Originally published as JCO Early Release 10.1200/JCO.2008.18.3152 on October 20 2008

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Google Scholar Articles by Bilchik, A. J. Articles by Choti, M. A. PubMed PubMed Citation Articles by Bilchik, A. J. Articles by Choti, M. A.

Prognostic Variables for Resection of Colorectal Cancer Hepatic Metastases: An Evolving Paradigm

Anton J. Bilchik

California Oncology Research Institute and David Geffen School of Medicine at University of California at Los Angeles, Los Angeles, CA

Graeme Poston

Royal Liverpool and Aintree University Hospitals, Liverpool, United Kingdom

Rene Adam

Assistance Publique—Hôpitaux de Paris Hôpital Paul Brousse, Université Paris Sud, Paris, France

Michael A. Choti

Johns Hopkins School of Medicine, Baltimore, MD

The management of colorectal cancer liver metastases has evolved over the past decade, as reflected by the integration of effective systemic therapies with advanced surgical techniques. Five-year survival rates after resection have increased from 20% to almost 60% in recent series.^1-3 Although the dramatic improvements may be due to improved selection using more sophisticated imaging technology, the introduction of multidrug regimens that include irinotecan, oxaliplatin, cetuximab, and bevacizumab as components of standard treatment has likely considerably improved response rates, resectability rates, and survival.

Standard selection criteria for hepatectomy, including size and number of tumors, are no longer routinely applied, leading to changes in definition as to what constitutes resectable or the potential to become resectable. The definitions of complete clinical response (cCR), radiographic response, and pathologic response (PR) are also evolving, and the potential integration of these into management algorithms is the subject of much debate. Several groups have proposed risk scores to improve the selection of patients for hepatectomy, with conflicting data. Both Nordlinger et al⁴ and Fong et al⁵ have demonstrated that carcinoembryonic antigen (CEA) levels, tumor size and number, disease-free interval, margin status, age, and nodal status of the primary tumor can be effectively applied as components of a risk score. Those with lower risk scores are better candidates for initial hepatectomy and those with higher risk scores are considered for neoadjuvant chemotherapy. Zakaria et al,⁶ however, applied these risk scores to a large group of patients at the Mayo Clinic and found that neither survival nor recurrence were correlated with any of the scoring systems described. The only significant prognostic factors were perioperative blood transfusions and tumor-involved hepatoduodenal lymph nodes.

In this edition of the Journal of Clinical Oncology, Blazer et al⁷ review a large cohort of 305 patients undergoing hepatectomy for colorectal liver metastases who were treated preoperatively with irinotecan- or oxaliplatin-based chemotherapy. Resected tumors were examined by pathologists blinded to clinical information, treatment regimens, and outcome. PR was reported as the mean of percentage of cancer cells remaining in each tumor. By multivariate analysis, only surgical margin status and PR were predictors of survival; a complete pathologic response (pCR) was associated with a 5-year survival rate of 75%. This outstanding survival, however, applied to a minority of patients (9%), but was significantly higher than those with a major or minor PR.

These data are encouraging, because they suggest that not only is PR a predictor of survival after preoperative chemotherapy but that the degree of response may be an important prognostic indicator. Furthermore, because biologic response can be measured, the selection of adjuvant chemotherapy for patients can be improved (eg, tailoring postoperative therapy based on PR). A major limitation, however, is that surgical resection is required to obtain this information and is therefore not applicable in the preoperative setting. A bigger challenge is to improve the selection and the timing of patients for resection based on preoperative assessment of response, as well as perhaps determining whether patients who have had a cCR can be observed without surgery. In this study, correlation between radiologic responses with PR was not reported. We know, however, that based on standard response criteria, preoperative imaging has not been accurate in determining a pCR:⁸ viable tumor cells are found in the majority of pathologic specimens whose tumors are not visualized radiographically. Furthermore, these tumors have been shown to recur if not resected.

Adam et al,⁹ in a recent analysis of 767 patients, demonstrated that a pCR was identified in only 4% patients, and none of these patients had a cCR (disappearance of initial tumor sites). Similar to the Blazer study,⁷ the 5-year survival of those with pCR was 76%, compared with a 45% 5-year survival rate in those without a pCR. The probability of a pCR was associated with several factors (age > 60 years, tumor size < 3 cm, and CEA levels < 30 ng/mL), suggesting that chemotherapy is likely to be more effective with a lower tumor burden. Another recent study by Tan et al¹⁰ at Washington University, St Louis, MO, a center with one of the largest experiences in the world using positron emission tomography imaging, found viable tumor in almost all patients with colorectal cancer liver metastases who had undergone a complete metabolic response and disappearance by computed tomography. These studies demonstrate that despite improvements in the sensitivity and accuracy of imaging, we do not have any accurate radiographic methods that can predict pCR, arguing in favor of resection even in those patients who have had a cCR.

It is also apparent from studies using preoperative therapy in other tumor types, such as esophageal and rectal cancer, that the duration of therapy and the period of waiting before resection can impact PR. Therefore, PR in retrospective analyses should be interpreted with caution. In addition, it is important to understand the intent of preoperative chemotherapy when interpreting PR. Patients with initially unresectable or borderline resectable cancer may be offered preoperative conversion chemotherapy to elicit a response and improve resectability. In such an approach, chemotherapy is administered for a variable duration with careful monitoring for radiologic response and conversion to a resectable state. Only those with a significant response are offered surgery, enriching those who may have a PR. In contrast, patients with initially resectable disease can be offered preoperative chemotherapy as a neoadjuvant approach, typically for a short duration before surgery, and most of these patients will undergo surgery, regardless of the extent of response. Depending on the intent of therapy, the regimen choice, the duration, and the selection for surgery can be quite different. This was not specified in the Blazer study.⁷ The true incidence of PR and its relationship with choice and duration of chemotherapy will need to await analyses of prospective trial data.

The poor correlation between preoperative studies and PR has also resulted in a more complex therapeutic dilemma for the hepatobiliary surgeon. What is the best approach to manage the disappearing liver metastases? Should a blind hepatectomy be performed based on the original imaging studies, or should the resection be performed once the tumor has recurred and is radiographically visible? It also complicates the informed consent process, in which the complications of a hepatectomy are discussed despite the possibility of no tumor being detected in the resected specimen. Some patients and physicians may construe the finding of a pCR as the measure of an unnecessary hepatectomy. This situation can be avoided by regular dialogue between the medical and surgical oncologist to optimize the timing of surgery and to resect the tumor while it is still visible. The shift in the paradigm for the treating oncologist is therefore not to treat until maximal response but rather to provide a limited course of chemotherapy until a response or stabilization has occurred or until a borderline resectable patient becomes resectable. This also has the advantage of reducing perioperative complications associated with the hepatotoxic effects of prolonged exposure to chemotherapy.¹¹

As the treatment options for patients with metastatic colorectal cancer continue to expand, prospective studies are needed to evaluate biochemical, pathologic, and possible molecular criteria to guide therapeutic decisions. The remarkable survival reported by two recent series^7,9 confirm that patients after hepatectomy and a response to chemotherapy have survival rates approaching stage III and even high-risk stage II colon cancer. Although this represents the minority of patients overall, it may include 30% responders (younger than 60 years of age with a maximum tumor size < 3 cm and a low CEA level⁹).

As chemotherapy is increasingly being recommended in patients with resectable disease and as more effective systemic chemotherapy and biologic agents are developed, the incidence of pCR will likely increase in patients with limited as well as advanced metastatic disease. Perhaps in the future, improvements in imaging techniques will be better predictors of the rare circumstance of pCR, thereby avoiding surgery in these patients. At present, however, surgery continues to be the only curative option in patients with colorectal cancer hepatic metastases.

AUTHORS’ DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST

The author(s) indicated no potential conflicts of interest.

AUTHOR CONTRIBUTIONS

Conception and design: Anton J. Bilchik, Graeme Poston, Rene Adam, Michael Choti

Financial support: Anton J. Bilchik

Administrative support: Anton J. Bilchik

Provision of study materials or patients: Anton J. Bilchik

Collection and assembly of data: Anton J. Bilchik, Graeme Poston, Rene Adam, Michael Choti

Data analysis and interpretation: Anton J. Bilchik, Graeme Poston, Rene Adam, Michael Choti

Manuscript writing: Anton J. Bilchik, Graeme Poston, Rene Adam, Michael Choti

Final approval of manuscript: Anton J. Bilchik, Graeme Poston, Rene Adam, Michael Choti

ACKNOWLEDGMENTS

Supported by the Davidow Charitable Fund, the Nancy and Bruce Newberg Charitable Fund, the Sequoia Foundation for Achievement in the Arts and Education, Ruth Weil, and Marguerite Perkins Mautner (Los Angeles, CA).

NOTES

published online ahead of print at www.jco.org on October 20, 2008

REFERENCES

1. Fernandez FG, Drebin JA, Linehan DC, et al: Five-year survival after resection of hepatic metastases from colorectal cancer in patients screened by positron emission tomography with F-18 fluorodeoxyglucose (FDG-PET). Ann Surg 240:438-447, 2004[CrossRef][Medline]

2. Choti MA, Sitzmann JV, Tiburi MF, et al: Trends in long-term survival following liver resection for hepatic colorectal metastases. Ann Surg 235:759-766, 2002[CrossRef][Medline]

3. Pawlik TM, Scoggins CR, Zorzi D, et al: Effect of surgical margin status on survival and site of recurrence after hepatic resection for colorectal metastases. Ann Surg 241:715-722, 2005[CrossRef][Medline]

4. 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 Francaise de Chirugie. Cancer 77:1254-1262, 1996[CrossRef][Medline]

5. Fong Y, Fortner J, Sun RL, et al: Clinical score for predicting recurrence after hepatic resection for metastatic colorectal: Analysis of 1001 consecutive cases. Ann Surg 230:309-318, 1999[CrossRef][Medline]

6. Zakaria S, Donohue JH, Que FG, et al: Hepatic resection for colorectal metastases: Value for risk scoring systems? Ann Surg 246:183-191, 2007[CrossRef][Medline]

7. Blazer D, Kishi Y, Maru DM, et al: Pathologic response to preoperative chemotherapy: A new outcome end point after resection of hepatic colorectal metastases. J Clin Oncol doi: 10.1200/JCO.2008.17.5299[Abstract/Free Full Text]

8. Benoist S, Brouquet A, Penna C, et al: Complete response of colorectal liver metastases after chemotherapy: Does it mean cure? J Clin Oncol 24:3939-3945, 2006[Abstract/Free Full Text]

9. Adam R, Wicherts DA, deHaas RJ, et al: Complete pathologic response after preoperative chemotherapy for colorectal liver metastases: Myth or reality? J Clin Oncol 26:1635-1641, 2008[Abstract/Free Full Text]

10. Tan MC, Linehan DC, Hawkins WG, et al: Chemotherapy-induced normalization of FDG uptake by colorectal liver metastases does not usually indicate complete pathologic response. J Gastrointest Surg 11:1112-1119, 2007[CrossRef][Medline]

11. Bilchik AJ, Poston G, Curley SA, et al: Neoadjuvant chemotherapy for metastatic colon cancer: A cautionary note. J Clin Oncol 23:9073-9078, 2005[Free Full Text]

This Article Full Text (PDF) Purchase Article View Shopping Cart Alert me when this article is cited Alert me if a correction is posted Services Email this article to a colleague Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Save to my personal folders Download to citation manager Rights & Permissions Google Scholar Articles by Bilchik, A. J. Articles by Choti, M. A. PubMed PubMed Citation Articles by Bilchik, A. J. Articles by Choti, M. A.