Colorectal cancer is the second leading cause of cancer-related

Transcription

1 ORIGINAL ARTICLES Long-Term Outcome of Initially Unresectable Metastatic Colorectal Cancer Patients Treated with 5-Fluorouracil/Leucovorin, Oxaliplatin, and Irinotecan (FOLFOXIRI) Followed by Radical Surgery of Metastases Gianluca Masi, MD,* Fotios Loupakis, MD,* Luca Pollina, MD, Enrico Vasile, MD,* Samanta Cupini, MD,* Sergio Ricci, MD, Isa Maura Brunetti, MD, Roberta Ferraldeschi, MD, Giuseppe Naso, MD, Franco Filipponi, MD, Andrea Pietrabissa, MD, Orlando Goletti, MD,** Giacomo Baldi, MD,* Lorenzo Fornaro, MD,* Michele Andreuccetti, CTA,* and Alfredo Falcone, MD* Objective/Background: The GONO-FOLFOXIRI regimen improved the rate of R0 secondary resection of metastases in initially unresectable metastatic colorectal cancer. The objective of this study was to evaluate the long-term outcome of resected patients and the impact of FOLFOXIRI on perioperative morbidities, mortality, and chemotherapy induced hepatotoxicity. Patients and Methods: Overall, 196 patients with initially unresectable metastatic colorectal cancer were treated with FOLFOXIRI in 2 phase II and 1 phase III trial. This regimen was associated with an elevated response rate (70.4%) and 37 patients (19%) could undergo a secondary R0 surgery on metastases. This study was registered with the Australian New Zealand Clinical Trials Registry Database at and has ID number ACTRN Results: Main characteristics of the 37 radically resected patients were: median age 64 years (45 73), Eastern Cooperative Oncology Group Performance Status (ECOG) PS 1 in 30%, synchronous metastases in 65%, multiple sites of disease in 22%, and metastases confined to the liver in 68%. Preoperative FOLFOXIRI was administered for a median of 5.5 months. There was no perioperative mortality and all morbidities (27% of patients) resolved without sequelae. After a median follow up of 67 months, 5-year and 8-year survival are 42% and 33% respectively. At 5 years, 29% of patients are free of disease. The analysis of treatmentinduced liver injury showed neither G3 vascular toxicity nor G4 steatosis, and steato-hepatitis in only 5% of patients. Conclusions: The GONO-FOLFOXIRI regimen allow an R0 surgery in approximately 1 out of 5 unselected patients with initially unresectable metastatic colorectal cancer, and the long-term survival of resected From the *Unità Operativa Oncologia Medica, Azienda USL-6, Istituto Toscano Tumori, Livorno, Italy; Unità Operativa Anatomia e Istologia Patologica I, Azienda Ospedaliero-Universitaria, Istituto Toscano Tumori, Pisa, Italy; Unità Operativa Oncologia Medica, Azienda Ospedaliero-Universitaria, Istituto Toscano Tumori, Pisa, Italy; Unità Operativa Oncologia Medica B, Dipartimento di Medicina Sperimentale, Università la Sapienza, Roma, Italy; Unità Operativa Chirurgia Generale e Trapianto Fegato Universitaria, Azienda Ospedaliero-Universitaria, Istituto Toscano Tumori, Pisa, Italy; Unità Operativa Chirurgia Generale I Universitaria, Azienda Ospedaliero-Universitaria, Istituto Toscano Tumori, Pisa, Italy; **Unità Operativa Chirurgia Generale, Azienda USL-5, Istituto Toscano Tumori, Pisa, Italy; and Dipartimento di Oncologia, dei Trapianti e Nuove Tecnologie in Medicina, Università degli Studi, Pisa, Italy. Acknowledgements of research support: Supported in part by a research grant of the Associazione Italiana Ricerca Cancro (A.I.R.C.) and by the Fondazione A.R.C.O. Correspondence and requests for reprints to: Prof. Alfredo Falcone, Department of Oncology, Transplants and New Techologies in Medicine, University of Pisa, Department of Oncology, Azienda USL-6 of Livorno, Viale Alfieri, 36, Livorno, Italy, Phone , Fax , a.falcone@med.unipi.it. Copyright 2009 by Lippincott Williams & Wilkins ISSN: /09/ DOI: /SLA.0b013e31819a patients is considerable. Neoadjuvant FOLFOXIRI for 3-6 months is safe and not associated with severe liver injury. (Ann Surg 2009;249: ) Colorectal cancer is the second leading cause of cancer-related death in Western countries. The prognosis of metastatic colorectal carcinoma profoundly changes based on the possibility to perform a radical surgical resection of metastases. 1 In fact, despite the recent advances in the medical treatment of such disease, the chance of long-term survival remains unacceptably low ( 5%) for patients with not resected metastatic colorectal cancer. On the contrary, surgical resection of liver metastases is associated with a chance of survival at 10 years as high as 20%. 2 Also, in patients with pulmonary metastases or limited extrahepatic disease, surgical resection can offer opportunities for long-term survival. 3,4 Therefore, today, a major goal of medical treatment in initially unresectable metastatic colorectal cancer is to induce a significant tumor shrinkage in the attempt to increase the number of radically resected patients. In fact, the prognosis of patients turned resectable, and radically resected after response to chemotherapy, is similar to the prognosis of patients resected at the onset. 5 Of interest, a pooled analysis demonstrated a correlation between the activity of chemotherapy in terms of response rate and the likelihood of surgical resection of metastases. 6 Mainly on the basis of these evidences the Gruppo Oncologico Nord Ovest (G.O.N.O.) developed a combination of infusional 5-fluorouracil/leucovorin (5FU/LV), irinotecan, and oxaliplatin (FOLFOXIRI) with the aim of improving the activity of the treatment and potentially to achieve an increased rate of secondary resection of metastases. Two consecutive phase II trials demonstrated that the FOLFOXIRI regimen had a manageable toxicity profile coupled with an elevated activity, 7,8 and a secondary radical resection of metastatic disease could be performed in 26% of patients. 9 Afterward, in a phase III trial, 10 the G.O.N.O. compared FOLFOXIRI to infusional 5-fluorouracil/leucovorin and irinotecan (FOLFIRI), demonstrating that the triple drug combination was associated with an improved activity and efficacy, and it significantly increased the rate of secondary radical surgery (15% versus 6%, P 0.03 in the overall population and 36% versus 12%, P 0.01 in patients with isolated liver disease). Two major concerns about the use of such an intensive strategy (triple drug chemotherapy followed by surgery) in metastatic colorectal cancer can be outlined. First, there is the possibility Annals of Surgery Volume 249, Number 3, March 2009

2 Annals of Surgery Volume 249, Number 3, March 2009 FOLFOXIRI and Resection of Colorectal Metastases that the increased rate of secondary resection of metastases could not translate in a longer long-term survival. And second, there is the risk of an increased perioperative mobility/mortality possibly due to an excessive chemotherapy-related liver injury. 11 We conducted the present pooled analysis, including the 196 initially unresectable metastatic colorectal cancer patients treated in firstline with FOLFOXIRI in theabove mentioned trials, to evaluate the long-term outcome of the patients undergoing a secondary radical resection of metastases and to asses the impact of FOLFOXIRI on perioperative morbidities, mortality, and chemotherapy-induced hepatotoxicity. PATIENTS AND METHODS Patient Selection From May 1999 to April 2005, the G.O.N.O. enrolled patients in 3 consecutive trials: a phase I-II (42 patients), a phase II (32 patients), and a phase III trial (244 patients, 122 of them treated with FOLFOXIRI). Main selection criteria were: colorectal adenocarcinoma, metastatic disease deemed unresectable, not previous palliative chemotherapy for metastatic disease, age years, Eastern Cooperative Oncology Group Performance Status (ECOG) performance status 2 if age 70 years or 0 if age years, measurable disease according to World Health Organization (WHO) criteria, adequate bone marrow, liver, cardiac, and renal functions. The Ethics Committee of all participating institutions approved the protocols. Patients were informed of the investigational nature of the studies and provided their written informed consent before registration onto the study. Assessments Pretreatment evaluation included history, physical examination, complete blood profile, carcino-embryonic antigen (CEA), urine analysis, electrocardiogram, chest and abdominal computed tomography (CT) scan, and any other appropriate diagnostic procedures to evaluate metastatic sites. Disease unresectability was established through a multidisciplinary assessment performed by a team that included surgeons and medical oncologists from the institutions involved in these studies. Sites of metastatic disease were evaluated at the onset of chemotherapy and after every 8 weeks. In case of response, the same team of surgeons and medical oncologists reconsidered disease resectability. Chemotherapy The treatment planned in the first study (42 patients) consisted of irinotecan 125 mg/m 2 in the initial 3 patients and then 175 mg/m 2, followed by oxaliplatin 100 mg/m2, I LV 200 mg/m 2, and 5-FU 3800 mg/m 2 administered as a 48-hour chronomodulated continuous infusion (FOLFOXIRI). The treatment planned in the subsequent phase II (32 patients) and phase III (122 patients) studies consisted of irinotecan 165 mg/m 2, followed by oxaliplatin 85 mg/m 2, I LV 200 mg/m 2, and 5-FU 3200 mg/m 2 administered as a 48-hour flat continuous infusion ( simplified FOLFOXIRI). The simplified FOLFOXIRI schedule is outlined in table 1 and figure 1. Both regimens were repeated every 2 weeks. Treatment was administered for a maximum of 12 cycles or until evidence of progression, unacceptable toxicity, or patient refusal. Surgery and Postoperative Outcomes Surgical resection of metastases was reconsidered during chemotherapy and was attempted when technically feasible and potentially curative. In patients with liver disease, a complete exploration of the abdomen, including intraoperative ultrasonography, was recommended. Radiofrequency ablation (RFA) was allowed only in association with surgery, to treat few ( 4) and small ( 3 cm) other- TABLE 1. Characteristics of Patients Who Underwent Curative Surgery After FOLFOXIRI Characteristic N % No. of patients Sex Male Female ECOG performance status Age, median (range) 64 yr yr Primary tumor Colon Rectum No. of metastatic sites Single Multiple 8 22 Metastases Synchronous Metachronous Sites of disease Liver only Liver and lymph nodes 5 14 Liver and peritoneum 1 3 Liver and lung 2 5 Lung only 4 11 Liver involvement 25% Reason of initial unresectability (according to ONCOSURGE criteria) Extensive liver involvement segments involved % liver parenchyma involved All 3 hepatic veins involved 2 5 Unresectable extra-hepatic disease 6 16 Patient unfit for surgery 1 3 Immediate resection not appropriate Inadequate radiological margins 4 11 Portal lymph nodes involvement 3 8 Number of metastases 4 or 4 but bi-lobar 4 11 FIGURE 1. Simplified FOLFOXIRI regimen Lippincott Williams & Wilkins 421

3 Masi et al Annals of Surgery Volume 249, Number 3, March 2009 wise unresectable liver metastases. Postoperative chemotherapy was not planned, but it was allowed. After surgery, follow-up was performed every 2 months with physical examination, complete blood profile, CEA, and CT scan of the chest and abdomen. The progression free survival was calculated from the day of registration/randomization into the study to the first observation of disease progression or death from any cause. The time to failure of strategy (TFS) was calculated from the day of registration/randomization into the study to the observation of disease progression deemed definitively unresectable or death from any cause. The overall survival time was calculated from the day of registration/randomization into the study until death from any cause, censoring patients who had not died at the last date known to be alive. Survival curves were estimated by the Kaplan-Meier method and compared with the log-rank test. Histopathologic Examination Histology was reviewed in 21 patients who underwent hepatic resection. Three sections of tumor and nontumoral liver from one large block of the resected specimen were sampled to evaluate pathologic changes. Sinusoidal dilatation was graded according to Rubbia-Brandt et al: 12 Grade 0 absent, Grade 1 mild (centrilobular involvement limited to 1 third of the lobular surface), Grade 2 moderate (centrilobular involvement extending to 2 thirds of the lobular surface) and Grade 3 severe (complete lobular involvement). Steatosis was assessed in the peri-tumoral liver, and the presence of macrovesicular and microvesicular steatosis was noted. The degree of macrovacuolar steatosis in the liver not involved with tumor was classified into 5 grades: Grade 0 none, Grade 1 steatosis in 10% of the parenchymal area, Grade 2 steatosis in 10% but 30% of the total parenchymal area, Grade 3 steatosis in 30% but 60% of the parenchymal area, and Grade 4 steatosis in 60% of the parenchymal area. Steatohepatitis was graded as defined by Kleiner et al 13 based on steatosis (score 0 5%; 1 5% to 33%; 2 33% to 66%; and 3 66%), lobular inflammation (score 0 no foci; 1 2 foci; 2 2 to 4 foci; and 3 4 foci per X200 field), and ballooning (score 0 none; 1 few balloon cells; 2 many cells/prominent ballooning). RESULTS Patients A total of 196 patients with metastatic colorectal cancer deemed unresectable were enrolled onto the above mentioned trials and treated with FOLFOXIRI. Among these 196 patients, 73 (37%) had metastases confined to the liver. Overall, 138 (70%) patients achieved an objective response to chemotherapy. In 67 patients, surgery was not reconsidered because of the extent of disease, while the remaining 71 patients (36% of total) were reevaluated for surgical resection of metastases. Among these 71 patients, 29 were unable to undergo operation with a curative intent because this was judged not feasible after a clinical and instrumental reevaluation (24 patients) or after an exploratory laparotomy (5 patients), whereas in 5 patients the operation was not radical (R1-R2). Finally, 37 of the initial 196 patients could undergo operation with curative intent (resection rate: 19%) that consisted of a radical surgical resection of all metastatic sites (R0; 29 patients) or an R0 resection of metastases combined with intraoperative RFA of small residual hepatic nodules (R0-RF; 8 patients). In particular, surgery with curative intent was performed in 25 of the 73 patients with disease confined to the liver at baseline (resection rate: 34%). A flow-chart of the treatments is summarized in Figure 2. Characteristics of the 37 resected patients are listed in table 1. Of note metastases, were synchronous in 65% of patients and 22% had multiple sites of disease. Baseline sites of disease were: liver only in 25 patients (68%), lung only in 4 patients (11%), liver and lung in 2 (5%), and liver and lymph nodes or 422 FIGURE 2. Treatments flow-chart. peritoneum in 6 patients (17%). Main reasons for initial unresectability were classified according to ONCOSURGE criteria 14 and reported in table 1. With regard to the 25 patients with liver only metastases, the reasons why they were deemed initially not resectable were: all 3 hepatic veins involved in 2 patients (8%), involvement of 7 or 8 liver segments in 7 patients (28%), inadequate future liver remnant ( 20%) in 7 patients (28%). Moreover, 1 patient (4%) was judged unfit for surgery at baseline and finally the resection was considered technically feasible, but oncologically not appropriate in 8 patients (32%) because of the presence on more than 4 bi-lobar synchronous metastases or because of the estimation of inadequate radiologic margins. Chemotherapy Administration Among the 37 patients to undergo surgery with curative intent, a median of 11 cycles of preoperative chemotherapy were administered (range, 9 15 cycles), with a median duration of preoperative chemotherapy of 5.5 months (range, 1.6 to 8.3 months). Only 5 patients (14%) were treated for more than 6 months. The median time to achieve the best objective response (WHO criteria) was 2.6 months. The median time between the end of chemotherapy and operation was 1.9 months. Ten patients received postoperative chemotherapy with systemic FOLFOXIRI (4 patients) or with intrahepatic floxuridine (6 patients). Surgical and local Treatments A major hepatectomy ( 3 segments) was performed in 19 patients (52%), a minor hepatectomy ( 3 segments) in 14 patients (38%), and multiple segmental lung resections in 4 patients (11%). Surgical removal of circumscribed extrahepatic disease was also performed in 8 patients (22%) with liver metastases (abdominal lymphonodes 5 patients, lung 2 patients, peritoneum 1 patient). Intraoperative RFA of hepatic metastases was used in combination with liver resection in 8 patients to treat small residual nodules that could not be resected, but that could be fully ablated. The histopathologic analysis of resected nodules demonstrated a complete pathologic response of all lesions (absence of any viable tumor cell irrespective of the proportion of necrosis and fibrosis) in 4 patients (11%). These 4 patients had primary colon cancer and liver only disease, number of metastases ranging from 2 to 7 and maximum diameter of metastases ranging from 3 to 5 cm. All these 4 patients had achieved an instrumental partial response to chemotherapy. Operative Morbitity-Mortality and Liver Injury There was neither intraoperative nor postoperative mortality within 3 months after surgery. The perioperative complication rate 2009 Lippincott Williams & Wilkins

4 Annals of Surgery Volume 249, Number 3, March 2009 FOLFOXIRI and Resection of Colorectal Metastases was 27% and included: transient liver failure (3 patients), biliary fistula (2 patients), wound infection (2 patients), bilioma (1 patient), and pneumonia (1 patient). All complications resolved without sequelae. One pathologist with hepatobiliary expertise and who was blinded to clinical data made a detailed microscopic assessment of liver tissue not involved by tumor in the resected specimen from 21 patients. Details of histo-pathologic evaluation are the followings: sinusoidal dilatation was identified in 100% of patients (grade 1: 52%; grade 2: 48%) but no grade 3 vascular toxicity was found; steatosis of any grade was demonstrated in 76% of patients (grade 1: 52%; grade 2: 19%; grade 3: 5%), but no grade 4 steatosis was found; steatohepatitis was observed only in 1 patients (5%). Disease Recurrence and Survival The median follow-up of patients who underwent surgery with curative intent is 67 months. Disease recurred in 31 of these 37 patients. The median PFS from study entry is 18 months, and at 5 years 6 patients (16%) are still free of progression; all these 6 patients had disease confined to the liver at baseline. After disease recurrence, 27 (87%) patients received second-line chemotherapy with FOLFOXIRI (9 patients), FOLFIRI (12 patients), infusional 5-fluorouracil/leucovorin and oxaliplatin (FOLFOX) (5 patients), or cetuximab (1 patient). Among these 31 patients who developed a recurrence of the disease, 11 (36%) were able to benefit from a new radical resection or RFA of metastases. As a consequence, the median TFS is 19 months, and 11 patients (29%) are free of disease at 5 years as shown in Figure 3. With regard to survival, 22 out of 37 patients have died (20 because of progressive disease and 2 for other causes). The median survival from study entry is 40 months, and the estimated 5-year and 8-year survival rates are 42% and 33% respectively (Fig. 3). For the 25 patients with liver only disease, the median survival is 61 months and the estimated 5-year and 8-year survival rate is 43%. For the 17 patients with truly unresectable liver only metastases (excluding 8 patients deemed oncologically not appropriate for resection), the median survival is 34 months and the estimated 5-year and 8-year survival rates are 32% and 22% respectively. For comparison, the median survival of the 14 patients who achieved a clinical complete response to chemotherapy, but did not undergo operation, is 30 months (5-year survival: 14%, log-rank P versus operated 0.006); the median survival of the 101 patients who responded to chemotherapy (complete or partial response), but did not undergo operation is 23 months (5 years survival: 3%, log-rank P versus operated 0.001) and the median survival of the 56 patients who did neither achieve an objective response to FOLFOXIRI (stable disease and progression) nor undergo operation is 14 months (5 years survival: 0%, log-rank P versus operated 0.001). Finally, the median survival of patients who achieved a complete pathologic response is 64 months (5 years survival: 75%). Survival curves are reported in Figure 4. DISCUSSION The treatment of metastatic colorectal cancer has changed mainly for improvements in the efficacy of chemotherapy, for the availability of targeted agents, and for an increased use of surgery on metastases. Unfortunately, survival beyond 5 years is still uncommon in patients with non resected metastatic disease. Therefore, surgical resection of metastases remains the best option to achieve long-term survival, and there is a considerable interest in expanding the criteria for resectability. There are growing evidences that an active neoadjuvant chemotherapy can shift patients from having unresectable to resectable disease, and the pooled analysis of Folprecht et al 6 demonstrated a strong correlation between the FIGURE 3. Kaplan-Meier estimates of time to progression (A), time to failure of strategy (B), and overall survival (C) of radically resected patients after response to FOLFOXIRI. activity of chemotherapy and the resection rate, being the response to chemotherapy almost a prerequisite for resection. In this context, we developed the FOLFOXIRI regimen also with the aim to expand the fraction of patients eligible for a potentially curative resection of metastases. The data we reported show that this triple-drug combination made possible a radical resection of initially unresectable metastases in approximately 1 out 5 patients. Moreover, with a 2009 Lippincott Williams & Wilkins 423

5 Masi et al Annals of Surgery Volume 249, Number 3, March 2009 FIGURE 4. Kaplan-Meier estimates of overall survival (OS) of patients treated with FOLFOXIRI. (A), Patients responders to FOLFOXIRI and radically resected versus patients with clinical complete response to FOLFOXIRI and not resected. (B), Patients responders to FOLFOXIRI and radically resected versus patients with clinical complete or partial response to FOL- FOXIRI and not resected versus patients not responders (stable disease or progression) and not resected. median follow up of more than 5 years, the long-term survival of radically resected patients is particularly remarkable, with actuarial 5-year and 8-year survival rates of 42% and 33% respectively. Of note, these results were achieved in a particularly difficult setting of patients with metastatic colorectal cancer who were not selected for a neoadjuvant approach and in whom the initial treatment had mainly a palliative purpose because of the extent of metastatic disease. In fact, although the 5-year survival reported in trials using oxaliplatin or irinotecan based doublets 5,15,16,17 has also ranged between 30% and 50%, these results were obtained in more selected populations of patients with liver-only disease. Unfortunately, to compare the results of studies of preoperative chemotherapy is problematic because of the heterogeneous definition of unresectability and because of the variable surgical skill and aggressiveness. Nevertheless, in our opinion, the FOLFOXIRI regimen is of particular interest in a neoadjuvant strategy due to its elevated activity coupled with a manageable toxicity profile. In addition, it is worth noting that with FOLFOXIRI we observed a considerable incidence of complete pathologic responses (11% of resected patients) and this condition is associated with a very favorable outcome, as also recently reported. 18 An aspect to be discussed is the duration of the preoperative chemotherapy. In fact, even if achievable in a small percentage of patients, a complete disappearance of radiologically evident disease, 424 in most cases, does not reflect a complete pathologic response, but may compromise surgery. 19 Furthermore, the long-term outcome of patients who achieved a complete radiologic remission of disease, but were not operated on, is not as good as that of patients who were radically operated on 20 (in the present analysis, 5-year survival 14% versus 42%). For these reasons, we recommended to reconsider surgical resection as soon as it was judged feasible, especially in patients for whom surgery was a real chance of cure and when all sites of metastases could be resected. Instead, in patients with more diffuse metastatic disease where, even after a major response, not all sites of initial disease could be resected, we considered a more prolonged chemotherapy (up to 6 months), in the attempt to induce a pathologic complete response of some metastatic lesions. It must be considered that the risk of liver toxicity, and possibly the incidence of surgical complication, pay increase with prolonged chemotherapy. The surgical complication rate, the perioperative morbility and mortality, and the degree of liver injury we reported compare favorably with those of other series. 21,22,23,24,25 Probably, the duration of preoperative treatment (no longer then 6 months) in the majority of patients and the non elevated dose of irinotecan we used can explain these data. A key point is the selection of patients who will benefit most from an intensive therapeutic approach including chemotherapy and surgery on metastases. In fact, a technically feasible resection is not always a curative resection. In the absence of validated biologic prognostic factors, several authors proposed clinico-pathologic scoring systems to predict survival after surgery in metastatic patients downstaged by chemotherapy. 26,27 However, recent data suggest that the presence of poor prognostic factors does not preclude the possibility of long-term survival. 2 Therefore today we should select patients by combining these prognostic models with advanced imaging techniques and with a careful clinical evaluation of the disease spread and aggressiveness, but every effort should be made to not deny a potentially curative treatment to our patients. In this scenario, the availability of a more active, but not substantially more toxic regimen like FOLFOXIRI, should be considered when a potentially curative resection is a reasonable option. Today, targeted agents like anti-vegf and anti-egfr monoclonal antibodies are routinely used to treat metastatic colorectal patients. Results of randomized trials 28,29 demonstrated that combining a biologic agent to an oxaliplatin or irinotecan based doublet can improve treatment activity and efficacy and also the rate of secondary resection of metastases. Therefore, on behalf of the G.O.N.O. we conducted a phase II trial in unresectable metastatic colorectal cancer by combining FOLFOXIRI with bevacizumab. 30 A phase III trial to compare this regimen to a combination of FOLFIRI and bevacizumab is ongoing. In conclusion, the G.O.N.O. FOLFOXIRI regimen not only improves the rate of radical surgery in initially unresectable metastatic colorectal cancer, but also long-term survival of resected patients, with considerable benefit. Furthermore, neoadjuvant FOLFOXIRI for 3 6 months is safe and not associated with unexpected adverse events or severe liver injury. ACKNOWLEDGMENTS The authors thank Mrs. Simona Giannace and Mrs. Etleva Beijta for data management and technical assistance. REFERENCES 1. Bismuth H, Adam R, Lévi F, et al. Resection of nonresectable liver metastases from colorectal cancer after neoadjuvant chemotherapy. Ann Surg. 1996;224(4): Tomlinson JS, Jarnagin WR, DeMatteo RP, et al. Actual 10-year survival after resection of colorectal liver metastases defines cure. J Clin Oncol. 2007;25(29): Lippincott Williams & Wilkins

6 Annals of Surgery Volume 249, Number 3, March 2009 FOLFOXIRI and Resection of Colorectal Metastases 3. Shah SA, Haddad R, Al-Sukhni W, et al. Surgical resection of hepatic and pulmonary metastases from colorectal carcinoma. J Am Coll Surg. 2006; 202(3): Elias D, Sideris L, Pocard M, et al. Results of R0 resection for colorectal liver metastases associated with extrahepatic disease. Ann Surg Oncol. 2004;11(3): Adam R, Avisar E, Ariche A, et al. Five-year survival following hepatic resection after neoadjuvant therapy for nonresectable colorectal. Ann Surg Oncol. 2001;8(4): Folprecht G, Grothey A, Alberts S, et al. Neoadjuvant treatment of unresectable colorectal liver metastases: correlation between tumour response and resection rates. Ann Oncol. 2005;16(5): Falcone A, Masi G, Allegrini G, et al. Biweekly chemotherapy with oxaliplatin, irinotecan, infusional Fluorouracil, and leucovorin: a pilot study in patients with metastatic colorectal cancer. J Clin Oncol. 2002;20(19): Masi G, Allegrini G, Cupini S, et al. First line treatment of metastatic colorectal cancer with irinotecan, oxaliplatin and 5-fluorouracil/leucovorin (FOLFOXIRI): results of phase II study with a simplified biweekly schedule. Ann Oncol. 2004;15: Masi G, Cupini S, Marcucci L, et al. Treatment with 5-fluorouracil/folinic acid, oxaliplatin and irinotecan (FOLFOXIRI) enables surgical resection of metastases in patients with initially unresectable metastatic colorectal cancer. Ann Surg Oncol. 2006;13(1): Falcone A, Ricci S, Brunetti I, et al. Phase III trial of infusional fluorouracil, leucovorin, oxaliplatin, and irinotecan (FOLFOXIRI) compared with infusional fluorouracil, leucovorin, and irinotecan (FOLFIRI) as first-line treatment for metastatic colorectal cancer: the Gruppo Oncologico Nord Ovest. J Clin Oncol. 2007;25(13): Nordlinger B, Benoist S. Benefits and risks of neoadjuvant therapy for liver metastases. J Clin Oncol. 2006;24(31): Rubbia-Brandt L, Audard V, et al. Severe hepatic sinusoidal obstruction associated with oxaliplatin-based chemotherapy in patients with metastatic colorectal cancer. Ann Oncol. 2004;15(3): Kleiner DE, Brunt EM, Van Natta M, et al. Design and validation of a histological scoring system for nonalcoholic fatty liver disease. Hepatology. 2005;41(6): Poston GJ, Adam R, Alberts S, et al. OncoSurge: a strategy for improving resectability with curative intent in metastatic colorectal cancer. J Clin Oncol. 2005;23(28): 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. 1999;10(6): Barone C, Nuzzo G, Cassano A, et al. Final analysis of colorectal cancer patients treated with irinotecan and 5-fluorouracil plus folinic acid neoadjuvant chemotherapy for unresectable liver metastases. Br J Cancer. 2007; 97(8): Leonard GD, Brenner B, Kemeny NE. Neoadjuvant chemotherapy before liver resection for patients with unresectable liver metastases from colorectal carcinoma. J Clin Oncol. 2005;23(9): Adam R, Wicherts DA, de Haas RJ, et al. Complete pathologic response after preoperative chemotherapy for colorectal liver metastases: myth or realty? J Clin Oncol. 2008;26(10): Benoist S, Brouquet A, Penna C, et al. Complete response of colorectal liver metastases after chemotherapy: does it mean cure? J Clin Oncol. 2006;24(24): Wicherts DA, Haas RJ, Levi F, et al. Complete pathological response of colorectal liver metastases after neoadjuvant chemotherapy: myth or reality? J Clin Oncol. 2007;25:18S, (suppl; abstr 4063) 21. Zorzi D, Laurent A, Pawlik TM, et al. Chemotherapy-associated hepatotoxicity and surgery for colorectal liver metastases. Br J Surg. 2007;94(3): Vauthey JN, Pawlik TM, Ribero D, et al. Chemotherapy regimen predicts steatohepatitis and an increase in 90-day mortality after surgery for hepatic colorectal metastases. J Clin Oncol. 2006;24(13): Aloia T, Sebagh M, Plasse M, et al. Liver histology and surgical outcomes after preoperative chemotherapy with fluorouracil plus oxaliplatin in colorectal cancer liver metastases. J Clin Oncol. 2006;24(31): Karoui M, Penna C, Amin-Hashem M, et al. Influence of preoperative chemotherapy on the risk of major hepatectomy for colorectal liver metastases. Ann Surg. 2006;243(1): Kooby DA, Fong Y, Suriawinata A, et al. Impact of steatosis on perioperative outcome following hepatic resection. J Gastrointest Surg. 2003;7(8): Fong Y, Fortner J, Sun RL, et al. Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer: analysis of 1001 consecutive cases. Ann Surg. 1999;230(3): 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. 2004;240(4): Saltz L, Clarke S, Diaz-Rubio E, et al. Bevacizumab in combination with oxaliplatin-based chemotherapy as first-line therapy in metastatic colorectal cancer: a randomized phase III study. J Clin Oncol. 2008;26(12): Van Cutsem E, Nowacki M, Lang I, et al. Randomized phase III study of irinotecan and 5-FU/FA with or without cetuximab in the first-line treatment of patients with metastatic colorectal cancer (mcrc): The CRYSTAL trial. J Clin Oncol. 2007;25 June 20 Supplement (abstract 4000). 30. Masi G, Loupakis F, Baldi G, et al. First-line irinotecan, oxaliplatin and infusional 5FU/LV (FOLFOXIRI) in combination with bevacizumab (BV) in metastatic colorectal cancer (mcrc) patients (pts): A phase II study by the GONO Group. J Clin Oncol. 2008;26 May 20 suppl (abstract 4074) Lippincott Williams & Wilkins 425