ORIGINAL ARTICLE
Chemotherapy or Liver Transplantation for Nonresectable Liver Metastases From Colorectal Cancer? Svein Dueland, MD, PhD,∗ Tormod K. Guren, MD, PhD,∗ Morten Hagness, MD, PhD,†‡ Bengt Glimelius, MD, PhD,§ P˚al-Dag Line, MD, PhD,† Per Pfeiffer, MD, PhD,¶ Aksel Foss, MD, PhD,†‡ and Kjell M. Tveit, MD, PhD∗ ‡ Objective: The primary objective was to compare overall survival (OS) in patients with colorectal cancer (CRC) with nonresectable liver-only metastases treated by liver transplantation or chemotherapy. Background: CRC is the third most common cancer worldwide. About 50% of patients will develop metastatic disease primarily to the liver and the lung. The majority of patients with liver metastases receive palliative chemotherapy, with a median OS of trial patients of about 2 years, and less than 10% are alive at 5 years. Methods: Patients with nonresectable liver-only CRC metastases underwent liver transplantation in the SECA study (n = 21). Disease-free survival (DFS) and OS of patients included in the SECA study were compared with progression-free survival (PFS) and OS in a similar cohort of CRC patients with liver-only disease included in a first-line chemotherapy study, the NORDIC VII study (n = 47). PFS/DFS and OS were estimated by the Kaplan-Meier method. Results: DFS/PFS in both groups were 8 to 10 months. However, a dramatic difference in OS was observed. The 5-year OS rate was 56% in patients undergoing liver transplantation compared with 9% in patients starting firstline chemotherapy. The reason for the large difference in OS despite similar DFS/PFS is likely different metastatic patterns at relapse/progression. Relapse in the liver transplantation group was often detected as small, slowly growing lung metastases, whereas progression of nonresectable liver metastases was observed in the chemotherapy group. Conclusions: Compared with chemotherapy, liver transplantation resulted in a marked increased OS in CRC patients with nonresectable liver-only metastases. Keywords: chemotherapy, colorectal cancer, liver metastases, liver transplantation, overall survival (Ann Surg 2015;261:956–960)
C
olorectal carcinoma (CRC) is one of the most frequent cancers worldwide, and about half of the patients develop metastases, primarily to the liver or lung. Surgical treatment of metastases is the only treatment option with curative potential; however, only about 10% to 20% of the patients seen at major hospitals are candidates
for surgical resection. Furthermore, the majority of patients undergoing liver resection develop relapse of the disease and 5-year survival is about 30% to 40% in most reported studies.1 Most patients with colorectal liver metastases have nonresectable disease; these patients have poor prognosis and only about 10% survive up to 5 years.2,3 Standard treatment option for patients with nonresectable liver-only metastases is palliative chemotherapy. Modern chemotherapy including the use of bevacizumab/EGFR antibodies has increased median survival of patients with metastatic CRC (mCRC) included in clinical trials from about 6 months to 2 years.4–6 Liver transplantation has been successfully developed as a treatment option for primary liver malignancy such as hepatocellular carcinoma with a 5-year overall survival (OS) after liver transplantation of about 70% to 80%.7,8 Before 1995, several liver transplants for colorectal liver metastases were performed and results were reported to the European Liver Transplant Registry. The 1- and 5-year survival rates were 62% and 18%, respectively.7 During this time period, the perioperative mortality after liver transplantation was high compared with the results obtained after 2005 and up to 30% of the patients treated before 1995 died from surgical or perioperative complications.9 On the background of improvements in outcome after liver transplantation for nonmalignant diseases and development of immunosuppressants with antimalignant properties, we initiated a study reexamining liver transplantation as a treatment option in mCRC patients with nonresectable liver-only disease in 2006 (SECA study). Of patients included in first-line chemotherapy studies, about 20% have liver-only disease and we considered liver transplantation to be a treatment option for a selected subgroup of these patients. We have recently reported a 5-year OS rate of 60% in 21 patients in the SECA study.10 Without liver transplantation, the patients in the SECA study would have been offered palliative chemotherapy. The aim of this article is to compare our results with liver transplantation of mCRC patients with the best available nonsurgical treatment. Thus, in this report, we have compared disease-free survival (DFS)/progression-free survival (PFS) and OS in mCRC patients with nonresectable liver-only disease, treated with liver transplantation (SECA study),10 and in the first-line chemotherapy study with Nordic FLOX with or without cetuximab (NORDIC VII).11
METHODS From the ∗ Department of Oncology, Oslo University Hospital, Oslo, Norway; †Section for Transplantation Surgery, Department of Transplantation Medicine, Oslo University Hospital, Oslo, Norway; ‡Institute of Clinical Medicine, University of Oslo, Oslo, Norway; §Department of Radiology, Oncology and Radiation Science, Uppsala University, Uppsala, Sweden; and ¶Department of Oncology, Odense University Hospital, Odense, Denmark. Disclosure: Supported by South-Eastern Norway Regional Health Authority, Oslo University Hospital, Norwegian Cancer Society, and Swedish Cancer Society. The authors declare no conflicts of interest. Reprints: Svein Dueland, MD, PhD, Department of Oncology, Oslo University Hospital, Postbox 4950 Nydalen, N-0424 Oslo, Norway. E-mail: [email protected]. C 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright ISSN: 0003-4932/14/26105-0956 DOI: 10.1097/SLA.0000000000000786
956 | www.annalsofsurgery.com
Study Design and Procedures The SECA study was an open, prospective pilot study with liver transplantation in patients with nonresectable CRC liver-only metastases.10 All patients (n = 21) who underwent liver transplantation are included in the present analysis. No patients in the SECA study turned out to have a BRAF mutated tumor, and they were 65 years or younger at the time of liver transplantation. The NORDIC VII study was a randomized, multicenter, 3-arm phase III trial in which patients were randomized 1:1:1 to Nordic FLOX (arm A), FLOX + cetuximab (arm B), and FLOX intermittently + cetuximab continuously (arm C). There were no significant differences between the treatment arms in terms of response rates, Annals of Surgery r Volume 261, Number 5, May 2015
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Liver Transplantation in Colorectal Cancer
PFS, or OS.11 Therefore, we combined the results from the 3 arms in this report. Of the entire NORDIC VII cohort, a total of 64 patients had nonresectable, BRAF nonmutated, liver-only metastases and were younger than 66 years. These patients were identified from the study database. Seventeen of these underwent liver resection after inclusion in the NORDIC VII study and were excluded from this comparison between chemotherapy and liver transplantation in nonresectable patients. Both SECA and NORDIC VII studies had obtained approvals from regional ethics committee and institutional review board and were registered in Clinical Trials.gov (SECA: NCT01311453; NORDIC VII: NCT00145314) before inclusion of patients.
DFS was determined by development of lung-only metastases in 13 of the 19 patients in the SECA study who had a relapse. The median size of lung metastases at the time of relapse was 7 mm. In comparison, of the patients in the NORDIC VII cohort who stopped chemotherapy because of progressive disease, 75% had progression of liver metastases. In the NORDIC VII study, the 5-year OS rate was 9% compared with 56% in the SECA study (P < 0.001 (Fig. 2A). The 5-year OS rate in the 21 patients with the longest OS in the NORDIC VII study was 19%, also significantly inferior to the SECA results (P = 0.012; Fig. 2B). Median OS in these 21 patients from the NORDIC VII population was 32 months compared with 21 months in all 47 patients from the NORDIC VII data set.
Statistical Analyses
OS After Progressive Disease
Data were continuously registered in case record forms. Survival data were estimated using the Kaplan-Meier method. Log-rank tests were used to compare outcomes between subgroups (the method used is noted in the text). For all tests, a 2-sided P < 0.05 was considered statistically significant. Analyses were performed with PASW statistics 18 (release 18.02) and SigmaPlot for Windows (version 11.0).
RESULTS
The 5-year OS rate after detection of relapse was 53% in the SECA group. Two patients in the NORDIC VII group had similar PFS and OS. Three of the 47 patients in the NORDIC VII study were alive 48 to 51 months after detection of progressive disease. All the other NORDIC VII patients were dead before 5 years after progressive disease was recorded (P < 0.001; Fig. 3). Median OS after the diagnosis of progressive disease in the 45 patients in NORDIC VII study with OS greater than PFS was 12 months.
Patients
OS and CEA Levels
The 21 patients undergoing liver transplantation according to the SECA protocol were included from November 2006 to March 2011. Median follow-up was 65 months (range, 32–84 months) by November 2013. No patient was lost to follow-up. Patients in the NORDIC VII study were included from May 2005 to October 2007. In this study, OS data from the 47 patients in the NORDIC VII study who did not undergo liver resection were collected up to 60 months after the last patient was included. Background characteristics of patients in the 2 studies are given in Table 1. They were similar except for baseline carcinoembryonic antigen (CEA) levels (median of 15μg/L in the SECA trial compared with that of 42 μg/L in the NORDIC VII trial). All patients in the NORDIC VII study received first-line chemotherapy. In contrast, 57% of patients included in the SECA study had received second- or thirdline chemotherapy. Actually, 6 patients in the SECA population had progressed on all standard lines of chemotherapy at the time of liver transplantation.
We have previously shown in the SECA study that patients with a CEA level of more than 80 μg/L at the time of liver transplantation had a significantly lower 5-year OS rate than patients with a CEA level of less than 80 μg/L.10 In the NORDIC VII study cohort, no significant differences in neither PFS nor OS were observed in patients with CEA levels above or below 80μg/L.
Clinical Outcome
Three patients were accepted for liver transplantation, and at the time of laparatomy, micro-metastatic disease was detected at frozen section examination of local lymph nodes. These patients did
Median DFS in the SECA (n = 21) and PFS in NORDIC VII (n = 47) studies were 10 and 8 months, respectively (Fig. 1).
Second- and Third-line Chemotherapy Of the 47 NORDIC VII patients, 39 started second-line chemotherapy. Median OS from start of second-line chemotherapy was 12 months. Of the 22 NORDIC VII patients with KRAS wildtype tumors who started second-line chemotherapy, 17 patients started third-line treatment with anti-EGFR antibodies. Median OS in these patients from start of third-line treatment was 7 months.
Survival in Patients With Microscopic Disease at the Time of Laparatomy
TABLE 1. Baseline Characteristics
Age, median (range), yr Sex (male/female) Site ECOG 0 CEA, median (range), μg/L ALP > UNL KRAS mutation Tumor size, median (range), cm Synchronous disease
Liver Transplantation SECA (n = 21)
NORDIC VII (n = 47)
NORDIC VII With the Longest Survival (n = 21)
56 (45–65) 62% male 52% colon 76% 15 (1–2002) 52% 42% 4.5 (2.8–13.0) 81%
58 (43–65) 72% male 66% colon 70% 42 (1–8260) 68% 32%∗ 4.7 (1.4–16.0) 96%
58 (34–64) 76% male 52% colon 86% 32 (1–1500) 48% 28%† 4.5 (1.4–16.0) 90%
∗ Six patients with unknown KRAS status excluded from the calculation. †Three patients with unknown KRAS status excluded from the calculation. ALP indicates alkaline phosphatase; ECOG, Eastern Cooperative Oncology Group; UNL, upper normal limit.
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.annalsofsurgery.com | 957
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Dueland et al
1.0
0.8
SECA (N=21)
0.6
Overall survival
Progression free survival
1.0
P=0.463
0.4 Nordic VII (N=47)
0.2 0.0 0
5
10
15
20
25
30
35
0.8
SECA (N=20)
0.6
0.4
P
Chemotherapy or Liver Transplantation for Nonresectable Liver Metastases From Colorectal Cancer? Svein Dueland, MD, PhD,∗ Tormod K. Guren, MD, PhD,∗ Morten Hagness, MD, PhD,†‡ Bengt Glimelius, MD, PhD,§ P˚al-Dag Line, MD, PhD,† Per Pfeiffer, MD, PhD,¶ Aksel Foss, MD, PhD,†‡ and Kjell M. Tveit, MD, PhD∗ ‡ Objective: The primary objective was to compare overall survival (OS) in patients with colorectal cancer (CRC) with nonresectable liver-only metastases treated by liver transplantation or chemotherapy. Background: CRC is the third most common cancer worldwide. About 50% of patients will develop metastatic disease primarily to the liver and the lung. The majority of patients with liver metastases receive palliative chemotherapy, with a median OS of trial patients of about 2 years, and less than 10% are alive at 5 years. Methods: Patients with nonresectable liver-only CRC metastases underwent liver transplantation in the SECA study (n = 21). Disease-free survival (DFS) and OS of patients included in the SECA study were compared with progression-free survival (PFS) and OS in a similar cohort of CRC patients with liver-only disease included in a first-line chemotherapy study, the NORDIC VII study (n = 47). PFS/DFS and OS were estimated by the Kaplan-Meier method. Results: DFS/PFS in both groups were 8 to 10 months. However, a dramatic difference in OS was observed. The 5-year OS rate was 56% in patients undergoing liver transplantation compared with 9% in patients starting firstline chemotherapy. The reason for the large difference in OS despite similar DFS/PFS is likely different metastatic patterns at relapse/progression. Relapse in the liver transplantation group was often detected as small, slowly growing lung metastases, whereas progression of nonresectable liver metastases was observed in the chemotherapy group. Conclusions: Compared with chemotherapy, liver transplantation resulted in a marked increased OS in CRC patients with nonresectable liver-only metastases. Keywords: chemotherapy, colorectal cancer, liver metastases, liver transplantation, overall survival (Ann Surg 2015;261:956–960)
C
olorectal carcinoma (CRC) is one of the most frequent cancers worldwide, and about half of the patients develop metastases, primarily to the liver or lung. Surgical treatment of metastases is the only treatment option with curative potential; however, only about 10% to 20% of the patients seen at major hospitals are candidates
for surgical resection. Furthermore, the majority of patients undergoing liver resection develop relapse of the disease and 5-year survival is about 30% to 40% in most reported studies.1 Most patients with colorectal liver metastases have nonresectable disease; these patients have poor prognosis and only about 10% survive up to 5 years.2,3 Standard treatment option for patients with nonresectable liver-only metastases is palliative chemotherapy. Modern chemotherapy including the use of bevacizumab/EGFR antibodies has increased median survival of patients with metastatic CRC (mCRC) included in clinical trials from about 6 months to 2 years.4–6 Liver transplantation has been successfully developed as a treatment option for primary liver malignancy such as hepatocellular carcinoma with a 5-year overall survival (OS) after liver transplantation of about 70% to 80%.7,8 Before 1995, several liver transplants for colorectal liver metastases were performed and results were reported to the European Liver Transplant Registry. The 1- and 5-year survival rates were 62% and 18%, respectively.7 During this time period, the perioperative mortality after liver transplantation was high compared with the results obtained after 2005 and up to 30% of the patients treated before 1995 died from surgical or perioperative complications.9 On the background of improvements in outcome after liver transplantation for nonmalignant diseases and development of immunosuppressants with antimalignant properties, we initiated a study reexamining liver transplantation as a treatment option in mCRC patients with nonresectable liver-only disease in 2006 (SECA study). Of patients included in first-line chemotherapy studies, about 20% have liver-only disease and we considered liver transplantation to be a treatment option for a selected subgroup of these patients. We have recently reported a 5-year OS rate of 60% in 21 patients in the SECA study.10 Without liver transplantation, the patients in the SECA study would have been offered palliative chemotherapy. The aim of this article is to compare our results with liver transplantation of mCRC patients with the best available nonsurgical treatment. Thus, in this report, we have compared disease-free survival (DFS)/progression-free survival (PFS) and OS in mCRC patients with nonresectable liver-only disease, treated with liver transplantation (SECA study),10 and in the first-line chemotherapy study with Nordic FLOX with or without cetuximab (NORDIC VII).11
METHODS From the ∗ Department of Oncology, Oslo University Hospital, Oslo, Norway; †Section for Transplantation Surgery, Department of Transplantation Medicine, Oslo University Hospital, Oslo, Norway; ‡Institute of Clinical Medicine, University of Oslo, Oslo, Norway; §Department of Radiology, Oncology and Radiation Science, Uppsala University, Uppsala, Sweden; and ¶Department of Oncology, Odense University Hospital, Odense, Denmark. Disclosure: Supported by South-Eastern Norway Regional Health Authority, Oslo University Hospital, Norwegian Cancer Society, and Swedish Cancer Society. The authors declare no conflicts of interest. Reprints: Svein Dueland, MD, PhD, Department of Oncology, Oslo University Hospital, Postbox 4950 Nydalen, N-0424 Oslo, Norway. E-mail: [email protected]. C 2014 Wolters Kluwer Health, Inc. All rights reserved. Copyright ISSN: 0003-4932/14/26105-0956 DOI: 10.1097/SLA.0000000000000786
956 | www.annalsofsurgery.com
Study Design and Procedures The SECA study was an open, prospective pilot study with liver transplantation in patients with nonresectable CRC liver-only metastases.10 All patients (n = 21) who underwent liver transplantation are included in the present analysis. No patients in the SECA study turned out to have a BRAF mutated tumor, and they were 65 years or younger at the time of liver transplantation. The NORDIC VII study was a randomized, multicenter, 3-arm phase III trial in which patients were randomized 1:1:1 to Nordic FLOX (arm A), FLOX + cetuximab (arm B), and FLOX intermittently + cetuximab continuously (arm C). There were no significant differences between the treatment arms in terms of response rates, Annals of Surgery r Volume 261, Number 5, May 2015
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Liver Transplantation in Colorectal Cancer
PFS, or OS.11 Therefore, we combined the results from the 3 arms in this report. Of the entire NORDIC VII cohort, a total of 64 patients had nonresectable, BRAF nonmutated, liver-only metastases and were younger than 66 years. These patients were identified from the study database. Seventeen of these underwent liver resection after inclusion in the NORDIC VII study and were excluded from this comparison between chemotherapy and liver transplantation in nonresectable patients. Both SECA and NORDIC VII studies had obtained approvals from regional ethics committee and institutional review board and were registered in Clinical Trials.gov (SECA: NCT01311453; NORDIC VII: NCT00145314) before inclusion of patients.
DFS was determined by development of lung-only metastases in 13 of the 19 patients in the SECA study who had a relapse. The median size of lung metastases at the time of relapse was 7 mm. In comparison, of the patients in the NORDIC VII cohort who stopped chemotherapy because of progressive disease, 75% had progression of liver metastases. In the NORDIC VII study, the 5-year OS rate was 9% compared with 56% in the SECA study (P < 0.001 (Fig. 2A). The 5-year OS rate in the 21 patients with the longest OS in the NORDIC VII study was 19%, also significantly inferior to the SECA results (P = 0.012; Fig. 2B). Median OS in these 21 patients from the NORDIC VII population was 32 months compared with 21 months in all 47 patients from the NORDIC VII data set.
Statistical Analyses
OS After Progressive Disease
Data were continuously registered in case record forms. Survival data were estimated using the Kaplan-Meier method. Log-rank tests were used to compare outcomes between subgroups (the method used is noted in the text). For all tests, a 2-sided P < 0.05 was considered statistically significant. Analyses were performed with PASW statistics 18 (release 18.02) and SigmaPlot for Windows (version 11.0).
RESULTS
The 5-year OS rate after detection of relapse was 53% in the SECA group. Two patients in the NORDIC VII group had similar PFS and OS. Three of the 47 patients in the NORDIC VII study were alive 48 to 51 months after detection of progressive disease. All the other NORDIC VII patients were dead before 5 years after progressive disease was recorded (P < 0.001; Fig. 3). Median OS after the diagnosis of progressive disease in the 45 patients in NORDIC VII study with OS greater than PFS was 12 months.
Patients
OS and CEA Levels
The 21 patients undergoing liver transplantation according to the SECA protocol were included from November 2006 to March 2011. Median follow-up was 65 months (range, 32–84 months) by November 2013. No patient was lost to follow-up. Patients in the NORDIC VII study were included from May 2005 to October 2007. In this study, OS data from the 47 patients in the NORDIC VII study who did not undergo liver resection were collected up to 60 months after the last patient was included. Background characteristics of patients in the 2 studies are given in Table 1. They were similar except for baseline carcinoembryonic antigen (CEA) levels (median of 15μg/L in the SECA trial compared with that of 42 μg/L in the NORDIC VII trial). All patients in the NORDIC VII study received first-line chemotherapy. In contrast, 57% of patients included in the SECA study had received second- or thirdline chemotherapy. Actually, 6 patients in the SECA population had progressed on all standard lines of chemotherapy at the time of liver transplantation.
We have previously shown in the SECA study that patients with a CEA level of more than 80 μg/L at the time of liver transplantation had a significantly lower 5-year OS rate than patients with a CEA level of less than 80 μg/L.10 In the NORDIC VII study cohort, no significant differences in neither PFS nor OS were observed in patients with CEA levels above or below 80μg/L.
Clinical Outcome
Three patients were accepted for liver transplantation, and at the time of laparatomy, micro-metastatic disease was detected at frozen section examination of local lymph nodes. These patients did
Median DFS in the SECA (n = 21) and PFS in NORDIC VII (n = 47) studies were 10 and 8 months, respectively (Fig. 1).
Second- and Third-line Chemotherapy Of the 47 NORDIC VII patients, 39 started second-line chemotherapy. Median OS from start of second-line chemotherapy was 12 months. Of the 22 NORDIC VII patients with KRAS wildtype tumors who started second-line chemotherapy, 17 patients started third-line treatment with anti-EGFR antibodies. Median OS in these patients from start of third-line treatment was 7 months.
Survival in Patients With Microscopic Disease at the Time of Laparatomy
TABLE 1. Baseline Characteristics
Age, median (range), yr Sex (male/female) Site ECOG 0 CEA, median (range), μg/L ALP > UNL KRAS mutation Tumor size, median (range), cm Synchronous disease
Liver Transplantation SECA (n = 21)
NORDIC VII (n = 47)
NORDIC VII With the Longest Survival (n = 21)
56 (45–65) 62% male 52% colon 76% 15 (1–2002) 52% 42% 4.5 (2.8–13.0) 81%
58 (43–65) 72% male 66% colon 70% 42 (1–8260) 68% 32%∗ 4.7 (1.4–16.0) 96%
58 (34–64) 76% male 52% colon 86% 32 (1–1500) 48% 28%† 4.5 (1.4–16.0) 90%
∗ Six patients with unknown KRAS status excluded from the calculation. †Three patients with unknown KRAS status excluded from the calculation. ALP indicates alkaline phosphatase; ECOG, Eastern Cooperative Oncology Group; UNL, upper normal limit.
C 2014 Wolters Kluwer Health, Inc. All rights reserved.
www.annalsofsurgery.com | 957
Copyright © 2014 Wolters Kluwer Health, Inc. Unauthorized reproduction of this article is prohibited.
Annals of Surgery r Volume 261, Number 5, May 2015
Dueland et al
1.0
0.8
SECA (N=21)
0.6
Overall survival
Progression free survival
1.0
P=0.463
0.4 Nordic VII (N=47)
0.2 0.0 0
5
10
15
20
25
30
35
0.8
SECA (N=20)
0.6
0.4
P
