Key points

Introduction

Colorectal cancer is the third most common cancer and cause of cancer-related deaths in the United States. Approximately 40% and 20% of patients have regional and distant metastases at the time of diagnosis, respectively. Approximately 4% to 5% of patients have synchronous peritoneal metastasis at the initial presentation, accounting for 20% to 25% of all stage IV disease. Another 3% to 5% of patients develop metachronous peritoneal metastasis after curative resection of stage I through III diseases. Although the incidence of developing peritoneal metastasis in patients who initially presented with liver metastasis alone is not well defined, 10% to 20% of patients with metastatic disease will have synchronous and/or metachronous peritoneal disease. Based on a large study pooling 2101 patients from the 2 North Central Cancer Treatment Group (NCCTG) N9741 and N9841 trials, 17% of patients with stage IV diseases had peritoneal metastases, whereas 2% of cases had isolated peritoneal metastases.

Peritoneal metastasis of colorectal origin poses a significant clinical challenge. Historically, when 5-fluorouracil (5-FU) was the only available treatment for colorectal cancer, the median overall survival was 6 to 9 months and the 5-year overall survival was 1%. Before the availability of multiple anticancer agents, investigators used cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS-HIPEC) for peritoneal metastasis of colorectal origin and reported a median survival for treated patients of between 12.8 to 60.1 months and a 5-year survival rate of 11% to 48%. Some treatment centers have also included early postoperative intraperitoneal chemotherapy (EPIC) as part of the regimen. With the addition of multiple anticancer agents, the median overall survival for patients with metastatic colon cancer is approximately 2 years, and 20% of patients will live for 5 years or longer. Given the improvement of current systemic treatment with various targeted therapies, the role of CRS-HIPEC and EPIC compared with systemic chemotherapy is debated.

Introduction

Colorectal cancer is the third most common cancer and cause of cancer-related deaths in the United States. Approximately 40% and 20% of patients have regional and distant metastases at the time of diagnosis, respectively. Approximately 4% to 5% of patients have synchronous peritoneal metastasis at the initial presentation, accounting for 20% to 25% of all stage IV disease. Another 3% to 5% of patients develop metachronous peritoneal metastasis after curative resection of stage I through III diseases. Although the incidence of developing peritoneal metastasis in patients who initially presented with liver metastasis alone is not well defined, 10% to 20% of patients with metastatic disease will have synchronous and/or metachronous peritoneal disease. Based on a large study pooling 2101 patients from the 2 North Central Cancer Treatment Group (NCCTG) N9741 and N9841 trials, 17% of patients with stage IV diseases had peritoneal metastases, whereas 2% of cases had isolated peritoneal metastases.

Peritoneal metastasis of colorectal origin poses a significant clinical challenge. Historically, when 5-fluorouracil (5-FU) was the only available treatment for colorectal cancer, the median overall survival was 6 to 9 months and the 5-year overall survival was 1%. Before the availability of multiple anticancer agents, investigators used cytoreductive surgery and hyperthermic intraperitoneal chemotherapy (CRS-HIPEC) for peritoneal metastasis of colorectal origin and reported a median survival for treated patients of between 12.8 to 60.1 months and a 5-year survival rate of 11% to 48%. Some treatment centers have also included early postoperative intraperitoneal chemotherapy (EPIC) as part of the regimen. With the addition of multiple anticancer agents, the median overall survival for patients with metastatic colon cancer is approximately 2 years, and 20% of patients will live for 5 years or longer. Given the improvement of current systemic treatment with various targeted therapies, the role of CRS-HIPEC and EPIC compared with systemic chemotherapy is debated.

Management with systemic chemotherapy alone

With the addition of oxaliplatin, irinotecan, and biologics including bevacizumab, cetuximab and panitumumab, the overall survival of patients with metastatic disease has notably increased. However, the impact of peritoneal metastasis on the treatment efficacy and overall survival is not clearly defined for several reasons: (1) most of the published series mixed all of the patients with stage IV disease, (2) peritoneal metastases appear in less than 20% of patients with stage IV, and (3) peritoneal metastases are often associated with other sites of metastasis. Nevertheless, a large study specifically examined the survival of patients with peritoneal metastasis by pooling 2101 with stage IV disease from the 2 NCCTG N9741 and N9841 trials. In this study, 364 patients (17%) had peritoneal metastasis and 1731 patients (83%) had nonperitoneal metastasis. No significant difference was seen between the groups in terms of age, gender, Eastern Cooperative Oncology Group status, and chemotherapeutic agents used. The only statistically significant differences were a higher number of metastatic sites and a lower number of associated liver and lung metastasis in patients with peritoneal metastasis. Patients with peritoneal metastasis had a lower median overall survival despite having a lower incidence of synchronous liver or lung metastasis: 12.7 versus 17.6 months in patients without peritoneal metastasis. The 5-year survival rate was 4% for patients with peritoneal metastasis versus 6% for patients without peritoneal metastasis. Among patients treated with first-line FOLFOX (5-FU, leucovorin, oxaliplatin), those with peritoneal metastasis had a median overall survival of 15.7 versus 20.9 months in those without peritoneal metastasis. Although many patients in these studies were accrued before the benefit of second-line bevacizumab and the correct use of cetuximab and panitumumab in patients with KRAS wild-type tumors were understood, the results highlight the poor prognosis for patients with peritoneal metastases.

Management with cytoreductive surgery, hyperthermic intraperitoneal chemotherapy, and early postoperative intraperitoneal chemotherapy

CRS-HIPEC was introduced as a treatment option for peritoneal carcinomatosis disseminated from colorectal cancer in 1990s. It involves surgical removal of all peritoneal surface tumors and involved organs. At the time of surgery, the abdominal cavity is fully examined to determine the extent of disease using the peritoneal carcinomatosis index (PCI) scoring system. PCI is calculated based on the number of regions involved and the maximum size of tumor deposits in each region; scores range from 0 to 36. The completeness of cytoreduction is defined as removal of all tumors more than 2.5 mm in maximum diameter (CC0 and CC1). CC0 indicates no residual disease, whereas CC2 indicates residual tumors more than or equal to 2.5 mm in size. After CRS, all intraperitoneal surfaces are exposed to heated chemotherapy. The chemotherapeutic agents, incubating temperature, duration of exposure, and techniques used vary among centers.

EPIC with various chemotherapeutic agents has also been used in some centers. In a multi-institutional registry study, Glehen and colleagues collected data on 506 patients who underwent CRS-HIPEC with or without EPIC from 28 institutions. Most of these patients had adenocarcinoma of colorectal origin, except 3% of the cohort who had an unknown primary site. Complete cytoreduction could be achieved in 75% of patients (CC0: N = 271, CC1: N = 106). HIPEC and EPIC were performed in 76% and 46% of patients, respectively; 22% of patients had both HIPEC and EPIC. HIPEC was commonly performed with mitomycin C (71%). Other regimens included mitomycin C with cisplatin (13%) and oxaliplatin (8%). The regimen of 5-FU with or without mitomycin C (96%) was commonly used for EPIC. With a median follow-up of 53 months, the median overall survival was 19.2 months and the overall 1-, 3-, and 5-year survival rates were 72%, 39%, and 19%, respectively. In patients with no residual disease after cytoreduction (ie, CC0) or low burden of disease at initial exploration, the median overall survival could be as high as 32.4 or 34.8 months, respectively. Notably, 54% of patients received preoperative chemotherapy, but no difference in overall survival was seen. In contrast, patients who received postoperative 5-FU–based chemotherapy (40%) had a better median overall survival (25.2 vs 15.6 months; P = .021). However, no statistically significant difference was seen among patients treated with HIPEC, EPIC, or combined HIPEC/EPIC (overall survival, 19.2, 19.2, and 21.6 months, respectively). Given the variability among different centers, it was difficult to conclude which regional therapy protocol was superior. Nonetheless, a consensus statement was published in 2006 to standardize the CRS-HIPEC treatment protocol and to establish patient selection criteria based on existing evidence.

In a retrospective multi-institutional analysis, Elias and colleagues reviewed 523 patients who underwent CRS-HIPEC/EPIC for colorectal peritoneal metastasis in 23 institutions between 1990 and 2007; 14% of these patients had an unknown primary site. The median overall survival reached 30.1 months and the 5-year survival rate was 27%. The apparent higher survival may have been influenced by better patient selection. In this study, 84% of patients had CC0 level of cytoreduction (vs 54% in the study reported by Glehen and colleagues ). In addition, 65% of patients had limited disease, indicated by a PCI of 12 or less (vs 34% in the study reported by Glehen and colleagues ). In fact, Glehen and colleagues also showed similar survival when the completeness of cytoreduction and low PCI were considered. In a systematic review and meta-analysis of all the CRS-HIPEC studies between 1995 and 2009, Chua and colleagues showed that the median overall survival of patients with complete cytoreduction (N = 722) was 33 months (range, 20–63 months) and the median 5-year survival rate was 43% (range, 20%–51%). In contrast, patients with incomplete cytoreduction had a much poorer outcome, with a median overall survival of only 8 months (range, 4–17 months) and a 5-year survival rate of 0%. These findings again showed that proper patient selection might be the key to success in CRS-HIPEC.

However, these retrospective studies did not address an important question. Given the fact that more than half of the patients received perioperative chemotherapy, the question remains as to whether CRS-HIPEC added a survival benefit. Verwaal and colleagues tested this treatment paradigm on peritoneal metastasis of colorectal cancer in a prospective randomized controlled trial in the Netherlands between 1998 and 2001. This trial randomly assigned 105 patients with isolated peritoneal metastasis of colorectal origin into 2 treatment groups: standard treatment with systemic 5-FU/leucovorin alone (N = 51) and CRS-HIPEC followed by systemic 5-FU/leucovorin (N = 54). With a median follow-up of 21.6 months, this trial showed a survival difference in favor of CRS-HIPEC followed by systemic treatment (22.4 vs 12.6 months in the systemic treatment alone arm; P = .032). With a median follow-up of 8 years, the median disease-specific survival was 22.2 months in the CRS-HIPEC treatment arm, versus 12.6 months in the standard treatment arm. However, with the longer follow-up, 4 patients remained alive in the standard treatment group (2 with disease and 2 without disease) and 5 remained alive in the CRS-HIPEC group (2 with disease and 3 without disease). This trial also demonstrated that complete cytoreduction (ie, residual tumor size<2.5 mm) and limited peritoneal disease were significant prognostic factors in the CRS-HIPEC group.

Although this trial showed an improvement in overall survival associated with CRS-HIPEC followed by systemic treatment, the study was limited by several factors. First, the validity of the trial was compromised by including 18 appendiceal cancers (17% of patient population). Appendiceal mucinous neoplasms and appendiceal adenocarcinomas are sometimes difficult to differentiate on pathologic analysis. The natural history of appendiceal cancers varies widely, and inclusion of any patients with appendiceal cancer as opposed to colorectal adenocarcinoma can compromise the results of the study. Second, although an incremental survival benefit of 10 months was shown in the CRS-HIPEC group, the added benefit may not be equivalent to that associated with modern systemic chemotherapeutic regimens including oxaliplatin or irinotecan. Third, although advocates of CRS-HIPEC emphasize the long-term survival seen in surgical series, the long-term survival in the control arm of the one randomized study was equivalent to the treatment arm. Lastly, whether the added survival benefit was a result of CRS, HIPEC, or both was unclear.

Surgically rendering patients free of all disease is associated with cure for patients with isolated liver or lung metastases. It is reasonable to assume that some patients who have isolated and limited peritoneal metastases that can be resected will experience a cure. Although the surgical series suggest that this may be the case, the independent effects of CRS from HIPEC are difficult to tease out because these treatments are almost always administered together. To address the true efficacy of HIPEC alone, a French multicenter, randomized, controlled trial (Prodige 7) is currently underway. This trial has the goal of randomizing 280 patients with complete cytoreduction into 2 arms (HIPEC using intraperitoneal oxaliplatin and intravenous 5-FU/leucovorin over 30 minutes at a minimum of 42C vs no HIPEC). The primary endpoint is 3- and 5-year overall survival. Although most of the existing data on CRS-HIPEC are based on the use of mitomycin C, the results from this trial may shed some light on whether HIPEC using intraperitoneal oxaliplatin with intravenous 5-FU/leucovorin improves survival. However, this trial will not be able to address the potential added benefit of HIPEC with mitomycin C to CRS in the existing published data. Unfortunately, a multicenter randomized controlled trial addressing this specific question, conducted by Walter Reed Army Medical Center and the American College of Surgeons, was closed prematurely because of a poor accrual.