Abstract
Background/Aim: To evaluate the efficacy of oxaliplatin plus fluorouracil and leucovorin (FOLFOX) on initially unresectable colorectal liver metastases (CRLM). Patients and Methods: From May 2005 to December 2008, FOLFOX was administered to 71 patients with initially unresectable CRLM. Hepatic resection was performed promptly after CRLM became resectable. Results: Twenty-six patients (37%) were downstaged as being resectable. The mean interval between the first FOLFOX and hepatic resection was six months (range, 3-7 months), and 7.1 courses (range, 2-12). Operative morbidity was 12% and mortality was nil. The median progression-free survival time was 19 and 7 months, and the median survival time was over 48 and 20 months, in finally resectable and unresectable patients, respectively. Multivariate analysis revealed that additional hepatic resection was the only independent prognostic factor (hazard ratio 4.80, p<0.01). Conclusion: FOLFOX is an effective chemotherapeutic regimen leading to successful hepatic resection and an excellent prognosis for patients with initially CRLM.
- Colorectal liver metastases
- hepatic resection
- chemotherapy with oxaliplatin plus fluorouracil and leucovorin
- FOLFOX
- Japanese patients
Combination chemotherapy including modulated infusional 5-fluorouracil (5-FU) plus irinotecan or oxaliplatin can achieve a response rate of 50% and a median survival of over 20 months (1-5). Oxaliplatin has been shown to improve the survival of patients with metastatic colorectal cancer, when given in combination with 5FU/LV, in first- or second-line therapy (1-5). Another phase III study has shown survival improvement using oxaliplatin plus 5-FU/LV over irinotecan plus 5-FU/leucovorin (LV) as a bolus administration (6).
In a phase III study to investigate two sequences of folinic acid, 5-FU, and irinotecan (FOLFIRI) followed by folinic acid, 5-FU, and oxaliplatin (FOLFOX6), and FOLFOX6 followed by FOLFIRI, hepatic resection of liver metastases was performed in 9% of patients after FOLFIRI versus 22% of patients in FOLFOX6 (p=0.02). R0 resection was performed in 7% of patients after FOLFIRI versus 13% after FOLFOX6 (3). Oxaliplatin-based chemotherapy, including the FOLFOX regimen, can lead to tumors being downstaged in some patients with initially unresectable colorectal liver metastases (CRLM), and allowed hepatic resection in 16-38 per cent patients (7). In a recent paper, FOLFOX4 resulted in tumor reduction in 60% patients and enabled surgical intervention in 40%, after a median of 6 months of chemotherapy in patients with liver-only CRLM (8). Therefore, many clinical oncologists and surgeons consider systemic chemotherapy with FOLFOX to be appropriate for CRLM (9).
It has been about 4 years from the introduction of oxaliplatin, which has been available for use in Japan since May 2005. The aim of this study was to assess the feasibility of use and the clinical value of preoperative FOLFOX in Japanese patients with initially unresectable CRLM.
Patients and Methods
From May 2005 to December 2008, 114 consecutive patients with CRLM were treated at the Department of Gastroenterological Surgery, Graduate School of Medical Sciences, Kumamoto University. The therapeutic strategy of CRLM in our institution after induction of FOLFOX is shown in Figure 1. A straightforward hepatic resection was selected for initially resectable 26 patients. Among 88 patients with initially unresectable CRLM or extrahepatic metastases, 71 patients treated with FOLFOX were entered into this study. Eight patients treated with FOLFOX and bevacizumab were excluded. There were 38 patients with liver-only metastases and 33 with liver plus extrahepatic metastases.
Therapeutic strategies for colorectal liver metastases. For patients with initially resectable disease and liver-only metastases, hepatic resection (HR) with 6 courses of adjuvant FOLFOX was performed. For these with initially unresectable or extrahepatic metastastases, induction chemotherapy with FOLFOX with or without bevacizumab was carried out. When curative hepatic resection became possible, HR with or without radiofrequency ablation (RFA) was immediately performed. Unresectable patients after FOLFOX were treated continuously with various regimens, including FOLFIRI, bevacizumab, cetuximab, or hepatic arterial infusion therapy.
The determination of initial resectability of CRLM before FOLFOX was based on the possibility of safe and curative (R0) resection. When CRLM became resectable after several courses of FOLFOX, hepatic resection was immediately performed. The final decision for hepatic resection after FOLFOX was based on the possibility for removing all metastases with resection and/or radiofrequency ablation therapy (RFA). Percutaneous transhepatic portal embolization was achieved preoperatively for two patients with an estimated volume of remnant functional liver parenchyma assessed by computed tomography (CT) were below 35%. The Institutional Review Board of the Graduate School of Medical Sciences, Kumamoto University, approved this clinical study.
Systemic chemotherapy. FOLFOX was administered mainly as outpatient chemotherapy with modified FOLFOX6 consisting of the biweekly regimen as follows: a 2-hour infusion of LV (200 mg/m2/d) and oxaliplatin 85 mg/m2 followed by a 5-FU bolus (400 mg/m2/d) and 46-hour infusion (2400 mg/m2/d) for 2 days every 2 weeks. Enrollment criteria in this study were as follows: age under 85 years, no organ dysfunction, and histologically proven adenocarcinoma. Treatment was continued until resectablility was achieved, disease progression, occurrence of unacceptable toxicity, or the patient's decision to discontinue treatment. After hepatic resection, the same regimen of preoperative systemic chemotherapy was continued postoperatively up to a total 12 cycles of pre- plus postoperative therapy.
Perioperative examination. Measurement of carcinoembryonic antigen (CEA) and carbohydrate antigen 19-9 (CA19-9) levels, abdominal ultrasonography (US), contrast-enhanced helical CT, magnetic resonance imaging (MRI) enhanced with superparamagnetic iron oxide (SPIO), and CT-angiography were performed routinely for preoperative staging. Patients with unresectable CRLM were assessed radiographically for resectability at every 3 cycles of FOLFOX, with measurement of tumor markers every month after starting therapy. The tumor regression effect was evaluated with CT according to the RECIST criteria (10).
Clinical characteristics of 26 patients with initially unresectable CRLM downstaged to be resectable after FOLFOX therapy.
Hepatic resection. The type of liver resection was based on the results of preoperative diagnostic imaging, intraoperative US, and careful attention to liver function. All detectable lesions were resected in principle or treated with RFA, for metastatic nodules smaller than 2 cm, especially deeper in the liver (11).
Histological examinations. In the patients who received hepatic resection after FOLFOX, the pathological effects of therapy of the tumor were determined in the resected specimens using grading criteria (12) as follows: grade 0: with no necrosis or cellular or structural change; grade 1a: with necrosis or disappearance of tumor in <1/3 of the entire lesion; grade 1b: with necrosis or disappearance of the tumor in <2/3 of the entire lesion; grade 2: with necrosis or disappearance of the tumor in >2/3 of the entire lesion, but with viable tumor cells remaining; and grade 3: with the entire lesion showing necrosis and/or fibrosis, and no viable tumor cells identified.
Complications. Operative morbidity and mortality were prospectively recorded.
Outcome. Cumulative progression-free survival (PFS) and overall survival (OS) after FOLFOX was recorded until June 2009, with the starting point being the day of initial FOLFOX therapy. Prognostic factors were evaluated by univariate and multivariate analysis.
Statistical analysis. Data are expressed as mean±standard deviation, and were compared between two groups by using the Mann-Whitney U-test. Categorical variables were compared by using the χ2 test or Fisher's exact test. PFS and OS were calculated by using the Kaplan-Meier method and were compared by using the log-rank test. The Cox proportional hazards regression model was used for the multivariate analysis. All statistical analyses were performed with StatView 5.0 computer software (SAS Institute Inc., Cary, NC, USA). Significance was defined as being a p-value of 0.05 or less.
Cumulative progression-free survival (PFS) (A) and overall survival (OS) (B) curve in 71 patients with initially unresectable liver metastases according to the existence or nonexistence of hepatic resection (solid line: patients with hepatic resection, dotted line: patients without hepatic resection). Cumulative PFS and OS in 26 patients with resectable disease after FOLFOX were significantly greater than that of 45 patients without hepatic resection (p<0.0001 and p<0.0005).
Results
Patient characteristics. Of 71 patients with initially unresectable CRLM, 26 (37%) of them became resectable after therapy. In patients with liver-only metastases and liver plus extrahepatic metastasis, the resection rates were 47% (18/38) and 24% (8/33), respectively. Clinical characteristics of the 26 patients are summarized in Table I. Baseline site of extrahepatic metastases which became resectable were the lung in 4 patients and para-aortic lymph nodes in 4 patients. The mean interval between starting FOLFOX and achieving hepatic resection was six months (range, 3 to 7), and a mean of 7.1 courses (range, 2 to 12) of FOLFOX were given. The mean size of maximal liver metastasis was downsized from 4.6±0.8 cm to 2.9±0.4 cm. RFA was additionally performed in 8 patients at the time of hepatic resection. Of 26 patients with CRLM, 18 (69%) experienced curative hepatic resection (R0). Portal embolization was performed preoperatively for two patients with 70% and 75% as an estimated resection of liver volume. Hepatic resections included 11 major resections larger than 3 segmentectomies, 8 resections with one to two segmentectomies, and 7 partial resections.
Response evaluation, tumor regression effect and histological examinations. According to the RECIST criteria, the response rate was 73% (19/26) in patients with CRLM finally resectable after FOLFOX. Regarding the 26 specimens resected after FOLFOX, the degree of histological effect in the tumor was classified as grade 1a in 13, 1b in 8, 2 in 3, and 3 in 2 patients. Two patients had no pathologically viable tumor cells in the liver metastases after FOLFOX therapy.
Postoperative complications. No patient died as a result of FOLFOX treatment. Postoperative complication was observed in 3 patients (pneumonia, prolonged jaundice, biliary leakage) in hepatic resection after FOLFOX but they smoothly recovered with medication and biliary drainage. Operative mortality within 3 months was nil.
Outcome. Cumulative PFS and OS in 71 CRLM patients treated with FOLFOX are shown in Figure 2. Cumulative PFS in 26 CRLM patients resectable after FOLFOX was significantly greater than that of 45 patients without hepatic resection (Figure 2A). Median PFS time was 19 and 7 months in finally resectable and unresectable patients, respectively. Cumulative OS of resectable patients was significantly greater compared to that of unresectable patients (Figure 2B). The mean observation period was 22 (range 9 to 48) months. Median survival time (MST) was over 48 months and 20 in finally resectable and unresectable patients, respectively. In a univariate analysis, hepatic resection (p<0.0001), response by RECIST criteria (p=0.02), and presence of bilateral liver metastases (p=0.03) significantly influenced OS. Multivariate analysis revealed that additional hepatic resection was the only independent prognostic factor (HR: 4.80, p<0.01) (Table II).
Discussion
Hepatic resection is the only curative treatment with long-term survival for patients with CRLM, although only approximately 20% of patients are candidates for surgery (13-16). Nowadays, hepatic resection is safe, despite the increasing complexity of resections, and the surgical mortality rate is less than 5% (13-16). In past years, almost all initially unresectable patients were treated with systemic or locoregional chemotherapy, resulting in a long-term survival of less than 5% (7, 13, 17).
Prognostic factors of 71 initially unresectable patients with CRLM treated with FOLFOX.
According to a review to determine the relationship between the rate of tumor response and the rate of resection in patients with initially unresectable liver metastases, disease in 24 to 54% of patients became resectable following chemotherapy and a strong correlation was found between response rates and the resection rates (r=0.96, p=0.002). The response rate of FOLFOX4 as first-line therapy for liver-only colorectal metastases was reported to be 60% (1, 8). In the present study, of the patients who received hepatic resection after FOLFOX therapy, 19 (73%) exhibited a partial response (PR).
The introduction of new chemotherapeutic and molecular targeting agents as standard treatments for metastatic colorectal cancer has resulted in better prognosis for patients with CRLM. Among patients with unresectable CRLM, chemotherapy can render some resectable, leading to the possibility of a prolonged survival (18, 19). Oxaliplatin-based regimen, mainly FOLFOX, downsized unresectable tumors or concomitant extrahepatic metastases to resectable in 16% to 51% of patients (7, 18, 20). In the present study, 37% (26/71) of patients with initially unresectable CRLM became resectable after a mean of 7.1 cycles of FOLFOX. The patients with liver-only metastases, the resection rates were still better, at 47% (18/38).
Eighteen patients excect for 8 patients treated combination with RFA, underwent histologically curative resection. R0 resection has been recommended to obtain good long-term prognosis (7). Nowadays, R1 resection provides similar survival rates compared to R0 resection in the era of new effective chemotherapy (21). Although RFA-combination resection is not R0 resection, PFS and OS were similar HR alone and RFA-combination (11).
Bismuth and colleagues (22) reported that the 5-year OS of 50% observed in patients with liver resection following neoadjuvant chemotherapy was comparable to 28% to 39% in primarily resectable patients (23, 24). In the current study, cumulative PFS and OS was significantly greater in finally resectable than unresectable patients. MST was over 40 months in finally resectable patients and 20 in unresectable patients even after FOLFOX treatment followed by other chemotherapy. Multivariate analysis demonstrated that additional hepatic resection was the only independent prognostic factor (HR: 4.80, p<0.01). Masi et al. (25) reported that in most patients, complete radiological remission does not reflect a complete pathological response, and the long-term outcome of patients who achieved a complete radiological remission without operation was not as good as that of patients who were radically operated without complete radiological remission (5-year survival 14% versus 42%). From these viewpoints, an alteration from unresectable to resectable disease by chemotherapy is quite important in the treatment strategy of CRLM.
The therapeutic dilemma faced by the hepatic surgeon is the timing of hepatic resection after chemotherapy. Surgery during chemotherapy must be performed immediately curative hepatic resection is possible. In addition, hepatic resection is recommended when complete response (CR), PR, and stable disease (SD) status following FOLFOX therapy is achieved. Three-year survival rates after surgery were 58% for patients with a PR and 45% with SD, while none of the patients with progressive disease (PD) were alive at three years (26). CR is usually defined as the disappearance of target lesions on imaging and is considered to be a good outcome in evaluating the efficacy of chemotherapy. Of 66 CRLM assessed as CR on CT scan before hepatic resection, persistent macroscopic or microscopic residual tumor or early recurrence were observed in 55 (83%) (27). In most patients receiving systemic chemotherapy for CRLM, a CR on diagnostic imaging does not indicate cure microscopically. In fact, 26 patients in the present study were rendered resectable, after being initially unresectable, although all but two tumors (92%) had viable components on histological examination.
In conclusion, FOLFOX is feasible and safe systemic chemotherapy for patients with CRLM, resulting in a high resectability rate and an excellent prognosis of patients with initially unresectable CRLM.
Footnotes
- Received December 4, 2009.
- Accepted February 26, 2010.
- Copyright© 2010 International Institute of Anticancer Research (Dr. John G. Delinassios), All rights reserved
References
In this issue
Jump to section
Related Articles
Cited By...
- Sinusoidal Obstruction Syndrome and Postoperative Complications Resulting from Preoperative Chemotherapy for Colorectal Cancer Liver Metastasis
- Post-chemotherapeutic CEA and CA19-9 Are Prognostic Factors in Patients with Colorectal Liver Metastases Treated with Hepatic Resection After Oxaliplatin-based Chemotherapy
- Liver Resectability of Advanced Liver-limited Colorectal Liver Metastases Following mFOLFOX6 with Bevacizumab (KSCC0802 Study)
- Carcinoembryonic Antigen Half-life Is an Early Predictor of Therapeutic Effects in Induction Chemotherapy for Liver Metastases from Colorectal Cancer