ABSTRACT: The treatment of resectable colorectal cancer metastases to the liver has undergone changes with the addition of active chemotherapeutic agents, innovations and definition in the surgical procedures, understanding of the benefits and toxicities of the surgical and chemotherapeutic (cytotoxic and biologic) interventions, and use of the team approach. Patients are initially evaluated for the overall risk of their disease, which includes the standard parameters for disease recurrence and blends in disease and patient comorbidities and likelihood of surgical success. Advanced imaging techniques are mandatory in the initial evaluation. Rather than approaching the patient with sequential, independent therapies and handoff from specialist to specialist, a continuous interaction is required. This article outlines the initial consultation, required team components, surgical decision-making, and use of cytotoxic and biologic agents. The implication is that the best outcomes can only be achieved with the use of all modalities.
The combined-modality care of the patient with colon or rectal cancer metastatic to the liver demands a team approach. It is little wonder that there is much confusion about this topic, given the number of unique treatment options that are delivered in a sequential and reiterative process. The concept of multidisciplinary approaches to complex cancer challenges has been adopted for a variety of tumor types and situations.
Questions of therapy sequencing, therapy-associated morbidity, and disease-related morbidity that limit the use of multiple interventions have necessitated an approach that is data-driven and response-modulated. For patients with colon or rectal cancer metastatic to the liver, both oncologic and functional outcomes are important. The interaction of the modalities must be considered as well as the relative benefits of any single one.
The components of the multidisciplinary team include those with expertise in:
• Diagnostic Radiology—This specialty is important for maintaining consistency during review of the initial films, recommending appropriate additional evaluative studies, and consistently reporting with standard oncologic criteria, such as Response Evaluation Criteria In Solid Tumors (RECIST), on the outcome of therapy. As a subset of the diagnostic radiologist, interventional radiologists may extend the skill set into the invasive procedures.
• Medical Oncology—These team members play a key role in assessing the drug or drug combinations that will have selected tumor responses and patient toxicity. Evaluation of specific toxicities such as those involving the gastrointestinal tract, related to age or comorbidities (hypertension) and the unique interaction of the biologic agents bevacizumab(Drug information on bevacizumab) (Avastin) and cetuximab(Drug information on cetuximab) (Erbitux) on wound healing and skin effects.
• Surgical Oncology—The surgical oncologist has a responsibility to assess resectability and properly weigh the surgical complications of bevacizumab associated with wound healing, which are well established. In addition, experience with cetuximab-induced rash and skin changes can be very important in planning a successful, uncomplicated operative procedure.
The overarching question posed to the surgeon is whether all tumors can be destroyed (resected or ablated). In a simpler era, this was a question of whether the tumor could be removed with a negative margin in a single operative procedure. Today, the primary consideration for the surgeon is not so much the location or size of the tumors but rather the volume and function of the liver postoperatively. Mathematical models and the use of computerized tomograms have been developed to estimate the functional liver remnant after surgical resection and correlate that with outcome. Once again, multidisciplinary expertise—including reliance on the radiologist in measuring total liver volume and postoperative liver volume—is critical.
The use of ablative tools such as cryoablation , radiofrequency ablation, or microwave ablation has expanded the population of patients who can be rendered metastasis-free (R0 resection) and in some cases make the need for large resections of otherwise normal tissue unnecessary. The introduction of the two-stage hepatectomy to achieve R0 status is at the same time radical, innovative, and appropriate. Surgeons specially trained in liver resection and hepatobiliary surgery are now accepting the challenge of designing operations and treatment plans that provide maximum resection and extremely low morbidity. The surgical procedures are coupled with other components of care for this patient population. Selection of a surgeon and surgical team with expertise in the assessment of postoperative liver remnant, multimodality surgical techniques, and intraoperative ultrasound is required.
Surgical organizations representing the expertise of surgical oncology and hepatobiliary surgery have come together to define a series of parameters that define resectability. Rather than focusing on the amount of liver removed, they have redirected the resectability analysis on the portion of the liver remaining at the end of the resection. The consensus panel recommended that there be two contiguous Couinaud’s segments with intact vascular inflow and outflow and biliary drainage. Such definitions of resectability often exceed the technical skills of the general surgeon and require specific training in surgical oncology or hepatobiliary surgery. Thus, it is important to recognize that these definitions are made with an expectation of advanced training and experience. Finally, the surgical decision of resectability must also be balanced against patient comorbidities and disease-based risk.
Several large studies have reviewed risk factors for operative morbidity and the recurrence of colorectal cancer metastatic to the liver.[7-9] The primary disease-specific risk factors include the original stage of the colon or rectal cancer and the disease-free interval between resection of the primary and identification of metastatic disease. Factors related to the metastatic disease include the size of the largest metastasis, number of metastases, patient’s age, CEA level, and ability to completely resect the metastases. These models all exclude patients with extrahepatic disease including metastasis to organs other than the liver, peritoneal implants, and lymph nodes such as those in the porta hepatis and retroperitoneum. The presence of metastatic nodal disease in the porta hepatis has been reported and debated as a relative and not absolute contraindication to curative resection.[10-13]
The above enumerated risk factors are accumulated to determine their overall negative impact on disease-free liver–specific recurrence-free and overall survival. Then, they are used as relative risk factors in modifying the multimodality decision-making (Table 1). For example, the patient with an unfavorable disease-free interval of less than 12 months, multiple tumors, and an age greater than 60 years might be considered for neoadjuvant chemotherapy to assess response of measureable disease, failure outside the liver, and in vivo sensitivity of the metastases to a given agent. These decision-making concepts will be further explored in this paper.
Synchronous Resection of Colorectal Primary and Liver Disease
Although uncommon, the synchronous presentation of resectable liver disease with an intact colon or rectal primary presents a unique scenario. Prior to the widespread use of preoperative chemotherapy and radiation for rectal cancers, the general approach to the primary tumor was resection at initial identification. Currently, with the recognition of lower morbidity and equivalent control with continuous neoadjuvant chemotherapy and concomitant radiation to the rectal tumor and pelvis, the thought process toward initial rectal resection has changed. Similarly, it is more common to consider the synchronous resection of primary colon cancer and liver metastases either at presentation or following neoadjuvant chemotherapy.
Major discussion has centered around the treatment of patients with this presentation, but controversy remains as to whether the operation on the liver and bowel should be performed as a single operative experience or in a staged fashion. Often this decision is based on the intraoperative finding of a liver metastasis when the surgeon is resecting the primary colon tumor. The major issues involved in this situation include the operative approach and skill of the surgeon. For example, it may be difficult to perform a complete, manual assessment of the liver and abdominal cavity intraoperatively due to limited access with the laparoscopic approach. The required skill with intraoperative ultrasound and availability of the correct probe type may also hinder complete evaluation. Finally, intraoperatively there is the inability to obtain fully informed consent for a procedure with known morbidity and mortality. In these situations, it is clearly preferable to complete the diagnostic evaluation postoperatively, seek expert opinion, and stage the patient fully (to include a positron-emission tomography [PET] scan) prior to resection of the liver disease. Preoperative staging with computed tomography [CT] scans that include the liver are more commonly done with larger tumors, and liver disease is often clearly identified and deemed resectable.
If all disease is encompassible at a single operation (based on patient characteristics, tumor characteristics, and surgical skill set—either a single surgeon with skill in resection of bowel tumors and resection of liver metastases or two separate teams), this approach is scientifically sound and preferable. Prospective studies have reviewed the associated morbidity with synchronous or metachronous approaches. The conclusions have clearly favored a combined approach in nearly all clinical scenarios. The total length of stay during a single hospitalization for a synchronous procedure is not thought to exceed that of the two hospitalizations associated with the metachronous approach.
The one legitimate decision point is related to a difficult operative encounter during the resection of the primary disease (ie, colon resection). Such encounters may include excessive blood loss, unprepped bowel, or hypotension. Another cautionary scenario involves a complex or extensive hepatic resection, where morbidity associated with liver failure or infection is of significant concern.
In most situations, a combined surgical and chemotherapy approach to the metastases is recommended. The timing of treatment with neoadjuvant chemotherapy is crucial, not only for the associated toxicity, but because of the impact of response on the operative intervention. Prior approaches of preoperative chemotherapy for high-risk, resectable metastases that have produced a “best response” led to the ironic conundrum of dealing with lesions that had disappeared. A more confounding situation is one in which a mixed response results in the disappearance of some lesions but persistence of others. One might consider simply following with serial CT and PET scans of the single lesion that has disappeared. However, this is not a reasonable strategy for patients in whom a curative resection could be performed if not for the fact that only a subset of the original multiple lesions requiring resection are visible.
An interesting retrospective review of patients treated to disappearance of disease demonstrated a remarkable recurrence rate or persistence rate of close to 80%. These data have led to an obligatory sequential evaluation at set time points, usually defined by number of chemotherapy cycles and the status of the lesions. Because the resolution in identification of lesions by CT scan, intraoperative ultrasound, and intraoperative palpation is approximately 1 cm, it is necessary to time the surgery with the change in tumor size. When lesions become smaller than 1 cm they are difficult to identify intraoperatively. In situations where a lesion is not identified, the surgeon should give serious consideration to a blind resection of the area where the tumor formerly resided. In some settings, the goal is reduction of size to change the operative procedure from a lobectomy to segmentectomy or to use ablation instead of resection. In this setting, the effect of therapy is carefully monitored to reach the desired endpoint (Figure 1).