Annals of Surgical Oncology 10:227-233 (2003)
© 2003 Society of Surgical Oncology
Resection of Locally Recurrent Colorectal Cancer in the Presence of Distant Metastases: Can It Be Justified?
J.E. Hartley, MD,
R.A. Lopez, MD,
P.B. Paty, MD,
W.D. Wong, MD,
A.M. Cohen, MD and
J.G. Guillem, MD, MPH
From the Colorectal Service, Memorial Sloan-Kettering Cancer Center, New York, New York.
Correspondence: Address correspondence and reprint requests to: J. G. Guillem, FACS, Colorectal Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, 1275 York Avenue, New York, NY 10021; Fax: 646-422-2318; E-mail: guillemj{at}mskcc.org
 |
ABSTRACT
|
|---|
Background: We aimed to determine the outcome of resections for local recurrence of colorectal carcinoma in the presence of distant (M1) disease.
Methods: Patients who underwent resection of local recurrence in the presence of potentially resectable M1 disease were identified from the colorectal database. Outcome was determined by chart review.
Results: Forty-two patients (23 men) of mean age 60 years (range, 34–88 years) underwent complete gross resection of their local recurrence in the presence of M1 disease. Thirteen of the 42 underwent synchronous M1 resections to render them free of gross disease (R0). Nine of the 29 patients who left with residual disease (R1) subsequently underwent staged M1 resection, so that 22 of 42 were rendered R0 by surgery. The median survival of all patients was 14.5 months (interquartile range, 6–30 months), and that of patients rendered R0 was 23 months (interquartile range, 10–37 months), in comparison with 7 months (interquartile range, 3–25 months) for those of R1 status (P = .006; log-rank method). Ability to achieve R0 status by synchronous or staged resection was the only factor predictive of survival.
Conclusions: The presence of M1 disease per se should not preclude resection of local recurrence, although case selection is problematic.
Key Words: Colorectal cancer • Distant metastasis • Local recurrence • Resection
|
INTRODUCTION
|
|---|
Approximately 1 in 20 of the population of North America and most Western countries will develop colorectal cancer (CRC), of whom perhaps half will be treated with curative intent.1 Nevertheless, approximately 20% to 30% of such patients will go on to develop locoregional recurrence, with half of these amenable to salvage surgery aimed at improving the duration and quality of survival.2–6 Extensive literature suggests that resections for recurrence can be undertaken with acceptable levels of operative mortality and major morbidity and that highly selected patients may go on to long-term survival, although few are cured of their disease.7–12
Conventional teaching is that such resections for recurrent disease should be reserved for patients in whom there is no evidence of distant metastatic disease. This is because additional sites of disease have been believed to render aggressive local therapy futile.13,14 However, the role for resection of distant metastases from CRC is supported by 5-year survival rates of 25% to 35% for hepatic15–17 and pulmonary18,19 resection in carefully selected patients. Recent years, have, therefore seen the increasing use of synchronous or staged resections of locoregional recurrence and metastatic disease in highly selected patients.11,12 However, the utility of such aggressive approaches remains unproven.
This study examined our institutional experience with resections for recurrent colorectal carcinoma in the presence of distant metastases. Our aims were to determine the outcome of synchronous and staged resection and, secondarily, to identify factors predictive of both resectability and favorable outcome.
|
METHODS
|
|---|
A retrospective review of the prospective Colorectal Database was undertaken for the years 1987 to 1999. Records of those patients who had undergone an abdominal resection for recurrent CRC (excluding isolated liver resection) were identified. Those patients who had undergone complete resection of locoregional recurrence in the presence of either liver or lung metastases and those who had peritoneal or visceral involvement beyond the field of resection were studied in detail. Office charts, operative details, and pathology reports were reviewed. Survival rates were calculated with the Kaplan-Meier method. Comparisons of survival rates between groups were made with the log-rank method. Multivariate and univariate analysis was undertaken with the Cox proportional hazard method. Variables were entered into the Cox model at the significance level of .10 and were retained at a significance level of .05. Significant factors in the Cox analysis were expressed as relative risk and 95% confidence intervals. Statistical analyses were undertaken with SPSS software (SPSS Inc., Chicago, IL).
|
RESULTS
|
|---|
Four hundred eleven patients underwent a colorectal resection for locoregional recurrence of colorectal carcinoma during the defined period. Of these, 42 patients (23 men) of mean age 60 years (range, 34–88 years) underwent a complete gross locoregional resection in the presence of distant disease and are the focus of this study. These patients had undergone a range of resections for primary lesions for a range of pathologic stages (Table 1). Thirty-two patients received adjuvant therapy either before or after primary resection; this consisted of chemotherapy in 26 patients, radiotherapy in 5, and combined-modality therapy in 5. The median disease-free interval was 14 months (interquartile range, 7–24 months). Nineteen patients were symptomatic at the time of presentation (Table 2). The carcinoembryonic antigen level was >5 ng/mL (median, 13.0 ng/mL; interquartile range, 7.0–58.8 ng/mL) in 28 of the 35 patients in whom this assay was performed.
The sites of distant disease are listed in Table 3. Thirteen of the 42 patients underwent synchronous resection of their distant disease to render them free of gross disease (termed R0 for the purposes of this study; Table 3). Most often, this involved small-bowel resection to remove peritoneal disease outside the field of resection, but liver and other visceral resections were undertaken, as was pulmonary wedge resection. The remaining 29 patients were left with residual distant disease after initial surgery for recurrence (termed R1). These resections were undertaken with a major morbidity rate of 17% (Table 4) and a median postoperative stay of 10 days (interquartile range, 8–14 days). There was no operative mortality. Nine of these 29 patients underwent staged resection of their metastatic disease (hepatic resection in 8 patients and pulmonary wedge resection in 1) so that a total of 22 patients (52%) were left free of gross disease (R0) after either single or staged surgical resection. Two of these patients had positive microscopic margins of excision. Of those 20 patients who were left with residual distant disease (R1), 8 were thought at the time of resection of their recurrence to be potential candidates for resection of their metastatic disease but were not subsequently resected (Table 5). The remaining 12 patients were left with residual gross peritoneal or distant nodal and/or visceral disease that was deemed unresectable at the time of resection of their recurrence. None of the data points examined were of any predictive value with regard to the ability to surgically free a patient of gross disease (Table 6).
View this table:
[in this window]
[in a new window]
|
TABLE 3. Details of the locoregional resections undertaken, the site of distant metastatic disease, and the nature of the synchronous resection of that disease
|
|
The median survival of the entire study population was 14.5 months (interquartile range, 6–30 months). Survival curves for all patients according to their principal site of distant disease are shown in Fig. 1. These curves were not significantly different. The median survival of patients rendered R0 was 23 months (interquartile range, 10–37 months), and that of patients left with residual disease was 7 months (interquartile range, 3–25 months); this value was significant on log-rank analysis (P = .006). These survival curves are shown in Fig. 2. The ability to achieve an R0 status via either a single or staged resection of local recurrence and metastatic disease was the only factor associated with survival on univariate and multivariate analysis (Table 7).
View larger version (13K):
[in this window]
[in a new window]
|
FIG. 1. Overall survival after resection of locoregional recurrence of colorectal cancer according to the site of distant metastasis.
|
|
View larger version (11K):
[in this window]
[in a new window]
|
FIG. 2. Cumulative survival after R1 and R0 resections for recurrent colorectal cancer in the present of distant metastatic disease.
|
|
View this table:
[in this window]
[in a new window]
|
TABLE 7. Prediction of survival after resection of local recurrence of colorectal cancer in the presence of distant metastases
|
|
The median disease-free survival of patients rendered R0 was 13.5 months (interquartile range, 9–19 months). Kaplan-Meier curves are shown in Fig. 3. The site of first recurrence was locoregional in three patients, peritoneal in one, hepatic in three, pulmonary in two, and in nonregional lymph nodes in one.
View larger version (10K):
[in this window]
[in a new window]
|
FIG. 3. Cumulative disease-free and overall survival after R0 resection of recurrent colorectal cancer in the presence of distant metastatic disease.
|
|
|
DISCUSSION
|
|---|
There is increasing interest in surgical techniques for the management of locally recurrent or metastatic CRC. These trends toward surgical excision are based on the results from specialist centers operating on highly selected series of patients.7–12,20 It is therefore difficult to determine the applicability of results from such series to the individual patient. However, it is generally accepted that resections for recurrence should be undertaken in patients without metastatic disease beyond the field of resection and, similarly, that metastatic disease should not be resected in the face of local recurrence.13,14 The question of whether local recurrence and metastatic disease should be approached surgically during either single or staged procedures has yet to be adequately examined. Thus, of 30 patients undergoing resection for recurrent rectal cancer reported in a recent series, 8 underwent synchronous resection of metastases, which were ovarian in 4 and hepatic in only 1 patient.12 Similarly, Maetani et al.11 reported similar synchronous resection of hepatic metastases in 4 of 59 patients undergoing resection for recurrent rectal cancer. Our data therefore represent the most significant attempt to date to address this issue.
This experience confirms that in highly selected patients, extended resections for recurrent disease, with either synchronous or staged resection of metastatic disease, can be undertaken with acceptable rates of operative morbidity and mortality. Our results in this regard are comparable to those reported after isolated resection of recurrent rectal cancer.8,9,12,20 However, although reassuring, comparisons between such heterogenous populations should be viewed with caution because case selection is clearly a major factor in the ability to tolerate radical surgery. Given the overtly aggressive stance taken in this series of patients, it is salutary to note that only half of those in whom it was thought that all gross disease might be technically resectable were in fact rendered free of disease by single or staged intervention. The inclusion of those patients with unresectable distant disease might be believed to represent a self-fulfilling prophecy in terms of survival analysis. However, we thought it important to document what most surgeons believe to be intuitive—that is, that those left with residual disease will do badly. In addition, only half of those patients who underwent a resection of their intra-abdominal recurrence and were thought to be candidates for resection of their metastatic disease were subsequently resected. It is possible that newer imaging modalities, such as positron emission tomography21,22 or magnetic resonance imaging,23 will affect case selection. However, at present these odds are important in counseling the patient before exploration.
The median survival of all patients in this selected series is, at 14.5 months, similar to that which might be expected in an unselected series of patients with stage IV CRC.13 It is interesting that the site of the distant disease per se did not seem to have a bearing on outcome in this series. However, this may be a reflection of the relatively small numbers involved in each subgroup, and we would certainly not suggest that peritoneal seedlings from CRC should be regarded as comparable to resectable liver or lung metastases in terms of prognosis. The ability to surgically render patients free of gross disease was the only factor found to be predictive of survival in this series. The median survival of those surgically cleared of disease was more than 3-fold that of those left with residual disease. This finding is consistent with other reports of surgery for local recurrence.10,20 In addition, other groups have reported such variables as high preoperative carcinoembryonic antigen7,9,11 and symptoms at the time of presentation of recurrence8 to be of prognostic significance.
Margin status seems to be of critical importance, as others have reported a median survival of only 10 months in patients with positive resection margins after abdominosacral resections for pelvic recurrence.9 The incidence of positive microscopic margins in this study is somewhat lower than has been reported elsewhere, for reasons that are unclear. However, these patients were highly selected, and it is possible that those with questionable margins at the time of resection of their recurrence did not go on to have their metastatic disease resected.
The overall survival of patients freed of gross disease by synchronous or staged resection of local recurrence and metastases is not dissimilar to that reported after resection of isolated hepatic16,17 or pulmonary18,19 metastases or locally recurrent rectal cancer.7–11,20 Unfortunately, although such patients may be approached aggressively and with curative intent, few will be cured of their disease. This is reflected in the correspondingly lower disease-free survival observed in this series, within which there was only one 5-year survivor. Thus, taking a more esoteric view, it is uncertain whether surgical intervention affected the natural history of the disease process or whether this represents an example of experienced clinicians selecting patients with a favorable tumor biology. It seems likely, therefore, that increasing understanding of the molecular markers of tumor behavior and, therefore, outcome will assist in case selection in years to come. However, given that few patients with recurrent CRC will be cured of their disease or, indeed, enjoy protracted disease-free survival, perhaps the most important justification for aggressive surgical intervention is to improve the quality of life in patients by alleviating the symptoms associated with recurrence. Unfortunately, there are few data pertaining to quality of life after surgery for recurrent CRC, and such assessments should be encouraged in future studies.
In conclusion, therefore, these data challenge traditional surgical oncology teaching and suggest that the presence of distant disease per se should not preclude patients with recurrent CRC from consideration for an aggressive surgical approach. Furthermore, in carefully selected patients, such aggressive surgical intervention with the aim of removing all gross disease may be associated with long-term survival. Continued advances in systemic and locoregional chemotherapeutic approaches can only improve the results of radical surgery in this otherwise dismal scenario.
|
Footnotes
|
|---|
Presented to the Society of Surgical Oncology, Washington, DC, March 2001.
Patients with locally recurrent colorectal cancer and distant metastatic disease can be resected to a disease-free status by synchronous or staged procedures and experience prolonged survival but are rarely cured. Although salvage surgery should be considered in these complex cases, selection is problematic.
Received for publication May 16, 2001.
Accepted for publication October 28, 2002.
|
REFERENCES
|
|---|
- Cohen AM, Minsky BD, Schilsky RL. Cancer of the colon. In: Devita VT, Hellman S, Rosenberg SA, eds. Cancer—Principles and Practice of Oncology. 5th ed. Philadelphia: Lippincott, 1997.
- Pilipshen SJ, Heilwell M, Quan SH, Sternberg SS, Enker WE. Patterns of pelvic recurrence following definitive resections of rectal cancer. Cancer 1984; 53: 1354–62.[CrossRef][Medline]
- Gunderson LL, Sosin H. Areas of failure found at reoperation (second or symptomatic look) following "curative surgery" for adenocarcinoma of the rectum. Cancer 1974; 34: 1278–92.[CrossRef][Medline]
- Rao AR, Kagan AR, Chan PM, Gilbert HA, Nussbaum H, Hintz BL. Patterns of recurrence following curative resection alone for adenocarcinoma of the rectum and sigmoid colon. Cancer 1981; 48: 1492–5.[CrossRef][Medline]
- Rich T, Gunderson LL, Lew R, Galbidini JJ, Cohen AM, Donaldson G. Patterns of recurrence of rectal cancer after potentially curative surgery. Cancer 1983; 52: 1317–29.[CrossRef][Medline]
- Welch JP, Donaldson GA. Detection and treatment of recurrent cancer of the colon and rectum. Am J Surg 1978; 135: 505–11.[CrossRef][Medline]
- Salo JC, Paty PB, Guillem J, Minsky BD, Harrison LB, Cohen AM. Surgical salvage of recurrent rectal carcinoma after curative resection: a ten year experience. Ann Surg Oncol 1999; 6: 171–7.[Abstract]
- Gagliardi G, Hawley PR, Hershman MJ, Arnott SJ. Prognostic factors in surgery for local recurrences of rectal cancer. Br J Surg 1995; 82: 1401–5.[Medline]
- Wanebo HJ, Koness RJ, Vezeridis MP, Cohen SI, Wrobleski DE. Pelvic resection of recurrent rectal cancer. Ann Surg 1994; 220: 586–97.[CrossRef][Medline]
- Tschmelitsch J, Kronberger P, Glaser K, Klingler A, Bodner E. Survival after surgical treatment of recurrent carcinoma of the rectum. J Am Coll Surg 1994; 179: 54–8.[Medline]
- Maetani S, Onodera H, Nishikawa T, et al. Significance of local recurrence of rectal cancer as a local or disseminated disease. Br J Surg 1998; 85: 521–5.[CrossRef][Medline]
- Huguier M, Houry S. Treatment of local recurrence of rectal cancer. Am J Surg 1998; 175: 288–92.[CrossRef][Medline]
- Eckhauser FE, Knol JA. Surgery for primary and metastatic colorectal cancer. Gastroenterol Clin North Am 1997; 26: 104–28.
- Sagar PM, Pemberton JH. Surgical management of locally recurrent rectal cancer. Br J Surg 1996; 83: 293–304.[Medline]
- Gayowski TJ, Iwatsuki S, Madariaga JR, et al. Experience in hepatic resection for metastatic colorectal cancer: analysis of clinical and pathologic risk factors. Surgery 1994; 116: 703–10;discussion 710–1.
- Hughes KS, Simon R, Songhorabodi S, et al. Resection of the liver for colorectal carcinoma metastases: a multi-institutional study of indications for resection. Registry of Hepatic Metastases. Surgery 1988; 103: 278–88.[Medline]
- Scheele J, Stangl R, Altendorf-Hofmann A. Hepatic metastases from colorectal carcinoma: impact of surgical resection on the natural history. Br J Surg 1990; 77: 1241–6.[Medline]
- Goya T, Miyazawa N, Kondo H, et al. Surgical resection of pulmonary metastases from colorectal cancer: 10-year follow-up. Cancer 1989; 64: 1418–21.[CrossRef][Medline]
- Sauter ER, Bolton JS, Willis GW, et al. Improved survival after pulmonary resection of metastatic colorectal carcinoma. J Surg Oncol 1990; 43: 135–8.[Medline]
- Suzuki K, Dozois RR, Devine RM, et al. Curative reoperations for locally recurrent rectal cancer. Dis Colon Rectum 1996; 39: 730–6.[CrossRef][Medline]
- Flanagan DL, Dehdashti F, Ogunbiyi OA, Kodner IJ, Siegel BA. Utility of FDG-PET for investigating unexplained plasma CEA elevation in patients with colorectal cancer. Ann Surg 1998; 27: 319–23.
- Guillem JG, Puig-La Calle J Jr, Akhurst T, et al. Prospective assessment of primary rectal cancer response to preoperative radiation and chemotherapy using 18-fluorodeoxyglucose positron emission tomography. Dis Colon Rectum 2000; 43: 18–24.[CrossRef][Medline]
- Dicle O, Obuz F, Cakmakci H. Differentiation of recurrent rectal cancer and scarring with dynamic MR imaging. Br J Radiol 1999; 72: 1155–9.[Abstract]
TOP
ABSTRACT
INTRODUCTION
