|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
Journal of Clinical Oncology, Vol 26, No 10 (April 1), 2008: pp. 1635-1641 © 2008 American Society of Clinical Oncology. DOI: 10.1200/JCO.2007.13.7471 Complete Pathologic Response After Preoperative Chemotherapy for Colorectal Liver Metastases: Myth or Reality?
From the Assistance Publique–Hop Corresponding author: Professor René Adam, MD, PhD, AP-HP Hôpital Paul Brousse, Centre Hépato-Biliaire, 12 Avenue Paul Vaillant Couturier, F-94804 Villejuif, France; e-mail: rene.adam{at}pbr.aphp.fr
Purpose Complete clinical response (CCR) of colorectal liver metastases (CLM) following chemotherapy is of limited predictive value for complete pathologic response (CPR) and cure of the disease. The objective of this study was to determine predictive factors of CPR as well as its impact on survival. Patients and Methods From January 1985 to July 2006, 767 consecutive patients with CLM underwent liver resection after systemic chemotherapy. Patients with CPR were compared with patients without CPR.
Results Twenty-nine of 767 (4%) patients had CPR, and none of these 29 patients had CCR. Patients with CPR (mean age, 54 years) had a mean number of 3.3 metastases at diagnosis (mean size, 29.3 mm). Objective response and stable disease were observed in 79% and 21% of cases, respectively. Postoperative mortality rate was 0%. After a median follow-up of 52.2 months (range, 1.1 to 193.0 months), overall 5-year survival was 76% for patients with CPR compared with 45% for patients without CPR (P = .004). Independent predictive factors for CPR were: age
Conclusion CPR was observed in 4% of patients with CLM treated with preoperative chemotherapy. However, CPR may occur in almost one-third of objective responders age
More than 50% of patients with colorectal cancer will develop liver metastases.1 Hepatic resection today offers the only chance for cure, with 5-year survival rates ranging from 21% to 48%.2,3 For patients with initially irresectable metastases, modern chemotherapy treatment allows the conversion of approximately 13% of these patients to a resectable situation, with 5-year survival rates after resection reaching 35%.4 In addition, preoperative chemotherapy is increasingly being administered to patients with resectable disease. The objective of this approach is to control the metastatic disease, in order to avoid surgery in patients with rapidly progressive disease associated with a poor outcome after hepatic resection.5 Furthermore, reduction of tumor size facilitates the possibility for curative resection. The combination of fluorouracil and leucovorin (LV) with oxaliplatin or irinotecan (FOLFOX/FOLFIRI) has increased the response rates of colorectal liver metastases (CLM) up to 60%.6-8 Preliminary results show that the addition of biologic agents to conventional chemotherapy may further increase these response rates and, subsequently, the proportion of patients referred for surgery.9-11 With these advances in chemotherapy efficacy, the frequency of a complete clinical response (CCR) of metastases has increased. Although still rare, CCR has recently created debate in the surgical and medical oncology communities.12,13 CCR has, however, been shown to be of limited predictive value for complete pathologic response (CPR) and disease cure.12 By contrast, a complete tumor destruction is observed in some cases, whether the initial tumor sites have disappeared (CCR) or still persist (no CCR) radiologically, and from an oncologic perspective, CPR may have much more clinical relevance than CCR. Currently, no data exists concerning the long-term outcome and predictive factors of CPR. The aim of our study was to determine the incidence of CPR, to assess its relationship with CCR, and to evaluate its predictive factors and long-term survival.
Patient Selection From January 1985 to July 2006, 945 consecutive patients with CLM were operated on at our institution and prospectively included in a database. Of all 945 patients, 767 (81%) received preoperative systemic chemotherapy and underwent hepatectomy. This group consisted of 450 (59%) men and 317 (41%) women with a median age of 59.5 years (range, 27.6 to 84.8 years). The median number of metastases at diagnosis was 3 (range, 1 to 25) with a median maximum size of 35.0 mm (range, 1 to 250 mm). Concomitant extrahepatic disease was present in 162 patients (21%).
At diagnosis, disease was irresectable in 262 (34%) patients. Chemotherapy for resectable metastases was indicated for synchronous metastases with the primary tumor in place, or for marginally resectable disease (
Preoperative Management
Hepatic Resection
Pathological Examination At microscopic level, CPR was defined by the absence of any viable tumor cell irrespective of the proportions of necrosis and fibrosis. In case of any doubt, especially for the differential diagnosis with dystrophic biliary structures, additional immunohistochemistry for CK7 and CK20 was performed. CPR was only considered in patients following a complete resection of all CLM.
Follow-Up
Statistical Analysis
Of all 767 resected patients receiving preoperative chemotherapy, 29 (4%) patients presented with CPR. The remaining 738 patients formed the comparison group for the study population.
Patient and Tumor Characteristics
In nine patients with CPR (31%), extrahepatic disease was diagnosed before or during liver surgery, compared with 153 patients (21%) in the group without CPR (P = .18). For patients with CPR, extrahepatic metastases were present in the lungs in four patients (44%) and intra-abdominal lymph nodes in two patients (22%). In the other three patients (33%), extrahepatic disease was located in the ovaries (one patient), bone (one patient) and peritoneum (one patient). Only one of the six patients with CPR that underwent extrahepatic disease resection also had a complete necrosis of resected extrahepatic metastases. For the three patients who did not undergo extrahepatic disease resection, chemotherapy (two patients) and radiotherapy (one patient) resulted in the radiologic disappearance of the disease. Localization and resection of extrahepatic disease were not different for both patient groups (Table 1).
Chemotherapy Details Eighteen patients (62%) received one line, six patients (21%) had two lines, and two patients (7%) had three lines of chemotherapy. Only three patients (10%) received more than three lines. The majority of patients (66%) received FOLFOX in the last preoperative line (Table 1). Two of 29 patients (7%) had received cetuximab. However, no patients were treated with bevacizumab before hepatectomy. The median number of cycles was eight (range, three to 29 cycles) for patients with CPR compared with nine (range, one to 54 cycles) for patients without CPR. In the majority of patients with CPR (79%), a partial response was observed before surgery. The other 21% of patients had disease stabilization. CPR was observed in 19 of 340 patients treated with FOLFOX (6%) and in 2 (2%) of 99 patients receiving FOLFIRI (P = .14). No significant differences in chemotherapy characteristics were observed between patients with and without CPR (Table 1).
Relation Between CCR and CPR
Hepatectomy Characteristics Portal vein embolization was performed before hepatectomy in three patients to increase the volume of the future remnant liver (Table A1, online only).18 For two patients of the CPR group the selected hepatic resection constituted the second step of a two-stage resection.19 The 60-day mortality rate was 0%, and hepatic and general complications were observed in 16% and 24% of patients with CPR, respectively. Chemotherapy was continued postoperatively in 82% and 84% of patients with and without CPR, respectively (P = .75).
Pathology Details
Survival
A comparison of both patient groups solely considering patients with liver-only metastases revealed 5-year overall survival rates of 65% and 44% (P = .05) with and without CPR (Fig 3). After a median follow-up of 52.2 months (range, 1.1 to 193.0 months), less recurrences occurred in patients with CPR (41% v 62% for patients without CPR; P = .03). Hepatic recurrences were less frequent in the CPR group (17% v 57%, respectively; P = .003). Isolated extrahepatic or combined intra- and extrahepatic recurrences both occurred in 42% of patients with CPR. Repeat hepatectomy was performed in six patients, two of whom also underwent extrahepatic disease resection. Three additional patients only had resection of extrahepatic disease recurrence. At last follow-up, 20 patients with CPR are alive disease-free (69%), of whom nine more than 5 years and 5 more than 10 years after hepatectomy. Two patients are alive with disease (7%) and seven patients (24%) have died from recurrence.
Analysis of Predictive Factors Univariate analysis identified seven preoperative variables with statistical difference (P  .10) between both groups (Table 2). At multivariate analysis, four of the univariate factors were found to be independent predictive factors of CPR: age 60 years, maximum size of metastases 3 cm at diagnosis, CEA level 30 ng/mL at diagnosis, and objective response following chemotherapy.
Predictive Model for CPR To predict the presence of CPR in individual patients treated with preoperative chemotherapy, the four multivariate predictive factors were each assigned one point. For patients without any of the predictive factors, the probability to present with CPR was 0.2%. The addition of subsequent factors increased this probability to 0.9% for one factor, 3.6% for two factors, 12.7% for three factors, and 30.9% for four factors (Table 3).
The use of highly effective chemotherapy regimens in patients with CLM has resulted in an increased frequency of CCR. Recent studies have shown that most of the lesions that disappear on imaging are still found to have viable tumor when resected, or recur in situ when not resected.12 Based on these findings, it appears that CPR could be of more important prognostic value than CCR. To date, outcomes in patients with CPR have remained unexplored. In this study, we evaluated the long-term survival of patients with CPR after hepatic resection. Overall, CPR was identified in 4% of resected patients. Noteworthy, CPR was more than 10 times more frequent than CCR (0.3%). Previously, we showed CCR and CPR percentages of 3% and 7%, respectively, in more selected patients with initially irresectable disease, converted to resection after chemotherapy treatment.20 By definition, these patients were high responders to chemotherapy, resulting in a logically higher frequency of CCR. The strict definition of CCR used in the present series might be another explanation for the lower percentage of CCR, and a proportion of remnant metastatic lesions can now be detected by better preoperative imaging modalities. The fact that in our experience CPR was more than 10 times more frequent than CCR indicates that total necrosis does not imply disappearance of metastases on preoperative imaging. Conversely, in this study, none of the patients with CCR had CPR. CPR was associated with a remarkable 5-year overall survival of 76% with a still not reached median survival. Survival was not only significantly higher than for patients without CPR, but constitutes the highest rate of 5-year survival ever reported after resection of CLM. To eliminate the influence of extrahepatic metastases on long-term outcome, survivals of both groups were also compared considering patients with liver-only metastases. Five-year survival for patients with CPR decreased to 65% due to the exclusion of five of 11 patients who survived more than 5 years. However, survival remained higher than that of patients without CPR, confirming that CPR is a strong predictor of prolonged survival and cure. The favorable long-term results for patients with CPR could question the utility of surgery in this setting. However, in our opinion surgery should be proposed for several reasons. First, confirmation of CPR ultimately depends on the accuracy of gross pathologic examination and on the exhaustivity of histologic sampling, and few undetected malignant cells could therefore still be present in resected lesions. By resecting all metastases, the possibility of leaving active tumor cells untreated is secured. Second, additional metastatic lesions may be found at laparotomy, that otherwise would have remained undetected.21 Finally, no imaging technique, even positron emission tomography–CT, is currently reliable to diagnose CPR.22 In the absence of any reliable preoperative assessment tool for CPR, only surgical resection with concomitant pathologic examination of the specimen is able to make the definitive diagnosis. In addition, the complete disappearance of metastases on imaging should not contraindicate surgery, since a majority of these patients will not have CPR. Paradoxically, patients treated with chemotherapy should be referred to surgeons before CCR to avoid the therapeutically difficult situation associated with the inability to localize previous radiologically apparent lesions. In these situations, remnant malignant lesions might be missed, risking disease progression when chemotherapy is withdrawn. Few studies have dealed with complete response: three concerning clinical response with controversial results, and two approaching pathologic findings. Recent data have clarified that for 83% of CLM showing CCR, macro- or microscopic disease persists or in situ recurrence occurs within 1 year postoperatively.12 On the contrary, another study, evaluating the evolution of nonresected disappeared CLM in 11 patients, found that 73% of patients did not present with in situ recurrence after a median follow-up of 31 months.13 However, the limited number of patients and the use of intra-arterial chemotherapy in a large percentage of these patients makes it difficult to draw definitive conclusions on the clinical significance of these disappeared lesions. Additionally, a recent analysis from the Intergroup N9741 trial found a remarkable median survival of 44 months for patients with CCR after systemic chemotherapy alone.23 This concerned, however, a mixed patient population with different types of advanced colorectal cancer. With regard to pathologic response, a first study concluded that different types of remnant lesions can be present after chemotherapy treatment with a still undefined impact on outcome and recurrence.24 In another study, five of 112 patients pretreated with chemotherapy (4%) presented with CPR, which is in agreement with our series.25 A higher degree of pathologic tumor regression was associated with improved overall and disease-free survival. However, both studies lack the evaluation of predictive factors of CPR and of the impact of CPR on long-term survival. Interestingly, in the latter study, patients treated with oxaliplatin experienced a higher degree of tumor regression compared with patients treated with irinotecan.25 In our experience, the number of lines, cycles, and type of chemotherapy treatment did not influence the incidence of CPR. CPR occurred more frequently in patients treated with oxaliplatin; however, this difference did not reach statistical significance.
In clinical practice, an important issue is to appreciate the possibility of CPR. Owing to the overall incidence of 4%, it could be argued that this situation is anecdotal. However, by analyzing the predictive factors of CPR, our study determined that patients presenting with small tumors ( In conclusion, CPR is observed in 4% of patients resected for CLM following preoperative chemotherapy. When combined with an aggressive oncosurgical approach, CPR is associated with an exceptional high survival. The emergence of still more active chemotherapy regimens and biotherapies will certainly increase the frequency of CPR in the near future raising the need of a better comprehension of this entity. Further work should be encouraged to identify chemotherapeutic factors that increase the chance of CPR in resected patients, which is obviously associated with a real hope of cure.
The author(s) indicated no potential conflicts of interest.
Conception and design: René Adam, Francis Lévi, Catherine Guettier, Francis Kunstlinger, Denis Castaing Provision of study materials or patients: René Adam, Dennis A. Wicherts, Robbert J. de Haas, Francis Lévi, Bernard Paule, Catherine Guettier, Daniel Azoulay, Denis Castaing Collection and assembly of data: Dennis A. Wicherts, Robbert J. de Haas, Francis Kunstlinger Data analysis and interpretation: René Adam, Dennis A. Wicherts, Robbert J. de Haas, Thomas Aloia, Valérie Delvart Manuscript writing: René Adam, Dennis A. Wicherts, Robbert J. de Haas, Thomas Aloia, Catherine Guettier, Valérie Delvart Final approval of manuscript: René Adam, Dennis A. Wicherts, Robbert J. de Haas, Thomas Aloia, Francis Lévi, Bernard Paule, Catherine Guettier, Francis Kunstlinger, Valérie Delvart, Daniel Azoulay, Denis Castaing
Presented at the 2007 Annual Meeting of the American Society of Clinical Oncology, June 1-5, Chicago, IL. Both R.A. and D.A.W. contributed equally to this work. Authors’ disclosures of potential conflicts of interest and author contributions are found at the end of this article.
1. Steele G Jr, Ravikumar TS: Resection of hepatic metastases from colorectal cancer: Biologic perspectives. Ann Surg 210:127-138, 1989[Medline] 2. Scheele J, Altendorf-Hofmann A: Resection of colorectal liver metastases. Langenbecks Arch Surg 384:313-327, 1999[CrossRef][Medline] 3. Fong Y, Cohen AM, Fortner JG, et al: Liver resection for colorectal metastases. J Clin Oncol 15:938-946, 1997 4. Adam R, Avisar E, Ariche A, et al: Five-year survival following hepatic resection after neoadjuvant therapy for nonresectable colorectal liver metastases. Ann Surg Oncol 8:347-353, 2001[Medline] 5. Adam R, Pascal G, Castaing D, et al: Tumor progression while on chemotherapy: A contraindication to liver resection for multiple colorectal metastases? Ann Surg 240:1052-1064, 2004[CrossRef][Medline] 6. Giacchetti S, Itzhaki M, Gruia G, et al: Long-term survival of patients with unresectable colorectal cancer liver metastases following infusional chemotherapy with 5-fluorouracil, leucovorin, oxaliplatin and surgery. Ann Oncol 10:663-669, 1999 7. Alberts SR, Horvath WL, Sternfeld WC, et al: Oxaliplatin, fluorouracil, and leucovorin for patients with unresectable liver-only metastases from colorectal cancer: A north central cancer treatment group phase II study. J Clin Oncol 23:9243-9249, 2005 8. Saltz LB, Cox JV, Blanke C, et al: Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer. N Engl J Med 343:905-914, 2000 9. Folprecht G, Lutz MP, Schöffski P, et al: Cetuximab and irinotecan/5-fluorouracil/folinic acid is a safe combination for the first-line treatment of patients with epidermal growth factor receptor expressing metastatic colorectal carcinoma. Ann Oncol 17:450-456, 2006 10. Cunningham D, Humblet Y, Siena S, et al: Cetuximab monotherapy and cetuximab plus irinotecan in irinotecan-refractory metastatic colorectal cancer. N Engl J Med 351:337-345, 2004 11. Hurwitz H, Fehrenbacher L, Novotny W, et al: Bevacizumab plus irinotecan, fluorouracil, and leucovorin for metastatic colorectal cancer. N Engl J Med 350:2335-2342, 2004 12. Benoist S, Brouquet A, Penna C, et al: Complete response of colorectal liver metastases after chemotherapy: Does it mean cure? J Clin Oncol 24:3939-3945, 2006 13. Elias D, Youssef O, Sideris L, et al: Evolution of missing colorectal liver metastases following inductive chemotherapy and hepatectomy. J Surg Oncol 86:4-9, 2004[CrossRef][Medline] 14. Therasse P, Arbuck SG, Eisenhauer EA: New guidelines to evaluate the response to treatment in solid tumors. J Natl Cancer Inst 92:205-216, 2000 15. Adam R, Bismuth H, Castaing D, et al: Repeat hepatectomy for colorectal liver metastases. Ann Surg 225:51-62, 1997[CrossRef][Medline] 16. Cox DR: Regression models and life tables (with discussion). J R Stat Soc B 34:187-202, 1972 17. Breslow NE: Covariance analysis of censored survival data. Biometrics 30:89-99, 1974[CrossRef][Medline] 18. Azoulay D, Castaing D, Krissat J, et al: Percutaneous portal vein embolization increases the feasibility and safety of major resection for hepatocellular carcinoma in injured liver. Ann Surg 232:665-672, 2000[CrossRef][Medline] 19. Adam R, Laurent A, Azoulay D, et al: Two-stage hepatectomy: A planned strategy to treat irresectable liver tumors. Ann Surg 232:777-785, 2000[CrossRef][Medline] 20. Adam R, Delvart V, Pascal G, et al: Rescue surgery for unresectable colorectal liver metastases downstaged by chemotherapy: A model to predict long-term survival. Ann Surg 240:644-657, 2004[Medline] 21. Elias D, Sideris L, Pocard M, et al: Incidence of unsuspected and treatable metastatic disease associated with operable colorectal liver metastases discovered only at laparotomy (and not treated when performing percutaneous radiofrequency ablation). Ann Surg Oncol 12:298-302, 2005[CrossRef][Medline] 22. Tan MC, Linehan DC, Hawkins WG, et al: Chemotherapy-induced normalization of FDG uptake by colorectal liver metastases does not usually indicate complete pathologic response. J Gastrointest Surg 11:1112-1119, 2007[CrossRef][Medline] 23. Dy GK, Krook JE, Green EM, et al: Impact of complete response to chemotherapy on overall survival in advanced colorectal cancer: Results from Intergroup N9741. J Clin Oncol 25:3469-3474, 2007 24. Znajda T, Hayashi S, Horton P, et al: Postchemotherapy characteristics of hepatic colorectal metastases: Remnants of uncertain malignant potential. J Gastrointest Surg 10:483-489, 2006[CrossRef][Medline] 25. Rubbia-Brandt L, Giostra E, Brezault C, et al: Importance of histological tumor response assessment in predicting the outcome in patients with colorectal liver metastases treated with neo-adjuvant chemotherapy followed by liver surgery. Ann Oncol 18:299-304, 2007 Submitted August 8, 2007; accepted December 6, 2007.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2008 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
taux de Paris Hôpital Paul Brousse, Centre Hépato-Biliaire; Asistance Publique–Hop
TOP
ABSTRACT
INTRODUCTION
5 bilateral nodules or limited concomitant extrahepatic disease). Irresectability was defined as the technical inability to completely remove all metastases while leaving at least 30% of normal liver parenchyma. Patients received a median number of two lines of chemotherapy (range, 1 to 6 lines), with 350 patients (46%) receiving one line, 329 patients (43%) receiving two lines, 58 patients (8%) receiving three lines, and 26 patients (4%) receiving more than three lines. Last preoperative chemotherapy consisted of fluorouracil and LV alone (25%) or combined with oxaliplatin (48%), irinotecan (14%) or both (7%). Miscellaneous systemic regimens were administered in the remaining 7% of patients. Patients received a median number of nine cycles (range, 1 to 54 cycles). 
2 test for categoric data and the independent-samples T test for continuous data. Overall and disease-free survival probabilities were determined with the Kaplan-Meier method and compared using the log-rank test. A P value 
