Health Articles

 

Drjeanmichelbutte.cl

Curr Surg Rep (2014) 2:50 Treatment of Unresectable Liver-Only Disease: Systemic Therapyversus Locoregional Therapy Jean M. Butte • Chad G. Ball • Elijah Dixon  Springer Science + Business Media New York 2014 Most patients with colorectal liver metastases has decreased dramatically; nonetheless, an important present with unresectable/disseminated disease and are proportion of patients present with metastatic disease treated with palliative therapies. Patients with unresectable The liver is the most frequent solid organ involved with liver only disease represent a large subset that similarly has metastases Approximately 15–25 % of patients diag- a poor prognosis. Numerous reports pertaining to local and nosed with a colorectal cancer present with liver metastasis systemic therapies have been published, there are however (CLM) at the moment of diagnosis and about 50 % develop few methodologically rigorous studies to define the best CLM during their follow-up. Moreover, patients may approach in these patients. Most information comes from present with a wide spectrum of clinical scenarios that retrospective reports at high volume centers, which does impact their treatment and survival []. To facilitate com- not necessarily represent the current treatment for the parisons, patients are usually divided in a group with liver- majority of patients. The aim of this review is to analyze only disease (LOD) and another with extrahepatic disease the current systemic and local treatments utilized in these (EHD) to better define the therapy and prognosis [].
patients with the aim of defining the ideal approach for Complete resection/ablation is the treatment of choice for patients with resectable CLM because it is the onlytherapy that may lead to long-term survival and cure Liver metastases
Systemic therapy
Survival is significantly worse in patients treated only with Locoregional therapy
Unresectable liver-only disease
Intra-arterial chemotherapy
Directed radiotherapy
selecting patients for surgery may be challenging and Colorectal cancer
Treatment
Survival depends on multiple factors such as extent of disease (intraand extra hepatic), surgical expertise, extent of future liverremnant, overall patient fitness and performance status, comorbid medical conditions, and tumor biology Patients with isolated liver disease may be approached Colorectal cancer is the fourth most common cancer and differently, depending on the extent of disease. While there the second cause of cancer death in the USA. Its mortality is consensus that those patients with limited disease (i.e.,one liver metastasis) should be treated with resection upfront [], it has been difficult to define the best treatmentfor patients who present with extensive but resectable This article is part of the Topical Collection on Colorectal Liver disease. Some centers define this group as ‘‘potentially'' unresectable, based more on the biology of disease than on J. M. Butte
C. G. Ball
E. Dixon (&) real anatomical factors of resectability, and suggest that Service of Hepatobiliary and Pancreatic Surgery, Division of neoadjuvant chemotherapy could help to select those General Surgery, Faculty of Medicine, Foothills Medical Centre, patients that may benefit from surgery []. In contrast, University of Calgary, EG-26, Foothills Medical Centre, patients with truly unresectable disease (related to exten- 1403-29 Street NW, Calgary, AB T2N 2T9, Canadae-mail: [email protected] sion of disease, anatomical factors or liver remnant) should Curr Surg Rep (2014) 2:50 be treated initially with systemic and/or regional therapies, Finally, it is important to note that patients with limited with the main aim of downsizing the tumors to achieve a CLM and potential good tumor biology may have an chance for complete resection, which is the treatment that adverse prognosis because their disease is unresectable may improve survival in this subgroup of patients •].
because of anatomical or functional problems. Since the This review is focused on patients with CLM who concept of unresectability may be variable among centers, present with unresectable isolated liver disease and evalu- every patient must be evaluated within the context of a ates local and systemic treatments. For the purpose of this multidisciplinary team, which should include a surgeon article, the role of (truly) neoadjuvant chemotherapy only with expertise and experience in liver surgery, before will be discussed to better understand the biology and deciding that the disease is unresectable because in some prognosis of patients with extensive resectable disease, patients a two-stage hepatectomy could be performed , which is also important in helping to define the prognosis of patients with unresectable liver metastases.
Natural History of Patients with Extensive/Unresectable Definition of Unresectable Disease Confined to the Liver Unresectable disease is defined by anatomic considerations Approximately 80 % of patients with CLM will present and/or functional factors. Liver resection must include the with unresectable disease In the past, the majority preservation of two contiguous segments of liver, the were treated with palliative chemotherapy, and the survival ability to preserve adequate vascular inflow, outflow, bili- was less than 1 year. Contemporary series that employ the ary drainage, and future liver remnant volume and func- combination of new drugs developed over the last 20 years tion. A margin-negative resection is also expected.
have shown marked improvements in both tumor response Some high-volume centers have shown that ultra- and survival []. Nevertheless, in patients treated exclu- selected candidates may undergo complete resection with sively with chemotherapy, the median survival is less than acceptable morbidity using non-conventional surgical 2 years, which is significantly lower than in patients treated techniques such as ex vivo, ante-situm, or ALPPS with complete resection, demonstrating that surgical resections []. The amount of liver remnant is resection provides the only chance for cure.
another factor that must be considered when defining who Several retrospective studies have analyzed the natural is a surgical candidate [Moreover, the quality of the history of patients with unresectable LOD. Bismuth et al.
liver, the number of cycles of chemotherapy received, ] evaluated 434 patients with CLM. The majority and the comorbidities are all important factors that must (n = 330, 76 %) were considered unresectable because the be considered when defining what an adequate liver authors considered that it was not possible to perform a remnant volume is. There is consensus that it is necessary complete resection and received systemic chemotherapy to have 25–30 % of the liver preserved as the future liver [5-fluorouracil (5-FU), folinic acid, and oxaliplatin (Ox)].
remnant volume when liver is normal and as high as Only 53 (16 %) patients had an adequate downsizing to 40 % in patients who have injured livers: those who have undergo resection. The majority required a major hepa- received extensive preoperative chemotherapy, have ste- tectomy (n = 37, 70 %) in one-stage (n = 46), and PVE atosis, or have diabetes mellitus []. Patients with dia- was uncommon (n = 5, 9 %). Importantly, after a median betes have a decreased rate of regeneration and therefore follow-up of 42 months, 34 (66 %) patients had recurrence should be evaluated carefully. Thus, every patient should in the liver, and 25 (47 %) patients had extrahepatic have a formal quantification of the future liver remnant recurrences. At last follow-up, 23 patients had died from before surgical exploration, and portal vein embolization disease and 19 were free of disease, but 14 of these 19 (PVE) should be considered in borderline cases [– patients required repeat liver resection after recurrence to A two-stage hepatectomy is another option in patients become NED. Five-year estimated overall survival (OS) with high risk of liver failure. In this procedure, a was 40 %. This series demonstrated that only a small compensatory liver regeneration after a first non-curative percentage (16 %) of patients with unresectable disease hepatectomy allows a second, potentially curative sur- will ever achieve surgical resection, and in this subset of gery. It has been observed that chemotherapy after PVE patients, multiple resections were needed to improve sur- does not decrease the hypertrophy of the remnant liver vival. Adam et al. [] updated the experience of the same nor increase the postoperative complication, but may be group and evaluated 1,439 patients treated over a period of useful to decrease the risk of developing new tumors 11 years (Fig. As in the previous experience, the during the month that it is necessary to wait before majority (n = 1,104, 77 %) presented with unresectable resection , •].
disease, based on the same author's definition ] and Curr Surg Rep (2014) 2:50 Fig. 1 Schematic representation of patients with CLM at the moment resection after chemotherapy will finally be free of disease at the of diagnosis. a About 75 % of patients will have unresectable disease.
moment of last follow-up. Thus, 1 of 30 (3 %) patients presenting b About 15 % will convert from unresectable to resectable disease with unresectable disease will be free of disease at the moment of last after chemotherapy; c only 25 % of those patients who converted to received systemic chemotherapy [5-FU plus leucovorin resections of that period). This subgroup clearly had an combined with Ox (70 %), irinotecan (Iri) (7 %), or both aggressive disease since the median number of tumors and (4 %)]. After a median number of ten cycles, 138 (12.5 %) size was 5 and 4.3 cm, respectively, and the CLM was patients responded adequately and underwent liver resec- diagnosed within 12 months of initial surgery in most tion (93 % with curative intent). Despite this, most patients patients. The majority (55 %) received neoadjuvant che- responded to first line chemotherapy, but 14 % required a motherapy (median = 7 months), and progression during second and 9 % a third line. This series also included 52 this treatment was documented in 28 %. The majority of (38 %) patients who presented with EHD, mainly involv- patients underwent an aggressive surgical treatment ing the lungs, and resection was performed in 41 of them.
[extended (45 %) or hemi (32 %) hepatectomy, and addi- Fifteen patients required a two-stage hepatectomy, 15 at tional resection/ablation of another hepatic lesion (46 %)].
least one ablation, and 9 % underwent PVE. After a mean Resection of EHD was performed in 18 patients, but for follow-up of 48 months, 111 out of 138 (80 %) patients local extension of the liver metastases in most of them.
Ninety-two percent received adjuvant chemotherapy.
(n = 12, 9 %), or both (n = 59, 43 %). A new hepatec- Importantly, most patients (57 %) recurred during the first tomy after liver recurrence was performed in 55 patients, year, and the median disease-free survival (DFS) from and a new extrahepatic recurrence was extirpated in 28 resection was 12 months. After a median follow-up of patients. At the moment of analysis, 99 patients had died 33 months, 51 % of patients had died of disease, 30 % and 25 were free of disease. Five-year DFS and OS were 22 were alive with disease, and 19 % had no evidence of and 33 %. Four factors (rectal primary, C3 metastases, disease. Five-year OS was 33 %, and patients who pro- preoperative CA19-9 [ 100 UI/l, and preoperative tumor gressed during chemotherapy had the worst survival. This size [10 cm) predicted OS, decreasing significantly from series included seven 5-year survivors, but all had recurred.
59 % (without any factor) to 0 % when 4 factors were This study outlines the natural history of those patients who present. Recently, Adam et al. [] focused their analysis on are not necessarily unresectable at the moment of diagno- 184 consecutive patients who presented with unresectable sis, but present with advanced disease.
disease, but underwent complete resection after tumor In another European study, Ardito et al. •] evaluated downsizing. As previously described, most patients had the chance of cure in patients who presented with unre- advanced disease (bilobar involvement in 76 %) and nee- sectable LOD. This series evaluated 61 patients without ded a median of ten cycles to convert. Despite this, most EHD and exclusively treated with liver resection. Most patients had a follow-up of 5 years or more, 112 (76 %) patients received irinotecan-based chemotherapy, and died from disease after this period, and only 24 (16 %) resectability was achieved after a mean number of 11 were considered cured (most after the first hepatectomy).
cycles. Thirty-one patients required a major hepatectomy, In another retrospective study, Komprat et al. [ which was associated with a PVE in seven, and nine evaluated 98 patients with four or more CLMs who patients were candidates for a two-stage hepatectomy underwent complete resection at Memorial Sloan-Kettering (completed in 5). Despite extensive treatment, only 45 Cancer Center (MSKCC) between 1998 and 2002 (17 % of (74 %) patients underwent an R0 resection. Forty-four Curr Surg Rep (2014) 2:50 Table 1 Selected series of patients included in retrospective studies Unresectable or aggressive Adequate response biology at presentation after chemotherapy Bismuth et al.
Hepatic = 34 (66 %) Extrahepatic = 25 (47 %) 138/1104 (12.5 %) Komprat et al.
Ardito et al.
*Only 54 (55 %) patients received chemotherapy before surgery (79 %) patients recurred, and a new resection was per- frequently used drugs are: 5-FU/leucovorin, oxaliplatin, formed in 15. After a median follow-up of 39 months, irinitecan, bevacizumab, and cetuximab.
5-year RFS and OS were 23 and 43 %, respectively. OS Alberts et al. [as part of the North Central Cancer was highly correlated with complete resection (68 months Treatment Group, evaluated the role of FOLFOX in in R0 resection). Despite 30 patients completing 5 years of patients with unresectable LOD. Forty-four patients were follow-up, only 11 were alive at the moment of analysis.
enrolled from 13 institutions, and 42 were treated. Eleven More recently, the University of Toronto evaluated 24 (26 %) patients had received previous treatment, and the patients who were considered initially unresectable. The main reason for unresectability was the number of lesions majority had advanced disease [bilobar disease (n = 23), in 19 (45 %) patients, which was assessed by a surgeon with experience in liver surgery, before starting the treat- tumor = 7 cm]. All patients received oxaliplatin/irinote- ment. Patients received chemotherapy biweekly until best can-based chemotherapy. Twenty out of 24 (83 %) response or progression/toxicity. Resectability was evalu- underwent an R0 resection, but most patients (n = 18) ated at 6 and 12 weeks after starting chemotherapy, and at recurred within 9 months. Three-year DFS and OS were 19 least two cycles of FOLFOX were planned after resection.
After a median number of ten cycles, a reduction of tumor These retrospective studies demonstrate the natural size was observed in 25 (60 %) patients, and 17 (40 %) history of patients who present with unresectable LOD and underwent surgical exploration. Complete resection was allow us to draw some conclusions about how best to treat obtained in 14 (33 %) patients, partial resection in 1, and 2 them (Table ). First, all patients should undergo com- were not resected. Ten (67 %) patients received adjuvant prehensive medical and surgical treatment to achieve chemotherapy. After a median follow-up of 22 months, complete resection at centers with expertise in complex recurrence was observed in 11 out of 15 (73 %) patients hepatic surgery. Second, despite our best efforts, only 17 % treated with resection. The majority were located in the of patients will achieve resection, and the majority of liver. Median time to recurrence was 19 months. At the patients will recur after resection, but the survival is sig- moment of analysis, 31 (74 %) patients had died. Median nificantly better than historical controls of patients treated OS was 26 months, and median OS for patients undergoing with chemotherapy alone. Thus, the effort should be resection was not determined because 67 % were alive at applied in every patient, and a prospective randomized trial is not needed and unethical at this point. Third, patients Ychou et al. ] in another phase II trial evaluated the who progress during chemotherapy have the worst prog- role of FOLFIRINOX in patients with unresectable LOD nosis. This situation defines a group of patients who have (defined by two liver surgeons and two radiologists) with tumors with aggressive biology and a high chance of pro- the aim of determining the rate of R0 resection. Every gression despite our best surgical efforts.
patient was treated every 2 weeks until completing 12cycles or having progression/toxicity. Thirty-four patientswere enrolled and evaluated, but 11 patients had minor Prospective Trials Including Systemic Treatment protocol violations that included the inclusion of one in Patients with Unresectable LOD patient with carcinomatosis, another with resectable dis-ease, five with lung metastases, and two who underwent Systemic chemotherapy is the most common treatment liver resection before entering in the study. Partial and utilized in the majority of patients who present with un- complete response was observed in 23 and 1 patients, resectable LOD. Different combinations of drugs have respectively. By contrast, only three patients had progres- been tested in prospective phase II trials, which have been sion of disease. The median time between treatment and designed to determine the rate of response. The most surgery was 4 months, and hepatic resection and/or Curr Surg Rep (2014) 2:50 ablation was performed in 28 (82.4 %) patients (liver closed prematurely (43 % of initially planed accrual) resection alone = 15, liver resection plus ablation = 10, because bevacizumab became a part of the standard of care and ablation alone = 3). Complete resection was per- for the authors. Median duration of treatment was 5 and formed in 9 out of 34 (26.5 %) patients. However, after a 5.1 months, respectively, and both treatments had a com- median follow-up of 31 months, 8 out of 9 (89 %) patients parable rate of toxicity. Both groups had a similar response recurred, mainly in the liver (n = 7). Median RFS and OS rate (XELIRI = 49 % vs. FOLFIRI = 48 %), probability were 13.9 and 36 months, and 2-year OS was 83 %.
of complete resection (XELIRI = 29 % vs. FOLFIR- Importantly, every patient had at least one adverse event I = 44 %, p = 0.16), and rate of R0 resection (24 %).
related to the treatment, and at least one grade 3 or 4 Complete radiologic response was observed in five (7 %) toxicity was observed in 26 (76.5 %) patients. The most patients treated with XELIRI and one (2 %) with FOLFIRI, frequent complications were neutropenia and diarrhea.
but they did not undergo resection. At the moment of Massi et al. ] evaluated the long-term outcome of 196 analysis, 37 % of patients treated with XELIRI and 26 % patients enrolled in three trials who presented with unre- with FOLFIRI did not have recurrence. Both treatments sectable disease and were treated with FOLFOXIRI fol- had a similar median PFS (XELIRI = 10.3 months vs.
lowed by radical surgery. The therapy was given for 12 FOLFIRI = 9.3 months, p = 0.78) and OS (XELIR- cycles or until evidence of progression/toxicity. This study I = 30.7 months vs. FOLFIRI = 16.6 months, p = 0.16).
included 73 (37 %) patients with liver metastases. This is However, it should be noted that there was a trend toward important because it evaluates the effectiveness of this better survival in the group treated with XELIRI. Since this treatment for metastatic colorectal cancer. Chemotherapy study was not powered to answer this question, a larger was effective in 138 of 196 (70 %) patients, but only 37 out phase III trial with an adequate statistical power is needed.
of 196 (19 %) patients underwent a surgery with curative Another phase II trial conducted by Zhao et al. ] intent. Adequate response that promoted complete resection focused only on the role of XELIRI in patients with un- was more common in patients with LOD (25 of 73 patients resectable LOD. Forty-eight patients were enrolled, and 47 with liver metastases, 34 %) and better than at other sites of were assessed for response. Twenty-nine patients had some metastatic disease (25 of 37 patients completely resected, grade of response, and 18 had a partial response. Surgical 68 %). The 37 patients who underwent surgical exploration exploration was performed in 23 patients (49 %) [R0 received a median number of 11 cycles, and the median resection = 20, incomplete resection = 2 (based on the time of preoperative chemotherapy was 5.5 months. Liver postoperative CT), no resection = 1]. After a median fol- resection included a major hepatectomy in 19 (52 %) low-up of 24 months, 13 out of 22 (59 %) resected patients patients, but 8 were also treated with radiofrequency abla- had recurred, mainly in the liver remnant. The median time tion. Complete pathological response was observed in four to progression for patients treated with complete resection patients with LOD. After a median follow-up of 67 months, was 23 months, while the median and 3-year OS was 31 out of 37 (84 %) patients recurred. PFS was 18 months 27.5 months and 29 %.
from study entry, but if we consider that the median time Wong et al. ] evaluated the role of a combination of between chemotherapy and surgery was 5.5 months, capecitabine, oxaliplatin, and bevacizumab in 45 patients patients recurred at a median time of 12 months after sur- who were not selected for upfront resection. Despite the gery. Importantly, the six patients who have not recurred fact that this trial included patients who were not neces- had LOD, suggesting that in this group of high-risk patients, sarily unresectable at the time of presentation, it is those with LOD have a better prognosis. A second resection important to know the relevance of this treatment in was performed in 11 out of 31 patients who presented with patients with a high risk of recurrence to better understand recurrence. Median and 5-year OS was 40 months and the biology of this disease. Ten of 15 (67 %) initially 42 %. However, the survival was better in patients with resectable patients (with their primary in situ) underwent LOD (median = 61 months, 5-years = 43 %). The best resection (R0 = 6 and R1 = 4). On the other hand, 12 of survival between the 196 patients evaluated was observed in 30 (40 %) initially unresectable patients converted to those patients with LOD who underwent R0 resection and resectable, but 8 underwent resection (2 other patients had had complete pathologic response (median = 64 months, complete response and 2 others were not surgical candi- 5-years = 75 %).
dates). Despite this positive result, only 3 out of 8 In another phase II trial, Skof et al. ] compared the underwent an R0 resection, and in total 18 out of 45 effectiveness of XELIRI (capecitabine plus irinotecan) vs.
(40 %) underwent a liver resection (R0 = 9, 20 %). The FOLFIRI (5-FU/LV plus irinotecan) in patients with un- authors report 1-year PFS and OS of 50 and 86 %, but it resectable LOD, with the aim of determining the rate of should be considered that the median follow-up was only response and resection. Forty-one patients were treated 12.5 months, which is a small interval to make a con- with XELIRI and 46 with FOLFIRI, but the study was clusion about survival. Thirty-eight grade 3 complications Curr Surg Rep (2014) 2:50 Table 2 Selected series of patients with unresectable liver disease treated in a prospective trial Number of patients treated Median OS (months) Alberts et al.
Ychou et al.
Takahashi et al.
OS overall survival related to cabecitabine/oxaliplatin and five grade 3 com- the liver, increasing the local activity, and decreasing the plications and two grade 4 complications associated with systemic toxicity. In this section of this review, we discuss bevacizumab were observed.
the most common regional treatment currently used; More recently, Takahashi et al. [] evaluated the role hepatic-arterial infusion pump (HAIP) chemotherapy and of modified FOLFOX in patients with unresectable LOD in directed radiotherapy are discussed.
a multicenter study (38 centers). The main objective was todetermine the rate of curative surgery. Thirty-six patientswithout ED and without a previous history of chemother- HAIP Chemotherapy apy with oxaliplatin/irinotecan were included. Mostpatients presented with advanced disease [more than 5 In patients with unresectable LOD, HAIP chemotherapy has tumors = 28 (78 %), 20 (56 %) patients had tumors larger the advantage that higher doses of chemotherapy can be used than 5 cm] and received six to eight cycles of FOLFOX.
without increasing the systemic toxicity ]. Different An additional six cycles were recommended after surgery.
modalities have been used to deliver the drugs, but Thirty-one (86 %) patients completed the treatment with a implantable pumps are the most common, since they are median of six cycles [partial response = 18 (50 %), stable associated with fewer complications when compared with disease = 12, progression = 4]. Fourteen out of 36 (39 %) other forms of delivery [Before inserting a pump, it is patients underwent surgical exploration, and 13 had an R0 important to rule out the presence of EHD and to define the resection. Survival was not mentioned in this study.
arterial anatomy. The main proponent of this technique in the Finally, Ji et al. [] evaluated 73 patients with unre- US has been MSKCC, which is the center with the greatest sectable LOD and K-RAS wild type, enrolled at eight experience with this treatment []. After exploring the Korean centers. Each patient received FOLFOX plus ce- abdominal cavity to exclude metastatic disease, a chole- tuximab every 2 weeks for a maximum of 12 cycles, and cystectomy and a complete ligation of all collateral arteries 46 completed the treatment. A partial response was to the duodenum, bile duct, pancreas, and stomach should be observed in 53 (73 %) patients, and 36 (49 %) underwent performed. The gastroduodenal artery (GDA) is identified, surgical exploration [R0 resection = 20 out of 73 (27 %), dissected out, and ligated distally to insert and place the including radiofrequency ablation in 6, R1 = 6, and catheter in the GDA-hepatic artery junction to decrease the R2 = 10]. Median time to progression in all patients and in risk of thrombosis. The catheter should be secured with two those treated with R0 resection was 9.8 and 14 months, silk ties and connected to the device located in the subcu- respectively. At the time of analysis, 56 (77 %) patients taneous pocket created. After finishing this procedure, the had progressed, and 23 (32 %) had died of disease. The catheter should be tested using fluorescein dye or methylene most common hematologic complication was thrombocy- blue to confirm bilobar perfusion and to rule out the presence topenia (49 %), while the most common non-hematologic of extrahepatic perfusion [ complication was skin rash (28 %). Table shows selected Postoperative complications are seen in 10–40 % of series of patients included in prospective trials.
patients and significantly decrease with the expertise andexperience of the surgical team. Allen et al. ] evaluatedthe complication rate of HAIP in 544 consecutive patients.
Regional Treatments One hundred twenty (22 %) patients had at least onecomplication related to the pump, and the incidence of The rationale for using regional therapies in patients with pump failure increased with the time (first year = 9 %; unresectable LOD is that the treatments may be focused in second year = 16 %). Early complications were usually Curr Surg Rep (2014) 2:50 related to the hepatic artery system and were frequently and patients could benefit from a combination of local and salvaged, while late complications were mainly related to systemic treatment.
the catheter. Complication rates were higher when there In another prospective trial, Kemeny et al. ] evaluated was variant arterial anatomy, catheter insertion into another the role of systemic chemotherapy combined with HAIP vessel (not the GDA), placement during the first half of a chemotherapy in patients with unresectable LOD. Twenty- study period, or when the surgeon had less experience (less one patients received HAIP chemotherapy with FUDR plus than 25 procedures). More recently, another study from systemic oxaliplatin and irinotecan, and 15 patients received MSKCC showed that the risk of biliary sclerosis associated HAIP chemotherapy with FUDR plus systemic FOLFOX.
with HAIP chemotherapy was very low (5.5 % in patients Nineteen out of 21 (90 %) patients of the first group had a receiving adjuvant therapy and 2 % in those with unre- partial response, 7 (33 %) underwent a liver resection, and 2 did not have residual tumor. Median time for liver and Despite multiple drugs being tested via HAIP chemo- extrahepatic progression was 16.4 and 16.9 months. Median therapy, floxuridine (FUDR) has been the most common survival was 35.8 months, and 2-year survival was 65 %.
and currently is the standard treatment. This drug has a Similarly, 13 out of 15 (87 %) patients had a partial response high extraction rate in the liver ]. Since most compli- in the second group. Median time for liver and extrahepatic cations of the FUDR affect the liver and/or bile ducts, liver progression was 9.4 and 10.8 months. Median survival was enzymes should be evaluated every 2 weeks to adjust the 22 months, and 2-year survival was 40 %. This promising dose. The use of concomitant dexamethasone has been study confirmed that the best response is obtained using a associated with a lower elevation of bilirubin levels and a combination of systemic and HAIP chemotherapy, and higher rate of partial and complete responses ].
suggested that the combination of oxaliplatin and irinotecan HAIP chemotherapy has been mainly used as adjuvant has the best response. Based on these findings, Kemeny et al.
therapy in patients with LOD who have undergone com- ] treated 49 patients with unresectable LOD with HAIP plete resection. Multiple randomized trials have shown a chemotherapy plus systemic chemotherapy with oxaliplatin benefit in improving survival in this setting [– and irinotecan. Twenty-three patients were chemotherapy- However, this discussion is limited to its role in unresec- naı¨ve at the moment of being included in this study and 90 % table disease.
had a clinical risk score C3. Forty-five out of 49 (92 %) Kemeny et al. [enrolled 162 patients and random- patients had a partial (84 %) or complete (8 %) response, ized 99 to receive systemic chemotherapy or HAIP with and 23 (47 %) patients underwent resection (R0 = 19, 3 FUDR. Patients with ED or resectable disease were not with complete response) after a median time of 7 months.
included in this trial. Forty-eight patients received HAIP Twelve patients required PVE, four patients underwent two- chemotherapy and 51 systemic chemotherapy. A signifi- stage surgery, and ten were treated with radiofrequency cantly better response was observed in patients receiving ablation in addition to surgery, confirming that these patients HAIP chemotherapy (50 vs. 19.6 %, p = 0.001). Thirty- usually need a complex surgical treatment. Median DFS for one patients initially treated with systemic chemotherapy patients undergoing resection was 7.6 months, but when all crossed over to the HAIP treatment. Interestingly, 27 out of patients were evaluated, median survival was significantly 48 (56 %) patients who initially received HAIP chemo- better in patients who were chemotherapy-naı¨ve compared therapy developed EHD compared with 19 out of 51 with those who had received chemotherapy previously (50.8 (37 %) patients initially treated with systemic disease vs. 35 months, p = 0.02).
(p = 0.09). Importantly, 11 out of these 19 (58 %) patients More recently, Ammori et al. ••] retrospectively developed EHD when they crossed over to HAIP chemo- evaluated the experience of MSKCC treating patients with therapy. Median time of progression in the HAIP group unresectable CLM who were treated with systemic and was 9 months and better than in the systemic group HAIP chemotherapy. This study is important in order to (5 months), p = 0.016), but liver progression was signifi- understand the natural history of this group of patients, cantly lower in patients receiving HAIP chemotherapy.
since most of them were treated outside a protocol. Three Median survival was 17 months in the HAIP group and hundred seventy-three patients were evaluated between 12 months in the systemic therapy (p = 0.4), but those 2000 and 2009. The majority had a clinical risk score C3, patients who were initially treated with systemic therapy bilobar disease, and C4 tumors. This study also included 60 and crossed over to HAIP chemotherapy had a better sur- (16 %) patients with EHD, but most of them were diag- vival (18 months) than those who did not cross over nosed during surgery, and only 18 patients had a previous (8 months). This study mainly showed that in patients with history of EHD (resected in 14 and anastomotic recurrence LOD, HAIP chemotherapy is associated with a better in 4). Two hundred ninety-six (79 %) patients had received response, but does not decrease the risk of developing chemotherapy previously (oxaliplatin = 199, iriniotec- EHD, as the natural history of this disease usually shows, an = 121, and bevacizumab = 121). Ninety-two (25 %) Curr Surg Rep (2014) 2:50 patients converted to complete resection/ablation after a catheter placement through the GDA. Eight (16 %) patients median time of 7.1 months. An exclusive liver resection had evidence of minor EHD at the moment of treatment.
was performed in 38 (41 %), while 46 (50 %) were treated Despite a CT scan performed 3 months after completing with resection plus ablation, and 8 (9 %) were only ablated.
the treatment that showed a reduction in tumor size in 32 Sixty-two out of these 92 (67 %) patients recurred after a patients, 23 (52 %) had developed EHD 6 months after median time of 16 months, and 14 underwent a new treatment. After a median follow-up of 25.5 months, 37 resection/ablation. Median survival for the conversion and patients had died of disease, most of them for progression non-conversion group was 59 months and 16 months, of EHD. Median OS was 9.8 months from treatment.
while 5-year survival was 47 and 6 %, respectively In another study, Gray et al. ••] randomized 70 (p = 0.001). At the moment of analysis, 38 patients were patients with bilobar and unresectable CLM to receive alive without evidence of disease, and 24 were alive with HAIP chemotherapy alone with FUDR or HAIP chemo- disease. Multivariate analysis showed that conversion to therapy with FUDR plus SIR-spheres. At the moment of resection, preoperative CEA 200 ng/ml, and being che- analysis, only four patients were alive. Patients receiving motherapy-naı¨ve were independent predictors factors of HAIP chemotherapy plus SIR-spheres had significant higher rate of complete or partial response (44 vs. 18 %, Unfortunately, there are no randomized trials to define p = 0.01), changes in tumor volume (50 vs. 24 %, the advantage in terms of survival between HAIP chemo- p = 0.03), and decrease of CEA levels (72 vs. 47 %, therapy and systemic chemotherapy in patients with unre- p = 0.004). In addition to this, patients treated with both sectable LOD. Most studies are phase II trials that evaluate modalities had a higher PFS, but the OS was similar the rate of response to this combination of treatments.
between the groups.
Since the best response was obtained with a combination of In a separate trial, Lim et al. ] evaluated 30 patients systemic and HAIP chemotherapy, including oxaliplatin who were enrolled in three Australian centers. It included and irinotecan, it is highly necessary to design and conduct patients with unresectable liver metastases previously a phase III trial with the aim of defining what is the best treated with 5-FU-based chemotherapy. Patients with EHD treatment to improve survival.
were included only if the liver was the dominant site of Studies from other institutions have been unable to metastasis. A decrease of more than 30 % of the target replicate the excellent results reported by MSKCC. Com- lesions at 2 months of follow-up defined partial response, plication rates in general have been higher. There are a while progression of disease was defined as an increase of number of potential reasons for this including: institutional more than 20 % of the target lesion. All patients had failed expertise, commitment to the protocol and the need for an 5-FU-based chemotherapy, and 22 (73 %) had failed oxa- institutional ‘champion,' individual expertise, the medica- liplatin/irinotecan regimens. Despite these failures, 21 out tion and dosages used, and ability to salvage and manage of 30 patients receive 5-FU concurrent with SIR-spheres.
complications when they do occur. The external validity of Partial response was observed in ten patients, but the the technology still needs to be confirmed at other expert response continued during the first year, and one patient with initial partial response had complete response at6 months of treatment. After a median follow-up of18 months, Directed Radiotherapy 8.3 months. Another 8 patients had stable disease, and 12progressed. Median PFS was 5.3 months in all patients, Selective internal radiation (SIR) spheres is a novel treat- and 9.2 months for patients who had achieved a partial ment that has been used in patients with CLM and unre- response. Four (13 %) patients had severe toxicity, defined sectable LOD [The microspheres contain Yttrium90, by gastric or duodenal ulcers that were managed medically.
which is a high-energy beta-emitting isotope that is em- More recently, Cianni et al. ] retrospectively evaluated bolized through the hepatic artery, delivering 200–300 Gy 41 symptomatic patients who were treated with SIR-spheres on average. This method has the advantage of delivering a between 2005 and 2008. Thirty-nine patients had bilobar higher dose to the liver without having liver toxicity metastases, and four (9.7 %) had EHD, involving porta compared with external beam radiation, which can only hepatis lymph nodes or bone. All patients had failed the first, deliver 30–35 Gy to the liver ]. Small studies have second, and third line of chemotherapy. Before starting the evaluated the role of SIR spheres in unresectable CLM, treatment, each patient had a complete evaluation of the mainly as third or fourth line of therapy.
vascular anatomy of the liver with angiography, and all Stubbs et al. ] treated 50 patients with extensive branches of the hepatic artery to the GI tract were coiled. In a CLM that were not considered for resection or ablation second step, a complete evaluation of possible arteriovenous between 1997 and 1999. Each patient underwent an HAI shunts from the hepatic artery to the pulmonary system or Curr Surg Rep (2014) 2:50 ectopic implantation into the GI tract was performed. The retrospective or small phase II trials that have been con- interval between lobar treatments was 4–6 weeks, and the ducted to determine the rate of response to local and/or possibility of re-treatment was evaluated after 8 weeks of systemic treatments. The lack of well-conducted phase III completing the initial treatment. Complications related to the trials does not allow us to define either the standard treat- procedure were low: five patients presented with mild ment or the impact on survival of current therapies for most abdominal pain and nausea after 12 h after procedure, one of these patients. In addition, it has been difficult to rep- patient had a medically treated acute cholecystitis 25 days licate these results at some expert centers. Thus, it is nec- after the procedure, two patients had gastritis 4 and 6 weeks essary to develop a multicenter phase III trial including after the procedure, and one patient had liver failure 40 days both systemic and local treatments. Since the combination after the procedure, which was the only major hepatic of three systemic drugs (FOLFOXIRI) associated with complication. Eight (4.8 %) patients had complete response, HAIP chemotherapy is the treatment that has shown the 17 (41.5 %) had partial response, 14 (36.2) had stable dis- best response into and outside the liver in phase II trials; a ease, and 8 (19.5 %) had progression of disease. At the phase III trial including two arms, FOLFOXIRI versus moment of analysis, ten patients were alive, and the others FOLFOXIRI plus HAIP chemotherapy, is needed not only died because of disease progression. Median OS was 1 year, to define the best treatment, but also to define the impact of and median PFS was 0.8 years.
all these therapies on survival.
In another study, Hong et al. ] utilized a different modality of treatment. They included 36 patients with liver Compliance with Ethics Guidelines dominant CLM treated with chemoembolization (TACE) Conflict of Interest Jean M. Butte, Chad G. Ball, and Elijah Dixon or Thera-spheres with Yttrium90. Twenty-one patients were declare that they have no conflict of interest.
treated with TACE, and 15 received Thera-spheres. At thetime of analysis, only five patients in the group treated with Human and Animal Rights and Informed Consent TACE were alive. Moreover, the follow-up after treatment does not contain any studies with human or animal subjects was too short to make assumptions about survival performed by any of the authors.
(6.3 months for TACE and 5.7 months for Thera-spheres).
Three patients died before 30 days and had disease pro- gression. The authors reported a median survival fromtreatment of 7.7 months for the TACE group and Recently published papers of particular interest have been 6.9 months for the Thera-spheres group.
Finally, Turkmen et al. [] evaluated 61 patients with unresectable metastases, and 23 patients had CLM. This •• Of major importance study does not describe the response obtained in this subsetof patients, but they reported a median OS of 14 months.
In summary, SIR spheres have been mainly used in 1. Siegel R, Naishadham D, Jemal A. Cancer statistics, 2013. CA Cancer J Clin. 2013;63(1):11–30.
patients with unresectable and extensive disease who have 2. House MG, et al. Survival after hepatic resection for metastatic received other modalities of treatment previously. The colorectal cancer: trends in outcomes for 1,600 patients during impact on survival is poor, and it should be used very at a single institution. J Am Coll Surg.
3. Butte JM, et al. Patterns of failure in patients with early onset (synchronous) resectable liver metastases from rectal cancer.
Cancer. 2012;118(21):5414–23.
4. Carpizo DR, et al. Liver resection for metastatic colorectal cancer in patients with concurrent extrahepatic disease: results in 127patients treated at a single center. Ann Surg Oncol. 2009;16(8): Patients with unresectable LOD represent the vast majority of individuals with CLM that are evaluated in a hepatob- 5. Adam R, et al. Rescue surgery for unresectable colorectal liver iliary unit. Many of them will not have a good response to metastases downstaged by chemotherapy: a model to predict long- systemic and/or local treatments and will progress and die term survival. Ann Surg. 2004;240(4):644–57 discussion 657-8.
6. Adam R, et al. Patients with initially unresectable colorectal liver from their disease. On the other hand, a small and highly metastases: is there a possibility of cure? J Clin Oncol.
selected subgroup of patients (12.5–20 %) will have enough response in the liver to achieve surgical exploration 7. Abdalla EK, et al. Improving resectability of hepatic colorectal and eventually complete resection. Unfortunately, most metastases: expert consensus statement. Ann Surg Oncol.
2006;13(10):1271–80.
patients treated with complete resection will recur at a 8. Adams RB, et al. Selection for hepatic resection of colorectal median time of 12 months and will need consideration for liver metastases: expert consensus statement. HPB (Oxford).
repeat resection or other therapy. Most studies are Curr Surg Rep (2014) 2:50 9. • Cardona K et al. Treatment of extensive metastatic colorectal 5-year survival. J Gastrointest Surg, 2013; 17(2): 352–9. ‘‘This cancer to the liver with systemic and hepatic arterial infusion study shows the possibility of cure in patients with unresectable chemotherapy and two-stage hepatic resection: the role of salvage disease, treated with chemotherapy and surgery.'' therapy for recurrent disease. Ann Surg Oncol 2013. This study 28. Alberts SR, et al. Oxaliplatin, fluorouracil, and leucovorin for analyzed the recurrence pattern, salvage rate, and survival after patients with unresectable liver-only metastases from colorectal two-stage hepatectomy combined with systemic and regional cancer: a North Central Cancer Treatment Group phase II study.
J Clin Oncol. 2005;23(36):9243–9.
10. Hemming AW, et al. Resection of the liver and inferior vena cava 29. Ychou M, et al. Tritherapy with fluorouracil/leucovorin, irino- for hepatic malignancy. J Am Coll Surg. 2013;217(1):115–24 tecan and oxaliplatin (FOLFIRINOX): a phase II study in colo- discussion 124-5.
rectal cancer patients with non-resectable liver metastases.
11. Lodge JP, et al. Ex vivo and in situ resection of inferior vena cava Cancer Chemother Pharmacol. 2008;62(2):195–201.
30. Masi G, et al. Long-term outcome of initially unresectable met- astatic colorectal cancer patients treated with 5-fluorouracil/leu- 12. Yamamoto Y. Ante-situm hepatic resection for tumors involving covorin, oxaliplatin, and irinotecan (FOLFOXIRI) followed by the confluence of hepatic veins and IVC. J Hepatobiliary Pancreat radical surgery of metastases. Ann Surg. 2009;249(3):420–5.
31. Skof E, et al. Capecitabine plus irinotecan (XELIRI regimen) 13. Malde DJ, et al. Inferior vena cava resection with hepatectomy: compared to 5-FU/LV plus Irinotecan (FOLFIRI regimen) as challenging but justified. HPB (Oxford). 2011;13(11):802–10.
neoadjuvant treatment for patients with unresectable liver-only 14. Mehrabi A, et al. Hypothermic ante situm resection in tumors of metastases of metastatic colorectal cancer: a randomised pro- the hepatocaval confluence. Dig Surg. 2011;28(2):100–8.
spective phase II trial. BMC Cancer. 2009;9:120.
15. Schnitzbauer AA, et al. Right portal vein ligation combined with 32. Zhao R, et al. A phase II study of irinotecan and capecitabine for in situ splitting induces rapid left lateral liver lobe hypertrophy patients with unresectable liver-only metastases from colorectal enabling 2-staged extended right hepatic resection in small-for- cancer. Jpn J Clin Oncol. 2010;40(1):10–6.
size settings. Ann Surg. 2012;255(3):405–14.
33. Wong R, et al. A multicentre study of capecitabine, oxaliplatin 16. Shindoh J, et al. Optimal future liver remnant in patients treated plus bevacizumab as perioperative treatment of patients with with extensive preoperative chemotherapy for colorectal liver poor-risk colorectal liver-only metastases not selected for upfront metastases. Ann Surg Oncol. 2013;20(8):2493–500.
resection. Ann Oncol. 2011;22(9):2042–8.
17. Huang SY, et al. Efficacy and safety of portal vein embolization 34. •• Takahashi T et al. Multicenter phase II study of modified for two-stage hepatectomy in patients with colorectal liver FOLFOX6 as neoadjuvant chemotherapy for patients with unre- metastasis. J Vasc Interv Radiol. 2013;. doi: sectable liver-only metastases from colorectal cancer in Japan: ROOF study. Int J Clin Oncol. 2013;18(2): 335–42. This trial 18. Shindoh J, et al. Safety and efficacy of portal vein embolization shows the role of FOLFOX in patients with unresectable, liver before planned major or extended hepatectomy: an institutional only disease.
experience of 358 patients. J Gastrointest Surg. 2013;. doi: 35. •• Ji JH et al. Prospective phase II study of neoadjuvant FOL- FOX6 plus cetuximab in patients with colorectal cancer and 19. Shindoh J, et al. Portal vein embolization improves rate of unresectable liver-only metastasis. Cancer Chemother Pharmacol.
resection of extensive colorectal liver metastases without wors- 2013; 72(1): 223–30. This trial shows the effect of FOLFOX plus ening survival. Br J Surg. 2013;100(13):1777–83.
cetuximab in patients with liver only, unresectable disease.
20. Shindoh J, et al. Analysis of the efficacy of portal vein emboli- 36. Kemeny N, Fata F. Hepatic-arterial chemotherapy. Lancet Oncol.
zation for patients with extensive liver malignancy and very low future liver remnant volume, including a comparison with the 37. Kingham TP, D'Angelica M, Kemeny NE. Role of intra-arterial associating liver partition with portal vein ligation for staged hepatic chemotherapy in the treatment of colorectal cancer hepatectomy approach. J Am Coll Surg. 2013;217(1):126–33 metastases. J Surg Oncol. 2010;102(8):988–95.
discussion 133-4.
38. Kemeny N, et al. Phase I trial of systemic oxaliplatin combination 21. Covey AM, et al. Combined portal vein embolization and neo- chemotherapy with hepatic arterial infusion in patients with un- adjuvant chemotherapy as a treatment strategy for resectable resectable liver metastases from colorectal cancer. J Clin Oncol.
hepatic colorectal metastases. Ann Surg. 2008;247(3):451–5.
22. • Fischer C et al. Chemotherapy after portal vein embolization to 39. Kemeny N, et al. Phase I study of hepatic arterial infusion of protect against tumor growth during liver hypertrophy before floxuridine and dexamethasone with systemic irinotecan for un- hepatectomy. JAMA Surg 2013. This study shows the impact of resectable hepatic metastases from colorectal cancer. J Clin chemotherapy on tumor growth after portal vein embolization.
23. Reddy SK, et al. Timing of multimodality therapy for resectable 40. Kemeny NE, et al. Conversion to resectability using hepatic synchronous colorectal liver metastases: a retrospective multi- artery infusion plus systemic chemotherapy for the treatment of institutional analysis. Ann Surg Oncol. 2009;16(7):1809–19.
unresectable liver metastases from colorectal carcinoma. J Clin 24. Brouquet A, et al. High survival rate after two-stage resection of advanced colorectal liver metastases: response-based selection 41. Allen PJ, et al. Technical complications and durability of hepatic artery infusion pumps for unresectable colorectal liver metasta- ses: an institutional experience of 544 consecutive cases. J Am 25. Bismuth H, et al. Resection of nonresectable liver metastases Coll Surg. 2005;201(1):57–65.
from colorectal cancer after neoadjuvant chemotherapy. Ann 42. Ito K, et al. Biliary sclerosis after hepatic arterial infusion pump Surg. 1996;224(4):509–20 discussion 520-2.
chemotherapy for patients with colorectal cancer liver metastasis: 26. Kornprat P, et al. Outcome after hepatectomy for multiple (four incidence, clinical features, and risk factors. Ann Surg Oncol.
or more) colorectal metastases in the era of effective chemo- therapy. Ann Surg Oncol. 2007;14(3):1151–60.
43. Kemeny N, et al. Hepatic arterial infusion of chemotherapy after 27. • Ardito F et al. Chance of cure following liver resection for resection of hepatic metastases from colorectal cancer. N Engl J initially unresectable colorectal metastases: analysis of actual Curr Surg Rep (2014) 2:50 44. Kemeny N, et al. Phase I trial of adjuvant hepatic arterial infusion 50. Gray BN, et al. Regression of liver metastases following treat- (HAI) with floxuridine (FUDR) and dexamethasone plus systemic ment with yttrium-90 microspheres. Aust N Z J Surg.
oxaliplatin, 5-fluorouracil and leucovorin in patients with resec- ted liver metastases from colorectal cancer. Ann Oncol.
51. •• Gray B et al. Randomised trial of SIR-Spheres plus chemo- therapy vs. chemotherapy alone for treating patients with liver 45. Kemeny NE, et al. Hepatic arterial infusion versus systemic metastases from primary large bowel cancer. Ann Oncol.
therapy for hepatic metastases from colorectal cancer: a ran- domized trial of efficacy, quality of life, and molecular markers 52. Lim L, et al. A prospective evaluation of treatment with Selective (CALGB 9481). J Clin Oncol. 2006;24(9):1395–403.
Internal Radiation Therapy (SIR-spheres) in patients with unre- 46. Kemeny N, et al. Intrahepatic or systemic infusion of fluorodeoxy- sectable liver metastases from colorectal cancer previously trea- uridine in patients with liver metastases from colorectal carcinoma.
ted with 5-FU based chemotherapy. BMC Cancer. 2005;5:132.
A randomized trial. Ann Intern Med. 1987;107(4):459–65.
53. Cianni R, et al. Selective internal radiation therapy with SIR- 47. •• Ammori JB et al. Conversion to complete resection and/or spheres for the treatment of unresectable colorectal hepatic ablation using hepatic artery infusional chemotherapy in patients metastases. Cardiovasc Intervent Radiol. 2009;32(6):1179–86.
with unresectable liver metastases from colorectal cancer: a 54. Hong K, et al. Salvage therapy for liver-dominant colorectal decade of experience at a single institution. Ann Surg Oncol.
metastatic adenocarcinoma: comparison between transcatheter 2013; 20(9): 2901–7. This study shows the natural history of arterial chemoembolization versus yttrium-90 radioembolization.
patients with unresectable liver metastases from colorectal can- J Vasc Interv Radiol. 2009;20(3):360–7.
cer treated initially with hepatic artery infusional chemotherapy.
55. Turkmen C, et al. Initial outcome after selective intraarterial 48. Abdalla EK, et al. Locoregional surgical and interventional radionuclide therapy with yttrium-90 microspheres as salvage therapies for advanced colorectal cancer liver metastases: expert therapy for unresectable metastatic liver disease. Cancer Biother consensus statements. HPB (Oxford). 2013;15(2):119–30.
49. Stubbs RS, Cannan RJ, Mitchell AW. Selective internal radiation therapy with 90yttrium microspheres for extensive colorectalliver metastases. J Gastrointest Surg. 2001;5(3):294–302.

Source: http://www.drjeanmichelbutte.cl/archivos/9f0215_Tratamientodelasmetastasishepaticasirresecables2014.pdf