). Thus, 101 patients were enrolled for the present retrospective study. The patient characteristics are summarized in Tables 1 and 2. The median age was 58 years (range, 28–77), and the male/ female ratio was 1.9. The postoperative CA 19-9 level was measured in 99 patients (98%), and 26 patients had consistently elevated values. The procedure for resection was determined by the primary tumor location (Table 1). R0 resection was achieved in 52 patients (51%), R1 resection in 37 (37%), and R2 resection in 12 (12%; Table 2). Adenocarcinoma was confirmed in all tumor specimens. Survival and prognostic factors The median follow-up time was 47 months (range, 20–84) for the surviving patients. Of the 101 patients, 35 were alive at the time of our retrospective analysis. The median OS and PFS of the entire group was 24 and 14 months, respectively. The 5-year OS and PFS rate was 34% and 27%, respectively (Fig. 1). For patients with R0 and R1 resection, the 5-year OS and PFS rate was 39% and 30%, respectively. Univariate analysis of the prognostic factors for OS (Table 3) showed that the postoperative CA 19-9 level
698
I. J. Radiation Oncology d Biology d Physics
Volume 79, Number 3, 2011
Table 1. Patients and treatment characteristics (n = 101)
Age (y) Range Median Gender Male Female Tumor location Proximal Distal Preoperative bilirubin (mg/dL) <10 $10 Unavailable Preoperative CA 19-9 (U/mL) <37 $37 Unavailable Postoperative CA 19-9 (U/mL) <37 $37 Unavailable Surgical procedure Proximal tumor Liver lobectomy with bile duct resection Bile duct resection Distal tumor PPPD Whipple procedure Bile duct resection Radiation dose (Gy) R0 resection <50 $50 R1 or R2 resection <50 $50 Concurrent chemotherapy Yes No
Value
n (%)
Characteristic
31 (31) 15 (15) 20 (20)
Residual disease R0 resection R1 resection R2 resection Histologic grade Well differentiated Moderately differentiated Poorly differentiated Unspecified Perineural invasion Negative Positive Lymphovascular invasion Negative Positive T stage T1 T2 T3 T4 N stage N0 N1 Unspecified TNM stage I II III Unspecified
5 (5) 47 (47)
Abbreviations: EHBDC = extrahepatic bile duct cancer; R0 resection = no residual tumor; R1 resection = microscopic residual tumor; R2 resection = macroscopic residual tumor.
28–77 58 66 (65) 35 (35) 35 (35) 65 (65) 64 (63) 24 (24) 13 (13) 36 (36) 59 (58) 6 (6) 73 (72) 26 (26) 2 (2) 28 (28) 7 (7)
4 (4) 45 (44) 85 (84) 16 (16)
Abbreviations: CA = carbohydrate antigen; PPPD = pylorus-preserving pancreaticoduodenectomy. Data presented as numbers, with percentages in parentheses, unless otherwise noted.
(p <.0001), extent of residual disease (p = .0135), and histologic tumor grade (p = .0261) were significantly associated with a poor prognosis. A significant difference was found in the survival rates of the R0/R1 and R2 resection groups; however, the difference was not significant between the R0 and R1 groups (44%, 33%, and 0% for R0, R1, and R2 resection, respectively; R0 vs. R1, p = .2779; R0 vs. R2, p = .0038; R1 vs. R2, p = .0427; Fig. 2). Multivariate analysis indicated that the postoperative CA 19-9 level (relative risk, 2.870; 95% confidence interval, 1.512–5.445; p = .001) was the only significant independent prognostic factor for OS (Fig. 3 and Table 4). Of the 101 patients, 68 (68%) had experienced disease progression at the time of our analysis. Univariate analysis (Table 3) indicated that the postoperative CA 19-9 level (p <.0001), extent of residual disease (p = .0408), presence
52 (51) 37 (37) 12 (12) 24 (24) 56 (55) 14 (14) 7 (7) 32 (32) 69 (68) 61 (60) 40 (40) 18 (18) 25 (25) 45 (44) 13 (13) 52 (52) 46 (45) 3 (3) 25 (25) 61 (60) 12 (12) 3 (3)
of lymphovascular invasion (p = .0459), histologic tumor grade (p = .0015), and N stage (p = .0279) were significant factors for PFS (Table 3). Multivariate analysis showed that the postoperative CA 19-9 level (relative risk, 2.066; 95% confidence interval, 1.059–4.032; p = .033) and histologic tumor grade (relative risk, 2.922; 95% confidence interval , 1.422–6.004; p = .004) were independent prognostic factors (Fig. 3 and Table 4). 100
Cumulative Survival Rate (%)
Characteristic
Table 2. Histopathologic characteristics of EHBDC
Locoregional control Overall survival
80
Progression-free survival
60
40
20
0
0
1
2
3
4
5
6
7
8
Time (years)
Fig. 1. Locoregional control, progression-free survival, and overall survival rates for all 101 patients who underwent postoperative radiotherapy.
Postoperative CCRT for extrahepatic bile duct cancer d J. PARK et al.
699
Table 3. Univariate analysis of overall survival, progression-free survival, and locoregional progression-free survival 5-y Survival rate (%) Prognostic factor Age (y) <60 $60 Gender Male Female Preoperative bilirubin (mg/dL) <10 $10 CA 19-9 (U/mL) Preoperative <37 $37 Postoperative <37 $37 Primary tumor location Proximal Distal Residual disease extent R0 resection R1 resection R2 resection Perineural invasion Negative Positive Lymphovascular invasion Negative Positive Histologic grade Grade 1-2 Grade 3 T stage T1-T2 T3-T4 N stage N0 N1 Chemotherapy Yes No
Patients (n)
OS
p
PFS
p
LRPFS
p
57 44
35 33
.3322
26 30
.9749
60 66
.7537
66 35
32 37
.9251
27 27
.6892
56 75
.0875
64 24
37 18
.4299
33 22
.4158
69 42
.1251
36 59
39 28
.1184
32 23
.1990
66 57
.4470
73 26
43 8
<.0001
34 6
<.0001
69 30
.0006
35 66
23 37
.9479
30 27
.7798
54 67
.4814
52 37 12
44 33 0
.0135
35 22 0
.0408
75 63 0
<.0001
32 69
47 27
.1157
43 19
.0616
82 53
.0136
61 40
41 23
.0521
33 19
.0459
66 55
.3245
80 14
36 18
.0261
34 0
.0015
60 65
.8285
43 58
27 37
.8963
27 27
.7064
54 69
.0969
52 46
39 28
.2023
39 13
.0279
63 60
.9375
84 17
36 25
.3963
29 12
.9483
66 37
.2359
Abbreviations: OS = overall survival; PFS = progression-free survival; LRPFS = locoregional progression-free survival; CA = carbohydrate antigen; other abbreviations as in Table 2.
Locoregional progression occurred in 32 patients (32%) during the follow-up period, and 29 (91%) of these 32 patients experienced locoregional failure within 18 months of surgical resection. The estimated LRPFS rate was 62% at 5 years (Fig. 1) and was 75% for R0 resection, 63% for R1 resection, and 0% for R2 resection (R0 vs. R1, p = .2784; R0 vs. R2, p <.0001; R1 vs. R2, p = .0010; Fig. 2). Patients with perineural invasion (p = .0136) and an elevated postoperative CA 19-9 level (p = .0006) had lower 5-year LRPFS (Table 3). R2 resection (relative risk, 4.031; 95% confidence interval, 4.447–11.232; p = .008) was the only significant prognostic factor for locoregional recurrence on multivariate analysis (Table 4). For the 26 patients with elevated CA 19-9 levels after resection, we analyzed the survival data as a function of change
in the post-RT CA 19-9 level by serial examination. After RT, the CA 19-9 level returned to normal in 10 patients but remained elevated in 16. The 10 patients whose CA 19-9 level had returned to normal had significantly better 3-year OS, PFS, and LRPFS rates compared with the 16 other patients (40% vs. 0%, p = .0013; 17% vs. 0%, p = .0034; and 53% vs. 0%, p = 0.0136, respectively; Fig. 4). Patterns of failure An analysis of the first site of recurrence in the 89 patients with R0 or R1 resection showed that treatment failure occurred in 58 (65%) 2–42 months after resection. Locoregional failure occurred in 7 patients (12%) and distant metastasis in 40 patients (69%). Synchronous locoregional and distant failure occurred in 11 patients (19%). The
I. J. Radiation Oncology d Biology d Physics
700
Volume 79, Number 3, 2011
(b) 100
(a)100
p = 0.0408
80
60
R0 resection
40
R1 resection R2 resection
20
0
0
1
2
3
4
5
80
Cumulative Survival Rate (%)
Cumulative Survival Rate (%)
p < 0.0001
6
7
R0 resection R1 resection
60
R2 resection 40
20
0
8
0
1
2
3
4
5
6
7
8
Time (years)
Time (years)
(c) 100 Cumulative Survival Rate (%)
p = 0.0135 80
R0 resection R1 resection
60
R2 resection
40
20
0
0
1
2
3
4
5
6
7
8
Time (years)
Fig. 2. Comparison of survival curves according to residual disease extent. (a) Locoregional progression-free survival. (b) Progression-free survival. (c) Overall survival. R0 resection = no residual disease; R1 resection = microscopic residual disease; R2 resection = macroscopic residual disease.
location of locoregional recurrence was the anastomosis in 4, the hepatic hilum or hepatoduodenal ligament in 8, the retroperitoneal lymph node in 6, and the pancreatic head in 2. The most common sites of distant failure were the peritoneum (n = 22) and liver (n = 18). An analysis of the proportion of locoregional failure and risk factors indicated that locoregional failure was more prevalent in patients with positive perineural invasion (39% vs. 7%, p = .032) and was unaffected by other factors.
Toxicity In most cases, postoperative RT was well tolerated, and the side effects were manageable. No treatment-related mortality or hospitalization occurred. No patient experienced weight loss greater than Grade 2 during the RT period. Two patients (2%), who had received concurrent chemotherapy, experienced Grade 3 or greater hematologic toxicity, but no neutropenia-related infection developed. Two patients (2%) had Grade 3 nausea and vomiting, which subsided with supportive management. Three patients developed late complications of Grade 3 or greater. Duodenal ulcers and bleeding occurred in 2 patients at 7 and 19 months after RT; these patients had received a radiation dose of 60 Gy and 50 Gy, respectively. Both patients recovered after endoscopic coagulation and transcatheter arterial embolization. Benign enterobiliary anastomotic
stricture developed in 1 patient 10 months after RT and was controlled by temporary percutaneous biliary drainage. No patient experienced radiation-induced liver disease. The treatment-related toxicities are summarized in Table 5. DISCUSSION Cholangiocarcinoma is a relatively rare disease and has been regarded as an oncologic challenge because of the poor prognosis. Although long-term survival in EHBDC patients could be possible after complete resection, the surgical candidates are limited and many patients experience disease recurrence (2, 3). Several studies have reported that postoperative RT improves the survival rate; however, controversy exists about the use of adjuvant RT for EHBDC (8–16). Nevertheless, the high rate of locoregional relapse after surgery has provided the rationale for the use of adjuvant RT (4, 5, 14). Jarnagin et al. (5) reported that the isolated locoregional recurrence rate was 59% for a group of patients in whom the R0 resection rate was 76%. Borghero et al. (14) reported a 37% 5-year locoregional recurrence rate for patients with good prognostic features (R0 resection and pN0 disease). In present study, the local recurrence rate was 20% for the R0 and R1 resection groups. The 5-year LRPFS rate was 62% for all patients combined and 75% for patients with R0 resection. Moreover, the analysis of the first site of recurrence among patients with R0 or R1 resection indicated
Postoperative CCRT for extrahepatic bile duct cancer d J. PARK et al.
(a) 100
701
(b) 100
60
CA 19-9 < 37 U/ml CA 19-9
40
80
Cumulative Survival Rate (%)
Cumulative Survival Rate (%)
p < 0.0001 p = 0.0006
80
37 U/ml
20
CA 19-9 < 37 U/ml CA 19-9
60
37 U/ml
40
20
0
0 0
1
2
3
4
5
6
7
8
0
1
2
3
4
Time (years)
5
6
7
8
Time (years)
(C)100 Cumulative Survival Rate (%)
p < 0.0001 80
CA 19-9 < 37 U/ml CA 19-9
60
37 U/ml
40
20
0 0
1
2
3
4
5
6
7
8
Time (years)
Fig. 3. Comparison of survival curves according to postoperative level of carbohydrate antigen (CA) 19-9, evaluated within 3 months of resection. (a) Locoregional progression-free survival. (b) Progression-free survival. (c) Overall survival.
In the present study, the 5-year OS rate for the entire group was 34%, and the 5-year OS rate for patients with R0 and R1 resection was 39%. These results are similar to those of several recent studies, which reported 5-year OS rates of 33– 36% (10, 11, 13, 14), and greater than those reported in our previous study, in which we found a 24.0% 5-year survival rate after surgery alone (excluding operative mortality) for treatment of proximal EHBDC (17). Also, several earlier studies that used a historical surgery-only control group demonstrated a survival benefit for adjuvant RT (9, 11, 14–16). However, it is difficult to compare the present results with those from surgery-only reports because of the retrospective nature of all studies, differences in patient populations, and the greater risk faced by patients who underwent RT.
that solitary locoregional failure was responsible for only 12% of recurrent disease. Because the patients who were referred for postoperative RT could have had more adverse features, which were not evident from the pathologic reports, our results suggest that adjuvant RT might be effective in decreasing locoregional recurrence. However, with a locoregional control rate of 75% for R0 resection, there is still room for improvement with locoregional therapy, and the poor locoregional control for patients with R2 resection (0%) indicates the need for treatment intensification. Because the predominant pattern of failure is distant failure, better systemic therapy will improve survival in this population and ultimately change the pattern of failure and increase the significance of locoregional control.
Table 4. Multivariate analysis of prognostic factors for overall survival 5-y OS
5-y PFS
5-y LRPFS
Prognostic factor
RR
p
RR
p
RR
p
Postoperative CA 19-9 ($37 U/mL) Histologic grade (Grade 3) Residual disease R1 resection R2 resection Lymphovascular invasion (positive) Perineural invasion (positive) N stage (N1)
2.870 2.003
.001 .054
2.066 2.922
.033 .004
2.033 —
.116 —
1.486 1.285 1.439 — —
.207 .560 .189 — —
1.728 2.155 1.569 1.082 1.479
.106 .115 .150 .818 .143
1.290 4.031 — 2.442 —
.556 .008 — .075 —
Abbreviations: RR = relative risk; other abbreviations as in Tables 2 and 3.
I. J. Radiation Oncology d Biology d Physics
702
Volume 79, Number 3, 2011
(b) 100 CA 19-9 < 37 U/ml
80
CA 19-9
Cumulative Survival Rate (%)
Cumulative Survival Rate (%)
(a) 100 37 U/ml
60
40
p = 0.0136
20
0 0
1
2
3
4
5
6
7
CA 19-9 < 37 U/ml
80
CA 19-9
37 U/ml
60
40
p = 0.0034
20
0
8
0
1
2
3
Time (years)
4
5
6
7
8
Time (years)
Cumulative Survival Rate (%)
(c) 100 CA 19-9 < 37 U/ml
80
CA 19-9
37 U/ml
60
40
p = 0.0013
20
0 0
1
2
3
4
5
6
7
8
Time (years)
Fig. 4. Comparison of survival curves according to postradiotherapy carbohydrate antigen (CA) 19-9 level for patients with elevated CA 19-9 levels after resection. (a) Locoregional progression-free survival. (b) Progression-free survival. (c) Overall survival.
However, it seems probable that adjuvant RT might have contributed to the improved survival rate observed in the present study, considering the greater R0 resection rate (78% vs. 51%) in the historical surgery-only group from our institution (17). It is also controversial whether adjuvant RT can compensate for suboptimal resection in EHBDC. Several investigators have demonstrated a benefit of RT after R1 resection (9, 10, 14, 15, 18), but these results have been contradicted by other reports (2, 8, 12). Ben-David et al. (12) reported that R0 resection was the only significant predictor of OS and PFS and found no difference in OS or PFS between patients with R1 and R2 resection. The investigators concluded that surgery provided little benefit if R0 resection could not be achieved. However, these results could have been biased Table 5. Treatment-related toxicities Patients (n) Toxicity Acute Weight loss Hematologic Gastrointestinal Late Gastrointestinal
Grade 2 or less Grade 3 Grade 4 or greater 101 (100) 99 (98) 99 (98)
0 1 (1) 2 (2)
0 1 (1) 0
98 (97)
3 (3)
0
Data in parentheses are percentages.
by the small sample size and the enrollment of patients with gall bladder cancer (50% of R0 and R1 resections), which has clinical characteristics different from those of EHBDC (5). Also, two other related studies (2, 8) have been criticized because of the small sample size, poor prognostic features, and suboptimal treatment given to the adjuvant RT group (14, 19). In the present study, we found no significant differences in the 5-year OS, PFS, and LRPFS between the R0 and R1 resection groups. Our multivariate analysis found that R1 resection was not a significant prognostic factor for OS, PFS, or LRPFS. Although even R2 resection was not an independent prognostic factor for OS and PFS, R2 resection became significant when we excluded the effect of the postoperative CA 19-9 level in the multivariate analysis (data not shown). Our data strongly suggest that postoperative RT could improve locoregional tumor control and OS in patients with R1 resection, even though RT might not fully compensate for suboptimal resection. Carbohydrate antigen 19-9 is a carbohydrate antigen originally identified in the culture medium of a colorectal cancer cell line and is currently one of the most widely used tumor markers for cholangiocarcinoma (20, 21). A recent prospective Phase III trial reported that the postresection CA 19-9 level was a significant predictor of OS in patients with pancreatic cancer (22); however, little is known about the effectiveness of CA 19-9 as a prognostic indicator for bile duct cancer. In the present study, our
Postoperative CCRT for extrahepatic bile duct cancer d J. PARK et al.
703
Table 6. Recent studies of adjuvant chemoradiotherapy for extrahepatic bile duct cancer Investigator
Year
Patients (n)
R0/R1/R2 resection (n)
Stage N1 (n)
Median radiation dose (Gy)
CCRT (%)
5-y OS (%)
5-y DFS (%)
5-y LC (%)
Present study* Borghero et al. (14) Nelson et al. (13) Hughes et al. (11) Kim et al. (10)
2009 2008 2008 2007 2002
81 42 45 34 84
52 (58)/37 (42)/— 15 (36)/ 27 (64)/— 36 (80)/6 (13)/ 3 (7) 25 (74)/9 (26)/— 47 (56)/25 (30)/12 (14)
43 (48) 19 (45) 11 (25) 28 (82) 27 (32)
50 45 50.4 50.4 40
83 100 98 100 85
39 36 33 35 31
30 34 37 30 26
70 62 78 70 —
Abbreviations: CCRT = concurrent chemoradiotherapy; OS = overall survival; DFS = disease-free survival; LC = local control; other abbreviations as in Table 2. * Patients with R2 resection were excluded from Table 6. Data in parentheses are percentages
multivariate analysis indicated that the postoperative CA 19-9 level was an independent predictor of OS and PFS. Moreover, patients whose CA 19-9 level was elevated after resection, but in whom the concentration had returned to normal after RT, had greater OS, PFS, and LRPFS rates than did the patients whose CA 19-9 level remained elevated after RT. An elevated postoperative CA 19-9 level might indicate the burden of residual disease after resection as a component of locoregional or systemic disease. It is also possible that patients in whom the CA 19-9 level returned to normal after RT had radiosensitive tumors that were limited to the locoregional area. Thus, we suggest that the CA 19-9 level might be an important prognostic indicator after surgery or postoperative RT in patients with EHBDC. Also, we suggest that more aggressive treatment should be considered for patients in whom the CA 19-9 level remains elevated after postoperative RT. Concurrent chemoradiotherapy has proved effective against other gastrointestinal cancers (23), but only a few reports on the use of adjuvant CCRT for EHBDC have been published (Table 6). Previous studies have reported 5-year OS rates of 33–36%, with median survival of 16–37 months (10, 11, 13, 14, 24). In the present report, patients who underwent CCRT tended to have better 5-year OS, PFS, and LRPFS than the RT-only group. However, we could not conclude that CCRT was more effective than RT alone, because our study was retrospective and the size of the RT-only group was very small (16 patients). Regardless, our data showed that adjuvant CCRT that included oral 5-fluorouracil was
well tolerated. Considering the radiosensitizing effect of chemotherapy and the low toxicity among our patients, we suggest that CCRT with 3D-CRT be considered as a reasonable adjuvant therapy for EHBDC. The present study was nonrandomized and retrospective and, thus, might have had unrecognized biases, just as all other previous reports of EHBDC. To the best of our knowledge, however, our study is one of the largest patient series exploring the effect of postoperative RT for EHBDC and also addressed locoregional control and patterns of failure. Moreover, this is the first report to indicate that the postoperative CA 19-9 level is an independent prognostic factor for EHBDC recurrence. Because all patients underwent 3DCRT and most patients underwent CCRT, our study provides a basis for implementation of modern adjuvant chemoradiotherapy for EHBDC. CONCLUSION Our results suggest that adjuvant CCRT with 3D-CRT after surgical resection of EHBDC might effectively control locoregional disease and improve patient survival, with low toxicity levels, especially for patients with R1 resection. Distant metastasis was the most common pattern of recurrence; thus, more effective systemic therapy after adjuvant RT is clearly needed. The CA 19-9 level might be an important prognostic factor after resection and after postoperative RT for the identification of patients who might need more aggressive therapy.
REFERENCES 1. Longmire WP, McArthur MS, Bastounis EA, et al. Carcinoma of the extrahepatic biliary tract. Ann Surg 1973;178:333–345. 2. Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma: A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg 1996;224:463–473. 465. 3. Reding R, Buard JL, Lebeau G, et al. Surgical management of 552 carcinomas of the extrahepatic bile ducts (gallbladder and periampullary tumors excluded): Results of the French Surgical Association Survey. Ann Surg 1991;213:236–241. 4. Kopelson G, Galdabini J, Warshaw AL, et al. Patterns of failure after curative surgery for extra-hepatic biliary tract carcinoma: Implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 1981;7:413–417. 5. Jarnagin WR, Ruo L, Little SA, et al. Patterns of initial disease recurrence after resection of gallbladder carcinoma and hilar
6. 7.
8. 9.
cholangiocarcinoma: Implications for adjuvant therapeutic strategies. Cancer 2003;98:1689–1700. Wang SJ, Fuller CD, Kim JS, et al. Prediction model for estimating the survival benefit of adjuvant radiotherapy for gallbladder cancer. J Clin Oncol 2008;26:2112–2117. Shinohara ET, Mitra N, Guo M, et al. Radiotherapy is associated with improved survival in adjuvant and palliative treatment of extrahepatic cholangiocarcinomas. Int J Radiat Oncol Biol Phys 2009;74:1191–1198. Pitt HA, Nakeeb A, Abrams RA, et al. Perihilar cholangiocarcinoma: Postoperative radiotherapy does not improve survival. Ann Surg 1995;221:788–797. Todoroki T, Ohara K, Kawamoto T, et al. Benefits of adjuvant radiotherapy after radical resection of locally advanced main hepatic duct carcinoma. Int J Radiat Oncol Biol Phys 2000;46:581–587.
704
I. J. Radiation Oncology d Biology d Physics
10. Kim S, Kim SW, Bang YJ, et al. Role of postoperative radiotherapy in the management of extrahepatic bile duct cancer. Int J Radiat Oncol Biol Phys 2002;54:414–419. 11. Hughes MA, Frassica DA, Yeo CJ, et al. Adjuvant concurrent chemoradiation for adenocarcinoma of the distal common bile duct. Int J Radiat Oncol Biol Phys 2007;68:178–182. 12. Ben-David MA, Griffith KA, Abu-Isa E, et al. External-beam radiotherapy for localized extrahepatic cholangiocarcinoma. Int J Radiat Oncol Biol Phys 2006;66:772–779. 13. Nelson JW, Ghafoori AP, Willett CG, et al. Concurrent chemoradiotherapy in resected extrahepatic cholangiocarcinoma. Int J Radiat Oncol Biol Phys 2008;72:1495–1501. 14. Borghero Y, Crane CH, Szklaruk J, et al. Extrahepatic bile duct adenocarcinoma: Patients at high-risk for local recurrence treated with surgery and adjuvant chemoradiation have an equivalent overall survival to patients with standard-risk treated with surgery alone. Ann Surg Oncol 2008;15:3147– 3156. 15. Cheng Q, Luo X, Zhang B, et al. Predictive factors for prognosis of hilar cholangiocarcinoma: Postresection radiotherapy improves survival. Eur J Surg Oncol 2007;33:202–207. 16. Gerhards MF, van Gulik TM, Gonzalez Gonzalez D, et al. Results of postoperative radiotherapy for resectable hilar cholangiocarcinoma. World J Surg 2003;27:173–179.
Volume 79, Number 3, 2011
17. Lee SG, Lee YJ, Park KM, et al. One hundred and eleven liver resections for hilar bile duct cancer. J Hepatobiliary Pancreat Surg 2000;7:135–141. 18. Schoenthaler R, Phillips TL, Castro J, et al. Carcinoma of the extrahepatic bile ducts: The University of California at San Francisco experience. Ann Surg 1994;219:267–274. 19. Heron DE, Stein DE, Eschelman DJ, et al. Cholangiocarcinoma: The impact of tumor location and treatment strategy on outcome. Am J Clin Oncol 2003;26:422–428. 20. Nehls O, Gregor M, Klump B. Serum and bile markers for cholangiocarcinoma. Semin Liver Dis 2004;24:139–154. 21. Goonetilleke KS, Siriwardena AK. Systematic review of carbohydrate antigen (CA 19-9) as a biochemical marker in the diagnosis of pancreatic cancer. Eur J Surg Oncol 2007;33:266–270. 22. Berger AC, Garcia M Jr., Hoffman JP, et al. Postresection CA 19-9 predicts overall survival in patients with pancreatic cancer treated with adjuvant chemoradiation: A prospective validation by RTOG 9704. J Clin Oncol 2008;26:5918–5922. 23. National Institutes of Health. NIH consensus conference: Adjuvant therapy for patients with colon and rectal cancer. JAMA 1990;264:1444–1450. 24. Minsky BD, Wesson MF, Armstrong JG, et al. Combined modality therapy of extrahepatic biliary system cancer. Int J Radiat Oncol Biol Phys 1990;18:1157–1163.
doi:10.1016/j.ijrobp.2009.12.031
CLINICAL INVESTIGATION
Bile Ducts
POSTOPERATIVE CHEMORADIOTHERAPY FOR EXTRAHEPATIC BILE DUCT CANCER JIN-HONG PARK, M.D.,* EUN KYUNG CHOI, M.D., PH.D.,* SEUNG DO AHN, M.D., PH.D.,* SANG-WOOK LEE, M.D., PH.D.,* SI YEOL SONG, M.D., PH.D.,* SANG MIN YOON, M.D., PH.D.,* YOUNG SEOK KIM, M.D.,* YU SUN LEE, M.D.,* SUNG-GYU LEE, M.D., PH.D.,y SHIN HWANG, M.D., PH.D.,y YOUNG-JOO LEE, M.D., PH.D.,y KWANG-MIN PARK, M.D., PH.D.,y TAE WON KIM, M.D., PH.D.,z HEUNG MOON CHANG, M.D., PH.D.,z JAE-LYUN LEE, M.D., PH.D.,z AND JONG HOON KIM, M.D., PH.D.* Departments of *Radiation Oncology, yGeneral Surgery, and zInternal Medicine, Asan Medical Center, University of Ulsan College of Medicine, Seoul, Korea Purpose: To evaluate the effect of postoperative concurrent chemoradiotherapy using three-dimensional conformal radiotherapy and to identify the prognostic factors that influence survival in patients with extrahepatic bile duct cancer. Methods and Materials: We retrospectively analyzed the data from 101 patients with extrahepatic bile duct cancer who had undergone postoperative concurrent chemoradiotherapy using three-dimensional conformal radiotherapy. Of the 101 patients, 52 (51%) had undergone complete resection (R0 resection) and 49 (49%) had microscopic or macroscopic residual tumors (R1 or R2 resection). The median radiation dose was 50 Gy. Also, 85 patients (84%) underwent concurrent chemotherapy with 5-fluorouracil. Results: The median follow-up period was 47 months for the surviving patients. The 5-year overall survival rate was 34% for all patients. A comparison between patients with R0 and R1 resection indicated no significant difference in the 5-year overall survival (44% vs. 33%, p = .2779), progression-free survival (35% vs. 22%, p = .3107), or locoregional progression-free survival (75% vs. 63%, p = .2784) rates. An analysis of the first failure site in the 89 patients with R0 or R1 resection indicated isolated locoregional recurrence in 7 patients. Elevated postoperative carbohydrate antigen 19-9 level was an independent prognostic factor for overall survival (p = .001) and progression-free survival (p = .033). A total of 3 patients developed Grade 3 or greater late toxicity. Conclusion: Adjuvant concurrent chemoradiotherapy using three-dimensional conformal radiotherapy appears to improve locoregional control and survival in extrahepatic bile duct cancer patients with R1 resection. The postoperative carbohydrate antigen 19-9 level might be a useful prognostic marker to select patients for more intensified adjuvant therapy. Ó 2011 Elsevier Inc. Extrahepatic bile duct cancer, postoperative chemoradiotherapy, carbohydrate antigen 19-9.
radiotherapy (RT). A recent report, using data from the Surveillance, Epidemiology, and End Results database, suggested that postoperative RT provides a survival benefit in biliary tract cancer patients (6, 7). However, the benefits of adjuvant RT in EHBDC is still a matter of debate (8–16). Because of the relatively low incidence of EHBDC, most reports concerning adjuvant RT have been retrospective analyses of data that included heterogeneous treatments over long periods. Very few reports have described the use of modern RT or concurrent chemoradiotherapy (CCRT) techniques as adjuvant therapy for EHBDC (10, 12–14). Thus, the aim of the present study was to evaluate the results of adjuvant CCRT using three-dimensional conformal RT
INTRODUCTION Cholangiocarcinoma is a relatively uncommon malignancy that arises from the epithelial cells of the biliary tract and has a tendency for extension along the biliary tract (1). Surgical resection of extrahepatic bile duct cancer (EHBDC) is the only treatment that allows extended survival, but complete resection is often limited because of locally advanced stage and its close proximity to adjacent critical structures, such as major vessels of the liver (2, 3). Even after curative resection, locoregional relapse is common and a significant cause of death (4, 5). To improve the poor results of surgical resection, numerous attempts have been made to find a role for adjuvant
Conflict of interest: none. Received Sept 13, 2009, and in revised form Dec 1, 2009. Accepted for publication Dec 3, 2009.
Reprint requests to: Jong Hoon Kim, M.D., Ph.D., Department of Radiation Oncology, Asan Medical Center, University of Ulsan College of Medicine, 388-1, PoongNap-Dong, SongPa-Gu, Seoul, Korea. Tel: (+82) 2-3010-4434; Fax: (+82) 2-486-7258; E-mail: [email protected] 696
Postoperative CCRT for extrahepatic bile duct cancer d J. PARK et al.
(3D-CRT) and to identify the prognostic factors that influence survival in EHBDC patients treated at a single institution. METHODS AND MATERIALS Eligibility criteria The eligibility criteria for the present study were as follows: histologically proven EHBDC, no distant metastasis, no history of malignancy other than skin cancer, Eastern Cooperative Oncology Group performance status of #2, radiation dose of >30 Gy. Patients who had undergone only a simple biliary bypass procedure without resection of gross tumor were excluded from the present study.
697
Chemotherapy Concurrent chemotherapy was administered to 85 (84%) of the 101 patients. Of the 85 patients, 7 received two cycles of 5-fluorouracil (375 mg/m2/d) and leucovorin (20 mg/m2/d) by rapid intravenous injection for 3 days per cycle from Days 1 to 28 of RT. Of the 85 patients, 78 (92%) received oral uracil/tegafur or capecitabine. Of these 78 patients, 42 were prescribed oral uracil/tegafur (300 mg/m2/ d) and leucovorin (90 mg/d) in three divided doses, and 36 received capecitabine (1,650 mg/m2/d) in two divided doses. Oral chemotherapy was administered throughout the entire RT course, without weekend breaks. Adjuvant chemotherapy based on 5-fluorouracil was given to 29 patients who showed postoperative residual disease or abnormal postoperative serum CA 19-9 levels.
Statistical analysis Patient evaluation The preoperative evaluation included measurement of serum bilirubin and carbohydrate antigen 19-9 (CA 19-9) levels, chest radiography, computed tomography of the abdomen, magnetic resonance imaging of the abdomen, and endoscopic retrograde cholangiography or percutaneous transhepatic cholangiography. All patients underwent disease staging according to the sixth edition of the American Joint Committee on Cancer System. We classified EHBDC patients into two groups using the system proposed by Nakeeb et al. (2). Proximal EHBDC was defined as tumor that involved the proximal bile duct, including the confluent portion of the bilateral hepatic ducts or the common hepatic duct. Distal EHBDC was defined as tumor that involved the common bile duct. A diffuse tumor that involved both the proximal and distal regions was classified as proximal EHBDC. The extent of residual disease after resection was classified as R0 (no residual tumor), R1 (microscopic residual tumor), or R2 (macroscopic residual tumor). A postoperative computed tomography scan was performed before RT, and 1–3 months after surgical resection, the CA 19-9 level was remeasured. Treatment-related toxicity was graded according to the National Cancer Institute Common Terminology Criteria for Adverse Events, version 3.0. After RT completion, computed tomography of the abdomen was performed every 3 months for the first 2 years and every 6 months thereafter.
Postoperative RT Postoperative RT was generally recommended for patients with incomplete resection (R1 or R2 resection) or R0 resection with nodal metastases and/or invasion to a neighboring organ or major vessel. The RT was typically started 4–6 weeks after resection. All patients were treated with 3D-CRT. The clinical target volume included the primary tumor bed with a 2–3-cm margin and the regional lymphatics. The planning target volume included the clinical target volume and a nonuniform 1–2-cm margin, considering setup uncertainty and breathing motion. External beam RT was delivered with multiple fields using megavoltage photon beams (6 or 15 MV) at 1.8 or 2.0 Gy daily for 5 days each week. The dose for the initial field was 50 Gy, followed by a boost dose in patients with R1 or R2 resection. An additional boost dose of 9–10 Gy, using three-to-six coplanar or noncoplanar beams, was delivered to the tumor bed and/or gross residual disease, with a 1–2-cm margin, in patients with R1 or R2 resection. The median total dose for all patients was 50 Gy (range, 32–60), and 92 patients (91%) received a dose of $50 Gy. The median dose for patients with R1 or R2 resection was 60 Gy (range, 35–60).
All events were measured from the date of surgery to the date of occurrence or the last follow-up visit. Locoregional recurrence was defined as any recurrence in the primary tumor bed and regional lymphatic areas. Distant metastasis was defined as any recurrence in a systemic organ, the peritoneum, or a distant lymph node. The chi-square test was used to compare failure rates. The Kaplan-Meier method was used to estimate overall survival (OS), progression-free survival (PFS), and locoregional progression-free survival (LRPFS). Survival curves were compared using the log– rank test. A Cox proportional hazards model was used to identify prognostic factors on multivariate analysis.
RESULTS Patient selection and characteristics A total of 236 patients with EHBDC underwent RT between January 1998 and October 2007 at the Asan Medical Center (Seoul, Korea). Of these, 126 patients were treated definitively without surgical resection or with only a palliative bypass operation. Of the 110 patients who received postoperative RT, 9 were ineligible (6 had received a radiation dose of <30 Gy, 2 had recurrent cancer, and 1 had simultaneous metastatic disease). Thus, 101 patients were enrolled for the present retrospective study. The patient characteristics are summarized in Tables 1 and 2. The median age was 58 years (range, 28–77), and the male/ female ratio was 1.9. The postoperative CA 19-9 level was measured in 99 patients (98%), and 26 patients had consistently elevated values. The procedure for resection was determined by the primary tumor location (Table 1). R0 resection was achieved in 52 patients (51%), R1 resection in 37 (37%), and R2 resection in 12 (12%; Table 2). Adenocarcinoma was confirmed in all tumor specimens. Survival and prognostic factors The median follow-up time was 47 months (range, 20–84) for the surviving patients. Of the 101 patients, 35 were alive at the time of our retrospective analysis. The median OS and PFS of the entire group was 24 and 14 months, respectively. The 5-year OS and PFS rate was 34% and 27%, respectively (Fig. 1). For patients with R0 and R1 resection, the 5-year OS and PFS rate was 39% and 30%, respectively. Univariate analysis of the prognostic factors for OS (Table 3) showed that the postoperative CA 19-9 level
698
I. J. Radiation Oncology d Biology d Physics
Volume 79, Number 3, 2011
Table 1. Patients and treatment characteristics (n = 101)
Age (y) Range Median Gender Male Female Tumor location Proximal Distal Preoperative bilirubin (mg/dL) <10 $10 Unavailable Preoperative CA 19-9 (U/mL) <37 $37 Unavailable Postoperative CA 19-9 (U/mL) <37 $37 Unavailable Surgical procedure Proximal tumor Liver lobectomy with bile duct resection Bile duct resection Distal tumor PPPD Whipple procedure Bile duct resection Radiation dose (Gy) R0 resection <50 $50 R1 or R2 resection <50 $50 Concurrent chemotherapy Yes No
Value
n (%)
Characteristic
31 (31) 15 (15) 20 (20)
Residual disease R0 resection R1 resection R2 resection Histologic grade Well differentiated Moderately differentiated Poorly differentiated Unspecified Perineural invasion Negative Positive Lymphovascular invasion Negative Positive T stage T1 T2 T3 T4 N stage N0 N1 Unspecified TNM stage I II III Unspecified
5 (5) 47 (47)
Abbreviations: EHBDC = extrahepatic bile duct cancer; R0 resection = no residual tumor; R1 resection = microscopic residual tumor; R2 resection = macroscopic residual tumor.
28–77 58 66 (65) 35 (35) 35 (35) 65 (65) 64 (63) 24 (24) 13 (13) 36 (36) 59 (58) 6 (6) 73 (72) 26 (26) 2 (2) 28 (28) 7 (7)
4 (4) 45 (44) 85 (84) 16 (16)
Abbreviations: CA = carbohydrate antigen; PPPD = pylorus-preserving pancreaticoduodenectomy. Data presented as numbers, with percentages in parentheses, unless otherwise noted.
(p <.0001), extent of residual disease (p = .0135), and histologic tumor grade (p = .0261) were significantly associated with a poor prognosis. A significant difference was found in the survival rates of the R0/R1 and R2 resection groups; however, the difference was not significant between the R0 and R1 groups (44%, 33%, and 0% for R0, R1, and R2 resection, respectively; R0 vs. R1, p = .2779; R0 vs. R2, p = .0038; R1 vs. R2, p = .0427; Fig. 2). Multivariate analysis indicated that the postoperative CA 19-9 level (relative risk, 2.870; 95% confidence interval, 1.512–5.445; p = .001) was the only significant independent prognostic factor for OS (Fig. 3 and Table 4). Of the 101 patients, 68 (68%) had experienced disease progression at the time of our analysis. Univariate analysis (Table 3) indicated that the postoperative CA 19-9 level (p <.0001), extent of residual disease (p = .0408), presence
52 (51) 37 (37) 12 (12) 24 (24) 56 (55) 14 (14) 7 (7) 32 (32) 69 (68) 61 (60) 40 (40) 18 (18) 25 (25) 45 (44) 13 (13) 52 (52) 46 (45) 3 (3) 25 (25) 61 (60) 12 (12) 3 (3)
of lymphovascular invasion (p = .0459), histologic tumor grade (p = .0015), and N stage (p = .0279) were significant factors for PFS (Table 3). Multivariate analysis showed that the postoperative CA 19-9 level (relative risk, 2.066; 95% confidence interval, 1.059–4.032; p = .033) and histologic tumor grade (relative risk, 2.922; 95% confidence interval , 1.422–6.004; p = .004) were independent prognostic factors (Fig. 3 and Table 4). 100
Cumulative Survival Rate (%)
Characteristic
Table 2. Histopathologic characteristics of EHBDC
Locoregional control Overall survival
80
Progression-free survival
60
40
20
0
0
1
2
3
4
5
6
7
8
Time (years)
Fig. 1. Locoregional control, progression-free survival, and overall survival rates for all 101 patients who underwent postoperative radiotherapy.
Postoperative CCRT for extrahepatic bile duct cancer d J. PARK et al.
699
Table 3. Univariate analysis of overall survival, progression-free survival, and locoregional progression-free survival 5-y Survival rate (%) Prognostic factor Age (y) <60 $60 Gender Male Female Preoperative bilirubin (mg/dL) <10 $10 CA 19-9 (U/mL) Preoperative <37 $37 Postoperative <37 $37 Primary tumor location Proximal Distal Residual disease extent R0 resection R1 resection R2 resection Perineural invasion Negative Positive Lymphovascular invasion Negative Positive Histologic grade Grade 1-2 Grade 3 T stage T1-T2 T3-T4 N stage N0 N1 Chemotherapy Yes No
Patients (n)
OS
p
PFS
p
LRPFS
p
57 44
35 33
.3322
26 30
.9749
60 66
.7537
66 35
32 37
.9251
27 27
.6892
56 75
.0875
64 24
37 18
.4299
33 22
.4158
69 42
.1251
36 59
39 28
.1184
32 23
.1990
66 57
.4470
73 26
43 8
<.0001
34 6
<.0001
69 30
.0006
35 66
23 37
.9479
30 27
.7798
54 67
.4814
52 37 12
44 33 0
.0135
35 22 0
.0408
75 63 0
<.0001
32 69
47 27
.1157
43 19
.0616
82 53
.0136
61 40
41 23
.0521
33 19
.0459
66 55
.3245
80 14
36 18
.0261
34 0
.0015
60 65
.8285
43 58
27 37
.8963
27 27
.7064
54 69
.0969
52 46
39 28
.2023
39 13
.0279
63 60
.9375
84 17
36 25
.3963
29 12
.9483
66 37
.2359
Abbreviations: OS = overall survival; PFS = progression-free survival; LRPFS = locoregional progression-free survival; CA = carbohydrate antigen; other abbreviations as in Table 2.
Locoregional progression occurred in 32 patients (32%) during the follow-up period, and 29 (91%) of these 32 patients experienced locoregional failure within 18 months of surgical resection. The estimated LRPFS rate was 62% at 5 years (Fig. 1) and was 75% for R0 resection, 63% for R1 resection, and 0% for R2 resection (R0 vs. R1, p = .2784; R0 vs. R2, p <.0001; R1 vs. R2, p = .0010; Fig. 2). Patients with perineural invasion (p = .0136) and an elevated postoperative CA 19-9 level (p = .0006) had lower 5-year LRPFS (Table 3). R2 resection (relative risk, 4.031; 95% confidence interval, 4.447–11.232; p = .008) was the only significant prognostic factor for locoregional recurrence on multivariate analysis (Table 4). For the 26 patients with elevated CA 19-9 levels after resection, we analyzed the survival data as a function of change
in the post-RT CA 19-9 level by serial examination. After RT, the CA 19-9 level returned to normal in 10 patients but remained elevated in 16. The 10 patients whose CA 19-9 level had returned to normal had significantly better 3-year OS, PFS, and LRPFS rates compared with the 16 other patients (40% vs. 0%, p = .0013; 17% vs. 0%, p = .0034; and 53% vs. 0%, p = 0.0136, respectively; Fig. 4). Patterns of failure An analysis of the first site of recurrence in the 89 patients with R0 or R1 resection showed that treatment failure occurred in 58 (65%) 2–42 months after resection. Locoregional failure occurred in 7 patients (12%) and distant metastasis in 40 patients (69%). Synchronous locoregional and distant failure occurred in 11 patients (19%). The
I. J. Radiation Oncology d Biology d Physics
700
Volume 79, Number 3, 2011
(b) 100
(a)100
p = 0.0408
80
60
R0 resection
40
R1 resection R2 resection
20
0
0
1
2
3
4
5
80
Cumulative Survival Rate (%)
Cumulative Survival Rate (%)
p < 0.0001
6
7
R0 resection R1 resection
60
R2 resection 40
20
0
8
0
1
2
3
4
5
6
7
8
Time (years)
Time (years)
(c) 100 Cumulative Survival Rate (%)
p = 0.0135 80
R0 resection R1 resection
60
R2 resection
40
20
0
0
1
2
3
4
5
6
7
8
Time (years)
Fig. 2. Comparison of survival curves according to residual disease extent. (a) Locoregional progression-free survival. (b) Progression-free survival. (c) Overall survival. R0 resection = no residual disease; R1 resection = microscopic residual disease; R2 resection = macroscopic residual disease.
location of locoregional recurrence was the anastomosis in 4, the hepatic hilum or hepatoduodenal ligament in 8, the retroperitoneal lymph node in 6, and the pancreatic head in 2. The most common sites of distant failure were the peritoneum (n = 22) and liver (n = 18). An analysis of the proportion of locoregional failure and risk factors indicated that locoregional failure was more prevalent in patients with positive perineural invasion (39% vs. 7%, p = .032) and was unaffected by other factors.
Toxicity In most cases, postoperative RT was well tolerated, and the side effects were manageable. No treatment-related mortality or hospitalization occurred. No patient experienced weight loss greater than Grade 2 during the RT period. Two patients (2%), who had received concurrent chemotherapy, experienced Grade 3 or greater hematologic toxicity, but no neutropenia-related infection developed. Two patients (2%) had Grade 3 nausea and vomiting, which subsided with supportive management. Three patients developed late complications of Grade 3 or greater. Duodenal ulcers and bleeding occurred in 2 patients at 7 and 19 months after RT; these patients had received a radiation dose of 60 Gy and 50 Gy, respectively. Both patients recovered after endoscopic coagulation and transcatheter arterial embolization. Benign enterobiliary anastomotic
stricture developed in 1 patient 10 months after RT and was controlled by temporary percutaneous biliary drainage. No patient experienced radiation-induced liver disease. The treatment-related toxicities are summarized in Table 5. DISCUSSION Cholangiocarcinoma is a relatively rare disease and has been regarded as an oncologic challenge because of the poor prognosis. Although long-term survival in EHBDC patients could be possible after complete resection, the surgical candidates are limited and many patients experience disease recurrence (2, 3). Several studies have reported that postoperative RT improves the survival rate; however, controversy exists about the use of adjuvant RT for EHBDC (8–16). Nevertheless, the high rate of locoregional relapse after surgery has provided the rationale for the use of adjuvant RT (4, 5, 14). Jarnagin et al. (5) reported that the isolated locoregional recurrence rate was 59% for a group of patients in whom the R0 resection rate was 76%. Borghero et al. (14) reported a 37% 5-year locoregional recurrence rate for patients with good prognostic features (R0 resection and pN0 disease). In present study, the local recurrence rate was 20% for the R0 and R1 resection groups. The 5-year LRPFS rate was 62% for all patients combined and 75% for patients with R0 resection. Moreover, the analysis of the first site of recurrence among patients with R0 or R1 resection indicated
Postoperative CCRT for extrahepatic bile duct cancer d J. PARK et al.
(a) 100
701
(b) 100
60
CA 19-9 < 37 U/ml CA 19-9
40
80
Cumulative Survival Rate (%)
Cumulative Survival Rate (%)
p < 0.0001 p = 0.0006
80
37 U/ml
20
CA 19-9 < 37 U/ml CA 19-9
60
37 U/ml
40
20
0
0 0
1
2
3
4
5
6
7
8
0
1
2
3
4
Time (years)
5
6
7
8
Time (years)
(C)100 Cumulative Survival Rate (%)
p < 0.0001 80
CA 19-9 < 37 U/ml CA 19-9
60
37 U/ml
40
20
0 0
1
2
3
4
5
6
7
8
Time (years)
Fig. 3. Comparison of survival curves according to postoperative level of carbohydrate antigen (CA) 19-9, evaluated within 3 months of resection. (a) Locoregional progression-free survival. (b) Progression-free survival. (c) Overall survival.
In the present study, the 5-year OS rate for the entire group was 34%, and the 5-year OS rate for patients with R0 and R1 resection was 39%. These results are similar to those of several recent studies, which reported 5-year OS rates of 33– 36% (10, 11, 13, 14), and greater than those reported in our previous study, in which we found a 24.0% 5-year survival rate after surgery alone (excluding operative mortality) for treatment of proximal EHBDC (17). Also, several earlier studies that used a historical surgery-only control group demonstrated a survival benefit for adjuvant RT (9, 11, 14–16). However, it is difficult to compare the present results with those from surgery-only reports because of the retrospective nature of all studies, differences in patient populations, and the greater risk faced by patients who underwent RT.
that solitary locoregional failure was responsible for only 12% of recurrent disease. Because the patients who were referred for postoperative RT could have had more adverse features, which were not evident from the pathologic reports, our results suggest that adjuvant RT might be effective in decreasing locoregional recurrence. However, with a locoregional control rate of 75% for R0 resection, there is still room for improvement with locoregional therapy, and the poor locoregional control for patients with R2 resection (0%) indicates the need for treatment intensification. Because the predominant pattern of failure is distant failure, better systemic therapy will improve survival in this population and ultimately change the pattern of failure and increase the significance of locoregional control.
Table 4. Multivariate analysis of prognostic factors for overall survival 5-y OS
5-y PFS
5-y LRPFS
Prognostic factor
RR
p
RR
p
RR
p
Postoperative CA 19-9 ($37 U/mL) Histologic grade (Grade 3) Residual disease R1 resection R2 resection Lymphovascular invasion (positive) Perineural invasion (positive) N stage (N1)
2.870 2.003
.001 .054
2.066 2.922
.033 .004
2.033 —
.116 —
1.486 1.285 1.439 — —
.207 .560 .189 — —
1.728 2.155 1.569 1.082 1.479
.106 .115 .150 .818 .143
1.290 4.031 — 2.442 —
.556 .008 — .075 —
Abbreviations: RR = relative risk; other abbreviations as in Tables 2 and 3.
I. J. Radiation Oncology d Biology d Physics
702
Volume 79, Number 3, 2011
(b) 100 CA 19-9 < 37 U/ml
80
CA 19-9
Cumulative Survival Rate (%)
Cumulative Survival Rate (%)
(a) 100 37 U/ml
60
40
p = 0.0136
20
0 0
1
2
3
4
5
6
7
CA 19-9 < 37 U/ml
80
CA 19-9
37 U/ml
60
40
p = 0.0034
20
0
8
0
1
2
3
Time (years)
4
5
6
7
8
Time (years)
Cumulative Survival Rate (%)
(c) 100 CA 19-9 < 37 U/ml
80
CA 19-9
37 U/ml
60
40
p = 0.0013
20
0 0
1
2
3
4
5
6
7
8
Time (years)
Fig. 4. Comparison of survival curves according to postradiotherapy carbohydrate antigen (CA) 19-9 level for patients with elevated CA 19-9 levels after resection. (a) Locoregional progression-free survival. (b) Progression-free survival. (c) Overall survival.
However, it seems probable that adjuvant RT might have contributed to the improved survival rate observed in the present study, considering the greater R0 resection rate (78% vs. 51%) in the historical surgery-only group from our institution (17). It is also controversial whether adjuvant RT can compensate for suboptimal resection in EHBDC. Several investigators have demonstrated a benefit of RT after R1 resection (9, 10, 14, 15, 18), but these results have been contradicted by other reports (2, 8, 12). Ben-David et al. (12) reported that R0 resection was the only significant predictor of OS and PFS and found no difference in OS or PFS between patients with R1 and R2 resection. The investigators concluded that surgery provided little benefit if R0 resection could not be achieved. However, these results could have been biased Table 5. Treatment-related toxicities Patients (n) Toxicity Acute Weight loss Hematologic Gastrointestinal Late Gastrointestinal
Grade 2 or less Grade 3 Grade 4 or greater 101 (100) 99 (98) 99 (98)
0 1 (1) 2 (2)
0 1 (1) 0
98 (97)
3 (3)
0
Data in parentheses are percentages.
by the small sample size and the enrollment of patients with gall bladder cancer (50% of R0 and R1 resections), which has clinical characteristics different from those of EHBDC (5). Also, two other related studies (2, 8) have been criticized because of the small sample size, poor prognostic features, and suboptimal treatment given to the adjuvant RT group (14, 19). In the present study, we found no significant differences in the 5-year OS, PFS, and LRPFS between the R0 and R1 resection groups. Our multivariate analysis found that R1 resection was not a significant prognostic factor for OS, PFS, or LRPFS. Although even R2 resection was not an independent prognostic factor for OS and PFS, R2 resection became significant when we excluded the effect of the postoperative CA 19-9 level in the multivariate analysis (data not shown). Our data strongly suggest that postoperative RT could improve locoregional tumor control and OS in patients with R1 resection, even though RT might not fully compensate for suboptimal resection. Carbohydrate antigen 19-9 is a carbohydrate antigen originally identified in the culture medium of a colorectal cancer cell line and is currently one of the most widely used tumor markers for cholangiocarcinoma (20, 21). A recent prospective Phase III trial reported that the postresection CA 19-9 level was a significant predictor of OS in patients with pancreatic cancer (22); however, little is known about the effectiveness of CA 19-9 as a prognostic indicator for bile duct cancer. In the present study, our
Postoperative CCRT for extrahepatic bile duct cancer d J. PARK et al.
703
Table 6. Recent studies of adjuvant chemoradiotherapy for extrahepatic bile duct cancer Investigator
Year
Patients (n)
R0/R1/R2 resection (n)
Stage N1 (n)
Median radiation dose (Gy)
CCRT (%)
5-y OS (%)
5-y DFS (%)
5-y LC (%)
Present study* Borghero et al. (14) Nelson et al. (13) Hughes et al. (11) Kim et al. (10)
2009 2008 2008 2007 2002
81 42 45 34 84
52 (58)/37 (42)/— 15 (36)/ 27 (64)/— 36 (80)/6 (13)/ 3 (7) 25 (74)/9 (26)/— 47 (56)/25 (30)/12 (14)
43 (48) 19 (45) 11 (25) 28 (82) 27 (32)
50 45 50.4 50.4 40
83 100 98 100 85
39 36 33 35 31
30 34 37 30 26
70 62 78 70 —
Abbreviations: CCRT = concurrent chemoradiotherapy; OS = overall survival; DFS = disease-free survival; LC = local control; other abbreviations as in Table 2. * Patients with R2 resection were excluded from Table 6. Data in parentheses are percentages
multivariate analysis indicated that the postoperative CA 19-9 level was an independent predictor of OS and PFS. Moreover, patients whose CA 19-9 level was elevated after resection, but in whom the concentration had returned to normal after RT, had greater OS, PFS, and LRPFS rates than did the patients whose CA 19-9 level remained elevated after RT. An elevated postoperative CA 19-9 level might indicate the burden of residual disease after resection as a component of locoregional or systemic disease. It is also possible that patients in whom the CA 19-9 level returned to normal after RT had radiosensitive tumors that were limited to the locoregional area. Thus, we suggest that the CA 19-9 level might be an important prognostic indicator after surgery or postoperative RT in patients with EHBDC. Also, we suggest that more aggressive treatment should be considered for patients in whom the CA 19-9 level remains elevated after postoperative RT. Concurrent chemoradiotherapy has proved effective against other gastrointestinal cancers (23), but only a few reports on the use of adjuvant CCRT for EHBDC have been published (Table 6). Previous studies have reported 5-year OS rates of 33–36%, with median survival of 16–37 months (10, 11, 13, 14, 24). In the present report, patients who underwent CCRT tended to have better 5-year OS, PFS, and LRPFS than the RT-only group. However, we could not conclude that CCRT was more effective than RT alone, because our study was retrospective and the size of the RT-only group was very small (16 patients). Regardless, our data showed that adjuvant CCRT that included oral 5-fluorouracil was
well tolerated. Considering the radiosensitizing effect of chemotherapy and the low toxicity among our patients, we suggest that CCRT with 3D-CRT be considered as a reasonable adjuvant therapy for EHBDC. The present study was nonrandomized and retrospective and, thus, might have had unrecognized biases, just as all other previous reports of EHBDC. To the best of our knowledge, however, our study is one of the largest patient series exploring the effect of postoperative RT for EHBDC and also addressed locoregional control and patterns of failure. Moreover, this is the first report to indicate that the postoperative CA 19-9 level is an independent prognostic factor for EHBDC recurrence. Because all patients underwent 3DCRT and most patients underwent CCRT, our study provides a basis for implementation of modern adjuvant chemoradiotherapy for EHBDC. CONCLUSION Our results suggest that adjuvant CCRT with 3D-CRT after surgical resection of EHBDC might effectively control locoregional disease and improve patient survival, with low toxicity levels, especially for patients with R1 resection. Distant metastasis was the most common pattern of recurrence; thus, more effective systemic therapy after adjuvant RT is clearly needed. The CA 19-9 level might be an important prognostic factor after resection and after postoperative RT for the identification of patients who might need more aggressive therapy.
REFERENCES 1. Longmire WP, McArthur MS, Bastounis EA, et al. Carcinoma of the extrahepatic biliary tract. Ann Surg 1973;178:333–345. 2. Nakeeb A, Pitt HA, Sohn TA, et al. Cholangiocarcinoma: A spectrum of intrahepatic, perihilar, and distal tumors. Ann Surg 1996;224:463–473. 465. 3. Reding R, Buard JL, Lebeau G, et al. Surgical management of 552 carcinomas of the extrahepatic bile ducts (gallbladder and periampullary tumors excluded): Results of the French Surgical Association Survey. Ann Surg 1991;213:236–241. 4. Kopelson G, Galdabini J, Warshaw AL, et al. Patterns of failure after curative surgery for extra-hepatic biliary tract carcinoma: Implications for adjuvant therapy. Int J Radiat Oncol Biol Phys 1981;7:413–417. 5. Jarnagin WR, Ruo L, Little SA, et al. Patterns of initial disease recurrence after resection of gallbladder carcinoma and hilar
6. 7.
8. 9.
cholangiocarcinoma: Implications for adjuvant therapeutic strategies. Cancer 2003;98:1689–1700. Wang SJ, Fuller CD, Kim JS, et al. Prediction model for estimating the survival benefit of adjuvant radiotherapy for gallbladder cancer. J Clin Oncol 2008;26:2112–2117. Shinohara ET, Mitra N, Guo M, et al. Radiotherapy is associated with improved survival in adjuvant and palliative treatment of extrahepatic cholangiocarcinomas. Int J Radiat Oncol Biol Phys 2009;74:1191–1198. Pitt HA, Nakeeb A, Abrams RA, et al. Perihilar cholangiocarcinoma: Postoperative radiotherapy does not improve survival. Ann Surg 1995;221:788–797. Todoroki T, Ohara K, Kawamoto T, et al. Benefits of adjuvant radiotherapy after radical resection of locally advanced main hepatic duct carcinoma. Int J Radiat Oncol Biol Phys 2000;46:581–587.
704
I. J. Radiation Oncology d Biology d Physics
10. Kim S, Kim SW, Bang YJ, et al. Role of postoperative radiotherapy in the management of extrahepatic bile duct cancer. Int J Radiat Oncol Biol Phys 2002;54:414–419. 11. Hughes MA, Frassica DA, Yeo CJ, et al. Adjuvant concurrent chemoradiation for adenocarcinoma of the distal common bile duct. Int J Radiat Oncol Biol Phys 2007;68:178–182. 12. Ben-David MA, Griffith KA, Abu-Isa E, et al. External-beam radiotherapy for localized extrahepatic cholangiocarcinoma. Int J Radiat Oncol Biol Phys 2006;66:772–779. 13. Nelson JW, Ghafoori AP, Willett CG, et al. Concurrent chemoradiotherapy in resected extrahepatic cholangiocarcinoma. Int J Radiat Oncol Biol Phys 2008;72:1495–1501. 14. Borghero Y, Crane CH, Szklaruk J, et al. Extrahepatic bile duct adenocarcinoma: Patients at high-risk for local recurrence treated with surgery and adjuvant chemoradiation have an equivalent overall survival to patients with standard-risk treated with surgery alone. Ann Surg Oncol 2008;15:3147– 3156. 15. Cheng Q, Luo X, Zhang B, et al. Predictive factors for prognosis of hilar cholangiocarcinoma: Postresection radiotherapy improves survival. Eur J Surg Oncol 2007;33:202–207. 16. Gerhards MF, van Gulik TM, Gonzalez Gonzalez D, et al. Results of postoperative radiotherapy for resectable hilar cholangiocarcinoma. World J Surg 2003;27:173–179.
Volume 79, Number 3, 2011
17. Lee SG, Lee YJ, Park KM, et al. One hundred and eleven liver resections for hilar bile duct cancer. J Hepatobiliary Pancreat Surg 2000;7:135–141. 18. Schoenthaler R, Phillips TL, Castro J, et al. Carcinoma of the extrahepatic bile ducts: The University of California at San Francisco experience. Ann Surg 1994;219:267–274. 19. Heron DE, Stein DE, Eschelman DJ, et al. Cholangiocarcinoma: The impact of tumor location and treatment strategy on outcome. Am J Clin Oncol 2003;26:422–428. 20. Nehls O, Gregor M, Klump B. Serum and bile markers for cholangiocarcinoma. Semin Liver Dis 2004;24:139–154. 21. Goonetilleke KS, Siriwardena AK. Systematic review of carbohydrate antigen (CA 19-9) as a biochemical marker in the diagnosis of pancreatic cancer. Eur J Surg Oncol 2007;33:266–270. 22. Berger AC, Garcia M Jr., Hoffman JP, et al. Postresection CA 19-9 predicts overall survival in patients with pancreatic cancer treated with adjuvant chemoradiation: A prospective validation by RTOG 9704. J Clin Oncol 2008;26:5918–5922. 23. National Institutes of Health. NIH consensus conference: Adjuvant therapy for patients with colon and rectal cancer. JAMA 1990;264:1444–1450. 24. Minsky BD, Wesson MF, Armstrong JG, et al. Combined modality therapy of extrahepatic biliary system cancer. Int J Radiat Oncol Biol Phys 1990;18:1157–1163.