Ann Surg Oncol (2013) 20:183–192 DOI 10.1245/s10434-012-2473-5

ORIGINAL ARTICLE – COLORECTAL CANCER

Definition of Patients Presenting a High Risk of Developing Peritoneal Carcinomatosis After Curative Surgery for Colorectal Cancer: A Systematic Review Charles Honore´, MD, Diane Goe´re´, MD, Amine Souadka, MD, Fre´de´ric Dumont, MD, and Dominique Elias, MD, PhD Department of Surgical Oncology, Institut Gustave Roussy, Cancer Center, Villejuif, France

ABSTRACT Background. In colorectal cancer, complete cytoreductive surgery associated with hyperthermic intraperitoneal chemotherapy achieves encouraging results in early peritoneal carcinomatosis (PC), but this early detection can only be accurately accomplished during a systematic second-look surgery. This costly and invasive approach can only be proposed to selected patients. The objective of this study was to identify risk factors predictive of developing PC after curative surgery for colorectal cancer. Methods. After a systematic review of the literature published between 1940 and 2011, all clinical studies reporting the incidence of PC after curative surgery for colorectal cancer were searched for factors associated with the primary tumor that were likely to influence the incidence of recurrent PC. Results. Sixteen clinical studies were considered informative, all nonrandomized, three prospective and 13 retrospective, including 4–395 patients. Overall, the methodological quality of the reported studies was low. Data were available for the following factors: synchronous PC, synchronous ovarian metastases, perforated primary tumor, serosal and/or adjacent organ invasion, histological subtype, and positive peritoneal cytology with reported incidences of recurrent PC between 8 and 75 %. No study was found that mentioned an impact of lymph node invasion, tumor location, laparoscopy, occlusive tumors, or bleeding tumor on recurrent PC.

 Society of Surgical Oncology 2012 First Received: 29 November 2011; Published Online: 23 October 2012 C. Honore´, MD e-mail: [email protected]

Conclusions. Evidence regarding the incidence of recurrent PC after curative surgery for colorectal cancer is poor. Emerging data indicate three situations that could result in a real higher risk of recurrent PC: synchronous PC, synchronous isolated ovarian metastases, and a perforated primary tumor.

Peritoneal carcinomatosis (PC) after curative surgery for colorectal cancer occurs in nearly 10 % of all patients.1–7 Single or multiple tumor deposits on the peritoneum characterize this specific kind of recurrence. It must be distinguished from other types of recurrences (local recurrence—i.e., anastomotic; lymph node and distant metastases) whose treatment and long-term results differ. Patients with PC have a very poor prognosis, and it is often considered a terminal disease. Median survival does not exceed 6 months with older chemotherapies in unselected patients, and survival is less than 24 months with new drug regimens in highly selected patients, corresponding to a 5-year overall survival rate of 5 and 13 %, respectively.7–10 Nevertheless, a recent approach combining complete cytoreductive surgery with hyperthermic intraperitoneal chemotherapy (HIPEC) changed the therapeutic landscape. The latest results demonstrate a 5-year overall survival rate of 42 % and a median survival of 41 months after a mean follow-up of 49 months.11 One major prognostic factor after complete cytoreductive surgery plus HIPEC is the extent of the peritoneal disease evaluated peroperatively with the peritoneal cancer index.8,12–17 In a recent series of 523 patients treated with complete cytoreductive surgery plus HIPEC, the 5-year overall survival rate dropped from 49 % in patients with a low peritoneal cancer index (below 7) to less than 10 % in patients with peritoneal cancer index above 20.12 All efforts should therefore be made to identify patients with PC at the earliest stage. However, the

C. Honore´ et al.

184

current noninvasive assessment (i.e., clinical examination, imaging, or biology) achieves poor results, and only a systematic second-look surgery can detect an early PC.17 This invasive approach cannot be proposed to all patients concerned, but only to high-risk populations if they can be defined. Multiple suspected risk factors for developing PC (Table 1) have been reported in the past and were recently summarized.18 However, we were unable to find corroborative data for most of these suspected risk factors in the literature, which in fact are often confused with prognostic factors and factors associated with PC at the initial diagnosis and therefore lacking in evidence-based data. The aim of this systematic review was to evaluate the influence of factors associated with the primary tumor likely to influence the emergence of recurrent PC after curative surgery for colorectal cancer. METHODS Search Strategy All retrospective and prospective studies explicitly describing the incidence of recurrent PC after curative resection of colorectal cancer since 1940 were eligible for inclusion in this systematic review. All case reports were excluded from the analysis. Study Inclusion and Data Extraction A systematic review of the literature was conducted by two independent reviewers using PubMed, Medline, Embase, Ovid, the Cochrane database, and Google Scholar Search to identify all clinical studies published in English between 1940 and 2011 reporting the incidence of recurrent PC. Combinations of the following search terms were used to identify the studies: ‘‘colorectal cancer,’’ ‘‘colon cancer,’’ ‘‘colon carcinoma,’’ ‘‘peritoneal recurrence,’’ TABLE 1 Characteristics of patients with primary colorectal cancer identified in the literature as being at high risk of developing peritoneal carcinomatosis • Visible evidence of peritoneal carcinomatosis

‘‘peritoneal carcinomatosis,’’ ‘‘ovarian metastasis,’’ ‘‘perforated colon cancer,’’ ‘‘obstructive cancer,’’ ‘‘mucinous adenocarcinoma,’’ ‘‘signet ring cell adenocarcinoma,’’ ‘‘serosal invasion,’’ ‘‘pattern of recurrent colorectal cancer,’’ ‘‘peritoneal cytology,’’ and ‘‘lavage.’’ A three-step eliminative selection was applied (a title-based selection, followed by an abstract-based selection among the studies selected in step 1, in turn followed by a full-text selection among the studies selected in step 2), and studies were excluded if they included noncolorectal cancer, if the site of the recurrence and/or the incidence of recurrent PC were not clearly specified, if patients having undergone a curative resection could not be isolated, and if the patients had received intraperitoneal chemotherapy. All data concerning disease-free and overall survival were excluded from the analysis. Whenever possible, the results of studies that used the Astler–Coller and Dukes classifications were converted into the tumor, node, metastasis staging classification, 6th edition (Union for International Cancer Control, 2002).19,20 When we found multiple publications from institutions reporting the same cohort, only the most recent and complete articles were included. Data Analysis and Quality of Included Studies Once studies specifically reporting the incidence of recurrent PC were identified, we categorized them according to the primary tumor characteristic suspected of influencing the incidence of recurrent PC. Results from both independent reviewers were pooled, and in discrepant cases, incriminated studies were analyzed by all the authors together to decide whether to select the study. The studies were then separately graded from 1 to 5 in terms of the methodological quality: according to the number of patients included, their randomized or nonrandomized nature, and whether they were prospective or retrospective following the Oxford Centre for Evidence Based Medicine levels of evidence (Table 2).21 A meta-analysis was not performed because of the heterogeneity of study methods and because of the techniques used to diagnose PC. RESULTS

• Ovarian metastasis • Positive cytology either before or after cancer resection • Adjacent organ involvement or cancer-induced fistula • pT4 tumors • Perforated cancer (spontaneous or iatrogenic perforation) • Obstructed cancer • pT3 mucinous cancer and pT3 signet ring cell cancer • Positive lateral margins of excision • Bleeding cancer

The initial search strategy identified 6,522 articles. The study selection process is reported in Fig. 1. Most studies were excluded because recurrent PC as the sole site of recurrence was either not reported or because it was mixed with other types of recurrence and could not be isolated. Numerous articles devoted to PC include an intraperitoneal treatment that modifies the natural history of the disease. Sixteen clinical studies were considered informative concerning the risk factors for developing PC and were

High Risk of Peritoneal Carcinomatosis

185

TABLE 2 Oxford Centre for Evidence Based Medicine levels of evidence, March 2009

TABLE 3 Influence of synchronous PC associated with the primary tumor on recurrent PC after a curative resection

Level Description

Study

Years

n

1a

1a SR (with homogeneity) of RCTs

1b

Individual RCT (with narrow confidence interval)

Fujiwara et al.22

2010

4

2a

SR (with homogeneity) of cohort studies

Elias et al.23

2011

28

2b

Individual cohort study (including low quality RCT)

PC peritoneal carcinomatosis

2c

Outcomes research

3a

SR (with homogeneity) of case–control studies

3b 4

Individual case–control study Case series (and poor-quality cohort and case–control studies)

5

Expert opinion without explicit critical appraisal or based on physiology, bench research, or first principles

SR systematic review, RCT randomized controlled trial

MeSH search:

6522 references Title selection 132 references Abstract selection

Recurrence PC incidence 75 % (3/4) 54 % (15/28)

second-look surgery at 12 months, Elias et al.23 showed that 54 % (15 of 28) of all patients with completely resected synchronous PC developed a peritoneal recurrence. The percentages at a greater or a shorter interval between the first surgical procedure and systematic secondlook surgery are not known but are probably strongly influenced by this delay. The other study concerned only four patients with an R0 resection.22 Three of these patients had experienced disease relapsed on the peritoneum after 20 months of follow-up. In conclusion, the incidence of recurrent PC in patients with completely resected synchronous PC ranged between 54 and 75 % with a level of evidence from 3b to 4. Synchronous Ovarian Metastasis

97 references Full text selection 16 references

FIG. 1 Study inclusion scheme

included in the review.5,22–36 All 16 studies were nonrandomized. Three studies were prospective.23,31, 32 Thirteen were retrospective.5,22,24–30,33–36 The total number of patients included was 598. According to the reported incidence of PC, only six influencing factors associated with curative surgery of the primary tumor were usable for further review: synchronous PC completely resected during the primary tumor surgery, synchronous ovarian metastases, a perforated primary tumor, the histological subtype, positive peritoneal cytology, and serosal invasion of the primary tumor.

Synchronous PC (Completely Resected) Synchronous PC is discovered serendipitously in 4.8 % of the patients scheduled for curative surgery.37 Two nonrandomized studies were selected for the peritoneal recurrence analysis; one was one prospective and the other retrospective (Table 3).22,23 After performing systematic

Ovarian metastases are associated in 0.8–7.4 % of all colorectal cancers.22,38,39 This rate increases to 5–9.7 % in autopsy series.40–42 Two studies, one each retrospective and prospective, including 8–16 patients, were selected for the recurrent PC analysis (Table 4).24,27 In the prospective series, 62 % (5 of 8) exhibited macroscopic PC at the systematic second-look procedure 12 months after initial surgery.23 In the retrospective studies, PC had occurred in 56 % (9 of 16) of cases diagnosed at imaging.24 These results only concerned cases without macroscopic PC. This is an important point because ovarian metastases are associated with synchronous PC in 29–72 % of cases.22,24,43 In conclusion, the incidence of recurrent PC in patients with isolated synchronous ovarian metastases ranged between 56 and 62 % with a level of evidence from 3b to 4.

TABLE 4 Influence of synchronous ovarian metastasis on recurrent PC after a curative resection Study

Years

n

Tan et al.24

2010

16

56 % (9/16)

Elias et al.23

2011

8

62 % (5/8)

PC peritoneal carcinomatosis

Recurrent PC incidence

C. Honore´ et al.

186

Perforated Primary Tumor The true incidence of perforated primary tumors is unknown because colonic perforations above the occlusive tumor are frequently mixed with tumor perforations responsible for peritoneal tumor seeding. The overall rate of perforations ranges between 1 and 8 %.27,44–47 Four studies including 11 to 89 patients with a perforated colorectal cancer were selected for the peritoneal recurrence analysis (three retrospective series and one prospective series) (Table 5).23,25–37 In the prospective series of systematic second-look surgery at 12 months, macroscopic PC was found in 27 % (3 of 11) of patients with a perforated tumor.23 Concerning the retrospective series, one older series reported a peritoneal recurrence rate of 19 % in 32 patients with a perforated tumor, and a more recent series reported a rate of 54 % in 13 patients.26 The fourth series, by Cheynel et al., compared 89 perforated tumors (excluding colonic perforations above the occlusive tumor) to 5,462 nonperforated tumors. PC had occurred in 14 % (12 of 89) of the patients with a perforated tumor, versus 6 % (344 of 5,462) of those with a nonperforated tumor (p = 0.04).27 No data were found in the literature concerning iatrogenic rupture of the primary tumor during surgery. However, one study reported the local recurrence rate (but not of PC), which attained 64 % in 17 patients.48 In conclusion, the incidence of recurrent PC after primary tumor perforation ranged between 14 and 54 % with a level of evidence from 3b to 4. Histological Subtype Mucinous adenocarcinoma accounts for 3–15 % of all colorectal cancers.28,49–53 An adenocarcinoma is said to be mucinous when it contains more than 50 % of mucin; if it contains less than 50 %, it is defined as a carcinoma with a mucinous component.28 The signet ring cell carcinoma accounts for 0.1–2.4 % of all colorectal cancers.28,49,54,55 An adenocarcinoma is said to be of the signet ring cell type when more than 50 % of its cells have intracytoplasmic mucin pushing the nucleus; if less than 50 % of the cells are involved, it is a carcinoma with a signet ring cell TABLE 5 Influence of primary tumor perforation on recurrent PC after a curative resection Study

Years

n

Recurrent PC incidence

Willett et al.25

1985

32

19 % (6/32)

Ogawa et al.26

2009

13

54 % (7/13)

Cheynel et al.

2009

89

14 % (12/89)

Elias et al.23

2011

11

27 % (3/11)

27

PC peritoneal carcinomatosis

component, which accounts for 5.1 % of all colorectal cancers.28 Synchronous PC at the time of primary surgery was frequently associated with these histological subtypes in 11–75 % of mucinous adenocarcinoma and in 71 % of signet ring cell carcinoma.28,37,56 When we focused on patients with stages 1–3 disease and after curative resection, only one retrospective study (including 127 patients) was selected for the peritoneal recurrence analysis.28 Peritoneal recurrence was reported in 36 % (5 of 14) when the tumor was a true mucinous adenocarcinoma and in 11 % (5 of 46) when there was a mucinous component. No data were found concerning peritoneal recurrences from signet ring cell carcinoma, and PC had occurred in 27 % (6 of 22) of cases with a signet ring cell component. In conclusion, the incidence of recurrent PC in patients with mucinous adenocarcinoma or with a mucinous/signet ring cell carcinoma component ranged between 11 and 36 % with a level of evidence of 4. Positive Peritoneal Cytology The incidence of positive cytology, all stages (1–4) and detecting methods (light examination, immunohistochemistry, and PCR) included, ranges between 2.2 and 42.7 %.29–34,57–70 A recent meta-analysis including nine studies reported a 15.2 % (125 of 1,182) incidence of positive cytology, demonstrating the poor efficacy of this technique.71 Six studies were selected for the recurrent PC analysis and are reported in Table 6. All studies were nonrandomized (two prospective and four retrospective) and included from 8 to 22 patients with positive cytology.29–34 They concerned patients without PC who had undergone a curative (R0) resection. They concerned patients exhibiting different disease stages. In addition, it is noteworthy that various sampling and analysis techniques were used because there is currently no well-recognized standardized technique. Four of the six studies did not detect a significant difference in the risk of developing PC between patients with positive cytology and those with negative cytology. In conclusion, the incidence of recurrent PC in patients with positive cytology ranged between 9 and 36 % with a level of evidence from 3b to 4. Primary Tumor with Serosal and/or Adjacent Organ Invasion Primary tumors with serosal and/or adjacent organ invasion accounted for 16 % of all colorectal cancers in the 2010 study by Gunderson et al.72 Four studies were selected for the recurrent PC analysis and are reported in Table 7. All studies were nonrandomized and retrospective and included from 6 to 395 patients without synchronous

High Risk of Peritoneal Carcinomatosis

187

TABLE 6 Influence of peritoneal cytology on recurrent PC after a curative resection Study

Years

Hase et al.29

n

Positive cytology (%)

Recurrence PC incidence when cytology is Positive

Negative

p

1998

87a

9

25 % (2/8)

6 % (5/79)

NS

Yamamoto et al.

2003

189b

6

36 % (4/11)

2 % (4/178)

\0.001

Kanellos et al.31

2003

110b

20

23 % (5/22)

8 % (7/88)

NS

c

30

32

Noura et al.

2009

374

3

33 % (5/15)

2 % (7/359)

\0.001

Fujii et al.33

2009

283a

4

9 % (1/11)

2 % (6/272)

NS

Katoh et al.34

2009

226d

4

33 % (3/10)

9 % (19/216)

NS

PC peritoneal carcinomatosis a

Concerns patients with stages 1–4 colorectal disease

b

Concerns patients with stages 1–3 colorectal disease

c

Concerns patients with stages 3–4 colorectal disease

d

Concerns patients with stages 2–4 colorectal disease

TABLE 7 Influence of serosal and/or organ invasion on recurrent PC after a curative resection Study

Years

n

Recurrent PC incidence Serosal and/or organ invasion

Macroscopic serosal invasion

Neighboring organ invasion

Willett et al.5

1984

395

8 % (32/395)

8 % (22/263)

8 % (10/132)

Shepherd et al.35 Luna-Pe´rez et al.36

1997

242

19 % (45/242)

–

–

2002 2009

28 6

– 17 % (1/6)

– –

29 % (8/28) –

Fujii et al.33 PC peritoneal carcinomatosis

PC.5,33,35,36 The rate of recurrent PC ranged between 8 and 19 %. We tried to distinguish two different subgroups among these patients: those with macroscopic serosal invasion, and those with macroscopic involvement of an adjacent organ. This distinction was only made in one of the four studies where there was no significant difference in the rate of recurrent PC between the two subgroups.5 One study that focused exclusively on primary tumors invading an adjacent organ found a higher rate of recurrent PC.36 In conclusion, the incidence of recurrent PC in patients with serosal and/or adjacent organ invasion ranged between 8 and 17 % with a level of evidence of 4. Other Factors We failed to find any evidence-based correlation for other factors suspected of leading to a high risk for recurrent PC. One retrospective study of nearly 700 patients clearly stated that although lymph node invasion was a strong prognostic factor, it was not a risk factor for peritoneal recurrence.32 This study also reported that the tumor location had no influence on the peritoneal recurrence rate. No study ever reported that a higher PC rate was associated with laparoscopic colorectal surgery.73–76

Although many experimental studies conducted on animals suggest that peritoneal tumor seeding is linked to surgical trauma, no human study has ever confirmed this relationship. No conclusion can be drawn regarding this point.77 None of the numerous large series of patients with an occlusive tumor directly analyzed its influence on the risk of developing recurrent PC. However, there was no mention of an increased peritoneal recurrence in that setting.44,78–88 Finally, no study was found that associated a bleeding tumor with the development of recurrent PC. A study by McArdle et al.89 in 2006 concerning 176 patients treated for a bleeding tumor under emergency conditions reported lower overall survival when these patients were operated on under emergency circumstances (43.8 vs. 61.5 % 5-year overall survival, p \ 0.001) without providing information about disease-free survival and the type of recurrence. DISCUSSION This systematic review aimed to identify patients who truly exhibit a high risk of PC after curative resection of their primary colorectal cancer on the basis of evidencebased data. When we looked at the extensive literature on

C. Honore´ et al.

188

the behavior of colorectal cancers, a considerable amount was in fact devoted to the analysis of overall or local recurrences, but few studies were specifically focused on peritoneal recurrences and influencing factors. We only found 16 publications that clearly specified the incidence of recurrent postoperative PC.5,22–36 The available literature is poor regarding this subject, and we were obliged to admit that many unfounded assertions on factors leading to the development of PC have been reported in the literature without solid proof. Indeed, we find it astonishing that factors such as the extent of lymph node involvement, a high carcinoembryonic antigen level, an occlusive or a bleeding tumor, presented in studies as predisposing factors to PC, were not substantiated by data to confirm such claims. These factors predispose patients to developing recurrences in general and not PC in particular, but these uncorroborated claims remain strongly anchored in surgeons’ minds. In fact, there is a paucity of scientific data in the literature concerning a high risk of developing recurrent PC, and to our knowledge, this is the first systematic review to address this issue. The main bias when studying the incidence of recurrent PC, and therefore of this review, is the poor reliability of the noninvasive diagnostic tools used to detect PC. The best results, achieved with computed tomographic (CT) scan, demonstrate a sensitivity of only 60–79 %, but this drops to below 30 % when peritoneal lesions are smaller than 5 mm.90 This fact is supported by the 56 % of PC found in a systematic second-look study in patients with a negative clinical, biological, and radiological assessment.23 In earlier studies, when CT scanning was not widely used, the diagnosis of peritoneal recurrence was mainly based on the clinical and radiological examinations available at the time. Although some studies included a diagnostic laparotomy (25 % of the cases), we can safely assume that the actual rate was highly underestimated.25 This underestimation is confirmed by a study that compared the PC rate at autopsy (55 %) to the clinical premortem diagnosis (7 %).4 As the diagnosis of recurrent PC is closely related to the follow-up strategy (tools applied and duration), we considered that the best option was a systematic laparotomy at a standardized time. To date, only three studies have proposed laparotomy as a diagnostic tool.17,91,92 Without systematically performing CT scans, Minton et al.92 performed second-look surgery based on an increase in the carcinoembryonic antigen level or on clinical symptoms and found recurrences in 98 % of patients, among whom 10 % were exclusively peritoneal. Before the era of imaging, Gunderson et al.91, based on Wangensteen’s93 pioneering work, performed systematic second-look surgery 6–12 months after curative surgery in 91 patients with a transmural colorectal cancer and/or with associated lymph node metastases. They found a 33 % rate of

recurrent PC in an overall 64 % intra-abdominal recurrence rate. The authors concluded that the incidence of locoregional and/or peritoneal failure in patients with clinically detected distant metastasis was underestimated in the absence of reoperation or autopsy.92 In our opinion, the most convincing data are to be found in the only prospective second-look surgery series, which explored patients with a negative preoperative assessment 1 year after resection of the primary tumor.17,23 Using the available evidence-based data and considering the quality of the studies, we were unable to identify strong evidence-based predictive factors for recurrent PC after curative surgery for colorectal cancer. Nevertheless, we noticed a wide disparity in the incidence of recurrent PC according to the reported risk factors, and although this is not strongly supported by the literature, in our opinion, patients’ risk categories can be identified as follows. High-Risk Patients Patients with isolated synchronous ovarian metastases and with synchronous PC (completely resected) run a risk of developing a peritoneal recurrence greater than 50 %.22,23 These data were clearly confirmed by a systematic laparotomy at 12 months.23 Patients with a perforated primary tumor run a lower risk of developing PC, ranging between 14 and 54 %.23,25–27 However, this probably underestimates the true incidence of recurrent PC. The only study with a systematic second-look laparotomy, which included the colonic perforations above the occlusive tumor theoretically associated with a lower risk, reported a 27 % peritoneal recurrence rate at 12 months.23 Iatrogenic perforations, because the way of peritoneal seeding occurs is similar, could be comparable in terms of the risk of recurrent PC to ‘‘naturally’’ perforated primary tumor. Low-Risk Patients No evidence of an increased incidence of recurrent PC was found for some patient categories; patients with a tumor invading the serosa and/or an adjacent organ run a risk of developing recurrent PC of between 8 and 19 %, which is comparable to the risk in the general population.5, 33,35,36 This risk is astonishingly low and one explanation might be the en bloc resection of these tumors aimed at preventing tumor seeding. In this regard, the two studies that mentioned recurrent PC after tumor invasion of an adjacent organ reported controversial results.5,36 Because the rate of tumor rupture during surgery was not specified in either study, we assume that it could be responsible for this disparity. Another possible explanation for this low rate might also be a higher risk of distant recurrence with

High Risk of Peritoneal Carcinomatosis

these locally advanced tumors, which could obliterate or supersede the onset of PC. There is no evidence to indicate that the risk of developing PC is increased in the case of occlusive primary tumors compared to nonocclusive tumors. During the 1970s, several large retrospective studies were published concerning occlusive tumors.44,78–89 Recurrence rates were reported, but PC was not singled out. For the 124 patients with an occlusive tumor (among 1,566 colorectal patients) in the study by Welch and Donaldson44, the 5-year survival rate was similar in patients who had undergone an R0 resection irrespective of whether they had an occlusive tumor. In the series reported by Ragland et al. 86 concerning 227 of 1,137 patients with occlusion, the prognosis for occlusive tumors was not different from that of nonocclusive tumors, and like these latter lesions was mainly correlated with lymph node involvement. The conclusions are similar for bleeding tumors: there is no evidence to indicate that the risk of recurrent PC is higher. Unknown-Risk Patients For the following subgroups of patients, the diagnostic unreliability combined with the low number of patients and the totally controversial results prevented us from considering these data as informative until further studies become available. Patients with positive peritoneal cytology run an average risk of 26 % of developing PC, but the range is wide: from 9 to 36 %, according to the studies.29–34,57–70 These controversial results could be because there are no standardized techniques for peritoneal cytology sampling, and interpretation with varying accuracies has not yet been compared. Another important point is that among the 1,250 patients reported in the six studies selected, only 73 patients exhibited positive cytology, raising the question of the limited efficacy of this technique in routine use for colorectal cancer.29–34,57–70 The histological subtypes, mucinous and signet ring cell, with an 11 to 36 % risk of recurrent PC, respectively, could be categorized in the intermediate risk group, but only one retrospective study was available, and it provided no clear explanation of the method used for the diagnosis of PC.28 With the categorization of the risk of recurrent PC, such information will only be beneficial if we offer an appropriate strategy that could prevent the onset of PC, detect it earlier, or treat it efficiently. One prevention option is to intensify systemic adjuvant chemotherapy, but no study in the literature has ever demonstrated the benefit of such a strategy. Nevertheless, in a high-risk population, the use of a standard adjuvant 5-fluorouracil plus oxaliplatin- or irinotecan-based chemotherapy regimens after the primary tumor resection did not prevent the onset of PC in half of the patients systematically reoperated on at 12 months.23

189

We believe that PC might somehow be resistant to systemic chemotherapy, but this is highly debated in the literature. Another preventive option would be to directly administer intraperitoneal therapy at the end of surgery to ‘‘high-risk’’ patients. This is controversial because the associated increased morbidity and mortality cannot be justified if a clear survival benefit is not demonstrated. No study in the literature has ever used such prophylactic treatment. Nevertheless, in our center, we believe that patients with synchronous PC and/or synchronous ovarian metastases run such a high risk of recurrent PC that they should receive HIPEC directly at the end of surgery of the primary tumor, or if intraperitoneal therapy is not possible, patients should be referred to an expert center without resection of the primary tumor (unless it is symptomatic). Further studies are needed in other situations, especially in case of iatrogenic tumor rupture. When prevention fails, another option could be a closer follow-up in order to diagnose recurrent PC early and treat it with complete cytoreductive surgery plus HIPEC. The main drawback of this approach is the poor reliability of current imaging. The best sensitivity results obtained with CT are below 30 % when the peritoneal lesions are smaller than 5 mm.90 Relying exclusively on this approach means that PC will be diagnosed at a later stage, which compromises long-term survival because the extent of peritoneal disease is a major prognostic factor. Considering the unreliability of radiological, clinical, and biological examinations in detecting PC, we believe that the best option is to propose systematic second-look surgery to diagnose and treat recurrent PC early.17 This strategy can only be proposed to patients who are truly at a high risk of developing recurrent PC, and a clear evidence-based definition of these patients is required. After the encouraging preliminary results of systematic second-look surgery plus HIPEC concerning selected high-risk patients and long-term results, we think that the time has come to validate this strategy with a prospective randomized trial comparing radiological surveillance alone, the current standard of care, with this second-look strategy.23 In conclusion, this study underlines the current scarcity of available scientific data. The current published literature does not allow identification of any strong evidence-based factors predictive of recurrent PC after curative surgery. Nevertheless, we found clues about a disparity in incidence of recurrent PC associated with the primary tumor. Emerging data indicate three situations that could be associated with a real higher risk of recurrent PC: synchronous PC, synchronous isolated ovarian metastases, and a perforated primary tumor. ACKNOWLEDGMENT editing.

The authors thank Lorna Saint Ange for

C. Honore´ et al.

190 CONFLICT OF INTEREST flicts of interest to disclose.

The authors have no potential con-

REFERENCES 1. Cass AW, Million RR, Pfaff WW. Patterns of recurrence following surgery alone for adenocarcinoma of the colon and rectum. Cancer. 1976;37:2861–5. 2. Minsky BD, Mies C, Recht A, Rich TA, Chaffey JT. Resectable adenocarcinoma of the rectosigmoid and rectum. I. Patterns of failure and survival. Cancer. 1988;61:1408–16. 3. Mendenhall WM, Million RR, Pfaff WW. Patterns of recurrence in adenocarcinoma of the rectum and rectosigmoid treated with surgery alone: implications in treatment planning with adjuvant radiation therapy. Int J Radiat Oncol Biol Phys. 1983;9:977–85. 4. Gilbert JM, Jeffrey I, Evans M, Kark AE. Sites of recurrent tumour after ‘‘curative’’ colorectal surgery: implications for adjuvant therapy. Br J Surg. 1984;71:203–5. 5. Willett CG, Tepper JE, Cohen AM, Orlow E, Welch CE. Failure patterns following curative resection of colonic carcinoma. Ann Surg. 1984;200:685–90. 6. Russell AH, Tong D, Dawson LE, Wisbeck W. Adenocarcinoma of the proximal colon. Sites of initial dissemination and patterns of recurrence following surgery alone. Cancer. 1984;53:360–7. 7. Jayne DG, Botterill I, Ambrose NS, Brennan TG, Guillou PJ, O’Riordain DS. Peritoneal carcinomatosis from colorectal cancer. Br J Surg. 2002;89:428–32. 8. Elias D, Lefevre JH, Chevalier J, et al. Complete cytoreductive surgery plus intraperitoneal chemohyperthermia with oxaliplatin for peritoneal carcinomatosis of colorectal origin. J Clin Oncol. 2009;27:681–5. 9. Verwaal VJ, Bruin S, Boot H, van Slooten G, van Tinteren H. 8-Year follow-up of randomized trial: cytoreduction and hyperthermic intraperitoneal chemotherapy versus systemic chemotherapy in patients with peritoneal carcinomatosis of colorectal cancer. Ann Surg Oncol. 2008;15:2426–32. 10. Quenet F, Goe´re´ D, Mehta S, Roca L, Dumont F, Essissen M. Two bi-institutional prospective phase II trials using intraperitoneal oxaliplatin with or without irinotecan during HIPEC following complete cytoreductive surgery for colorectal carcinomatosis. Ann Surg. 2011;254:294–301. 11. Elias D, Gilly F, Boutitie F, et al. Peritoneal colorectal carcinomatosis treated with surgery and perioperative intraperitoneal chemotherapy: retrospective analysis of 523 patients from a multicentric French study. J Clin Oncol. 2010;28:63–8. 12. Elias D, Glehen O, Pocard M, et al.; Association Franc¸aise de Chirurgie. A comparative study of complete cytoreductive surgery plus intraperitoneal chemotherapy to treat peritoneal dissemination from colon, rectum, small bowel, and nonpseudomyxoma appendix. Ann Surg. 2010;251:896–901. 13. Sugarbaker PH, Schellinx ME, Chang D, Koslowe P, von Meyerfeldt M. Peritoneal carcinomatosis from adenocarcinoma of the colon. World J Surg. 1996;20:585–92. 14. Sugarbaker PH, Chang D. Results of treatment of 385 patients with peritoneal surface spread of appendiceal malignancy. Ann Surg Oncol. 1999;6:727–31. 15. Glehen O, Kwiatkowski F, Sugarbaker PH, et al. Cytoreductive surgery combined with intraperitoneal chemotherapy for management of peritoneal carcinomatosis from colorectal cancer: a multi-institutional study. J Clin Oncol. 2004;22:3284–92. 16. Elias D, Raynard B, Farkhondeh F, et al. Peritoneal carcinomatosis of colorectal origin. Long term results of intraperitoneal chemohyperthermia with oxaliplatin following complete cytoreductive surgery. Gastroenterol Clin Biol. 2006;30:1200–4.

17. Elias D, Goe´re´ D, Di Pietrantonio D, et al. Results of systematic second-look surgery in patients at high risk of developing colorectal peritoneal carcinomatosis. Ann Surg. 2008;247:445–50. 18. Sugarbaker PH. Revised guidelines for second-look surgery in patients with colon and rectal cancer. Clin Transl Oncol. 2010;12:621–8. 19. American Joint Comitee on Cancer. AJCC cancer staging manual. 6th ed. New York: Springer; 2002. 20. Sobin LH, Wittekind C, editors. UICC TNM classification of malignant tumors. 6th ed. New York: Wiley-Liss; 2002. 21. Howick J, Phillips B, Ball C, et al. Oxford Centre for EvidenceBased Medicine. Levels of evidence. March 2009. http://www.cebm.net/index.aspx?o=1025. Accessed 1 Oct 2011 22. Fujiwara A, Noura S, Ohue M, et al. Significance of the resection of ovarian metastasis from colorectal cancers. J Surg Oncol. 2010;102:582–7. 23. Elias D, Honore´ C, Dumont F, et al. Results of systematic secondlook surgery plus HIPEC in asymptomatic patients presenting a high risk of developing colorectal peritoneal carcinomatosis. Ann Surg. 2011;254:289–93. 24. Tan KL, Tan WS, Lim JF, Eu KW. Krukenberg tumors of colorectal origin: a dismal outcome—experience of a tertiary center. Int J Colorectal Dis. 2010;25:233–8. 25. Willett C, Tepper JE, Cohen A, Orlow E, Welch C. Obstructive and perforative colonic carcinoma: patterns of failure. J Clin Oncol. 1985;3:379–84. 26. Ogawa M, Watanabe M, Eto K, et al. Clinicopathological features of perforated colorectal cancer. Anticancer Res. 2009;29:1681–4. 27. Cheynel N, Cortet M, Lepage C, Ortega-Debalon P, Faivre J, Bouvier AM. Incidence, patterns of failure, and prognosis of perforated colorectal cancers in a well-defined population. Dis Colon Rectum. 2009;52:406–11. 28. Pande R, Sunga A, Levea C, et al. Significance of signet-ring cells in patients with colorectal cancer. Dis Colon Rectum. 2008;51:50–5. 29. Hase K, Ueno H, Kuranaga N, Utsunomiya K, Kanabe S, Mochizuki H. Intraperitoneal exfoliated cancer cells in patients with colorectal cancer. Dis Colon Rectum. 1998;41:1134–40. 30. Yamamoto S, Akasu T, Fujita S, Moriya Y. Long-term prognostic value of conventional peritoneal cytology after curative resection for colorectal carcinoma. Jpn J Clin Oncol. 2003;33:33–7. 31. Kanellos I, Demetriades H, Zintzaras E, Mandrali A, Mantzoros I, Betsis D. Incidence and prognostic value of positive peritoneal cytology in colorectal cancer. Dis Colon Rectum. 2003;46:535–9. 32. Noura S, Ohue M, Seki Y, Yano M, Ishikawa O, Kameyama M. Long-term prognostic value of conventional peritoneal lavage cytology in patients undergoing curative colorectal cancer resection. Dis Colon Rectum. 2009;52:1312–20. 33. Fujii S, Shimada H, Yamagishi S, et al. Evaluation of intraperitoneal lavage cytology before colorectal cancer resection. Int J Colorectal Dis. 2009;24:907–14. 34. Katoh H, Yamashita K, Sato T, Ozawa H, Nakamura T, Watanabe M. Prognostic significance of peritoneal tumour cells identified at surgery for colorectal cancer. Br J Surg. 2009;96:769–77. 35. Shepherd NA, Baxter KJ, Love SB. The prognostic importance of peritoneal involvement in colonic cancer: a prospective evaluation. Gastroenterology. 1997;112:1096–102. 36. Luna-Pe´rez P, Rodrı´guez-Ramı´rez SE, de la Barrera MG, Zeferino M, Labastida S. Multivisceral resection for colon cancer. J Surg Oncol. 2002;80:100–4. 37. Lemmens VE, Klaver YL, Verwaal VJ, Rutten HJ, Coebergh JW, de Hingh IH. Predictors and survival of synchronous peritoneal carcinomatosis of colorectal origin: a population-based study. Int J Cancer. 2011;128:2717–25.

High Risk of Peritoneal Carcinomatosis 38. Birnkrant A, Sampson J, Sugarbaker PH. Ovarian metastasis from colorectal cancer. Dis Colon Rectum. 1986;29:767–71. 39. Hanna NN, Cohen AM. Ovarian neoplasms in patients with colorectal cancer: understanding the role of prophylactic oophorectomy. Clin Colorectal Cancer. 2004;3:215–22. 40. Buirge RE. Carcinoma of the large intestine. Review of four hundred and sixteen autopsy records. Arch Surg. 1941;42:801–18. 41. Abrams HL, Spiro R, Goldstein N. Metastases in carcinoma; analysis of 1000 autopsied cases. Cancer. 1950;3:74–85. 42. Ko¨ves I, Va´mosi-Nagy I, Besznya´k I. Ovarian metastases of colorectal tumours. Eur J Surg Oncol. 1993;19:633–5. 43. Chung TS, Chang HJ, Jung KH, et al. Role of surgery in the treatment of ovarian metastases from colorectal cancer. J Surg Oncol. 2009;100:570–4. 44. Welch JP, Donaldson GA. Perforative carcinoma of colon and rectum. Ann Surg. 1974;180:734–40. 45. Mandava N, Kumar S, Pizzi WF, Aprile IJ. Perforated colorectal carcinomas. Am J Surg. 1996;172:236–8. 46. Lee IK, Sung NY, Lee YS, et al. The survival rate and prognostic factors in 26 perforated colorectal cancer patients. Int J Colorectal Dis. 2007;22:467–73. 47. Schwenter F, Morel P, Gervaz P. Management of obstructive and perforated colorectal cancer. Expert Rev Anticancer Ther. 2010;10:1613–9. 48. Slanetz CA. The effect of inadvertent intraoperative perforation on survival and recurrence in colorectal cancer. Dis Colon Rectum. 1984;27:792–7. 49. Kang H, O’Connell JB, Maggard MA, Sack J, Ko CY. A 10-year outcomes evaluation of mucinous and signet-ring cell carcinoma of the colon and rectum. Dis Colon Rectum. 2005;48:1161–8. 50. Du W, Mah JT, Lee J, Sankila R, Sankaranarayanan R, Chia KS. Incidence and survival of mucinous adenocarcinoma of the colorectum: a population-based study from an Asian country. Dis Colon Rectum. 2004;47:78–85. 51. Umpleby HC, Ranson DL, Williamson RC. Peculiarities of mucinous colorectal carcinoma. Br J Surg. 1985;72:715–8. 52. Halvorsen TB, Seim E. Influence of mucinous components on survival in colorectal adenocarcinomas: a multivariate analysis. J Clin Pathol. 1988;41:1068–72. 53. Yamamoto S, Mochizuki H, Hase K, et al. Assessment of clinicopathologic features of colorectal mucinous adenocarcinoma. Am J Surg. 1993;166:257–61. 54. Wolfman EF, Astler VB, Coller FA. Mucoid adenocarcinoma of the colon and rectum. Surgery. 1957;42:846–52. 55. Okuno M, Ikehara T, Nagayama M, Kato Y, Yui S, Umeyama K. Mucinous colorectal carcinoma: clinical pathology and prognosis. Am Surg. 1988;54:681–5. 56. Nissan A, Guillem JG, Paty PB, Wong WD, Cohen AM. Signetring cell carcinoma of the colon and rectum: a matched control study. Dis Colon Rectum. 1999;42:1176–80. 57. Moore GE, Sako K. The spread of carcinoma of the colon and rectum: a study of invasion of blood vessels, lymph nodes and the peritoneum by tumors cells. Dis Colon Rectum. 1959;2:92–7. 58. Quan SH. Cul-de-sac smears; a follow-up study. Surgery. 1966;60:1170–4. 59. Urdaneta LF, Hoyne RF, Loh PM, Fidler R, McGrady D. A search for malignant cells in peritoneal fluid from patients with colorectal adenocarcinoma. Am Surg. 1983;49:76–81. 60. Martin JK Jr, Goellner JR. Abdominal fluid cytology in patients with gastrointestinal malignant lesions. Mayo Clin Proc. 1986;61:467–71. 61. Ambrose NS, MacDonald F, Young J, Thompson H, Keighley MR. Monoclonal antibody and cytological detection of free malignant cells in the peritoneal cavity during resection of colorectal cancer—Can monoclonal antibodies do better? Eur J Surg Oncol. 1989;15:99–102.

191 62. Horattas MC, Evasovich MR, Topham N. Colorectal carcinoma and the relationship of peritoneal cytology. Am J Surg. 1997; 174:334–7. 63. Leather AJ, Kocjan G, Savage F, et al. Detection of free malignant cells in the peritoneal cavity before and after resection of colorectal cancer. Dis Colon Rectum. 1994;37:814–9. 64. Wind P, Norklinger B, Roger V, Kahlil A, Guin E, Parc R. Longterm prognostic value of positive peritoneal washing in colon cancer. Scand J Gastroenterol. 1999;34:606–10. 65. Vogel P, Ru¨schoff J, Ku¨mmel S, et al. Prognostic value of microscopic peritoneal dissemination: comparison between colon and gastric cancer. Dis Colon Rectum. 2000;43:92–100. 66. Yang SH, Lin JK, Lai CR, et al. Risk factors for peritoneal dissemination of colorectal cancer. J Surg Oncol. 2004;87:167–73. 67. Lloyd JM, McIver CM, Stephenson SA, Hewett PJ, Rieger N, Hardingham JE. Identification of early-stage colorectal cancer patients at risk of relapse post-resection by immunobead reverse transcription-PCR analysis of peritoneal lavage fluid for malignant cells. Clin Cancer Res. 2006;12:417–23. 68. Gozalan U, Yasti AC, Yuksek YN, Reis E, Kama NA. Peritoneal cytology in colorectal cancer: incidence and prognostic value. Am J Surg. 2007;193:672–5. 69. Lee IK, Kim do H, Gorden DL, et al. Prognostic value of CEA and CA 19-9 tumor markers combined with cytology from peritoneal fluid in colorectal cancer. Ann Surg Oncol. 2009;16: 861–70. 70. Nishikawa T, Watanabe T, Sunami E, Tsuno NH, Kitayama J, Nagawa H. Prognostic value of peritoneal cytology and the combination of peritoneal cytology and peritoneal dissemination in colorectal cancer. Dis Colon Rectum. 2009;52:2016–21. 71. Rekhraj S, Aziz O, Prabhudesai S, et al. Can intra-operative intraperitoneal free cancer cell detection techniques identify patients at higher recurrence risk following curative colorectal cancer resection? A meta-analysis. Ann Surg Oncol. 2008;15: 60–8. 72. Gunderson LL, Jessup JM, Sargent DJ, Greene FL, Stewart A. Revised tumor and node categorization for rectal cancer based on Surveillance, Epidemiology, and End Results and rectal pooled analysis outcomes. J Clin Oncol. 2010;28:256–63. 73. Lujan HJ, Plasencia G, Jacobs M, Viamonte M III, Hartmann RF. Long-term survival after laparoscopic colon resection for cancer: complete five-year follow-up. Dis Colon Rectum. 2002;45: 491–501. 74. Patankar SK, Larach SW, Ferrara A, et al. Prospective comparison of laparoscopic vs. open resections for colorectal adenocarcinoma over a ten-year period. Dis Colon Rectum. 2003;46:601–11. 75. Kim SH, Park IJ, Joh YG, Hahn KY. Laparoscopic resection of rectal cancer: a comparison of surgical and oncologic outcomes between extraperitoneal and intraperitoneal disease locations. Dis Colon Rectum. 2008;51:844–51. 76. Wong MT, Ng KH, Lim JF, Ooi BS, Tang CL, Eu KW. 418 Cases of laparoscopic colorectal resections: a single-institution experience and literature review. Singap Med J. 2010;51:650–4. 77. Oosterling SJ, van der Bij GJ, van Egmond M, van der Sijp JR. Surgical trauma and peritoneal recurrence of colorectal carcinoma. Eur J Surg Oncol. 2005;31:29–37. 78. Goligher JC, Smiddy FG. The treatment of acute obstruction or perforation with carcinoma of the colon and rectum. Br J Surg. 1957;45:270–4. 79. Ulin AW, Ehrlich EW. Current views related to management of large bowel obstruction caused by carcinoma of the colon. Am J Surg. 1962;104:463–7. 80. Chang WY, Burnett WE. Complete colonic obstruction due to adenocarcinoma. Surg Gynecol Obstet. 1962;114:353–6. 81. Minster JJ. Comparison of obstructing and nonobstructing carcinoma of the colon. Cancer. 1964;17:242–7.

192 82. Loefler I, Hafner CD. Survival rate in obstructing carcinoma of colon. Arch Surg. 1964;89:716–8. 83. Hickey RC, Hyde HP. Neoplastic obstruction of the large bowel. Surg Clin N Am. 1965;45:1157–63. 84. Floyd CE, Cohn I Jr. Obstruction in cancer of the colon. Ann Surg. 1967;165:721–31. 85. Watters NA. Survival after obstruction of the colon by carcinoma. Can J Surg. 1969;12:124–8. 86. Ragland JJ, Londe AM, Spratt JS Jr. Correlation of the prognosis of obstructing colorectal carcinoma with clinical and pathologic variables. Am J Surg. 1971;121:552–6. 87. Glenn F, McSherry CK. Obstruction and perforation in colorectal cancer. Ann Surg. 1971;173:983–92. 88. Korenaga D, Ueo H, Mochida K, et al. Prognostic factors in Japanese patients with colorectal cancer: the significance of large bowel obstruction—univariate and multivariate analyses. J Surg Oncol. 1991;47:188–92.

C. Honore´ et al. 89. McArdle CS, McMillan DC, Hole DJ. The impact of blood loss, obstruction and perforation on survival in patients undergoing curative resection for colon cancer. Br J Surg. 2006;93:483–8. 90. Dromain C, Leboulleux S, Auperin A, et al. Staging of peritoneal carcinomatosis: enhanced CT vs. PET/CT. Abdom Imaging. 2008;33:87–93. 91. Gunderson LL, Sosin H, Levitt S. Extrapelvic colon—areas of failure in a reoperation series: implications for adjuvant therapy. Int J Radiat Oncol Biol Phys. 1985;11:731–41. 92. Minton JP, Hoehn JL, Gerber DM, et al. Results of a 400-patient carcinoembryonic antigen second-look colorectal cancer study. Cancer. 1985;55:1284–90. 93. Wangensteen OH. Cancer of the colon and rectum; with special reference to earlier recognition of alimentary tract malignancy; secondary delayed re-entry of the abdomen in patients exhibiting lymph node involvement; subtotal primary excision of the colon; operation in obstruction. Wis Med J. 1949;48:591–7.