OncoSurge: A Strategy for Improving Resectability With Curative Intent in Metastatic Colorectal Cancer
the Royal Liverpool University Hospital, Liverpoolayn, http://www.100md.com
University of Southampton, Southampton, UKayn, http://www.100md.com
Hopital Ambroise Pare, Boulogneayn, http://www.100md.com
Hopital Paul Brousse, Villejuifayn, http://www.100md.com
Evidis, Paris, Franceayn, http://www.100md.com
Mayo Clinic, Rochester, MNayn, http://www.100md.com
The University of Texas M.D. Anderson Cancer Center, Houston, TXayn, http://www.100md.com
L'Hospitalet de Llobregat, Barcelona, Spainayn, http://www.100md.com
University of Pennsylvania Cancer Center, Philadelphia PAayn, http://www.100md.com
Hopital Cantonal, Geneva, Switzerlandayn, http://www.100md.com
Marlene and Stewart Greenebaum Cancer Center, Baltimore MDayn, http://www.100md.com
Allegheny General Hospital, Pittsburgh PAayn, http://www.100md.com
University Medical Center, Nijmegenayn, http://www.100md.com
RAND Europe, Leiden, the Netherlandsayn, http://www.100md.com
Martin Luther University, Halle, Germanyayn, http://www.100md.com
ABSTRACTayn, http://www.100md.com
PURPOSE: Most patients with colorectal liver metastases present to general surgeons and oncologists without a specialist interest in their management. Since treatment strategy is frequently dependent on the response to earlier treatments, our aim was to create a therapeutic decision model identifying appropriate procedure sequences.
METHODS: We used the RAND Corporation/University of California, Los Angeles Appropriateness Method (RAM) assessing strategies of resection, local ablation and chemotherapy. After a comprehensive literature review, an expert panel rated appropriateness of each treatment option for a total of 1,872 ratings decisions in 252 cases. A decision model was constructed, consensus measured and results validated using 48 virtual cases, and 34 real cases with known outcomes.ll5\l, 百拇医药
RESULTS: Consensus was achieved with overall agreement rates of 93.4 to 99.1%. Absolute resection contraindications included unresectable extrahepatic disease, more than 70% liver involvement, liver failure, and being surgically unfit. Factors not influencing treatment strategy were age, primary tumor stage, timing of metastases detection, past blood transfusion, liver resection type, pre-resection carcinoembryonic antigen (CEA), and previous hepatectomy. Immediate resection was appropriate with adequate radiologically-defined resection margins and no portal adenopathy; other factors included presence of 4 or > 4 metastases and unilobar or bilobar involvement. Resection was appropriate postchemotherapy, independent of tumor response in the case of 4 metastases and unilobar liver involvement. Resection was appropriate only for > 4 metastases or bilobar liver involvement, after tumor shrinkage with chemotherapy. When possible, resection was preferred to local ablation.
CONCLUSION: The results were incorporated into a decision matrix, creating a computer program (OncoSurge). This model identifies individual patient resectability, recommending optimal treatment strategies. It may also be used for medical education.{h0, http://www.100md.com
INTRODUCTION{h0, http://www.100md.com
Colorectal cancer is the third most prevalent cancer in the world, after lung and stomach cancer. More than 940,000 cases occur annually worldwide and nearly 500,000 people will die from it each year.1 The most common site of metastases is the liver, involved in 15% to 25% of patients at the time of diagnosis.2,3 An additional 35% to 45% of patients will develop liver metastases during the course of their disease.2 Interestingly, 20% to 30% of patients with advanced colorectal cancer have liver-only metastases,4 whereas recurrences after resection of the primary tumor are confined to the liver in nearly 50% of cases.5{h0, http://www.100md.com
Hepatic resection is the only potentially curative option for liver metastases, with contemporary operative mortality rates as low as 1% to 2%.6-10 Five-year survival rates following resection range between 25% and 40% in most large series.6-11 Even though eligibility for liver surgery continues to expand,12-15 80% of patients with metastatic disease remain unresectable at presentation. The recent development of more effective chemotherapeutic agents such as oxaliplatin and irinotecan are capable of inducing significant tumor shrinkage,16-20 prolong survival in nonoperable disease,21 and also appear to allow an additional 10% to 20% of patients thought to be initially unresectable for cure to undergo metastectomy.22-27 Long-term survival rates for these patients previously treated with palliative intent alone are comparable to those of primarily resected patients.22,23
Although there are a number of published clinical staging systems that predict prognosis based on available preoperative parameters,28-30 the subset of patients eligible for potentially curative resection is increasing; however, this proportion is not yet clearly defined. Definitions of surgical resectability of liver metastases are changing at the same time that new chemotherapeutic agents are evolving. In addition, prognostic factors intended to determine groups of patients able to benefit from resection need to be further defined. Moreover, data on the relative roles of surgery and chemotherapy have generally come from retrospective analyses from single centers, without a clear definition of selection criteria for surgery or chemotherapy. These data need to be confirmed by large-scale prospective randomized trials. Furthermore, many scoring systems for resection of liver metastases with curative intent were published in the era before the widespread adoption of the use of irinotecan and oxaliplatin in the management of these patients and similar improvements in both diagnostic and staging radiographic imaging.16-28 Ongoing studies of adjuvant chemotherapy for patients with resected or ablated liver metastases in both North America (National Surgical Adjuvant Breast and Bowel project [NSABP] C-09) and Europe (European Organisation for Research and Treatment of Cancer [EORTC] studies 40983 and 40004) will not be reported for several years. This delay poses a major problem for the present management of patients with colorectal liver metastases, who worldwide are frequently under the care of general oncologists and surgeons.31
These key contemporary developments in surgical technique, radiographic imaging, and chemotherapy are moving the treatment of colorectal liver metastases to a new multidisciplinary level, involving surgical, medical, and radiologic subspecialists. Although such multidisciplinary teams are the norm in tertiary centers of academic excellence, access to such expertise is not usually the case for most busy primary- and secondary-level practices across the world. Although it is accepted that in the present Internet era, data on the use of the latest chemotherapeutic regimens are readily available, the integration of these regimens (with evolving definitions of surgical resectability into multistage treatment strategies) remain in reality beyond the capabilities of the busy, overworked general oncologist or surgeon, who may see relatively few patients with potentially curable liver metastases./':, http://www.100md.com
The purpose of this project was to undertake an appropriateness study (a validated and accepted scientific method) using the RAND Corporation/University of California, Los Angeles Appropriateness Method (RAM)32-36 and to create a therapeutic decisional model (OncoSurge) to assist general medical oncologists and surgeons in defining optimal treatment strategies for individual patients with colorectal liver metastases.
METHODS\;2, http://www.100md.com
For the purpose of this study, the RAM was employed to assess the appropriateness of different treatment sequences for specific patient profiles.\;2, http://www.100md.com
The RAM is a statistically validated method32,33 that is a modified Delphi process.34 In using the RAM, experts overlay their consensus judgments on the available scientific evidence to assess the appropriateness of treatments. The model considers the average patient within each indication category, treated by the average provider in the average hospital.32-36 For this purpose the term "average" is explicitly defined clearly within a scientific context.32-36 The concept of appropriateness refers to the relative weight of the benefits and risks of a medical procedure. A procedure is "appropriate" for a given indication when expected benefits for the patient outweigh possible negative consequences by a sufficient margin. It is accepted that, in reality, the definition of a "sufficient margin" must be based on a synthesis of the perceptions of the individual patient, their clinician(s) and their respective individual thresholds for taking risks. Therefore, to be of clinical use, the decision model must rank and stratify benefit/risk and benefits ratios for each potential given treatment strategy in each potential individual clinical presentation. A procedure is "inappropriate" when possible negative consequences exceed benefits and "uncertain" when benefits and risks are nearly equal, or if there is disagreement within the expert panel on risk/benefit.34-36
The appropriateness of different treatments was scored on a scale of 1 (highly inappropriate) to 9 (clearly appropriate) against each clinical presentation. For the purpose of analysis, ratings from 1-3 for any given treatment are considered inappropriate, 4 to 6 represented uncertainty on the part of the expert rater, and 7 to 9 meant that the expert rated the given treatment as appropriate. The group rating was taken as the median judgment of the panelists, if there was no statistically defined disagreement among the panel members.32-36 Uncertainty here can mean that the risks and benefits are approximately equivocal, that the experts are uncertain, or that there is disagreement among the experts; in any of these instances, uncertainty is a call for further clinical research. Bimodal disagreement (1 to 3 v 7 to 9) meant that a group of the experts misunderstood the proposed treatment options, and in all cases where this phenomenon occurred during the first round of ratings, changed to unimodal agreement (1 to 3 or 7 to 9) after the discussion preceding the second round of ratings. Unimodal agreement between 4 and 6 and linear disagreement (ranging between 1 and 9) represents true uncertainty on the part of the expert panelists and identifies areas for future clinical research.
Uncertainty here can mean that the risks and benefits are approximately equivocal, that the experts are uncertain, or that there is disagreement among the experts; in any of these instances, uncertainty is a call for further clinical research. For the purpose of this study, an international panel of 16 oncologists, surgeons and radiologists with large experience in the diagnosis and treatment of liver metastases participated in the study (Table 1). As a first step, the experts were provided with a systematic review prepared specifically for this study by two of the authors of this study and funded independently by the British Cancer Research Campaign (now CRUK).37 This review was intended to summarize scientific evidence and list major indications for surgical resection of colorectal liver metastases. Any patient could be assigned to only one indication, but considered for each possible treatment.!y%nf, 百拇医药
A preliminary panel meeting established three treatment segments to be considered: liver resection, local ablation, and chemotherapy. The panel recognized that treatment by tumor ablation is presently in evolution and its utility in comparison to chemotherapy alone remains be proven. For the purposes of the RAM, the panel agreed that ablation could be considered for low-volume, liver-only disease in a patient deemed to be either unsuitable for liver resection by an experienced liver surgeon, or as an adjunct to liver resection (see the Results section). For each treatment segment, all relevant possible patient characteristics were identified and included in a comprehensive list of indications. Negative characteristics considered to be exclusively inappropriate for any treatment segment strategy were also compiled. Panelists performed the first ratings, and these ratings were subsequently analyzed for appropriateness and for areas for disagreement.
A second meeting was held to present and discuss the results of the first round of ratings, and modifications were made to the list of both treatment segment options and indications. Panelists then privately rated each of the three treatment segments separately. The second round of ratings provided the basis for assessing the appropriateness of the treatment segments.cv, 百拇医药
The different treatment strategies that resulted from the second round of ratings were incorporated into a decision matrix in the form of a computer program, the OncoSurge decision model. Validation of the model by the expert panel was accomplished using 48 representative virtual theoretical cases using the real decision characteristics identified by the appropriateness analysis. These 48 cases were created specifically to ensure that the expert panel had considered the full range of all potential clinical presentations, making the validation process span the full range of potential patients. Furthermore, the validation process using the 48 virtual theoretical cases was reinforced by the panelists using 34 real cases from their panelists' own clinical practices (with known outcomes). The purpose of the virtual theoretical cases was to test the rigor accuracy of the methodology and the rating process, with the intention of measuring consensus (appropriateness or inappropriateness of each treatment strategy) and disagreement (uncertainty) within the panel. This methodology has been utilized and tested in previous uses of the RAM in health care decision making.32-36,41
Each case was validated by comparing the group modal treatment choice with the second-round appropriateness ratings within each treatment segment of the case. If that option had been rated appropriate, then the group was scored as consistent for the treatment segment for that case. If the option had been rated inappropriate, then the group was scored as inconsistent. If the appropriateness rating for the choice was uncertain, then the group was rated consistent only if "no other choice" had been rated appropriate. In other words, a consistent choice was one in which there was no option considered better by the appropriateness ratings.}f6|, 百拇医药
RESULTS}f6|, 百拇医药
Initial Lists of Indications and Treatment Options}f6|, 百拇医药
On the basis of the systematic literature review, separate lists of indications and treatment options were drawn up for each of the three treatment segments considered for the study: resection, local ablation, and chemotherapy. A list of absolute contraindications to resection with curative intent was also defined, including the following clinical conditions: unresectable extrahepatic disease (such as peritoneal carcinomatosis, multifocal lung metastases, and distant lymph nodes in the preaortic and celiac axis N2 nodal basin for the liver, bone, or brain metastases), extensive liver involvement (more than six liver segments involved, > 70% liver invasion or all three hepatic veins involved), major liver insufficiency, or Child B or C liver cirrhosis with complications, and patients unfit for or declining surgery. Although it was accepted that tumor involvement at the confluence of two hepatic veins, inferior vena cava, or contralateral portal pedicle might indicate that only an R1 rather than R0 resection was possible, this was not considered an absolute contraindication to liver resection. Although patient age, stage of primary tumor, timing of detection of metastases, blood transfusion during surgery, type of resection (anatomic versus nonanatomic), presurgical carcinoembryonic antigen level, and previous hepatectomy are known to be relative prognostic factors for survival,28-30,37 they were not considered in the decision making for different treatment strategies.
Although triphasic computed tomography scanning of the liver is now used routinely in detecting metastatic liver disease, the assessment of the extent of resection necessary to achieve negative resection margins will be available only to experienced liver surgeons working closely with their hepatobiliary radiologists (usually in tertiary center). For example, if a trisegmentectomy was required but would leave a small functioning hepatic remnant, then this would be important preoperative information to justify neoadjuvant chemotherapy, with or without preoperative portal vein embolization. However, decisions at this strategic level are the province of the specialist center, and beyond the scope of this project, which was to advise physicians working at the primary and secondary levels of care (who do not deal frequently with advanced colorectal cancer), on the appropriateness of complex interdependent treatment strategies.qw+]o$p, http://www.100md.com
The indications for rating each treatment segment were drawn from the clinical characteristics shown in Table 2. A total of seven characteristics were considered, of which five had two possible values and two had three possible values. Therefore, there were potentially 288 (3 x 3 x 2 x 2 x 2 x 2 x 2) indications for each treatment option. The initial therapeutic options considered for each treatment segment are presented in Table 3. For resection, eight different possibilities were considered, depending on whether surgery was combined with preoperative chemotherapy and on whether surgery was performed (or not) in one or two stages. For local ablation, the panelists considered cryosurgical and radiofrequency ablation together, yielding four local ablation possibilities differing with respect to their possible combination with resection. For chemotherapy, 12 different administrations were considered, depending on the goal and timing of chemotherapy in the treatment strategy (palliative, preoperative, postoperative), and the chemotherapy regimen when chemotherapy was to be considered (fluorouracil plus leucovorin [FU/LV], FU/LV plus oxaliplatin, FU/LV plus irinotecan, FU/LV plus both oxaliplatin and irinotecan). Thus, there were 384 (8 x 4 x 12) different treatment sequences to consider. If each indication were assessed for each treatment sequence, there would have been 110,592 (288 x 384) different ratings to be made, clearly an impossible situation.
The ratings task was brought back to tractability by employing two different tactics.38 First, instead of assessing each sequence (combination of resection, local ablation, and chemotherapy) as a whole, appropriateness was assessed for each segment separately, and later for different sequences if each segment of the sequence was itself appropriate. Second, within each treatment segment, clinical characteristics which were not considered germane were identified (Table 2), thereby providing a large reduction in indications.4, 百拇医药
More specifically, for resection, tumor size was not considered germane,39 so only 144 indications were rated for each of the eight resection possibilities, for a total of 1,152 ratings. For local ablation, radiologically suspicious portal lymph node involvement and lobar (uni- v bilobar) involvement were not considered germane, so only 72 indications were rated for each of the four local ablation possibilities, for a total of 288 ratings. For chemotherapy, at the time of these discussions (August 2003) data on biologic therapies were not sufficient enough to be included as treatment options. Similarly, the use of regionally delivered hepatic chemotherapy was confined largely to the United States, and was therefore not considered specifically. Radiologically-defined resection margins, portal lymph node involvement and lobar involvement were not considered relevant to the decision concerning chemotherapy options, so only 36 indications were rated for each chemotherapy possibility, for a total of 432 ratings.
Therefore, instead of the total possible 110,592 ratings, there were only 1,872 (1,152 + 288 + 432) to be performed by each panelist. This is well within the range of reasonableness, and indeed, the panelists complied with the task. However, as shown in Table 2, not all characteristics were used for each treatment segment.ko.p@, 百拇医药
Results of the First-Round Ratingsko.p@, 百拇医药
The first-round ratings were examined, less for the description of appropriateness for each treatment segment than for the presence of inter-rater disagreement, logical consistency, and whether the clinical characteristics made a difference in appropriateness.ko.p@, 百拇医药
For the resection segment, there was disagreement in 86 (7.5%) of the 1,152 ratings indications. However, this is a possible overstatement of consensus, because treatments were scored as rated inappropriate for 70% of the ratings indications. Thus, the two-stage procedure was always inappropriate, as well as surgery in the presence of portal lymph node involvement. Moreover, the ratings for immediate resection without previous chemotherapy and resection following optimal chemotherapy were almost identical. Similarly, the lymph node status of the primary colorectal tumor made no difference in appropriateness for liver resection.
For the local ablation segment, there was disagreement in 29 (10.0%) of the 288 indications. Again, statistics overstated true consensus, because the local destruction was considered appropriate in only 22 of 288 indications. Again, the lymph node status of the primary tumor was not taken into account for ratings.f$7[.7, 百拇医药
For the chemotherapy segment, the rate of disagreement was very high, in 30 (2.8%) of 144 ratings indications. In addition, many other indications had dispersions that were close to the statistical border for disagreement. Consequently, it was decided to entirely rebuild the indications structure for this treatment segment.f$7[.7, 百拇医药
Final Indicationsf$7[.7, 百拇医药
On the basis of the results of the first round of ratings, the panelists made major revisions to the indications structure, changing both the treatment options to be assessed and the indications matrix. All of these changes were in the direction of simplifying potential treatment strategies.f$7[.7, 百拇医药
As with the first round of ratings, second-round ratings were performed separately for each treatment segment of an overall treatment strategy encompassing resection, local ablation and chemotherapy. Regarding the intention to perform liver resection, the panel recommended to first resect the primary tumor before performing the liver resection. The final decision to proceed with potentially curative treatment of liver metastases assumed that the primary tumor could be removed completely (R0 resection of primary tumor) (including local lymph node metastases). The panel accepted that occasionally a patient will present with extensive but potentially resectable liver disease in the presence of a small and asymptomatic primary tumor. However, such presentations are uncommon, and the optimal treatment strategy for such patients would ideally reside within a tertiary center. Staged liver resections and the combination of portal vein embolization with surgery had to be decided by the performing liver surgeon at the time of surgery. As such these factors were not a consideration relating to the decision to refer the patient to the specialist hepatobiliary center with the intention to operate, and therefore were not part of the indications structure presented for rating. Chemotherapy could be used before resection, either with the intent to render unresectable tumors resectable or to improve the status of initially resectable tumors. Thus, three resection options were finally considered: immediate resection with no previous chemotherapy; resection following preoperative chemotherapy, independent of the tumor response; and resection only after tumor shrinkage with preoperative chemotherapy.
All clinical characteristics considered for the first round, except for lymph node status of the primary tumor, were employed (Table 2), so that there were 48 indications for each of the three new treatment options, or 144 total ratings for resection.'5, 百拇医药
Regarding the intention to perform local ablation, the four treatment options shown in Table 3 were not modified, apart from a decision to restrict the maximum diameter of the largest treatable liver metastasis to < 4 cm. The panel accepted that there was no consensus on the maximum number and size of liver metastases that could be treated using this technique. Similarly, the panel accepted that this treatment could be used repeatedly, depending on the response to previous tumor ablations. Therefore the panel agreed to rate recommendations on ablation therapy on intention to treat at the outset, but with the proviso that if referred on, then the final decision would reside with the specialist liver resection center (allowing the possibility of resection instead of ablation after review at the specialist center). Thus, the second round considered 24 ratings for each treatment option, for a total of 96 ratings.
Chemotherapy strategies were different, depending on palliative intent (where, in the view of the panel, there was little or no hope of reducing tumor burden to the point of resection with curative intent) or in combination with surgery in the preoperative and/or postoperative setting. The choice of postoperative chemotherapy also depended on the curative (R0) intent or not (R1) of the resection. Finally, nine indications resulted from the combination of these clinical conditions: preoperative chemotherapy—unresectable metastases, and unlikely to be rendered resectable (> 70% liver replacement, > 6 segments tumor involvement, unresectable extra-hepatic disease) and unresectable metastases, possibly rendered resectable (all three hepatic veins involved, radiologic involvement of portal pedicle lymph nodes, radiologic concern about resection margins); postoperative chemotherapy—initially resectable metastases, R0 resection, without preoperative chemotherapy, R1 resection, without preoperative chemotherapy, R0 resection, after effective (regression or stabilization) preoperative chemotherapy, R1 resection after effective preoperative chemotherapy, R0 resection, after ineffective (tumor progression while on treatment) preoperative chemotherapy, and R1 resection, after ineffective preoperative chemotherapy.
Of the initial clinical characteristics, only the presence of treatable extrahepatic disease was considered. As a consequence, there were 18 indications rated for each of four generic chemotherapy regimens, for a total of 72 ratings.k7, 百拇医药
For each treatment segment, panelists privately rated the new indications structure and these 312 (144 + 96 + 72) second-round ratings provided the basis for assessing the appropriateness of the treatment segments. These second-round ratings provided the basis for assessing the appropriateness of the treatment segments.k7, 百拇医药
Analysis of the Second Round of Ratingsk7, 百拇医药
The first step of the second round of ratings was to analyze the disagreements among the panelists. For all three treatment segments, disagreements were low: 11 of 144 ratings indications (7.6%) for resection, one of 96 ratings indications (0.9%) for local destruction, and two of 72 ratings indications (2.8%) for chemotherapy. The decrease in disagreement is due to the revision of the indications list that evolved from discussion during the second round of ratings. Therefore, analysis of the appropriateness rating was performed using a classification tree technique. Classification was based on the splitting of the data into increasingly homogeneous subsets, separating the indication-treatment combinations rated as appropriate from those rated inappropriate or uncertain. This classification was based on a dichotomous analysis of the results (classification and regression tree analysis), using a classification risk < 10% for appropriateness.38 Each split was conditioned by an optimal cutoff in the levels of a single factor and the process was continued as long as it was feasible or statistically significant.
For resection, immediate liver resection depended primarily on the presence of adequate radiologic resection margins. The panel accepted that there was no universally agreed upon definition of adequate resection margins and that the previously accepted standard of at least 1 cm has now been largely discarded.28-30,37,39 Therefore for the purpose of the study, the exact definition of acceptable resection margins (as with the type of liver resection) would be left to the expert hepatobiliary surgical team. In cases of radiologically inadequate resection margins on preoperative liver scans, immediate resection was considered inappropriate. Immediate resection depended secondarily on the radiologic absence of portal lymph node (PLN). If there were adequate resection margins with PLN involvement, immediate resection was inappropriate. Numbers of metastases and lobar involvement were additional criteria. In cases of adequate resection margins without PLN involvement, immediate resection was appropriate if the number of metastases was 4. If there were more than 4 metastases, immediate resection was appropriate or uncertain only if the metastases were unilateral, and inappropriate if they were bilateral. A decision to resect was considered appropriate after preoperative chemotherapy, independent of tumor response, primarily if the number of metastases was 4 and secondarily if there was unilobar involvement. If the number of metastases was > 4 with bilobar liver involvement, even if immediately amenable to trisegmentectomy, resection was considered appropriate only after tumor shrinkage with chemotherapy.
Although triphasic CT scanning is now used routinely in detecting liver metastases, the ability to assess the extent of resection required to achieve negative resection margins, and consequently the potential indication for neoadjuvant chemotherapy, will only be available to experienced liver surgeons working closely with their hepatobiliary radiologists and oncologists within a multidisciplinary team. The panel concluded that worldwide, most general oncologists and surgeons who do not deal with metastatic colorectal cancer on a day-to-day basis, would not feel comfortable in making decisions of this degree of complexity at this strategic level.}+hyp6-, 百拇医药
For local ablation, few indications (9.4%) were considered appropriate, generally if there were fewer than four small metastases in a patient for whom resection was not considered appropriate because of poor general condition or other excessive surgical risk. Whenever possible, resection was preferred to local ablation.}+hyp6-, 百拇医药
For chemotherapy, postoperative chemotherapy was indicated in all patients who received preoperative chemotherapy, using the same regimen if this was effective in the preoperative setting or changing the chemotherapy combination if this was ineffective in the preoperative setting. Analysis of chemotherapy regimens when chemotherapy was indicated is summarized in Table 4. By the end of 2003 (time of the second panel meeting), FU/LV alone was rarely rated appropriate in any given circumstance (preoperative, post R0 resection, post R1 resection).16,17,20-25 The three regimens using FU/LV in combination with oxaliplatin and/or irinotecan were generally appropriate, except following complete resection (R0) when postoperative chemotherapy was perceived as uncertain instead of positive. Differences between the choice of oxaliplatin and irinotecan were small, with a slight preference in favor of oxaliplatin. A triple combination of FU/LV with both oxaliplatin and irinotecan (which at the time the panelists met was an unlicensed regimen) was restricted to incomplete resection (R1).
Validation\62*[s, 百拇医药
The validation process began with an examination of the 48 virtual theoretical cases that covered all the factors previously identified in determining treatment strategy. Of these, 47 were validated, and one was not. That single case was related to the recommendation of an uncertain treatment when an appropriate one was available. Looking next at the 47 remaining cases, for four of these the local ablation choice was inconsistent with the appropriateness ratings. Again, all of the inconsistent ratings were related to the choice of an uncertain treatment when an appropriate one was available. Finally, with regards to the choices of chemotherapy for the 43 remaining cases, only six of them were inconsistent, and again because an uncertain choice was preferred to an appropriate one. Thus, looking at the entire treatment strategy, 37 (77%) of 48 were validated. The panel therefore identified uncertainty at the present time for an appropriate treatment strategy in 23% of clinical scenarios.\62*[s, 百拇医药
DISCUSSION\62*[s, 百拇医药
The management of patients with metastatic colorectal cancer has improved significantly over the last few years. Medical and surgical advances are the foundation of a multidisciplinary approach for patients with colorectal liver metastases. This approach is now mainstream practice in most major academic centers. However, in worldwide day-to-day practice, this integrated approach is still not used in many patients who currently might be curable with optimal multidisciplinary therapy.31 With the OncoSurge concept, emerging developments in surgery, chemotherapy and radiology to optimally define the management for colorectal liver metastases are brought together into an integrated, combined modality framework.
The present study has attempted to address the question of appropriateness of three treatment options: resection; local ablation; and chemotherapy, used alone or in combination, depending on each patient's specific individual tumor characteristics. In combination treatment strategies, each sequence depended on the response to the previous therapy strategy.o{, http://www.100md.com
The statistically validated RAM32-36 was used to assess the clinical/radiological characteristics of the patients ("indications") and to rate each treatment approach according to its appropriateness. Previous uses of the RAM in health care decision making have focused on clinical conditions for which the assessment of the appropriateness of a particular treatment has been relatively straightforward, such as prophylaxis of deep vein thrombosis, the use of ACE 2 inhibitors in the management of hypertension, the role of surgery in lumbar disc prolapse, emergency room triage, and the use of high-dose chemotherapy in the treatment of lymphoma.40-42 None to date used the segmentation tactic employed here. In the present model, both the indications and the possible treatment segments were multiple. The number of indications for treatment multiplied by the number of individual treatment strategies could easily result in literally millions of ratings.
The OncoSurge study overcomes these difficulties by adapting the RAM to consider limited treatment strategies composed of treatment segments. This study has demonstrated the feasibility of applying the concept of appropriateness to the assessment of complex treatment strategies.42 Similarly, other non-RAM Delphi-panel analyses have addressed treatment strategies ranging from the management of low back pain,43,44 through emergency room triage,45 to the use of high-dose chemotherapy in the treatment of lymphoma46; again, none to date used the segmentation tactic employed in the present study.*d*, 百拇医药
Clearly, important limitations emerge when considering what constitutes "appropriate" care. Answers to this question depend on which clinicians are asked, where they work and live, what weight is given to the different types and levels of evidence and end points, the weight given to the preferences of patients and their families, the level of resources in a given health care system, and the prevailing values of both the health system and the society in which it operates (especially relevant in the setting of a national health care system). The RAM methodology draws heavily on expert opinions, which can vary with use between different sets of panels.47 It is accepted that without extremely rigorous application, variability can be allowed to occur within all the standard steps of the RAM,48 and this variability can explain why the sensitivity and specificity of this method might be less than perfect.49
Moreover, when clinical decisions are being made, both patient and family requests (such as proximity of treatment center to home) and physician factors (relationships between oncologists and surgeons) are very important influences. Hence an individual clinician's approach to a particular medical condition is molded by a complex interplay of their understanding of the scientific evidence, clinical circumstances, and personal values or preferences, which are not adequately evaluated by the RAM. Even for patients with colorectal liver metastases (where the decision-making strategies are potentially life saving), all these issues come into play when formulating a treatment plan.:i, 百拇医药
The OncoSurge decision model allows a clinician to derive a choice of treatment strategy for an individual patient with colorectal liver metastases and compares this choice with the expert's view. Thus, the OncoSurge study combines the best available scientific evidence with the collective judgment of worldwide experts to yield a statement regarding the appropriateness of performing liver resection and/or local destruction, with or without chemotherapy for each patient. The methodology used to reach these conclusions can be applied equally efficiently across health care systems worldwide.50
The OncoSurge decision model identifies and individualizes patient resectability and makes recommendations for the optimal treatment of colorectal liver metastases, and may be used as an innovative approach for therapeutic decision support and medical education. Multimodality therapy is fast becoming the mainstay in the management of cancer. This methodology now requires further exploration in other indications, such as rectal cancer or adjuvant therapy for breast, lung, and pancreatic cancer.er-, http://www.100md.com
Authors' Disclosures of Potential Conflicts of Interester-, http://www.100md.com
Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.er-, http://www.100md.com
NOTES
Supported by an unrestricted educational grant from the Sanofi-Synthelabo Group.p;, 百拇医药
Presented previously at the 40th Annual Meeting of the American Society of Clinical Oncology, New Orleans, LA, June 5-8, 2004; the 29th Annual Meeting of the Euproean Society of Medical Oncology, Vienna, Austria, October 29-November 2, 2004; and the American Society of Clinical Oncology Gastrointestinal Cancers Symposium, Hollywood, FL, January 27-29, 2005.p;, 百拇医药
The results of this study have not been subject to approval by the Sanofi-Synthelabo Group. The recommendations given in this study are solely the responsibility of the authors and do not represent the views of the Sanofi-Synthelabo Group.p;, 百拇医药
Authors' disclosures of potential conflicts of interest are found at the end of this article.p;, 百拇医药
REFERENCESp;, 百拇医药
WHO. Global cancer rates could increase by 50% to 15 million by 2020. http://www.who.int/mediacentre/releases/2003/pr27/en/print.htmlp;, 百拇医药
Kemeny N, Fata F: Arterial, portal or systemic chemotherapy for patients with hepatic metastasis of colorectal carcinoma. J Hepatobiliary Pancreat Surg 6:39-49, 1999
Seifert JK, Junginger T, Morris DL: A collective review of the world literature on hepatic cryotherapy. J R Coll Surg Edinb 43:141-154, 1998]7t, 百拇医药
Borner MM: Neoadjuvant chemotherapy for unresectable liver metastases of colorectal cancer: Too good to be true? Ann Oncol 10:623-626, 1999]7t, 百拇医药
Fong Y, Cohen AM, Fortner JG, et al: Liver resection for colorectal metastases. J Clin Oncol 15:938-946, 1997]7t, 百拇医药
Adson MA: Resection of liver metastases: When is it worthwhile? World J Surg 11:511-520,1987]7t, 百拇医药
Wilson SM, Adson MA: Surgical treatment of hepatic metastases from colorectal cancers. Arch Surg 111:330-334,1976]7t, 百拇医药
Rees M, John TG: Current status of surgery in colorectal metastases to the liver. Hepatogastroenterology 48:341-344, 2001]7t, 百拇医药
Hohenberger P: Colorectal cancer: What is standard surgery? Eur J Cancer 37:S173-S187, 2001 (suppl 7)]7t, 百拇医药
Papadimitriou JD, Fotopoulos AC, Prahalias AA, et al: The impact of new technology on hepatic resection for malignancy. Arch Surg 136:1307-1313, 2001
Malafosse R, Penna C, Sa Cunha A, et al: Surgical management of hepatic metastases from colorectal malignancies. Ann Oncol 12:887-894, 2001}, 百拇医药
Scheele J, Stang R, Altendorf-Hofmann A, et al: Resection of colorectal liver metastases. World J Surg 19:59-71, 1995}, 百拇医药
Gayowski TJ, Iwatsuki S, Madariaga SR, et al: Experience in hepatic resection for metastatic colorectal cancer: Analysis of clinical and pathological risk factors. Surgery 116:703-711, 1994}, 百拇医药
Fortner JG, Silva JS, Golbey RB, et al: Multivariate analysis of a personal series of 247 consecutive patients with liver metastases from colorectal cancer: Treatment by hepatic resection. Ann Surg 199:306-316, 1984}, 百拇医药
Doci R, Gennari L, Bignami P, et al: 100 patients with hepatic metastases from colorectal cancer treated by resection: Analysis of prognostic determinants. Br J Surg 78:797-801, 1991}, 百拇医药
de Gramont A, Figer A, Seymour M, et al: Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 18:2938-2947, 2000
Grothey A, Sargent D, Goldberg RM, et al: Survival of patients with advanced colorectal cancer improves with the availability of fluorouracil-leucovorin, irinotecan and oxaliplatin in the course of treatment. J Clin Oncol 22:1209-1214, 2004(, 百拇医药
De Gramont A, Cervantes A, Andre T et al: OPTIMOX study: FOLFOX7/LV5FU2 compared to FOLFOX4 in patients with advanced colorectal cancer. Proc Am Soc Clin Oncol 22:251S, 2004 (abstr 3525)(, 百拇医药
Douillard JY, Cunningham D, Roth AD, et al: Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: A multicentre randomised trial. Lancet 355:1041-1047, 2000(, 百拇医药
Saltz LB, Cox JV, Blanke C, et al: Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer: Irinotecan Study Group. N Engl J Med, 343:905-914, 2000(, 百拇医药
Bismuth H, Adam R, Levi F, et al: Resection of nonresectable liver metastases from colorectal cancer after neoadjuvant chemotherapy. Ann Surg 224:509-520, 1996(, 百拇医药
Adam R, Avisar E, Ariche A, et al: Five-year survival following hepatic resection after neoadjuvant therapy for nonresectable colorectal. Ann Surg Oncol 8:347-353, 2001
Giacchetti S, Itzhaki M, Gruia G, et al: Long-term survival of patients with unresectable colorectal cancer liver metastases following infusional chemotherapy with 5-fluorouracil, leucovorin, oxaliplatin and surgery. Ann Oncol 10:663-669, 1999:\8, 百拇医药
Alberts SR, Donohue JH, Mahoney WL, et al: Liver resection after 5- fluorouracil, leucovorin, and oxaliplatin for patients with metastatic colorectal cancer limited to the liver: A North Central Cancer Treatment Group phase II study. Proc Am Soc Clin Oncol 22:263, 2003 (abstr 1053):\8, 百拇医药
Adam R: Chemotherapy and surgery: New perspectives on the treatment of unresectable liver metastases. Ann Oncol 14:ii13-ii16, 2003:\8, 百拇医药
Tournigand C, Andre T, Achille E, et al: FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: A randomised GERCOR study. J Clin Oncol 22:229-237, 2004:\8, 百拇医药
Pozzo C, Basso M, Cassano A, et al: Neoadjuvant treatment of unresectable liver disease with irinotecan, and 5-fluorouracil plus folinic acid in colorectal cancer patients. Ann Oncol 15:933-939, 2004
Nordlinger B, Guiguet M, Vaillant JC, et al: Surgical resection of colorectal carcinoma metastases to the liver: A prognostic scoring system to improve case selection, based on 1568 patients: Association Fran?aise de Chirurgie. Cancer 77:1254-1262,1996k;sqh3, 百拇医药
Fong Y, Fortner J, Sun RL, et al: Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer. Ann Surg 230:309-321, 1999k;sqh3, 百拇医药
Lise M, Bacchetti S, Da Pian P, et al: Patterns of recurrence after resection of colorectal liver metastases: Prediction by models of outcome analysis. World J Surg 25:638-644, 2001k;sqh3, 百拇医药
Heriot AG, Reynolds J, Marks CG, et al: Hepatic resection for colorectal metastases: A national perspective. Ann Royal Coll Surg Eng 86:420-424, 2004k;sqh3, 百拇医药
Brook RH, Chassin MR, Fink A, et al: A method for the detailed assessment of the appropriateness of medical technologies. Int J Technol Assess Health Care 2:53-63, 1986k;sqh3, 百拇医药
Fitch K, Bernstein SJ, Aguilar MD, et al: The RAND/UCLA Appropriateness User's Manual. Santa Monica, CA, RAND, MR-1269-DG XII/RESISTANCE, 2001
Linstone HA, Turoff M: The Delphi Method: Techniques and Applications. Reading, MA, Addisson-Wesley, 1975m, 百拇医药
Downs SH, Black N: The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 52:377-384, 1998m, 百拇医药
Cowley DE: Prostheses for primary total hip replacement: A critical appraisal of the literature. Int J Technol Assess Health Care 11:770-778, 1995m, 百拇医药
Simmonds PC, Colquitt JL, Primrose JN, et al. Surgical resection of hepatic metastases from colorectal cancer: A systematic review of published series. Colorectal hepatic metastases review group. J Clin Oncol (in press)m, 百拇医药
Cornelis M, Kahan JP, de Vries H, Poston GJ. From the appropriateness of procedures to the appropriateness of treatment strategies. Submitted Health Policy 2004m, 百拇医药
Cascinu S, Catalano V, Scartozzi M, et al: Liver metastases from colorectal cancer. Ann Oncol 8:393-399, 1997m, 百拇医药
Kahan JP, Van het Loo M: Defining appropriate health care. Eurohealth 5:16-18, 1999
Kahan JP, Cornelis M, De Vries H, et al: Preventing venous thromboembolism: Prophylactic options for patients at different risk levels. Leiden, the Netherlands, RAND Europe, MR-1689-EVIDIS/Sanofi, 2003[], 百拇医药
Park RE, Fink A, Brook RH, et al: Physician ratings of appropriate indications for six medical and surgical procedures. Am J Public Health 76:766-772, 1986[], 百拇医药
Wietlisbach V, Vader JP, Porchet F, et al: Statistical approaches in the development of clinical practice guidelines from expert panels: The case of laminectomy in sciatica patients. Med Care 37:785-797, 1999[], 百拇医药
Shekelle PG, Schriger DL: Evaluating the use of the appropriateness method in the Agency for Health Care Policy and Research Clinical Practice Guideline Development process. Health Serv Res 31:453-468, 1996[], 百拇医药
Washington DL, Stevens CD, Shekelle PG, et al: Safely directing patients to appropriate levels of care: Guideline-driven triage in the emergency service. Ann Emerg Med 36:15-22, 2000[], 百拇医药
Gale RP, Park RE, Dubois R, et al: Delphi-panel analysis of appropriateness of high-dose chemotherapy and blood cell or bone marrow autotransplants in diffuse large-cell lymphoma. Leuk Lymphoma 32:139-149, 1998[], 百拇医药
Shekelle PG, Kahan JP, Bernstein SJ, et al: The reproducibility of a method to identify the overuse and underuse of medical procedures. N Engl J Med 338:1888-1895, 1998[], 百拇医药
Shekelle PG: Are appropriateness criteria ready for use in clinical practice? N Engl J Med 344:677-678, 2001[], 百拇医药
Shekelle PG, Park RE, Kahan JP, et al: Sensitivity and specificity of the RAND/UCLA Appropriateness Method to identify the overuse and underuse of coronary revascularization and hysterectomy. J Clin Epidemiol 54:1004-1010, 2001[], 百拇医药
Marshall MN, Shekelle PG, McGlynn EA, et al: Can health care quality indicators be transferred between countries? Qual Saf Health Care 12:8-12, 2003(Graeme J. Poston, René Ad)
University of Southampton, Southampton, UKayn, http://www.100md.com
Hopital Ambroise Pare, Boulogneayn, http://www.100md.com
Hopital Paul Brousse, Villejuifayn, http://www.100md.com
Evidis, Paris, Franceayn, http://www.100md.com
Mayo Clinic, Rochester, MNayn, http://www.100md.com
The University of Texas M.D. Anderson Cancer Center, Houston, TXayn, http://www.100md.com
L'Hospitalet de Llobregat, Barcelona, Spainayn, http://www.100md.com
University of Pennsylvania Cancer Center, Philadelphia PAayn, http://www.100md.com
Hopital Cantonal, Geneva, Switzerlandayn, http://www.100md.com
Marlene and Stewart Greenebaum Cancer Center, Baltimore MDayn, http://www.100md.com
Allegheny General Hospital, Pittsburgh PAayn, http://www.100md.com
University Medical Center, Nijmegenayn, http://www.100md.com
RAND Europe, Leiden, the Netherlandsayn, http://www.100md.com
Martin Luther University, Halle, Germanyayn, http://www.100md.com
ABSTRACTayn, http://www.100md.com
PURPOSE: Most patients with colorectal liver metastases present to general surgeons and oncologists without a specialist interest in their management. Since treatment strategy is frequently dependent on the response to earlier treatments, our aim was to create a therapeutic decision model identifying appropriate procedure sequences.
METHODS: We used the RAND Corporation/University of California, Los Angeles Appropriateness Method (RAM) assessing strategies of resection, local ablation and chemotherapy. After a comprehensive literature review, an expert panel rated appropriateness of each treatment option for a total of 1,872 ratings decisions in 252 cases. A decision model was constructed, consensus measured and results validated using 48 virtual cases, and 34 real cases with known outcomes.ll5\l, 百拇医药
RESULTS: Consensus was achieved with overall agreement rates of 93.4 to 99.1%. Absolute resection contraindications included unresectable extrahepatic disease, more than 70% liver involvement, liver failure, and being surgically unfit. Factors not influencing treatment strategy were age, primary tumor stage, timing of metastases detection, past blood transfusion, liver resection type, pre-resection carcinoembryonic antigen (CEA), and previous hepatectomy. Immediate resection was appropriate with adequate radiologically-defined resection margins and no portal adenopathy; other factors included presence of 4 or > 4 metastases and unilobar or bilobar involvement. Resection was appropriate postchemotherapy, independent of tumor response in the case of 4 metastases and unilobar liver involvement. Resection was appropriate only for > 4 metastases or bilobar liver involvement, after tumor shrinkage with chemotherapy. When possible, resection was preferred to local ablation.
CONCLUSION: The results were incorporated into a decision matrix, creating a computer program (OncoSurge). This model identifies individual patient resectability, recommending optimal treatment strategies. It may also be used for medical education.{h0, http://www.100md.com
INTRODUCTION{h0, http://www.100md.com
Colorectal cancer is the third most prevalent cancer in the world, after lung and stomach cancer. More than 940,000 cases occur annually worldwide and nearly 500,000 people will die from it each year.1 The most common site of metastases is the liver, involved in 15% to 25% of patients at the time of diagnosis.2,3 An additional 35% to 45% of patients will develop liver metastases during the course of their disease.2 Interestingly, 20% to 30% of patients with advanced colorectal cancer have liver-only metastases,4 whereas recurrences after resection of the primary tumor are confined to the liver in nearly 50% of cases.5{h0, http://www.100md.com
Hepatic resection is the only potentially curative option for liver metastases, with contemporary operative mortality rates as low as 1% to 2%.6-10 Five-year survival rates following resection range between 25% and 40% in most large series.6-11 Even though eligibility for liver surgery continues to expand,12-15 80% of patients with metastatic disease remain unresectable at presentation. The recent development of more effective chemotherapeutic agents such as oxaliplatin and irinotecan are capable of inducing significant tumor shrinkage,16-20 prolong survival in nonoperable disease,21 and also appear to allow an additional 10% to 20% of patients thought to be initially unresectable for cure to undergo metastectomy.22-27 Long-term survival rates for these patients previously treated with palliative intent alone are comparable to those of primarily resected patients.22,23
Although there are a number of published clinical staging systems that predict prognosis based on available preoperative parameters,28-30 the subset of patients eligible for potentially curative resection is increasing; however, this proportion is not yet clearly defined. Definitions of surgical resectability of liver metastases are changing at the same time that new chemotherapeutic agents are evolving. In addition, prognostic factors intended to determine groups of patients able to benefit from resection need to be further defined. Moreover, data on the relative roles of surgery and chemotherapy have generally come from retrospective analyses from single centers, without a clear definition of selection criteria for surgery or chemotherapy. These data need to be confirmed by large-scale prospective randomized trials. Furthermore, many scoring systems for resection of liver metastases with curative intent were published in the era before the widespread adoption of the use of irinotecan and oxaliplatin in the management of these patients and similar improvements in both diagnostic and staging radiographic imaging.16-28 Ongoing studies of adjuvant chemotherapy for patients with resected or ablated liver metastases in both North America (National Surgical Adjuvant Breast and Bowel project [NSABP] C-09) and Europe (European Organisation for Research and Treatment of Cancer [EORTC] studies 40983 and 40004) will not be reported for several years. This delay poses a major problem for the present management of patients with colorectal liver metastases, who worldwide are frequently under the care of general oncologists and surgeons.31
These key contemporary developments in surgical technique, radiographic imaging, and chemotherapy are moving the treatment of colorectal liver metastases to a new multidisciplinary level, involving surgical, medical, and radiologic subspecialists. Although such multidisciplinary teams are the norm in tertiary centers of academic excellence, access to such expertise is not usually the case for most busy primary- and secondary-level practices across the world. Although it is accepted that in the present Internet era, data on the use of the latest chemotherapeutic regimens are readily available, the integration of these regimens (with evolving definitions of surgical resectability into multistage treatment strategies) remain in reality beyond the capabilities of the busy, overworked general oncologist or surgeon, who may see relatively few patients with potentially curable liver metastases./':, http://www.100md.com
The purpose of this project was to undertake an appropriateness study (a validated and accepted scientific method) using the RAND Corporation/University of California, Los Angeles Appropriateness Method (RAM)32-36 and to create a therapeutic decisional model (OncoSurge) to assist general medical oncologists and surgeons in defining optimal treatment strategies for individual patients with colorectal liver metastases.
METHODS\;2, http://www.100md.com
For the purpose of this study, the RAM was employed to assess the appropriateness of different treatment sequences for specific patient profiles.\;2, http://www.100md.com
The RAM is a statistically validated method32,33 that is a modified Delphi process.34 In using the RAM, experts overlay their consensus judgments on the available scientific evidence to assess the appropriateness of treatments. The model considers the average patient within each indication category, treated by the average provider in the average hospital.32-36 For this purpose the term "average" is explicitly defined clearly within a scientific context.32-36 The concept of appropriateness refers to the relative weight of the benefits and risks of a medical procedure. A procedure is "appropriate" for a given indication when expected benefits for the patient outweigh possible negative consequences by a sufficient margin. It is accepted that, in reality, the definition of a "sufficient margin" must be based on a synthesis of the perceptions of the individual patient, their clinician(s) and their respective individual thresholds for taking risks. Therefore, to be of clinical use, the decision model must rank and stratify benefit/risk and benefits ratios for each potential given treatment strategy in each potential individual clinical presentation. A procedure is "inappropriate" when possible negative consequences exceed benefits and "uncertain" when benefits and risks are nearly equal, or if there is disagreement within the expert panel on risk/benefit.34-36
The appropriateness of different treatments was scored on a scale of 1 (highly inappropriate) to 9 (clearly appropriate) against each clinical presentation. For the purpose of analysis, ratings from 1-3 for any given treatment are considered inappropriate, 4 to 6 represented uncertainty on the part of the expert rater, and 7 to 9 meant that the expert rated the given treatment as appropriate. The group rating was taken as the median judgment of the panelists, if there was no statistically defined disagreement among the panel members.32-36 Uncertainty here can mean that the risks and benefits are approximately equivocal, that the experts are uncertain, or that there is disagreement among the experts; in any of these instances, uncertainty is a call for further clinical research. Bimodal disagreement (1 to 3 v 7 to 9) meant that a group of the experts misunderstood the proposed treatment options, and in all cases where this phenomenon occurred during the first round of ratings, changed to unimodal agreement (1 to 3 or 7 to 9) after the discussion preceding the second round of ratings. Unimodal agreement between 4 and 6 and linear disagreement (ranging between 1 and 9) represents true uncertainty on the part of the expert panelists and identifies areas for future clinical research.
Uncertainty here can mean that the risks and benefits are approximately equivocal, that the experts are uncertain, or that there is disagreement among the experts; in any of these instances, uncertainty is a call for further clinical research. For the purpose of this study, an international panel of 16 oncologists, surgeons and radiologists with large experience in the diagnosis and treatment of liver metastases participated in the study (Table 1). As a first step, the experts were provided with a systematic review prepared specifically for this study by two of the authors of this study and funded independently by the British Cancer Research Campaign (now CRUK).37 This review was intended to summarize scientific evidence and list major indications for surgical resection of colorectal liver metastases. Any patient could be assigned to only one indication, but considered for each possible treatment.!y%nf, 百拇医药
A preliminary panel meeting established three treatment segments to be considered: liver resection, local ablation, and chemotherapy. The panel recognized that treatment by tumor ablation is presently in evolution and its utility in comparison to chemotherapy alone remains be proven. For the purposes of the RAM, the panel agreed that ablation could be considered for low-volume, liver-only disease in a patient deemed to be either unsuitable for liver resection by an experienced liver surgeon, or as an adjunct to liver resection (see the Results section). For each treatment segment, all relevant possible patient characteristics were identified and included in a comprehensive list of indications. Negative characteristics considered to be exclusively inappropriate for any treatment segment strategy were also compiled. Panelists performed the first ratings, and these ratings were subsequently analyzed for appropriateness and for areas for disagreement.
A second meeting was held to present and discuss the results of the first round of ratings, and modifications were made to the list of both treatment segment options and indications. Panelists then privately rated each of the three treatment segments separately. The second round of ratings provided the basis for assessing the appropriateness of the treatment segments.cv, 百拇医药
The different treatment strategies that resulted from the second round of ratings were incorporated into a decision matrix in the form of a computer program, the OncoSurge decision model. Validation of the model by the expert panel was accomplished using 48 representative virtual theoretical cases using the real decision characteristics identified by the appropriateness analysis. These 48 cases were created specifically to ensure that the expert panel had considered the full range of all potential clinical presentations, making the validation process span the full range of potential patients. Furthermore, the validation process using the 48 virtual theoretical cases was reinforced by the panelists using 34 real cases from their panelists' own clinical practices (with known outcomes). The purpose of the virtual theoretical cases was to test the rigor accuracy of the methodology and the rating process, with the intention of measuring consensus (appropriateness or inappropriateness of each treatment strategy) and disagreement (uncertainty) within the panel. This methodology has been utilized and tested in previous uses of the RAM in health care decision making.32-36,41
Each case was validated by comparing the group modal treatment choice with the second-round appropriateness ratings within each treatment segment of the case. If that option had been rated appropriate, then the group was scored as consistent for the treatment segment for that case. If the option had been rated inappropriate, then the group was scored as inconsistent. If the appropriateness rating for the choice was uncertain, then the group was rated consistent only if "no other choice" had been rated appropriate. In other words, a consistent choice was one in which there was no option considered better by the appropriateness ratings.}f6|, 百拇医药
RESULTS}f6|, 百拇医药
Initial Lists of Indications and Treatment Options}f6|, 百拇医药
On the basis of the systematic literature review, separate lists of indications and treatment options were drawn up for each of the three treatment segments considered for the study: resection, local ablation, and chemotherapy. A list of absolute contraindications to resection with curative intent was also defined, including the following clinical conditions: unresectable extrahepatic disease (such as peritoneal carcinomatosis, multifocal lung metastases, and distant lymph nodes in the preaortic and celiac axis N2 nodal basin for the liver, bone, or brain metastases), extensive liver involvement (more than six liver segments involved, > 70% liver invasion or all three hepatic veins involved), major liver insufficiency, or Child B or C liver cirrhosis with complications, and patients unfit for or declining surgery. Although it was accepted that tumor involvement at the confluence of two hepatic veins, inferior vena cava, or contralateral portal pedicle might indicate that only an R1 rather than R0 resection was possible, this was not considered an absolute contraindication to liver resection. Although patient age, stage of primary tumor, timing of detection of metastases, blood transfusion during surgery, type of resection (anatomic versus nonanatomic), presurgical carcinoembryonic antigen level, and previous hepatectomy are known to be relative prognostic factors for survival,28-30,37 they were not considered in the decision making for different treatment strategies.
Although triphasic computed tomography scanning of the liver is now used routinely in detecting metastatic liver disease, the assessment of the extent of resection necessary to achieve negative resection margins will be available only to experienced liver surgeons working closely with their hepatobiliary radiologists (usually in tertiary center). For example, if a trisegmentectomy was required but would leave a small functioning hepatic remnant, then this would be important preoperative information to justify neoadjuvant chemotherapy, with or without preoperative portal vein embolization. However, decisions at this strategic level are the province of the specialist center, and beyond the scope of this project, which was to advise physicians working at the primary and secondary levels of care (who do not deal frequently with advanced colorectal cancer), on the appropriateness of complex interdependent treatment strategies.qw+]o$p, http://www.100md.com
The indications for rating each treatment segment were drawn from the clinical characteristics shown in Table 2. A total of seven characteristics were considered, of which five had two possible values and two had three possible values. Therefore, there were potentially 288 (3 x 3 x 2 x 2 x 2 x 2 x 2) indications for each treatment option. The initial therapeutic options considered for each treatment segment are presented in Table 3. For resection, eight different possibilities were considered, depending on whether surgery was combined with preoperative chemotherapy and on whether surgery was performed (or not) in one or two stages. For local ablation, the panelists considered cryosurgical and radiofrequency ablation together, yielding four local ablation possibilities differing with respect to their possible combination with resection. For chemotherapy, 12 different administrations were considered, depending on the goal and timing of chemotherapy in the treatment strategy (palliative, preoperative, postoperative), and the chemotherapy regimen when chemotherapy was to be considered (fluorouracil plus leucovorin [FU/LV], FU/LV plus oxaliplatin, FU/LV plus irinotecan, FU/LV plus both oxaliplatin and irinotecan). Thus, there were 384 (8 x 4 x 12) different treatment sequences to consider. If each indication were assessed for each treatment sequence, there would have been 110,592 (288 x 384) different ratings to be made, clearly an impossible situation.
The ratings task was brought back to tractability by employing two different tactics.38 First, instead of assessing each sequence (combination of resection, local ablation, and chemotherapy) as a whole, appropriateness was assessed for each segment separately, and later for different sequences if each segment of the sequence was itself appropriate. Second, within each treatment segment, clinical characteristics which were not considered germane were identified (Table 2), thereby providing a large reduction in indications.4, 百拇医药
More specifically, for resection, tumor size was not considered germane,39 so only 144 indications were rated for each of the eight resection possibilities, for a total of 1,152 ratings. For local ablation, radiologically suspicious portal lymph node involvement and lobar (uni- v bilobar) involvement were not considered germane, so only 72 indications were rated for each of the four local ablation possibilities, for a total of 288 ratings. For chemotherapy, at the time of these discussions (August 2003) data on biologic therapies were not sufficient enough to be included as treatment options. Similarly, the use of regionally delivered hepatic chemotherapy was confined largely to the United States, and was therefore not considered specifically. Radiologically-defined resection margins, portal lymph node involvement and lobar involvement were not considered relevant to the decision concerning chemotherapy options, so only 36 indications were rated for each chemotherapy possibility, for a total of 432 ratings.
Therefore, instead of the total possible 110,592 ratings, there were only 1,872 (1,152 + 288 + 432) to be performed by each panelist. This is well within the range of reasonableness, and indeed, the panelists complied with the task. However, as shown in Table 2, not all characteristics were used for each treatment segment.ko.p@, 百拇医药
Results of the First-Round Ratingsko.p@, 百拇医药
The first-round ratings were examined, less for the description of appropriateness for each treatment segment than for the presence of inter-rater disagreement, logical consistency, and whether the clinical characteristics made a difference in appropriateness.ko.p@, 百拇医药
For the resection segment, there was disagreement in 86 (7.5%) of the 1,152 ratings indications. However, this is a possible overstatement of consensus, because treatments were scored as rated inappropriate for 70% of the ratings indications. Thus, the two-stage procedure was always inappropriate, as well as surgery in the presence of portal lymph node involvement. Moreover, the ratings for immediate resection without previous chemotherapy and resection following optimal chemotherapy were almost identical. Similarly, the lymph node status of the primary colorectal tumor made no difference in appropriateness for liver resection.
For the local ablation segment, there was disagreement in 29 (10.0%) of the 288 indications. Again, statistics overstated true consensus, because the local destruction was considered appropriate in only 22 of 288 indications. Again, the lymph node status of the primary tumor was not taken into account for ratings.f$7[.7, 百拇医药
For the chemotherapy segment, the rate of disagreement was very high, in 30 (2.8%) of 144 ratings indications. In addition, many other indications had dispersions that were close to the statistical border for disagreement. Consequently, it was decided to entirely rebuild the indications structure for this treatment segment.f$7[.7, 百拇医药
Final Indicationsf$7[.7, 百拇医药
On the basis of the results of the first round of ratings, the panelists made major revisions to the indications structure, changing both the treatment options to be assessed and the indications matrix. All of these changes were in the direction of simplifying potential treatment strategies.f$7[.7, 百拇医药
As with the first round of ratings, second-round ratings were performed separately for each treatment segment of an overall treatment strategy encompassing resection, local ablation and chemotherapy. Regarding the intention to perform liver resection, the panel recommended to first resect the primary tumor before performing the liver resection. The final decision to proceed with potentially curative treatment of liver metastases assumed that the primary tumor could be removed completely (R0 resection of primary tumor) (including local lymph node metastases). The panel accepted that occasionally a patient will present with extensive but potentially resectable liver disease in the presence of a small and asymptomatic primary tumor. However, such presentations are uncommon, and the optimal treatment strategy for such patients would ideally reside within a tertiary center. Staged liver resections and the combination of portal vein embolization with surgery had to be decided by the performing liver surgeon at the time of surgery. As such these factors were not a consideration relating to the decision to refer the patient to the specialist hepatobiliary center with the intention to operate, and therefore were not part of the indications structure presented for rating. Chemotherapy could be used before resection, either with the intent to render unresectable tumors resectable or to improve the status of initially resectable tumors. Thus, three resection options were finally considered: immediate resection with no previous chemotherapy; resection following preoperative chemotherapy, independent of the tumor response; and resection only after tumor shrinkage with preoperative chemotherapy.
All clinical characteristics considered for the first round, except for lymph node status of the primary tumor, were employed (Table 2), so that there were 48 indications for each of the three new treatment options, or 144 total ratings for resection.'5, 百拇医药
Regarding the intention to perform local ablation, the four treatment options shown in Table 3 were not modified, apart from a decision to restrict the maximum diameter of the largest treatable liver metastasis to < 4 cm. The panel accepted that there was no consensus on the maximum number and size of liver metastases that could be treated using this technique. Similarly, the panel accepted that this treatment could be used repeatedly, depending on the response to previous tumor ablations. Therefore the panel agreed to rate recommendations on ablation therapy on intention to treat at the outset, but with the proviso that if referred on, then the final decision would reside with the specialist liver resection center (allowing the possibility of resection instead of ablation after review at the specialist center). Thus, the second round considered 24 ratings for each treatment option, for a total of 96 ratings.
Chemotherapy strategies were different, depending on palliative intent (where, in the view of the panel, there was little or no hope of reducing tumor burden to the point of resection with curative intent) or in combination with surgery in the preoperative and/or postoperative setting. The choice of postoperative chemotherapy also depended on the curative (R0) intent or not (R1) of the resection. Finally, nine indications resulted from the combination of these clinical conditions: preoperative chemotherapy—unresectable metastases, and unlikely to be rendered resectable (> 70% liver replacement, > 6 segments tumor involvement, unresectable extra-hepatic disease) and unresectable metastases, possibly rendered resectable (all three hepatic veins involved, radiologic involvement of portal pedicle lymph nodes, radiologic concern about resection margins); postoperative chemotherapy—initially resectable metastases, R0 resection, without preoperative chemotherapy, R1 resection, without preoperative chemotherapy, R0 resection, after effective (regression or stabilization) preoperative chemotherapy, R1 resection after effective preoperative chemotherapy, R0 resection, after ineffective (tumor progression while on treatment) preoperative chemotherapy, and R1 resection, after ineffective preoperative chemotherapy.
Of the initial clinical characteristics, only the presence of treatable extrahepatic disease was considered. As a consequence, there were 18 indications rated for each of four generic chemotherapy regimens, for a total of 72 ratings.k7, 百拇医药
For each treatment segment, panelists privately rated the new indications structure and these 312 (144 + 96 + 72) second-round ratings provided the basis for assessing the appropriateness of the treatment segments. These second-round ratings provided the basis for assessing the appropriateness of the treatment segments.k7, 百拇医药
Analysis of the Second Round of Ratingsk7, 百拇医药
The first step of the second round of ratings was to analyze the disagreements among the panelists. For all three treatment segments, disagreements were low: 11 of 144 ratings indications (7.6%) for resection, one of 96 ratings indications (0.9%) for local destruction, and two of 72 ratings indications (2.8%) for chemotherapy. The decrease in disagreement is due to the revision of the indications list that evolved from discussion during the second round of ratings. Therefore, analysis of the appropriateness rating was performed using a classification tree technique. Classification was based on the splitting of the data into increasingly homogeneous subsets, separating the indication-treatment combinations rated as appropriate from those rated inappropriate or uncertain. This classification was based on a dichotomous analysis of the results (classification and regression tree analysis), using a classification risk < 10% for appropriateness.38 Each split was conditioned by an optimal cutoff in the levels of a single factor and the process was continued as long as it was feasible or statistically significant.
For resection, immediate liver resection depended primarily on the presence of adequate radiologic resection margins. The panel accepted that there was no universally agreed upon definition of adequate resection margins and that the previously accepted standard of at least 1 cm has now been largely discarded.28-30,37,39 Therefore for the purpose of the study, the exact definition of acceptable resection margins (as with the type of liver resection) would be left to the expert hepatobiliary surgical team. In cases of radiologically inadequate resection margins on preoperative liver scans, immediate resection was considered inappropriate. Immediate resection depended secondarily on the radiologic absence of portal lymph node (PLN). If there were adequate resection margins with PLN involvement, immediate resection was inappropriate. Numbers of metastases and lobar involvement were additional criteria. In cases of adequate resection margins without PLN involvement, immediate resection was appropriate if the number of metastases was 4. If there were more than 4 metastases, immediate resection was appropriate or uncertain only if the metastases were unilateral, and inappropriate if they were bilateral. A decision to resect was considered appropriate after preoperative chemotherapy, independent of tumor response, primarily if the number of metastases was 4 and secondarily if there was unilobar involvement. If the number of metastases was > 4 with bilobar liver involvement, even if immediately amenable to trisegmentectomy, resection was considered appropriate only after tumor shrinkage with chemotherapy.
Although triphasic CT scanning is now used routinely in detecting liver metastases, the ability to assess the extent of resection required to achieve negative resection margins, and consequently the potential indication for neoadjuvant chemotherapy, will only be available to experienced liver surgeons working closely with their hepatobiliary radiologists and oncologists within a multidisciplinary team. The panel concluded that worldwide, most general oncologists and surgeons who do not deal with metastatic colorectal cancer on a day-to-day basis, would not feel comfortable in making decisions of this degree of complexity at this strategic level.}+hyp6-, 百拇医药
For local ablation, few indications (9.4%) were considered appropriate, generally if there were fewer than four small metastases in a patient for whom resection was not considered appropriate because of poor general condition or other excessive surgical risk. Whenever possible, resection was preferred to local ablation.}+hyp6-, 百拇医药
For chemotherapy, postoperative chemotherapy was indicated in all patients who received preoperative chemotherapy, using the same regimen if this was effective in the preoperative setting or changing the chemotherapy combination if this was ineffective in the preoperative setting. Analysis of chemotherapy regimens when chemotherapy was indicated is summarized in Table 4. By the end of 2003 (time of the second panel meeting), FU/LV alone was rarely rated appropriate in any given circumstance (preoperative, post R0 resection, post R1 resection).16,17,20-25 The three regimens using FU/LV in combination with oxaliplatin and/or irinotecan were generally appropriate, except following complete resection (R0) when postoperative chemotherapy was perceived as uncertain instead of positive. Differences between the choice of oxaliplatin and irinotecan were small, with a slight preference in favor of oxaliplatin. A triple combination of FU/LV with both oxaliplatin and irinotecan (which at the time the panelists met was an unlicensed regimen) was restricted to incomplete resection (R1).
Validation\62*[s, 百拇医药
The validation process began with an examination of the 48 virtual theoretical cases that covered all the factors previously identified in determining treatment strategy. Of these, 47 were validated, and one was not. That single case was related to the recommendation of an uncertain treatment when an appropriate one was available. Looking next at the 47 remaining cases, for four of these the local ablation choice was inconsistent with the appropriateness ratings. Again, all of the inconsistent ratings were related to the choice of an uncertain treatment when an appropriate one was available. Finally, with regards to the choices of chemotherapy for the 43 remaining cases, only six of them were inconsistent, and again because an uncertain choice was preferred to an appropriate one. Thus, looking at the entire treatment strategy, 37 (77%) of 48 were validated. The panel therefore identified uncertainty at the present time for an appropriate treatment strategy in 23% of clinical scenarios.\62*[s, 百拇医药
DISCUSSION\62*[s, 百拇医药
The management of patients with metastatic colorectal cancer has improved significantly over the last few years. Medical and surgical advances are the foundation of a multidisciplinary approach for patients with colorectal liver metastases. This approach is now mainstream practice in most major academic centers. However, in worldwide day-to-day practice, this integrated approach is still not used in many patients who currently might be curable with optimal multidisciplinary therapy.31 With the OncoSurge concept, emerging developments in surgery, chemotherapy and radiology to optimally define the management for colorectal liver metastases are brought together into an integrated, combined modality framework.
The present study has attempted to address the question of appropriateness of three treatment options: resection; local ablation; and chemotherapy, used alone or in combination, depending on each patient's specific individual tumor characteristics. In combination treatment strategies, each sequence depended on the response to the previous therapy strategy.o{, http://www.100md.com
The statistically validated RAM32-36 was used to assess the clinical/radiological characteristics of the patients ("indications") and to rate each treatment approach according to its appropriateness. Previous uses of the RAM in health care decision making have focused on clinical conditions for which the assessment of the appropriateness of a particular treatment has been relatively straightforward, such as prophylaxis of deep vein thrombosis, the use of ACE 2 inhibitors in the management of hypertension, the role of surgery in lumbar disc prolapse, emergency room triage, and the use of high-dose chemotherapy in the treatment of lymphoma.40-42 None to date used the segmentation tactic employed here. In the present model, both the indications and the possible treatment segments were multiple. The number of indications for treatment multiplied by the number of individual treatment strategies could easily result in literally millions of ratings.
The OncoSurge study overcomes these difficulties by adapting the RAM to consider limited treatment strategies composed of treatment segments. This study has demonstrated the feasibility of applying the concept of appropriateness to the assessment of complex treatment strategies.42 Similarly, other non-RAM Delphi-panel analyses have addressed treatment strategies ranging from the management of low back pain,43,44 through emergency room triage,45 to the use of high-dose chemotherapy in the treatment of lymphoma46; again, none to date used the segmentation tactic employed in the present study.*d*, 百拇医药
Clearly, important limitations emerge when considering what constitutes "appropriate" care. Answers to this question depend on which clinicians are asked, where they work and live, what weight is given to the different types and levels of evidence and end points, the weight given to the preferences of patients and their families, the level of resources in a given health care system, and the prevailing values of both the health system and the society in which it operates (especially relevant in the setting of a national health care system). The RAM methodology draws heavily on expert opinions, which can vary with use between different sets of panels.47 It is accepted that without extremely rigorous application, variability can be allowed to occur within all the standard steps of the RAM,48 and this variability can explain why the sensitivity and specificity of this method might be less than perfect.49
Moreover, when clinical decisions are being made, both patient and family requests (such as proximity of treatment center to home) and physician factors (relationships between oncologists and surgeons) are very important influences. Hence an individual clinician's approach to a particular medical condition is molded by a complex interplay of their understanding of the scientific evidence, clinical circumstances, and personal values or preferences, which are not adequately evaluated by the RAM. Even for patients with colorectal liver metastases (where the decision-making strategies are potentially life saving), all these issues come into play when formulating a treatment plan.:i, 百拇医药
The OncoSurge decision model allows a clinician to derive a choice of treatment strategy for an individual patient with colorectal liver metastases and compares this choice with the expert's view. Thus, the OncoSurge study combines the best available scientific evidence with the collective judgment of worldwide experts to yield a statement regarding the appropriateness of performing liver resection and/or local destruction, with or without chemotherapy for each patient. The methodology used to reach these conclusions can be applied equally efficiently across health care systems worldwide.50
The OncoSurge decision model identifies and individualizes patient resectability and makes recommendations for the optimal treatment of colorectal liver metastases, and may be used as an innovative approach for therapeutic decision support and medical education. Multimodality therapy is fast becoming the mainstay in the management of cancer. This methodology now requires further exploration in other indications, such as rectal cancer or adjuvant therapy for breast, lung, and pancreatic cancer.er-, http://www.100md.com
Authors' Disclosures of Potential Conflicts of Interester-, http://www.100md.com
Although all authors completed the disclosure declaration, the following authors or their immediate family members indicated a financial interest. No conflict exists for drugs or devices used in a study if they are not being evaluated as part of the investigation. For a detailed description of the disclosure categories, or for more information about ASCO's conflict of interest policy, please refer to the Author Disclosure Declaration and the Disclosures of Potential Conflicts of Interest section in Information for Contributors.er-, http://www.100md.com
NOTES
Supported by an unrestricted educational grant from the Sanofi-Synthelabo Group.p;, 百拇医药
Presented previously at the 40th Annual Meeting of the American Society of Clinical Oncology, New Orleans, LA, June 5-8, 2004; the 29th Annual Meeting of the Euproean Society of Medical Oncology, Vienna, Austria, October 29-November 2, 2004; and the American Society of Clinical Oncology Gastrointestinal Cancers Symposium, Hollywood, FL, January 27-29, 2005.p;, 百拇医药
The results of this study have not been subject to approval by the Sanofi-Synthelabo Group. The recommendations given in this study are solely the responsibility of the authors and do not represent the views of the Sanofi-Synthelabo Group.p;, 百拇医药
Authors' disclosures of potential conflicts of interest are found at the end of this article.p;, 百拇医药
REFERENCESp;, 百拇医药
WHO. Global cancer rates could increase by 50% to 15 million by 2020. http://www.who.int/mediacentre/releases/2003/pr27/en/print.htmlp;, 百拇医药
Kemeny N, Fata F: Arterial, portal or systemic chemotherapy for patients with hepatic metastasis of colorectal carcinoma. J Hepatobiliary Pancreat Surg 6:39-49, 1999
Seifert JK, Junginger T, Morris DL: A collective review of the world literature on hepatic cryotherapy. J R Coll Surg Edinb 43:141-154, 1998]7t, 百拇医药
Borner MM: Neoadjuvant chemotherapy for unresectable liver metastases of colorectal cancer: Too good to be true? Ann Oncol 10:623-626, 1999]7t, 百拇医药
Fong Y, Cohen AM, Fortner JG, et al: Liver resection for colorectal metastases. J Clin Oncol 15:938-946, 1997]7t, 百拇医药
Adson MA: Resection of liver metastases: When is it worthwhile? World J Surg 11:511-520,1987]7t, 百拇医药
Wilson SM, Adson MA: Surgical treatment of hepatic metastases from colorectal cancers. Arch Surg 111:330-334,1976]7t, 百拇医药
Rees M, John TG: Current status of surgery in colorectal metastases to the liver. Hepatogastroenterology 48:341-344, 2001]7t, 百拇医药
Hohenberger P: Colorectal cancer: What is standard surgery? Eur J Cancer 37:S173-S187, 2001 (suppl 7)]7t, 百拇医药
Papadimitriou JD, Fotopoulos AC, Prahalias AA, et al: The impact of new technology on hepatic resection for malignancy. Arch Surg 136:1307-1313, 2001
Malafosse R, Penna C, Sa Cunha A, et al: Surgical management of hepatic metastases from colorectal malignancies. Ann Oncol 12:887-894, 2001}, 百拇医药
Scheele J, Stang R, Altendorf-Hofmann A, et al: Resection of colorectal liver metastases. World J Surg 19:59-71, 1995}, 百拇医药
Gayowski TJ, Iwatsuki S, Madariaga SR, et al: Experience in hepatic resection for metastatic colorectal cancer: Analysis of clinical and pathological risk factors. Surgery 116:703-711, 1994}, 百拇医药
Fortner JG, Silva JS, Golbey RB, et al: Multivariate analysis of a personal series of 247 consecutive patients with liver metastases from colorectal cancer: Treatment by hepatic resection. Ann Surg 199:306-316, 1984}, 百拇医药
Doci R, Gennari L, Bignami P, et al: 100 patients with hepatic metastases from colorectal cancer treated by resection: Analysis of prognostic determinants. Br J Surg 78:797-801, 1991}, 百拇医药
de Gramont A, Figer A, Seymour M, et al: Leucovorin and fluorouracil with or without oxaliplatin as first-line treatment in advanced colorectal cancer. J Clin Oncol 18:2938-2947, 2000
Grothey A, Sargent D, Goldberg RM, et al: Survival of patients with advanced colorectal cancer improves with the availability of fluorouracil-leucovorin, irinotecan and oxaliplatin in the course of treatment. J Clin Oncol 22:1209-1214, 2004(, 百拇医药
De Gramont A, Cervantes A, Andre T et al: OPTIMOX study: FOLFOX7/LV5FU2 compared to FOLFOX4 in patients with advanced colorectal cancer. Proc Am Soc Clin Oncol 22:251S, 2004 (abstr 3525)(, 百拇医药
Douillard JY, Cunningham D, Roth AD, et al: Irinotecan combined with fluorouracil compared with fluorouracil alone as first-line treatment for metastatic colorectal cancer: A multicentre randomised trial. Lancet 355:1041-1047, 2000(, 百拇医药
Saltz LB, Cox JV, Blanke C, et al: Irinotecan plus fluorouracil and leucovorin for metastatic colorectal cancer: Irinotecan Study Group. N Engl J Med, 343:905-914, 2000(, 百拇医药
Bismuth H, Adam R, Levi F, et al: Resection of nonresectable liver metastases from colorectal cancer after neoadjuvant chemotherapy. Ann Surg 224:509-520, 1996(, 百拇医药
Adam R, Avisar E, Ariche A, et al: Five-year survival following hepatic resection after neoadjuvant therapy for nonresectable colorectal. Ann Surg Oncol 8:347-353, 2001
Giacchetti S, Itzhaki M, Gruia G, et al: Long-term survival of patients with unresectable colorectal cancer liver metastases following infusional chemotherapy with 5-fluorouracil, leucovorin, oxaliplatin and surgery. Ann Oncol 10:663-669, 1999:\8, 百拇医药
Alberts SR, Donohue JH, Mahoney WL, et al: Liver resection after 5- fluorouracil, leucovorin, and oxaliplatin for patients with metastatic colorectal cancer limited to the liver: A North Central Cancer Treatment Group phase II study. Proc Am Soc Clin Oncol 22:263, 2003 (abstr 1053):\8, 百拇医药
Adam R: Chemotherapy and surgery: New perspectives on the treatment of unresectable liver metastases. Ann Oncol 14:ii13-ii16, 2003:\8, 百拇医药
Tournigand C, Andre T, Achille E, et al: FOLFIRI followed by FOLFOX6 or the reverse sequence in advanced colorectal cancer: A randomised GERCOR study. J Clin Oncol 22:229-237, 2004:\8, 百拇医药
Pozzo C, Basso M, Cassano A, et al: Neoadjuvant treatment of unresectable liver disease with irinotecan, and 5-fluorouracil plus folinic acid in colorectal cancer patients. Ann Oncol 15:933-939, 2004
Nordlinger B, Guiguet M, Vaillant JC, et al: Surgical resection of colorectal carcinoma metastases to the liver: A prognostic scoring system to improve case selection, based on 1568 patients: Association Fran?aise de Chirurgie. Cancer 77:1254-1262,1996k;sqh3, 百拇医药
Fong Y, Fortner J, Sun RL, et al: Clinical score for predicting recurrence after hepatic resection for metastatic colorectal cancer. Ann Surg 230:309-321, 1999k;sqh3, 百拇医药
Lise M, Bacchetti S, Da Pian P, et al: Patterns of recurrence after resection of colorectal liver metastases: Prediction by models of outcome analysis. World J Surg 25:638-644, 2001k;sqh3, 百拇医药
Heriot AG, Reynolds J, Marks CG, et al: Hepatic resection for colorectal metastases: A national perspective. Ann Royal Coll Surg Eng 86:420-424, 2004k;sqh3, 百拇医药
Brook RH, Chassin MR, Fink A, et al: A method for the detailed assessment of the appropriateness of medical technologies. Int J Technol Assess Health Care 2:53-63, 1986k;sqh3, 百拇医药
Fitch K, Bernstein SJ, Aguilar MD, et al: The RAND/UCLA Appropriateness User's Manual. Santa Monica, CA, RAND, MR-1269-DG XII/RESISTANCE, 2001
Linstone HA, Turoff M: The Delphi Method: Techniques and Applications. Reading, MA, Addisson-Wesley, 1975m, 百拇医药
Downs SH, Black N: The feasibility of creating a checklist for the assessment of the methodological quality both of randomised and non-randomised studies of health care interventions. J Epidemiol Community Health 52:377-384, 1998m, 百拇医药
Cowley DE: Prostheses for primary total hip replacement: A critical appraisal of the literature. Int J Technol Assess Health Care 11:770-778, 1995m, 百拇医药
Simmonds PC, Colquitt JL, Primrose JN, et al. Surgical resection of hepatic metastases from colorectal cancer: A systematic review of published series. Colorectal hepatic metastases review group. J Clin Oncol (in press)m, 百拇医药
Cornelis M, Kahan JP, de Vries H, Poston GJ. From the appropriateness of procedures to the appropriateness of treatment strategies. Submitted Health Policy 2004m, 百拇医药
Cascinu S, Catalano V, Scartozzi M, et al: Liver metastases from colorectal cancer. Ann Oncol 8:393-399, 1997m, 百拇医药
Kahan JP, Van het Loo M: Defining appropriate health care. Eurohealth 5:16-18, 1999
Kahan JP, Cornelis M, De Vries H, et al: Preventing venous thromboembolism: Prophylactic options for patients at different risk levels. Leiden, the Netherlands, RAND Europe, MR-1689-EVIDIS/Sanofi, 2003[], 百拇医药
Park RE, Fink A, Brook RH, et al: Physician ratings of appropriate indications for six medical and surgical procedures. Am J Public Health 76:766-772, 1986[], 百拇医药
Wietlisbach V, Vader JP, Porchet F, et al: Statistical approaches in the development of clinical practice guidelines from expert panels: The case of laminectomy in sciatica patients. Med Care 37:785-797, 1999[], 百拇医药
Shekelle PG, Schriger DL: Evaluating the use of the appropriateness method in the Agency for Health Care Policy and Research Clinical Practice Guideline Development process. Health Serv Res 31:453-468, 1996[], 百拇医药
Washington DL, Stevens CD, Shekelle PG, et al: Safely directing patients to appropriate levels of care: Guideline-driven triage in the emergency service. Ann Emerg Med 36:15-22, 2000[], 百拇医药
Gale RP, Park RE, Dubois R, et al: Delphi-panel analysis of appropriateness of high-dose chemotherapy and blood cell or bone marrow autotransplants in diffuse large-cell lymphoma. Leuk Lymphoma 32:139-149, 1998[], 百拇医药
Shekelle PG, Kahan JP, Bernstein SJ, et al: The reproducibility of a method to identify the overuse and underuse of medical procedures. N Engl J Med 338:1888-1895, 1998[], 百拇医药
Shekelle PG: Are appropriateness criteria ready for use in clinical practice? N Engl J Med 344:677-678, 2001[], 百拇医药
Shekelle PG, Park RE, Kahan JP, et al: Sensitivity and specificity of the RAND/UCLA Appropriateness Method to identify the overuse and underuse of coronary revascularization and hysterectomy. J Clin Epidemiol 54:1004-1010, 2001[], 百拇医药
Marshall MN, Shekelle PG, McGlynn EA, et al: Can health care quality indicators be transferred between countries? Qual Saf Health Care 12:8-12, 2003(Graeme J. Poston, René Ad)